- Email: [email protected]
Extracorporeal Life Support After Pulmonary Endarterectomy as a Bridge to Recovery or Transplantation: Lessons From 31 Consecutive Patients
Extracorporeal Life Support After Pulmonary Endarterectomy as a Bridge to Recovery or Transplantation: Lessons From 31 Consecutive Patients David Boulate, MD, Olaf Mercier, MD, PhD, Sacha Mussot, MD, Dominique Fabre, MD, Franc¸ois Stephan, MD, PhD, Franc¸ois Haddad, MD, Xavier Jaïs, MD, Philippe Dartevelle, MD, and Elie Fadel, MD, PhD Department of Thoracic, Vascular, and Heart-Lung Transplantation, and Adult Intensive Care Unit, Marie Lannelongue Hospital, Le Plessis-Robinson, Paris South University, France; Cardiovascular Medicine, Stanford Hospital, Stanford University, California; and Department of Pulmonology and Respiratory Intensive Care, Bicetre Hospital, Le Kremlin-Bicetre, Paris South University, France ˇ
ˇ
Background. Extracorporeal life support (ECLS) can be used to sustain patients having cardiorespiratory failure after pulmonary endarterectomy (PEA). We aimed to assess outcomes and to identify factors associated with short-term survival among patients who required ECLS after PEA. Methods. We reviewed the charts of consecutive patients who required ECLS after PEA between 2005 and 2013 at our institution. Patients with failed PEA were scheduled for heart-lung transplantation, and patients with potentially reversible hemodynamic or respiratory failure were given appropriate supportive care until recovery. Results. Of the 829 patients who underwent PEA, 31 (3.7%) required postoperative ECLS. Of these, 23 continued to receive support, and 8 were listed for heartlung transplantation during ECLS. Overall inhospital
survival was 48.4% (15 of 31). Of patients listed for transplantation, 2 died while on support; 4 of the 6 patients undergoing transplantation lived to hospital discharge. Of the 23 supportive care patients, 11 (47.8%) were alive at hospital discharge. The factors associated with survival were younger age (p ¼ 0.02), larger post-PEA decrease in mean pulmonary artery pressure (p ¼ 0.020), lower post-PEA total pulmonary resistance (p ¼ 0.008), and pure respiratory failure related to reperfusion edema or airway bleeding (p ¼ 0.003). Conclusions. Extracorporeal life support may be useful to support patients with complications after PEA either to recovery or to salvage transplantation.
T
(ECLS) as a bridge to recovery (BTR) or bridge to transplantation (BTT). In a recent European multicenter registry study, ECLS was used in 3.1% of patients after PEA [6]. To date, only two studies have reported outcomes after ECLS as BTR in patients with complications of PEA. In the first study [10], venovenous (VV) extracorporeal membrane oxygenation (ECMO) was used in 20 patients with respiratory failure, 6 (30%) of whom survived. In the second study [11], 7 patients with cardiorespiratory failure were managed with venoarterial (VA) ECMO, and 4 (57%) survived. Although these studies established the feasibility and benefits of ECLS as BTR in patients with cardiorespiratory failure after PEA, they were not able to identify factors associated with short-term survival. Furthermore, no studies have assessed ECLS as BTT in
he survival of patients with chronic thromboembolic pulmonary hypertension has improved with the advent of pulmonary endarterectomy (PEA). After PEA, 5-year survival rates range from 75% to 82% [1–3], a substantial improvement over the rates of 20% to 50% seen in the pre-PEA era [4]. Operative mortality has also decreased significantly, from 13% in the first case series study reported in 1983 [5] to 2.2% to 4.7% in recent studies [3, 6]. Despite these major advances, high-risk patients [7–9] may have life-threatening postoperative complications related to either pulmonary artery (PA) revascularization, such as airway bleeding and pulmonary edema, or to suboptimal or failed PEA with persistent pulmonary hypertension and right ventricular failure. These complications may require extracorporeal life support
(Ann Thorac Surg 2016;-:-–-) Ó 2016 by The Society of Thoracic Surgeons
Accepted for publication Jan 28, 2016. Address correspondence to Dr Mercier, Marie Lannelongue Hospital, Department of Thoracic and Vascular Surgery and Heart-Lung Transplantation, 133 Ave de la Resistance, Le Plessis Robinson 92350, France; email: [email protected].
Ó 2016 by The Society of Thoracic Surgeons Published by Elsevier
The Appendix can be viewed in the online version of this article [http://dx.doi.org/10.1016/j.athoracsur.2016. 01.103] on http://www.annalsthoracicsurgery.org.
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2016.01.103
2
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Abbreviations and Acronyms BTR BTT ECLS ECMO
= = = =
IQR LA MPAP PA PEA VA VV
= = = = = = =
bridge to recovery bridge to transplantation extracorporeal life support extracorporeal membrane oxygenation interquartile range left atrium mean pulmonary artery pressure pulmonary artery pulmonary endarterectomy venoarterial venovenous
patients with refractory right ventricular failure after failed PEA. Here, we report our experience of patients managed with ECLS using a BTR or BTT strategy after complicated PEA.
Patients and Methods Our Institutional Review Board approved the study and waived the requirement for written informed consent in compliance with French law on retrospective studies consisting of analyses of anonymous data.
Patients We retrospectively identified all patients who underwent PEA at the Marie Lannelongue Hospital between January 2005 and July 2013. A multidisciplinary team specialized in chronic thromboembolic pulmonary hypertension management selected patients for PEA. We identified those who had received postoperative ECLS and reviewed their records for data on medical history and perioperative data.
Pulmonary Endarterectomy Pulmonary endarterectomy was performed as described previously [9]. Briefly, cardiopulmonary bypass with bicaval and ascending aorta cannulations was established; body temperature was decreased to 20 C before cross-clamping of the aorta. Right, and then left PEA was performed with sequential circulatory arrests for distal PA recanalization. After PA closure, a Swan-Ganz catheter was introduced into the main PA trunk for postoperative hemodynamic assessments.
Classification of Cardiorespiratory Failure After PEA We retrospectively classified cardiorespiratory failures requiring ECLS as follows: (1) pure respiratory failure defined as hypoxia with pulse oximetry oxygen saturation less than 90% despite mechanical ventilation with 100% fraction of inspired oxygen, without preexisting hemodynamic failure (ie, without inotropic support or systemic hypotension, defined as mean systemic pressure less than 60 mm Hg or cardiocirculatory arrest); (2) pure hemodynamic failure defined as circulatory failure precluding
Ann Thorac Surg 2016;-:-–-
weaning off cardiopulmonary bypass or new-onset cardiogenic shock requiring inotropic support initiation or VA ECLS, without prior respiratory failure; and (3) mixed respiratory and hemodynamic failure defined as any combination of signs of both respiratory and hemodynamic failure.
Strategy and Timing of ECLS When ECLS was started in the operating room immediately after PEA, preference was given to central cannulation and use of VA-ECMO or central PA-to-left atrium (LA) Novalung (Novalung GmbH, Heilbronn, Germany), the latter being preferred for BTT. The sternum was left open in patients with major postoperative ventricular enlargement and hemodynamic deterioration at attempted sternum closure. In patients whose sternum was already closed at the time of decision for ECLS implantation, peripheral VA-ECMO was generally used. When ECLS was started after postoperative intensive care unit admission, we generally used peripheral VA-ECMO. We used VV-ECMO in 2 patients with pure respiratory failure occurring at least 7 days after PEA. Additional embolization of the systemic vasculature to the lung was successfully performed in 4 patients with airway bleeding. In case of pulmonary edema occurring during VA-ECMO, an additional LA vent line was surgically implanted to decrease pulmonary vein pressures. Excluding PA-LA Novalung, an ECLS weaning trial was performed after 48 hours of hemodynamic and respiratory stabilization, assessing the possibility of removing the ECLS by echocardiography, clinical monitoring, and laboratory tests. When weaning from ECLS proved impossible and complications such as refractory bleeding or multiorgan failure arose, a conference among staff members in charge of the patient was held and an ethical evaluation performed to determine the appropriateness of ECLS system removal. The ECLS was removed after informed consent was obtained from the family. Patients considered for transplantation fell into two categories. For patients with preoperative hemodynamic compromise due to distal PA occlusions, the need for a BTT strategy was anticipated before the PEA procedure, leading to full pretransplantation workup, with the informing of the patient and the family. In the other category of patients, the need for a BTT strategy became obvious only when weaning from ECLS proved impossible after failed PEA with refractory hemodynamic failure, persistent pulmonary hypertension, and right ventricular failure. Heart-lung transplantation was usually preferred over double-lung transplantation in case of persistent right ventricular failure and because of proximal PA frailty related to the recent PEA. However, because of donor scarcity, we now try to perform doublelung transplantation as often as possible considering the good results of double-lung transplantation in pulmonary hypertension patients. Of interest, a large amount of PA should be harvested to overcome any problem at the level of the PA anastomosis. We administered anticoagulation therapy to patients according to the institutional standard protocol for the
Ann Thorac Surg 2016;-:-–-
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
3
treatment of chronic thromboembolic pulmonary hypertension regardless of the use of ECLS. Anticoagulation therapy management was not different from all other surgical procedures requiring short-term ECLS. In case of VA-ECMO, pulmonary blood flow was preserved to decrease the risk of post-PEA pulmonary thrombosis. The maximum VA-ECMO output was determined based on the maintenance of pulsatile flow in the PA according to the PA wave pressure of the Swan-Ganz catheter and on the exhaled carbon dioxide partial pressure higher than 15 mm Hg.
were fewer than 20 patients. Otherwise, continuous variables are expressed as mean SD and compared using the Student t test. Categoric variables were described as n (%) and compared using Fisher’s exact test or the c2 test. Odds ratio of variables associated with inhospital mortality in the BTR group were estimated using univariate logistic regression. All tests were two sided, and p values less than 0.05 were considered significant. Statistical tests were performed using software R version 3.0.1 (http://www.R-project.org).
Perioperative and Midterm Hemodynamic Evaluations
Results
The preoperative hemodynamic evaluation was performed by right-side heart catheterization, except in 2 patients with a thrombus in the right heart chambers. Early postoperative hemodynamics were assessed before ECLS implantation. Early postoperative hemodynamic values were not available in patients who were not weaned from cardiopulmonary before ECLS implantation. We measured the postoperative hemodynamic effect of PEA by assessing the percentage change in cardiac index, total PA resistance, and mean PA pressure (MPAP) as follows: % change ¼ [(early postoperative value preoperative value)/preoperative value] $ 100. For the midterm hemodynamic evaluation of BTR survivors, systolic PA pressure was assessed by echocardiography at hospital discharge. The MPAP was computed as reported by Chemla and associates [12]: MPAP ¼ (systolic PA pressure $ 0.61) þ 2 mm Hg. Rightside heart catheterization was performed routinely 6 months after PEA.
Patient Characteristics
Statistical Analysis Continuous variables are described as median and interquartile range (IQR) and compared using the nonparametric Wilcoxon rank sum test when groups
Of 829 patients who underwent PEA during the study period, 31 (3.7%) required postoperative ECLS (Fig 1), including 8 listed for transplantation immediately after PEA (BTT group) and 23 given supportive care while awaiting recovery (BTR group). Tables 1 and 2 report patients’ characteristics. Preoperative dobutamine was required for 6 patients (19.4%; Table 2).
BTT and BTR Group Comparisons Compared with the BTR patients, patients in the BTT group were younger and more often had risk factors for poor PEA outcomes (Table 2). All BTT patients had Jamieson type 4 thromboembolic disease [13], whereas 91.3% of BTR patients had Jamieson type 2 or 3. A transplantation workup was performed before PEA in 2 BTT patients (Appendix Table 1). The ECLS was implanted in the operating room immediately after PEA in 7 patients (87.5%) of the BTT group and in 6 patients (26%; Figs 2 and 3) of the BTR group. The ECLS was implanted after intensive care unit admission in 1 BTT patient (12.5%), 10 days after PEA, and in 17 BTR patients (74%) at a median of 3 days (IQR: 0 to 15) after PEA. In all 8 BTT patients, ECLS was implanted because of hemodynamic failure, secondary to persistent pulmonary Fig 1. Flowchart of patients requiring surgery for chronic thromboembolic pulmonary hypertension (CTEPH). (DLT ¼ doublelung transplantation; ECLS ¼ extracorporeal life support; HLT ¼ heart-lung transplantation; Mty ¼ 30-day mortality; PEA ¼ pulmonary endarterectomy.)
4
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Ann Thorac Surg 2016;-:-–-
Table 1. Characteristics of Overall Pulmonary Endarterectomy Cohort Compared With Patients Who Underwent Post–Pulmonary Endarterectomy Extracorporeal Life Support Between 2005 and 2013 Characteristics Age, years Male Body surface area, m2 NYHA status (n ¼ 395) Class I Class II Class III Class IV Reoperative hemodynamics MPAP, mm Hg TPR, dynes $ s $ cm5 Cardiac index, L $ min-1 $ m-2 Central venous pressure, mm Hg PH-targeted therapy before PEA Preoperative dobutamine Operative times, minutes Cardiopulmonary bypass Aorta cross-clamp time Circulatory arrest time Jamieson classification Type 1 Type 2 Type 3 Type 4
PEA w/o ECLS (n ¼ 798)
Post-PEA ECLS (n ¼ 31)
p Value
59.33 15.4 397 1.83 0.22
49.96 16.11 14 1.83 0.19
0.01 0.87 0.99
4 (1.0) 80 (20.3) 250 (63.3) 59 (14.9)
0 (0) 4 (12.9) 11 (35.5) 16 (51.6) 53.0 9.02 1116 339.7 2.21 0.47 10.9 6.7 101 6
<0.001 <0.001 0.04 0.03 0.21 <0.001
219 42 101 27 29 12
243 56 116 32 35 13
0.02 0.002 0.012 <0.001
33 234 67 16
2 11 10 8
42.8 797.9 2.56 7.9
12.8 366.4 0.63 5.1 77 2
Values are mean SD, n, or n (%). ECLS ¼ extracorporeal life support; MPAP ¼ mean pulmonary artery pressure; endarterectomy; PH ¼ pulmonary hypertension; SD ¼ standard deviation;
hypertension and right ventricular failure (Fig 2A). However, in the BTR group, ECLS was implanted mainly because of respiratory failures (Fig 2B). Figure 3 shows the devices used for primary ECLS. Individual data on ECLS devices, device changes, and patient outcomes in the BTT and BTR groups are reported in Appendix Tables 1 and 2.
Short-Term Outcomes Of the 31 patients, 15 (48.4%) survived to hospital discharge. In the BTT group, inhospital survival was 50% (4 of 8); 6 patients underwent transplantation a median of 6.5 days (IQR: 1 to 8) after PEA, and 4 of them survived to hospital discharge. The 2 patients who were listed but not receiving transplants died before organ allocation; causes of death are reported in Appendix Table 1. In the BTR group, inhospital survival was 47.8% (11 of 23) and median ECLS duration was 6.5 days (IQR: 2 to 18). Of the 12 patients who died in the BTR group, 2 underwent repeat VA-ECMO implantation 2 days after weaning off the first device but subsequently died, 1 of refractory sepsis and 1 of massive stroke (patients 18 and 24, respectively; Appendix Table 2). The other 10 patients died either during ECLS (n ¼ 1) of multiorgan failure, or after ECLS removal, of refractory bleeding (n ¼ 7), multiorgan failure (n ¼ 2), or recurrent PA
NYHA ¼ New York Heart Association; TPR ¼ total pulmonary artery resistance;
PEA ¼ pulmonary w/o ¼ without.
thrombosis (n ¼ 1). The most common adverse events during ECLS were bleeding and infections (Table 3).
Factors Affecting Inhospital Survival The BTT group was too small for statistical analysis. In the BTR group, survivors were younger, had larger early MPAP decreases, lower post-PEA total PA resistance values, and a higher rate of pure respiratory failure compared with nonsurvivors (all p < 0.05; Table 4). The odds ratios for inhospital death were 1.87 (95% confidence interval: 1.16 to 3.74) per additional 100 dyne $ s $ cm-5 of early postoperative total PA resistance (p ¼ 0.03) and 0.66 (95% confidence interval: 0.41 to 0.91) per additional 10 mm Hg of early postoperative MPAP decrease (p ¼ 0.04). None of the other preoperative or intraoperative variables differed significantly between survivors and nonsurvivors (Table 4).
Midterm Hemodynamic Factors After Weaning From ECLS The MPAP values at hospital discharge or 6 months after PEA were available for the 11 BTR patients who survived to hospital discharge. The early postoperative decrease in MPAP persisted in all patients and even tended to become greater over time.
Ann Thorac Surg 2016;-:-–-
5
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Table 2. Preoperative and Intraoperative Extracorporeal Life Support Patient Characteristics Characteristics
Patients With ECLS (n ¼ 31)
BTT Group (n ¼ 8)
BTR Group (n ¼ 23)
52 (39–61) 14 1.8 (1.7–2.0)
28 (21–52) 4 1.8 (1.7–1.9)
54 (50–68) 10 1.8 (1.7–2.0)
0 (0) 4 (12.9) 11 (35.5) 16 (51.6)
0 0 3 5
0 4 8 11
52 (45–61) 1,017 (855–1,431) 2.3 (1.8–2.6) 9 (6–14.5) 10/301 6 8 22 11 1 5 4 1 2
50 (45–61) 1,100 (883–1,150) 2.1 (1.9–2.6) 6 (5–16) 3 1 2 4 6
53 (46–60) 1,016 (849–1,438) 2.4 (1.8–2.6) 9 (6–14) 7 5 6
3 2 1 2
0 0
237 (198–283) 112 (91–130) 34 (29–45)
268 (214–304) 105 (83–124) 32 (20–46)
227 (198–268) 119 (95–132) 34 (30–46)
2 11 10 8
0 0 0 8
2 11 10 0
0.13 0.30 0.59 . . . . .
3 1
0 0
3 1
. .
Age, years Male Body surface area, m2 NYHA status (n ¼ 395) Class I Class II Class III Class IV Preoperative hemodynamics MPAP, mm Hg TPR, dynes $ s $ cm5 Cardiac index, L $ min1 $ m2 Central venous pressure, mm Hg PH-targeted therapy Preoperative dobutamine Patent foramen ovale History of pulmonary embolism Risk factors Splenectomy Central intravenous line Thrombophilia Prior PEA Anticipated need for HLT Operative times, minutes Cardiopulmonary bypass Aorta cross-clamp time Circulatory arrest time Jamieson classification Type 1 Type 2 Type 3 Type 4 Associated surgical procedures Persistent foramen ovale closure Coronary artery bypass graft a
5 1 2
p Value 0.04 0.99 0.61 0.69a
0.76 0.97 0.76 0.33 0.99 0.99 0.99 0.18 0.01 . 0.09 0.27 . .
Fisher’s exact test, number of class IV patients versus other patients.
Values are median (interquartile range), n, or n (%). BTR ¼ bridge to recovery; BTT ¼ bridge to transplant; ECLS ¼ extracorporeal life support; HLT ¼ heart-lung transplantation; MPAP ¼ mean pulmonary artery pressure; NYHA ¼ New York Heart Association; PEA ¼ pulmonary endarterectomy; PH ¼ pulmonary hypertension; TPR ¼ total pulmonary artery resistance.
Long-Term Outcome of Transplanted Patients Among the 8 patients who received transplants who were bridged using ECLS, 4 were discharged from the hospital and followed. Two patients, died at 3 and 45 months, of graft dysfunction and breast cancer, respectively. The remaining 2 were still alive at 38 and 6 months.
Comment In patients managed with ECLS as BTR, factors associated with inhospital survival were younger age, greater early improvement in hemodynamic variables, and respiratory failure as the reason for ECLS. In addition, we
demonstrated that ECLS as BTT strategy was successful in 4 of 8 patients with failed PEA. Pneumonia and airway bleeding were major causes of morbidity and mortality in patients requiring ECLS after PEA. Thistlesthwaite and colleagues [10] provided the historical series of 20 patients managed with VV-ECMO after PEA complicated with respiratory failure, with a survival of 30%. Berman and associates [11] reported 7 patients with severe cardiorespiratory failure after PEA who were managed using VA-ECMO as BTR. In these studies, pneumonia and airway bleeding were the leading causes of morbidity and mortality. Consistently, we found that pneumonia was a major cause of morbidity in
6
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Ann Thorac Surg 2016;-:-–-
Fig 2. Indication for extracorporeal life support (ECLS) implantation: (A) in operating room (OR); and (B) in intensive care unit (ICU). Dark gray areas indicate hemodynamic failure; lighter gray areas indicate respiratory failure; patterned areas indicate respiratory and hemodynamic failure. (BTR ¼ bridge to recovery; BTT ¼ bridge to transplantation; PEA ¼ pulmonary endarterectomy.)
patients with post-PEA ECLS. Bleeding was also a leading cause of morbidity but, differing from the study by Thistlesthwaite and colleagues [10], bleeding in the airway was not the most common site of bleeding. That is potentially due to the pulmonary circulation discharge with our preference for VA-ECMO decreasing the risk for airway hemorrhage; also, higher levels of anticoagulation agents required for VA-ECMO compared with VV-ECMO increased the risk for overall hemorrhage. Of importance, the early hemodynamic improvement in patients with successful BTR persisted in the midterm, confirming the benefit of PEA in patients with
Fig 3. Type of devices depending on strategies and implantation sites as well as inhospital survival. (BTR ¼ bridge to recovery; BTT ¼ bridge to transplantation; ECMO ¼ extracorporeal membrane oxygenation; ICU ¼ intensive care unit; OR ¼ operating room; PA-LA ¼ pulmonary artery to left atrium; PEA ¼ pulmonary endarterectomy; VA ¼ venoarterial; VV ¼ venovenous.)
severe chronic thromboembolic pulmonary hypertension [14]. This observation is consistent with microvascular disease reversal within a few weeks after PEA as previously suggested in human and animal models [15–19]. Furthermore, good outcomes obtained in patients with pure respiratory failure were obtained chiefly using VA-ECMO instead of VV-ECMO. Our results suggest that VA-ECMO may also benefit patients with pure respiratory failure. As VA-ECMO decreases the transpulmonary blood flow, it may help to control both airway bleeding from the pulmonary circulation and pulmonary edema secondary to reperfusion, which are
Ann Thorac Surg 2016;-:-–-
7
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Table 3. Complications in the Bridge to Transplantation and Bridge to Recovery Groups Complications Bleeding Thoracic Airways Cannulation sites Gastrointestinal Infections Lung Cannulation sites Other Renal replacement therapy Limb ischemia Stroke Heparin-induced thrombocytopenia Hemolysis a
All Patients
BTT Group (n ¼ 8)
BTR Group (n ¼ 23)
p Value
17 (54.8) 8 5 4 1 16 (51.6) 13 2 2 8 (25.8) 4 (12.9) 4 (12.9) 3 (9.7) 6 (19.3)
2 (25.0) 2 0 0 0 2 (25.0) 2 0 0 1 (12.5) 1 (12.5) 2 (25.0) 1 (12.5) 1 (15.5)
15 (65.2) 6 5 4 1 14 (60.9) 11 2 2 7 (30.4) 3 (13.0) 2 (8.7) 2 (8.9) 5 (21.7)
0.10a . . . . 0.11a . . . . . . . .
Bridge to recovery (BTR) versus bridge to transplantation (BTT).
Values are n (%) or n.
the leading causes of morbidity and mortality among patients requiring ECLS after PEA [10, 11]. Moreover, VA-ECMO is more likely to protect the right ventricle from the increased PA pressure observed in severe acute respiratory distress syndrome [20]. However, the benefits of VA-ECMO should be balanced with the heightened risk of bleeding, stroke, and limb ischemia related to VA cannulations and the higher level of anticoagulation therapy. Heparin-coated circuits, systematic limb reperfusion, and ischemia screening should help to decrease these risks. We used PA-LA Novalung in the BTT strategy. Potential advantages of this strategy are the absence of peripheral vessel cannulation and facilitated rehabilitation of conscious patients owing to central cannulation. The PA-LA Novalung was used as a BTT strategy for patients with end-stage pulmonary hypertension and refractory hemodynamic failure [21, 22] but has not been yet compared with VA-ECMO. For patients with poor early hemodynamic response to PEA, factors responsible for hemodynamic failure were not reversible despite ECLS, explaining poor outcomes. For patients potentially eligible for salvage transplantation, a transplantation workup before PEA seemed beneficial by facilitating the listing and organ allocation process. Heart-lung transplantation was our primary choice because we considered right ventricular failure to be irreversible—heart-lung transplantation decreases the chance for organ allocation compared with double-lung transplantation in the French allocation organ system. That is one of the reasons leading our team to move through heart-lung transplantation. Indeed, our experience with VA-ECLS before heart-lung transplantation for patients with severe pulmonary hypertension increased during the study (unpublished data) and allowed us to think that heart-lung transplantation is feasible in patients with severe pulmonary hypertension and right
ventricle failure. However, posttransplantation ECLS is frequent, potentially increasing morbidity. Our results help to target unmet needs for managing life-threatening complications after PEA. For patients with a suboptimal response to PEA who cannot be weaned from short-term ECLS and are not eligible for transplantation, the development of destination therapies based on adapted ECLS devices might provide new therapeutic options in the future. Also, anticipating the need for potential post-PEA transplantation should help to shorten the time to organ allocation. Finally, the place of transluminal PA angioplasty has to be defined in patients with hemodynamic failure and persistent pulmonary hypertension after PEA. We acknowledge the limitations of our study, which was retrospective and included a small number of patients. Respiratory failure and hemodynamic failure were defined retrospectively. The midterm hemodynamic outcome was limited because many patients were from foreign countries. In addition, comparison between VV-ECMO and VA-ECMO was not feasible owing to the small number of patients treated with VV-ECMO. Therefore, our positive results obtained using VA-ECMO in patients with respiratory failure should be confirmed by further studies. In conclusion, ECLS has an important role in the management of patients with severe chronic thromboembolic pulmonary hypertension, allowing them to benefit from lasting hemodynamic improvement. Extracorporeal life support is efficient as a bridge to salvage transplantation after failed PEA. A pre-PEA transplantation workup may improve the chances of survival through salvage transplantation in young patients with distal disease deemed operable. Patients with suboptimal PEA who have hemodynamic failure and are not eligible for transplantation have poor prognoses with short-term ECLS. Further development of destination therapies and
8
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Ann Thorac Surg 2016;-:-–-
Table 4. Comparison of Survivors and Nonsurvivors in Bridge to Recovery Group Variables Age, years Body surface area, m2 Male MPAP, mm Hg Preoperative Percent decrease TPR, dyne $ s $ cm5 Preoperative Percent decrease Cardiac index, L $ min1 $ m2 Preoperative Percent increase Preoperative CVP, mm Hg Operative times, minutes Cardiopulmonary bypass Patent foramen closure Thromboembolic disease Type 2 Preoperative characteristics NYHA class I/II/III/IV Pulmonary embolism history Preoperative dobutamine Postoperative course Pure respiratory failure Mixed cardiorespiratory failure Pure hemodynamic failure
BTR group (n ¼ 23)
Survivors (n ¼ 11)
Nonsurvivors (n ¼ 12)
p Value
54 (49.5–67.5) 1.79 (1.67–1.98) 13
52 (46.0–53.5) 1.83 (1.71–1.96) 6
67.5 (56–70.3) 1.775 (1.67–2.03) 7
0.012 0.64 0.99
53.0 (45.5–59.8) 27.2 (3.1–50)
58.5 (47.5–64) 46.8 (44.4–57.1)
50.5 (46.25–54) 4.2 (11.9–27.2)
0.15 0.02
1,016 (849–1,438) 25.7 (5.4–42.3)
996 (802–1,338) 35.8 (27.1–46.2)
1,124 (890–1,447) 11.2 (4.0 to 31.7)
0.54 0.11
2.36 (1.82;2.59) 5.4 (21.1 to 5.9) 9.0 (6–14)
2.54 (2.16–2.68) 7.1 (17.3 to 0.7) 9 (8–12)
2.23 (1.59–2.49) 3.4 (28.8 to 11.4) 11 (6–18.5)
0.08 0.82 0.54
227 (198–268) 3
217 (197–237) 3
263 (200–287) 0
0.15 0.10
11
6
5
0.68
0/4/8/11 18 18
0/4/3/4 8 10
0/0/5/7 10 8
0.097 0.59 0.32
13 7 3
10 0 1
3 7 2
0.003 0.005 0.99
Values are median (interquartile range) or n. BTR ¼ bridge to recovery; CVP ¼ central venous pressure; Association; TPR ¼ total pulmonary artery resistance.
MPAP ¼ mean pulmonary artery pressure;
strategies decreasing anticoagulation therapy and lung infections should improve outcomes.
6.
The authors wish to thank Tai Pham, MD, for his contribution to the statistical analysis review. 7.
References 1. Jamieson SW, Kapelanski DP, Sakakibara N, et al. Pulmonary endarterectomy: experience and lessons learned in 1, 500 cases. Ann Thorac Surg 2003;76:1457–62. 2. Freed DH, Thomson BM, Berman M, et al. Survival after pulmonary thromboendarterectomy: effect of residual pulmonary hypertension. J Thorac Cardiovasc Surg 2011;141: 383–7. 3. Madani MM, Auger WR, Pretorius V, et al. Pulmonary endarterectomy: recent changes in a single institution’s experience of more than 2,700 patients. Ann Thorac Surg 2012;94:97–103. 4. Riedel M, Stanek V, Widimsky J, Prerovsky I. Longterm follow-up of patients with pulmonary thromboembolism. Late prognosis and evolution of hemodynamic and respiratory data. Chest 1982;81:151–8. 5. Moser KM, Spragg RG, Utley J, Daily PO. Chronic thrombotic obstruction of major pulmonary arteries: results of
8. 9. 10.
11. 12. 13.
NYHA ¼ New York Heart
thromboendarterectomy in 15 patients. Ann Intern Med 1983;99:299–305. Mayer E, Jenkins D, Lindner J, et al. Surgical management and outcome of patients with chronic thromboembolic pulmonary hypertension: results from an international prospective registry. J Thorac Cardiovasc Surg 2011;141: 702–10. Bonderman D, Skoro-Sajer N, Jakowitsch J, et al. Predictors of outcome in chronic thromboembolic pulmonary hypertension. Circulation 2007;115:2153–8. Condliffe R, Kiely DG, Gibbs JS, et al. Prognostic and aetiological factors in chronic thromboembolic pulmonary hypertension. Eur Respir J 2009;33:332–8. Dartevelle P, Fadel E, Mussot S, et al. Chronic thromboembolic pulmonary hypertension. Eur Respir J 2004;23:637–48. Thistlethwaite PA, Madani MM, Kemp AD, Hartley M, Auger WR, Jamieson SW. Venovenous extracorporeal life support after pulmonary endarterectomy: indications, techniques, and outcomes. Ann Thorac Surg 2006;82:2139–45. Berman M, Tsui S, Vuylsteke A, et al. Successful extracorporeal membrane oxygenation support after pulmonary thromboendarterectomy. Ann Thorac Surg 2008;86:1261–7. Chemla D, Castelain V, Humbert M, et al. New formula for predicting mean pulmonary artery pressure using systolic pulmonary artery pressure. Chest 2004;126:1313–7. Thistlethwaite PA, Mo M, Madani MM, et al. Operative classification of thromboembolic disease determines outcome after pulmonary endarterectomy. J Thorac Cardiovasc Surg 2002;124:1203–11.
Ann Thorac Surg 2016;-:-–-
14. Thistlethwaite PA, Kemp A, Du L, Madani MM, Jamieson SW. Outcomes of pulmonary endarterectomy for treatment of extreme thromboembolic pulmonary hypertension. J Thorac Cardiovasc Surg 2006;131:307–13. 15. Moser KM, Bloor CM. Pulmonary vascular lesions occurring in patients with chronic major vessel thromboembolic pulmonary hypertension. Chest 1993;103:685–92. 16. Moser KM, Metersky ML, Auger WR, Fedullo PF. Resolution of vascular steal after pulmonary thromboendarterectomy. Chest 1993;104:1441–4. 17. Fadel E, Michel RP, Eddahibi S, et al. Regression of postobstructive vasculopathy after revascularization of chronically obstructed pulmonary artery. J Thorac Cardiovasc Surg 2004;127:1009–17. 18. Fadel E, Wijtenburg E, Michel R, et al. Regression of the systemic vasculature to the lung after removal of pulmonary
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
19.
20. 21.
22.
9
artery obstruction. Am J Respir Crit Care Med 2006;173: 345–9. Boulate D, Perros F, Dorfmuller P, et al. Pulmonary microvascular lesions regress in reperfused chronic thromboembolic pulmonary hypertension. J Heart Lung Transplant 2015;34:457–67. Zapol WM, Snider MT. Pulmonary hypertension in severe acute respiratory failure. N Engl J Med 1977;296: 476–80. Strueber M, Hoeper MM, Fischer S, et al. Bridge to thoracic organ transplantation in patients with pulmonary arterial hypertension using a pumpless lung assist device. Am J Transplant 2009;9:853–7. Schmid C, Philipp A, Hilker M, et al. Bridge to lung transplantation through a pulmonary artery to left atrial oxygenator circuit. Ann Thorac Surg 2008;85:1202–5.
ˇ
Background. Extracorporeal life support (ECLS) can be used to sustain patients having cardiorespiratory failure after pulmonary endarterectomy (PEA). We aimed to assess outcomes and to identify factors associated with short-term survival among patients who required ECLS after PEA. Methods. We reviewed the charts of consecutive patients who required ECLS after PEA between 2005 and 2013 at our institution. Patients with failed PEA were scheduled for heart-lung transplantation, and patients with potentially reversible hemodynamic or respiratory failure were given appropriate supportive care until recovery. Results. Of the 829 patients who underwent PEA, 31 (3.7%) required postoperative ECLS. Of these, 23 continued to receive support, and 8 were listed for heartlung transplantation during ECLS. Overall inhospital
survival was 48.4% (15 of 31). Of patients listed for transplantation, 2 died while on support; 4 of the 6 patients undergoing transplantation lived to hospital discharge. Of the 23 supportive care patients, 11 (47.8%) were alive at hospital discharge. The factors associated with survival were younger age (p ¼ 0.02), larger post-PEA decrease in mean pulmonary artery pressure (p ¼ 0.020), lower post-PEA total pulmonary resistance (p ¼ 0.008), and pure respiratory failure related to reperfusion edema or airway bleeding (p ¼ 0.003). Conclusions. Extracorporeal life support may be useful to support patients with complications after PEA either to recovery or to salvage transplantation.
T
(ECLS) as a bridge to recovery (BTR) or bridge to transplantation (BTT). In a recent European multicenter registry study, ECLS was used in 3.1% of patients after PEA [6]. To date, only two studies have reported outcomes after ECLS as BTR in patients with complications of PEA. In the first study [10], venovenous (VV) extracorporeal membrane oxygenation (ECMO) was used in 20 patients with respiratory failure, 6 (30%) of whom survived. In the second study [11], 7 patients with cardiorespiratory failure were managed with venoarterial (VA) ECMO, and 4 (57%) survived. Although these studies established the feasibility and benefits of ECLS as BTR in patients with cardiorespiratory failure after PEA, they were not able to identify factors associated with short-term survival. Furthermore, no studies have assessed ECLS as BTT in
he survival of patients with chronic thromboembolic pulmonary hypertension has improved with the advent of pulmonary endarterectomy (PEA). After PEA, 5-year survival rates range from 75% to 82% [1–3], a substantial improvement over the rates of 20% to 50% seen in the pre-PEA era [4]. Operative mortality has also decreased significantly, from 13% in the first case series study reported in 1983 [5] to 2.2% to 4.7% in recent studies [3, 6]. Despite these major advances, high-risk patients [7–9] may have life-threatening postoperative complications related to either pulmonary artery (PA) revascularization, such as airway bleeding and pulmonary edema, or to suboptimal or failed PEA with persistent pulmonary hypertension and right ventricular failure. These complications may require extracorporeal life support
(Ann Thorac Surg 2016;-:-–-) Ó 2016 by The Society of Thoracic Surgeons
Accepted for publication Jan 28, 2016. Address correspondence to Dr Mercier, Marie Lannelongue Hospital, Department of Thoracic and Vascular Surgery and Heart-Lung Transplantation, 133 Ave de la Resistance, Le Plessis Robinson 92350, France; email: [email protected].
Ó 2016 by The Society of Thoracic Surgeons Published by Elsevier
The Appendix can be viewed in the online version of this article [http://dx.doi.org/10.1016/j.athoracsur.2016. 01.103] on http://www.annalsthoracicsurgery.org.
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2016.01.103
2
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Abbreviations and Acronyms BTR BTT ECLS ECMO
= = = =
IQR LA MPAP PA PEA VA VV
= = = = = = =
bridge to recovery bridge to transplantation extracorporeal life support extracorporeal membrane oxygenation interquartile range left atrium mean pulmonary artery pressure pulmonary artery pulmonary endarterectomy venoarterial venovenous
patients with refractory right ventricular failure after failed PEA. Here, we report our experience of patients managed with ECLS using a BTR or BTT strategy after complicated PEA.
Patients and Methods Our Institutional Review Board approved the study and waived the requirement for written informed consent in compliance with French law on retrospective studies consisting of analyses of anonymous data.
Patients We retrospectively identified all patients who underwent PEA at the Marie Lannelongue Hospital between January 2005 and July 2013. A multidisciplinary team specialized in chronic thromboembolic pulmonary hypertension management selected patients for PEA. We identified those who had received postoperative ECLS and reviewed their records for data on medical history and perioperative data.
Pulmonary Endarterectomy Pulmonary endarterectomy was performed as described previously [9]. Briefly, cardiopulmonary bypass with bicaval and ascending aorta cannulations was established; body temperature was decreased to 20 C before cross-clamping of the aorta. Right, and then left PEA was performed with sequential circulatory arrests for distal PA recanalization. After PA closure, a Swan-Ganz catheter was introduced into the main PA trunk for postoperative hemodynamic assessments.
Classification of Cardiorespiratory Failure After PEA We retrospectively classified cardiorespiratory failures requiring ECLS as follows: (1) pure respiratory failure defined as hypoxia with pulse oximetry oxygen saturation less than 90% despite mechanical ventilation with 100% fraction of inspired oxygen, without preexisting hemodynamic failure (ie, without inotropic support or systemic hypotension, defined as mean systemic pressure less than 60 mm Hg or cardiocirculatory arrest); (2) pure hemodynamic failure defined as circulatory failure precluding
Ann Thorac Surg 2016;-:-–-
weaning off cardiopulmonary bypass or new-onset cardiogenic shock requiring inotropic support initiation or VA ECLS, without prior respiratory failure; and (3) mixed respiratory and hemodynamic failure defined as any combination of signs of both respiratory and hemodynamic failure.
Strategy and Timing of ECLS When ECLS was started in the operating room immediately after PEA, preference was given to central cannulation and use of VA-ECMO or central PA-to-left atrium (LA) Novalung (Novalung GmbH, Heilbronn, Germany), the latter being preferred for BTT. The sternum was left open in patients with major postoperative ventricular enlargement and hemodynamic deterioration at attempted sternum closure. In patients whose sternum was already closed at the time of decision for ECLS implantation, peripheral VA-ECMO was generally used. When ECLS was started after postoperative intensive care unit admission, we generally used peripheral VA-ECMO. We used VV-ECMO in 2 patients with pure respiratory failure occurring at least 7 days after PEA. Additional embolization of the systemic vasculature to the lung was successfully performed in 4 patients with airway bleeding. In case of pulmonary edema occurring during VA-ECMO, an additional LA vent line was surgically implanted to decrease pulmonary vein pressures. Excluding PA-LA Novalung, an ECLS weaning trial was performed after 48 hours of hemodynamic and respiratory stabilization, assessing the possibility of removing the ECLS by echocardiography, clinical monitoring, and laboratory tests. When weaning from ECLS proved impossible and complications such as refractory bleeding or multiorgan failure arose, a conference among staff members in charge of the patient was held and an ethical evaluation performed to determine the appropriateness of ECLS system removal. The ECLS was removed after informed consent was obtained from the family. Patients considered for transplantation fell into two categories. For patients with preoperative hemodynamic compromise due to distal PA occlusions, the need for a BTT strategy was anticipated before the PEA procedure, leading to full pretransplantation workup, with the informing of the patient and the family. In the other category of patients, the need for a BTT strategy became obvious only when weaning from ECLS proved impossible after failed PEA with refractory hemodynamic failure, persistent pulmonary hypertension, and right ventricular failure. Heart-lung transplantation was usually preferred over double-lung transplantation in case of persistent right ventricular failure and because of proximal PA frailty related to the recent PEA. However, because of donor scarcity, we now try to perform doublelung transplantation as often as possible considering the good results of double-lung transplantation in pulmonary hypertension patients. Of interest, a large amount of PA should be harvested to overcome any problem at the level of the PA anastomosis. We administered anticoagulation therapy to patients according to the institutional standard protocol for the
Ann Thorac Surg 2016;-:-–-
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
3
treatment of chronic thromboembolic pulmonary hypertension regardless of the use of ECLS. Anticoagulation therapy management was not different from all other surgical procedures requiring short-term ECLS. In case of VA-ECMO, pulmonary blood flow was preserved to decrease the risk of post-PEA pulmonary thrombosis. The maximum VA-ECMO output was determined based on the maintenance of pulsatile flow in the PA according to the PA wave pressure of the Swan-Ganz catheter and on the exhaled carbon dioxide partial pressure higher than 15 mm Hg.
were fewer than 20 patients. Otherwise, continuous variables are expressed as mean SD and compared using the Student t test. Categoric variables were described as n (%) and compared using Fisher’s exact test or the c2 test. Odds ratio of variables associated with inhospital mortality in the BTR group were estimated using univariate logistic regression. All tests were two sided, and p values less than 0.05 were considered significant. Statistical tests were performed using software R version 3.0.1 (http://www.R-project.org).
Perioperative and Midterm Hemodynamic Evaluations
Results
The preoperative hemodynamic evaluation was performed by right-side heart catheterization, except in 2 patients with a thrombus in the right heart chambers. Early postoperative hemodynamics were assessed before ECLS implantation. Early postoperative hemodynamic values were not available in patients who were not weaned from cardiopulmonary before ECLS implantation. We measured the postoperative hemodynamic effect of PEA by assessing the percentage change in cardiac index, total PA resistance, and mean PA pressure (MPAP) as follows: % change ¼ [(early postoperative value preoperative value)/preoperative value] $ 100. For the midterm hemodynamic evaluation of BTR survivors, systolic PA pressure was assessed by echocardiography at hospital discharge. The MPAP was computed as reported by Chemla and associates [12]: MPAP ¼ (systolic PA pressure $ 0.61) þ 2 mm Hg. Rightside heart catheterization was performed routinely 6 months after PEA.
Patient Characteristics
Statistical Analysis Continuous variables are described as median and interquartile range (IQR) and compared using the nonparametric Wilcoxon rank sum test when groups
Of 829 patients who underwent PEA during the study period, 31 (3.7%) required postoperative ECLS (Fig 1), including 8 listed for transplantation immediately after PEA (BTT group) and 23 given supportive care while awaiting recovery (BTR group). Tables 1 and 2 report patients’ characteristics. Preoperative dobutamine was required for 6 patients (19.4%; Table 2).
BTT and BTR Group Comparisons Compared with the BTR patients, patients in the BTT group were younger and more often had risk factors for poor PEA outcomes (Table 2). All BTT patients had Jamieson type 4 thromboembolic disease [13], whereas 91.3% of BTR patients had Jamieson type 2 or 3. A transplantation workup was performed before PEA in 2 BTT patients (Appendix Table 1). The ECLS was implanted in the operating room immediately after PEA in 7 patients (87.5%) of the BTT group and in 6 patients (26%; Figs 2 and 3) of the BTR group. The ECLS was implanted after intensive care unit admission in 1 BTT patient (12.5%), 10 days after PEA, and in 17 BTR patients (74%) at a median of 3 days (IQR: 0 to 15) after PEA. In all 8 BTT patients, ECLS was implanted because of hemodynamic failure, secondary to persistent pulmonary Fig 1. Flowchart of patients requiring surgery for chronic thromboembolic pulmonary hypertension (CTEPH). (DLT ¼ doublelung transplantation; ECLS ¼ extracorporeal life support; HLT ¼ heart-lung transplantation; Mty ¼ 30-day mortality; PEA ¼ pulmonary endarterectomy.)
4
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Ann Thorac Surg 2016;-:-–-
Table 1. Characteristics of Overall Pulmonary Endarterectomy Cohort Compared With Patients Who Underwent Post–Pulmonary Endarterectomy Extracorporeal Life Support Between 2005 and 2013 Characteristics Age, years Male Body surface area, m2 NYHA status (n ¼ 395) Class I Class II Class III Class IV Reoperative hemodynamics MPAP, mm Hg TPR, dynes $ s $ cm5 Cardiac index, L $ min-1 $ m-2 Central venous pressure, mm Hg PH-targeted therapy before PEA Preoperative dobutamine Operative times, minutes Cardiopulmonary bypass Aorta cross-clamp time Circulatory arrest time Jamieson classification Type 1 Type 2 Type 3 Type 4
PEA w/o ECLS (n ¼ 798)
Post-PEA ECLS (n ¼ 31)
p Value
59.33 15.4 397 1.83 0.22
49.96 16.11 14 1.83 0.19
0.01 0.87 0.99
4 (1.0) 80 (20.3) 250 (63.3) 59 (14.9)
0 (0) 4 (12.9) 11 (35.5) 16 (51.6) 53.0 9.02 1116 339.7 2.21 0.47 10.9 6.7 101 6
<0.001 <0.001 0.04 0.03 0.21 <0.001
219 42 101 27 29 12
243 56 116 32 35 13
0.02 0.002 0.012 <0.001
33 234 67 16
2 11 10 8
42.8 797.9 2.56 7.9
12.8 366.4 0.63 5.1 77 2
Values are mean SD, n, or n (%). ECLS ¼ extracorporeal life support; MPAP ¼ mean pulmonary artery pressure; endarterectomy; PH ¼ pulmonary hypertension; SD ¼ standard deviation;
hypertension and right ventricular failure (Fig 2A). However, in the BTR group, ECLS was implanted mainly because of respiratory failures (Fig 2B). Figure 3 shows the devices used for primary ECLS. Individual data on ECLS devices, device changes, and patient outcomes in the BTT and BTR groups are reported in Appendix Tables 1 and 2.
Short-Term Outcomes Of the 31 patients, 15 (48.4%) survived to hospital discharge. In the BTT group, inhospital survival was 50% (4 of 8); 6 patients underwent transplantation a median of 6.5 days (IQR: 1 to 8) after PEA, and 4 of them survived to hospital discharge. The 2 patients who were listed but not receiving transplants died before organ allocation; causes of death are reported in Appendix Table 1. In the BTR group, inhospital survival was 47.8% (11 of 23) and median ECLS duration was 6.5 days (IQR: 2 to 18). Of the 12 patients who died in the BTR group, 2 underwent repeat VA-ECMO implantation 2 days after weaning off the first device but subsequently died, 1 of refractory sepsis and 1 of massive stroke (patients 18 and 24, respectively; Appendix Table 2). The other 10 patients died either during ECLS (n ¼ 1) of multiorgan failure, or after ECLS removal, of refractory bleeding (n ¼ 7), multiorgan failure (n ¼ 2), or recurrent PA
NYHA ¼ New York Heart Association; TPR ¼ total pulmonary artery resistance;
PEA ¼ pulmonary w/o ¼ without.
thrombosis (n ¼ 1). The most common adverse events during ECLS were bleeding and infections (Table 3).
Factors Affecting Inhospital Survival The BTT group was too small for statistical analysis. In the BTR group, survivors were younger, had larger early MPAP decreases, lower post-PEA total PA resistance values, and a higher rate of pure respiratory failure compared with nonsurvivors (all p < 0.05; Table 4). The odds ratios for inhospital death were 1.87 (95% confidence interval: 1.16 to 3.74) per additional 100 dyne $ s $ cm-5 of early postoperative total PA resistance (p ¼ 0.03) and 0.66 (95% confidence interval: 0.41 to 0.91) per additional 10 mm Hg of early postoperative MPAP decrease (p ¼ 0.04). None of the other preoperative or intraoperative variables differed significantly between survivors and nonsurvivors (Table 4).
Midterm Hemodynamic Factors After Weaning From ECLS The MPAP values at hospital discharge or 6 months after PEA were available for the 11 BTR patients who survived to hospital discharge. The early postoperative decrease in MPAP persisted in all patients and even tended to become greater over time.
Ann Thorac Surg 2016;-:-–-
5
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Table 2. Preoperative and Intraoperative Extracorporeal Life Support Patient Characteristics Characteristics
Patients With ECLS (n ¼ 31)
BTT Group (n ¼ 8)
BTR Group (n ¼ 23)
52 (39–61) 14 1.8 (1.7–2.0)
28 (21–52) 4 1.8 (1.7–1.9)
54 (50–68) 10 1.8 (1.7–2.0)
0 (0) 4 (12.9) 11 (35.5) 16 (51.6)
0 0 3 5
0 4 8 11
52 (45–61) 1,017 (855–1,431) 2.3 (1.8–2.6) 9 (6–14.5) 10/301 6 8 22 11 1 5 4 1 2
50 (45–61) 1,100 (883–1,150) 2.1 (1.9–2.6) 6 (5–16) 3 1 2 4 6
53 (46–60) 1,016 (849–1,438) 2.4 (1.8–2.6) 9 (6–14) 7 5 6
3 2 1 2
0 0
237 (198–283) 112 (91–130) 34 (29–45)
268 (214–304) 105 (83–124) 32 (20–46)
227 (198–268) 119 (95–132) 34 (30–46)
2 11 10 8
0 0 0 8
2 11 10 0
0.13 0.30 0.59 . . . . .
3 1
0 0
3 1
. .
Age, years Male Body surface area, m2 NYHA status (n ¼ 395) Class I Class II Class III Class IV Preoperative hemodynamics MPAP, mm Hg TPR, dynes $ s $ cm5 Cardiac index, L $ min1 $ m2 Central venous pressure, mm Hg PH-targeted therapy Preoperative dobutamine Patent foramen ovale History of pulmonary embolism Risk factors Splenectomy Central intravenous line Thrombophilia Prior PEA Anticipated need for HLT Operative times, minutes Cardiopulmonary bypass Aorta cross-clamp time Circulatory arrest time Jamieson classification Type 1 Type 2 Type 3 Type 4 Associated surgical procedures Persistent foramen ovale closure Coronary artery bypass graft a
5 1 2
p Value 0.04 0.99 0.61 0.69a
0.76 0.97 0.76 0.33 0.99 0.99 0.99 0.18 0.01 . 0.09 0.27 . .
Fisher’s exact test, number of class IV patients versus other patients.
Values are median (interquartile range), n, or n (%). BTR ¼ bridge to recovery; BTT ¼ bridge to transplant; ECLS ¼ extracorporeal life support; HLT ¼ heart-lung transplantation; MPAP ¼ mean pulmonary artery pressure; NYHA ¼ New York Heart Association; PEA ¼ pulmonary endarterectomy; PH ¼ pulmonary hypertension; TPR ¼ total pulmonary artery resistance.
Long-Term Outcome of Transplanted Patients Among the 8 patients who received transplants who were bridged using ECLS, 4 were discharged from the hospital and followed. Two patients, died at 3 and 45 months, of graft dysfunction and breast cancer, respectively. The remaining 2 were still alive at 38 and 6 months.
Comment In patients managed with ECLS as BTR, factors associated with inhospital survival were younger age, greater early improvement in hemodynamic variables, and respiratory failure as the reason for ECLS. In addition, we
demonstrated that ECLS as BTT strategy was successful in 4 of 8 patients with failed PEA. Pneumonia and airway bleeding were major causes of morbidity and mortality in patients requiring ECLS after PEA. Thistlesthwaite and colleagues [10] provided the historical series of 20 patients managed with VV-ECMO after PEA complicated with respiratory failure, with a survival of 30%. Berman and associates [11] reported 7 patients with severe cardiorespiratory failure after PEA who were managed using VA-ECMO as BTR. In these studies, pneumonia and airway bleeding were the leading causes of morbidity and mortality. Consistently, we found that pneumonia was a major cause of morbidity in
6
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Ann Thorac Surg 2016;-:-–-
Fig 2. Indication for extracorporeal life support (ECLS) implantation: (A) in operating room (OR); and (B) in intensive care unit (ICU). Dark gray areas indicate hemodynamic failure; lighter gray areas indicate respiratory failure; patterned areas indicate respiratory and hemodynamic failure. (BTR ¼ bridge to recovery; BTT ¼ bridge to transplantation; PEA ¼ pulmonary endarterectomy.)
patients with post-PEA ECLS. Bleeding was also a leading cause of morbidity but, differing from the study by Thistlesthwaite and colleagues [10], bleeding in the airway was not the most common site of bleeding. That is potentially due to the pulmonary circulation discharge with our preference for VA-ECMO decreasing the risk for airway hemorrhage; also, higher levels of anticoagulation agents required for VA-ECMO compared with VV-ECMO increased the risk for overall hemorrhage. Of importance, the early hemodynamic improvement in patients with successful BTR persisted in the midterm, confirming the benefit of PEA in patients with
Fig 3. Type of devices depending on strategies and implantation sites as well as inhospital survival. (BTR ¼ bridge to recovery; BTT ¼ bridge to transplantation; ECMO ¼ extracorporeal membrane oxygenation; ICU ¼ intensive care unit; OR ¼ operating room; PA-LA ¼ pulmonary artery to left atrium; PEA ¼ pulmonary endarterectomy; VA ¼ venoarterial; VV ¼ venovenous.)
severe chronic thromboembolic pulmonary hypertension [14]. This observation is consistent with microvascular disease reversal within a few weeks after PEA as previously suggested in human and animal models [15–19]. Furthermore, good outcomes obtained in patients with pure respiratory failure were obtained chiefly using VA-ECMO instead of VV-ECMO. Our results suggest that VA-ECMO may also benefit patients with pure respiratory failure. As VA-ECMO decreases the transpulmonary blood flow, it may help to control both airway bleeding from the pulmonary circulation and pulmonary edema secondary to reperfusion, which are
Ann Thorac Surg 2016;-:-–-
7
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Table 3. Complications in the Bridge to Transplantation and Bridge to Recovery Groups Complications Bleeding Thoracic Airways Cannulation sites Gastrointestinal Infections Lung Cannulation sites Other Renal replacement therapy Limb ischemia Stroke Heparin-induced thrombocytopenia Hemolysis a
All Patients
BTT Group (n ¼ 8)
BTR Group (n ¼ 23)
p Value
17 (54.8) 8 5 4 1 16 (51.6) 13 2 2 8 (25.8) 4 (12.9) 4 (12.9) 3 (9.7) 6 (19.3)
2 (25.0) 2 0 0 0 2 (25.0) 2 0 0 1 (12.5) 1 (12.5) 2 (25.0) 1 (12.5) 1 (15.5)
15 (65.2) 6 5 4 1 14 (60.9) 11 2 2 7 (30.4) 3 (13.0) 2 (8.7) 2 (8.9) 5 (21.7)
0.10a . . . . 0.11a . . . . . . . .
Bridge to recovery (BTR) versus bridge to transplantation (BTT).
Values are n (%) or n.
the leading causes of morbidity and mortality among patients requiring ECLS after PEA [10, 11]. Moreover, VA-ECMO is more likely to protect the right ventricle from the increased PA pressure observed in severe acute respiratory distress syndrome [20]. However, the benefits of VA-ECMO should be balanced with the heightened risk of bleeding, stroke, and limb ischemia related to VA cannulations and the higher level of anticoagulation therapy. Heparin-coated circuits, systematic limb reperfusion, and ischemia screening should help to decrease these risks. We used PA-LA Novalung in the BTT strategy. Potential advantages of this strategy are the absence of peripheral vessel cannulation and facilitated rehabilitation of conscious patients owing to central cannulation. The PA-LA Novalung was used as a BTT strategy for patients with end-stage pulmonary hypertension and refractory hemodynamic failure [21, 22] but has not been yet compared with VA-ECMO. For patients with poor early hemodynamic response to PEA, factors responsible for hemodynamic failure were not reversible despite ECLS, explaining poor outcomes. For patients potentially eligible for salvage transplantation, a transplantation workup before PEA seemed beneficial by facilitating the listing and organ allocation process. Heart-lung transplantation was our primary choice because we considered right ventricular failure to be irreversible—heart-lung transplantation decreases the chance for organ allocation compared with double-lung transplantation in the French allocation organ system. That is one of the reasons leading our team to move through heart-lung transplantation. Indeed, our experience with VA-ECLS before heart-lung transplantation for patients with severe pulmonary hypertension increased during the study (unpublished data) and allowed us to think that heart-lung transplantation is feasible in patients with severe pulmonary hypertension and right
ventricle failure. However, posttransplantation ECLS is frequent, potentially increasing morbidity. Our results help to target unmet needs for managing life-threatening complications after PEA. For patients with a suboptimal response to PEA who cannot be weaned from short-term ECLS and are not eligible for transplantation, the development of destination therapies based on adapted ECLS devices might provide new therapeutic options in the future. Also, anticipating the need for potential post-PEA transplantation should help to shorten the time to organ allocation. Finally, the place of transluminal PA angioplasty has to be defined in patients with hemodynamic failure and persistent pulmonary hypertension after PEA. We acknowledge the limitations of our study, which was retrospective and included a small number of patients. Respiratory failure and hemodynamic failure were defined retrospectively. The midterm hemodynamic outcome was limited because many patients were from foreign countries. In addition, comparison between VV-ECMO and VA-ECMO was not feasible owing to the small number of patients treated with VV-ECMO. Therefore, our positive results obtained using VA-ECMO in patients with respiratory failure should be confirmed by further studies. In conclusion, ECLS has an important role in the management of patients with severe chronic thromboembolic pulmonary hypertension, allowing them to benefit from lasting hemodynamic improvement. Extracorporeal life support is efficient as a bridge to salvage transplantation after failed PEA. A pre-PEA transplantation workup may improve the chances of survival through salvage transplantation in young patients with distal disease deemed operable. Patients with suboptimal PEA who have hemodynamic failure and are not eligible for transplantation have poor prognoses with short-term ECLS. Further development of destination therapies and
8
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
Ann Thorac Surg 2016;-:-–-
Table 4. Comparison of Survivors and Nonsurvivors in Bridge to Recovery Group Variables Age, years Body surface area, m2 Male MPAP, mm Hg Preoperative Percent decrease TPR, dyne $ s $ cm5 Preoperative Percent decrease Cardiac index, L $ min1 $ m2 Preoperative Percent increase Preoperative CVP, mm Hg Operative times, minutes Cardiopulmonary bypass Patent foramen closure Thromboembolic disease Type 2 Preoperative characteristics NYHA class I/II/III/IV Pulmonary embolism history Preoperative dobutamine Postoperative course Pure respiratory failure Mixed cardiorespiratory failure Pure hemodynamic failure
BTR group (n ¼ 23)
Survivors (n ¼ 11)
Nonsurvivors (n ¼ 12)
p Value
54 (49.5–67.5) 1.79 (1.67–1.98) 13
52 (46.0–53.5) 1.83 (1.71–1.96) 6
67.5 (56–70.3) 1.775 (1.67–2.03) 7
0.012 0.64 0.99
53.0 (45.5–59.8) 27.2 (3.1–50)
58.5 (47.5–64) 46.8 (44.4–57.1)
50.5 (46.25–54) 4.2 (11.9–27.2)
0.15 0.02
1,016 (849–1,438) 25.7 (5.4–42.3)
996 (802–1,338) 35.8 (27.1–46.2)
1,124 (890–1,447) 11.2 (4.0 to 31.7)
0.54 0.11
2.36 (1.82;2.59) 5.4 (21.1 to 5.9) 9.0 (6–14)
2.54 (2.16–2.68) 7.1 (17.3 to 0.7) 9 (8–12)
2.23 (1.59–2.49) 3.4 (28.8 to 11.4) 11 (6–18.5)
0.08 0.82 0.54
227 (198–268) 3
217 (197–237) 3
263 (200–287) 0
0.15 0.10
11
6
5
0.68
0/4/8/11 18 18
0/4/3/4 8 10
0/0/5/7 10 8
0.097 0.59 0.32
13 7 3
10 0 1
3 7 2
0.003 0.005 0.99
Values are median (interquartile range) or n. BTR ¼ bridge to recovery; CVP ¼ central venous pressure; Association; TPR ¼ total pulmonary artery resistance.
MPAP ¼ mean pulmonary artery pressure;
strategies decreasing anticoagulation therapy and lung infections should improve outcomes.
6.
The authors wish to thank Tai Pham, MD, for his contribution to the statistical analysis review. 7.
References 1. Jamieson SW, Kapelanski DP, Sakakibara N, et al. Pulmonary endarterectomy: experience and lessons learned in 1, 500 cases. Ann Thorac Surg 2003;76:1457–62. 2. Freed DH, Thomson BM, Berman M, et al. Survival after pulmonary thromboendarterectomy: effect of residual pulmonary hypertension. J Thorac Cardiovasc Surg 2011;141: 383–7. 3. Madani MM, Auger WR, Pretorius V, et al. Pulmonary endarterectomy: recent changes in a single institution’s experience of more than 2,700 patients. Ann Thorac Surg 2012;94:97–103. 4. Riedel M, Stanek V, Widimsky J, Prerovsky I. Longterm follow-up of patients with pulmonary thromboembolism. Late prognosis and evolution of hemodynamic and respiratory data. Chest 1982;81:151–8. 5. Moser KM, Spragg RG, Utley J, Daily PO. Chronic thrombotic obstruction of major pulmonary arteries: results of
8. 9. 10.
11. 12. 13.
NYHA ¼ New York Heart
thromboendarterectomy in 15 patients. Ann Intern Med 1983;99:299–305. Mayer E, Jenkins D, Lindner J, et al. Surgical management and outcome of patients with chronic thromboembolic pulmonary hypertension: results from an international prospective registry. J Thorac Cardiovasc Surg 2011;141: 702–10. Bonderman D, Skoro-Sajer N, Jakowitsch J, et al. Predictors of outcome in chronic thromboembolic pulmonary hypertension. Circulation 2007;115:2153–8. Condliffe R, Kiely DG, Gibbs JS, et al. Prognostic and aetiological factors in chronic thromboembolic pulmonary hypertension. Eur Respir J 2009;33:332–8. Dartevelle P, Fadel E, Mussot S, et al. Chronic thromboembolic pulmonary hypertension. Eur Respir J 2004;23:637–48. Thistlethwaite PA, Madani MM, Kemp AD, Hartley M, Auger WR, Jamieson SW. Venovenous extracorporeal life support after pulmonary endarterectomy: indications, techniques, and outcomes. Ann Thorac Surg 2006;82:2139–45. Berman M, Tsui S, Vuylsteke A, et al. Successful extracorporeal membrane oxygenation support after pulmonary thromboendarterectomy. Ann Thorac Surg 2008;86:1261–7. Chemla D, Castelain V, Humbert M, et al. New formula for predicting mean pulmonary artery pressure using systolic pulmonary artery pressure. Chest 2004;126:1313–7. Thistlethwaite PA, Mo M, Madani MM, et al. Operative classification of thromboembolic disease determines outcome after pulmonary endarterectomy. J Thorac Cardiovasc Surg 2002;124:1203–11.
Ann Thorac Surg 2016;-:-–-
14. Thistlethwaite PA, Kemp A, Du L, Madani MM, Jamieson SW. Outcomes of pulmonary endarterectomy for treatment of extreme thromboembolic pulmonary hypertension. J Thorac Cardiovasc Surg 2006;131:307–13. 15. Moser KM, Bloor CM. Pulmonary vascular lesions occurring in patients with chronic major vessel thromboembolic pulmonary hypertension. Chest 1993;103:685–92. 16. Moser KM, Metersky ML, Auger WR, Fedullo PF. Resolution of vascular steal after pulmonary thromboendarterectomy. Chest 1993;104:1441–4. 17. Fadel E, Michel RP, Eddahibi S, et al. Regression of postobstructive vasculopathy after revascularization of chronically obstructed pulmonary artery. J Thorac Cardiovasc Surg 2004;127:1009–17. 18. Fadel E, Wijtenburg E, Michel R, et al. Regression of the systemic vasculature to the lung after removal of pulmonary
BOULATE ET AL ECLS AFTER PULMONARY ENDARTERECTOMY
19.
20. 21.
22.
9
artery obstruction. Am J Respir Crit Care Med 2006;173: 345–9. Boulate D, Perros F, Dorfmuller P, et al. Pulmonary microvascular lesions regress in reperfused chronic thromboembolic pulmonary hypertension. J Heart Lung Transplant 2015;34:457–67. Zapol WM, Snider MT. Pulmonary hypertension in severe acute respiratory failure. N Engl J Med 1977;296: 476–80. Strueber M, Hoeper MM, Fischer S, et al. Bridge to thoracic organ transplantation in patients with pulmonary arterial hypertension using a pumpless lung assist device. Am J Transplant 2009;9:853–7. Schmid C, Philipp A, Hilker M, et al. Bridge to lung transplantation through a pulmonary artery to left atrial oxygenator circuit. Ann Thorac Surg 2008;85:1202–5.