Abstract
Background:
Acute pulmonary embolism (PE) is a major threat to the health and lives of hospitalized patients. This study was conducted to clarify the real-world outcomes of pulmonary embolectomy.
Methods and Results:
Retrospective investigation of 355 patients who underwent pulmonary embolectomy for acute PE was conducted using the Japanese Cardiovascular Surgery Database. Risk factors for operative death within 30 days after pulmonary embolectomy and major adverse cardiovascular events (MACE), including operative death, postoperative stroke and postoperative coma, were analyzed. Cardiopulmonary resuscitation (CPR) was required preoperatively in 27.6%, and preoperative veno-arterial extracorporeal membrane oxygenation was performed in 26.5%. Urgent or emergency operation was performed in 93% of patients. Operative mortality rate was 73/355 (20.6%). Incidence of MACE was 97/355 (27.3%). In univariate analysis, preoperative predictors of death were obesity, renal dysfunction, chronic obstructive pulmonary disease, liver injury, recent myocardial infarction, shock, refractory shock, CPR, heart failure, inotrope use, poor left ventricular function, preoperative arrhythmia and tricuspid regurgitation. In multivariate analysis, independent risk factors for operative death were heart failure (P=0.013), poor left ventricular function (P=0.007), and respiratory failure (P=0.001). Poor left ventricular function (P=0.033), preoperative CPR (P=0.002) and respiratory failure (P=0.007) were independent risk factors for MACE.
Conclusions:
The outcomes of pulmonary embolectomy were acceptable, considering the urgency and preoperative comorbidities of patients. Early triage of patients with hemodynamically unstable PE is important.
Acute pulmonary embolism (PE) is a major threat to the health and lives of hospitalized patients. Although PE is an adverse outcome of deep vein thrombosis (DVT), DVT is frequently asymptomatic in bed-ridden patients. Thrombolysis for acute massive PE improves the right heart burden, but the life-salvage effect of thrombolytic agents is questionable because of the risk of major hemorrhagic complications, such as intracranial hemorrhage, especially in aged patients.1–3
Friedrich Trendelenburg in Leipzig developed pulmonary embolectomy as a life-saving operation for massive PE approximately 100 years ago.4
Pulmonary embolectomy is an important therapeutic option for massive PE with unstable hemodynamics, massive PE after failed thrombolysis and massive PE contraindicated to thrombolytic therapy. It is also effective for submissive PE with relative contraindication to anticoagulant therapy. Recent reports show a variable mortality rate for pulmonary embolectomy, ranging from 3.6% to 27.2%.5–7
The annual volume of pulmonary embolectomy in each cardiovascular institution is very small. A multi-institutional study of 214 patients undergoing pulmonary embolectomy, which included 82% submassive PEs, indicated a mortality rate of 11.7%.6
The same study showed preoperative cardiopulmonary arrest was an important risk factor for death. However, little is known about the risk factors for mortality and morbidity following pulmonary embolectomy for acute PE.
The Japan Cardiovascular Surgery Database (JCVSD) Organization was founded in 2001 to analyze the clinical outcomes of adult cardiovascular surgery. As of 2014, 538 hospitals were part of this organization and the total number of registered adult cardiovascular surgical cases at the time of this study was 280,000. We analyzed data from the JCVSD to investigate the risk factors for the mortality and morbidity of pulmonary embolectomy.
Methods
Data Collection
The data collection form of the JCVSD for adult cardiovascular surgical cases has more than 300 variables, the definitions of which (available at: http://www.jacvsd.umin.jp) are based on those of the Society of Thoracic Surgeons (STS) National Clinical Database. We identified 355 patients who underwent pulmonary embolectomy for PE from January 2008 to December 2014 in the JCVSD. Patients who underwent pulmonary endarterectomy for chronic thromboembolic pulmonary hypertension were excluded from this study. Patients who underwent concomitant cardiovascular surgery with pulmonary embolectomy were included. The average number of patients registered annually was 51. There were 186 female patients and 169 male patients. Average patient age was 62.1±15.7 years, ranging from 15 to 91 years. Patient characteristics are shown in
Table 1. Operative death was defined as in-hospital or 30-day death (whichever was later), which is equivalent to “the 30-day operative mortality” defined in the STS National Adult Cardiac Surgery Database. Data from survivors and non-survivors were compared to analyze the risk factors for operative death. We also analyzed morbidity after pulmonary embolectomy as a major adverse cardiovascular event (MACE). MACE included early death, postoperative stroke and postoperative coma.
Table 1.
Characteristics of Survivors and Non-Survivors Among Patients With Acute Massive Pulmonary Embolism Who Underwent Pulmonary Embolectomy
| |
Total
(n=355) |
Survivors
(n=282) |
Non-survivors
(n=73) |
P value |
| Male sex (%) |
169 (47.6%) |
138 (48.9%) |
31 (42.5%) |
0.359 |
| Age (years) |
62.1±15.7 |
61.2±15.8 |
63.2±14.7 |
0.313 |
| History of smoking |
92 (25.9%) |
76 (27.0%) |
16 (21.9%) |
0.454 |
| Recent smoker (<1 month) |
51 (14.4%) |
42 (14.9%) |
9 (12.3%) |
0.709 |
| Diabetes |
52 (14.6%) |
38 (13.5%) |
14 (19.1%) |
0.264 |
| Obesity (BMI >30) |
42 (11.8%) |
27 (9.6%) |
15 (20.5%) |
0.014 |
| Renal dysfunctiona |
46 (13.0%) |
29 (10.3%) |
17 (23.3%) |
0.006 |
| Dialysis |
3 (0.8%) |
2 (0.7%) |
1 (1.4%) |
0.50 |
| Dyslipidemia |
66 (18.6%) |
47 (16.7%) |
19 (26.0%) |
0.090 |
| Hypertension |
159 (44.8%) |
120 (42.6%) |
39 (53.4%) |
0.113 |
| History of CVAb |
66 (18.6%) |
60 (21.3%) |
6 (8.2%) |
0.011 |
| Recent CVA |
32 (9.0%) |
28 (9.9%) |
4 (5.5%) |
0.358 |
| COPD (>moderate)c |
53 (14.9%) |
33 (11.7%) |
21 (28.8%) |
0.001 |
| Liver injuryd |
45 (12.7%) |
29 (10.3%) |
16 (6.3%) |
0.016 |
| Recent (≤12 months) AMI |
2 (0.6%) |
0 |
2 (2.7%) |
0.042 |
| Shocke |
178 (50.1%) |
128 (45.4%) |
50 (68.5%) |
0.001 |
| Refractory shockf |
63 (17.7%) |
41 (14.5%) |
23 (31.5%) |
0.002 |
| CPR before embolectomy |
98 (27.6%) |
66 (23.4%) |
32 (43.8%) |
0.001 |
| NYHA IV |
163 (45.9%) |
113 (40.1%) |
50 (68.5%) |
<0.001 |
| Inotrope use |
41 (11.5%) |
26 (9.2%) |
15 (20.5%) |
0.012 |
| Poor LV functiong |
42 (11.8%) |
23 (8.1%) |
19 (26.0%) |
<0.001 |
| Arrhythmia <2 weeks of embolectomyh |
55 (15.5%) |
37 (13.1%) |
18 (24.7%) |
0.019 |
| VA-ECMO |
94 (26.5%) |
71 (25.1%) |
23 (31.5%) |
0.298 |
| Tricuspid regurgitation >grade 2 |
128 (36.1%) |
93 (33.0%) |
35 (47.9%) |
0.020 |
| Previous cardiotomy |
7 (2.0%) |
5 (1.8%) |
2 (2.7%) |
0.636 |
| CABG |
3 (0.8%) |
3 (1.1%) |
0 (0%) |
|
| Valve surgery |
2 (0.6%) |
2 (0.7%) |
0 (0%) |
|
| Previous thoracic aorta surgery |
7 (2.0%) |
4 (1.4%) |
3 (4.1%) |
|
aPositive urinary protein, S-Cr ≥1.3 mg/dL, eGFR <60 mL/min/1.73 cm2. bTransient ischemic attack, stroke (symptoms for >72 h). cFEV1.0 <75%, PaO2
<60, or PaCO2
>50 (room air), steroid use for pulmonary disease. dDiagnosis of liver cirrhosis, GOT >100 IU/L, serum total bilirubin level >1.5 g/dL. eSBP <80 mmHg, and/or cardiac index <1.8 L/min/m2
or maximal dose of inotrope use to keep SBP >80 mmHg or clinically apparent shock. fSBP <80 mmHg and/or cardiac index <1.8 L/min/m2. gPreoperative LV function indicates LV ejection fraction <30% on echocardiography or left ventriculography. hVentricular tachycardia, ventricular fibrillation, complete atrioventricular block, medication for arrhythmia. AMI, acute myocardial infarction; BMI, body mass index; CABG, coronary artery bypass grafting; COPD, chronic obstructive pulmonary disease; CPR, cardiopulmonary resuscitation; CVA, cerebrovascular accident; LV, left ventricular; NYHA, New York Heart Association classification for congestive heart failure; SBP, systolic blood pressure; VA-ECMO, veno-arterial extracorporeal membrane oxygenator.
This study was approved by the Institutional Review Board of Hirosaki University (No. 2017-1050). The requirement for informed patient consent was waived because of the anonymous nature of the data. Written informed consent was given by each patient for registration in the JCVSD.
Statistical Analysis
Discrete variables are expressed as frequencies and percentages. Univariate analysis of death or MACE was performed by the Mann-Whitney test, Chi-square test, or Fisher’s exact test, as appropriate. Continuous variables with normal distribution are expressed as mean and standard deviation. Multivariate analysis for early death and MACE was performed to elucidate risk factors for operative mortality and morbidity. Independent risk factors for hospital death and MACE were examined by multivariate logistic regression analysis using risk factors with P values <0.1 by prior univariate analysis. Statistical analysis was performed using the SPSS software for Windows, version 13.0 (SPSS, Inc., Chicago, IL, USA).
Results
Patient characteristics and deaths are shown in
Table 1. The preoperative condition of all patients was critical; half the patients were in shock, 27.6% required preoperative cardiopulmonary resuscitation (CPR), and 26.5% of patients required veno-arterial extracorporeal membrane oxygenation (VA-ECMO) therapy. The operative mortality rate was 73/355 (20.6%, 95% confidence interval (CI) [16.4, 22.7]). The incidence of MACE was 97/355 (27.3%, 95% CI [22.7, 29.7]). In the univariate analysis, the following factors were found to be statistically significant predictors of operative death: obesity, renal dysfunction, COPD, liver injury, recent acute myocardial infarction (AMI), shock, refractory shock, CPR before embolectomy, heart failure, inotrope use, poor left ventricular (LV) function, preoperative arrhythmia and tricuspid regurgitation. The incidence of preoperative VA-ECMO use was not different between the 2 groups. History of cerebrovascular accident (CVA) was a favorable factor for survival (present in 21.3% of survivors and 8.2% of non-survivors, P=0.011). Most patients (93.0%) underwent emergency pulmonary embolectomy and urgency did not affect survival (Table 2). Cardiopulmonary bypass time and aortic cross-clamp time were significantly longer in non-survivors than in survivors.
Table 2.
Operative Data of the Study Subjects
| |
Total
(n=355) |
Survivors
(n=282) |
Non-survivors
(n=73) |
P value |
| Urgent or emergency op. (<24 h) |
330 (93.0%) |
266 (94.3%) |
64 (87.7%) |
0.068 |
| Concomitant CABG |
13 (3.7%) |
12 (4.2%) |
1 (1.4%) |
|
| Concomitant valve surgery |
21 (5.9%) |
12 (4.2%) |
9 (12.3%) |
|
| PFO or ASD repair |
7 (2.0%) |
7 (2.0%) |
0 (0%) |
|
| Aortic cross-clamp |
206 (58.0%) |
160 (56.7%) |
46 (63.0%) |
0.334 |
| Blood transfusion |
318 (89.6%) |
246 (87.2%) |
72 (98.6%) |
0.005 |
| CPB time (min) |
135.0±94.2 |
121.7±68.7 |
184.2±146.3 |
0.001 |
| Aortic clamp time (min) |
71.8±45.4 |
69.2±37.7 |
88.6±60.8 |
0.047 |
ASD, atrial septal defect; CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass; PFO, patent foramen ovale.
There were various postoperative complications, including re-exploration for bleeding in 13.0%, stroke in 5.4%, coma in 9.9%, renal dysfunction in 12.7% and sepsis in 4.5% (Table 3). The incidences of re-exploration for bleeding, postoperative coma, postoperative renal dysfunction, use of hemodialysis, sepsis and longer ventilator support were significantly higher in non-survivors than in survivors. Although the incidence of ICU stay longer than 8 days was not different between survivors and non-survivors, postoperative hospital stay was significantly shorter in non-survivors than in survivors.
Table 3.
Postoperative Complications Following Pulmonary Embolectomy
| |
Total
(n=355) |
Survivors
(n=282) |
Non-survivors
(n=73) |
P value |
| Re-exploration for bleeding |
46 (13.0%) |
19 (6.7%) |
27 (37.0%) |
<0.001 |
| Stroke |
19 (5.4%) |
13 (4.6%) |
6 (8.2%) |
0.243 |
| Postoperative coma |
35 (9.9%) |
19 (6.7%) |
16 (21.9%) |
0.001 |
| Postoperative renal dysfunctiona |
45 (12.7%) |
22 (7.8%) |
23 (31.5%) |
<0.001 |
| Hemodialysis |
29 (8.1%) |
13 (4.6%) |
16 (21.9%) |
<0.001 |
| Sepsis |
16 (4.5%) |
8 (2.8%) |
8 (11.0%) |
0.007 |
| Ventilator use >72 h |
94 (26.5%) |
59 (20.9%) |
35 (47.9%) |
<0.001 |
| ICU stay (days) |
10.3±15.0 |
11.0±15.8 |
8.0±10.8 |
0.13 |
| ICU stay > 8days |
128 (36.1%) |
107 (37.9%) |
21 (28.8%) |
0.146 |
| Postoperative hospital stay (days) |
38.1±42.5 |
39.8±40.9 |
10.9±14.6 |
<0.001 |
aSerum creatinine level >2.0 mg/dL or serum creatine level doubled from preoperative level, or use of dialysis for renal dysfunction. ICU, intensive care unit.
Result of the univariate analysis of MACE and operative data related to MACE are shown in
Table 4
and
Table 5, respectively. Postoperative complications in patients with MACE are shown in
Table 6.
Table 4.
Univariate Analysis of MACE
| |
Total
(n=355) |
No MACE
(n=258) |
MACE
(n=97) |
P value |
| Male sex |
169 (47.6%) |
121 (46.9%) |
48 (49.5%) |
0.376 |
| Age |
62.1±15.7 |
61.2±15.8 |
62.8±14.9 |
0.371 |
| Recent smoker (<1 month) |
51 (14.3%) |
38 (14.7%) |
13 (13.4%) |
0.866 |
| Diabetes |
52 (14.6%) |
35 (13.6%) |
17 (17.5%) |
0.400 |
| Obesity (BMI >30) |
42 (11.8%) |
26 (10.1%) |
16 (16.5%) |
0.100 |
| Renal dysfunctiona |
46 (13.0%) |
25 (9.7%) |
21 (21.6%) |
0.004 |
| Dialysis |
3 (0.8%) |
2 (0.8%) |
1 (1.0%) |
1.00 |
| Dyslipidemia |
66 (18.6%) |
44 (17.1%) |
22 (22.7%) |
0.225 |
| Hypertension |
159 (44.8%) |
104 (40.3%) |
55 (56.7%) |
0.006 |
| History of CVAb |
66 (18.6%) |
56 (21.7%) |
10 (10.3%) |
0.014 |
| Recent CVA |
32 (9.0%) |
26 (10.1%) |
6 (6.2%) |
0.303 |
| COPD (>moderate)c |
24 (6.8%) |
10 (3.9%) |
14 (14.4%) |
0.001 |
| Liver injuryd |
45 (12.7%) |
24 (9.3%) |
21 (21.6%) |
0.004 |
| Recent (≤12 months) AMI |
2 (0.3%) |
0 (0%) |
2 (2.1%) |
0.074 |
| Shocke |
178 (40.1%) |
109 (42.2%) |
69 (6.2%) |
<0.001 |
| Refractory shockf |
64 (18.0%) |
31 (12.0%) |
33 (34.0%) |
<0.001 |
| CPR before embolectomy |
98 (27.6%) |
50 (19.4%) |
48 (49.5%) |
<0.001 |
| NYHA IV |
163 (45.9%) |
102 (40.0%) |
61 (62.9%) |
<0.001 |
| Inotrope use |
41 (11.5%) |
22 (8.5%) |
19 (19.6%) |
0.005 |
| Poor LV functiong |
42 (11.8%) |
19 (7.4%) |
23 (23.7%) |
<0.001 |
| Arrhythmia <2 weeks of embolectomyh |
55 (15.5%) |
32 (12.4%) |
23 (23.7%) |
0.013 |
| VA-ECMO |
94 (26.5%) |
60 (23.3%) |
34 (35.1%) |
0.031 |
| Tricuspid regurgitation > grade 2 |
128 (36.1%) |
89 (34.5%) |
39 (40.2%) |
0.324 |
| Previous cardiotomy |
| CABG |
3 (0.8%) |
3 (1.2%) |
0 (0%) |
|
| Valve surgery |
2 (0.6%) |
2 (0.8%) |
0 (0%) |
|
| Previous thoracic aorta surgery |
7 (2.0%) |
2 (0.8%) |
5 (5.2%) |
|
aPositive urinary protein, S-Cr ≥1.3 mg/dL, eGFR <60 mL/min/1.73 cm2. bTransient ischemic attack, stroke (symptoms for >72 h). cFEV1.0 <75%, PaO2
<60, or PaCO2
>50 (room air), steroid use for pulmonary disease. dDiagnosis of liver cirrhosis, GOT >100 IU/L, serum total bilirubin level >1.5 g/dL. eSBP <80 mmHg, and/or cardiac index <1.8 L/min/m2
or maximal dose of inotrope use to keep SBP >80 mmHg or clinically apparent shock. fSBP <80 mmHg and/or cardiac index <1.8 L/min/m2. gPreoperative LV function indicates LV ejection fraction <30% on echocardiography or left ventriculography. hVentricular tachycardia, ventricular fibrillation, complete atrioventricular block, medication for arrhythmia. MACE, major adverse cardiovascular events. Other abbreviations as in Table 1.
Table 5.
Operative Data Related to MACE in Univariate Analysis
| |
Total
(n=355) |
No MACE
(n=258) |
MACE
(n=97) |
P value |
| Urgent or emergent op. (<24 h) |
330 (93.0%) |
243 (94.1%) |
87 (89.7%) |
0.163 |
| Concomitant CABG |
13 (3.7%) |
11 (4.3%) |
2 (2.1%) |
|
| Concomitant valve surgery |
21 (5.9%) |
10 (3.9%) |
11 (11.3%) |
|
| PFO or ASD repair |
7 (2.0%) |
7 (2.7%) |
0 (0%) |
|
| Aortic cross-clamp |
206 (50.0%) |
145 (56.2%) |
61 (62.9%) |
0.279 |
| Blood transfusion |
318 (89.6%) |
223 (86.4%) |
95 (97.9%) |
0.001 |
| CPB time (min) |
135.0±94.2 |
120.8±69.8 |
170.5±131.5 |
0.001 |
| Aortic clamp time (min) |
71.8±45.4 |
69.9±38.7 |
82.0±55.5 |
0.126 |
Aabbreviations as in Tables 2,4.
Table 6.
Postoperative Complications in Patients With MACE
| |
Total
(n=355) |
No MACE
(n=258) |
MACE
(n=97) |
P value |
| Re-exploration for bleeding |
46 (13.0%) |
14 (5.4%) |
32 (33.0%) |
<0.001 |
| Stroke |
19 (5.3%) |
0 (0%) |
19 (19.6%) |
<0.001 |
| Postoperative coma |
35 (9.9%) |
0 (0%) |
35 (36.1%) |
<0.001 |
| Postoperative renal dysfunctiona |
45 (12.7%) |
12 (4.7%) |
33 (34.0%) |
<0.001 |
| Hemodialysis |
29 (8.2%) |
6 (2.3%) |
23 (23.7%) |
<0.001 |
| Sepsis |
16 (4.5%) |
6 (2.3%) |
10 (10.3%) |
0.003 |
| Ventilator use >72 h |
94 (26.5%) |
48 (18.6%) |
46 (47.4%) |
<0.001 |
| ICU stay (days) |
10.3±15.0 |
9.7±14.4 |
12.0±16.0 |
0.200 |
| ICU stay ≥8days |
128 (36.1%) |
88 (34.1%) |
40 (41.2%) |
0.217 |
| Postoperative hospital stay (days) |
38.1±42.5 |
36.7±36.8 |
26.0±43.1 |
0.033 |
aSerum creatinine level >2.0 mg/dL or serum creatine level doubled from preoperative level, or use of dialysis for renal dysfunction. ICU, intensive care unit; MACE, major adverse cardiovascular events.
Multivariate analysis revealed the following factors as significant independent preoperative predictors of death: heart failure (P=0.013), poor LV function (P=0.007) and respiratory failure (P=0.001) (Table 7). History of stroke was a negative predictor of death (P=0.009, Exp(B)=0.277).
Table 7.
Independent Risk Factors for Operative Death and Postoperative MACE
| |
B |
Wald |
P |
Exp(B) |
95% CI |
| Risk factor for early death |
| History of stroke |
−1.285 |
6.87 |
0.009 |
0.277 |
0.106, 0.723 |
| Heart failure |
0.765 |
6.133 |
0.013 |
2.149 |
1.173, 3.937 |
| Poor LV function |
1.063 |
7.173 |
0.007 |
2.895 |
1.330, 6.302 |
| Respiratory failure |
1.740 |
11.643 |
0.001 |
5.700 |
2.098, 15.490 |
| Constant |
−2.494 |
74.414 |
|
0.083 |
|
| Risk factor for MACE |
| History of stroke |
−1.049 |
6.683 |
0.01 |
0.350 |
0.158, 0.776 |
| Poor LV function |
0.815 |
4.558 |
0.033 |
2.259 |
1.069, 4.775 |
| CPR |
1.016 |
9.772 |
0.002 |
2.762 |
1.461, 5.221 |
| Respiratory failure |
1.357 |
7.354 |
0.007 |
3.884 |
1.457, 10.358 |
| Constant |
−2.058 |
67.394 |
|
0.128 |
|
CI, confidence interval; LV, left ventricular; MACE, major adverse cardiovascular events.
Multivariate analysis for MACE revealed poor LV function (P=0.033), CPR before pulmonary embolectomy (P=0.002) and respiratory failure (P=0.007) as independent risk factors of MACE (Table 7). History of stroke was a negative risk factor of MACE (P=0.010, Exp(B)=0.350).
Discussion
Many guidelines recommend surgical pulmonary embolectomy as an important therapeutic option for patients with massive PE.8,9
The natural course of massive PE is poor, especially in patients with cardiopulmonary arrest and those requiring CPR; therefore, early recognition and early reperfusion before circulatory collapse is important for saving the lives of patients with massive PE. In our study, approximately one-quarter of the patients required CPR. Most patients underwent emergency pulmonary embolectomy. Because VA-ECMO is widely used in the emergency wards and cardiology departments in Japan, one-quarter of the patients underwent VA-ECMO for respiratory and hemodynamic support. This clinical scenario with regard to PE raises 3 clinical questions. First is a mortality rate of 20.6% and morbidity of 27.3% in patients with acute PE acceptable? Second, does this outcome reflect real-world outcomes of pulmonary embolectomy? Third, is VA-ECMO effective in saving critically ill patients with massive PE?
Stein et al performed a meta-analysis of 1,300 patients in 46 reports and found a linear relationship between death and the prevalence of cardiopulmonary arrest before pulmonary embolectomy in each study.10
Applying the linear correlation curve demonstrated by Stein et al, the mortality rate of 20.6% among 27.6% CPR patients in our study is lower than the predicted mortality in their study. In the International Cooperative Pulmonary Embolism Registry study, the mortality rate of patients experiencing cardiopulmonary arrest was 65%. The reported mortality rate of PE patients with shock was also high, at 25%.11
Keeling et al conducted a multicenter study of pulmonary embolectomy and showed that the mortality rate of patients receiving preoperative CPR was 32.1% (9/28).6
In our study, the mortality rate of patients requiring CPR before pulmonary embolectomy was similar, at 32.7% (32/98 patients).
Kilic et al7
performed a retrospective investigation of pulmonary embolectomy in the USA. They collected data from the Nationwide Inpatient Sample database using the International Classification of Disease, 9th version, and analyzed both patient and hospital factors. They obtained data from 1,050 institutions in 44 states, representing a 20% stratified sample of community hospitals. Theirs was the largest analysis of pulmonary embolectomy for acute PE. They showed a mortality rate of 27.2% among 2,709 patients over 10 years. They revealed that a high comorbidity index, including cardiac, pulmonary, hepatic, neoplastic and metabolic comorbidities, significantly increased the odds of inpatient death.
The Japanese Association of Thoracic Surgery (JATS) has been conducting an annual survey of the outcomes of thoracic and cardiovascular surgeries in Japan. It covers 97% of hospitals performing cardiovascular surgery. During the same period as our study, 591 patients underwent pulmonary embolectomy for acute PE. Therefore, the present study covered 60.1% of real-world data in Japan. Data from the JATS Annual Report showed a mortality rate of 18.8% in patients who underwent pulmonary embolectomy from 2008 to 2014 (Figure).12–18
Thus, we believe our study reflects real-world data on pulmonary embolectomy in Japan.
VA-ECMO (i.e., portable cardiopulmonary bypass) was initially proposed by Cooley et al in the 1960 s as a bridge to pulmonary embolectomy.19
Because VA-ECMO is widely used in tertiary emergency centers and teaching hospitals in Japan,20
94 patients (26.5%) in our study underwent VA-ECMO, with a survival rate of 75.5% (71/94 patients). This result concurs with previous reports on using VA-ECMO as cardiopulmonary support for acute massive PE.21
A national survey in Japan by Aso et al showed a survival rate with use of the VA-ECMO for patients with PE of 36% (127/353 patients), and 120 of the patients were discharged from the hospital after weaning from the VA-ECMO.20
Therefore, VA-ECMO and pulmonary embolectomy are very important in saving critically ill patients with acute massive PE. However, although VA-ECMO has a protective effect on cerebral function, we must note the high incidence of complications, such as limb ischemia, puncture site bleeding, retroperitoneal bleeding etc., with its use. We believe that early consultation with cardiac surgeons and a team approach including cardiac surgeons is crucially important, as is early triage of critically ill patients with acute massive PE.8,21,22
Functional recovery following treatment of acute PE is another important issue. Most studies of pulmonary embolectomy have only investigated survival after PE. Recovery of right ventricular function after pulmonary embolectomy is better in the early stages post procedure. Azari et al demonstrated a significant decrease in right ventricular diameter and systolic pulmonary arterial pressure in the early period after pulmonary embolectomy compared with thrombolysis.23
Keeling et al, using echocardiographic follow-up, showed that the improvement in right ventricular function was preserved in the midterm after pulmonary embolectomy.24
A major concern related to massive PE is cerebral complications because of prolonged preoperative anoxia and intracranial bleeding caused by thrombolytic therapy. In our study, we also focused on the neurologic outcomes of pulmonary embolectomy by assessing postoperative stroke as part of the MACE and observed that the incidence of MACE after pulmonary embolectomy was 27.3%. We also showed that poor LV function, the need for preoperative CPR and respiratory failure are predictors of MACE following pulmonary embolectomy for acute massive PE.
An interesting and somewhat contrary observation in our study was that a history of CVA was favorable for survival. Although we do not have a definitive explanation for this observation, we presume that this could be because thrombolytic therapy for massive PE is contraindicated in patients with a history of stroke according to many guidelines. Our study included 66 patients (18.6%) with a history of CVA and 32 (9.0%) patients with recent history of CVA. The study by Keeling et al also included 7.0% patients with previous CVA and 9.8% with cerebrovascular disease.6
Patients with a history of CVA may be evaluated by cardiovascular surgeons early after onset because of the contraindication to thrombolysis.
A recent study showed that the risk of bleeding in patients with intermediate-risk PEs is not negligible. Recent studies on pulmonary embolectomy have included patients with submassive PE and a risk of bleeding, such as patients with perioperative PE.5,6
Because the incidence of re-exploration for bleeding is high in many studies, early consultation with the surgical team before the use of thrombolytic agents is important.6
Study Limitations
Because data on the preoperative use of thrombolytic agents was not recorded in the JCVSD, the effect of thrombolysis on postoperative bleeding could not be analyzed. Hospital case volume and surgeon’s experience are important factors in the outcome of pulmonary embolectomy.7–9
However, analysis of that data was not possible because of the anonymous nature of the data. Because this study was retrospective, the indication for pulmonary embolectomy in each hospital may be heterogeneous.
Conclusions
The outcome of pulmonary embolectomy is acceptable considering the urgency of the situation and the preoperative comorbidities of patients. Independent risk factors for death were heart failure, poor LV function and respiratory failure by multivariate analysis. Although the incidence of MACE was as high as 27.3% in this study, this is also acceptable because of the grave preoperative condition of the patients. Independent risk factors for MACE were poor LV function, preoperative CPR and respiratory failure by multivariate analysis. History of stroke was negative risk factor for death and MACE. Early triage of patients with severe PE is important to save critically ill patients with hemodynamic instability.
Acknowledgments
The authors deeply appreciate the support of Ms. Eriko Fukuchi (Tokyo University) with the statistical analysis.
Conflict of Interest
None.
References
- 1.
Chatterjee S, Chakraborty A, Weinberg I, Kadakia M, Wilensky RL, Sardar P, et al. Thrombolysis for pulmonary embolism and risk of all-cause mortality, major bleeding, and intracranial hemorrhage: A meta-analysis. JAMA 2014; 311: 2414–2421.
- 2.
Riera-Mestre A, Jiménez D, Muriel A, Lobo JL, Moores L, Yusen RD, et al. Thrombolytic therapy and outcome of patients with an acute symptomatic pulmonary embolism. J Thromb Haemost 2012; 10: 751–759.
- 3.
Meyer G, Vicaut E, Danays T, Agnelli G, Becattini C, Beyer-Westendorf J, et al. Fibrinolysis for patients with intermediate-risk pulmonary embolism. N Engl J Med 2014; 370: 1402–1411.
- 4.
Meyer JA. Friedrich Trendelenburg and the surgical approach to massive pulmonary embolism. Arch Surg 1990; 125: 1202–1205.
- 5.
Aymard T, Kadner A, Widmer A, Basciani R, Tevaearai H, Weber A, et al. Massive pulmonary embolism: Surgical embolectomy versus thrombolytic therapy – should surgical indications be revisited? Eur J Cardiothorac Surg 2013; 43: 90–94.
- 6.
Keeling WB, Sundt T, Leacche M, Okita T, Binongo J, Lasajanak Y, et al. Outcomes after surgical pulmonary embolectomy for acute pulmonary embolus: A multi- institutional study. Ann Thorac Surg 2016; 102: 1498–1502.
- 7.
Kilic A, Shah AS, Conte JV, Mandal K, Baumgartner WA, Cameron DE, et al. Nationwide outcomes of surgical embolectomy for acute pulmonary embolism. J Thorac Cardiovasc Surg 2013; 145: 373–377.
- 8.
Konstantinides SV, Torbicki A, Agnelli G, Danchin N, Fitzmaurice D, Galiè N, et al; Task Force for the Diagnosis and Management of Acute Pulmonary Embolism of the European Society of Cardiology (ESC). 2014 ESC guidelines on the diagnosis and management of acute pulmonary embolism. Eur Heart J 2014; 35: 3033–3069.
- 9.
Kearon C, Akl EA, Comerota AJ, Prandoni P, Bounameaux H, Goldhaber SZ, et al. Antithrombotic therapy for VTE disease: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141(Suppl): e419S–e494S.
- 10.
Stein PD, Alnas M, Beemath A, Patel NR. Outcome of pulmonary embolectomy. Am J Cardiol 2007; 99: 421–423.
- 11.
Kasper W, Konstantinides S, Geibel A, Olschewski M, Heinrich F, Grosser KD, et al. Management strategies and determinants of outcome in acute major pulmonary embolism: Results of a multicenter registry. J Am Coll Cardiol 1997; 30: 1165–1171.
- 12.
Sakata R, Fujii Y, Kuwano H. Thoracic and cardiovascular surgery in Japan during 2008: Annual report by The Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg 2010; 58: 356–383.
- 13.
Sakata R, Fujii Y, Kuwano H. Thoracic and cardiovascular surgery in Japan during 2009: Annual report by the Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg 2011; 59: 636–667.
- 14.
Kuwano H, Amano J, Yokomise H. Thoracic and cardiovascular surgery in Japan during 2010: Annual report by The Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg 2012; 60: 680–708.
- 15.
Amano J, Kuwano H, Yokomise H. Thoracic and cardiovascular surgery in Japan during 2011: Annual report by The Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg 2013; 61: 578–607.
- 16.
Masuda M, Kuwano H, Okumura M, Amano J, Arai H, Endo S, et al. Thoracic and cardiovascular surgery in Japan during 2012: Annual report by The Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg 2014; 62: 734–764.
- 17.
Masuda M, Kuwano H, Okumura M, Arai H, Endo S, Doki Y, et al. Thoracic and cardiovascular surgery in Japan during 2013: Annual report by The Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg 2015; 63: 670–701.
- 18.
Masuda M, Okumura M, Doki Y, Endo S, Hirata Y, Kobayashi J, et al. Thoracic and cardiovascular surgery in Japan during 2014: Annual report by The Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg 2016; 64: 665–697.
- 19.
Cooley DA, Beall AC, Alexander JK. Acute massive pulmonary embolism: Surgical treatment using temporary cardiopulmonary bypass. JAMA 1961; 177: 283–286.
- 20.
Fukuda I, Daitoku K. Surgical embolectomy for acute pulmonary thromboembolism. Ann Vasc Dis 2017; 10: 371–377.
- 21.
Aso S, Matsui H, Fushimi K, Yasunaga H. In-hospital mortality and successful weaning from venoarterial extracorporeal membrane oxygenation: Analysis of 5,263 patients using a national inpatient database in Japan. Crit Care 2016; 20: 80.
- 22.
Goldhaber SZ. Surgical pulmonary embolectomy: The resurrection of an almost discarded operation. Tex Heart Inst J 2013; 40: 5–8.
- 23.
Azari A, Beheshti AT, Moravvej Z, Bigdelu L, Salehi M. Surgical embolectomy versus thrombolytic therapy in the management of acute massive pulmonary embolism: Short and long-term prognosis. Heart Lung 2015; 44: 335–339.
- 24.
Keeling WB, Leshnower BG, Lasajanak Y, Binongo J, Guyton RA, Halkos ME, et al. Midterm benefits of surgical pulmonary embolectomy for acute pulmonary embolus on right ventricular function. J Thorac Cardiovasc Surg 2016; 152: 872–878.
