Preoperative Assessment of Endothelial Function for Prediction of Adverse Events After Cardiovascular Surgery

Abstract

Background: Cardiovascular surgery is one of the highest risk procedures in the field of surgery. Preoperative assessment of endothelial function has been reported as useful for predicting postoperative adverse events (AEs). The aim of this study was to investigate the relationship between endothelial function assessed by reactive hyperemia index (RHI) and AEs after cardiovascular surgery.

Methods and Results: A prospective observational study of 197 patients who underwent cardiovascular surgery was conducted. RHI was measured before the surgery. The primary endpoint was a composite of postoperative death, reoperation, stroke, newly required dialysis, deep sternum infection, and prolonged ventilation within 30 days. The secondary endpoint was new-onset atrial fibrillation (AF) within 30 days. Following cardiovascular surgery, 19 patients (9.6%) had AEs. New-onset AF was documented in 42 (25.9%) of 162 patients without a prior history of AF. In the receiver-operating characteristic curve analysis, RHI significantly predicted AEs (area under the curve [AUC] 0.67, best cutoff value 1.64, P=0.03), whereas RHI did not predict new-onset AF (AUC 0.53, P=0.93). Patients with RHI ≤1.64 had more AEs than those with RHI >1.64 (16.3% vs. 4.5%, P=0.005). Multiple logistic regression analysis showed the number of surgical procedures and RHI ≤1.64 as significant predictors of AEs.

Conclusions: Preoperative endothelial dysfunction assessed by RHI was associated with postoperative AEs in patients with cardiovascular surgery.

Despite recent advances in perioperative care and management, patients undergoing cardiovascular surgery represent a more challenging group than other various surgeries.^1,2 Thus, risk stratification of patients prior to a surgical procedure is useful and important for determining patient management and improving quality control.¹ Although prognostic risk models for cardiac and aortic surgery, such as EuroSCORE II and JapanSCORE,^3–5 have been introduced into current clinical practice, there is still room for improvement in terms of their predictive ability.⁶

Endothelial function is known to be associated with maintenance of vascular homeostasis and regulation of the coagulation system.^7,8 Impairment of endothelial function leads to progression of atherosclerosis and consequently, increased risk of cardiovascular events.^9,10 Preoperative endothelial dysfunction has previously been demonstrated to be a predictor of postoperative adverse events (AEs) in patients with peripheral vascular surgery,¹¹ general thoracic surgery,¹² and total hip or knee arthroplasty.¹³ However, there are no data about this issue in patients undergoing cardiovascular surgery. Furthermore, endothelial dysfunction is also associated with the pathogenesis of atrial fibrillation (AF),¹⁴ the new onset of which is one of the major concerns after cardiovascular surgery.¹⁵

Although endothelial function has been evaluated by flow-mediated dilation (FMD) in a number of studies, the recently developed reactive hyperemia peripheral arterial tonometry (RH-PAT) can non-invasively assess endothelial function as a reactive hyperemia index (RHI) with excellent reproducibility.¹⁶ Both RHI and FMD are reported to predict cardiovascular events with similar prognostic magnitude.¹⁷ Thus, preoperative assessment of endothelial function using RH-PAT may be able to predict postoperative events in patients scheduled for cardiovascular surgery. The aim of this study was to investigate the relationship between RHI and AEs after cardiovascular surgery.

Methods

Study Population

Between October 2014 and April 2016, a total of 200 patients who were scheduled to undergo elective cardiovascular surgery with thoracotomy and/or laparotomy, such as coronary artery bypass grafting, cardiac valve and aortic surgery, under general anesthesia at Chiba University Hospital were prospectively enrolled. Patients on hemodialysis were considered ineligible; 2 patients who could not complete measurement of RHI because of upper arm pain during the test, and 1 patient who declined surgery at the last minute were excluded. Thus, 197 patients were included. The present study was approved by the institutional ethical committee, and written informed consent was given by all patients. This study was registered at the University Hospital Medical Information Network Clinical Trials Registry (UMIN000015135).

RH-PAT

Endothelial function was assessed by RH-PAT using the EndoPAT2000 device (Itamar Medical, Caesarea, Israel). The principle of RH-PAT measurement has been previously described.^13,18,19 In brief, the patients fasted and refrained from caffeine, tobacco or medications for at least 8 h. The test was conducted in a quiet and temperature-controlled room in the early morning. A finger probe was placed on each index finger, and the air cuff was connected by a pneumatic tube to the inflation device. Pulsatile volume changes in the distal digit were recorded by the EndoPAT2000 device. After a 5-min baseline measurement, the blood pressure cuff on the test arm was inflated to obtain complete occlusion for 5 min, then the cuff was deflated and the PAT tracing was continued for another 5 min. The RH-PAT data were automatically analyzed to calculate the RHI, which is the ratio of amplitude of the PAT signal after cuff deflation divided by that before cuff inflation, indexed to the contralateral arm. In this study, RHI was measured before cardiovascular surgery during hospitalization.

Endpoints

The primary endpoint was a composite of postoperative death, reoperation for any reason, stroke, newly required dialysis, deep sternum infection, and prolonged ventilation (≥72 h) within 30 days of the surgical procedure according to JapanSCORE.^4,5 The secondary endpoint was the development of new-onset AF within 30 days in patients without a prior history of AF. These events were documented by cardiovascular surgeons who were unaware of the RHI data.

Statistical Analysis

Statistical analysis was performed with SAS statistical software package version 9.4 (SAS Institute, Cary, NC, USA). Data are expressed as the mean±standard deviation or frequency (%). Continuous and categorical variables were compared using Student’s t-test and Fisher’s exact test. The receiver-operating characteristic (ROC) curve analysis was performed based on the occurrence of the primary endpoint. Thresholds for RHI were established by finding the values that corresponded to the maximum average sensitivity and specificity. The associated variables in univariable analyses (P<0.05) were included in the stepwise backward selection method in the multivariable model to identify the independent predictors of the primary endpoint, presented as odds ratio (OR) with 95% confidence intervals (CI). A value of P<0.05 was considered statistically significant.

We assumed a 1:1 distribution of patients with lower and higher RHI divided by best cutoff value for the primary endpoint and, based on a previous study in patients with peripheral vascular surgery,¹¹ we anticipated a 5% event rate for the primary endpoint in patients with higher RHI. Allowing for a type I error rate of 0.05, a sample size of 282 patients was expected to provide 80% power to detect an event rate of 15% in patients with lower RHI. However, patients with coronary artery bypass grafting, cardiac valve, and aortic surgery are assumed to have more events corresponding to the primary endpoint of this study, ranging from 21.1% to 34.5%.⁴ Thus, the sample size of 200 patients was determined.

Results

In a total of 197 patients, 19 patients (9.6%) had primary endpoint AEs, comprising death (n=2), stroke (n=5), reoperation (n=12), newly required dialysis (n=3), deep sternum infection (n=2), and prolonged ventilation (n=3) within 30 days of the surgical procedure. Reoperation was mainly because of cardiac tamponade (n=6). New-onset AF was documented in 42 (25.9%) of 162 patients without a prior history of AF. Baseline characteristics of patients with and without events are shown in Table 1. In the ROC curve analysis, RHI significantly predicted AEs (area under the curve [AUC] 0.67, best cutoff value 1.64, P=0.03), whereas RHI did not predict new-onset AF in patients without a preoperative history of AF (AUC 0.53, P=0.93). Patients with RHI ≤1.64 had more AEs than those with RHI >1.64 (16.3% vs. 4.5%, P=0.005) (Table 2). With the cutoff value, the diagnostic performance of RHI for postoperative AEs had a sensitivity of 74%, specificity of 60%, positive predictive value of 16%, negative predictive value of 96%, and accuracy of 61%. Multiple logistic regression analysis showed the number of surgical procedures and RHI ≤1.64 as significant predictors of AEs (Table 3).

Table 1. Characteristics of the Study Patients Undergoing Cardiovascular Surgery

Variable	All (n=197)	Event (−) (n=178)	Event (+) (n=19)	P value
Age (years)	70.1±10.7	70.0±10.8	71.3±11.8	0.62
Men	129 (65%)	117 (66%)	12 (63%)	0.82
Body mass index (kg/m2)	23.1±3.3	23.2±3.3	22.3±3.2	0.25
Hypertension	140 (71%)	125 (70%)	15 (79%)	0.43
Diabetes mellitus	46 (23%)	43 (24%)	3 (16%)	0.42
Dyslipidemia	131 (66%)	120 (67%)	11 (58%)	0.41
Current smoker	29 (15%)	28 (16%)	1 (5%)	0.22
Prior myocardial infarction	24 (12%)	21 (12%)	3 (16%)	0.62
History of AF	35 (18%)	28 (16%)	7 (37%)	0.02
eGFR (mL/min/1.73 m2)	55.8±19.2	56.8±18.9	46.4±19.7	0.02
LVEF	58.6±11.4	58.8±11.6	56.7±9.4	0.45
Medication
ACE-I or ARB	108 (55%)	96 (54%)	12 (63%)	0.45
β-blocker	63 (32%)	57 (32%)	6 (32%)	0.97
Statin	89 (45%)	79 (44%)	10 (53%)	0.49
Type of surgery
CABG	70 (36%)	62 (35%)	8 (42%)	0.53
Aortic valve surgery	73 (37%)	65 (37%)	8 (42%)	0.63
Mitral valve surgery	56 (28%)	45 (25%)	11 (58%)	0.003
Tricuspid valve surgery	31 (16%)	25 (14%)	6 (32%)	0.046
Thoracic aortic surgery	33 (17%)	28 (16%)	5 (26%)	0.24
Abdominal aortic surgery	10 (5%)	10 (6%)	0 (0%)	0.29
LAA closure and/or Maze procedure	27 (14%)	23 (13%)	4 (21%)	0.33
Other	16 (8%)	14 (8%)	2 (8%)	0.97
No. of surgical procedures	1.6±0.9	1.5±0.8	2.3±1.1	0.001
Preoperative risk score
EuroSCORE II	3.4±4.2	3.1±3.9	6.2±5.8	0.002
JapanSCORE
30-day operative mortality	4.5±4.9	4.3±4.1	6.6±9.5	0.049
Major complication	18.3±11.3	17.9±10.9	21.7±14.7	0.16
RHI	1.81±0.48	1.83±0.48	1.58±0.39	0.03

ACE-I, angiotensin-converting enzyme inhibitor; AF, atrial fibrillation; ARB, angiotensin-receptor blocker; CABG, coronary artery bypass grafting; eGFR, estimated glomerular filtration rate; LAA, left atrial appendage; LVEF, left ventricular ejection fraction; RHI, reactive hyperemia index.

Table 2. Postoperative AEs in the Study Patients Undergoing Cardiovascular Surgery

Variable	All (n=197)	RHI ≤1.64 (n=86)	RHI >1.64 (n=111)	P value
AEs within 30 days	19 (12.2%)	14 (16.3%)	5 (4.5%)	0.005
Death	2 (1.0%)	2 (2.3%)	0 (0%)	0.11
Stroke	5 (2.5%)	4 (4.7%)	1 (0.9%)	0.10
Reoperation	12 (6.1%)	8 (9.3%)	4 (3.6%)	0.10
Newly required dialysis	3 (1.5%)	2 (2.3%)	1 (0.9%)	0.42
Deep sternum infection	2 (1.0%)	1 (1.2%)	1 (0.9%)	0.86
Prolonged ventilation	3 (1.5%)	2 (2.3%)	1 (0.9%)	0.42

AE, adverse events; RHI, reactive hyperemia index.

Table 3. Predictors of AEs in the Study Patients Undergoing Cardiovascular Surgery

Variable	Univariable		Multivariable
	OR (95% CI)	P value	OR (95% CI)	P value
History of AF	3.13 (1.13–8.63)	0.03
eGFR (mL/min/1.73 m2)	0.97 (0.94–0.996)	0.03
Mitral valve surgery	4.07 (1.54–10.74)	0.005
No. of surgical procedures	2.28 (1.42–3.65)	<0.001	2.13 (1.30–3.49)	0.003
EuroSCORE II	1.11 (1.02–1.21)	0.01
RHI ≤1.64	4.12 (1.42–11.94)	0.009	3.38 (1.13–10.08)	0.03

CI, confidence interval; OR, odds ratio. Other abbreviations as in Tables 1,2.

Discussion

The present study demonstrated that the preoperative RHI had a predictive value for postoperative AEs in patients undergoing cardiovascular surgery. The number of surgical procedures was also demonstrated as an independent predictor of AEs. On the other hand, the RHI did not predict postoperative AF.

Endothelial Dysfunction and Postoperative AEs

RHI assessed by RH-PAT can evaluate endothelial function non-invasively, reproducibly, and operator independently.^16,20 It has been previously shown that RHI <1.67 indicates endothelial dysfunction,²⁰ which is consistent with our best cutoff value for predicting postoperative AEs (i.e., 1.64). In the present study, patients with RHI ≤1.64 had more AEs, which consisted of death, reoperation, stroke, newly required dialysis, deep sternum infection, and prolonged ventilation. Although, the difference was not statistically significant in each component, the consistent trend to greater event rates was shown in patients with endothelial dysfunction. Endothelial dysfunction is related to an increased risk of cardiovascular events.^9,10 Preoperative assessment of endothelial function is reportedly useful to predict postoperative events in patients undergoing various types of surgery.^11–13 Schier et al demonstrated that patients with low FMD had more postoperative events that consisted of cardiac events, pulmonary events (e.g., prolonged ventilation), wound infection, or reoperative intervention among subjects undergoing general thoracic surgery, including esophagectomy and major lung surgery.¹² Cardiovascular surgery is one of the most challenging among the various surgical fields, and the postoperative event rate is substantially high.⁴ Therefore, it is important to clarify the predictors and establish risk stratification after cardiovascular surgery. Although the AUC of the RHI for the primary endpoint was relatively low, endothelial dysfunction may be one of the predictors for postoperative AEs from the results of our study.

Role of Endothelial Function

Endothelial function is reported to be associated with atherosclerotic progression,⁷ regulation of the coagulation system,⁸ wound healing,²¹ inflammation and renal dysfunction.²² Collectively, these factors may be the mechanisms explaining why the event rate of all components was greater in patients with endothelial dysfunction than in those without. In particular, reoperation because of cardiac tamponade was the major cause of AEs in this study (3.0%). Endothelial cells play important roles during the process of wound healing, such as recruitment of leukocytes into the wounded tissue and promotion of angiogenesis via nitric oxide (NO) and chemokines.²¹ Additionally, a previous study showed a negative correlation between levels of high-sensitivity C-reactive protein and FMD.²² Damage to the endothelium promotes the expression of cellular adhesion molecules, overexpression of which may aggravate the inflammatory reaction and tissue injury.²³ Thus, more cases of cardiac tamponade in patients with endothelial dysfunction might be related to impaired wound healing and inflammation. Although NO bioavailability plays an important role in the methodologies, including FMD, used for assessment of endothelial function, other substances, such as prostaglandin, adenosine, and hydrogen peroxide, can also influence vasodilation response to shear stress and ischemia.¹⁷ Endothelium-derived NO might have a more important role in FMD than in the RHI,¹⁷ and thus, further studies are needed to clarify the detailed mechanism of the RHI.

Endothelial dysfunction can be recovered by several interventions such as diet, exercise and administration of angiotensin-converting enzyme inhibitors and statins.^24–26 In addition, cardiovascular risk can be reduced in patients with improved endothelial function compared with those with persistent endothelial dysfunction.²⁷ Therefore, it is suggested that preoperative management to recover endothelial function might improve postoperative outcomes in patients with endothelial dysfunction who have to undergo cardiovascular surgery.

Endothelial Function and New-Onset AF

New-onset AF, which is the most common complication after cardiac surgery with an incidence of 20–50%, is also one of the important issues because of the increased rates of death and readmission, and increased medical costs.¹⁵ In the general population, the association of endothelial dysfunction with incidence of AF has been reported.¹⁴ In the present study, however, there was no obvious relation between RHI and new-onset AF after cardiovascular surgery. Various factors, such as patients’ clinical characteristics, blood biomarkers, echocardiographic and electrocardiographic measurements, and genetic variants, are known to predict AF in the general population.²⁸ Although several scoring systems using patients’ clinical factors or echocardiographic measurements have been reported as useful tools to predict new-onset AF after cardiac surgery,²⁹ a prediction method for postoperative AF has not been well established. Because the underlying mechanisms of postoperative AF remain widely unknown, preoperative assessment of endothelial function may be less likely to predict new-onset AF after the surgical procedure.

Study Limitations

There are some limitations in the present study. First, this was a single-center study. Second, the number of patients was relatively small. Third, endothelial function was assessed using RH-PAT in this study, whereas FMD was used in a number of the previous studies. Although early reports showed a positive correlation between RHI and FMD,¹⁸ a recent study did not.³⁰ Nevertheless, RHI, which represents digital microvessel dilation, is considered to evaluate endothelial function from a different aspect than FMD, which assesses conduit artery vasodilation.³⁰ In addition, both RHI and FMD are reported to predict cardiovascular events with similar prognostic magnitude.¹⁷ Fourth, RHI was measured once before cardiovascular surgery in this study. Although the test-retest reliability of RHI measurement has been reported to be good,³¹ physiological variation of the subjects may influence the RHI.

Conclusions

Preoperative endothelial dysfunction assessed by RHI was associated with postoperative AEs in patients undergoing cardiovascular surgery, and may be useful in the risk stratification of future cardiovascular surgical patients.

Disclosures

None.

References

• 1.   Motomura N, Miyata H, Tsukihara H, Takamoto S. Risk model of thoracic aortic surgery in 4707 cases from a nationwide single-race population through a web-based data entry system: The first report of 30-day and 30-day operative outcome risk models for thoracic aortic surgery. Circulation 2008; 118: S153–S159.
• 2.   Motomura N, Miyata H, Tsukihara H, Okada M, Takamoto S. First report on 30-day and operative mortality in risk model of isolated coronary artery bypass grafting in Japan. Ann Thorac Surg 2008; 86: 1866–1872.
• 3.   Nashef SA, Roques F, Sharples LD, Nilsson J, Smith C, Goldstone AR, et al. EuroSCORE II. Eur J Cardiothorac Surg 2012; 41: 734–744.
• 4.   Miyata H, Motomura N, Tsukihara H, Takamoto S. Risk models including high-risk cardiovascular procedures: Clinical predictors of mortality and morbidity. Eur J Cardiothorac Surg 2011; 39: 667–674.
• 5.   Saito A, Motomura N, Miyata H, Takamoto S, Kyo S, Ono M. Age-specific risk stratification in 13488 isolated coronary artery bypass grafting procedures. Interact Cardiovasc Thorac Surg 2011; 12: 575–580.
• 6.   Chalmers J, Pullan M, Fabri B, McShane J, Shaw M, Mediratta N, et al. Validation of EuroSCORE II in a modern cohort of patients undergoing cardiac surgery. Eur J Cardiothorac Surg 2013; 43: 688–694.
• 7.   Widlansky ME, Gokce N, Keaney JF Jr, Vita JA. The clinical implications of endothelial dysfunction. J Am Coll Cardiol 2003; 42: 1149–1160.
• 8.   Yau JW, Teoh H, Verma S. Endothelial cell control of thrombosis. BMC Cardiovasc Disord 2015; 15: 130.
• 9.   Yeboah J, Crouse JR, Hsu FC, Burke GL, Herrington DM. Brachial flow-mediated dilation predicts incident cardiovascular events in older adults: The Cardiovascular Health Study. Circulation 2007; 115: 2390–2397.
• 10.   Rubinshtein R, Kuvin JT, Soffler M, Lennon RJ, Lavi S, Nelson RE, et al. Assessment of endothelial function by non-invasive peripheral arterial tonometry predicts late cardiovascular adverse events. Eur Heart J 2010; 31: 1142–1148.
• 11.   Gokce N, Keaney JF Jr, Hunter LM, Watkins MT, Nedeljkovic ZS, Menzoian JO, et al. Predictive value of noninvasively determined endothelial dysfunction for long-term cardiovascular events in patients with peripheral vascular disease. J Am Coll Cardiol 2003; 41: 1769–1775.
• 12.   Schier R, Hinkelbein J, Marcus H, Mehran R, El-Zein R, Hofstetter W, et al. Preoperative microvascular dysfunction: A prospective, observational study expanding risk assessment strategies in major thoracic surgery. Ann Thorac Surg 2012; 94: 226–233.
• 13.   Suzuki H, Matsuzawa Y, Konishi M, Akiyama E, Takano K, Nakayama N, et al. Utility of noninvasive endothelial function test for prediction of deep vein thrombosis after total hip or knee arthroplasty. Circ J 2014; 78: 1723–1732.
• 14.   O’Neal WT, Efird JT, Yeboah J, Nazarian S, Alonso A, Heckbert SR, et al. Brachial flow-mediated dilation and incident atrial fibrillation: The multi-ethnic study of atherosclerosis. Arterioscler Thromb Vasc Biol 2014; 34: 2717–2720.
• 15.   Gillinov AM, Bagiella E, Moskowitz AJ, Raiten JM, Groh MA, Bowdish ME, et al. Rate control versus rhythm control for atrial fibrillation after cardiac surgery. N Engl J Med 2016; 374: 1911–1921.
• 16.   Selamet Tierney ES, Newburger JW, Gauvreau K, Geva J, Coogan E, Colan SD, et al. Endothelial pulse amplitude testing: Feasibility and reproducibility in adolescents. J Pediatr 2009; 154: 901–905.
• 17.   Matsuzawa Y, Kwon TG, Lennon RJ, Lerman LO, Lerman A. Prognostic value of flow-mediated vasodilation in brachial artery and fingertip artery for cardiovascular events: A systematic review and meta-analysis. J Am Heart Assoc 2015; 4: e002270.
• 18.   Kuvin JT, Patel AR, Sliney KA, Pandian NG, Sheffy J, Schnall RP, et al. Assessment of peripheral vascular endothelial function with finger arterial pulse wave amplitude. Am Heart J 2003; 146: 168–174.
• 19.   Axtell AL, Gomari FA, Cooke JP. Assessing endothelial vasodilator function with the Endo-PAT 2000. J Vis Exp 2010; 44: 2167.
• 20.   Akiyama E, Sugiyama S, Matsuzawa Y, Konishi M, Suzuki H, Nozaki T, et al. Incremental prognostic significance of peripheral endothelial dysfunction in patients with heart failure with normal left ventricular ejection fraction. J Am Coll Cardiol 2012; 60: 1778–1786.
• 21.   Laing T, Hanson R, Chan F, Bouchier-Hayes D. The role of endothelial dysfunction in the pathogenesis of impaired diabetic wound healing: A novel therapeutic target? Med Hypotheses 2007; 69: 1029–1031.
• 22.   Recio-Mayoral A, Banerjee D, Streather C, Kaski JC. Endothelial dysfunction, inflammation and atherosclerosis in chronic kidney disease: A cross-sectional study of predialysis, dialysis and kidney-transplantation patients. Atherosclerosis 2011; 216: 446–451.
• 23.   Nolte D, Kuebler WM, Muller WA, Wolff KD, Messmer K. Attenuation of leukocyte sequestration by selective blockade of PECAM-1 or VCAM-1 in murine endotoxemia. Eur Surg Res 2004; 36: 331–337.
• 24.   Matsuzawa Y, Sugiyama S, Sugamura K, Sumida H, Kurokawa H, Fujisue K, et al. Successful diet and exercise therapy as evaluated on self-assessment score significantly improves endothelial function in metabolic syndrome patients. Circ J 2013; 77: 2807–2815.
• 25.   Shahin Y, Khan JA, Samuel N, Chetter I. Angiotensin converting enzyme inhibitors effect on endothelial dysfunction: A meta-analysis of randomised controlled trials. Atherosclerosis 2011; 216: 7–16.
• 26.   Dupuis J, Tardif JC, Cernacek P, Théroux P. Cholesterol reduction rapidly improves endothelial function after acute coronary syndromes: The RECIFE (REduction of Cholesterol in Ischemia and Function of the Endothelium) trial. Circulation 1999; 99: 3227–3233.
• 27.   Kitta Y, Obata JE, Nakamura T, Hirano M, Kodama Y, Fujioka D, et al. Persistent impairment of endothelial vasomotor function has a negative impact on outcome in patients with coronary artery disease. J Am Coll Cardiol 2009; 53: 323–330.
• 28.   Alonso A, Norby FL. Predicting atrial fibrillation and its complications. Circ J 2016; 80: 1061–1066.
• 29.   Mariscalco G, Biancari F, Zanobini M, Cottini M, Piffaretti G, Saccocci M, et al. Bedside tool for predicting the risk of postoperative atrial fibrillation after cardiac surgery: The POAF score. J Am Heart Assoc 2014; 3: e000752.
• 30.   Hamburg NM, Palmisano J, Larson MG, Sullivan LM, Lehman BT, Vasan RS, et al. Relation of brachial and digital measures of vascular function in the community: The Framingham heart study. Hypertension 2011; 57: 390–396.
• 31.   Hasin T, Matsuzawa Y, Guddeti RR, Aoki T, Kwon TG, Schettle S, et al. Attenuation in peripheral endothelial function after continuous flow left ventricular assist device therapy is associated with cardiovascular adverse events. Circ J 2015; 79: 770–777.