Part 1. Basic Knowledge and Recommendations for Perioperative Cardiovascular Management Based on Comprehensive Review of Evidence
I. Overview of Perioperative Cardiovascular Management
1. Epidemiology
Perioperative complications occur in approximately 10% of surgeries, and the crude death rate is approximately 1%.4 The complication can be various,5 and in approximately 42% of cases may comprise cardiovascular and cerebrovascular complications, such as myocardial infarction (MI) and cerebral infarction.4 More importantly, approximately half of the cardiovascular complications are reportedly preventable,6,7 which indicates the importance of preoperative cardiovascular risk assessment and therapeutic planning to reduce the risk.
2. Preoperative Management8
2.1 Preoperative Evaluation
2.1.1 The Purpose of Preoperative Evaluation
The purpose of preoperative cardiovascular evaluation is not to provide cardiac clearance, but rather to evaluate perioperative risk and to assess the management of that risk over perioperative; preoperative, intraoperative and postoperative periods. This assessment should be shared among patients, surgeons, and anesthesiologists.
2.1.2 Medical History and Physical Findings
Preoperative evaluation of the medical history and the physical examination may reveal various, previously undiscovered cardiovascular and non-cardiovascular diseases. Routine evaluation of medication history, lifestyle habits (smoking, alcohol consumption), allergy history, and physical findings is essential. The preoperative evaluation should be considered as a good opportunity for improving long-term prognosis as well.4 For example, instructing smokers to stop smoking immediately would not only reduce perioperative respiratory complications but potentially lower the long-term risk of respiratory and cardiovascular diseases as well.
2.1.3 Preoperative Testing
In considering whether to perform cardiac tests preoperatively to assess perioperative cardiac risk, it is important to consider whether its result will adequately stratify the perioperative risk further, whether it will affect perioperative management as well as the potential delay of non-cardiac surgery and its influence on the outcome of the disease requiring the non-cardiac surgery. This clinical practice guideline does not recommend the routine performance of 12-lead echocardiography (ECG), transthoracic echocardiography or the cardiac stress test.
2.1.4 Importance of Risk Scores
Recent clinical practice guidelines and reviews of preoperative evaluation recommend calculating a risk score from the patient’s background information obtained from a basic medical history, physical findings and basic laboratory tests.4,9–11 The present clinical practice guideline also refers to the Revised Cardiac Risk Index (RCRI), the National Surgical Quality Improvement Program Myocardial Infarction and Cardiac Arrest (NSQIP MICA), and American College of Surgeons NSQIP Surgical Risk Calculator, which will be described in detail (see Chapter II.6 Perioperative Risk Score).
2.2 Preoperative Optimization
After risk stratification, the patient’s medical condition is optimized in preparation for non-cardiac surgery. For example, preoperative optimization for heart failure consists of optimizing hemodynamics, evaluating the primary disease, and introducing maintenance therapy. Of note, chronic heart failure patients undergoing non-cardiac surgery have higher risks of both cardiac and non-cardiac complications (see Section 24).12
3. Potential Interventions for Risk Reduction
The following are examples of assessments that can be performed preoperatively, intraoperatively, or postoperatively to reduce the risk of complications. Examination for coronary artery disease and revascularization tend to be the main focuses of the preoperative cardiovascular assessment, but a comprehensive approach is important for reducing risk.
(1) Preoperative therapeutic interventions for cardiac disease (e.g., pharmacotherapeutic optimization, invasive cardiac intervention)
(2) Close intraoperative monitoring (use of transesophageal echocardiography, arterial line, and pulmonary artery catheters) or re-assessment of the method of anesthesia
(3) Change the type of surgery (i.e., less invasive approach) or considering alternatives to surgical therapy if the risk is very high or prohibitive
(4) Step-up to the intensive care unit (ICU) for postoperative care instead of the general ward
(5) Surveillance of myocardial injury by ECG or troponin monitoring after surgery
4. Preoperative Communication With the Patient on Perioperative Risks and Shared Decision-Making
Shared decision-making (SDM) is particularly important, especially for high-risk surgeries (Figure 1A). The benefits and harms of each treatment option, the patient’s and family’s values (goal of care, fears/worries, unacceptable conditions, trade-offs, and surrogate) (Figure 1B) should be shared between the physician, patient, and family.
In SDM, it is important to ensure that patients and families understand the actual clinical information that is relevant to them. In order to transmit the clinical information accurately, using expressions such as “high” or “low” risk (e.g., “Your risk is high”) or risk ratios (e.g., “Your risk is twice the average”) is not sufficient. Systematic review of RCTs showed that patients understand probabilistic information more accurately in the format of natural frequencies than as words (low, intermediate, high risk), probabilities or summarized as effect measures such as relative risk reduction.14 For example, expressions such as “How many people per hundred” are easier to understand (e.g., “Out of 100 people with the same underlying condition who undergo the same surgery, 2 will die of cardiac complications, and 5 will experience heart failure complications”).10,14 In line with this principle, the risks are presented as predicted incidence (%) to calculate natural frequencies whenever possible throughout the clinical practice guidelines. Furthermore, the Best Case/Worst Case model as a tool of communication is suggested for use in SDM of high-risk surgery.15,16
5. Assessment of the Indication of Invasive Interventions for Cardiac Disease Before Non-Cardiac Surgery
Sometimes, cardiac disease, with a clear-cut indication of invasive treatment, is found during the preoperative evaluation for non-cardiac surgery. For example, multivessel coronary artery disease and symptomatic severe aortic stenosis, which are a Class I indication of revascularization with coronary artery bypass grafting (CABG) and aortic valve replacement, respectively.17,18 The options are listed below.
Option 1: Perform the invasive cardiac procedure first, followed by non-cardiac surgery at an appropriate time
Option 2: Perform the non-cardiac surgery first, followed by the invasive cardiac procedure at an appropriate time
Option 3: Perform either of them
Option 4: Perform neither of them
In selecting treatment in such cases, it is essential to assess the following:
(1) Urgency of the invasive cardiac procedure (e.g., percutaneous coronary intervention, CABG, valve replacement)
(2) Risk of the invasive cardiac procedure (e.g., STS score,19 JapanSCORE,20 EuroSCORE II21)
(3) Urgency of the non-cardiac surgery: how long the non-cardiac surgery can be delayed for the cardiac intervention and to assess the risks and harms of delay. For example, delayed cancer surgery runs the risk of progression, and delaying surgery for a femoral fracture in an older patient beyond 48 h increases the risk of death,22 dependency,23 and complications,22,23 as well as prolonging pain
(4) Risk of non-cardiac surgeries without prophylactic cardiac procedures
(5) Effect of antithrombotic agents, which are necessary following invasive cardiac procedures; that is, the risk of bleeding, and thromboembolism associated with continuation or discontinuation of antithrombotic therapy in the perioperative period
Such a comprehensive assessment from multiple therapeutic standpoints is of paramount importance (Figure 2). For example, among patients who have cardiac disease associated with increased perioperative risk, when estimated risk of delaying the non-cardiac surgery is high and/or the risk of the cardiac procedure is high, so the total risk of Option 1 (cardiac procedure followed by non-cardiac surgery) can be higher than with Option 2 (non-cardiac surgery without invasive cardiac procedure) (Patterns 1, 2 in Figure 2). When the risks of non-cardiac surgery and the invasive cardiac procedure are too high, we should avoid both procedures and consider alternatives. To reiterate, sufficient sharing of clinical information with the patient and taking into consideration the patient’s values and preferences such as the goals of care and burden or suffering they can tolerate, are important in SDM (Figure 1).
6. Perioperative Palliative Care
Palliative care focuses on improving quality of life (QOL), and consists of pain management, communication, such as advance care planning, and coordination of care.24 There have been reports that suggest palliative care improved QOL and the quality of communication as well as perioperative mortality for perioperative patients.25,26
Perioperative complications (cardiac and non-cardiac) can lead to death, impose a burden and suffering on the patient, such as prolonged ICU stay, and result in poor functional status. Serious illness is defined as “a health condition that carries a high risk of mortality and either negatively impacts a person’s daily function or quality of life or excessively strains their caregivers”.27,28 High-risk conditions for serious illness caused by surgery are shown in Table 6.29 Of note, patients with heart failure are included in this condition with a high rate of perioperative complication (see Chapter IV.5 Chronic Heart Failure). Furthermore, even in cases with serious illness, some surgery can improve QOL; for example, surgery for hip fracture in elderly patients. Not only mortality but also QOL is an important factor in SDM30 (Figure 1).
Table 6. Serious Illness
Serial
No. |
Definition |
| 1. |
ASA Risk: Class IV or V |
| 2. |
Vulnerable elder
• Older adult >84 years old
• Older adult >64 years old with any functional or cognitive disability |
| 3. |
Advanced cancer
• Stages III and IV solid cancers OR hematologic malignancies
• AND ≥1 hospitalization in prior year |
| 4. |
Oxygen-dependent pulmonary disease |
| 5. |
Heart failure diagnosis with any all-cause hospitalization or ≥2 ED visits in past 6 months |
| 6. |
Cirrhosis with any Child-Turcotte-Pugh (CTP) class or Model for Endstage Liver Disease (MELD) score |
| 7. |
Endstage renal disease on dialysis or eligible for dialysis |
| 8. |
Dementia with impaired daily function and ≥1 hospitalization in prior year |
| 9. |
Frailty |
| 10. |
Trauma patient
• Severe traumatic brain injury with Abbreviated Injury Scale score of ≥3
• Critical injury (Injury Severity Score >25 or >24 h intensive care unit admission) |
| 11. |
Nursing home resident |
(Modified from Lee KC, et al. 2019.29). ASA, American Society of Anesthesiology; ED, emergency department.
It has been reported that a palliative care consultation was associated with better ratings of overall end-of-life care, communication, and support, as reported by families of patients who died within 90 days of high-risk surgery.31 Preoperative advance care planning (goal of care, patient’s preference, advance directives and designation of surrogate) is recommended for geriatric patients undergoing surgery, according to the ACS NSQIP and the American Geriatrics Society.16,32 Although it has not been acknowledged yet in Japan, palliative care should be considered in cases with serious illness perioperatively.
II. Preoperative Evaluation of Cardiovascular Risk (Table 7)
Table 7. Recommendation and Level of Evidence on Patient Communication and Team Care in the Perioperative Period
| |
COR |
LOE |
Comprehensive discussion of perioperative management among the surgeon,
anesthesiologist, and cardiologist and shared decision making with the patient are
recommended when a cardiac condition requiring invasive cardiac procedure in
non-perioperative settings* is diagnosed preoperatively |
I |
C |
*If the situation is not perioperative, the procedure is idicated. In perioperative seting, whether to perform that procedure preoperative should require comprehensive discussion. COR, class of recommendation; LOE, level of evidence.
The anesthetic considerations to be made in the preoperative period are summarized below.
1.1 The American Society of Anesthesiologists Physical Status (ASA PS)
This is a classification of preoperative patient risk used by the anesthesiologist33 (Table 8). The issue of inter-rater variability has been raised with regard to this classification system, as it is a subjective index. The ASA has since improved the consistency of assessments between raters by providing more specific examples.33
Table 8. ASA PS Classification and ASA-Approved Examples
ASA PS
classification |
Definition |
Adult examples, including, but not limited to: |
| ASA I |
A normal healthy patient |
Healthy, non-smoking, no or minimal alcohol use |
| ASA II |
A patient with mild systemic disease |
Mild diseases only without substantive functional limitations. Current
smoker, social alcohol drinker, pregnancy, obesity (30<BMI<40),
well-controlled DM/HTN, mild lung disease |
| ASA III |
A patient with severe systemic disease |
Substantive functional limitations; One or more moderate to severe
diseases. Poorly controlled DM or HTN, COPD, morbid obesity (BMI ≥40),
active hepatitis, alcohol dependence or abuse, implanted pacemaker,
moderate reduction of ejection fraction, ESRD undergoing regularly
scheduled dialysis, history (>3 months) of MI, CVA, TIA, or CAD/stents |
| ASA IV |
A patient with severe systemic disease
that is a constant threat to life |
Recent (<3 months) MI, CVA, TIA or CAD/stents, ongoing cardiac ischemia
or severe valve dysfunction, severe reduction of ejection fraction, shock,
sepsis, DIC, ARD or ESRD not undergoing regularly scheduled dialysis |
| ASA V |
A moribund patient who is not expected to
survive without the operation |
Ruptured abdominal/thoracic aneurysm, massive trauma, intracranial bleed
with mass effect, ischemic bowel in the face of significant cardiac pathology
or multiple organ/system dysfunction |
| ASA VI |
A declared brain-dead patient whose organs
are being removed for donor purposes |
|
“E” denotes Emergency surgery: (An emergency is defined as existing when delay in treatment of the patient would lead to a significant increase in the threat to life or body part).33
(Excerpt from reference 33 with permission.) A copy of the full text can be obtained from the American Society of Anesthesiologists, 1061 American Lane Schaumburg, IL 60173–4973, USA or online at www.asahq.org.
1.2 Methods of Anesthesia and Hemodynamics
1.2.1 General Anesthesia
Intravenous and inhaled anesthetics are cardiodepressants, and opioids suppress the sympathetic nervous system to cause vasodilation. Thus, administering general anesthesia sometimes results in lower blood pressure, which can be compensated by fluid management or administration of vasopressors as needed. Furthermore, respiration during general anesthesia is generally managed by positive-pressure ventilation. The elevation of intrathoracic pressure by positive-pressure ventilation reduces venous return, thereby decreasing the cardiac output and hypotension.35 Intraoperative fluid infusions for maintaining systemic perfusion that leak into the extravascular tissues36 return to the intravascular space several days after surgery, and may result in pulmonary edema or heart failure. Therefore, adequate fluid management is required to prevent volume overload postsurgically (see Chapter IV.5 Heart Failure).
1.2.2 Neuraxial Anesthesia (Spinal/Epidural Anesthesia) and Hemodynamics
Blocking the sympathetic nerve pathway by neuraxial anesthesia lowers peripheral vascular resistance, thus hypotension occurs in 15–33%37,38 and bradycardia in 9–13%37,38 of cases. The latter occurs especially when blocking the T1–T5 thoracic nerves from which the sympathetic cardiac nerves are derived.39 Furthermore, severe hypotension caused by sympathetic nerve block by spinal anesthesia has been reported in some cases of aortic stenosis, so due caution is needed.40
1.2.3 Body Position and Hemodynamics
Arterial pressure, central venous pressure (CVP), pulmonary arterial wedge pressure (PAWP), and systemic vascular pressure increase in the Trendelenburg position.41,42 Systemic perfusion may appear to be maintained based on blood pressure and CVP in Trendelenburg position, but these parameters may decrease when the patient is returned to the horizontal position.41,42
1.3 Risk of Intraoperative Cardiac Arrest
Since 1992, the Japanese Society of Anesthesiologists (JSA) has conducted studies investigating life-threatening episodes under anesthesia managed by an anesthesiologist at JSA-certified hospitals. In the 2016 report investigating approximately 2.13 million cases, cardiac arrest occurred in 2.30/10,000 individuals, and 1.37/10,000 died within 30 days (≈60%).44 Anesthesia-related cardiac arrest occurred in 0.15/10,000.44 In a study from a different year, the leading preoperative complications associated with death were hemorrhagic shock (26.2%), major bleeding caused by surgery (16.8%), multiple organ failure/sepsis (12.8%), and cardiac complications (12.0%).45 In the USA, the incidence of intraoperative cardiac arrest was 7.2/10,000 according to data from 360,000 surgery patients, and the risk factors were investigated by multivariate analysis,46 which revealed that intraoperative blood transfusion volume was the most important risk factor. The odds ratios (OR) for an emergency procedure, functional dependency, and heart disease were 2.04 (95% confidence interval [CI] 1.45–2.86), 2.33 (95% CI 1.69–3.22) and 1.48 (95% CI 1.11–1.96), respectively.46 These data also suggest the importance of managing heart disease in the perioperative period.
2. Perioperative Cardiovascular Risk According to the Type of Surgery
2.1 Surgical Subspecialty and Risk
The incidence of major cardiovascular/cerebrovascular events and acute myocardial infarction (AMI) in the perioperative period are stratified by surgical subspecialty in Table 9.47,48 Risk is highest in vascular surgery, followed by pulmonary surgery and transplant surgery.
Table 9. Incidence of Risk in Non-Cardiac Surgery by Subspecialty
| |
Major cardiovascular
and cerebrovascular
event* |
Acute myocardial
infarction |
| Vascular |
7.7% |
2.0% |
| Thoracic |
6.5% |
1.5% |
| Transplant |
6.2% |
1.6% |
| Neurosurgery |
4.6% |
0.5% |
| General |
3.8% |
1.0% |
| Skin/burn |
3.8% |
Skin and breast cancer
0.8% |
| Otolaryngology |
1.8% |
0.61% |
| Genitourinary |
1.6% |
0.67% |
| Orthopedic |
0.96% |
0.36% |
| Endocrine |
0.96% |
0.36% |
| Breast |
0.35% |
|
| Gynecology |
0.31% |
0.13% |
| Obstetrics |
0.13% |
0.08% |
*In-hospital all cause of death, myocardial infarction and ischemic stroke. (Adapted from Smilowitz NR, et al. 2017.47,48)
2.2 Type of Surgery and Risk
The incidence of cardiac death and non-fatal AMI within the first 30 postoperative days are shown by type of surgery in Table 10. Event rates <1%, 1–5%, and >5% are classified as low-, intermediate-, and high-risk, respectively.49 The Table was rearranged for this clinical practice guideline by subspecialty or field, with the subspecialties and type of surgery listed in the upper rows associated with higher risk (Table 10).
Table 10. Risk Classification of Non-Cardiac Surgery (Various Departments) by Rate of Cardiac Complications
| |
High risk (>5%) |
Intermediate risk (1–5%) |
Low risk (<1%) |
| Vascular |
Aorta
Thromboembolectomy |
Endovascular repair
Peripheral vascular surgery |
|
| Thoracic |
Pneumonectomy
Lobectomy |
Minor intrathoracic surgery |
|
| Transplant |
Pulmonary or liver transplant |
Kidney transplant |
|
| Neurology |
|
Major neurological surgery
Carotid symptomatic (CEA or CAS) |
Carotid asymptomatic (CEA or CAS) |
Gastroenterology
or
Hepatopancreases |
Esophagectomy
Pancreatoduodenectomy
Liver resection, bile duct surgery
Repair of perforated bowel |
Intraperitoneal; gastrectomy,
colectomy, splenectomy, hiatal
hernia repair, cholecystectomy |
Inguinal hernia repair |
| Urology |
Total cystectomy |
Urological major |
Urological minor (transurethral
resection of prostate) |
| Orthopedic |
Leg amputation |
Major (hip or spine surgery) |
Minor (menisectomy) |
| Endocrinology |
Adrenal resection |
|
Thyroid |
| Gynecology |
|
Major |
Minor |
| Others |
|
Head and neck surgery |
Breast
Reconstructive
Oral surgery
Eye
Superficial surgery |
Surgical risk estimate is 30-day risk of cardiovascular death and myocardial infarction, which takes into account only the specific surgical intervention, without considering the patient’s comorbidities. CAS, carotid artery stenting; CEA, carotid endarterectomy. (Adapteded from Kristensen SD, et al. 20144 and Smilowitz NR, et al. 2017.47)
All vascular surgeries, except for stent placement and minor surgery, are considered as high-risk interventions.4 In pulmonary surgery, lobectomy is considered high-risk due to its risk of myocardial injury.50 Lung and liver transplants are highly invasive and are considered as high-risk procedures.51 In gastrointestinal surgery, interventions involving the esophagus, pancreas, and liver are considered high-risk, and gastrectomy and colectomy are considered as intermediate-risk surgeries.49 Generally, neurosurgeries involve higher risk than gastrointestinal surgeries,47,48 and urology, orthopedics, and endocrine surgeries are considered as relatively lower risk.4 Gynecological surgeries have a low-risk of cardiovascular and cerebrovascular event (0.3%).47 Other interventions under local anesthesia and surgeries of superficial organs, even if performed under general anesthesia, are considered relatively safe. The incidence of AMI in day surgery is 0.03%, which is not different from the adjusted incidence during the non-perioperative period.52
2.3 Surgical Approach and Risk
Laparoscopic surgery requiring only small incisions, that has recently become widespread as a minimally invasive approach, enables early recovery and discharge.53 A systematic review concluded that the incidence of perioperative cardiac complications was lower in laparoscopic surgery than in open surgery in older adults (≥70 years).54 A meta-analysis concluded that the complication rate of video-assisted thoracic surgery (VATS) was lower than that of thoracotomy, but the mortality rate was comparable.55 However, reduced diastolic function, increased peripheral vascular resistance, and decreased cardiac output during the pneumoperitoneum used for laparoscopic surgery have been observed in severely obese patients.56 Laparoscopic intrapelvic surgeries (rectum, uterus, bladder) that place the patient in the Trendelenburg position increase the mean blood pressure, CVP, and peripheral vascular resistance; thus, the risk of cardiac complications is potentially no lower than that for open surgery.57,58 In this guideline, no consensus has been reached as to whether endoscopic surgery has a lower rate of cardiac complications than laparotomy or thoracotomy.
2.4 Urgency of Surgery and Risk
2.4.1 Classification by Urgency of Surgery
According to the 2014 AHA/ACC clinical practice guideline, an emergency procedure is one in which life or limb is threatened and needs to be performed as soon as possible.59 An urgent procedure is one which should be performed within 24 h. A time-sensitive procedure should be performed within 6 weeks. An elective procedure could be delayed for up to 1 year.59 The urgency of surgery is important. The preoperative assessment of cardiovascular risk must be made quickly for very urgent surgeries. If time allows for urgent surgery, an assessment for unstable cardiac condition should be performed* (Chapter II.11 Preoperative Testing Algorithms). Close perioperative monitoring is also important because cardiac complications affect patients undergoing emergency procedures at a 2–5-fold higher rate than those undergoing elective surgeries.60 The retrospective data of the ACS NSQIP has shown that the risks of death and complications increase with the level of urgency.61 However, whether the risk is inherent to the urgency itself or is related to the general or preoperative condition of the patient needing an emergency procedure remains unknown.
3. Self-Reported Functional Capacity
3.1 Utility of Assessment of Self-Reported Functional Capacity of the Patient62–66
The ACC/AHA clinical practice guideline 2014 and ESC clinical practice guidelines 2014 recommended assessment of self-reported functional capacity to investigate perioperative cardiac risk.4,59 However, the Canadian Cardiovascular Society Guidelines 201710 decided not to make a recommendation on self-reported functional capacity to estimate perioperative cardiac risk.
The studies below suggest that functional capacity ≥4 METS assessed by the physician based on patient self-report is associated with lower frequency of myocardial ischemic events, but the certainty of the evidence is limited.
In a study of 600 patients undergoing non-cardiac surgery,62 the relationship between self-reported functional capacity and postoperative complications was evaluated. Patients with poor functional capacity (i.e., unable to walk 4 blocks [≈400 m] on flat land or unable to climb two flights of stairs; both corresponding to ≤4 METS) had a significantly higher incidence of total serious complications, myocardial ischemia, and neurological events (mainly delirium).62
In a retrospective study of a large-scale database consisting of 5,939 patients, the functional capacity evaluated subjectively by an anesthesiologist based on patient self-reporting had great inter-rater variability and could not predict postoperative cardiac complications or death.63
A study investigating the relationship between ability to climb two flights of stairs (4 METS) and postoperative mortality rate in non-cardiac surgery found an increased risk of death in patients undergoing thoracic surgery, but the relationship was not observed for non-thoracic surgery.64
3.2 Duke Activity Status Index
The Duke Activity Status Index (DASI) is a 12-item questionnaire (Table 11) developed to objectively evaluate functional capacity and has been shown to correlate significantly with maximum oxygen consumption.67
Table 11. Duke Activity Status Index
| Activity |
Weight |
| Can YOU... |
|
| 1. Take care of yourself, that is, eating, dressing, bathing, or using the toilet? |
2.75 |
| 2. Walk indoors such as around your house? |
1.75 |
| 3. Walk a block or 2 on level ground? |
2.75 |
| 4. Climb a flight of stairs or walk up a hill? |
5.50 |
| 5. Run a short distance? |
8.00 |
| 6. Do light work around the house like dusting or washing dishes? |
2.70 |
| 7. Do moderate work around the house like vacuuming, sweeping floors, or carrying in groceries? |
3.50 |
| 8. Do heavy work around the house like scrubbing floors, or lifting or moving heavy furniture? |
8.00 |
| 9. Do yardwork like raking leaves, weeding or pushing a power mower? |
4.50 |
| 10. Have sexual relations? |
5.25 |
11. Participate in moderate recreational activities like golf, bowling, dancing, double tennis, or
throwing a baseball or football? |
6.00 |
| 12. Participate in strenuous sports like swimming, singles tennis, football, basketball, skiing? |
7.50 |
(Adapted from Hlatky MA, et al. 1989.67)
The METS trial published in 2018 showed that patient-reported functional capacity subjectively evaluated by an anesthesiologist and functional capacity by cardiopulmonary testing (CPX) did not predict either 30-day postoperative death after major non-cardiac surgery or non-fatal MI, whereas the DASI assessment alone predicted the risk.65 In the subanalysis, a DASI score of 34 points (corresponding to 7 METS) was reported as the optimal cutoff value for predicting perioperative cardiac complications.66
Recommendations for self-reported functional capacity are shown in Table 12.
Table 12. Recommendations and Levels of Evidence on Managing Self-Reported Functional Capacity
| |
COR |
LOE |
DASI score should be considered as an assessment of self-reported functional
capacity |
IIa |
C |
Further preoperative cardiovascular examinations are not needed if self-reported
functional capacity ≥7 METS |
IIa |
C |
Further preoperative cardiovascular evaluation may not be needed if self-reported functional
capacity ≥4 METS |
IIb |
C |
COR, class of recommendation; LOE, level of evidence.
4.1 Risk Stratification by Abnormal Preoperative 12-Lead ECG Findings
The association between preoperative 12-lead ECG abnormalities and in-hospital death of patients undergoing non-cardiac surgeries was retrospectively investigated in a total of 28,457 cases.68 The incidence of in-hospital cardiovascular death was 0.7%, and the risk increased in the presence of ECG abnormalities (1.8% vs. 0.3%) (adjusted odds ratio: 4.5, 95% confidence interval [CI] 3.3–6.0). The incidence of in-hospital cardiovascular death was higher among patients with ECG abnormalities than those without it (0.7% vs 0.2%) in low- to low intermediate-risk surgeries, but the difference was only 0.5%. Whereas the gap was much higher at 3.9% (1.1% vs. 5%) in surgeries of intermediate high- to high-risk, suggesting the potential utility of the 12-lead ECG in this group of patients.68
4.2 Effects on Management by Performing Preoperative 12-Lead ECG
Munro et al. reported that abnormal findings on ECG were observed in 4.6–31.7%, and patient management was altered accordingly in 0–2.2% of the cases.69 Bhuripanyo et al. reported that preoperative ECG abnormalities were observed in 31.7% of patients aged 40 years or older with no significant medical history and no abnormal physical findings who underwent ECG as a routine test and that it had clinical significance in 6.4% and an effect on management in 2.2%.70
4.3 Effects on Cardiovascular Event Risk by Performing Preoperative 12-Lead ECG
A study of 1,061 patients undergoing day surgeries (orthopedic surgery, plastic surgery, general surgery, urological surgery, ophthalmologic surgery [except cataract surgery], and spine surgery) were randomly assigned to a group that underwent basic blood tests, 12-lead ECG, and chest X-ray or a group that did not.71 The study found no difference in complication rates (including AMI, heart failure, and sudden death) between groups. ECG abnormalities were observed in 7.2%, but were not associated with postoperative complications.71 Another study that also randomly assigned 19,000 individuals who underwent cataract surgery to those who did and did not undergo preoperative testing (12-lead ECG, blood test) found no difference in the number of events (cardiovascular events) in the perioperative period (31.3 events/1,000 surgeries).72
Although there is little clinical significance of routine preoperative 12-lead ECG in low-risk surgeries such as those mentioned above, ECG should be considered for patients whose medical history or physical findings suggest cardiovascular disease.
The recommendation of preoperative 12-lead ECG is shown in Table 13.
Table 13. Recommendation and Level of Evidence on Preoperative 12-Lead ECG
| |
COR |
LOE |
| Routine preoperative 12-lead ECG for low-risk surgery is not recommended |
III (No
benefit) |
B |
COR, class of recommendation; LOE, level of evidence.
5.1 Preoperative TTE and Perioperative Outcomes
Approximately 15% of patients undergoing elective intermediate- and high-risk non-cardiac surgeries in the Canadian health insurance database (n=≈260,000) had preoperative echocardiography.73 Comparing the propensity score-matched groups (presence of preoperative TTE vs. absence of it), preoperative TTE was not associated with a decrease in mortality after non-cardiac surgery.73
5.2 Preoperative Abnormal TTE Findings and Risk Stratification of Perioperative Cardiovascular Events
TTE findings did not add value to the clinical model for predicting postoperative cardiovascular event in patients with coronary artery disease or those at high-risk of developing coronary artery disease.74 On the other hand, in patients undergoing intermediate- and high-risk surgeries, abnormal findings on preoperative TTE* were associated with perioperative cardiovascular complications (MI and congestive heart failure).75
To date, the data on the relationship between preoperative TTE findings and risk stratification have been inconsistent.
*Left ventricular systolic dysfunction (left ventricular ejection fraction <55%), left ventricular hypertrophy (septal or posterior wall thickness ≥13 mm, moderate or severe mitral regurgitation, and peak pressure gradient ≥40 mmHg at the aortic valve).
5.3 RCRI and Preoperative TTE
Among RCRI Class III (RCRI=2 items) and IV (RCRI ≥3 items) patients without abnormal TTE findings, as noted above*, had significantly lower rates of perioperative cardiovascular complications (4% vs. 16%).75 However, no significant association was observed between abnormal TTE findings and events in Class I (RCRI=0 item) or II (RCRI=1 item) patients.75 Thus, preoperative TTE may be useful to reclassify RCRI-estimated elevated risk group into low-risk group based on normal TTE findings, but may not be useful to reclassify RCRI-estimated low-risk group into an elevated group based on abnormal TTE findings.
5.4 Indications for Preoperative TTE and Cautions
Recommendations regarding preoperative TTE evaluations in other international clinical practice guidelines are as follows.
AHA 2014: Routine TTE is not recommended for assessment of left ventricular function9
ESC 2014: Routine preoperative echocardiography is not recommended for low- to intermediate-risk surgeries4
Canada 2017: Echocardiography is not recommended for perioperative risk stratification10
We also do not recommend routine preoperative TTE (Table 14). As mentioned in Chapter I, TTE can be performed for patients with a clinical indication, regardless of whether surgery is performed. The examples are patients with dyspnea of unknown etiology or exacerbation of symptoms of known heart failure.9 Conversely, performing echocardiography can delay necessary surgery significatly,76 which should be avoided.
Table 14. Recommendation and Level of Evidence on Preoperative Transthoracic Echocardiography
| |
COR |
LOE |
| Routine preoperative transthoracic echocardiography is not recommended |
III (No
benefit) |
C |
Practical consideration: TTE can be performed if the patient has a clinical indication for the test, but caution should be exercised so as to not delay any necessary surgery for the sole purpose of echocardiography. COR, class of recommendation; LOE, level of evidence.
Various scoring systems for predicting perioperative cardiovascular complications for non-cardiac surgery have been developed. Their goals are to (1) calculate a score from basic patient characteristics to predict perioperative cardiovascular events and (2) avoid unnecessary preoperative cardiac testing (e.g., echocardiography, myocardial stress test, etc.) by accurately screening for low-risk patients.
6.1 Revised Cardiac Risk Index
The RCRI77 (Table 15) is the most widely used scoring system. In the original study by Lee et al., having 1 items indicated 0.9% of perioperative major cardiac event risk. Having ≥3 items indicated 11.0% of the risk.77 The cardiovascular events concerned were AMI (diagnosed by creatinine phosphokinase [CPK] and creatinine phosphokinase-MB [CPK-MB], not contemporary troponin-based diagnosis), heart failure, ventricular fibrillation/cardiac arrest, and complete atrioventricular block.
Table 15. Revised Cardiac Risk Index (RCRI)
| History of ischemic heart disease* |
| History of congestive heart failure |
| History of cerebrovascular accident or TIA |
| Insulin-dependent diabetes mellitus |
| Serum creatinine >2 mg/dL |
| High-risk surgery** |
*Defined as a history of myocardial infarction, positive exercise test, current complaint of ischemic chest pain or nitrate use, or ECG with pathological Q waves; patients with previous coronary bypass surgery or angioplasty meet the criteria if they have such findings after their procedure. **Defined as intrathoracic, intra-abdominal, or suprainguinal vascular surgery. TIA, transient ischemic attack. (Adapted from Lee TH, et al. 1999.77)
Numerous follow-up studies of the RCRI have been conducted, and a meta-analysis has shown its high validity to predict major perioperative cardiac events during the hospital stay or by 30 days after non-cardiac surgery, specifically by an area under the curve (AUC) 0.75 (95% CI 0.72–0.79), and sensitivity 65%, specificity 76%, positive likelihood ratio 2.78, negative likelihood ratio 0.45, where ≥2 items indicates positivity.78 At the same time, the validity was slightly lower for vascular surgery, with an AUC 0.64 (95% CI 0.61–0.68).78
Studies on the RCRI where MI was diagnosed based on high-sensitivity troponin, rather than CK, were comprehensively searched and evaluated.79–90 The RCRI and perioperative cardiovascular event rates for non-vascular surgeries and vascular surgeries are shown in Table 16.
Table 16. RCRI Score and Perioperative Risk of Cardiovascular Event After Non-Cardiac Surgery
| (a) Non-vascular surgery |
| RCRI |
Risk estimate (95% CI) |
| 0 |
0.91% (0.70–1.2%) |
| 1 |
2.9% (2.5–3.4%) |
| 2 |
7.2% (6.0–8.6%) |
| ≥3 |
13.7% (10.7–17.4%) |
| (b) Vascular surgery |
| RCRI |
Risk estimate (95% CI) |
| 0 |
3.2% (2.7–3.7%) |
| 1 |
7.7% (6.9–8.5%) |
| 2 |
11.9% (10.6–13.4%) |
| ≥3 |
19.0% (16.6–21.6%) |
RCRI, Revised Cardiac Risk Index. (Adapted from references 79–90.)
6.2 American College of Surgeons National Surgical Quality Improvement Program Myocardial Infarction and Cardiac Arrest Risk-Prediction Rule (NSQIP MICA)
The ACS NSQIP MICA91 is a scoring system created based on the ACS NSQIP database. A specialized calculator using a statistical equation is necessary for the evaluation (http://www.surgicalriskcalculator.com/miorcardiacarrest). Although the validity of the score has been tested in 2 studies, both of which were single-center, retrospective studies,92,93 it is also important to note that the validity of the scoring system has not been tested in Japan.
6.3 ACS NSQIP Surgical Risk Calculator
The ACS NSQIP Surgical Risk Calculator94 is another scoring system that was developed based on the ACS NSQIP database and requires the use of a specialized calculator (www.riskcalculator.facs.org). Although the score has been tested extensively,94–109 most studies were retrospective, single-center studies. It is also important to note that the validity of the scoring system has not been assessed in Japan.
6.4 Recommendations on the Use of Risk Scores
The 2017 Canadian Cardiovascular Society clinical practice guidelines10 recommended using the RCRI for the assessment of perioperative risk after conducting a systematic review. Follow-up studies of the ACS NSQIP MICA risk-prediction rule and ACS NSQIP Surgical Risk Calculator are limited compared to the follow-up studies evaluating the RCRI, and the RCRI is easier to use in routine clinical care because it does not require a specialized calculator. Thus, this clinical practice guideline also recommends the use of the RCRI as a valid tool for perioperative risk assessment.
6.5 Cutoff Point for the Definition of Low-Risk Patients
The AHA 2014 clinical practice guideline59 defined an estimated risk of <1% as low-risk. However, an analysis of recent data has shown that the incidence of events in non-vascular surgeries was 0.91% (95% CI 0.70–1.2%) for 0 items in the RCRI and was >1% if the 95% CI was taken into account (Table 16). For vascular surgeries, the incidence increased to 3.2% (95% CI 2.7–3.7%) for 0 items in the RCRI (Table 16). The diagnosis of AMI in the original study proposing the RCRI77 used creatinine kinase (CK) as a biomarker. In contrast, the detection of the incidence of postoperative AMI has increased in recent clinical studies,110 and this has been attributed to the use of the more sensitive myocardial troponin for diagnosing AMI.10 Therefore, the definition of low-risk should also be revised according to these changes. The 2017 Canadian Cardiovascular Society clinical practice guidelines estimated the risk of events at 3.9% (95% CI 2.8–5.4%) for 0 items in the RCRI and 6.0% (95% CI 4.9–7.4%) for 1 item in the RCRI.10 In this guideline, RCRI ≥1 was considered to be elevated risk.
Considering the above, a predictive risk <5% instead of 1% is defined as low risk and ≥5% is defined as elevated risk in this guidelines as contemporary clinical circumstances use high-sensitivity troponin for the diagnosis of AMI in our country. Based on this cutoff value, 0–1 item on the RCRI would be considered a low-risk non-vascular surgery, whereas 0 items would be considered a low-risk vascular surgery in this clinical practice guideline.
6.6 Advantages and Disadvantages of Performing Additional Cardiovascular Evaluation in Low-Risk Patients
The validity of additional preoperative cardiovascular evaluation examinations for patients classified as low-risk by the RCRI has been investigated. Among patients classified as low-risk by a clinical prediction model, 1.1% of them were accurately reclassified as high-risk by a high level of BNP, and 8% were incorrectly classified as high-risk, demonstrating the ineffectiveness of risk stratification for patients classified as low-risk by the clinical prediction model.111 Similar findings have been reported with coronary computer tomography angiography and echocardiography75,112 (see Chapter II.5 Transthoracic Echocardiography (TTE), Chapter II.9 Coronary Computed Tomography Angiography). Therefore, the routine addition of these cardiac examinations is not recommended for patients classified as low-risk by the RCRI (Class III, no benefit, Level C) (Table 17).
Table 17. Recommendations and Levels of Evidence on Perioperative Risk Scores
| |
COR |
LOE |
| Use of the RCRI is recommended for preoperative risk assessment |
I |
B |
Additional routine preoperative cardiac examinations are not recommended for low-risk
patients (i.e., non-vascular surgery patients with RCRI 0–1 items and vascular
surgery patients with RCRI 0 items) |
III (No
benefit) |
B |
COR, class of recommendation; LOE, level of evidence; RCRI, Revised Cardiac Risk Index.
7.1 Clinical Risk Stratification and Restratification by BNP and NT-pro BNP
Earlier studies have reported that adding BNP or NT-pro BNP to preoperative evaluation with clinical prediction rule may improve perioperative risk stratification in some situations as shown below.
• Adding preoperative BNP/NT-pro BNP to perioperative event prediction with age, RCRI ≥3 items, type of surgery (vascular surgery or other), and surgery urgency (emergency/semi-emergency or elective) for perioperative event prediction was shown to improve the accuracy of event prediction111
• In a meta-analysis, 32% of patients classified as elevated risk in a risk assessment model that included RCRI were correctly reclassified as low risk by adding low level of BNP to the assessment. However, it should be noted that 17.6% were also inappropriately reclassified as higher risk by adding BNP even though no event occurred and that among those who were originally appropriately classified as high risk, 23% had been inappropriately reclassified into the low-risk group by adding BNP even though actually they had an event.111 Furthermore, among patients classified as low risk, a high BNP reclassified them appropriately as high risk in 1.1%, while it inappropriately reclassified patients as high risk in 8%, suggesting the insignificance of BNP in risk stratification
• In the Vascular Events in Non-cardiac Surgery Patients Cohort Evaluation (VISION) subanalysis, for prediction of composite endpoints including postoperative 30-day death and myocardial injury after non-cardiac surgery (MINS), the addition of NT-pro BNP to RCRI was reported to increase prediction precision113
It remains an option to measure BNP and NT-pro BNP preoperatively for risk stratification to determine whether to reclassify to low risk when the RCRI suggests elevated risk (≥2 items in non-vascular surgeries and ≥1 item in vascular surgeries) (Table 18).
Table 18. Recommendation and Level of Evidence on Preoperative BNP and NT-pro BNP
| |
COR |
LOE |
In patients with elevated risk by clinical prediction model who undergo intermediate-
or high-risk surgery (e.g., RCRI ≥1 item for vascular surgery or ≥2 items for non-vascular
surgery), the addition of BNP or NT-pro BNP evaluation may be considered for further risk
stratification |
IIb |
C |
Practical considerations: Angiotensin receptor-neprilysin inhibitor can affect BNP levels and, therefore, should be interpreted with caution when patients are taking that medicine. There is little significance in adding BNP measurement for low-risk patients. BNP, B-type natriuretic peptide; COR, class of recommendation; LOE, level of evidence; NT-proBNP, N-terminal fragment of BNP precursor; RCRI, Revised Cardiac Risk Index.
8.1 Aims of the Preoperative Cardiac Stress Test
Various stress tests have been considered for risk stratification of perioperative cardiovascular events when performing non-cardiac surgery and testing. However, numerous studies, including a meta-analysis, have shown no difference in mortality risk within 30 days of surgery, whether or not preoperative stress testing was performed.114
8.2 Accuracy of Testing in Predicting Perioperative Risks (Table 19)
Table 19. Summary of Various Cardiac Stress Tests and Prediction of Perioperative Cardiovascular Events
| |
Sensitivity |
Specificity |
Positive predictive
value |
Negative predictive
value |
| Functional capacity by exercise stress testing |
64–89% |
34–53% |
10–24% |
88–97% |
| Exercise stress ECG |
37–67% |
49–83% |
10–25% |
83–94% |
| Stress myocardial perfusion imaging |
87–100% |
49–71% |
13–50% |
98–100% |
| Dobutamine stress echocardiography |
30–100% |
44–93% |
9–43% |
95–100% |
8.2.1 Evaluation of Exercise Tolerance by Exercise Stress Testing
High exercise tolerance assessed by an exercise stress test is associated with lower risk of cardiovascular events, although low functional capacity does not necessarily indicate high risk. Moreover, studies to date have not evaluated cardiovascular events by distinguishing ischemic and nonischemic events (e.g., heart failure).
Low exercise tolerance evaluated by an exercise stress test (with or without ischemic ECG changes) is associated with adverse cardiovascular events perioperatively; validity has achieved 64–89% sensitivity, 34–53% specificity, 10–24% positive predictive rate, and negative predictive rate of 88–97%.115–118 Further, when the “positive” is defined by exercised tolerance ≤4 METS with or without ECG change by exercise stress test, test validity to predict perioperative cardiovascular events was evaluated in 2 studies: sensitivity was 66%, 82%, specificity 49%, 53%, positive predictive value 21%, 12%, and negative predictive value 88%, 97%, respectively.116,117
McPhail et al.118 performed cardiopulmonary exercise testing (CPX) for 100 patients undergoing major vascular surgery to predict postoperative cardiac complications and found a lower rate of cardiac complications in patients who were capable of achieving 85% of the target heart rate than patients who could not (6% vs. 24%, P=0.04). In this study, the cardiac event rate was not significantly different between the patients with and without exercise-induced ST depression. Another study showed exercise stress testing was ineffective for predicting perioperative death and MI.116
Two clinical studies evaluating patients undergoing surgery for lung cancer found a higher rate of postoperative cardiopulmonary complications in patients who could not perform an exercise stress test due to musculoskeletal disease, neurologic disease, peripheral vascular disease, or psychiatric disease, etc. than patients who were able to complete the test;119 however, the maximum oxygen consumption assessed by the CPX failed to predict the onset of postoperative cardiopulmonary complications.120
This clinical practice guideline concludes that exercise stress testing may be considered in patients whose functional capacity is not known by self-report.
8.2.2 Exercise Stress ECG Test
The validity of the exercise stress ECG test to predict perioperative cardiac death or MI in patients undergoing vascular, thoracic, or abdominal surgeries was reported as: sensitivity 37–67%, specificity 49–83%, positive predictive value 10–25%, and negative predictive value 83–94%115,116,118,121 (Table 19). This suggests that a negative result means low risk, but a positive result does not always mean high risk. Adding ECG to the exercise test does not seem to be more valid for predicting perioperative cardiovascular risk.
8.2.3 Stress Myocardial Perfusion Imaging by Scintigraphy
The validity of stress myocardial perfusion imaging to predict perioperative cardiac death and MI was reported as: sensitivity 87–100%, specificity 49–71%, positive predictive value 13–50%, and negative predictive value 98–100%122–127 (Table 19). The results of a meta-analysis showed higher occurrence of perioperative cardiovascular events in patients with moderate or high ischemia detected on stress myocardial imaging performed before non-cardiac surgery.128
8.2.4 Dobutamine Stress Echocardiography
The validity of dobutamine stress echocardiography to predict perioperative cardiac death and MI was reported as: sensitivity 30–100%, specificity 44–93%, positive predictive value 9–43%, and negative predictive value 95–100%129–134 (Table 19). The results of a meta-analysis also showed that stress echocardiography has a low probability of false negatives and can more accurately predict perioperative cardiovascular events than stress myocardial perfusion imaging as a preoperative test for non-cardiac surgery.128 Moreover, it has been reported to be useful for predicting long-term prognosis.135
8.2.5 Recommendation of Preoperative Cardiac Stress Test
There has not been a study that clearly demonstrates that the stress test (exercise stress ECG test, myocardial perfusion imaging by scintigraphy, and dobutamine stress echocardiography) can improve the accuracy of preoperative risk assessment using risk scores, including the RCRI. Therefore, their routine use is not recommended (Table 20).
Table 20. Recommendations and Levels of Evidence on Preoperative Stress Electrocardiography, Stress Myocardial Perfusion Imaging
| |
COR |
LOE |
Preoperative stress electrocardiography for risk assessment should not be performed
routinely for non-cardiac surgery* |
III (No
benefit) |
B |
Preoperative stress myocardial perfusion imaging and stress echocardiography should
not be performed routinely for non-cardiac surgery** |
III (No
benefit) |
B |
Practical considerations: *If functional capacity of the patient is unknown from self-report, an exercise stress test can be used to evaluate functional capacity. **These tests can be used for risk stratification for perioperative cardiovascular events, but there is no evidence supporting that they will improve the precision of the preoperative risk assessment by the RCRI. COR, class of recommendation; LOE, level of evidence; RCRI, Revised Cardiac Risk Index.
There are a limited number of studies examining the usefulness of coronary computed tomography angiography as a preoperative cardiovascular assessment. A recent meta-analysis showed that the incidence of perioperative cardiovascular events could be stratified depending on the presence or absence and the degree of coronary artery stenosis on coronary computer tomography angiography.112 It has also been reported that even in cases with RCRI score of ≥3, if there are no multivessel lesions on coronary computer tomography angiography, the onset of perioperative cardiovascular events is rare (negative predictive value 96%).112 This high negative predictive value of coronary computer tomography angiography is noteworthy.
However, a large-scale observational study of patients scheduled for non-cardiac surgery showed that when coronary computer tomography angiography findings were considered in addition to the risk score (RCRI score), 17 of 1,000 patients whose risk was assessed as “low” by the RCRI were appropriately reclassified into the high-risk group, but 98 patients were inappropriately reclassified into the high-risk group based on the coronary computer tomography angiography results.81 These study results indicate that in patients considered as low-risk preoperatively, additional evaluation by coronary computer tomography angiography might be inappropriate and may have certain disadvantages, such as complications associated with the intervention and delay of non-cardiac surgery. Therefore, routine use of preoperative coronary computer tomography angiography is not recommended (Table 21).
Table 21. Recommendation and Level of Evidence on Preoperative Coronary Computer Tomography Angiography
| |
COR |
LOE |
Routine preoperative coronary computer tomography angiography is not recommended
for non-cardiac surgery |
III (No
benefit) |
C |
Practical considerations: Performing coronary computer tomography angiography for patients classified as low-risk by the RCRI may lead to the inappropriate reclassification into high-risk of a significant number of patients. The risk of perioperative cardiovascular events is very low in the absence of significant findings by coronary computer tomography angiography; thus, it could be useful for further stratification of perioperative risk in patients with elevated risk as assessed by the RCRI. COR, class of recommendation; LOE, level of evidence; RCRI, Revised Cardiac Risk Index.
The incidence of serious complications is said to be ≈1%, and it is necessary to acknowledge that coronary angiography is an invasive test. Moreover, when coronary stenosis is detected by angiography before non-cardiac surgery, coronary revascularization for such stenosis has not been shown to be useful (see CQ1).136,137 Therefore, routine use of coronary angiography is not recommended (Table 22).
Table 22. Recommendation and Level of Evidence on Preoperative Coronary Angiography
| |
COR |
LOE |
Coronary angiography should not be performed routinely for the purpose of preoperative
assessment for non-cardiac surgery |
III (No
benefit) |
C |
COR, class of recommendation; LOE, level of evidence.
Figure 3 shows summary for preoperative cardiac assessments for non-cardiac surgery.
Step 1: Urgency of Non-Cardiac Surgery
Emergency surgery (i.e., for life-threatening conditions) should be performed as soon as possible. A detailed examination for cardiovascular risk usually cannot be performed for such patients. Physicians need to identify unstable cardiac conditions by medical history, physical findings, and basic tests such as ECG within the time allowed to avoid delaying the non-cardiac surgery. For non-emergency procedures, go to Step 2.
Step 2: Checking for Unstable Cardiac Condition
Unstable cardiac conditions include acute coronary syndrome, critical arrhythmia, acute heart failure, and symptomatic valvular heart disease. Further examinations should be added as needed if any of these conditions are suspected through medical history and physical findings. Physicians should stabilize the patient first if any of these conditions are diagnosed. Possible interventions for the cardiac disease include pharmacotherapy, invasive intervention (coronary artery revascularization [PCI, CABG], and valve replacement, etc). Whether or not to perform invasive procedures for the cardiac condition prior to the non-cardiac surgery needs comprehensive assessment with a multidisciplinary team. Addressing the following factors is necessary: (1) urgency of invasive procedure for cardiac condition (disease severity, symptoms severity), (2) urgency of the non-cardiac surgery (i.e., how long can it be postponed?), (3) risk of invasive procedure for cardiac conditions (e.g. STS score , EuroSCORE II, JapanSCORE), (4) risk of the non-cardiac surgery and (5) necessity of antithrombotic therapy after the invasive treatment for the cardiac condition; that is, the bleeding risk or thrombotic risk during the perioperative period of the non-cardiac surgeries with continuation or discontinuation of antithrombotic agents after the cardiovascular invasive intervention. In the absence of unstable cardiac condition, go to Step 3.
Step 3: Risks of the Non-Cardiac Surgical Procedure Itself
Perform the surgery if it is low risk (see Table 10 in Chapter II.2). For intermediate- and high-risk surgeries, go to Step 4.
Step 4: Risk Prediction for Perioperative Cardiovascular Events
Estimate the risk using the RCRI. For low-risk patients (non-vascular surgery: RCRI=0–1 items, vascular surgery: RCRI=0 items), perform surgery. For elevated risk patients (non-vascular surgery RCRI ≥2 items, vascular surgery RCRI ≥1 items), go to Step 5.
Definition of low risk: cardiovascular event risk <5% (if using troponin to diagnose AMI).
Definition of elevated risk: cardiovascular event risk ≥5%.
Step 5: Evaluation of Functional Capacity or BNP and NT-proBNP Levels
For patients at elevated risk, their functional capacity should be assessed objectively using the DASI score or another method. It is reasonable to consider surgery without any additional cardiac examinations for patients with ≥7 METS (DASI score ≥34 points, Class IIa). Performing the surgery without any additional cardiac examinations may also be considered for patients with ≥4 METS (DASI score ≥10 points, Class IIb). For patients with a METS score <4 or an unknown METS, the indications for preoperative cardiac examinations and overall management should be discussed comprehensively. As an alternative measure of functional capacity, the BNP or NT-proBNP values may be used (Class IIb), where BNP <92 pg/mL and NT-pro BNP <300 pg/mL can classify a patient as low-risk with proceeding to surgery.
It is important to note that the validity of this algorithm has not been tested.
III. Perioperative Pharmacotherapy
1. Pharmacotherapy (Except Anticoagulant Therapy)
1.1 Beta-Blockers (Table 23)
Table 23. Recommendations and Levels of Evidence on Perioperative Use of Beta-Blockers
| |
COR |
LOE |
Starting administration of β-blockers within 24 h prior to non-cardiac surgery is not
recommended |
III
(Harm) |
A |
Continuing long-term β-blocker therapy during the perioperative period of non-cardiac
surgery is recommended |
I |
B |
COR, class of recommendation; LOE, level of evidence.
Myocardial ischemia can be induced due to the stress of surgery and hypotension. Beta-blockers exert cardioprotective effects and are known to suppress arrhythmic events. However, the results of large-scale clinical trials, including randomized controlled trials (RCTs), on perioperative use of β-blockers for non-cardiac surgery have not yielded consistent results.
1.1.1 Preoperative Initiation
In the POISE (PeriOperative ISchemic Evaluation) trial, oral metoprolol administration 2–4 h before the induction of anesthesia reduced the incidence of myocardial infarction (MI) while significantly increasing mortality and cerebral infarction in the first 30 days after surgery compared with placebo.138 In a post-hoc analysis, hypotension was independently associated with increased mortality and cerebral infarction.
Two meta-analyses reported that preoperative initiation of β-blockers decreased the rate of MI, but increased mortality, stroke, hypotension, and bradycardia.139,140 A meta-analysis of 14,967 patients further indicated that preoperative administration of β-blockers lowered MI and atrial fibrillation/flutter but increased mortality, hypotension, and bradycardia.141
However, the POISE trial of 8,351 patients138 contributed the majority of the patients in those meta-analyses,139–141 warranting caution in interpreting the results. The POISE trial has been criticized for starting metoprolol at a high dose (100 mg), increased up to 400 mg in 24 h, and administered immediately before surgery without dose adjustment. If considering starting β-blockers preoperatively for non-cardiac surgery, it is important to introduce and adjust the dose cautiously to avoid hypotension and bradycardia.
Indeed, meta-analyses of studies in which metoprolol therapy was started within 24 h before the surgery have also shown increased risk of death.140,142 Given these facts, it is not recommended to start β-blockers within 24 h prior to surgery (Table 23).
1.1.2 Discontinuation or Continuation of Long-Term β-Blocker Therapy
Several observational studies have reported on whether patients under long-term oral β-blocker therapy should continue β-blockers in the perioperative period of non-cardiac surgery.143–148 A large-scale observational study of 12,105 patients undergoing non-cardiac surgery144 that used propensity score analysis identified an association between the perioperative discontinuation of β-blockers and increased 30-day (odds ratio [OR] 3.93, 95% confidence interval [CI] 2.57–6.01) and 1-year mortality (OR 1.96, 95% CI 1.49–2.58). Thus, long-term β-blocker therapy should be continued during the perioperative period for non-cardiac surgery (Table 23).
1.2 Statins (Table 24)
Table 24. Recommendations and Levels of Evidence on the Perioperative Use of Statins
| |
COR |
LOE |
| Consider starting statins at least 2 weeks (preferably 1 month) prior to vascular surgery |
IIa |
B |
Continuing long-term oral statin therapy in the perioperative period for non-cardiac surgery is
recommended |
I |
C |
COR, class of recommendation; LOE, level of evidence.
The efficacy of statins in primary and secondary prevention of ischemic heart disease has been demonstrated in many clinical trials.149
1.2.1 Preoperative Initiation
In a subanalysis of the VISION (Vascular Events in Noncardiac Surgery Patients Cohort Evaluation) trial, statins were associated with a significant lower risk of all-cause death, and cardiovascular death, although there were no statistically significant differences in the risk of MI or stroke.150
An RCT limited to vascular surgery showed a significant decrease in cardiovascular events among patients administered statins ≈1 month before surgery when compared with those administered a placebo.151,152 In a meta-analysis of 23,536 vascular surgery patients, the preoperative administration of statins significantly reduced the incidence of cardiovascular events.153
These results suggest that statins should be administered at least 2 weeks (preferably 1 month) prior to surgery for patients undergoing undergo vascular surgery4 (Table 24).
1.2.2 Discontinuation or Continuation of Long-Term Statin Therapy
Observational studies have reported that the perioperative discontinuation of statins is independently associated with increased subsequent cardiovascular events in patients receiving statins before the surgery.154,155 Therefore, statin therapy should be continued in the perioperative period for non-cardiac surgery if the patient is receiving long-term statins.
1.3 Angiotensin-Converting Enzyme Inhibitors and Angiotensin II Receptor Blockers (Table 25)
Table 25. Recommendations and Levels of Evidence on the Perioperative Use of Angiotensin-Converting Enzyme Inhibitors and Angiotensin Receptor Blockers
| |
COR |
LOE |
| Withholding ACE inhibitors/ARBs may be considered on the day of surgery |
IIb |
C |
ACE inhibitors/ARBs should be restarted as soon as possible after the surgery if these are
used for HFrEF and discontinued before the surgery |
I |
C |
COR, class of recommendation; HFrEF, heart failure with reduced ejection fraction; LOE, level of evidence.
Angiotensin-converting enzyme (ACE) inhibitors are mainly used for patients with hypertension or heart failure with reduced ejection fraction (HFrEF). Patients receiving oral ACE inhibitor therapy have an increased risk of hypotensive events requiring vasopressors shortly after induction of anesthesia.156 In a large retrospective observational study, however, patients who withheld ACE inhibitors only in the morning of the surgery had similar risks of hypotension, neurological events such as stroke, and cardiovascular events as those not taking any ACE inhibitors at all.157 Therefore, avoiding taking the ACE inhibitor on the morning of the surgery may be considered. However, concerning non-perioperative settings, discontinuing ACE inhibitors for HFrEF has been associated with worsening clinical prognosis;158 thus, ACE inhibitors should be re-administered as soon as possible after surgery if it is discontinued before surgery. Although there is little evidence of angiotensin II receptor blockers (ARBs) in this clinical setting, it is reasonable to deal with ARBs in the same way as ACE inhibitors in the perioperative settings, given their similar pharmacological properties.
1.4 α2 Agonists (Table 26)
Table 26. Recommendation and Level of Evidence on Perioperative Use of α
2 Agonists
| |
COR |
LOE |
| Starting α2 agonist therapy prior to non-cardiac surgery is not recommended |
III (No
benefit) |
B |
COR, class of recommendation; LOE, level of evidence.
In the POISE-2 (PeriOperative ISchemic Evaluation-2) trial of 10,010 non-cardiac surgery patients in 23 countries, clonidine did not reduce all-cause death or MI at 30 days after surgery compared with placebo (hazard ratio [HR] 1.08, 95% CI 0.93–1.26). In contrast, clinically significant hypotension (HR 1.3, 95% CI 1.24–1.4) and non-fatal cardiac arrest (HR 3.2, 95% CI 1.17– 8.76) were more frequently observed in the clonidine group.159
Based on these results, starting α2
agonists in the perioperative period of non-cardiac surgery is not recommended (Table 26). However, due caution is required for patients under long-term α2
agonist therapy, as withholding it can cause rebound phenomenon, such as hypertension, headache, or tremor.160
1.5 Calcium-Channel Blockers
A 2003 meta-analysis of 11 RCTs including 1,007 patients161 showed that the perioperative use of calcium-channel blockers significantly reduced myocardial ischemia and supraventricular tachycardia. There was also a trend toward reduction in all-cause death and MI. A subgroup analysis of the meta-analysis suggested that diltiazem contributed to a decrease in cardiovascular events.161
On the other hand, a retrospective observational study of 1,000 patients in Hungary suggested that the use of dihydropyridine calcium-channel blockers in emergency or elective thoracic or abdominal aortic aneurysm surgery was independently associated with an increase in perioperative death.162
1.6 Aspirin (Table 27)
Table 27. Recommendations and Levels of Evidence on Perioperative Use of Aspirin
| |
COR |
LOE |
Starting aspirin for primary prevention of cardiovascular event before non-cardiac
surgery should not be performed |
III
(Harm) |
B |
Whether to continue or discontinue aspirin as chronic maintenance therapy in the
perioperative period should be decided by weighing the risks of thrombosis and bleeding* |
I |
B |
*See Chapter IV.4. Management of Antiplatelet Agents for Coronary Artery Disease. COR, class of recommendation; LOE, level of evidence.
1.6.1 Effects of Continuation or Discontinuation of Long-Term Aspirin Therapy
A meta-analysis of RCTs (5 trials with 666 patients) showed no significant differences in the rates of bleeding requiring transfusion, repeated surgery for hemostasis, or incidence of acute MI and cerebral infarction between perioperative continuation and discontinuation of antiplatelet agents.163 The patients included in the meta-analysis had high-risk features such as prior history of coronary artery disease, cerebral infarction, and arteriosclerosis. Indeed, the rates of post-coronary artery stent placement in patients in each study were 25%,164 65%,165 13%,166 unknown,167 and 0%.168
Whether to continue or discontinue long-term oral aspirin therapy should be determined by weighing the risk of thrombosis and bleeding. Aspirin should be withheld in surgeries with high bleeding risk, such as spinal, intracranial and posterior chamber eye surgeries.
1.6.2 Preoperative Initiation
In the POISE-2 trial, the preoperative initiation of aspirin did not significantly reduce all-cause death or MI, while increasing major bleeding events at 30 days after non-cardiac surgery in patients with high-risk features for perioperative cardiovascular complications such as a history of coronary artery disease and stroke.169 This result was consistent regardless of oral aspirin therapy status before enrollment. However, the trial excluded patients at high risk of bleeding, such as those undergoing intracranial or retinal surgeries or those with a history of cerebral hemorrhage in the past 6 months. Preoperative administration of aspirin for primary prevention is not recommended.
1.7 SGLT-2 Inhibitors
Sodium glucose cotransporter 2 (SGLT-2) inhibitors are used for diabetes mellitus and chronic heart failure. The risk of ketoacidosis associated with SGLT-2 inhibitors increases perioperatively and is, thus, contraindicated as indicated on the package insert.
2. Pharmacotherapy (Anticoagulant Therapy)
2.1 Anticoagulant Therapy in the Perioperative Period
For patients receiving chronic anticoagulation therapy, the following issues must be assessed: (1) whether to continue or discontinue anticoagulant therapy during the perioperative period; and (2) if discontinued, whether heparin bridging is used to minimize the discontinuation period. Assessment of bleeding and thrombosis risk is important for determining the above.
2.2 Assessment of Bleeding Risk
2.2.1 Factors Related to Patients
The HAS-BLED score assesses the risk of bleeding for patients with atrial fibrillation.170 Both in non-perioperative setting and perioperative setting, a score ≥3 points indicates a higher risk of bleeding.171,172 However, risk factors in the HAS-BLED score such as hypertension, brain stroke, and advanced age are also common to the CHADS2
score.173 Thus, it is important to note that patients at higher risk of bleeding are also inherently at a higher risk of thrombosis. Furthermore, unstable INR in the HAS-BLED score assumes patients are receiving warfarin therapy. The ATRIA bleeding score174 and ORBIT score,175 which were simplified by eliminating the INR, were developed for patients receiving direct oral anticoagulants (DOACs), but the predictive ability was not significantly different.
2.2.2 Factors Related to Surgery
Bleeding risk factors related to surgery include (1) bleeding tendency related to surgical technique, (2) organ injury caused by bleeding, and (3) easiness in hemostasis in case of bleeding.176,177 For example, intracranial surgery and supine surgery are high-risk because bleeding at the surgical sites can result in sequelae such as paralysis leading to significant decline in quality of life, whereas superficial surgeries are lower risk because the wounds are small and hemostasis can be easily achieved. The surgical procedures classified by bleeding risk are shown in Table 28.178
Table 28. Classification of Surgeries by Bleeding Risk in Patients Receiving Anticoagulant Therapy for Atrial Fibrillation [Partially Modified From the JCS/Japanese Heart Rhythm Society (JHRS) 2020 Guideline on Pharmacotherapy of Cardiac Arrhythmias Japanese Circulation Society]
171
| Low bleeding risk |
Intermediate bleeding risk |
High bleeding risk |
Dental surgery
[extraction, abscess incision, paradontal
surgery, implant positioning, etc.]
Cataract surgery
Diagnostic gastroenterological endoscopic
procedures without biopsy
[upper/lower gastroenterological
endoscopy, capsule endoscopy, ERCP,
etc.]
Superficial surgery
[abscess incision, dermatologic excisions,
etc.]
Breast biopsy, mammotome biopsy |
Gastroenterological endoscopic procedures
with low bleeding risk
[balloon-assisted endoscopy,
gastroenterological pancreatic duct/biliary
duct stenting, endoscopic papillary balloon
dilation, etc.]
Endoscopic mucosal biopsy
Prostate biopsy
Transurethral surgery
[bladder biopsy, TUR-Bt, PVP, TUL, etc.]
Percutaneous nephrostomy
Glaucoma or vitreous surgery
Arthroscopic surgery
Mastectomy
Otorhinolaryngological surgery, head and
neck surgery
Cardiac device implantation
Angiography, intravascular surgery
Electrophysiological study or catheter
ablation (except AF ablation) |
Gastroenterological endoscopic procedures
with high bleeding risk
[polypectomy, ESD, etc.]
Transbronchial lung biopsy
Spinal or epidural anesthesia
Craniotomy, spinal cord surgery
Carotid endarterectomy
Thoracic surgery (including thoracoscopic
surgery)
Abdominal/pelvic surgery (including
laparoscopic surgery)
Breast cancer surgery
Major orthopedic surgery
Reconstructive surgery for head and neck
cancer
Lower extremity artery bypass surgery
Liver biopsy
Kidney biopsy
Transrectal prostate biopsy
TUR-P
ESWL
PNL |
AF, atrial fibrillation; ERCP, endoscopic retrograde cholangiopancreatography; ESD, endoscopic submucosal dissection; ESWL, extracorporeal shockwave lithotripsy; EUS-FNA, endoscopic ultrasonography guided fine-needle aspiration; PNL, percutaneous nephrolithotripsy; PVP, photoselective vaporization of the prostate; TUL, transurethral ureterolithotripsy; TUR-Bt, transurethral resection of the bladder tumor; TUR-P, transurethral resection of the prostate. Adapted as follows: Minor risk, and low risk in the original document were termed low bleeding risk and intermediate risk, respectively. Endoscopic duodenal papillectomy, endoscopic treatment of esophageal and gastric varices, EUS-FNA were removed. High bleeding and thromboembolic risk and AF ablation were removed.
Overall perioperative bleeding risk should be assessed on an individual case basis. An anesthesiologist should be involved in assessing bleeding risk related to regional anesthesia (see Chapter III.3).
2.3 Assessment of Risk of Thromboembolism
The perioperative risk of thromboembolism in various clinical settings is summarized in Table 29, based the American College of Chest Physicians (ACCP) evidence-based clinical practice guidelines 2012,179 and modified for clinical use in Japan.
Table 29. Risk of Thromboembolism for Atrial Fibrillation, Venous Thromboembolism, and Mechanical Valves
Risk of thromboembolism
(annual incidence) |
Atrial fibrillation |
Venous thromboembolism
(VTE) |
Mechanical heart valve |
| High (>10%) |
Rheumatic valvular heart disease*1
Recent cerebral infarction or TIA
(within 3 months) |
Recent VTE (within 3–6 months)*3
Severe thrombophilia*4 |
Mitral valve
History of brain infarction or TIA
within the past 6 months (high-risk
for 3 months after replacement with
bioprosthetic valve) |
| Moderate (5–10%) |
CHADS2 score*2: 4–6 points |
VTE (within 3–6 to 12 months)*2
Non-severe thrombophilia*5
VTE recurrence
Active cancer
Absence of provoking factors*6 |
Aortic valve (bileaflet valve) and
presence of risk factors*7 |
| Low (<5%) |
CHADS2 score*2: 0–3 points and
no history of cerebral infarction or
TIA |
VTE (>12 months after the event) |
Aortic valve (bileaflet valve) and
absence of risk factors*7 |
*1Moderate to severe mitral valve stenosis.
*2Classified according to the risk of thromboembolism of Japanese patients.180
*3Account for the high rate of recurrence within the 6 months after the onset of VTE.181
*4Protein C, S deficiency, AT-III deficiency, and antiphospholipid syndrome.
*5Factor V Leiden mutation, prothrombin gene mutation.
*6Transient risk or reversible risk factors such as surgery within 3 months, trauma, cast fixation, pregnancy, oral contraceptives, and hormone replacement therapy (Guidelines for diagnosis, treatment and prevention of pulmonary thromboembolism and deep vein thrombosis, Japanese Circulation Society [JCS] 2017).182
*7History of atrial fibrillation, cerebral infarction or TIA, hypertension, diabetes mellitus, heart failure, and age ≥75 years.179
TIA, transient ischemic attack; VTE, venous thromboembolism. (Adapted from Douketis JD, et al. 2012,179 Suzuki S, et al. 2015,180 Schulman S, et al. 1995,181 and JCS 2017.182)
2.3.1 Risk of Thromboembolism in Atrial Fibrillation
The JHRS 2020 Guideline on pharmacotherapy of cardiac arrhythmias of the Japanese Circulation Society recommends the CHADS2
score for assessing the risk of thrombosis in non-valvular atrial fibrillation.171 In Japan, CHADS2
score ≤3 can be classified as low-risk (<5%/year) for thromboembolism, and 4–6 points as moderate risk (5–10%/year) of thromboembolism if applied the ACCP classifications (Table 29).
2.3.2 Risk of Recurrence in Venous Thromboembolism
The ACCP guidelines 2012 states that the risk of recurrence is higher in the first year after VTE than in subsequent years for patients with a history of VTE who discontinue anticoagulants183 (Table 30). Furthermore, the recurrence rate after discontinuing anticoagulant therapy within 6 months after VTE is approximately 10%.181 The recurrence risk is moderate–high in patients with active cancer, thrombophilia or without provoking factor (Table 29).
Table 30. Recurrence Rate With Discontinuation of Anticoagulants in Venous Thromboembolism (VTE) Patients
| |
Within 1 year
after VTE |
After 1 year
(/year) |
| First VTE after surgery |
1.0% |
0.5% |
| First VTE provoked by non-surgical factor* |
5.0% |
2.5% |
| First unprovoked VTE |
10% |
5.0% |
| Second unprovoked VTE |
15% |
7.5% |
*Estrogen therapy, pregnancy, leg injury, flight of 8 h. (Adapted from Kearon C, et al. 2012.183)
2.3.3 Risk of Thrombosis With Artificial Valves
After mechanical valve replacement patients require lifelong anticoagulant therapy with warfarin. Notably, the occurrence of thromboembolism is reported to be 1–2%/year with warfarin,184 and 8%/year without anticoagulant therapy.185 In particular, patients with replacement of the mitral valve with a mechanical valve, reduced left ventricular ejection fraction, atrial fibrillation, and a history of thromboembolism are at high risk.18 Moreover, the risk of thromboembolism is higher (13%) in the first 3 postoperative months after bioprosthetic valve replacement without warfarin therapy,186,187 but anticoagulant therapy can be discontinued after 3 months in the absence of other indications for anticoagulant therapy.18
2.4 Decision-Making for Continuation or Withholding of Anticoagulant Therapy (Table 31)
Table 31. Recommendations and Levels of Evidence on Continuation or Withholding of Anticoagulant Therapy
| |
COR |
LOE |
Continuation of anticoagulant therapy is recommended in surgical interventions with a
low risk of bleeding |
I |
A |
Continuation or withholding of anticoagulant therapy should be determined by weighing
the risks of thromboembolism and bleeding in surgical interventions with a moderate risk of bleeding |
I |
C |
Withholding of anticoagulant therapy is recommended in surgical interventions with
high risk of bleeding |
I |
C |
Practical considerations: It is important to weigh up the thrombosis and bleeding risk carefully and share the risks and benefits of continuation or withholding of anticoagulant therapy with the patient. The withholding period of anticoagulant therapy should also be determined individually. COR, class of recommendation; LOE, level of evidence.
Temporary perioperative withholding of anticoagulant therapy should be decided after weighing the risks of thrombosis and bleeding for each patient.
2.5 Strategies for Withholding Anticoagulant Therapy
The method for perioperative withholding of anticoagulant therapy varies according to the type of anticoagulant drug. The ACCP Guidelines 2012 recommend withholding warfarin 5 days before surgery, but suspending warfarin 3 days before surgery is also accepted depending on the type of surgery or bleeding risk.179 This recommendation is based on a study reporting that only 7% of patients had INR ≥1.5 if warfarin was withheld 5 days before the surgery,188 and on another reporting that the median INR was 1.8 on the day of surgery when warfarin was discontinued 2–3 days prior to surgery.189 Table 32 summarizes the methods used to withhold warfarin.190 The duration of anticoagulant interruption must be adjusted for patients on DOACs depending on the type of surgical procedure, bleeding risk, and renal function177,178 (Table 33).
Table 32. Discontinuation of Warfarin Before Surgery
| Day |
Warfarin dose |
INR monitoring |
| −5 |
Consider stopping warfarin |
Laboratory tests (INR, hemoglobin, platelet count,
serum creatinine) |
| −4 |
Consider stopping warfarin |
|
| −3 |
Stop warfarin |
|
| −2 |
Stop warfarin |
|
| −1 |
Stop warfarin |
Check INR
INR <1.5: go to surgery
INR 1.5–1.8: consider oral vitamin K reversal |
| 0* or +1 |
Resume the maintenance dose of warfarin on the
evening of the procedure or the morning after the
procedure |
|
*Day of surgery. INR, international normalized ratio. (Adapted from Spyropoulos AC, et al. 2016.190)
Table 33. Time Frames for Discontinuation and Reimplementation of Anticoagulant Therapy in Patients Undergoing Elective Surgery
| A. Low surgical bleeding risk |
| |
Day −5 |
Day −4 |
Day −3 |
Day −2 |
Day −1 |
Day of
surgery |
Day +1 |
Day +2 |
Day +3 |
| DOAC |
〇 |
〇 |
〇 |
〇 |
△
≥12 h |
△
Resume
≥6–8 h p
ost
surgery |
〇 |
〇 |
〇 |
| B. Intermediate surgical bleeding risk |
| |
CCr
(mL/min) |
Day −5 |
Day −4 |
Day −3 |
Day −2 |
Day −1 |
Day of
surgery |
Day +1 |
Day +2 |
Day +3 |
| Dabigatran |
≥80 |
〇 |
〇 |
〇 |
〇 |
△
≥24 h |
△
Resume
≥6–8 h
post
surgery |
〇 |
〇 |
〇 |
| 50~79 |
〇 |
〇 |
〇 |
△
≥36 h |
╳ |
〇 |
〇 |
〇 |
| 30~49 |
〇 |
〇 |
〇 |
△
≥48 h |
╳ |
〇 |
〇 |
〇 |
Rivaroxaban
Apixaban
Edoxaban |
≥30 |
〇 |
〇 |
〇 |
〇 |
△
≥24 h |
〇 |
〇 |
〇 |
| 15~29 |
〇 |
〇 |
〇 |
△
≥36 h |
╳ |
〇 |
〇 |
〇 |
| C. High surgical bleeding risk |
| |
CCr
(mL/min) |
Day −5 |
Day −4 |
Day −3 |
Day −2 |
Day −1 |
Day of
surgery |
Day +1 |
Day +2 |
Day +3 |
| Dabigatran |
≥80 |
〇 |
〇 |
〇 |
△
≥48 h |
╳ |
△
Resume ≥6–8 h post
surgery (as soon as
possible) with
consideration of
bleeding status |
△
Resume ≥48–72 h
post surgery in
case of uncontrolled
bleeding |
| 50–79 |
〇 |
〇 |
△
≥72 h |
╳ |
╳ |
| 30–49 |
〇 |
△
≥96 h |
╳ |
╳ |
╳ |
Rivaroxaban
Apixaban
Edoxaban |
〇 |
〇 |
〇 |
△
≥48 h |
╳ |
〇: Intake. △: Decision to take or not take medicine based on the timing of surgery and the patient’s condition.
The time within the parenthesis denotes the recommended timing of last intake before surgery. ╳: Stop therapy.
Although guidelines for resumption are listed, consensus of the surgeons and anesthesiologists (during spinal anesthesia) is important for the actual timing of resumption. When there is some concerns of bleeding risk after surgery, heparin bridging may be considered, which potentially allows prevention of thromboembolism as well as easier management of bleeding.
Table 26 in the “JCS 2020 Guideline focused update on antithrombotic therapy in patients with coronary artery disease” was modified as follows:
• Descriptions related to warfarin were deleted.
• A. Minor surgical bleeding risk or procedures where bleeding is easily controlled (e.g., teeth extractions, superficial surgery) → Low surgical bleeding risk.
• B. Low surgical bleeding risk → Intermediate surgical bleeding risk.
• C. Intermediate or high surgical bleeding risk → High surgical bleeding risk.
• The following annotations were deleted. “*Heparin bridging is not generally recommended. However, if strict control of anticoagulation is required (e.g., patients who have an artificial valve), heparin bridging may be considered. In addition,”
• The following annotations were described. “Although guidelines for resumption are listed, consensus with the surgeons and anesthesiologists (during spinal anesthesia) is important for the actual timing of resumption.”
• Asterisks were deleted.
DOAC, direct oral anticoagulant. (Adapted from Nakamura M, et al. 2020.178)
2.6 Heparin Bridging for Withholding of Anticoagulant Therapy
Whether to perform heparin bridging while withholding anticoagulant therapy should be decided.
2.6.1 Warfarin
A systematic review and meta-analysis were performed to evaluate this topic for patients with atrial fibrillation on warfarin in Part 2 CQ 6 of the present clinical practice guidelines (see details in the corresponding section). The analysis included 1 RCT191 and 7 observational studies.192–198 The systematic review ultimately did not recommend preoperative heparin bridging in non-cardiac surgery for patients taking warfarin for atrial fibrillation when the risk of thromboembolism was not high (CHADS2
score ≤4 points) (GRADE 2B).
a. Patients Who Could Benefit From Heparin Bridging
Heparin bridging can be considered for patients who meet the following conditions with a high-risk of thromboembolism (Table 29) and low-risk of bleeding.
(1) Atrial fibrillation: CHADS2
score 5–6 points or complicated with moderate to severe mitral stenosis
(2) Pulmonary thromboembolism: within 3–6 months of onset
(3) Cerebral infarction: within 3 months after embolic cerebral infarction
(4) Artificial valve: within 3 months after bioprosthetic valve replacement
With regard to mechanical valves, heparin bridging should be considered (JCS Guideline 202018). The ESC Guidelines 2017 recommended heparin bridging as a Class I recommendation.199 Similarly, the 2017 AHA/ACC Guideline recommends heparin bridging as a Class I recommendation in post-mitral valve replacement patients and post-aortic valve replacement patients with risk of thrombosis.200
2.6.2 Direct Oral Anticoagulants
Given the short half-life of DOACs, they can be managed by short-term preoperative withholding of the drugs. In a subanalysis of the RE-LY trial,192 major bleeding was significantly increased in patients who underwent heparin bridging following the withholding of dabigatran compared with those not on heparin bridging. On the other hand, there were no significant differences in the incidence of thromboembolism between the groups.192 Therefore, heparin bridging with preoperative withholding of DOACs is not recommended.177
Table 34 shows the general guidance for withholding anticoagulants and heparin bridging according to the risk of thrombosis and bleeding.190 Although withholding anticoagulants is generally recommended in patients who have moderate bleeding risk, it should be carefully considered in each patient. Table 35 shows the recommendations of a heparin bridge in the case of withholding anticoagulants.
Table 34. Suggested Overall Periprocedural Anticoagulant and Bridging Management for Patients Receiving Chronic Oral Anticoagulants (Including Vitamin K Antagonists and Direct Oral Anticoagulants [DOACs]) Based on Thromboembolic and Procedural Bleeding Risks
| |
|
High bleeding risk |
Intermediate bleeding risk |
Low bleeding
risk |
Thromboembolic risk:
High |
Warfarin |
• Stop warfarin
• Heparin bridge |
• Stop warfarin
• Consider heparin bridge in many cases |
Continue oral
anticoagulant |
| DOAC |
• Stop DOAC
• No heparin bridge |
• Stop DOAC
• No heparin bridge |
Thromboembolic risk:
Intermediate |
Warfarin |
• Stop warfarin
• Consider heparin bridge
in some case |
• Stop warfarin
• No heparin bridge in many cases except for
mechanical valves |
Continue oral
anticoagulant |
| DOAC |
• Stop DOAC
• No heparin bridge |
• Stop DOAC
• No heparin bridge |
Thromboembolic risk:
Low |
Warfarin |
• Stop warfarin
• No heparin bridge |
• Stop warfarin
• No heparin bridge except for mechanical valves |
Continue oral
anticoagulant |
| DOAC |
• Stop DOAC
• No heparin bridge |
• Stop DOAC
• No heparin bridge |
(Adapted from Spyropoulos AC, et al. 2016.190)
Table 35. Recommendations and Levels of Evidence on the Withholding of Anticoagulants and Heparin Bridging
| |
COR |
LOE |
| Routine heparin bridging while withholding warfarin is not recommended |
III (No
benefit) |
C |
Heparin bridging should be considered while withholding warfarin in patients with
post-mechanical valve replacement or with valvular atrial fibrillation |
IIa |
C |
Heparin bridging should be considered while withholding warfarin in patients who have
high-risk of embolism and low-risk of bleeding |
IIa |
C |
| Heparin bridging while withholding DOACs is not recommended |
III (No
benefit) |
B |
COR, class of recommendation; DOAC, direct oral anticoagulant; LOE, level of evidence.
Table 36. Recommendation and Level of Evidence on Antithrombotic Agents to Be Considered for Regional Anesthesia
| |
COR |
LOE |
A multidisciplinary discussion including an anesthesiologist is recommended to determine
whether to continue or discontinue antithrombotic agents in a patient receiving
antithrombotic therapy and undergoing non-cardiac surgery where regional anesthesia is
an option |
I |
C |
COR, class of recommendation; LOE, level of evidence.
Nerve injury linked to spinal hematomas is a serious complication when administering spinal or epidural anesthesia to patients on antithrombotic therapy, including antiplatelet agents and anticoagulants.201
Hematoma following spinal anesthesia is rare in comparison with epidural anesthesia. Although nonsteroidal anti-inflammatory drugs (NSAIDs), including aspirin, are not contraindications for spinal anesthesia, some patients require the suspension of these medications and thus require individual assessment.201
Conversely, suspension of antithrombotic drugs is recommended when administering epidural anesthesia, which often requires catheter insertion, to patients on anticoagulant therapy.201 Of the antiplatelet agents, withholding NSAIDs is not necessary, but the suspension of ticlopidine and clopidogrel is recommended.201 As for epidural catheter placement in a patient treated with antithrombotic agents, it is important to assess the best decision, namely, (1) continuation or discontinuation before catheter placement, (2) whether or not to reintroduce after catheter placement and its timing if reintroducing the agents, and (3) the timing of the discontinuation or reintroduction of treatment following catheter removal.
As for performing a deep peripheral nerve block, withholding antithrombotic agents is ideal, although they can be continued for superficial peripheral nerve blocks.201
IV. Perioperative Management of Cardiovascular Disease
1. Acute Coronary Syndrome (ACS)
1.1 Epidemiology of Non-Cardiac Surgery in Post-ACS Patients
Cardiac death is a major cause of death within 30 days after ACS. Causes of death beyond 30 days are non-cardiac in two-thirds of cases,202 suggesting that long-term prognosis is more influenced by non-cardiac diseases. A sizable number of patients require open surgery in the chronic post-ACS stage due to the development of non-cardiac disease. A study of 3,942 post-ST-elevation myocardial infarction (MI) patients in Japan reported that patients underwent cardiac surgery other than coronary artery bypass grafting (CABG) or non-cardiac surgery within 1, 6, and 12 months at rates of 4.2%, 8.4%, and 11.6%, respectively, and this frequency increases with age.203
1.2 Timing of Post-ACS Elective Non-Cardiac Surgery and Perioperative Cardiovascular Complications
Several studies have reported on the interval between acute MI (AMI) to non-cardiac surgery and perioperative risk.
• Risk is highest immediately after MI, subsequently decreases gradually to the minimum level at 6 months, and then increases again204
• Risk is highest immediately after MI, decreases gradually, and essentially plateaus after 3 months205
• Risk is higher at ≤42 days than at >42 days206
• Risk is highest at ≤30 days, gradually decreases, and essentially plateaus at 3 months and onwards207,208
Collectively, the perioperative cardiovascular risk is highest within the first month after the MI event, and thereafter, the risk decreases gradually over 3–6 months (Figure 4, Table 37).
Table 37. Recommendations and Levels of Evidence on Post-ACS Non-Cardiac Surgery
| |
COR |
LOE |
| Elective non-cardiac surgery should be avoided until after 1 month from ACS onset* |
I |
C |
Emergency revascularization should be performed for patients scheduled for
non-cardiac surgery who have an ACS |
I |
C |
Performing complete revascularization before the non-cardiac surgery may be considered
for patients scheduled for non-cardiac surgery who have had an ACS |
IIb |
B |
*See also Chapter IV.3. Timing of Non-Cardiac Surgery After Percutaneous Coronary Intervention for the duration of the delay from ACS onset. ACS, acute coronary syndrome; COR, class of recommendation; LOE, level of evidence.
The JCS 2020 Guideline advises to “delay non-cardiac surgery for patients who underwent PCI for ACS until after ≥6 months”.178 The ESC 2017 Guidelines on dual antiplatelet therapy (DAPT) for coronary artery disease recommend delaying elective procedures for 6 months if DAPT is required after AMI (Class IIb).210
Individual assessments must be made to consider when to perform non-cardiac surgery according to the level of urgency of the non-cardiac surgery, myocardial ischemic risk, hemorrhagic risk, and the pros and cons of withholding antiplatelet therapy. Postponing surgery for a long time is not always an option, depending on the primary disease, as in the case of malignancy.
1.3 Revascularization and Perioperative Cardiovascular Risk for Patients With History of MI
1.3.1 Revascularization Status and Risk of Postoperative MI
When patients undergo non-cardiac surgery within 1 month after MI, patients with stent implantation for MI have a higher risk than patients without revascularization. In contrast, patients undergoing CABG for MI have lower risk than patients without revascularization208 (Figure 5).
1.3.2 Revascularization Status and Risk of Postoperative Death
When patients undergo non-cardiac surgery 2 months after MI, either revascularization with stent implantation or CABG for MI decreased perioperative mortality208 (Figure 5).
1.3.3 Influence of Complete Revascularization After MI on Perioperative Risk
In the non-perioperative setting, complete revascularization reduces the risk of cardiovascular death or MI in comparison with culprit-lesion-only percutaneous coronary intervention (PCI).211 In terms of patients undergoing non-cardiac surgeries, there have been no prospective studies evaluating the effect of complete revascularization on the perioperative outcomes for patients with a history of MI.
2. Coronary Artery Revascularization Before Non-Cardiac Surgery
2.1 Mechanism of Perioperative AMI
According to the 4th Universal Definition of MI, there are 5 types of MI.212 Two major mechanisms underlying the development of perioperative MI have been proposed: (1) plaque rupture due to plaque vulnerability from the release of inflammatory cytokines and increased catecholamines and increased shear stress to vessels,213 and thrombosis due to hypercoagulability (Type 1); and (2) myocardial ischemia due to imbalance between myocardial oxygen supply and demand, which is caused by bleeding, associated anemia and increased catecholamines that accelerate heart rate with resulting increased myocardial oxygen consumption, hypotension related to anesthesia and anemia with resulting decreased oxygen supply (Type 2).213,214 Otherwise, prior history of stent implantation can induce stent thrombosis-related AMI perioperatively (Type 4b).
2.2 Features of Perioperative AMI
Of 146 patients who underwent coronary angiography for AMI occurring within the 30 days following non-cardiac surgery, Types 1, 2, and 4b accounted for 25%, 73%, and 2.1%, respectively, with the majority being Type 2.215 An optical coherence tomography study of coronary artery lesions revealed that only 13.3% of perioperative AMI cases can be attributed to thrombosis,216 also providing further evidence that Type 2, rather than Type 1, is the main mechanism underlying AMI. Type 1 AMI has been reported to occur more frequently postoperatively rather than intraoperatively (intraoperative onset: 2.7% vs. postoperative onset 91.1%).215 The incidences of intra- and postoperative Type 2 AMI are 20.8% and 77.4%, respectively.215
2.3 Perioperative Risks of Ischemic Heart Disease
The risk of perioperative cardiovascular events increases in the presence of ischemic heart disease. A meta-analysis reported that perioperative cardiovascular events occurred in 2.0% of patients with no evidence of stenotic lesions on coronary computer tomography angiography, whereas the rate increased to 4.1% in patients with non-significant stenotic lesions, to 7.3% in patients with a single-vessel lesion, and to 23.1% in patients with multivessel lesions.112 Furthermore, the rate of perioperative cardiovascular events in patients with ischemic findings on preoperative exercise stress myocardial perfusion imaging was higher than that of patients without preoperative ischemic findings (16% vs. 1.7%).123
2.4 Preoperative Revascularization
Evidence on whether preoperative revascularization reduces perioperative risk, that are also adopted in Part 2 of the present clinical practice guidelines will be presented below.
• CARP trial:217 510 patients at increased risk for perioperative cardiac complications and clinically significant coronary artery disease based on coronary angiography were randomly assigned to a group that underwent revascularization before major vascular surgery (n=258) and a group that did not (n=252). From the date of randomization to postoperative day 30, 17 patients who were assigned to revascularization died (6.6%) compared with 9 patients (3.6%) who were assigned to no revascularization, showing in a numerically higher perioperative mortality in revascularization group. Of note, 10 of the 258 (3.9%) patients assigned to revascularization (4 deaths were associated with revascularization) and 1 of the 252 (0.39%) patients assigned to no revascularization died before non-cardiac surgery. The potential cause is the significantly longer period from randomization to surgery for the patients who underwent revascularization (median: 54 vs. 18 days). There was no difference in the 30-day postoperative mortality rate (3.1% vs. 3.4%), nor was there a difference in the incidence of AMI (8.4% vs. 8.4%). There was no significant difference in long-term mortality of 22% vs. 23% at a median of 2.7 years (risk ratio [RR], 0.98; 95% confidence interval [CI] 0.70–1.37; P=0.92)
The CARP trial showed that performing prophylactic revascularization before major vascular surgery in patients with stable coronary disease improved neither the perioperative nor the long-term clinical course.217,218 Notably, 10 patients in revascularization group died between revascularization and the vascular surgery suggesting the importance of recognizing both the risks associated with delaying the non-cardiac surgery to perform revascularization and the risks of revascularization itself. However, the CARP trial excluded patients with left main trunk disease, severe aortic stenosis (AS), and ejection fraction (EF) <20%. Indeed, a subanalysis of the CARP trial found that revascularization improves long-term prognosis in patients with left main trunk disease revascularization.218 Aside from the CARP trial, the following findings have been reported.
• A retrospective cohort study found that PCI performed before non-cardiac surgery for patients with coronary artery disease decreased rates of perioperative angina (6.4% vs. 12%, P<0.001) and the risk of heart failure (7.9% vs. 16.9%, P<0.001) compared with patients who did not undergo PCI, but that the rate of AMI (2.2% vs. 2.8%, P=0.304) and risk of death remained unchanged (2.6% vs. 2.9%, P=0.436)219
• A retrospective cohort study that analyzed patients with moderate and severe reversible ischemia detected by exercise stress myocardial perfusion imaging before vascular surgery found no significant difference in the 30-day postoperative survival between patients who underwent preoperative revascularization and those who did not (98.7% vs. 97.8%)220
• A retrospective study found no difference in the incidence of 30-day postoperative events (cardiovascular death, ACS, revascularization) in patients with coronary artery disease who underwent coronary revascularization before renal transplant and those who did not undergo revascularization (6% vs. 4%, P=0.54)221
• Two other studies also failed to find significant differences in the incidence of perioperative events according to whether or not revascularization was performed222,223
The results of the systematic review in Part 2 of the present clinical practice guidelines also found that neither the perioperative mortality rate nor the incidence of AMI was decreased by performing preoperative revascularization for patients with stable coronary artery disease and therefore concluded that coronary revascularization was ineffective. The present clinical practice guidelines, therefore, do not recommend prophylactic revascularization (Table 38) (GRADE 1B). This is consistent with a previous meta-analysis.224
Table 38. Recommendations and Levels of Evidence on Preoperative Revascularization to Non-Cardiac Surgery for Patients With Stable Coronary Artery Disease
| |
COR |
LOE |
Preoperative prophylactic revascularization to lower the surgical risk should not be performed
for patients with stable coronary artery disease undergoing non-cardiac surgery |
III
(Harm) |
B |
For patients with a lesion that is a Class I indication for revascularization, it is
recommended whether or not to perform revascularization before non-cardiac surgery
should be discussed by a multidisciplinary team involving a surgeon, an internal
medicine physician, a cardiologist and an anesthesiologist and also shared decision
making with the patient should be performed |
I |
C |
COR, class of recommendation; LOE, level of evidence.
2.5 Other Clinical Practice Guidelines
According to the 2014 ESC guidelines, “revascularization can be considered for patients with severe ischemia undergoing high-risk surgery (Class IIb, B)”.4 According to the AHA guideline 2014, “It is not recommended that routine coronary revascularization be performed before noncardiac surgery exclusively to reduce perioperative cardiac events” and “Revascularization before noncardiac surgery is recommended in circumstances in which revascularization is indicated according to existing clinical practice guidelines”.59 However, it states the importance of evaluating the risk of revascularization and functional capacity in patients for the ultimate decision.59 The 2017 Canadian Cardiovascular Society (CCS) clinical practice guidelines recommend against preoperative prophylactic coronary revascularization (Strong Recommendation).10
Further discussion on the indications for revascularization is expected to continue, with additional findings from the ISCHEMIA trial.225
2.6 Decision-Making for Preoperative Revascularization
There are several options for patients with severe coronary artery disease for which revascularization is Class I indication (left main trunk disease and coronary artery lesions for which symptoms are observed with mild effort (CCS III, IV), such as (1) revascularization before non-cardiac surgery, (2) revascularization after non-cardiac surgery, (3) performing either revascularization or non-cardiac surgery, or (4) performing neither due to high risks of both (Figure 6). For decisions that involve dilemmas, such as a non-cardiac surgery that cannot be delayed sufficiently or very high risk of revascularization, a comprehensive assessment by a multidisciplinary team and understanding the patient’s values and preferences through shared decision-making is important (Table 38) (see Chapter I. Overview of Perioperative Cardiovascular Management) (Figure 1).226,227
3. Timing of Non-Cardiac Surgery After Percutaneous Coronary Intervention
Given the recent advancement in PCI techniques, devices, and related pharmacotherapy, the present clinical practice guidelines reviewed reports from 2008 and later.204–209,228–250
3.1 Risks of Non-Cardiac Surgery for Patients With History of PCI
In a retrospective study comparing post-PCI patients and age-, sex-, and surgery-matched controls without ischemic heart disease, the cardiac mortality within the first 30 postoperative days was 1.1% vs. 0.2% (odds ratio [OR] 5.87) and MI occurred in 1.5% vs. 0.2% (OR 4.82), revealing that the risk was higher in post-PCI patients.232 Another study found a higher rate of cardiovascular events (death, MI, revascularization) within 30 postoperative days in patients with stent placement than controls without a history of stent placement after matching surgery type and patient background (5.7% vs. 3.6%).233 Higher rates of cardiovascular events within 30 postoperative days (death, MI, cardiac arrest, brain stroke) have also been reported in patients with stent placement in another study (3.7% vs. 1.5%).234
3.2 Timing of Surgery After PCI
Of the 29 relevant articles, 27 (excluding references 249, 250) reported better perioperative outcomes after waiting longer to perform the surgery after PCI.204–209,228–248 Although the interval between PCI and surgery varies among the studies, high rates of major adverse cardiac events (11–43%) are reported with surgery performed within 1 to 1.5 months after PCI, with an OR ranging from 2.3 to 11.5.
The results of the systematic review in Part 2 CQ2 were similar, and elective non-cardiac surgery within 1 to 1.5 months of coronary artery stent placement is not recommended (Class III harm, Level of Evidence B) (GRADE 1C). These previous studies suggest that if the patient’s condition allows for a delay of surgery, risk can be reduced by postponing surgery until 1-year post-PCI whenever possible. Comprehensive assessment by a surgeon, anesthesiologist, and a cardiologist for the best timing, and then shared decision-making with the patient (Figure 1), is recommended for time-sensitive surgeries.
3.3 Potential Differences by Indication for PCI (ACS vs. Non-ACS) and Perioperative Risks
Previous studies have been inconsistent, whereby some studies have reported that patients who underwent PCI for ACS had higher rates of perioperative events than patients who underwent PCI for non-ACS,204,206,207,229,232,242,250 and others have reported that whether it is ACS or non-ACS, was not a predictor of events.230,237–239,243,249
Perioperative cardiovascular event risk such as the risk of ACS, also decline during longer waiting periods until surgery (see Chapter IV.1 Acute Coronary Syndrome: 1.2 Timing of Post-ACS Elective Non-Cardiac Surgery and Perioperative Cardiovascular Complications). If possible, a delay in performing surgery decreases risk for both ACS and non-ACS.
Given the above, the present clinical practice guidelines will not specify the recommendation whether or not the indication for PCI was ACS.
4. Management of Antiplatelet Agents for Coronary Artery Disease
Lifelong aspirin therapy (or P2Y12 receptor antagonist therapy for patients with contraindications for aspirin) is recommended as a secondary prevention for patients with coronary artery disease, regardless of whether they have undergone PCI.178,251 However, antiplatelet therapy is associated with perioperative cardiovascular risk on its discontinuation and with bleeding risk by continuing the treatment.
4.1 Coronary Artery Disease Patients Without a History of PCI
There have not been any reports investigating long-term use of aspirin only for this population. Meta-analysis, including 5 randomized controlled trials, reported there was no significant difference between the group with perioperative continuation of long-term use aspirin and the group with discontinuation of the treatment, in terms of major bleeding requiring transfusion, reoperation for hemostasis, AMI and ischemic stroke.163
The POISE-2 trial, which was excluded from this meta-analysis, included a subgroup (n=4,382) of patients on long-term aspirin. The perioperative cardiovascular event rate was not significantly different between the aspirin continuation group and withholding group.169
Perioperative continuation or discontinuation of chronic maintenance antiplatelet therapy for coronary artery disease without stent placement should be decided upon after careful assessment of the risks of thrombosis and bleeding.
4.2 Advantages and Disadvantages of Withholding Antiplatelet Agents in Patients With a History of PCI
Withholding antiplatelet agents and subsequent cardiovascular events and bleeding risk was evaluated. The investigation in Part 2 CQ3 led us to conclude that continuation of aspirin does not lower the mortality risk, but does lower the perioperative AMI risk, without increasing bleeding risk; thus, the present guidelines suggest performing non-cardiac surgery under aspirin therapy continuation for patients on antiplatelet therapy for coronary stent placement if the risk of bleeding is not high (GRADE 2C).
4.2.1 Patients Receiving DAPT Therapy
a. Advantages and Disadvantages of Continuing DAPT
3 studies reported significant increases in bleeding events in surgeries performed with continuing DAPT,240,252,253 and 1 study reported an increased tendency in bleeding events.231 Meanwhile, no study has reported that continuing DAPT contributed to reducing perioperative cardiovascular events. A meta-analysis reported that clopidogrel use, mostly with concomitant use of aspirin (therefore, DAPT), had no benefit in lowering cardiovascular events, but increased bleeding events by 2.05-fold in comparison with no clopidogrel (most cases are on aspirin, therefore, Single-Antipaltelet Therapy [SAPT]).254 Thus, elective surgery should not be performed in post-PCI patients who still require DAPT therapy (Table 39).
Table 39. Recommendations and Levels of Evidence on Timing of Surgery for Post-Coronary Artery Stenting Patients
| |
COR |
LOE |
Elective non-cardiac surgery within 1–1.5 months of coronary artery stent placement
should not be performed due to the high rate of perioperative cardiovascular events* |
III
(Harm) |
B |
If surgery under continuation of DAPT therapy is not feasible, it should be considered to
postpone the surgery until appropriate DAPT duration after coronary artery stent placement
is completed |
IIa |
B |
*Practical considerations:
1. Cardiovascular risk decreases steadily with delay up to 1 year. Comprehensive assessment by a surgeon, an anesthesiologist, and a cardiologist for the best timing and shared decision making with the patient is recommended for time-sensitive surgeries
2. This recommendation is irrespective of whether the reason for the PCI is ACS or stable coronary artery disease. It is also irrespective of differences between devices used for PCI
COR, class of recommendation; LOE, level of evidence.
b. Duration of Withholding P2Y12 Receptor Antagonists
For patients who require DAPT but are switched to aspirin only for surgery, withholding prasugrel, clopidogrel, and ticagrelor starting 7, 5, and 3 days before surgery, respectively, is preferable.255–259 However, it is important to note that these recommendations are based on foreign data and differ from the information in Japanese package inserts regarding duration of withholding these medicines before surgery; prasugrel is used at different dosages and the indication of ticagrelor is limited in Japan.
c. Choice of Aspirin or P2Y12 Receptor Antagonist When Switching From DAPT to SAPT
Schoos et al.260 reported that of 490 patients who had discontinued DAPT (594 surgeries: both antiplatelet drugs discontinued for 254 cases, clopidogrel for 192, and aspirin for 148), thrombotic events occurred in 12. The event rate was 8/252=3.2% among both discontinued-both group, 4/192 (2.1%) in the aspirin monotherapy (clopidogrel-discontinued) group, and 0/148=0% in the clopidogrel monotherapy (aspirin-discontinued) group. The report did not provide any information on hemorrhagic events. Therefore, a definitive conclusion on which is the better medication to discontinue when switching DAPT patients to SAPT remains unknown.
4.2.2 Strategy on Whether to Withhold Antiplatelet Agents
Whether to withhold antiplatelet agents is determined by evaluating the risk of bleeding and the risk of stent thrombosis and the risk of thrombosis in areas other than the stented site (e.g. incidence of thrombosis and extent of MI when it occurs).
a. Bleeding Risk
Bleeding risk is evaluated by the patient’s predisposition to bleeding, bleeding-related organ injury, and difficulty of hemostasis in the event of bleeding. The bleeding risk in non-cardiac surgeries and procedures of the JCS 2020 Guideline are reproduced here as Table 40.178 As also mentioned in Chapter III.2 on anticoagulants, Table 28 is only a rule of thumb, and each patient should be analyzed individually through multidisciplinary discussion with the surgeon. The bleeding risk in neuraxial anesthesia and nerve block should also be discussed with an anesthesiologist (see Chapter III.3 Regional Anesthesia).
Table 40. Surgical Bleeding Risks With Non-Cardiac Surgery
| |
Low surgical bleeding risk |
Intermediate surgical bleeding risk |
High surgical bleeding risk |
General
surgery |
Hernioplasty, plastic surgery of incisional
hernias, cholecystectomy, appendectomy,
colectomy, gastric resection,
intestinal resection, breast surgery,
superficial surgery (e.g. abscess
incision, small dermatologic excisions) |
Hemorrhoidectomy, splenectomy,
gastrectomy, obesity surgery, rectal
resection, thyroidectomy |
Hepatic resection,
duodenocefalopancreasectomy |
Vascular
surgery |
Carotid endarterectomy, bypass or
endarterectomy of the lower extremity,
thoracic endovascular aortic repair
(TEVAR), endovascular aortic repair
(EVAR), limb amputations |
Open abdominal aorta surgery |
Open thoracic and thoracoabdominal
surgery |
Orthopedic
surgery |
Hand surgery, shoulder and knee
arthroscopy |
Prosthetic shoulder surgery, major
spine surgery, knee surgery (anterior
cruciate ligament, osteotomies), foot
surgery |
Major prosthetic surgery (hip or knee),
major traumatology (pelvis, long
bones), spin surgery |
Urology
surgery |
Flexible cystoscopy, ureteral
catheterization, ureteroscopy |
Prostate biopsy, orchiectomy,
circumcision |
Radical and partial nephrectomy,
percutaneous nephrostomy, percutaneous
lithotripsy, cystectomy, radical
prostatectomy, transurethral resection
of prostate (TURP), transurethral
resection of bladder tumor (TURBT),
penectomy, partial orchiectomy |
Thoracic
surgery |
Wedge resection, diagnostic
videothoracoscopy, chest wall
resection |
Lobectomy, pneumonectomy,
mediastinoscopy, sternotomy,
mediastinal mass excision |
Esophagectomy, pleuropneumonectomy,
decortication of lung |
Digestive
endoscopy |
Upper gastrointestinal endoscopy or
colonoscopy with or without biopsy,
echoendoscopy without biopsy, capsule
endoscopy, endoscopic retrograde
cholangiopancreatography (ERCP),
balloon endoscopy, endoscopic
marking, stent placement (pancreatic,
biliary, etc.), dilated papilla without
sphincterotomy |
Polypectomy, endoscopy with fine
needle aspiration biopsy for solid
lesions, stenosis dilatation (esophageal,
colorectal), argon plasma coagulation
treatment, percutaneous endoscopic
gastrostomy (PEG), endoscopic
treatment of gastroesophageal varices |
Dilatation in achalasia, mucosectomy,
submucosal resection, ampullectomy
of the ampulla of Vater, echography
with fine needle aspiration biopsy of
pancreatic cystic lesions |
| Others |
Dental interventions (teeth extraction,
periodontal surgery, incision of abscess,
implant positioning), cataract surgery,
bronchoscopy, etc. |
Bronchial biopsy, transbronchial
needle aspiration, etc. |
Spinal or epidural anesthesia, lumbar
diagnostic puncture, spinal and cranial
neurosurgery, eye posterior chamber
surgery, etc. |
Data from the supplementary profile of reference 261 as a reference for orthopedic surgery [bleeding risk by surgery according to the American Academy of Orthopedic Surgeons] to reclassify spinal surgery into high-risk, and proximal femoral fracture surgery in older adults as moderate-risk surgery. (Adapted as below from Table 25 of the JCS 2020 Guideline focused update on antithrombotic therapy in patients with coronary artery diseases.178)
b. Duration of Suspension of Antiplatelet Agents
The cutoff point for period of withholding medication that poses a cardiovascular event risk and the OR of cardiovascular events was: ≥3 days: OR 2.36,243 ≥5 days: OR 25.8,237 ≥6 days: OR 2.11,236 ≥8 days: OR 6.9,246 and ≥9 days: OR 3.38.248 These data suggest that the risk is lower with a shorter duration of withholding antiplatelet agents.
Thus, continuing antiplatelet therapy is desirable for surgeries with low bleeding risk that can be performed while the patient is on continued antiplatelet therapy, and withholding antiplatelet agents for the shortest duration possible is important for surgeries that require withholding.
As such, the present clinical practice guidelines recommends multidisciplinary discussion among the surgeon, anesthesiologist, and cardiologist and surgeries under continued aspirin therapy if they can be performed while the patient is on antiplatelet agents, and to perform the surgery with the minimum duration of withholding antiplatelet medications if withholding is needed (Class 1, Level of Evidence B, Table 41).
Table 41. Recommendations and Levels of Evidence on Perioperative Antiplatelet Therapy in Patients With Coronary Artery Disease
| |
COR |
LOE |
Perioperative continuation or discontinuation of chronic maintenance antiplatelet therapy
for coronary artery disease without stent placement should be decided upon careful
assessment of the risks of thrombosis and bleeding |
I |
C |
Newly introducing antiplatelet agents to reduce perioperative cardiovascular events should not
be performed |
III
(Harm) |
B |
Risks of thrombosis and bleeding in surgery performed on post-coronary artery stent
placement patients should be discussed by multidisciplinary team. Aspirin should be
continued for patients who are not at high risk of bleeding. Aspirin should be discontinued for
the shortest time possible in patients at high risk of bleeding |
I |
B |
There is no evidence supporting that heparin bridging while withholding antiplatelet agents
reduces cardiovascular events; rather this approach increases the risk of bleeding and
therefore, should not be performed |
III
(Harm) |
B |
COR, class of recommendation; LOE, level of evidence.
4.2.3 Heparin Bridging for Patients With a Coronary Artery Stent
Heparin bridging had been conventionally performed in Japan for patients with a coronary artery stent when it is needed to withhold antiplatelet agents for surgery. One study reported that preoperative heparin use was a predictor of stent thrombosis and that the OR of cardiovascular events was 3.22.242 Moreover, another study reported that heparin use did not contribute to reducing cardiovascular events but rather increased bleeding events slightly, although not to significant levels (OR 1.52, 95% CI 0.81–2.87).243 In addition, a retrospective study comparing 251 patients with coronary stent who were switched from antiplatelet to low-molecular-weight heparin and 264 patients who were not switched suggested the harms of heparin bridging, because it was associated with a 10.7 OR of cardiovascular events (P=0.001) and 1.86 OR of hemorrhagic events (P=0.017).262
Thus, the present clinical practice guidelines do not recommend the use of heparin bridging (Class III, harm, Level of evidence B, Table 41).
4.3 Patients With a History of CABG
Very few studies evaluating the perioperative antiplatelet status of patients with a history of CABG needing non-cardiac surgery have been reported.
Tokushige et al. compared 2,389 patients who needed surgery after PCI and 560 who needed surgery after CABG and found no difference in cardiovascular events between the 2 groups. However, they determined that hemorrhagic events were fewer in the patients who underwent surgery after CABG.263 Although 53% of patients in both groups were off antiplatelet agents at the time of surgery, 27% of the post-PCI patients were still on DAPT, compared with only 3.1% among the post-CABG patients, which may explain the lower incidence of hemorrhagic events among the patients with a history of CABG who underwent surgery.
There does not seem to be strong enough evidence to recommend either continuing or discontinuing antiplatelet agents for surgery. Thus, as of the present, whether to continue or discontinue antiplatelet therapy should be decided through multidisciplinary discussion.
5. Chronic Heart Failure
5.1 Epidemiology of Heart Failure in Non-Cardiac Surgery Patients
According to a database of 1.26 million non-cardiac surgery patients in the USA in 2009–2013, 0.61% had congestive heart failure (HF).264 According to another US database of 2.156 million non-cardiac surgery patients, HF was identified in 4.9% of patients undergoing non-cardiac surgery.265
5.2 Significance of Diagnosing Heart Failure Before Non-Cardiac Surgeries
In the perioperative period of non-cardiac surgery, HF increases the risk of cardiovascular and non-cardiovascular complications, and death.266 Diagnosing HF is important for preoperative risk stratification, so careful preoperative history taking and physical examination is important.
5.3 Perioperative Risk in Heart Failure Patients
The perioperative mortality in patients with HF, regardless of ischemic or non-ischemic origin, is significantly higher than that of patients with coronary artery disease267 (Figure 7). Heart failure is associated with higher risk of cardiovascular events, as well as non-cardiovascular events; including acute kidney injury, bleeding, infectious disease, reoperation, pulmonary embolism, delirium, stroke, anastomotic leak in gastrointestinal surgeries, etc.267–269 (Table 42).
Table 42. Cardiac and Non-Cardiac Perioperative Risk Among Patients With Heart Failure Undergoing Non-Cardiac Surgeries
| |
With heart
failure |
Without heart
failure |
Adjusted odds ratio |
Reference |
| Cardiac events |
| 30-day cardiac arrest, requiring CPR |
3.47% |
0.31% |
2.00 (1.74–2.30) |
264 |
| 30-day myocardial infarction |
2.61% |
0.43% |
1.36 (1.17–1.58) |
264 |
| In-hospital cardiac arrest |
1.5% |
0.22% |
2.22 (2.10–2.35) |
265 |
| In-hospital acute myocardial infarction |
4.2% |
0.33% |
3.55 (3.41–3.69) |
265 |
| Mortality |
| 30-day mortality |
12.37% |
0.99% |
1.96 (1.80–2.13) |
264 |
| 90-day mortality |
5.49% |
1.22% |
1.67 (1.57–1.76) |
270 |
| 90-day mortality after ambulatory surgery |
2.00% |
0.39% |
1.95 (1.69–2.44) |
271 |
| In-hospital mortality |
4.8% |
0.78% |
2.15 (2.09–2.22) |
265 |
| Complications |
| 30-day acute renal failure |
3.09% |
0.32% |
1.60 (1.38–1.85) |
264 |
| 30-day bleeding transfusion |
25.27% |
7.45% |
1.39 (1.31–1.47) |
264 |
| 30-day septic shock |
4.27% |
0.59% |
1.37 (1.21–1.55) |
264 |
| 30-day pneumonia |
5.89% |
1.11% |
1.32 (1.19–1.46) |
264 |
| 30-day urinary tract infection |
4.19% |
1.59% |
1.25 (1.11–1.41) |
264 |
| 30-day CVA with neurological deficit |
1.02% |
0.23% |
1.23 (0.97–1.56) |
264 |
| 30-day return to operating room |
10.02% |
3.22% |
1.23 (1.14–1.34) |
264 |
| In-hospital pulmonary embolism |
2.3% |
0.68% |
1.62 (1.56–1.69) |
265 |
| In-hospital acute ischemic stroke |
1.9% |
0.50% |
1.39 (1.33–1.45) |
265 |
| Anastomotic leak after gastrointestinal surgery |
NA |
NA |
31.54 (2.61–381.40) |
272 |
CPR, cardiopulmonary resuscitation; CVA, cerebrovascular accident.
(Adapted from references Turrentine FE, et al. 2016,264 Smilowitz NR, et al. 2021,265 Lerman BJ, et al. 2019,270 Lerman BJ, et al. 2019,271 and Turrentine FE, et al. 2015.272)
5.4 Left Ventricular Ejection Fraction and Perioperative Risk
Heart failure with reduced ejection fraction (HFrEF) with left ventricular ejection fraction (LVEF) <40% has higher risk of perioperative cardiovascular complications (death, acute heart failure, AMI)270 (Figure 8). The lower the EF, the higher the perioperative mortality (Figure 8). Furthermore, LVEF <30% is an independent factor of not only perioperative death but also the postoperative long-term mortality rate.273,274 Heart failure with preserved ejection fraction (HFpEF) is also associated with significantly increased mortality risk270 and postoperative cardiovascular complications.275–277
5.5 Preoperative Management of Heart Failure Patients
5.5.1 Importance of Preoperative Optimization of Hemodynamics
Patients with asymptomatic HF stabilized by medical therapy and without any physical findings suggesting HF are reported to have comparable perioperative mortality rates as patients without HF.278 If signs of HF are present, hemodynamics should be optimized preoperatively.4
5.5.2 Initiation of Chronic Maintenance Therapy
Chronic maintenance therapy with angiotensin-converting enzyme (ACE) inhibitors/angiotensin-receptor blockers (ARBs), β-blockers, aldosterone receptor antagonists, and other medications in line with clinical practice guidelines is important for improving the prognosis of HFrEF patients (Table 43).279 Elective non-cardiac surgery for a patient newly preoperatively diagnosed with HFrEF should be performed after introducing optimal medical therapy for chronic HF to attempt stabilization if delaying the surgery is not harmful. Although there is no supporting evidence available, the ESC Guidelines 2014 recommend postponing elective procedures for ≥3 months to introduce these therapies first and to wait for an improvement in heart function.4 However, it is not recommended to forcibly introduce or increase the dosage of medical treatment preoperatively in the case of an emergency or urgent procedure that does not allow time for optimizing treatment with ACE inhibitors/ARBs or β-blockers before surgery. If surgery must be performed without stabilizing the HF symptoms first, it is important to prevent worsening of HF by careful monitoring for signs of congestion and tissue hypoperfusion in the perioperative period, and to introduce medical treatment for HF postoperatively (see Chapter 15 for more on ACE inhibitors, ABR and β-blockers as maintenance therapy for chronic heart failure).
Table 43. Recommendations and Levels of Evidence Related to the Management of Chronic Heart Failure Patients
| |
COR |
LOE |
Sharing the knowledge that heart failure increases both perioperative cardiovascular
and non-cardiovascular event risk with the surgeon, anesthesiologist and the patient is
recommended |
I |
C |
It is recommended to introduce guideline-directed therapy of heart failure with reduced
EF newly diagnosed in the preoperative evaluation for intermediate- or high-risk
surgery to allow for dose titration and recovery of cardiac function before elective procedures
if that can be delayed for ≥3 months |
I |
C |
ACE inhibitors/ARBs as maintenance therapy for HFrEF discontinued preoperatively should
be restarted as early as possible postoperatively |
I |
C |
Beta-blockers administered as chronic maintenance therapy for HFrEF should not be
discontinued during the perioperative period |
III
(Harm) |
B |
ACE angiotensin-converting enzyme; ARB, angiotensin receptor blocker; COR, class of recommendation; HFrEF, heart failure with reduced ejection fraction; LOE, level of evidence.
5.6 Causes of Postoperative Acute Heart Failure (Table 44)
Table 44. Causes of Postoperative Acute Heart Failure
| • Excessive fluid infusion |
| • Myocardial injury (MINS: Myocardial Injury After Surgery) |
| • Myocardial ischemia (type 1 and 2 Acute Myocardial Infarction, Stent thrombosis) |
| • Stress related cardiomyopathy |
| • Atrial fibrillation |
| • Refilling of fluid extravasated by surgical invasion into blood vessel |
The patient is placed in the Trendelenburg position (with the head lowered) for laparoscopic lower abdominal surgeries, thus increasing preload.58,280,281 This position can cause HF in patients with compromised cardiac function.281 Because vasodilation can be induced by anesthesia, intraoperative fluid is often administered to maintain blood pressure and systemic perfusion. When peripheral vascular tone normalizes after discontinuing anesthesia, venous return to the heart increases, thereby causing postoperative HF in some patients.36
Residual irrigated fluid in the surgical space is absorbed later and can provoke acute HF in some cases of transurethral resection of the prostate and hysteroscopic hysterectomy.36 In others, water extravasated to extravascular tissues due to increased permeability of the local vessels during surgery can refill the vessels postoperatively, resulting in acute HF several days later.36 In particular, patients with HFpEF have increased stiffness of the left ventricle, which makes them more prone to pulmonary edema due to the excess fluid and thus require appropriate perioperative monitoring, caution with fluid replacement volumes, and afterload control.
One of the perioperative risks is myocardial injury after non-cardiac surgery (MINS),282 which can be a cause of HF. During the first few postoperative days, patients may be less likely to notice symptoms such as chest pain due to the effects of surgery or anesthesia, thus surveillance with troponin or ECG is necessary for the diagnosis (see Chapter V. Postoperative Surveillance of Myocardial Injury).282 Stress-induced cardiomyopathy associated with surgery should also be included in the differential diagnosis.283 Timely diagnosis of HF by close monitoring of its signs and symptoms is important.
6. Hypertrophic Cardiomyopathy
6.1 Epidemiology
The prevalence of hypertrophic cardiomyopathy (HCM) in Japan was 17.3/100,000 (≈1/5,000) according to a 1998 survey,284 in contrast to the ≈1/500 prevalence in foreign reports. It is considered a disease that can affect individuals of all age groups.284–286 Another Japanese study reported that the prevalence of HCM was 30/66,000 among patients undergoing non-cardiac surgery.285
6.2 Utility of Preoperative Echocardiography
Obstructive HCM is reportedly associated with a higher perioperative cardiac event risk than non-obstructive HCM,287 especially when the left ventricular outflow tract pressure gradient is ≥30 mmHg.288 In contrast, another study showed that outflow obstruction was not associated with perioperative adverse cardiac event risk.289 However, these studies involved only small sample sizes, suggesting that the findings should be interpreted with caution. The echocardiographic findings as perioperative risk factors are shown in Table 45.
Table 45. Echocardiographic Red Flags for Hypertrophic Cardiomyopathy Patients Undergoing Non-Cardiac Surgery Due to Increased Risk for Perioperative Complications
| • Left ventricular outflow tract pressure gradient ≥30 mmHg at rest, ≥50 mmHg with provocation (Valsalva maneuver, etc.) |
| • Moderate to severe mitral valve regurgitation with mitral valve systolic anterior motion |
| • Left ventricular systolic dysfunction (LVEF <50%) |
| • Decreased left ventricular diastolic performance with restrictive mitral inflow pattern |
(Adapted from Hensley N, et al. 2015.290)
6.3 Risk of Perioperative Cerebrovascular and Cardiovascular Complications
The perioperative risks of HCM are shown in Table 46. One of the studies, a 2006 cohort study of 227 patients, reported a high incidence of perioperative adverse events (death 6.7% and MI 2.2%).292 Another cohort study, which compared 92 HCM to non-HCM patients, found a 22% composite endpoint for 30-day postoperative death, brain stroke, AMI, and acute HF (vs. 12% in patients without HCM), which consisted predominantly of acute HF.288 A Japanese retrospective single-center analysis of 72 HCM patients who underwent non-cardiac surgery reported that surgery-related deaths and cardiovascular adverse events during hospital stay (death and exacerbation of HF) occurred in only 3% of patients, and the post-discharge mortality rate over an average 3.8-year follow-up was also low (4.2%), concluding that non-cardiac surgery in patients with HCM can be performed safely.287
Table 46. Perioperative Risks of Hypertrophic Cardiomyopathy
| Event |
Incidence |
| Death |
0–6.7% |
| Acute heart failure |
1–16% |
| Myocardial infarction |
0–2.9% |
| Atrial fibrillation |
4% |
6.4 Preoperative Preparation and Perioperative Management in Non-Cardiac Surgery
The severity of disease should be ascertained based on symptoms, cardiac function, and family history in the preoperative evaluation. It is important to carefully survey new development or worsening of left ventricular outflow tract (LVOT) obstruction and hypotension that result from decreased preload and peripheral vascular resistance caused by induction of anesthesia and positive-pressure ventilation. Therefore, β-blockers, non-dihydropyridine calcium-channel blockers and drugs such as cibenzoline and disopyramide used to prevent LVOT obstruction should be continued perioperatively. Furthermore, it is necessary to optimize the circulating blood volume by maintaining sufficient preload. A pulmonary artery catheter may need to be placed for stringent hemodynamic monitoring in patients with significantly decreased left ventricular diastolic performance.
Although extremely rare, acquired von Willebrand disease should be considered as a possible complication of HCM with left ventricular obstruction when patients have a history of bleeding diathesis.295 This disease is considered to have the same underlying mechanism as Heyde syndrome, which is a characteristic manifestation of severe AS. The etiology is believed to involve increased circulatory shear forces due to LVOT obstruction and subsequent cleavage of von Willebrand multimers.
7. Pulmonary Hypertension
Patients with pulmonary hypertension have lower reserve for hemodynamic changes than healthy individuals and thus, are more prone to perioperative right-sided heart failure or pulmonary hypertensive crises, which exhibit sustained hypoxemia and shock.296,297
It is important to prevent a pulmonary hypertensive crisis in the perioperative period. Once pronounced hypotension develops, reduced right coronary artery perfusion pressure causes ischemia of the right ventricle with resultant dysfunction, tricuspid valve regurgitation and septal compression towards the left ventricle. This leads to irreversible hemodynamic collapse, which is the mechanism of a pulmonary hypertensive crisis. Therefore, prompt treatment for hypoxemia and low systemic blood pressure, which trigger the disruption, are important in the perioperative period.
7.1 Complications and Risk Factors
7.1.1 Perioperative Complications
The perioperative mortality rate of patients with pulmonary hypertension is approximately 1–18%, and the complication rate ranges from 6.1% to 42%.298–310 The risks are 3–4-fold higher than for those without pulmonary hypertension.298,304–306 Sustained hypoxemia (0.5–28%),302,308 congestive heart failure (1.5–11%),302,310 right-sided heart failure (3.5–10.7%),300,310 arrhythmia (0.5–12%),301,302 hemodynamic instability/hypotension (1.1–10.7%),298,300 and sepsis (0.9–10.4%)306,308 have been reported as perioperative complications.
7.1.2 Patient Risk Factors (Table 47)
Table 47. Risk Factors for Perioperative Death and Complication Among Patients With Pulmonary Hypertension
| Patient factors |
Surgical factors |
Age ≥75 years
NYHA class ≥II
ASA class ≥III
6MWD <300 m
Right ventricular hypertrophy
RVSP/SBP ≥0.66
RVSP ≥70 mmHg
RAP >7 mmHg
Coronary artery disease
History of pulmonary embolism
Chronic renal insufficiency |
Emergency surgery
Intermediate-/high-risk surgery
Duration of anesthesia >3 h
Intraoperative use of vasopressors
Perioperative management without
inhalation of nitric oxide |
NYHA, New York Heart Association; ASA class, American Society of Anesthesiologist; 6MWD, 6-minute walking distance; RAP, right atrial pressure; RVSP, right ventricular systolic pressure; SBP, systolic blood pressure.
(Adapted from Minai OA, et al. 2013,296 Pilkington SA, et al. 2015,313 and Olsson KM, et al. 2018.314)
Reported patient risk factors associated with perioperative death or complications are age ≥75 years (odds ratio [OR] 5.0–6.3),305 NYHA class ≥III (OR 2.9),302 6-minute walking distance ≤399 m (OR 2.2),310 American Society of Anesthesiologists (ASA)311 physical status ≥Class III (OR 4.2),306 right ventricular hypertrophy on ECG,302 systolic pulmonary artery pressure >70 mmHg on echocardiography,304 and right ventricular systolic pressure/systemic systolic blood pressure ≥0.66.302,306
7.1.3 Surgical Risk Factors (Table 47)
Perioperative complications in patients with pulmonary hypertension undergoing low-risk surgery4 were 16%, but reached 47% in intermediate- and high-risk surgery in one report.302 Of note, it was as high as 61.5% in thoracic surgery.302 Surgery lasting ≥3 h300,302,308 and emergency procedures301,304,305,307,310 are also reported to be risk factors. Although laparoscopic surgeries are less invasive, increased airway pressure and decreased pulmonary compliance due to pneumoperitoneum, and hypercapnia-induced pulmonary vasoconstriction can also worsen pulmonary hypertension, and thus require due caution.312
7.2 Preoperative Considerations
7.2.1 Main Precautions for Surgical Patients With Pulmonary Hypertension
(1) Risk stratification, reconsideration of indication of the surgery and perioperative strategy
The risk factors shown in Table 47 should be assessed preoperatively for patients with pulmonary hypertension and the necessity of surgery and methods of perioperative management should be evaluated in discussion with a multidisciplinary team including the anesthesiologist, surgeon, and pulmonary hypertension specialist4
(2) Preoperative optimization of the treatment for pulmonary hypertension
Patients are prone to hemodynamic destabilization in the perioperative period. Thus, preoperative treatment for pulmonary hypertension should be optimized as much as possible. Pulmonary vasodilators that the patient is already taking should not be discontinued, even on the day of surgery59
(3) Referral to sepcialized institutions for surgery
Postoperative complication rates in facilities specializing in pulmonary hypertension are approximately 0.4-fold that in non-specialized facilities.4 In the absence of a specialist in pulmonary hypertension management, transferring the patient to a specialized institution should be considered
7.3 Intraoperative Management
7.3.1 Influence of Anesthetic Technique and Agents on Pulmonary Hypertension
To date, there have been no randomized controlled trials evaluating anesthetic techniques and agents used for patients with pulmonary hypertension. General anesthesia requires induction of anesthesia, intubation maneuvers, and positive-pressure ventilation with a mechanical ventilator at the start of surgery; thus, it has a higher rate of hemodynamic destabilization than regional anesthesia.315 Many intravenously administered anesthetics, such as propofol and thiopental, and volatile inhaled anesthetics have cardiosuppressant effects and induce hypotension upon induction. However, neuraxial anesthesia (spinal and epidural anesthesia) also have a risk of hypotension by causing systemic vasodilation. Both general and neuraxial anesthesia requires close monitoring of hemodynamics. Nitrous oxide316 and ketamine317 should be used carefully because they have been reported to increase pulmonary vascular resistance. Opioids have relatively minor effects on hemodynamics; thus, their combination with general anesthesia can reduce the total dose of anesthetic agent needed and can help to achieve a stable induction and maintenance of anesthesia.313
7.3.2 Respiratory Management With Artificial Ventilation
The following cautions and changes in hemodynamics at the time of starting artificial respiration are worth noting.
(1) Hypoxemia and hypercapnia induce pulmonary vasoconstriction and should be avoided as much as possible296
(2) Compared with spontaneous breathing, which uses negative pressure, positive-pressure ventilation with a mechanical ventilator increases pulmonary vascular resistance. During the inspiratory phase with positive pressure, as the intrathoracic pressure increases, the afterload of the right ventricle increases and preload decreases, and consequently cardiac output is lowered
(3) Pulmonary vascular resistance is lowest when the lung capacity is at end-expiratory level. At maximum inspiratory level, the alveolar capillaries may collapse, which increases pulmonary vascular resistance.318,319 To avoid hyperinflation of the lungs, the inspiratory plateau pressure should be maintained at ≤30 cmH2O296 and tidal volume should be 6–8 mL/kg296,315
(4) Lung collapse may increase pulmonary vascular resistance due to atelectasis. To avoid lung collapse, a certain level of positive end-expiratory pressure (PEEP) is needed
(5) Excessive PEEP has the risk of enhancing all effects described in (2). Thus, caution is warranted when applying PEEP
7.4 Postoperative Management
Monitoring of respiratory status and hemodynamic status should be continued postoperatively while correcting factors that worsen pulmonary hypertension and performing fluid management properly, with additional use of vasopressors, inotropic agents, and pulmonary vasodilators as needed (Figure 9).
7.4.1 Optimization of Right Ventricular Preload
Right atrial pressure is an important index for fluid management. Fluid administration is a common treatment for patients with hypotension due to hypovolemia, but it does not effectively increase cardiac output in right-sided heart failure patients with already high right ventricular filling pressure and can even lower it due to worsening tricuspid regurgitation and left ventricular compression by the septum in some patients.320 Such patients benefit from removing fluid with diuretics or dialysis.314 Optimal right atrial pressure depends on right cardiac function and on whether the development of the pulmonary hypertension is acute or chronic, but it usually ranges between 8 and 12 mmHg.319
7.4.2 Inotropic Agents for Right Ventricular Systolic Dysfunction
Inotropic agents are necessary for patients with reduced cardiac output, and the phosphodiesterase III inhibitor milrinone and the β1/β2 agonist dobutamine are used for this purpose.318 Both drugs increase cardiac output and reduce pulmonary vascular resistance, and also improve the pulmonary vascular resistance/systemic vascular resistance ratio.314 Although there are no randomized studies to date, good clinical outcomes have been achieved using these drugs for pulmonary hypertension.318,319 A small prospective randomized trial reported that the calcium sensitizer levosimendan was more effective in pediatric patients with pulmonary hypertension undergoing cardiac surgeries than dobutamine,321 but that drug is not commercially available in Japan.
7.4.3 Pulmonary Vasodilators
The use of pulmonary vasodilators should be considered when the above treatments are not effective. Although strong pulmonary vasodilator effects can be anticipated with continuous intravenous infusion of epoprostenol322 or treprostinil323 or systemic administration of prostacyclins by subcutaneous injection, these drugs also have the risk of lowering systemic blood pressure. Inhaled therapies are beneficial because they act directly on the pulmonary vessels supplying ventilated regions, have fewer side effects that decrease systemic blood pressure, and also improve ventilation–perfusion imbalance,324 making them useful in urgent perioperative situations. Epoprostenol,325 iloprost,326 and nitric oxide327 have been reportedly used as inhaled therapies. Phosphodiesterase V inhibitors328,329 and endothelin antagonists330,331 are also viable options for patients who are able to take oral medication.
8. Valvular Heart Disease
8.1 Epidemiology
Valvular heart disease has a relatively high prevalence, affecting about 2–3% of the general population and 10–20% of adults aged ≥75 years.332 The compensatory mechanisms of the heart often conceal symptoms of valvular heart disease until the disease progresses significantly, such that even severe valvular heart disease goes undiagnosed in some cases.333
8.2 Importance of Preoperative Assessment of Medical History and Physical Findings
The patient should be interviewed for history of valvular heart disease and symptoms of heart failure, as moderate and severe valvular heart disease is found in 11.3% and 2.7%, respectively, of patients with suspected heart failure.334 Heart murmurs are an important physical finding, as moderate or severe valvular heart disease is found in 18% patients with heart murmurs on auscultation.335
8.3 Perioperative Management
Hemodynamic monitoring and anesthetic technique should be selected according to the type, severity, and cardiac function in valvular heart disease patients. Intraoperative pulmonary artery catheterization and transesophageal echocardiography should be considered, especially for patients with severe valvular disease undergoing a high-risk non-cardiac surgery. It is important to avoid tachycardia in patients with aortic or mitral valve stenosis, to use vasodilators cautiously, and to maintain adequate preload. Conversely, bradycardia should be avoided in patients with aortic or mitral regurgitation, and optimization of afterload with vasodilators should be maintained.
8.4 Approaches in Deciding Whether to Perform Invasive Treatment for Valvular Heart Disease Before Non-Cardiac Surgery
As long as valvular disease is asymptomatic, the non-cardiac surgery can be performed safely without preoperative valve intervention in the contemporary era with recent advances in anesthetic techniques.336–338 However, valve replacement is a Class I recommendation for symptomatic severe valvular heart disease in the non-perioperative setting.18 There are also some time-sensitive non-cardiac surgeries that cannot be delayed excessively.
The following are the options for non-cardiac surgery for patients complicated with severe valvular heart disease: (1) perform the intervention for the valvular heart disease first, (2) perform the non-cardiac surgery first, (3) perform only one or the other, or (4) perform neither. When considering the therapeutic strategy, the followings need to be assessed and weighed: (1) urgency of the intervention for the valvular heart disease (symptoms, stability of hemodynamics, cardiac function), (2) risks of the non-cardiac surgery, (3) risks of the intervention for valvular heart disease, (4) urgency of the non-cardiac surgery (effect of postponing the non-cardiac surgery on the primary disease), and (5) risks of bleeding or thrombosis associated with the perioperative continuation or discontinuation of antithrombotic agents following valve replacement. It is, therefore, important for a multidisciplinary team of surgeon, anesthesiologist, and cardiologist to assess the risks and benefits of each option comprehensively and include the patient’s values and preferences for shared decision-making (Figure 1).
8.5 Various Valvular Heart Diseases
8.5.1 Aortic Stenosis (Table 48)
Table 48. Recommendations and Levels of Evidence on Valvular Heart Disease in the Perioperative Period
| |
COR |
LOE |
Consider non-cardiac surgery under appropriate hemodynamic monitoring for
asymptomatic severe AS patients |
IIa |
B |
Consider non-cardiac surgery under appropriate hemodynamic monitoring for
patients with asymptomatic severe AR or MR with maintained left ventricular function |
IIa |
C |
See CQ4 for Practical Considerations. Also see 8.4 Approaches in Deciding Whether to Perform Invasive Treatment for Valvular Heart Disease Before Non-Cardiac Surgery. AS, aortic stenosis; COR, class of recommendation; LOE, level of evidence; MR, mitral regurgitation.
a. Perioperative Cardiovascular Event Risk in Non-Cardiac Surgery in Patients With Aortic Stenosis
A 2017 meta-analysis reported severe aortic stenosis (AS) was not significantly associated with increased mortality (risk ratio [RR] 1.49, 95% confidence interval [CI] 0.85–2.61), myocardial infarction (MI) (RR 1.65, 0.66–4.13), heart failure (RR 1.42, 0.47–4.24), or stroke (RR 0.44, 0.14–1.34).339
Another meta-analysis also reported that asymptomatic severe AS was not significantly associated with increased mortality (OR 0.96, 95% CI 0.40–2.30), MI (OR 1.68, 0.76–3.70), or heart failure (OR 1.04, 0.49–2.22).340 That study concluded that prophylactic aortic valve replacement (AVR) or transcatheter aortic valve implantation (TAVI) could be omitted without increasing the perioperative risks for asymptomatic severe AS.
The studies of Raymer et al., 1998,341 Calleja et al., 2010,336 Agarwal et al., 2013,337 Tashiro et al., 2014,338 and MacIntyre et al., 2018342 were analyzed in a systematic review and meta-analysis performed to answer the key question 4.1 in Part 2 CQ4, and the result was consistent with the above findings.
b. Benefits and Harms of Preoperative AVR for Non-Cardiac Surgery
Part 2 of the present clinical practice guidelines examines the benefits of preoperative AVR for non-cardiac surgery in a systematic review of two retrospective observational studies (Taniguchi et al., 2020,343 and Luis et al., 2020.344 See Part 2 CQ4.2 for details).
c. Evidence Related to TAVI and Percutaneous Balloon Aortic Valvuloplasty
In recent years, transcatheter therapies for AS have rapidly increased in popularity.
The outcomes of 60 cases of preoperative balloon aortic valvuloplasty (BAV) and 58 cases of preoperative TAVI have been reported.345 Congestive heart failure complicated 33% of patients who underwent BAV and 74% of patients who underwent TAVI. The non-cardiac surgery resulted in a 1.7% 30-day mortality rate (BAV: 1.7%, TAVI: 1.7%), 0% incidence of MI, 0.8% brain infarction (BAV: 0%, TAVI: 1.7%) and 3.4% incidence of heart failure (BAV: 3.3%, TAVI: 3.4%). Non-cardiac surgery was performed on median 11 days and 28 days after BAV and TAVI, respectively.
Okuno et al. performed TAVI before non-cardiac surgery in 7 patients and reported that the non-cardiac surgery was performed on average 37 days after TAVI without any in-hospital deaths or adverse cardiac events.346
There have been no randomized controlled trials that have compared preoperative TAVI or BAV with no cardiac intervention before non-cardiac surgery; thus, it is difficult to make one-size-fits-all recommendation related to preoperative prophylactic BAV and TAVI. However, the treatments can be options for symptomatic hemodynamically unstable AS before non-cardiac surgery.
d. Other Relevant Clinical Practice Guidelines
(i) JCS/JATS/JSVS/JSCS 2020 Guideline on the Management of Valvular Heart Disease18
There is no recommendation, but the following is stated.
• Emergency non-cardiac surgery in patients with severe AS should be performed by a cardiovascular anesthesiologist under hemodynamic or transesophageal echocardiographic monitoring
• The indications for elective non-cardiac surgery should be determined by the presence of symptoms of AS and the type of surgery
• In symptomatic severe AS, surgical AVR (SAVR)/TAVI is recommended prior to non-cardiac surgery
• If SAVR/TAVI is high risk because of anatomic characteristics and patient background, percutaneous BAV should be considered
• In asymptomatic severe AS, it is possible to safely perform elective non-cardiac surgery. For high-risk non-cardiac surgery with significant volume overload, SAVR/TAVI prior to non-cardiac surgery may be considered in the absence of symptoms
(ii) 2020 ACC/AHA Guideline for the Management of Patients With Valvular Heart Disease347
It is reasonable to perform elective procedures for asymptomatic moderate to severe AS with normal systolic function (Class IIa, B).
(iii) ESC 2017 ESC/EACTS Guidelines for the Management of Valvular Heart Disease199
• In symptomatic patients, AVR should be considered before non-cardiac surgery
• In asymptomatic patients, elective non-cardiac surgery can be performed safely, albeit with a risk of worsening heart failure. If non-cardiac surgery implies large volume shifts, aortic valve replacement should be considered first
(iv) Part 2 CQ4 of the Present Clinical Practice Guidelines
• For patients with severe AS who undergo non-cardiac surgeries, it is suggested that non-cardiac surgery is performed without preoperative prophylactic AVR, either SAVR or TAVI if it is asymptomatic (GRADE 2D)
• Symptomatic cases should be assessed on an individual basis
(v) Recommendations in Part 1
Recommendations made in Part 2 should be followed. That is, non-cardiac surgery should be considered without performing prophylactic SAVR/TAVI for asymptomatic AS (Class IIa, B). If severe AS is symptomatic, AVR is a Class I indication for the non-perioperative period, but assessment should be made on the individual level for the perioperative period.
8.5.2 Mitral Stenosis
Mitral stenosis (MS) is a risk factor in non-cardiac surgery.348,349 For patients with severe MS with symptoms or pulmonary hypertension (systolic pulmonary blood pressure ≥50 mmHg) undergoing intermediate- to high-risk non-cardiac surgery, preoperative transcatheter mitral valve commissurotomy or mitral valve replacement may be considered to lower the perioperative risk. As mentioned earlier, risk assessment is important in determining whether to treat the valvular heart disease first. Appropriate intraoperative and postoperative hemodynamic monitoring is important if performing the non-cardiac surgery first.
8.5.3 Aortic Regurgitation (Table 48)
Aortic regurgitation (AR) is associated with increased perioperative risk for non-cardiac surgeries.348,349 According to 1 retrospective study including older patients (mean age 75 years) undergoing mainly intermediate-risk surgeries (68.9%), severe AR patients had a higher perioperative mortality rate than non-AR patients (9% vs. 0%), and also had more complications (prolonged ventilator use [13.8% vs. 3.0%], pulmonary edema [4.8% vs. 0.5%], severe arrhythmia [3.0% vs. 0.6%]).350 It is unknown whether these complications or the mortality rate will improve by performing valve replacement preoperatively. Nonetheless, severe AR with symptoms or left cardiac dysfunction is a Class I indication for valve replacement in non-perioperative settings,18 and the items discussed in section 8.4 “Approaches in Deciding Whether to Perform Invasive Treatment for Valvular Heart Disease Before Non-Cardiac Surgery” should be assessed to determine whether or not to perform AVR before non-cardiac surgery.
8.5.4 Mitral Regurgitation (Table 48)
Severe mitral regurgitation (MR) is associated with increased perioperative risk for non-cardiac surgery.348,349 A retrospective observational study showed that moderate and severe MR was associated with higher incidence of MI (2.7% vs. 0.9%, P=0.02) and heart failure (17.5% vs. 12.8%, P<0.001) than no MR, but there was not significant increase in in-hospital mortality (1.7% vs. 1.1%, P=0.43).351 Of note, only 21% of the MR group had organic MR. In cases of severe organic MR which is an indication for surgery, the items discussed in section 8.4 should be assessed to determine whether to perform surgery for the valvular heart disease before the non-cardiac surgery.
9. Hypertension
9.1 Perioperative Risks of Hypertension
Hypertension is widely known as a risk for perioperative cardiovascular events.352–354 A systematic review and meta-analysis of 30 observational studies reported that hypertension was associated with increased risk of cardiovascular complications (odds ratio 1.35).355
9.2 Preoperative Evaluation
When untreated hypertension is detected in the preoperative evaluation, basic medical history, physical findings, blood and urine tests, and ECG should be used to screen for hypertensive organ injury or complications, and detailed examination of secondary hypertension should be performed if indicated.354
Grade 1 or 2 hypertension354 (systolic blood pressure <180 mmHg, diastolic blood pressure <110 mmHg) is not grounds for postponing surgery356,357 (Table 49). However, in cases of Grade 3 hypertension, the benefits and disadvantages of postponing surgery until blood pressure is stabilized by medication should be assessed on an individual basis to determine the timing of the surgery.
Table 49. Recommendations and Levels of Evidence on Perioperative Blood Pressure Management
| |
COR |
LOE |
Not postponing surgery can be considered for Grade 1 or 2 hypertension (systolic blood
pressure <180 mmHg, diastolic blood pressure <110 mmHg) |
IIb |
B |
Consider maintaining blood pressure at 70–100% of baseline in the perioperative
period |
IIa |
B |
COR, class of recommendation; LOE, level of evidence.
9.3 Perioperative Management
Patients with hypertension are at higher risk of intraoperative changes in blood pressure, which can provoke myocardial ischemia.358 Intraoperative hypotension increases the risk of perioperative ischemic stroke, myocardial ischemia, and acute kidney injury as well as death.359–365 It is recommended to manage blood pressure adequately to maintain it at 70–100% of baseline in the perioperative period and to avoid excessive fluctuations (Table 49). A postoperative increase in blood pressure frequently occurs due to anxiety and pain after awakening from anesthesia; however, it will improve by treating these issues.
10. Pulmonary Artery Catheterization
A large randomized controlled trial (RCT)366 assessed the perioperative benefits of pulmonary artery catheterization in 1,994 patients with ASA risk score III or IV and aged ≥60 years undergoing emergency or elective non-cardiac surgery (abdominal, thoracic, vascular surgeries, or proximal femoral fracture). Patients were randomly assigned to preoperative pulmonary artery catheterization and the study found no improvement in the in-hospital mortality rate but rather an increase in pulmonary thromboembolism (P=0.004).
A subsequent meta-analysis of 13 RCTs (n=5,686) that tested the benefits of pulmonary artery catheterization in patients in the intensive care unit for various reasons such as surgery, acute heart failure, ARDS or sepsis also did not find improvement in mortality rate or shortened hospital stay by performing pulmonary artery catheterization.367
Based on these findings, routine pulmonary artery catheterization is not recommended for non-cardiac surgery, even in high-risk patients (Table 50).
Table 50. Recommendation and Level of Evidence on Pulmonary Artery Catheterization
| |
COR |
LOE |
| Routine use of pulmonary artery catheterization for non-cardiac surgery is not recommended |
III (No
benefit) |
A |
COR, class of recommendation; LOE, level of evidence.
In the present clinical practice guidelines, the recommendation is limited to considering the use of pulmonary artery catheterization for patients with diseases that may have a significant effect on perioperative hemodynamics, such as severe valvular heart disease or severe left ventricular systolic dysfunction.
11. Arrhythmia
The 3 following factors are possible components of new arrhythmia that may occur in the perioperative period.
(1) Latent arrhythmias discovered incidentally in the perioperative period
(2) Arrhythmia triggered by the use of medicine, autonomic nerve activity, electrolyte imbalance or new cardiomyopathy (myocardial ischemia, stress cardiomyopathy, etc) in the perioperative period
(3) Surgical maneuvers or direct contact with the heart during thoracic surgery (lungs, esophagus, etc)
11.1 Perioperative Risks in Patients With Atrial Fibrillation
11.1.1 Epidemiology
The 30-day postoperative mortality rate of non-cardiac surgery was 6.4% in patients with atrial fibrillation (AF), which is 1.69-fold higher than in patients with ischemic heart disease.267 A subanalysis of the VISION trial involving patients aged ≥45 years found a higher cardiovascular event risk in patients complicated with AF than in those without (26.6% vs. 9.0%; hazard ratio 1.58).
11.1.2 Perioperative Management of Chronic Maintenance Treatment for AF
Whether to continue or suspend anticoagulants, β-blockers or antiarrhythmic drugs in the perioperative period should be discussed with the cardiologist. Extra caution is needed in decision-making for anticoagulants. Anticoagulant therapy cannot be discontinued routinely post-ablation, even if sinus rhythm is maintained.171
11.2 New Onset of Perioperative/Postoperative AF (POAF)
11.2.1 Definition and Frequency
There are several definitions of POAF, such as new AF that develops intraoperatively or during the postoperative hospital stay and lasts at least 30 s or lasts for the duration of the ECG recording if only recorded for ≤30 s,368 or new onset of AF that occurs intraoperatively or within 30 postoperative days without specifying the duration.369
The incidence of POAF in Japan is 10%, and patients who develop POAF have a higher incidence of postoperative complications.370 The higher rates reported in Japan compared with other countries may be related to the longer duration of ECG monitoring in Japan during hospital stay.370 The incidence of AF by surgery is given in Table 51.370–373
Table 51. Incidence of Postoperative Atrial Fibrillation
| Surgery type |
Incidence |
| Cardiac thoracic surgery |
16–46%371 |
| Non-cardiac thoracic surgery |
0.4–12%371 |
| Lung surgery (wedge resection) |
2–4%370 |
| Lung surgery (lobectomy) |
10–15%371 |
| Pneumonectomy (unilateral pneumonectomy) |
20–30%371 |
| Esophagectomy for esophageal cancer |
4–10%371 |
| Colorectal cancer |
4.4%372 |
| Femoral fracture in older adults |
3.7%373 |
(Adapted from Fabiani I, et al. 2019,371 Higuchi S, et al. 2019,370 Leibowitz D, et al. 2017,372 and Siu CW, et al. 2005.373)
11.2.2 Factors Associated With POAF
The main risks of POAF are age, comorbid cardiac disease, and history of hypertension.374 Although there is no difference between thoracic and abdominal surgeries,375 laparotomic surgery has a higher risk than laparoscopic surgery.373 Surgical duration exceeding 600 min is also a risk.376
11.2.3 Short- and Long-Term Prognoses
a. POAF and Mortality Rate
POAF occurs most frequently on postoperative day 2, and is associated with prolonged hospital stay, and increases in the in-hospital and long-term mortality rates.374,377,378
b. POAF and Postoperative Complications
POAF after surgery for cancer in Japan is associated with perioperative complications. Complications within 30 postoperative days (i.e., acute myocardial infarction, congestive heart failure, bleeding, thrombosis, infection, and acute renal failure) were observed in 50% of POAF patients (odds ratio 2.84).370
c. POAF and Late AF Recurrence
POAF is associated with a higher rate of late AF recurrence. According to 1 Japanese report, AF recurrence occurred in 31% of patients within 1 year, which was a significantly higher rate than in non-POAF patients; however, the majority of patients were asymptomatic.379
d. Incidence of Ischemic Stroke in POAF Patients
The risk of ischemic stroke within 1 year is 1.47% in POAF patient compared with 0.36% in non-POAF patients, showing an association of POAF with ischemic stroke (hazard ratio [HR] 2.0; 95% confidence interval [CI] 1.7–2.3).380 Furthermore, meta-analyses have reported associations between POAF and the incidence of early postoperative and long-term stroke, and with higher long-term mortality rates.377,378
11.2.4 Prophylaxis
According to a Cochrane systematic review updated in 2019, β-blockers may reduce AF, but may also increase bradycardia and hypotension during the perioperative period, and it is unknown whether β-blockers could lower the postoperative 30-day mortality rate.141 This suggests that the use of β-blockers for the sole purpose of AF prophylaxis is not recommended (Class III, harm) (Table 52). Angiotensin-converting enzyme inhibitors and aldactone are also reported to be ineffective in preventing AF.381,382
Table 52. Recommendations and Levels of Evidence on POAF in the Intraoperative or Postoperative Period
| |
COR |
LOE |
| Anticoagulant therapy should be considered if complicated with POAF |
IIa |
B |
| Beta-blockers should not be administered for the sole purpose of preventing POAF |
III
(Harm) |
B |
COR, class of recommendation; LOE, level of evidence; POAF, perioperative/postoperative atrial fibrillation.
11.2.5 Perioperative Management of POAF
The therapeutic strategy for POAF is same in the non-perioperative period; that is, the indication of rate control, anticoagulant therapy, and rhythm control should be considered. Electrical or pharmacological defibrillation should be attempted in cases of severe symptoms or unstable hemodynamics.171
a. Anticoagulant Therapy for POAF
In 1 study, anticoagulant therapy was initiated for 24.4% of POAF cases and was continued 1 year later in 35.5% of these patients. Anticoagulant therapy was associated with reduced risk of thromboembolism (adjusted HR 0.52, 95% CI 0.40–0.67).383
There have been no randomized clinical trials on POAF with indications for anticoagulant therapy. In the present clinical practice guidelines, anticoagulant therapy should be considered according to the patient’s risk of thrombosis, as is done for non-valvular AF (Class IIa, B) (Table 52).
11.3 Ventricular Arrhythmia
New onset ventricular tachycardia (VT)/ventricular fibrillation (VF) in the perioperative period of non-cardiac surgery is rare, as it only occurs in 0.5% of patients.384 The severity or prognosis of ventricular arrhythmia is determined by the presence or absence of underlying heart disease. When VT/VF occurs, physician should consider the possibilities of concomitant cardiac disease (i.e., ischemic heart disease, valvular heart disease, cardiomyopathy) not identified preoperatively, or of electrolyte imbalance, effects of drugs used, and postoperative stress cardiomyopathy. Ventricular extrasystole or non-sustained VT observed before surgery is not a factor of poor perioperative or later prognosis without underlying heart disease, such as myocardial infarction.385,386
a. Brugada Syndrome
The mechanism of Brugada syndrome is associated with parasympathetic nerve activity. VF has also been reported to develop with fever.387 Patients with Brugada syndrome should be given strict body temperature and electrolyte control. In the event of VF, sympathetic stimulation with isoproterenol infusion is effective for preventing recurrence.387
11.4 Bradycardia Arrhythmia
The perioperative incidence of bradycardia arrhythmia is reported to be 0.4%,384 and its etiology is reported to involve drugs (β-blockers, digitalis, calcium-channel blockers, and amiodarone) electrolytes (hyperkalemia), parasympathetic nerve activity (spinal anesthesia, intubation, irrigation), and myocardial ischemia.388 It is treated as needed by drugs (e.g., atropine, dopamine, isoproterenol) or temporary cardiac pacing.
11.5 Cautions in Perioperative Use of Antiarrhythmic Drugs
Renal, hepatic and cardiac function may be impaired by perioperative complications. Antiarrhythmic drugs should be used with caution due to the potential for interactions with various other drugs. Accordingly, pharmacokinetics should be considered for drug selection and dosage adjustments should be made carefully.368
12. Pacemakers and Implantable Cardioverter Defibrillators
Cardiac implantable electronic devices (CIEDs) include a wide variety of devices, ranging from implantable electrocardiographs to pacemakers, cardiac resynchronization therapy pacemakers (CRT-Ps), implantable cardioverter defibrillators (ICDs), and cardiac resynchronization therapy defibrillators (CRT-Ds). Given appropriate perioperative CIED management, patients with these devices can undergo surgery as safely as regular patients, but they require the support of physicians and other medical professionals specializing in CIEDs.
12.1 Electromagnetic Interference
CIEDs sense electric excitation generated by the heart for rhythm diagnosis and treatment. Regular pacemakers inhibit pacing if intrinsic waveforms are sensed. When the CIED detects electronic noise from electric devices used for surgery, such as electrocautery devices, the necessary pacing is inhibited, which can cause bradycardia or cardiac arrest. Furthermore, ICD can malfunction, causing inappropriate antitachycardia pacing and electrical shock, when they misinterpret electronic noise as ventricular arrhythmia. Appropriate programming of CIEDs is thus recommended in the perioperative period.
12.2 Preoperative Preparation (Table 53)
Table 53. Checklist for the Perioperative Management of Patients With Cardiac Implanted Electronic Devices (CIEDs)
| CIED team |
| Indication of CIED |
□ |
SSS, AV block, prevention of sudden cardiac death,
resynchronization for heart failure, investigation for
arrhythmia, others ( ) |
| Device model |
□ |
PM · ICD · SICD · CRT-P · CRT-D · ICM |
| Manufacturer |
□ |
|
| Date of last interrogation |
□ |
|
| Pacing and sensing threshold, Impedance |
□ |
|
| Battery longevity |
□ |
≥3 months · <3 months |
| Unipolar lead |
□ |
Yes / No |
| Intrinsic heartbeat |
□ |
Yes / No |
| Pacing mode |
□ |
DDD · DDI · VDD · VVI |
| Programmed lower rate |
□ |
beats/min |
| Rate response mode |
□ |
Yes / No |
| ICD therapy |
□ |
VF zone beats/min |
| |
□ |
VT zone: VT1 beats/min, VT2 beats/min |
| Surgical team |
| Type of surgery |
□ |
|
| Date of surgery |
□ |
|
| Surgical site |
□ |
|
| Patient positioning during surgery |
□ |
|
| Use of unipolar electrosurgical knife |
□ |
Yes / No |
AV block, atrioventricular block; CRT-D, cardiac resynchronization therapy defibrillator; CRT-P, cardiac resynchronization therapy pacemaker; ICD, implantable cardioverter defibrillator; ICM, insertable cardiac monitor; PM, pacemaker; SICD, subcutaneous implantable defibrillator; SSS, sick sinus syndrome.
As soon as a patient with a CIED is scheduled for a surgery, perioperative measures should be discussed in advance between the cardiologist or other healthcare professionals with knowledge of CIEDs, the surgeon, and the anesthesiologist. For example, a checklist as shown in Table 53 should be prepared, and the relevant information should be shared with the entire team involved in the surgery. Furthermore, the baseline device parameters should be evaluated immediately before surgery for postoperative comparison.
In case of an unanticipated CIED problem interrupting surgery, a backup transcutaneous pacing or external defibrillator should be prepared.
12.3 Intraoperative Measures
12.3.1 Pacemaker Management
Intraoperative pacemaker settings are decided by the patient’s level of dependency on the pacemaker. When a patient is completely dependent on the pacemaker and unable to generate an intrinsic heart rate such as with complete atrioventricular block, inhibiting pacing by electromagnetic interference (EMI) can cause bradycardia or cardiac arrest, thus it is necessary to change the pacemaker program to asynchronous pacing mode (AOO, VOO, or DOO), which paces at a fixed rate to ensure that pacing is not inhibited by external electric noise. Conversely, the asynchronous mode may be inappropriate for patients with sufficient intrinsic heart rate, in which case mode setting changes are unnecessary.
12.3.2 Cardiac Resynchronization Therapy Pacemaker Management
CRT-Ps are more important for their role in treating heart failure by improving dyssynchrony, rather than as a treatment for bradycardic arrhythmia. Most patients implanted with a CRT have an intrinsic heart rate, and generally do not require changing of the mode.
12.3.3 ICD and CRT−D Management
An ICD diagnoses ventricular arrhythmia by heart rate. Electrocautery generates extremely high frequencies (i.e., 100 kHz to 4 MHz), and the cycles of electric noise can enter the zone of ventricular fibrillation and trigger a malfunction (inappropriate electrical shock or antitachycardia pacing); thus, antitachyarrhythmia therapy (defibrillation and antitachycardia pacing) should be turned off immediately before surgery. The entire surgical team should be aware that the defibrillation function has been turned off, and transcutaneous pacing/defibrillator pads should be applied to the patient in advance. To minimize damage to the ICD during external defibrillation, ideal positioning of the pads is important. The standard anteroposterior pad position is preferred to than anterolateral pad position. Furthermore, the anterior pad should be placed as far from the device as possible.
Perioperative management of patients with a SICD is no different than for patients with a standard ICD; however, because the shock lead passes subcutaneously from the area near the sternum to the lateral aspect of the chest, surgery should be performed with due caution to lead damage.
12.3.4 Intraoperative Measures by Surgeons
Risk of EMI can be reduced greatly by using bipolar rather than unipolar electrocautery.389 However, although bipolar electrocautery is suitable for achieving hemostasis of small areas with low output, it takes time for hemostasis of a large area and is much less effective in cutting tissue. Thus, unipolar electrocautery is more often used in surgical practice, in which case it is necessary to minimize the energy output to minimize the influence of EMI and to keep the current short and intermittent (e.g., currents within ≤5 s and ≥5 s apart).389 High-frequency currents that flow into the body from unipolar electrocautery are collected by the return electrode; thus, its position is also important. Normally, the adhesive return electrode is placed on the back or thigh. However, the current can pass through the device during surgery of the upper limbs, chest, head or neck. Therefore, it is necessary to adjust the position of the return electrode so that the current pathway from the electrosurgery will not cross the CIED generator or leads. ECG monitoring is necessary intraoperatively, but can be sometimes difficult to analyze due to EMI influence; therefore, other devices such as pulse oximeter or arterial line, which are not influenced by EMI, are necessary to monitor heart rate and rhythm intraoperatively.
12.4 Postoperative Measures
Use of electrocautery near the device can cause damage to the main body, the leads or myocardium in contact with the leads. The pacing threshold, sensitivity, resistance value, and battery lifespan should be assessed postoperatively to ensure there are no effects from EMI. Device settings should be returned to the preoperative settings, and most crucially, defibrillation activity on the ICD needs to be turned back on. Until then, ECG monitoring and the transcutaneous pacing/defibrillator pads should not be removed from patient.
12.5 Device Infection Prophylaxis
Device infection refers to infection of the implanted CIED or of the electrode leads, and is a serious complication that can cause infectious endocarditis and sepsis. There is no evidence supporting the effectiveness of perioperative prophylactic antibiotics. The American College of Cardiology’s statement on CIED infections and treatment does not recommend prophylactic antibiotic therapy for invasive procedures, including invasive dental, gastrointestinal, or genitourinary procedures that do not directly manipulate the device (Class III).390
V. Postoperative Surveillance of Myocardial Injury
1. Myocardial Injury After Non-Cardiac Surgery
It has been reported that myocardial injury that does not meet the diagnostic criteria of myocardial infarction (MI) can occur in the perioperative period.391 Myocardial ischemia-induced myocardial injury that develops intraoperatively or within 30 postoperative days of non-cardiac surgery is defined as MINS (Table 54).391 Furthermore, MINS has been reported to be independently and significantly associated with subsequent death.392
Table 54. Definition and Diagnostic Criteria of Myocardial Injury After Non-Cardiac Surgery (MINS)
| Definition |
| Myocardial injury due to ischemia that occurs during or within 30 days after non-cardiac surgery |
| Diagnostic criteria |
| Rise and fall pattern in postoperative troponin level |
| No evidence of non-ischemic postoperative complications (e.g., sepsis, pulmonary embolus, atrial fibrillation) |
No requirement for ischemic features fulfilling the universal definition of myocardial infarction*: ECG changes, new ventricular
wall motion abnormalities on echocardiography, or new ischemic findings on stress myocardial scintigraphy |
*Third universal definition of myocardial infarction. Thygesen K, et al. 2012.397 (Adapted from Smilowitz NR, et al. 2019.282)
1.1 Diagnostic Criteria
The definition and diagnostic criteria of MINS were established from inclusion/exclusion criteria of previously reported clinical trials on MINS393,394 (Table 54).
1.2 Clinical Features
MINS occurs in approximately 8–18% of non-cardiac surgeries, but approximately 84–93% of cases are asymptomatic or lacking in ECG changes.392,394 A reason for this is that chest symptoms are masked by analgesics used postoperatively and pain from the wound.395 MINS is broadly classified into types 1 and 2 according to disease condition396 (see 20.1. Mechanisms of perioperative AMI for types 1 and 2). History of heart failure, chronic kidney disease, coronary artery disease and MI have been reported as risk factors of MINS.282
1.3 Prognosis of MINS
The 2014 large-scale observational VISION study reported that MINS was independently associated with increased risk of 30-day death (hazard ratio [HR] 3.87, 95% confidence interval [CI] 2.96–5.08).392
The BASEL-PMI study of 2,018 high cardiovascular risk patients (2,546 surgeries) found that MINS was independently and significantly associated with increased risk of 30-day (9.8% vs. 1.6%; HR 2.73, 95% CI 1.54–4.84) and 1-year deaths (22.5% vs. 9.3%; HR 1.58, 95% CI 1.16–2.15).398 Furthermore, it was also associated with increased risk of 30-day (4.9% vs. 0.5%) and 1-year cardiovascular mortality rates (9.1% vs. 2.6%).398
A meta-analysis of 14 studies of 3,318 patients showed that increased troponin was an independent predictor of 1-year postoperative overall mortality (odds ratio 6.7, 95% CI 4.1–10.9).399
An analysis dividing peak high-sensitivity troponin T levels in the 3 postoperative days into 3 groups (20–65 ng/L, 65–1,000 ng/L, 1,000 ng/L) and comparing their prognosis to the group with the normal value (<5 ng/L) found that, regardless of presence or absence of ischemia symptoms, elevated troponin was associated with 30-day death. Furthermore, higher peak troponin T values were significantly associated with a higher 30-day mortality rate.394
Accordingly, MINS is suggested to be associated with both short- and long-term poor prognosis.
1.4 Management of MINS
Several clinical studies have reported that aspirin and statins are effective for the secondary prevention of ischemic heart disease.149,400 However, observational studies on the effects of oral aspirin and statin therapy for patients who developed MINS reported that these medications were associated with decreased rates of cardiovascular events and 30-day death; however, the evidence is limited.393,395
The MANAGE trial, a 2018 randomized controlled trial that investigated the efficacy of dabigatran after developing MINS, showed that dabigatran 110 mg twice daily for post-MINS patients decreased major cardiovascular events such as cardiovascular death and non-fatal MI without significantly increasing hemorrhagic events (HR 0.72, 95% CI, 0.55–0.93) at a mean follow-up of 16 months.401 However, the data were insufficient for making a recommendation based on this trial alone; thus, accumulation of evidence is awaited.
It can be considered reasonable to consult a cardiologist and discuss management of patients who have developed MINS on an individual basis.
1.5 Postoperative Monitoring of Myocardial Troponin
As mentioned earlier, MINS is often asymptomatic and is only recognized by postoperative surveillance of troponin levels.392,398 The Canadian Cardiovascular Society Guidelines 201710 lists high preoperative B-type natriuretic peptide, Revised Cardiac Risk Index ≥1 point, age 45–65 years with a history of coronary artery disease and age ≥65 years as high risk for MINS, and recommends perioperative surveillance with troponin monitoring. In high-risk patients, postoperative myocardial troponin monitoring can be considered (Class IIb, C) (Table 55), but the management of MINS diagnosed by this routine surveillance has yet to be established.
Table 55. Recommendations and Levels of Evidence on Postoperative Troponin and ECG Monitoring
| |
COR |
LOE |
Postoperative MINS surveillance by myocardial troponin may be considered for
high-risk patients |
IIb |
C |
Myocardial ischemia surveillance with postoperative 12-lead ECG may be
considered for high-risk patients |
IIb |
C |
COR, class of recommendation; LOE, level of evidence.
A prospective observational study of 3,564 non-cardiac surgeries reported that ischemic findings on postoperative ECG were an independent predictor of perioperative major cardiovascular events (OR 2.20, 95% CI 1.1–3.7).402
A prospective observational study that monitored continuous 12-lead ECG in the 48–72 postoperative hours of 185 patients who underwent vascular surgery reported that transient ischemic findings occurred in 20.5% and sustained postoperative MI occurred in 6.5%, most cases developing in the early postoperative period (<60 min).403
Myocardial ischemia surveillance with 12-lead ECG can be considered in the postoperative 48–72 h for high-risk patients (Class IIb, C) (Table 55).
Part 2: Recommendations According to the Grading of Recommendation Assessment, Development, and Evaluation (GRADE) System
VI. CQ1–6
CQ1: Is Preoperative Coronary Revascularization Recommended for Patients With Stable Coronary Artery Disease Undergoing Non-Cardiac Surgery?
Recommendation
It is recommended not to perform preoperative prophylactic coronary revascularization for patients with stable coronary artery disease undergoing non-cardiac surgery (GRADE 1B) (Strength of recommendation: Strong, Certainty of evidence: Moderate).
Practical Considerations
1. For patients with severe coronary artery disease, such as left main trunk disease or disease with Canadian Cardiovascular Society (CCS) Class III or IV symptoms, which is a general indication for revascularization in the non-perioperative setting, the options in the perioperative setting are to (1) perform revascularization (percutaneous coronary intervention [PCI] or coronary artery bypass graft [CABG]) first, then the non-cardiac surgery, (2) perform the non-cardiac surgery first, then, revascularization, (3) perform either revascularization or non-cardiac surgery and (4) perform neither. It is important for the multidisciplinary team to comprehensively assess the relevant issues, and engage in shared decision-making with the patient based on the patient’s values and preferences
2. Because the perioperative risk increases with the complication of coronary artery disease, it is important to work together with a multidisciplinary team consisting of the surgeon, anesthesiologist, and cardiologist in the preoperative, intraoperative, and postoperative periods (see also Chapter I.3 Potential Interventions for Risk Reduction) to reduce the risk
3. Patients with unstable angina, aortic stenosis, and heart failure (left ventricular dysfunction) were not included in the main clinical trial related to the present recommendation, and thus the management should be individualized
A. Background and Clinical Significance
When patients with stable coronary artery disease undergo non-cardiac surgery, myocardial oxygen imbalance due to bleeding, pain, and inflammatory cytokine production etc. can result in myocardial ischemia. Reportedly, coronary artery disease increases the perioperative cardiovascular risk. In patients without any stenotic lesions on coronary computed tomography angiography, the cardiovascular event rate was 2.0%, but this rate increased to 4.1% in patients with non-significant stenotic lesions, to 7.3% in patients with a single-vessel lesion, and to 23.1% in patients with multivessel lesions.112 Revascularization may be sometimes performed preoperatively for this type of coronary artery disease to prevent perioperative cardiovascular events. However, there are some concerns related with preoperative preventive revascularization.
First, because myocardial infarction (MI) is generally caused by plaque rupture, PCI, which is often performed for revascularization, is considered ineffective in preventing MI due to plaque rupture. Second, perioperative discontinuation of antiplatelet therapy for patients with coronary stent implantation can also generate new risk of MI triggered by stent thrombosis. Third, PCI and CABG can produce serious complications, including MI and death. Fourth, performing revascularization with PCI or CABG delays the non-cardiac surgery, which can be harmful in some cases such as disease progression of malignancy and rupture of aortic aneurysm, etc. Whether revascularization performed prior to non-cardiac surgery contributes to reducing perioperative cardiovascular risks is, therefore, an important clinical question to be addressed.
B. Summary of Available Evidence
B.1 PICO or PECO
P: Patient with stable coronary artery disease, I or E: preoperative revascularization, C: no revascularization, Outcome chosen and approved in panel session is shown in B.2. Its rating of importance was discussed and approved in panel session.
B.2 Outcomes (Rating of Importance)
Desirable effect (benefit): perioperative mortality (9), perioperative MI (7), 1-year mortality (9)
Undesirable effect (harm): perioperative bleeding complications (8), cost (7), length of hospital stay (7), length of delay of non-cardiac surgery (7)
B.3 Systematic Review
Systematic review of preoperative revascularization resulted in 1 randomized controlled trial (RCT: the CARP trial [McFalls et al., 2004217]), and 4 observational studies (Posner et al., 1999;219 Takahashi et al., 2002;222 Yamaguchi et al., 2004;223 Felix et al., 2016221). Landesberg et al. (2003220) was not included in the meta-analysis because it showed only survival rate and did not show the number of deaths.
C. Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
C.1 Desirable Effects (Benefit) (See Table 56 Summary of Findings)
Table 56. Summary of Findings on the Effect of Prophylactic Revascularization on Perioperative Cardiovascular Risk Among Stable Coronary Artery Patients Undergoing Non-Cardiac Surgeries
| |
Design
No. of research
No. of patients |
No
revascularization*
(Control) |
Revascularization**
(95% CI) |
Relative
risk***
(95% CI) |
Absolute risk
difference**
(95% CI) |
Certainty |
Importance
of outcome |
Plain language
summary |
Perioperative
mortality |
RCT
1
510 |
3.6% |
6.6%
(3.0–14.5%) |
1.84
(0.84–4.06) |
3.0% higher
(0.6% lower to
10.9% higher) |
B1 |
B3 |
9 |
Preoperative
revascularization
does not decrease
mortality |
Observational
4
747 |
3.4% |
3.6%
(1.6–8.6%) |
1.08
(0.49–2.40) |
0.27% higher
(1.7% lower to
4.7% higher) |
C2 |
Perioperative
myocardial
infarction |
RCT
1
510 |
9.3% |
8.6%
(4.7–15.8%) |
0.93
(0.51–1.70) |
0.7% lower
(4.5% lower to
6.5% higher) |
B4 |
B6 |
7 |
Preoperative
revascularization
does not decrease
MI |
Observational
1
284 |
3.5% |
7.0%
(2.5–20.0%) |
2.0
(0.70–5.70) |
3.5% higher
(1.1% lower to
16.5% higher) |
C5 |
Long-term
mortality |
RCT
1
510
Median follow-up
time 2.8 years |
27% |
27.1%
(20.5–36.2%) |
1.02
(0.77–1.36) |
0.53% higher
(6.1% lower to
9.6% higher) |
B7 |
B9 |
9 |
Preoperative
revascularization
does not decrease
long-term mortality |
Observational
1
296
Mean follow-up
time 8.6 years |
20.8% |
23.7%
(15.0–37.4%) |
1.14
(0.72–1.80) |
2.9% higher
(5.8% lower to
16.6% higher) |
C8 |
*Event rate of control group.
**Calculated by event rate of control group and relative risk in the meta-analysis.
***Relative risk calculated in the meta-analysis using Review Manager.
1Certainty downgraded by one to B due to indirectness because there was only 1 RCT that included only vascular surgery.
2Certainty judged to be C due to observational studies only.
3Certainty of mode of evidence was judged to be B according to the certainty of the RCT because of no heterogeneity of results among RCTs and observational studies.
4Certainty downgraded by one to B due to indirectness because there was only one RCT that included only vascular surgery.
5Certainty judged to be C due to observational studies only.
6Certainty of mode of evidence was judged to be B according to the certainty of the RCT because of no heterogeneity of results among RCTs and observational studies.
7Certainty downgraded by one to B due to indirectness because there was only 1 RCT that included only vascular surgery.
8Certainty judged to be C due to observational studies only.
9Certainty of mode of evidence judged to be B according to the certainty of the RCT because of no heterogeneity of results among RCTs and observational studies.
CI, confidence interval; RCT, randomized controlled trial. (Adapted from references 217, 219, 221–223.)
C.1.1 Perioperative Mortality Rate (Figure 10)
Perioperative mortality rate was investigated in 1 RCT217 and 4 observational studies.219,221–223 According to the RCT, the mortality rate of patients who did not undergo revascularization vs. those who did was 3.6% vs. 6.6% (95% CI 3.0–14.5%), the relative risk (RR) was 1.84 (95% CI 0.84–4.06), and the absolute risk difference for death was 3.0% higher in patients who underwent revascularization (95% CI 0.6% lower to 10.9% higher). In the meta-analysis of 4 observational studies (n=747), the corresponding data were 3.4% vs. 3.6% (95% CI 1.6–8.6%), the RR was 1.08 (95% CI 0.49–2.40), and the absolute risk difference was 0.27% higher (95% CI 1.7% lower to 4.7% higher). This difference was not statistically significant. The guideline panel concluded that revascularization did not lower the mortality rate.
C.1.2 Perioperative Myocardial Infarction (Figure 10)
Perioperative MI was studied in 1 RCT217 and 1 observational study.219 In the RCT (n=510), the incidence of perioperative MI in patients who did not undergo revascularization vs. those who did was 9.3% vs. 8.6% (95% CI 4.7–15.8%), the RR was 0.93 (95% CI 0.51–1.70), and the absolute difference was 0.7% lower (95% CI 4.5% lower to 6.5% higher), resulting in no significant difference. In the observational study (n=284), the incidences were 3.5% vs. 7.0%, the RR was 2.0 (95% CI 0.70–5.7), and the absolute difference was 3.5% higher (1.1% lower to 16.5% higher). Thus, the risk tended to be slightly higher in patients who underwent revascularization, although the difference was not significant. The guideline panel concluded that revascularization did not lower the incidence of perioperative MI.
C.1.3 Long-Term Mortality Rate (Figure 10)
This was studied in 1 RCT217 and 1 observational study.221 The RCT (n=510) had a median follow-up of 2.8 years, and there was no significant difference between patients who did not undergo revascularization vs. those who did. The mortality was 27.0% vs. 27.1% (95% CI 20.5–36.2%), and RR was 1.02 (95% CI 0.77–1.36). The same was true for the observational study (n=296), where the late mortality rate was 20.8% vs. 23.7% (95% CI 15.0–37.4%), the RR was 1.14 (95% CI 0.72–1.80), with an absolute risk difference of 2.9% higher (5.8% lower to 16.6% higher); thus, there was no significant difference. The guideline panel concluded that preoperative revascularization did not lower the long-term mortality rate.
C.1.4 Summary of Desirable Effects
The guideline panel concluded that there are no desirable effects (benefit) to revascularization before non-cardiac surgery.
C.2 Undesirable Effects (Harm)
The outcomes to be studied included perioperative bleeding complications (8), cost (7), length of hospital stay (7), and length of surgery delay associated with revascularization (7). Rating of importance is given in parentheses. According to the systematic review, there were no studies investigating these outcomes.
C.2.1 Additional Considerations on Undesirable Effects
a. Delay of the Non-Cardiac Surgery Associated With Revascularization
We recommended in CQ2 to delay elective procedures by at least 1–1.5 months after performing PCI. Among those who underwent revascularization in the CARP trial, 41% and 59% underwent CABG and PCI, respectively, and non-cardiac surgery was delayed by 36 more days than in patients who did not undergo revascularization.217 Before non-cardiac surgery, 3.9% (10/258) of patients who underwent revascularization and 0.39% (1/252) of those who did not died.217 This underlines the importance of carefully considering the risk of revascularization itself and the risk of postponing surgery.
The guideline panel concluded that there are minor to moderate undesirable effects to revascularization before non-cardiac surgery.
C.3 Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
Although coronary artery disease increased the perioperative risks, revascularization did not lower the mortality rate or the incidence of MI. Considering the risks of progression of the primary disease associated with delay of the non-cardiac surgery for revascularization and the risks of revascularization itself, the guideline panel concluded that there are no benefits to prophylactic revascularization for lowering the risk of surgery in stable coronary artery disease patients (has no prophylactic effect), and that the harms outweigh the benefits.
D. Certainty of Body of Evidence
The RCT and the observational studies both showed no benefits of preoperative prophylactic revascularization. There was only 1 RCT; thus, the certainty of evidence was downgraded to B. The RCT and the observational studies had the same results for all outcomes; thus, the certainty of evidence was concluded to be B, using the assessment of the RCT.
E. Values and Preferences
There were no studies of patients’ values or priority of the chosen outcomes. The panel gave the following opinions for additional consideration. If revascularization and the non-cardiac surgery are both Class I indications, the patient’s level of interest in the cardiac disease or primary disease (to be treated by non-cardiac surgery) will determine which should be prioritized. If coronary artery disease is stable and the non-cardiac surgery is time-sensitive (e.g., malignancy) or an urgent procedure (e.g., femoral neck fracture, cholecystitis), then many patients probably prioritize the non-cardiac surgery.
F. Cost
We could not identify any studies performing a cost–benefit analysis from the perspectives of patients. The out-of-pocket costs of patients are estimated below.
In Japan, the inpatient costs of coronary artery bypass surgery and PCI are estimated to be ≈3 million Japanese yen (JPY)404,405 and JPY1.4–1.5 million,406 respectively.
In Japan, medical costs paid by patients are capped within the High-Cost Medical Expense Benefit system. For example, the monthly out-of-pocket costs of a patient with annual income JPY3.7 million–7.7 million is calculated as follows:407
JPY80,100 + (medical costs − 267,000) × 1%
A JPY3 million coronary artery bypass surgery will be calculated as JPY107,000, and a JPY1.4 million PCI is calculated as JPY91,000. A moderate cost is thus incurred.
G. Acceptability
Prophylactic revascularization preoperative to non-cardiac surgery requires delaying of the surgery; thus, the acceptability would be low for patients who are undergoing a time-sensitive procedure, and the acceptability would be high for an elective procedure. Acceptability for the surgeon or anesthesiologist is unknown. Therefore, acceptability was graded as “vary.”
H. Feasibility
A specialized institution is needed for prophylactic revascularization. A patient in a non-specialized institution can still undergo the procedure by being referred to a specialized institution. When patients are known to have coronary artery disease, the perioperative risk is increased. Therefore, it might be safer to refer patients to an institute for surgery where a cardiologist can be involved in the perioperative period. Whether performing preoperative revascularization or performing non-cardiac surgery without revascularization for patients with stable coronary artery disease, feasibility is relatively low.
I. Grading Recommendation
Panel members voted for “recommending against performing prophylactic revascularization.” The RCT did not include patients with unstable angina, left main trunk disease, ejection fraction <20%, or aortic stenosis,217 suggesting that those patients should be assessed on an individual basis, and it should be reiterated that the recommendation only applies to patients with stable coronary artery disease without above conditions.
For patients with severe coronary artery disease such as left main trunk disease or CCS Class III or IV symptoms, that would be an indication for revascularization in the non-perioperative period. The options are (1) to perform revascularization (PCI or CABG) first, (2) to perform non-cardiac surgery first, (3) to perform only revascularization or the non-cardiac surgery, or (4) to perform neither. It is important to assess the indication and the timing of the procedures (non-cardiac surgery, cardiac procedures, and their alternatives) through comprehensive discussion with a multidisciplinary team including the surgeon, anesthesiologist, and cardiologist etc. and to have shared decision-making with the patient. In this process, (1) the urgency of revascularization, (2) the urgency of non-cardiac surgery (harms of delaying surgery), (3) the risks of revascularization, (4) the risks of non-cardiac surgery, and (5) the risk of thrombosis or bleeding by discontinuing or continuing antiplatelet therapy in the perioperative period after revascularization are taken into consideration.
The recommendation was approved with a 92% voting rate, median 8, disagreement index 0.132, and 92% agreement by modified Delphi method (RAND appropriateness method).
J. Recommendation by Other Related Clinical Practice Guidelines
The ESC 2014,4 AHA 2014,59 and Canadian Cardiovascular Society 201710 have provided recommendations (see Chapter I.4.2 Prophylactic Revascularization).
K. Monitoring Risks After Non-Cardiac Surgery
Stable coronary artery disease increases the risk of postoperative cardiovascular events. It is important to diagnose myocardial ischemia timely and to intervene as soon as possible.
L. Surveillance and Evaluation of the Recommendation
The perioperative mortality rate and incidence of acute MI in the hospital should be surveyed in patients with known coronary artery disease.
M. Future Research Directions
Although already evaluated in an RCT on vascular surgery, an RCT should be considered to determine whether the findings are similar in non-vascular intermediate- to high-risk surgeries (P: non-vascular surgery patients, I: prophylactic revascularization, C: no prophylactic revascularization, O: mortality rate, cardiovascular events).
CQ2: Is It Recommended to Perform Non-Cardiac Surgery >1–1.5 Months After PCI With Coronary Stent Placement?
Recommendation
It is recommended not to perform non-cardiac surgery within 1–1.5 months after PCI with coronary stent placement (GRADE 1C) (Strength of the recommendation: Strong, Certainty of evidence: Low).
Practical Considerations
It is difficult to conclude whether the shortest length of delay should be 1 month or 1.5 months. How long surgery should be delayed beyond 1.5 months was not investigated within the scope of the systematic review in this CQ.
The optimal timing and perioperative management for time-sensitive procedures that cannot be delayed for a long period should be discussed comprehensively between multidisciplinary experts in surgery, anesthesiology and cardiology by evaluating the (1) risks of stent thrombosis (incidence of thrombosis, extent of MI in case of thrombosis), (2) risks of bleeding by continuing antiplatelet therapy (bleeding rate, organ injury due to bleeding, and difficulty in hemostasis), and (3) harms of delaying surgery.
A. Background and Clinical Significance
Patients with a coronary stent undergoing non-cardiac surgery have a risk of acute MI (AMI) through stent thrombosis by discontinuation of antiplatelet agents and hypercoagulability induced by surgery.11 The risk of perioperative cardiovascular events is reported to be particularly high in the early period after PCI.11 Specifying the optimal timing of non-cardiac surgery needed after PCI is therefore an important clinical question to be addressed.
B. Summary of Available Evidence
B.1 PICO or PECO
P: patients with a coronary stent undergoing non-cardiac surgery, I or E: perform surgery >1–1.5 months after PCI, C: surgery within 1–1.5 months after PCI
Outcome chosen and approved in panel session is shown in B.2. Its rating of importance was discussed and approved in panel session.
B.2 Outcomes (Rating of Importance)
Desirable effects (Benefits): perioperative mortality (9), perioperative AMI (8), perioperative cardiac death (9)
Undesirable effects (Harms): anxiety (7) (risk of patients becoming anxious due to delaying surgery)
B.3 Systematic Review
According to the systematic review, there were no RCTs, and 9 observational studies were identified.204,206,231,232,235,238,240,408,409 Additionally, a meta-analysis was performed using the following studies: van Kuijk et al., 2009;240 Chia et al., 2010;408 Tokushige et al., 2012;231 Tokushige et al., 2013;206 Hawn et al., 2013;204 Holcomb et al., 2014;235 Bangalore et al., 2015;409 Egholm et al., 2016;232 and Rossini et al., 2017.238
C. Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
C.1 Desirable Effects (Benefits) (See Table 57 Summary of Findings)
Table 57. Summary of Findings of the Effect of Delaying Non-Cardiac Surgery >1–1.5 Months After Coronary Stent Placement on Perioperative Outcomes
| |
Design
No. of research
No. of patients |
Non-cardiac
surgery ≤1–1.5
months after
coronary stent
placement
(Control)* |
Non-cardiac
surgery >1–1.5
months after
coronary stent
placement**
(95% CI) |
Relative
risk***
(95% CI) |
Absolute risk
difference**
(95% CI) |
Certainty |
Importance
of outcome |
Plain language
summary |
Perioperative
mortality |
Observational
3
68,471 |
4.5% |
1.3%
(0.8–2.0%) |
0.28
(0.18–0.45) |
3.2% lower
(2.5% lower to
3.7% lower) |
C1 |
9 |
Timing of surgery
>1–1.5 months
after PCI with stent
placement reduces
mortality |
Perioperative
cardiac
mortality |
Observational
1
4,303 |
4.9% |
0.3%
(0.2–0.7%) |
0.07
(0.04–0.14) |
4.5% lower
(4.2% lower to
4.7% lower) |
D2 |
9 |
Timing of surgery
>1–1.5 months
after PCI with stent
placement reduces
cardiac mortality |
Perioperative
acute
myocardial
infarction |
Observational
4
69,204 |
6.9% |
1.5%
(0.6–4.2%) |
0.22
(0.08–0.61) |
5.3% lower
(2.7% lower to
6.3% lower) |
D3 |
8 |
Timing of surgery
>1–1.5 months
after PCI with stent
placement reduces
MI |
*Event rate of control group.
**Calculated by event rate of control group and relative risk in the meta-analysis.
***Relative risk calculated in the meta-analysis using Review Manager.
1Certainty judged to be C due to observational studies only.
2Certainty downgraded by one to D because there was only 1 observational study.
3Certainty downgraded by one to D because there was significant heterogeneity of the result and significant risk of bias.
CI, confidence interval; PCI, percutaneous coronary intervention. (Adapted from references 206, 232, 235, 408.)
C.1.1 Overall Mortality (Figure 11)
Overall mortality was investigated in 3 observational studies (n=68,471).206,235,232 Estimated overall mortality of patients who underwent non-cardiac surgery >1–1.5 months after PCI was 3.2% lower than those who underwent non-cardiac surgery within 1–1.5 months (95% CI 2.5% lower to 3.7% lower).
C.1.2 Cardiac Mortality (Figure 11)
Cardiac mortality was investigated in 1 observational study (n=4,303).232 The overall cardiac mortality of patients who underwent non-cardiac surgery >1–1.5 months after PCI was 4.5% lower than those who underwent non-cardiac surgery within 1–1.5 months (95% CI 4.2% lower to 4.7% lower).
C.1.3 Perioperative MI (Figure 11)
Perioperative MI was investigated in 4 observational studies (n=69,204).206,232,235,408 Perioperative MI in patients who underwent non-cardiac surgery >1–1.5 months after PCI was 5.3% lower than in patients who underwent non-cardiac surgery within 1–1.5 months (95% CI 2.7% lower to 6.3% lower).
C.1.4 Perioperative Major Cardiac Adverse Events (MACE)
MACE was investigated in 6 observational studies.204,231,235,238,240,409 Because the events investigated as MACE were not same among the studies, we did not perform a meta-analysis. Incidence of MACE in patients who underwent non-cardiac surgeries >1–1.5 months after PCI was lower than in those who underwent it within 1–1.5 months in 4 studies; the relative risk (95% CI) was 0.21 (0.13–0.33),231 0.38 (0.33–0.43),204 0.31 (0.28–0.35),235 and 0.59 (0.40–0.87).240 No significant difference was noted in 2 studies; the RR (95% CI) was 0.83 (0.65–1.06),409 and 0.32 (0.08–1.24).238
C.1.5 Summary of Desirable Effects
Moderate
C.2 Undesirable Effects (Harms)
There was no research or evidence investigating anxiety.
Additional Considerations
Patients may not experience anxiety when their elective surgery can be delayed. On the other hand, delaying time-sensitive non-cardiac procedures such as surgery for malignancies can lead to progression of the primary disease and patients may experience anxiety. The undesirable effect of delaying elective surgeries for >1–1.5 months was determined to be “minor.”
C.3 Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
If the surgery for the primary disease is an elective procedure that can be delayed for a given time period, then the benefits will exceed the harms by delaying it by >1–1.5 months.
D. Certainty of Body of Evidence
All outcomes examined improved very significantly in the intervention group. The certainty of evidence related to death, the most important outcome, was cited to give a quality of evidence of C (low).
E. Values and Preferences
There were no studies on priorities and values related to the outcomes.
Additional Considerations
In cases of elective procedures that usually can be delayed, there should be no substantial uncertainty or variation of patients’ values and preferences even if delaying surgery is selected. However, there may be variation in values and preferences for time-sensitive procedures such as surgery for malignancies.
F. Cost
We could not identify any studies performing cost–benefit analysis from the perspectives of patients. It is estimated that there is no significant difference in cost by duration of delay.
G. Acceptability
Patients will likely accept delaying elective procedures, but acceptability will probably decrease for time-sensitive procedures.
H. Feasibility
This should be feasible in any institution because it does not involve an invasive intervention.
I. Grading Recommendation
The panel session concluded on the addition of the following practical considerations.
It is difficult to conclude whether the shortest length of delay should be 1 month or 1.5 months. How long it should be delayed for beyond 1.5 months was not investigated within the scope of the systematic review in this CQ.
The optimal timing and perioperative management for time-sensitive procedures that cannot be delayed for a long period should be discussed comprehensively by the multidisciplinary team of members from surgery, anesthesiology and cardiology to evaluate (1) the risks of stent thrombosis (incidence of thrombosis, extent of MI in case of thrombosis), (2) the risks of bleeding by continuing antiplatelet therapy (bleeding rate, organ injury due to bleeding, and difficulty in hemostasis in cases of bleeding), and (3) the harms of delaying surgery.
The recommendation was approved with 100% voting rate, median 8.0, disagreement index 0.056, and 100% agreement rate by modified Delphi method (RAND appropriate method).
J. Recommendations by Other Related Clinical Practice Guidelines
J.1 The JCS 2020 Guideline focused update on antithrombotic therapy in patients with coronary artery disease178 recommends
Elective non-cardiac surgery should not be performed for patients within 1 month after coronary stent deployment (Class III, harm, B).
J.2 The ESC 2017 focused update on dual antiplatelet therapy on coronary artery disease developed in collaboration with the EACTS210 recommends
Non-cardiac surgery should not be performed within 1 month after coronary stent placement (Class III, B).
J.3 The AHA 2016 ACC/AHA guideline focused update on duration of dual antiplatelet therapy in patients with coronary artery disease410 recommends
Elective non-cardiac surgery should be delayed 30 days after bare metal stent (BMS) implantation and optimally 6 months after drug-eluting stent (DES) implantation (Class I, B, non-randomized).
Elective non-cardiac surgery should not be performed within 30 days after BMS implantation or within 3 months after DES implantation in patients in whom dual antiplatelet therapy will need to be discontinued perioperatively (Class III, harm, B, non-randomized).
Elective non-cardiac surgery after DES implantation in patients for whom P2Y12 inhibitor therapy will need to be discontinued may be considered after 3 months if the risk of further delay of surgery is greater than the expected risks of stent thrombosis (Class IIb, C, expert opinion).
K. Monitoring Risks After Non-Cardiac Surgery
The patient should be monitored for MI in the perioperative period. The surveillance of MI with postoperative ECG and troponin may be considered for 48–72 h after the surgery for high-risk patients.
L. Surveillance and Evaluation of the Recommendation
Incidence of MI, mortality rate, cardiac mortality rate, and negative effects of delaying surgery on progression of the primary disease and long-term prognosis should be evaluated.
M. Future Research Directions
An RCT is awaited to examine how long the surgery should be delayed beyond 1–1.5 months (e.g., P: elective procedure patients, I: 3 months and later, C: 1–3 months, O: cardiovascular events).
CQ3: Is It Recommended to Perform Non-Cardiac Surgery With Continuation of Aspirin to Patients on Antiplatelet Therapy for Coronary Stent Placement?
Recommendation
Performing non-cardiac surgery under aspirin therapy continuation is suggested for patients on antiplatelet therapy for coronary stent placement if the risk of bleeding is not high (GRADE 2C) (Strength of recommendation: Weak, Certainty of evidence: Low).
A. Background and Clinical Significance
Within 1 and 2 years following percutaneous coronary intervention (PCI), approximately 10%204,411,412 and 17% of patients, respectively, require non-cardiac surgery.231 Post-PCI patients are at risk of stent thrombosis, which usually presents as acute myocardial infarction (AMI), and has a mortality rate of 10–25%.413 After PCI, dual antiplatelet therapy (DAPT) is needed for a given period, followed by a period on single antiplatelet therapy (SAPT) with aspirin.178 Discontinuing antiplatelet therapy when DAPT is essential is a high risk for stent thrombosis.413 Hypercoagulability is often observed in the perioperative period;414 thus, the risk of stent thrombosis should be considered,413 although bleeding risk is a concern if antiplatelet agents are continued. It is, consequently, important to assess the safety of non-cardiac surgery under antiplatelet therapy, particularly with aspirin treatment.
B. Summary of Available Evidence
B.1 PICO or PECO
P: Patients with a coronary artery stent undergoing non-cardiac surgery, I or E: Surgery while on aspirin, C: Surgery after suspending aspirin, Outcome chosen and approved in panel session is shown in B.2. Its rating of importance was discussed and approved in panel session.
B.2 Outcomes (Rating of Importance)
Desirable outcomes (benefits): perioperative mortality (9), perioperative AMI (8), brain infarction (7)
Undesirable outcomes (harms): major bleeding (9), minor bleeding (7)
B.3 Systematic Review
Systematic review adopted 1 randomized controlled trial (RCT) subgroup analysis (Graham et al., 2018415) and 2 observational studies (Tokushige et al., 2012,231 and Howell et al., 2019253).
C. Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
C.1 Desirable Effects (Benefits) (SeeTable 58 for Summary of Findings)
Table 58. Summary of Findings on the Desirable Effect (Benefit) of Perioperative Aspirin Continuation for Patients With Coronary Artery Stents
| |
Design
No. of research
No. of patients |
No aspirin*
(control) |
Aspirin**
(95% CI) |
Relative
risk***
(95% CI) |
Absolute risk
difference**
(95% CI) |
Certainty |
Importance
of outcome |
Plain
language
summary |
Perioperative
mortality |
RCT subanalysis
1
470 |
1.3% |
0.85%
(0.14–5.1%) |
0.67
(0.11–3.99) |
0.42% lower
(1.1% lower to
3.8% higher) |
C1 |
C5 |
9 |
Aspirin does not
decrease
mortality |
Observational
1
1,604 |
2.1% |
0.91%
(0.31–2.6%) |
0.44
(0.15–1.25) |
1.2% lower
(1.8% lower to
0.52% higher) |
C2 |
Perioperative
myocardial
infarction |
RCT subanalysis
1
470 |
11% |
5.1%
(2.6–9.9%) |
0.47
(0.24–0.90) |
5.9% lower
(8.4% lower to
1.1% lower) |
C3 |
C6 |
9 |
Aspirin
decreases MI7 |
Observational
1
1,604 |
0.43% |
0.22%
(0.03–1.9%) |
0.52
(0.06–4.46) |
0.21% lower
(0.41% lower to
1.5% higher) |
D4 |
*Event rate of control group.
**Calculated by event rate of control group and relative risk in the meta-analysis
***Relative risk calculated in the meta-analysis using Review Manager.
1Certainty judged to be C because of subanalysis of RCT only.
2Certainty judged to be C due to observational studies only.
3Certainty judged to be C because of subanalysis of RCT only.
4Certainty downgraded by one to D because of significant difference in background, failure to adequately control confounding factors and wide 95% CI.
5Certainty judged to be C because both the RCT and observational study showed the same tendency in the result and both of them had certainty C.
6Certainty judged to be C because both the RCT and observational study showed the same tendency in the result and both of them had certainty C.
7RCT subanalysis showed significant difference and a cohort study showed no significant difference. Both suggest a similar tendency and both are consistent with each other.
CI, confidence interval; RCT, randomized controlled trial. (Adapted from references 231, 415.)
C.1.1 Perioperative Mortality Rate (Figure 12)
The perioperative mortality rate of patients who discontinued aspirin or continued aspirin was 1.3% vs. 0.85% (95% CI 0.14–5.1%); with RR 0.67 (95% CI 0.11–3.99) in the subgroup analysis of the RCT (n=470)415 and 2.1% vs. 0.91% (95% CI 0.31–2.6%) with RR 0.44 (95% CI 0.15–1.25) in one observational study (n=1,604),231 indicating no significant difference in mortality rate.
C.1.2 Perioperative Incidence of AMI (Figure 12)
According to a subgroup analysis of the RCT415 (n=470), the perioperative incidence of MI of patients who discontinued aspirin and continued aspirin was 11% vs. 5.1% (95% CI 2.6–9.9%), and the absolute risk difference was 5.9% lower in patients who continued aspirin (95% CI 8.4% lower to 1.1% lower), showing a significantly lower risk by continuing aspirin. In 1 of the observational studies231 (n=1,604), there was no significant difference (RR 0.52 [95% CI 0.06–4.46]), but the point estimate of RR was low. Therefore, the panel committee concluded that continuation of aspirin reduces the risk of AMI.
C.1.3 Summary of Desirable Effects
The desirable effect of continuing aspirin was judged to be “small.”
C.2 Undesirable Effects (Harms) (See Table 59 for Summary of Findings)
Table 59. Summary of Findings on the Undesirable Effect (Harms) of Perioperative Aspirin Continuation for Patients With Coronary Artery Stents
| |
Design
No. of research
No. of patients |
No aspirin*
(control) |
Aspirin**
(95% CI) |
Relative
risk***
(95% CI) |
Absolute risk
difference**
(95% CI) |
Certainty |
Importance
of outcome |
Plain language
summary |
Perioperative
major
bleeding |
RCT subanalysis
1
470 |
3.8% |
3.4%
(1.3–8.7%) |
0.90
(0.35–2.28) |
0.38% lower
(2.5% lower to
4.9% higher) |
C1 |
C3 |
9 |
Aspirin does not
increase
bleeding |
Observational
2
2,226 |
2.4% |
1.3%
(0.64–2.7%) |
0.55
(0.27–1.13) |
1.1% lower
(1.3% lower to
0.31% higher) |
C2 |
*Event rate of control group.
**Calculated by event rate of control group and relative risk in the meta-analysis.
***Relative risk calculated in the meta-analysis using Review Manager.
1Certainty judged to be C due to RCT subanalysis only.
2Certainty judged to be C due to observational studies only.
3Certainty judged to C because both the RCT and observational studies showed the same tendency in the result and both of them had certainty C.
CI, confidence interval; RCT, randomized controlled trial. (Adapted from references 231, 253, 415.)
C.2.1 Perioperative Major Bleeding (Figure 13)
The incidence of perioperative major bleeding in patients who discontinued or continued aspirin was 3.8% vs. 3.4% (95% CI 1.3–8.7%) with RR 0.90 (95% CI 0.35–2.3) in the subgroup analysis of the RCT (n=470),415 and 2.4% vs. 1.3% (95% CI 0.64–2.7%) with RR 0.55 (95% CI 0.27–1.13) in the 2 observational studies (n=1,604),231,253 showing no significant difference.
Additional Considerations
Although there was no overall difference in bleeding risk, individual assessment is necessary for procedures that were excluded from the main study, such as intracranial and retinal surgery (i.e., surgeries that can cause serious complications such as paralysis or blindness even from slight bleeding).415
A meta-analysis based on the systematic review of 5 RCTs (n=666) comparing continuation/discontinuation of 1 antiplatelet agent in patients on antiplatelet therapy, including post-PCI patients, also did not find a significant difference in the rate of bleeding requiring transfusion with RR 1.37 (95% CI 0.83–2.26), or repeat surgery for hemostasis with RR 1.54 (95% CI 0.31–7.58).163
Undesirable effects were determined to be “minor.”
C.3 Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
Because there was no difference in bleeding rates, and the risk of AMI decreased, intervention (continuing aspirin) is considered superior.
D. Certainty of Body of Evidence
The certainty of body of evidence was judged to be C because all evidence showed certainty C, which consistently favors continuation of aspirin.
E. Values and Preferences
There were no studies on the values and preferences on the priority of outcomes.
F. Cost
We could not identify any studies performing a cost–benefit analysis from the perspectives of patients, but aspirin is inexpensive, and the out-of-pocket costs are estimated to be low, provided that bleeding is not increased.
G. Acceptability
The acceptability can be presumed to be high for surgeries with low bleeding risk. The acceptability can be expected to be low for surgeries with high bleeding risk (i.e., surgeries associated with high incidence of bleeding, and surgeries that can result in serious sequelae associated with bleeding). Therefore, acceptability was graded as “vary”.
H. Feasibility
Aspirin itself is inexpensive and widely available, so feasibility is high.
I. Grading Recommendation
There were no objections in the panel session. The recommendation was approved with 100% voting rate, median 8.0, disagreement index 0.056, and 100% agreement rate by modified Delphi method (RAND appropriate method).
J. Recommendation by Other Related Clinical Practice Guidelines
J.1 JCS 2020 guideline focused update on antithrombotic therapy in patients with coronary artery disease178
In patients treated with DAPT after coronary stent implantation who must undergo surgical procedures that mandate the discontinuation of P2Y12 inhibitor therapy, it is recommended that aspirin be continued if possible and the P2Y12 platelet receptor inhibitor be restarted as soon as possible after surgery (Class I, B).
J.2 2017 ESC Focused Update on DAPT in Coronary Artery Disease210 (p.32)
It is recommended to continue aspirin perioperatively if the bleeding risk allows, and to resume the recommended antiplatelet therapy as soon as possible postoperatively (Class I, B).
J.3 2016 ACC/AHA Guideline Focused Update on Duration of Dual Antiplatelet Therapy in Patients with Coronary Artery Disease410 (p.18)
In patients treated with DAPT after coronary stent implantation who must undergo surgical procedures that mandate the discontinuation of P2Y12 inhibitor therapy, it is recommended that aspirin be continued if possible and the P2Y12 platelet receptor inhibitor be restarted as soon as possible after surgery (Class I, C-E0).
K. Monitoring Risks After Non-Cardiac Surgery
Postoperative bleeding should be monitored by vital signs, observation of the surgical site, and blood tests (Hb level). In rare cases that bleeding cannot be controlled, transfusion of antiplatelets to reverse the effect of antiplatelet agents,416–418 hemostasis by interventional radiology, and surgical hemostasis should be considered.
L. Surveillance and Evaluation of the Recommendation
Presence or absence of antiplatelet agents, risk of transfusion, repeat of surgery for bleeding, mortality risk related to bleeding, in-stent thrombosis, and incidence of AMI are monitored.
M. Future Research Directions
An RCT is necessary (P: subjects who are matched for number of days since stent placement (e.g., 1.5–3 months, 3–6 months, >6 months) between an intervention group and comparison group, I: antiplatelet monotherapy, C antiplatelet agent discontinuation, and O: cardiovascular events).
CQ4: Is Aortic Valve Replacement (AVR) (Surgical Aortic Valve Replacement [SAVR] or Transcutaneous Aortic Valve Implantation [TAVI]) Recommended Before Non-Cardiac Surgery for Patients Complicated With Severe Aortic Stenosis?
Recommendation
For patients with severe aortic stenosis (AS) who undergo non-cardiac surgeries, it is suggested that the non-cardiac surgery is performed without preoperative prophylactic AVR, either surgical AVR or TAVI, if it is asymptomatic. (GRADE 2D) (Strength of recommendation: Weak, Certainty of evidence: Extremely low).
Practical Considerations
In the non-perioperative setting, symptomatic and asymptomatic AS with decreased left ventricular ejection fraction, is an indication for AVR (SAVR, TAVI). However, for these patients, it is difficult to make a general recommendation in the perioperative period for non-cardiac surgery. Preoperative balloon aortic valvuloplasty can be an option in some unstable cases. It is important to comprehensively discuss the indications and the order for interventions to the aortic valve and non-cardiac surgery by the multidisciplinary team and understand the patient’s values and preferences for shared decision-making.
A. Background and Clinical Significance
Reflecting the current super-aging society, the prevalence of AS is increasing. As high as 12% of individuals aged ≥75 years are reported to have moderate or severe AS,332,419 so it is highly possible that a non-cardiac surgery patient is complicated with AS.
Patients with symptomatic severe AS have a high mortality risk in the non-perioperative setting and AVR is recommended.18 However, the recommendation varies for asymptomatic severe AS patients.
For patients with severe AS who need non-cardiac surgery, clinicians must decide whether to perform prophylactic AVR or non-cardiac surgery first. The following aspects should be addressed. (1) If performing AVR first, the effect on primary disease progression by delaying the non-cardiac surgery (e.g., progression of cancer). (2) If performing the non-cardiac surgery first, the risks in delaying AVR especially for symptomatic AS. (3) The risk of death and complications of AVR. (4) TAVI might possibly shorten the delay of non-cardiac surgery.
Several studies have reported that asymptomatic severe AS does not increase perioperative mortality risk.339,340 It is, therefore, important to assess whether (1) the severe AS is associated with increased risk of the non-cardiac surgery or (2) whether preoperative TAVI or SAVR for severe AS will improve the perioperative outcome even after accounting for their risks.
B. Summary of Available Evidence
Due to the lack of studies that have investigated the clinical question directly, the following two key questions (KQ) were formulated.
B.1 PICO or PECO
KQ4-1: Does severe AS increase the perioperative risks of non-cardiac surgery?
P: Non-cardiac surgery patients, E: with severe AS, C: without severe AS
KQ4-2: Does previous AVR (TAVI, SAVR) for severe AS lower the risk of non-cardiac surgery?
P: Non-cardiac surgery patients complicated with severe AS, E: post-AVR (including previously performed AVR), C: No history of AVR
Outcome chosen and approved in panel session is shown in B.2. Its rating of importance was discussed and approved in panel session.
B.2 Outcomes (Rating of Importance)
Desirable outcomes (benefits): 30-day mortality (9), cardiac mortality (9), AMI (8), acute heart failure (AHF) (7), 1-year mortality (8), 1-year cardiac mortality (8), 1-year AMI (7), one-year AHF (7)
Undesirable outcomes (harms): activities of daily living (ADL) 1 year later (9), bleeding (7), days of surgery delay (7)
The reason for selecting ADL 1 year later was because in patients who undergo non-cardiac surgery after undergoing preoperative AVR would be undergoing 2 surgeries, which could lower their functional capacity. We selected bleeding, because antithrombotic therapy is required after AVR and because continuing this in the perioperative period can increase the risk of bleeding but on the other hand discontinuing it can increase the risk of thromboembolism.
B.3 Systematic Review
In a systematic review, 5 retrospective observational studies336–338,341,342 on KQ1 were identified, and a meta-analysis of these studies was performed. There were 2 retrospective observational studies343,344 on KQ2. One (Luis et al., 2020)344 matched their subjects for propensity score, and the other (Taniguchi et al., 2020)343 did not adjust for background factors. Hence, the data from the individual studies were presented rather than performing a meta-analysis.
Adopted studies
KQ4-1: Raymer et al., 1998,341 Calleja et al., 2010,336 Agarwal et al., 2013,337 Tashiro et al., 2014,338 MacIntyre et al., 2018342
KQ4-2: Luis et al., 2020,344 Taniguchi et al., 2020.343
C. Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
In the panel session it was decided to study asymptomatic and symptomatic AS separately because they involve different risks.
C.1 Desirable Effects (Benefits) of AVR in Asymptomatic AS Patients
C.1.1 KQ4-1 (See Table 60 for Summary of Findings) (Figure 14)
Table 60. Summary of Findings on the Influence of Severe Aortic Stenosis (AS) on Perioperative Outcomes for Non-Cardiac Surgeries
| |
Design
No. of research
No. of patients |
No severe
AS*
(control) |
Severe AS**
(95% CI) |
Relative
risk***
(95% CI) |
Absolute risk
difference**
(95% CI) |
Certainty |
Importance
of outcome |
Plain language
summary |
| Asymptomatic severe AS |
30-day
mortality |
Observational
4
1,765 |
1.8% |
2.6%
(1.2–5.6%) |
1.44
(0.68–3.06) |
0.8% higher
(0.58% lower to
3.7% higher) |
C1 |
9 |
Asymptomatic
severe AS does
not increase
mortality9 |
30-day
cardiac death
(age matched) |
Observational
2
496 |
0.63% |
0.5%
(0.1–5.9%) |
0.85
(0.08–9.33) |
0.09% lower
(0.58% lower to
5.3% higher) |
D2 |
9 |
Asymptomatic
severe AS does
not increase
cardiac death10 |
30-day
myocardial
infarction
(age matched) |
Observational
3
1,644 |
1.6% |
2.7
(1.3–5.9%) |
1.66
(0.77–3.60) |
1.1% higher
(0.4% lower to
4.2% higher) |
D3 |
8 |
Asymptomatic
severe AS may
increase MI,
although no
significant
difference11 |
30-day acute
heart failure
(age matched) |
Observational
3
1,644 |
9.6% |
10%
(5.0–20.1%) |
1.04
(0.52–2.09) |
0.38% higher
(4.6% lower to
10.5% higher) |
D4 |
8 |
Asymptomatic
severe AS does
not increase
AHF12 |
| Symptomatic severe AS |
30-day
mortality |
Observational
3
1,522 |
1.8% |
5.9%
(3.0–11.5%) |
3.23
(1.66–6.31) |
4.1% higher
(1.2% higher to
9.7% higher) |
C5 |
9 |
Symptomatic
severe AS
increases
perioperative
mortality |
30-day
cardiac death
(age matched) |
Observational
1
362 |
0.78% |
1.9%
(0.3–13.2%) |
2.42
(0.34–16.92) |
1.11% higher
(0.52% lower to
12.4% higher) |
D6 |
9 |
Symptomatic
severe AS may
increase cardiac
death, although
no significant
difference13 |
30-day
myocardial
infarction
(age matched) |
Observational
2
1,410 |
1.5% |
3.4%
(0.9–13.0%) |
2.22
(0.58–8.42) |
1.9% higher
(0.65% lower to
11.4% higher) |
D7 |
8 |
Symptomatic
severe AS may
increase MI,
although no
significant
difference14 |
30-day acute
heart failure
(age matched) |
Observational
2
1,410 |
10.1% |
19.7%
(5.0–76.5%) |
1.96
(0.5–7.6) |
9.6% higher
(5.0% lower to
66.4% higher) |
D8 |
8 |
Symptomatic
severe AS may
increase AHF,
although no
significant
difference15 |
*Event rate of control group.
**Calculated by event rate of control group and relative risk in the meta-analysis.
***Relative risk calculated in the meta-analysis using Review Manager.
1Certainty judged to be C due to observational studies only.
2–4Certainty downgraded by one to D because of wide range of 95% CI.
5Certainty judged to be C due to observational studies only. Although 95% CI was wide, certainty was not downgraded because the lowest limit of absolute risk difference still showed clinically significant difference.
6–8Certainty downgraded by one to D because of wide range of 95% CI.
9Severe AS did not increase the risk with relative risk 0.87, 0.66, and 1.07 in 3 age-adjusted reports. One report with no background adjustment showed AS significantly increased the risk with relative risk 4.1. However, a meta-analysis of the 4 studies did not show significant increase in the risk; therefore, the panel committee concluded asymptomatic AS did not increase in the risk.
10,11Although there is no significant difference in the risk, the point estimate of the relative risk showed a tendency of clinically significant increase in the risk. Therefore, the panel committee concluded that AS may increase in each risk.
12There was not statistically significant difference in the risk and the point estimate did not show a clinically significant increase in risk. Therefore, the panel committee concluded that AS does not increase in the risk.
13–15Although there is no significant difference in the risk, the point estimate of the relative risk showed a tendency of clinically significant increase in risk. Therefore, the panel committee concluded that AS may increase in each risk.
CI, confidence interval. (Adapted from references 336–338, 341, 342.)
A meta-analysis of the perioperative risks of non-cardiac surgery in asymptomatic AS revealed that the RR of overall mortality (4 observational studies n=1,765),336–338,342 cardiac mortality (2 observational studies, n=496),336,338 and heart failure (3 observational studies, n=1,644)336–338 were 1.44 (95% CI 95% CI 0.68–3.06), 0.85 (95% CI 0.08–9.33), and 1.04 (95% CI 0.52–2.09), respectively, indicating no significant difference. The RR of MI (3 observational studies, n=1,644)336–338 was 1.66 (95% CI 0.77–3.60), indicating no significant difference, but the absolute difference was 1.1% higher (95% CI 0.4% lower to 4.2% higher), suggesting that it could increase risk.
C.1.2 Desirable Effect (Benefits) of Preoperative AVR Before Non-Cardiac Surgery
According to KQ1, asymptomatic severe AS does not increase the risk of all-cause death, cardiac death, and heart failure during the perioperative period, but it may increase the risk of AMI. On the other hand, preoperative AVR may decrease the risk of AMI.
Desirable effects (benefits) are concluded to be subtle.
C.2 Desirable Effects (Benefits) of AVR in Symptomatic AS Patients
C.2.1 KQ1 (See Table 60 for Summary of Findings) (Figure 15)
Symptomatic AS increased overall mortality (based on 3 observational studies, n=1,522, 1.8% vs. 5.9% [95% CI 3–11.5%]), and the absolute difference was 4.1% higher (95% CI 1.2% higher to 9.6% higher).337,338,342 There were no significant differences in cardiac mortality (1 observational study, n=362; 0.78% vs. 1.9% [95% CI 0.3–13.2%]; RR 2.42 [95% CI 0.34–16.92]),338 MI (2 observational studies, n=1,410; 1.5% vs. 3.4% [95% CI 0.9–13.0%]; RR 2.22 [95% CI 0.58–8.42])337,338 and AHF (2 observational studies, n=1,410; 10.1% vs. 19.7% [95% CI 5.0–76.5%]; RR 1.96 [0.5–7.6]).337,338
C.2.2 KQ2
Two retrospective observational studies compared patients who did and did not undergo preoperative AVR before non-cardiac surgery among both asymptomatic and symptomatic AS patients (Taniguchi et al., 2020,343 Luis et al., 2020344). In the panel session, the limitations of the 2 studies were discussed as key points determining the trade-off between benefits and harms of preoperative AVR.
1. The AVR groups of the studies were limited to those who had undergone both AVR and non-cardiac surgery
2. The studies did not include any subjects who could not undergo non-cardiac surgery due to AVR-related death or complications
3. The studies did not include any subjects who could not receive surgical interventions due to progression of the primary disease or death because of delaying the non-cardiac surgery for AVR
4. The studies likely included patients who could not undergo AVR (due to patient-related or institution-related reasons) and had to undergo non-cardiac surgery without treating AS
5. The AVR groups in the studies included subjects who had undergone AVR in the past, which is an indirect bias
The panel session determined the importance of noting that the benefits of preoperative AVR could be overestimated by the above possible biases.
C.2.3 Perioperative Mortality Rate (See Table 61 for Summary of Findings) (Figure 16)
Table 61. Summary of Findings on Perioperative Outcomes in Patients With and Without Aortic Valve Replacement (AVR)
| |
Design
No. of patients |
No AVR*
(control) |
AVR**
(95% CI) |
Relative
risk***
(95% CI) |
Absolute risk
difference**
(95% CI) |
Certainty |
Importance
of outcome |
Plain language
summary |
30-day mortality
rate
(Luis et al, 2020)
Propensity score-
matched analysis |
Observational
266 |
2.3% |
2.3%
(0.48–11.0%)* |
1.0
(0.21–4.87) |
0%
(1.8% lower to
8.7% higher) |
C1 |
C3 |
9 |
AVR does not
decrease
mortality4 |
30-day mortality
rate
(Taniguchi et al,
2020)
Unadjusted |
Observational
348 |
4.3% |
0.30%
(0–5.0%) |
0.07
(0.00–1.17) |
4.0% lower
(4.3% lower to
0.7% higher) |
D2 |
30-day cardiac
death
(Luis et al, 2020)
Unadjusted |
Observational
491 |
2.8% |
0.5%
(0.06–3.9%) |
0.18
(0.02–1.41) |
2.3% lower
(2.7% lower to
1.1% higher) |
D3 |
D5 |
9 |
AVR may
decrease cardiac
death, although
no significant
difference6 |
30-day cardiac
death
(Taniguchi et al,
2020)
Unadjusted |
Observational
348 |
3.2% |
0.29%
(0.03–5.0%) |
0.09
(0.01–1.57) |
2.9% lower
(3.2% lower to
1.8% higher) |
D4 |
30-day
myocardial
infarction
(Luis et al, 2020 )
Propensity score-
matched analysis |
Observational
266 |
3.0% |
0.33%
(0.3–6.1%) |
0.11
(0.01–2.04) |
2.7% lower
(3.0% lower to
3.1% higher) |
D7 |
8 |
AVR may
decrease MI,
although no
significant
difference8 |
30-day acute
heart failure
(Luis et al, 2020)
Propensity score-
matched analysis |
Observational
266 |
15.8% |
3.8%
(1.5–9.7%) |
0.24
(0.09–0.61) |
12% lower
(14.3% lower to
6.2% lower) |
C9 |
8 |
AVR decreased
AHF |
1-year mortality
(Luis et al, 2020)
Propensity score-
matched analysis |
Observational
266 |
24.1% |
12.8%
(7.5–22%) |
0.53
(0.31–0.91) |
11% lower
(16.6% lower to
2.1% lower) |
C10 |
9 |
AVR decreased
1-year mortality |
*Event rate of control group.
**Calculated by event rate of control group and relative risk in the meta-analysis.
***Relative risk calculated in the meta-analysis using Review Manager.
1,9,10Certainty judged to be C because of observational studies with propensity score matching only.
2–4Certainty downgraded to D because there was no background adjustment and a wide range of 95% CI.
5Certainty of body of evidence was judged to be D because both were D.
7Certainty downgraded to D because of the wide range of 95% CI.
4Taniguchi et al, 2020 with no adjustment showed no statistical difference in the risk. Luis et al, 2020, with propensity score matching, showed similar mortality. The panel committee concluded AVR did not decrease the risk.
6Although there was no statistically significant difference (95% CI of relative risk included 1.0), the point estimate was much lower than 1.0. Therefore, the panel committee concluded that after AVR the risk of cardiac death may decrease.
8Although there was no statistically significant difference (95% CI of relative risk included 1.0), the point estimate was much lower than 1.0. Therefore, the panel committee concluded that after AVR the risk of MI may decrease.
**The risk of the intervention group was calculated by the risk of the control group multiplied by the relative risk (95% CI).
**Calculated by Review Manager
1,9,10Observational studies with propensity score matching.
2–4No adjustment of background factors and wide range of 95% CI.
5Both studies are D.
7Irrespective of propensity score matching, the 95% CI was wide.
CI, confidence interval. (Adapted from references 343, 344.)
Propensity score-matching analysis by Luis et al., 2020, indicated no significant difference in perioperative mortality between patients who did not and did undergo AVR (2.3% vs. 2.3% [95% CI 0.48–11.0%]; RR 1.0 [95% CI 0.21–4.87], absolute difference 0% [95% CI 1.8% lower to 8.7% higher]).344
Taniguchi et al., 2020, found the corresponding data to be 4.3% vs. 0.3% (95% CI 0–5%), RR 0.07 (95% CI 0.00–1.17), absolute risk difference 4.0% lower (95% CI 4.3% lower to 0.7% higher).343 However, their findings were biased because their data was unadjusted for background factors; for instance, the patients who did not undergo AVR were older and included more dialysis patients.
As such, it is possible that perioperative mortality may be lower after AVR or no difference between AVR group and no AVR group. In the case of no difference, then the mortality risk of AVR itself translates into a net increase in mortality risk. In conclusion by the panel committee, the effect of preoperative AVR on perioperative mortality rate “varies.”
a. Perioperative Cardiac Mortality
There are no studies that have adjusted background factors to analyze this outcome. Taniguchi et al., 2020, reported 3.2% vs. 0.29% (95% CI 0.03–5.0%) and absolute risk difference 2.9% lower (95% CI 3.2% lower to 1.8% higher) in the AVR group.343 Although there was no significant difference, the panel committee concluded that AVR might possibly decrease this risk.
b. Perioperative MI
Luis et al., 2020, reported 3.0% vs. 0.33% (95% CI 0.3–6.1%) in the no AVR group and AVR group, respectively.344 Absolute risk difference was 2.7% lower (95% CI 3.0% lower to 3.1% higher) in the AVR group on propensity score-matched data. Although the difference was not significant, the report suggested that AVR might decrease the risk of MI.
c. Perioperative HF
Luis et al., 2020, reported 15.8% vs. 3.8% (95% CI 1.5–9.7%), absolute risk difference 12% lower (95% CI 6% lower to 14% lower) in the AVR group in their propensity score-matched data,344 indicating AVR significantly decreased this risk.
d. 1-Year Mortality Rate
Luis et al., 2020, reported 24.1% vs. 12.8% (95% CI 7.5–22%) and absolute risk difference 11% lower (95% CI 2% lower to 17% lower) in AVR group in their propensity score-matched data,344 indicating AVR significantly decreased this risk.
C.2.4 Desirable Effects of AVR
The panel committee concluded that desirable effects by AVR “vary,” considering the effects on overall mortality.
C.3 Undesirable Effects (Harms) of Preoperative AVR
According to the systematic review, there were no studies that investigated the outcomes of ADL 1 year later (9), bleeding (7), days of surgery delay (7) (number in parentheses denotes the importance of the outcome).
Additional Considerations
a. Days of Delay of Non-Cardiac Surgery
This is a key concern for urgent and time-sensitive procedures (e.g., malignancy). In patients diagnosed with AS and who underwent prophylactic AVR preoperatively, the median days from AVR to non-cardiac surgery was 105.344 Although not a finding from a systematic review, in 1 study the median duration from TAVI to non-cardiac surgery was 28.0 days (interquartile range 16.8–54.5 days), and the median duration from balloon aortic valvuloplasty (BAV) to non-cardiac surgery was 11 days (interquartile range 4–18 days).345
b. Complications of AVR
When prophylactic AVR is performed, the complications of AVR also should be considered. The rates of mortality and complications in SAVR are predicted by the Society of Thoracic Surgeons (STS) score, JapanSCORE or EuroSCORE II.
There were 2 Japanese studies of TAVI. The study of 1,613 patients reported a 1.7% 30-day mortality rate, 1.7% with disabling stroke, 0.8% acute coronary syndrome, 5.8% critical bleeding, tamponade 1.6%, and pacemaker 8.5%.420 The other study of 299 subjects reported 1.3% mortality rate, 2.3% brain infarction rate, 5.4% pacemaker, and 2.3% critical bleeding.421
c. Bleeding and Thrombosis
AVR requires antithrombotic therapy for at least the first 3 months postoperatively, whether SAVR (bioprosthetic valve) or TAVI,18 so non-cardiac surgery in this period potentially involves harm of bleeding by continuing antithrombotic agents or thrombosis by discontinuing it. With regard to the non-perioperative settings, the risk of brain infarction and systemic embolism without antithrombotic therapy is as high as 7% and 13%, respectively, in the 3 months after bioprosthetic valve surgery.187 Furthermore, hypercoagulability is expected in the perioperative period.
d. No Evidence Related to ADL
C.4 Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms) of Preoperative AVR for Asymptomatic AS Patients
As mentioned above, there is potential benefit of preoperative AVR in reducing AMI (desirable effect determined to be “a little”). However, considering the risk of AVR itself, and the risk of delaying the non-cardiac surgery (undesirable effects determined to be “moderate”), preoperative AVR likely has more harms than benefits.
C.5 Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms) of Preoperative AVR for Symptomatic AS Patients
According to a valvular heart disease clinical practice guidelines,18 symptomatic severe AS is a Class I indication for AVR in the non-perioperative setting, but preoperative AVR can potentially increase or reduce the risk of mortality rate, as explained above. Although the risk of AHF and 1-year mortality decreased, the trade-off of the benefits and harms of preoperative AVR was considered to “vary”, given its effects on the risk of perioperative mortality.
D. Certainty of Body of Evidence
The certainty of body of evidence of the various outcomes was C or D. Due to the high indirectness, the certainty of evidence was graded as extremely low (D).
E. Values and Preferences
According to the systematic review, there were no studies that investigated patients’ perioperative values and preferences.
Additional Considerations
The following opinions were stated in the panel session.
Perioperative mortality, cardiac mortality, MI, and HF are important outcomes. Because TAVI and SAVR are also associated with serious complications, patients are likely to value a total risk assessment, including the risk of AVR. Furthermore, the prognosis of both the primary disease and AS are important for the patient.
F. Cost
We could not identify any studies that investigated the cost–benefit from the perspectives of patients. The out-of-pocket costs of patients are estimated below.
Additional Considerations
The hospitalization costs of TAVI and SAVR are JPY6.8 million and JPY5.9 million, respectively.422 The High-Cost Medical Expensive Benefit system in Japan places a cap on out-of-pocket medical fees.407 For example, for a patient with an annual income of ≈JPY3.7–7.7 million, the maximum out-of-pocket fee is calculated as:
JPY80,100 + (medical fees − 267,000) × 1%
That is, the maximum out-of-pocket costs for TAVR and SAVR are JPY145,000 and JPY139,000, respectively, which was considered to be a “moderate cost” for patients.
G. Acceptability
There is no research evidence investigating this issue.
Additional Considerations
The acceptability for patients of performing AVR prior to non-cardiac surgery is expected to be influenced by the risks of delaying the surgery for the primary disease and by whether the patient has symptoms of AS. For the physician (surgeon, anesthesiologist), successful AVR eliminates the concerns of the risks by AS, so performing AVR might increase acceptability. At the same time, other clinical practice guidelines have already recommended that AVR does not necessarily need to be performed preoperatively if AS is asymptomatic,18,199,347 suggesting the acceptability of foregoing AVR. The acceptability is considered to “vary”.
H. Feasibility
SAVR and TAVI require specialized institutions. Non-cardiac surgery without AVR requires strict hemodynamic management; thus, it is ideal to perform the surgery at a specialized institution that can provide treatment in case of hemodynamic instability. Neither option can be performed in a non-specialized institution.
I. Grading Recommendation
I.1 Panel Session
There were no RCTs that investigated the effects of preoperative AVR on perioperative outcomes, including complications of AVR. Consequently, recommendations had to be based on indirect evidence. Therefore, reaching a consensus required a significant amount of discussion.
I.2 Asymptomatic AS
A consensus was reached that non-cardiac surgery should be performed without prophylactic AVR (SAVR, TAVI) for asymptomatic AS, then proceeded to voting.
I.3 Symptomatic AS
For the above reasons, it was determined that it is difficult to make general recommendations. AVR is recommended for symptomatic AS in non-perioperative settings. In perioperative cases, the options are to (1) perform AVR (SAVR, TAVI) first, with BAV also an option, and then to perform non-cardiac surgery; (2) perform the non-cardiac surgery first, and then perform AVR (SAVR, TAVI); (3) perform either of the surgeries only; or (4) perform neither surgery. The panel committee concluded that (1) the urgency of valve replacement (risk of delaying AVR), (2) urgency of the non-cardiac surgery (harms of delaying the non-cardiac surgery), (3) risk of AVR or BAV, (4) risk of non-cardiac surgery without AVR, and (5) risk of thrombosis or bleeding by discontinuing or continuing thrombotic agents initiated after valve replacement should be discussed comprehensively by a multidisciplinary team consisting of a surgeon, anesthesiologist and cardiologist. The panel committee highlighted the importance of understanding the patient’s values and preferences for shared decision-making. If the risk of AVR is low and the non-cardiac surgery can be delayed sufficiently, then AVR is performed first. Even in cases of symptomatic AS, if the patient is hemodynamically stable, and at high-risk for AVR, and if the non-cardiac surgery cannot be delayed (e.g., femoral neck fracture discussed in CQ5), then the non-cardiac surgery may be performed first.
I.4 After the Panel Session
The recommendation was approved with 100% vote rate, median 8.0, disagreement index 0.192, and 92% agreement rate by modified Delphi method (RAND appropriateness method).
J. Recommendation by Other Related Clinical Practice Guidelines
See Part 1 Chapter IV, 8.5.1 Interventions for severe AS to non-cardiac surgery.
K. Monitoring Risks After Non-Cardiac Surgery
Particularly in symptomatic patients who will not undergo preoperative AVR, hemodynamics should be stabilized preoperatively as much as possible. Preload and afterload fluctuations should be avoided intraoperatively and postoperatively. Onset of atrial fibrillation can also destabilize hemodynamics, so strict monitoring and early interventions are needed. Until stabilization is confirmed, postoperative management in a high-care unit or intensive care unit is desirable.
L. Surveillance and Evaluation of the Recommendation
If AVR precedes non-cardiac surgery: the rates of mortality and complications of AVR, the rate of being unable to perform the non-cardiac surgery due to adverse events of AVR, and the rates of bleeding and thrombotic complications.
If non-cardiac surgery precedes AVR: the short- and long-term postoperative mortality rates, cardiac mortality rate, incidence of MI and HF, length of hospital stay, and prognosis of the primary disease.
M. Future Research Directions
A RCT of patients assigned to preoperative AVR and no preoperative AVR is necessary.
If the non-cardiac surgery is for a malignancy, then not only the perioperative prognosis, but long-term (several years) prognosis should be investigated.
CQ5: Is Preoperative Aortic Valve Replacement (AVR) (Surgical Aortic Valve Replacement [SAVR] or Transcatheter Aortic Valve Implantation [TAVI]) Recommended Before Surgeries for Femoral Neck or Trochanteric Fracture in Older Patients Complicated With Severe Aortic Stenosis (AS)?
Recommendation
Performing surgery for the fracture under strict management of hemodynamics* without performing AVR (SAVR, TAVI) is suggested in older-aged patients with hip fracture complicated by severe AS if patients are hemodynamically stable. (Grade 2D) (Strength of recommendation: Weak, Certainty of evidence: Extremely low).
Practical Considerations
If hemodynamic stabilization is not possible, preoperative TAVI or balloon aortic valvuloplasty are options for the patient.
*Strict hemodynamic management includes the following:
(1) Pre-, intra-, and postoperative multidisciplinary management by a team of experts in anesthesiology, cardiovascular medicine, and orthopedic surgery
(2) Management in an appropriate place where strict hemodynamic monitoring is possible as needed, such as high-care unit or intensive care unit (ICU), particularly for the postoperative period until stability of hemodynamics is confirmed
A. Background and Clinical Significance
Due to Japan’s super-aging population, the incidence of femoral neck and trochanteric fracture is reported to be 170,000 per year and continues to increase.423 In a foreign study, 2% of patients with these fractures were reported to be complicated with severe AS.424 This corresponds to 3400 individuals who will encounter this problem each year in Japan.
Early surgery (within 48 h of admission) improves various outcomes, including mortality, independent living, pressure ulcers, and major or minor complications.23,425 Furthermore, hip fracture is associated with severe pain and early surgery has analgesic benefits.426,427 The UK Clinical Practice Guidelines recommend surgery on the day of admission or the next day.23 Other guidelines in the USA,428 Canada,429 and Australia430 recommended early surgery. Japanese clinical practice guidelines also recommend surgery as early as possible (Japan Orthopedic Association Clinical Practice Guideline 2021).431
AVR itself can be a significant risk in older patients (e.g., an 83-year-old woman, [average age of Japanese hip fracture patient], 155 cm, 45 kg, SAVR STS score,432 mortality 6.5%, morbidity and mortality 18%). TAVI also has risk (K TAVI Registry: 30-day mortality rate 1.3%, stroke 3.8%, OCEAN TAVI registry: mortality rate 1.7%, disabling stroke 1.7%).420,421
It is important to assess whether AS is a perioperative risk, and whether AVR should be performed preoperatively for this subgroup. The reasons for treating hip fracture surgery separately from other surgeries are advanced patient age and urgency of the procedure; it was considered important to make specific recommendations for this type of surgery.
B. Summary of Available Evidence
B.1 PICO or PECO
There are no studies that directly answered CQ5; thus, an analytic framework was made to reformulate it into 2 separate key questions (KQ).
KQ5-1: Does severe AS increase the perioperative risks of femoral neck/trochanteric fracture surgery?
PECO: P: Older femoral neck/trochanteric fracture patients, E: with severe AS, C: without AS
KQ5-2: Does performing AVR (SAVR, TAVI) for patients undergoing femoral neck/trochanteric fracture surgery decrease the perioperative risk?
PECO: P: Older femoral neck/trochanteric fracture patients with severe AS, E: post-AVR C: no history of AVR Outcome chosen and approved in panel session is shown in B.2. Its rating of importance was discussed and approved in panel session.
B.2 Outcomes (Rating of Importance)
Desirable outcomes: 30-day mortality (9), cardiac mortality (9), heart failure (7), MI (7), 1-year mortality rate (9)
Undesirable outcomes: pain (8), delirium (8), pressure ulcer (8), activities of daily living (ADL) independency (9), infection (8), walking prognosis (8), bleeding, complications (perioperative, 1-year) (7), days of non-cardiac surgery delay (7)
Furthermore, background knowledge was summarized as a Background question (BQ): When is the optimal timing for femoral neck/trochanteric fracture surgery?
B.3 Systematic Review
B.3.1 Studies Adopted in KQ5-1 and KQ 5-2
In a systematic review, three retrospective observational studies on the topic were identified: Leibowitz et al., 2009,433 McBrien et al., 2009,434 Rostagno et al., 2019.435
KQ2: We discussed whether KQ4-2 in CQ4 could be used, but the panel session concluded that it could not be used because of the substantial time taken between AVR and surgery in the studies and the fact that the subjects were not restricted to older adults with hip fracture.
B.3.2 Summary of BQ
In general, delaying surgery for hip fracture increases the mortality rate, lowers functional capacity, and increases the risk of severe pain, delirium, pressure sores and infections in the acute phase. The Japan Orthopedic Association Clinical Practice Guidelines emphasize the importance of early surgical intervention.431 Indeed, surgery is performed in a median of 3 days in Japan.
B.3.3 KQ5-1
Presence of severe AS does not necessary increase the risk of mortality or myocardial infarction (MI) (D), but may increase the risk of heart failure (D).
B.3.4 KQ5-2
There is no research evidence.
C. Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
C.1 Desirable Effects (Benefits) of Preoperative AVR
C.1.1 KQ1 (See Table 62 for a Summary of Findings)
Table 62. Summary of Findings on the Influence of Severe Aortic Stenosis (AS) on Perioperative Outcomes for Hip Fracture Surgery
| |
Design
No. of research
No. of patients |
No AS*
(control) |
Severe AS**
(95% CI) |
Relative
risk***
(95% CI) |
Absolute risk
difference**
(95% CI) |
Certainty |
Importance
of outcome |
Plain language
summary |
30-day
mortality |
Observational
3
Severe AS group:
94
No AS group:
4,069 |
7.1%* |
13.1%
(5.6–30.4%) |
1.84
(0.79–4.27) |
6.0% higher
(1.5% lower to
23.3% higher) |
D1 |
D3 |
9 |
Severe AS does
not increase
mortality4 |
Mortality in
Japan 1%** |
1.84
(0.79–4.27%) |
0.84% higher
(0.21% lower to
3.3% higher) |
30-day
mortality
(age matched) |
Observational
1
Severe AS group:
32
No AS group:
88 |
6.8%* |
6.3%
(1.3–29%) |
0.92
(0.19–4.31) |
0.55% lower
(5.5% lower to
22% higher) |
D2 |
9 |
Mortality in
Japan 1%** |
0.92%
(0.19–4.3%) |
0.08% lower
(0.81% lower to
3.3% higher) |
30-day
myocardial
infarction |
Observational
2
Severe AS group:
64
No AS group:
371 |
1.3%* |
7.4%
(0.75–73%) |
5.51
(0.56–54.45) |
6% higher
(0.6% lower to
72% higher) |
D5 |
D7 |
8 |
Severe AS does
not increase MI8 |
30-day
myocardial
infarction
(age matched) |
Observational
1
Severe AS group:
32
No AS group:
88 |
2.3%* |
3.1%
(0.3–33%) |
1.38
(0.13–14.65) |
0.86% higher
(2% lower to
31% higher) |
D6 |
8 |
30-day acute
heart failure |
Observational
2
Severe AS group:
64
No AS group:
371 |
1.1% |
10.7%
(2.1–55.3) |
9.91
(1.92–51.30) |
9.6% higher
(0.99% higher to
54.2% higher) |
D9 |
D11 |
8 |
Severe AS might
possibly increase
acute heart
failure although
no significant
difference12 |
30-day acute
heart failure
(age matched) |
Observational
1
Severe AS group:
32
No AS group:
88 |
2.3%* |
9.4%
(1.6–53.6%) |
4.13
(0.7–23.57) |
7.1% higher
(0.64% lower to
51.3% higher) |
D10 |
*Event rate of control group.
**Calculated by event rate of control group and relative risk in the meta-analysis.
***Relative risk calculated in the meta-analysis using Review Manager.
1,2Certainty downgraded to D due to small sample size in AS group with low event rate and wide range of 95% CI of relative risk.
3Certainty judged to be D because both were D.
5,6Certainty downgraded to D due to small sample size in AS group with low event rate and wide range of 95% CI of relative risk.
4Certainty judged to be D because both were D.
5Because there was no statistically significant difference and the point estimate was close to 1.0, panel committee concluded that AS does not increase myocardial infarction.
9,10Certainty downgraded to D due to small sample size in AS group with low event rate and wide range of 95% CI of relative risk.
11Certainty judged to be D because both were D.
12Although the difference was not statistically different, the point estimate of relative risk was much higher than 1.0 (4.1), panel committee concluded that AS may increase the risk of heart failure.
CI, confidence interval. (Adapted from references 433–435.)
According to the systematic review, there were 3 retrospective observational studies comparing patients with severe AS and without AS.433–435 The analysis of 2 studies were not age-adjusted, and in fact the patients with AS had advanced age (McBrien et al., 2009:434 age of patients with AS=86 years, patients without AS=78 years; Rostagno et al., 2019:435 87 vs. 83 years). Patients were age-adjusted in 1 study (Leibowitz et al., 2010433). None of the studies analyzed the symptomatic and asymptomatic AS separately.
a. 30-Day Mortality (Figure 17)
In the meta-analysis of 3 observational studies (AS group n=94, non-AS group=4,069), the RR was 1.84 (95% CI: 0.79–4.27), showing no significant difference.433–435 In the single age-adjusted study (n=120), the RR was 0.92 (95% CI 0.19–4.31), again showing no significant difference.433 Therefore, the panel concluded that severe AS does not increase mortality.
b. 30-Day AMI (Figure 17)
This was analyzed in 2 retrospective observational studies (n=435).433,435 In the meta-analysis, RR was 5.51 (95% CI 0.56–54.54). In the 1 age-matched study (n=120), RR was 1.38 (95% CI 0.13–14.7), again without significant difference.433 The panel concluded that severe AS does not increase the risk of AMI.
c. 30-Day AHF (Figure 17)
A meta-analysis of 2 retrospective studies (n=435) found the RR of patients with AHF among those with AS to be 9.91 (95% CI 1.92–51.30).433,435 The RR in the age-adjusted study (n=120) was 4.13 (95% CI 0.72–23.6).433 Although there was no statistical significance, the point estimate of RR was very high, so the panel concluded that severe AS may increase the risk of AHF.
d. Answer to KQ1
In summary, severe AS does not increase risk of perioperative death or AMI in patients undergoing surgery for femoral neck/trochanteric fractures, but may increase the risk of AHF (Certainty of body of evidence: D, D, D, respectively).
C.1.2 Desirable Effects (Benefits) of Preoperative AVR (Benefits)
KQ 5-1 suggests that AS increases the risk of AHF after surgery for hip fracture (importance of outcome: 7). Although there is no research or evidence, successful AVR might possibly lower this risk (a benefit); thus, desirable effects were considered to be “minor”.
C.2 Undesirable Effects of Preoperative AVR (Harms)
The outcomes related to harms (importance of outcomes)
Pain (8), delirium (8), pressure ulcers (8), ADL independence (9), infection (8), walking prognosis (8), bleeding complications (perioperative period, 1 year) (7), days of delaying non-cardiac surgery (7). The systematic review found no studies that investigated the effects of preoperative AVR on these outcomes.
C.3 Additional Considerations
C.3.1 Number of Days and Harms of Delaying Surgery
Performing TAVI is expected to delay surgery for at least several days. In 1 report, the median duration from TAVI to non-cardiac surgery was 28.0 days (quartiles 16.8–54.5 days),345 and that from balloon aortic valvotomy (BAV) to non-cardiac surgery was 11 days (quartiles 4, 18 days).345
Median time from AVR to surgery in patients preoperatively diagnosed with AS and who underwent prophylactic AVR was 105 days.344
As explained in the Background Question, delaying surgery increases the risks of death, worsening pain, infection, pressure sores, delirium, decreased functional capacity, and becoming dependent in ADL.
C.3.2 Risks of AVR Itself
a. Risks of TAVI
We searched Japanese data on TAVI and found 2 studies.420,421 One was based on the data of 1,613 patients, and reported a 30-day mortality rate 1.7%, disabling stroke 1.7%, acute coronary syndrome 0.8%, critical bleeding 5.8%, tamponade 1.6%, and pacemaker implantation 8.5%; moreover, the length of hospital stay was 11 days.420 The other analyzed the data of 299 patients, and reported the following rates: mortality 1.3%, brain infarction 2.3%, pacemaker implantation 5.4%, and critical bleeding 2.3%.421
b. Risks of SAVR
The risks of AVR for Japanese women of age 83, which was mean age of hip fracture in our country, were estimated by STS score,432 the mortality rate, and mortality and morbidity rate were 6.5% and 18%, respectively.
C.3.3 Bleeding
The issues of perioperative bleeding and thromboembolic risk associated with antithrombotic agents after AVR are the same as discussed in CQ4.
C.3.4 Undesirable Effects of Preoperative AVR
Undesirable effects were judged to be “moderate”.
C.4 Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
As mentioned above, there is potential benefit of preoperative AVR in reducing AHF (desirable effect determined to be “minor”). However, considering the risk of AVR itself, and the risk of delaying the hip fracture surgery (undesirable effect determined to be “moderate”), preoperative AVR likely has more harms than benefits.
C.4.1
Thus, performing preoperative AVR for older-aged hip fracture patients likely has few benefits and many harms, and the control (no preoperative AVR) is likely to be superior.
D. Certainty of Body of Evidence
These were retrospective observational studies of small samples; thus, the certainty of evidence was judged to be extremely low (D).
E. Values and Preferences
According to the systematic review, there were no studies on values related to the priority of the outcomes.
Additional Considerations
Approximately 25% of older adults who have femoral fractures lose the ability to walk and become unable to live independently, and many move into nursing facilities.436
An Australian scenario-based study assessed QOL as lowered when people become dependent in ADL due to a femoral fracture, and that 80% of the individuals evaluated this to be worse than dying.30
We judged that there were few serious variations in values.
F. Cost
We could not identify any studies performing a cost–benefit analysis from the perspectives of patients according to the systematic review. See CQ4 for the estimation of out-of-pocket inpatient cost for AVR.
G. Acceptability
There is no research evidence. As additional considerations, panel session members discussed that (1) fracture patients complain of severe pain; thus, the acceptability of performing AVR before surgery for the fracture would be low; (2) the acceptability of AVR was estimated to be particularly low for asymptomatic AS; (3) some patients have dementia, in which case the family will be the proxy decision-maker, but considering the pain and functional prognosis, the acceptability of performing preoperative AVR was also considered to be low for the family; (4) for surgeons and anesthesiologists, eliminating concerns of severe AS-related risk by preoperative AVR would increase their acceptability. Thus, the acceptability of performing preoperative AVR was determined to be low for patients and families, and to be “vary” for physicians.
H. Feasibility
AVR and TAVI require specialized institutions. Non-cardiac surgery without AVR requires strict hemodynamic management, and it is ideal to perform surgery at a specialized institution that can provide treatment in case of hemodynamic instability. Thus, the feasibility of neither option is low.
I. Grading Recommendation
I.1 Panel Session
AS may increase the risk of AHF, so there are concerns that it might be difficult to recommend against preoperative AVR. At the same time, delaying surgery increases the risk of death, severe pain, decline of functional capacity, and delirium, etc., thus, early surgery is ideal. The panel concluded that the outcomes of death, pain, and dysfunction are more important than prevention of AHF.
The recommendation was approved with 100% voting rate, median 8.0, disagreement index 0.192, and agreement rate 100% by modified Delphi method (RAND appropriate method).
I.2 Public Comment
1. There were suggestions to clearly state the ICU as the place for postoperative strict hemodynamic management. This was included under practical considerations on “strict hemodynamic management” in the recommendation
2. The importance of considering the indications for hip fracture surgery in severe AS patients by a multidisciplinary team was mentioned
Relieving pain, improving function (independently rolling on the bed, sitting, walking), pain during care (pain on transportation and elimination care), improvement of mortality rate and prevention of pressure ulcers and infection were some of the benefits expected with early surgery for fracture. This could mean that there are no benefits to surgery when patients do not have pain, and when improvement of function cannot be expected by surgery because of being too frail. Discussion by a multidisciplinary team about the goals anticipated by the patients and their families (caregivers), and whether the goals can be achieved is important. However, writing a recommendation on this is not within the scope of the present clinical practice guidelines, and will not be discussed beyond this mention here.
J. Recommendation by Other Related Clinical Practice Guidelines
Clinical practice guidelines related to recommendations on severe AS for older hip fracture patients were not found.
K. Monitoring Risks After Non-Cardiac Surgery
Hemodynamics should be stabilized preoperatively as much as possible when patients are not undergoing preoperative AVR. The patient should be strictly monitored preoperatively, intraoperatively, and postoperatively, and preload and afterload fluctuations should be avoided. Onset of atrial fibrillation can also destabilize hemodynamics. Strict monitoring and early interventions are needed.
L. Surveillance and Evaluation of the Recommendation
Perioperative incidence of hemodynamic instability, heart failure, and cardiac death should be monitored and evaluated.
M. Future Research Directions
1. Research to date has consisted of small-scale, retrospective studies, and thus has lower levels of certainty. Whether severe AS is a perioperative risk of hip fracture surgery should be investigated in a large-scale prospective observational study
2. Japanese patients and families’ values should be studied
3. An RCT testing the effects of preoperative TAVI (e.g., P: older hip fracture patients, I: TAVI, C: no TAVI, O: mortality rate, MI rate, AHF rate, functional capacity, walking prognosis, ADL independence, delirium, pain, infection, pressure sores, QOL, days of surgery delay) is needed. However, patients are of advanced age and often dealing with severe pain, so the feasibility may be low.
CQ6: Is Preoperative Heparin Bridging Recommended for Patients Taking Warfarin for Atrial Fibrillation Undergoing Non-Cardiac Surgery?
Recommendation
For patients on warfarin for atrial fibrillation (AF) undergoing non-cardiac surgery, if the risk of thrombosis is not high (CHADS2
score ≤4 points), it is recommended not to perform preoperative heparin bridging (GRADE 2B) (Strength of recommendation: Weak, Certainty of evidence: Moderate).
Practical Considerations
In patients with CHADS2
score ≥5 points (at high thromboembolic risk), heparin bridging should be considered if the risk of bleeding is not high. Even patients with CHADS2
score ≤4 points, heparin bridging can be considered in patients with other risk of thrombosis (e.g., history of recurrent embolism).
A. Background and Clinical Significance
Approximately 15–20% of patients on anticoagulant therapy for AF or other conditions undergo invasive procedures and surgeries.437 There are some concerns that discontinuing anticoagulant therapy for invasive procedures and surgical interventions is associated with a risk of thrombosis, and continuing the therapy is associated with risk of bleeding,437 Heparin bridging is the administering of heparin while discontinuing warfarin preoperatively to minimize the duration of withholding anticoagulant therapy. This is done to reduce the risk of perioperative embolic events, but anticoagulant therapy until immediately before surgery is a concern because it increases the risk of perioperative hemorrhagic complications. It is important to investigate whether heparin bridging should be given to patients on warfarin for AF who need to undergo non-cardiac surgery.
B. Summary of Available Evidence
B.1 PICO or PECO
P: Patients on warfarin for AF undergoing non-cardiac surgery that requires discontinuing of warfarin, I or E: use heparin bridging, C: not use heparin bridging
Outcome chosen and approved in panel session is shown in B.2. Its rating of importance was discussed and approved in panel session.
B.2 Outcomes (Rating of Importance)
Desirable outcomes: perioperative brain infarction (9), systemic embolism (8), mortality (8), cardiovascular event (8)
Undesirable outcomes: major bleeding (9), minor bleeding (7)
B.3 Systematic Review
According to the systematic review, there was 1 RCT (Douketis et al., 2015)191 and 7 observational studies (Wysokinski et al., 2008,195 Ahmed et al., 2010,198 Billett et al., 2010,197 Smoyer-Tomic et al., 2012,196 Douketis et al., 2015,192 Steinberg et al., 2015,194 Sjögren et al., 2017193) identified.
C. Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
C.1 Desirable Effects (Benefits) (See Table 63 for Summary of Findings)
Table 63. Summary of the Findings on the Desirable Effects of Heparin Bridging Among Patients on Warfarin for Atrial Fibrillation Undergoing Non-Cardiac Surgery
| |
Design
No. of research
No. of patients |
No heparin
bridge
(control)* |
Heparin
bridge**
(95% CI) |
Relative
risk***
(95% CI) |
Absolute risk
difference**
(95% CI) |
Certainty |
Importance
of outcome |
Plain language
summary |
| Mortality |
RCT
1
1,813 |
0.54% |
0.45%
(0.12–1.7%) |
0.82
(0.22–3.05) |
0.1% lower
(0.4% lower to
1.1% higher) |
B1 |
B3 |
8 |
Heparin bridge
does not
decrease
mortality |
Observational
4
12,308 |
4.74% |
2.9%
(0.5–19%) |
0.62
(0.1–4.04) |
1.8% lower
(4.3% lower to
14.4% higher) |
C2 |
| 0.2%**** |
0.12%
(0.02–0.81%) |
0.08% lower
(0.18% lower to
0.6% higher) |
| Brain infarction |
RCT
1
1,813 |
0.21% |
0.34%
(0.06–2.0%) |
1.54
(0.26–9.19) |
0.12% higher
(0.16% lower to
1.78% higher) |
B1 |
B4 |
9 |
Heparin bridge
does not
decrease cerebral
infarction |
Observational
3
5,420 |
0.70% |
0.34%
(0.08–1.4%) |
0.49
(0.12–1.98) |
0.36% lower
(0.62% lower to
0.69% higher) |
C2 |
Systemic
embolism |
RCT
1
1,813 |
0% |
0% |
|
|
B1 |
B4 |
8 |
Heparin bridge
does not
decrease
systemic
embolism |
Cardiovascular
event |
RCT
1
1,813 |
1.20% |
1.90%
(0.9–4.0%) |
1.59
(0.75–3.37) |
0.71% higher
(0.3% lower to
2.84% higher) |
B1 |
B4 |
8 |
Heparin bridge
does not
decrease
cardiovascular
events |
Observational
2
2,613 |
2.36% |
3.5%
(1.4–9.3%) |
1.5
(0.57–3.95) |
1.2% higher
(1.0% lower to
7.0% higher) |
C2 |
*Event rate of control group.
**Calculated by event rate of control group and relative risk in the meta-analysis.
***Relative risk calculated in the meta-analysis using Review Manager.
****Mortality rates in 4 cohort studies were 8.5%, 1.4%, 0.2%, 0%, respectively. As representative of the low-risk group, the second lowest 0.2% was adopted.
1Certainty downgraded by one to B due to indirectness because there was only one RCT.
2Certainty judged to be C due to observational studies only.
3Certainty judged to be B because both RCT and observational studies had similar tendency in the result and the certainty of the RCT, B, was adopted.
4Certainty judged to be B because both RCT and observational studies had similar tendency in the result and the certainty of the RCT, B, was adopted.
CI, confidence interval; RCT, randomized controlled trial. (Adapted from references 191–198.)
C.1.1 Mortality (Figure 18)
Mortality was investigated in 1 RCT191 and 4 observational studies.193,194,196,197 In the RCT (n=1,813), the mortality rate in patients who did and did not undergo heparin bridging was 0.54% vs. 0.45% (95% CI 0.12–1.7%), and according to the meta-analysis of observational studies (n=12,308), the corresponding data were 4.7% vs. 2.9% (95% CI 0.5–19%); the differences were not significant.
C.1.2 Brain Infarction (Figure 18)
The RCT191 and 3 observational studies195,197,198 were reviewed. In the RCT (n=1,813), brain infarction in patients who underwent and did not undergo heparin bridging occurred at 0.21% vs. 0.34% (95% CI 0.06–2.0%), and according to the meta-analysis of the observational studies (n=5,420), the corresponding data were 0.7% vs. 0.34% (95% CI 0.08–1.4%); the differences were not significant.
C.1.3 Systemic Thrombosis
This was reviewed in the RCT only and the rates were 0% vs. 0% (i.e., no significant difference).191
C.1.4 Cardiovascular Events (Figure 18)
Cardiovascular events were investigated in the RCT191 and 2 observational studies.194,195 In the RCT (n=1,813), the rates for patients who did and did not undergo heparin bridging were 1.20% vs. 1.90% (95% CI 0.9–4.0%), and according to the meta-analysis of the observational studies (n=2,613), the corresponding data were 2.36% vs. 3.5% (95% CI 1.4–9.3%); there was no significant difference.
C.1.5 Desirable Effects (Benefit)
We determined that there were no desirable effects of heparin bridging.
C.2 Undesirable Effects (Harms) (See Table 64 for Summary of Findings)
Table 64. Summary of Findings on the Undesirable Effects of Heparin Bridging Among Patients on Warfarin for Atrial Fibrillation Undergoing Non-Cardiac Surgery
| |
Design
No. of research
No. of patients |
No heparin
bridge*
(control) |
Heparin
bridge**
(95% CI) |
Relative
risk***
(95% CI) |
Absolute risk
difference**
(95% CI) |
Certainty |
Importance
of outcome |
Plain language
summary |
Major
bleeding |
RCT
1
1,813 |
1.31% |
3.24%
(1.7–6.3%) |
2.48
(1.28–4.83) |
1.93% higher
(0.37% higher to
5.01% higher) |
B1 |
B3 |
9 |
Heparin bridge
increases major
bleeding |
Observational
3
4,081 |
1.34% |
4.3%
(2.5–7.2%) |
3.17
(1.89–5.34) |
2.91% higher
(1.19% higher to
5.8% higher) |
C2 |
Minor
bleeding |
RCT
1
1,813 |
12.0% |
20.9%
(16.8–26.0%) |
1.74
(1.4–2.17) |
8.9% higher
(4.8% higher to
14.0% higher) |
B1 |
B4 |
7 |
Heparin bridge
increases minor
bleeding |
Observational
1
386 |
1.09% |
4.4%
(0.97–20.1%) |
4.02
(0.88–18.34) |
3.32% higher
(0.13% lower to
19.05% higher) |
C2 |
*Event rate of control group.
**Calculated by event rate of control group and relative risk in the meta-analysis.
***Relative risk calculated in the meta-analysis using Review Manager.
1Certainty downgraded to B because there was only 1 RCT.
2Certainty judged to be C because of observational studies only.
3Certainty judged to be B because both the RCT and observational studies had a similar tendency in the result and the certainty of the RCT, B, was adopted.
4Certainty judged to be B because both the RCT and observational studies had a similar tendency in the result and the certainty of the RCT, B, was adopted.
CI, confidence interval; RCT, randomized controlled trial. (Adapted from references 191, 192, 194, 195.)
C.2.1 Major Bleeding (Figure 19)
This was examined in 1 RCT191 and 3 observational studies.192,194,195 In the RCT, major bleeding occurred in patients who did and did not undergo heparin bridging at rates of 1.31% vs. 3.24% (95% CI 1.7–6.3%), the absolute difference was 1.93% higher for patients who underwent heparin bridging (95% CI 0.37% higher to 5.01% higher), and according to the meta-analysis of the observational studies (n=4,081), the corresponding data were 1.34% vs. 4.3% (95% CI 2.5–7.2%), the absolute difference was 2.91% higher for patients who underwent heparin bridging (95% CI 1.19% higher to 5.8% higher), and the difference was significant.
C.2.2 Minor Bleeding (Figure 19)
Minor bleeding is also significantly increased by heparin bridging.
Additional Considerations
The following were opinions from the panel.
Because it requires continuous intravenous infusion, there is a risk of delirium for older adult patients.438 Continuous intravenous infusion can hinder patients’ ADL, and is inconvenient for patients, as well as having the risk of line infections.
C.2.3 Undesirable Effects
Undesirable effects were judged to be “moderate”.
C.3 Trade-Off Between Desirable Effects (Benefits) and Undesirable Effects (Harms)
Heparin bridging has no benefits, and increases harms. The panel determined that the control “no heparin bridging” was superior.
Additional Considerations
The risk of thrombosis in patients in the cited studies was as follows.
In Douketis et al. 2015 (RCT),191 62% had CHADS ≤2 points, and 34% had 3 or 4 points. CHADS 5, 6 points were a minority (2.7%). Aside from that, Douketis et al. 2015192 reported INR: mean 2.1 (SD 1.1), Sjögren et al., 2017: 2.4 (1.4),193 Steinberg et al., 2015: 2.5 (1.4),194 Smoyer-Tomic et al., 2012: 1.3 (1.6),196 Billett et al., 2010: CHADS 5, 6 points in ≈8%,197 and Wysokinski et al., 2008: CHADS 2.2 points.195 CHADS 5, 6 points only represented a small percentage of patients. The effects of heparin bridging in these patients at high risk of thrombosis is unknown.
D. Certainty of Body of Evidence
The RCT was not biased on any of the investigated outcomes, but the 95% CI of the RR of results was wide, and only one RCT existed, thus, the certainty of body of evidence was downgraded to B. The observational studies were graded C. The results were consistent between the RCT and observational studies, thus, the certainty of body of evidence was graded B, based on the certainty of RCT.
E. Values and Preferences
According to the systematic review, there were no studies related to patients’ values and preferences on whether to bridge warfarin with heparin.
Additional Consideration
If bridging with heparin in Japan, fixed dose of subcutaneous low-molecular-weight heparin cannot be used due to not being covered by medical health insurance. It requires dose adjustment of intravenous injection of unfractionated heparin by monitoring the activated partial thromboplastin time (APTT). Therefore, heparin bridging requires hospitalization for several days before the surgery and regular blood tests, increasing the patient’s burden. Avoiding heparin bridging should thus lower the burden on patients and is expected to be in line with their values and preferences.
F. Cost
We could not identify any studies performing a cost–benefit analysis from the perspectives of patients according to the systematic review. An example of estimated out-of-pocket costs is presented below.
Assuming the cost of heparin 5,000 units=JPY137
15 thousand units/day=JPY410, 5 days=JPY2,050, cost of APTT testing=JPY290, assuming it is measured 4 times/day, JPY290 × 4=JPY1,160/day, × 5 days=JPY5,800, which is a total of JPY7,850. Assuming the patient pays 30% out-of-pocket fees, the cost for the patient is JPY2,355, and the fees of hospitalizations are added to this sum.
In the Diagnosis Procedure Combination (DPC) payment system, costs would be calculated based on the DPC points quick chart, April 2020 version.439 For example, if a patient is hospitalized for colectomy for colorectal cancer, and is hospitalized for 5 extra days perioperatively, then the total fees are calculated by the following equation:
DPC hospitalization fees = per day points × medical institution’s coefficient (for each hospital) × days
Assuming the medical institution’s coefficient to be 1,
2,809 points × 1 × 5 (days of hospital stay) × 3 (assuming patient pays 30% out-of-pocket) = JPY42,135 (per day cost: JPY8,427) (in actuality, the medical institution’s coefficient is often >1, increasing the total fees more).
G. Acceptability
If using unfractionated heparin, which is covered under the National Health Insurance in Japan, approximately 5 days of hospitalization and blood tests are needed; thus, the acceptability for patients is expected to be low. The acceptability for clinicians is unknown.
H. Feasibility
Because heparin bridging does not require any special technology, it can be managed in a non-specialized institution. Therefore, the feasibility is high.
I. Grading Recommendation
There were no objections in the panel session. The recommendation was approved with 100% voting rate, median 8.0, disagreement index 0.176, and 100% agreement rate by the modified Delphi method (RAND appropriateness method).
J. Recommendation by Other Related Clinical Practice Guidelines
J.1 JCS/JHRS 2020 Guideline on pharmacotherapy of cardiac arrhythmias Japanese Circulation Society171
Heparin bridging may be considered during the interruption of warfarin perioperatively (Class IIb, Level B). However, in the main text, it is indicated that, “generally, routine heparin bridging is not necessary for patients with atrial fibrillation during interruption of warfarin for an elective invasive procedure with high bleeding risk”. It also states that, “heparin bridging should be considered for patients using warfarin for valvular atrial fibrillation related to a mechanical valve or rheumatic mitral valve stenosis, and non-valvular atrial fibrillation at extremely high risk of thromboembolism (e.g., history of brain infarction within 3 months, extremely high CHADS2
score)”.
J.2 American College of Chest Physicians 2021 clinical practice guidelines179
Regarding patients with a mechanical heart valve, atrial fibrillation, or venous thromboembolism on warfarin, the clinical practice guidelines recommends heparin bridging during interruption of warfarin based on the risk of thromboembolism.
Patients at high risk for thromboembolism, bridging anticoagulation is suggested (GRADE 2C).
Patients at low risk for thromboembolism, no bridging is suggested (GRADE 2C).
Patients at moderate risk for thromboembolism, the bridging or no-bridging should be individualized based on an assessment of individual patient- and surgery-related factors.
K. Monitoring Risks After Non-Cardiac Surgery
If heparin bridging is not used, monitoring is unnecessary.
L. Surveillance and Evaluation of the Recommendation
Consider evaluating the rates of perioperative brain infarction, embolism, and bleeding in each institution.
M. Future Research Directions
An RCT of Japanese patients on warfarin analyzing by whether or not warfarin was bridged with heparin is needed and should be tested on patients with atrial fibrillation as well as pulmonary thromboembolism and mechanical valves.
