Severe pulmonary hypertension (PH) and in particular pulmonary artery hypertension (PAH) are characterized by poor prognosis and 3-year mortality exceeding 30% despite modern therapy.1-4 Thus, lung transplantation remains important for patients with refractory disease.5 Combined heart-lung transplantation (HLTx) or bilateral lung transplantation (BLTx) on cardiopulmonary bypass (CPB) have been used.6-9 However, early survival after transplantation for PAH is impaired as compared to other indications, such as chronic obstructive pulmonary disease, idiopathic pulmonary fibrosis, or cystic fibrosis.6 Primary graft dysfunction (PGD) is frequently cited as causing early morbidity and mortality in PAH patients.10,11 Some have speculated that PGD in PAH patients might be attributed to endothelial injury from shear-stress forces applied by a well-trained right ventricle (RV) resulting in subsequent pulmonary edema.12,13
Contrarily, we hypothesize that long-standing underfilling of the left ventricle (LV) in the presence of reduced cardiac output secondary to elevated pulmonary vascular resistance may result in “deconditioning” of the LV, rendering the LV incapable of handling a normal preload in the early postoperative period. Herein, the postoperative use of venoarterial extracorporeal membrane oxygenation (ECMO) could be a useful to provide time for gradual LV adaptation to altered hemodynamics.
Previously, ECMO was used intraoperatively instead of CPB14-16 and postoperatively in patients with severe PGD.17-19 We have also successfully used ECMO in awake, nonintubated patients as a bridge to BLTx, leading to improved outcomes compared with mechanical ventilation.20,21 In the present study, we attempted early extubation in patients with severe PH undergoing BLTx facilitated by the postoperative use of venoarterial ECMO (BLTx-ECMO group). Extracorporeal membrane oxygenation weaning was guided by echocardiography and invasive hemodynamics. The outcomes were compared with two historic control groups of patients with severe PH treated by combined HLTx group or BLTx without postoperative ECMO support (BLTx-ventilation group).
Demographic Data and Treatment Groups
Between June 2005 and April 2013, 53 patients underwent transplantation for severe PH at a single center. Patients with idiopathic PAH, pulmonary venoocclusive disease, chronic thromboembolic PH, or sarcoidosis with severe PH were included. Before March 2010, combined HLTx group (n = 17) or BLTx without postoperative ECMO (BLTx ventilation group, n = 13) was performed, depending on rather subjective parameters of estimated right ventricular (RV) dysfunction. The BLTx-ECMO group comprises 23 consecutive patients transplanted from March 2010 onward (Table 1). Four BLTx-ECMO patients presented with right ventricular failure before transplantation, and were bridged by implantation of venoarterial (v/a ECMO) for 5, 11, 12, and 20 days, respectively, while remaining nonintubated (“awake ECMO” strategy).20,21 Five BLTx ventilation patients were bridged to transplantation, three on awake ECMO for 16, 24, and 34 days, and two with ECMO after intubation for 1 and 8 days.
Early Extubation and ECMO Weaning in the BLTx-ECMO Group
Postoperatively, median ECMO support was 8 days (range, 5-19 days; Figure 1B). Median duration of postoperative mechanical ventilation was shorter in the BLTx-ECMO and HLTx groups compared to the BLTx-ventilation group at 3, 9 and 32 days, respectively (range, 0.6-42, 0.5-24, and 0.4-94 days; P < 0.0001). These results are reflected in the median intensive care unit (ICU) stays of 14, 29, and 38 days (range, 5-43, 4-73, and 2-94 days; P = 0.0005) in the BLTx-ECMO, HLTx, and BLTx ventilation groups, respectively. During the early ICU stay, all patients in the BLTx-ECMO group required intravenous β blocker and antihypertensive medication to control afterload. Contrarily, in the BLTx ventilation and HLTx groups, inotropic and vasopressor support was necessary in all patients for a minimum of 3 postoperative days. None of the 23 patients in BLTx-ECMO group required secondary implantation of an ECMO after weaning. In contrast, three patients in the BLTx ventilation and two patients in the HLTx group required implantation of v/a ECMO after initial weaning from the respirator.
Hemodynamic Outcome After Transplantation in the BLTx-ECMO Group
Mean pulmonary artery pressure decreased from 67 ± 17 mm Hg preoperatively to 18 ± 7 mm Hg on the second postoperative day (P < 0.0001) and did not change significantly thereafter at postoperative day 5 (Figure 2A). Mean right ventricular ejection fraction (RVEF) before transplantation was severely impaired (34 ± 10%), but increased after surgery to 42% ± 10% and 49% ± 7% at days 2 and 5, respectively (Figure 2B). Left ventricular ejection fraction (LVEF) was normal before lung transplantation (65% ± 13%). At the second postoperative day, LVEF decreased to 47% ± 17% (P = 0.02). On day 5, LVEF recovered to 60% ± 10% (P = 0.04; Figure 2B), and cardiac output increased from preoperative 4 ± 1.6 to 5.6 ± 0.7 L per min (Figure 2C). The mean preoperative left ventricular end-diastolic diameter (LVEDD) was 37 ± 10 mm, representing rather low values. Under minimized ECMO support, mean LEVDD increased to 46 ± 0 mm on day 2, stabilized at 42 ± 6 mm (P = 0.0008) on day 5 and did not change significantly thereafter. The same tendency was observed with left atrial (LA) pressures: 12 ± 5 to 6 ± 3 mm Hg on days two and five, respectively (Figure 3), suggesting transient, reversible LV dysfunction.
PGD and Risk Factors
We assessed those six independent clinical donor and recipient risk factors for grade 3 PGD that Diamond and colleagues10 recently defined in a multicenter effort. Pulmonary artery hypertension and sarcoidosis as underlying disease and the use of CPB (or v/a ECMO) are not variables in our study. Single lung transplantations were not performed in our collective. Recipient overweight (body mass index BLTx-ECMO group, 22 ± 4.2; BLTx-ventilation group, 22 ± 3.9; HLTx group, 24 ± 4.4) and donor smoking (smokers/all donors BLTx-ECMO group, 6/23; BLTx-ventilation group, 3/13; HLTx group, 5/17) were not significantly different among groups.
At arrival in the ICU, the PGD score was lower in the BLTx-ECMO group compared to historic BLTx-ventilation and HLTx controls and remained so at 24 hr. At 72 hr, the PGD scores in the BLTx-ECMO and HLTx groups were significantly lower than those in the BLTx-ventilation group (Figure 4A).
Survival and Midterm Follow-Up
The median inhospital stay after transplantation was 35, 65, and 84 days (range, 21–96, 26–227, and 28–148 days; P = 0.0023) in the BLTx-ECMO, HLTx, and BLTx-ventilation groups, respectively. The 90-day survival was 100%, 82%, 85%, respectively (Figure 4B; P = 0.12). In the BLTx-ECMO group, in those patients that were transplanted more than 1 year ago (n = 16 patients at risk), survival and BOS-free survival at 12 months both were 94%. The remaining seven patients are currently alive and nonhospitalized. During the first year, three patients each died in the HLTx (massive haemoptysis, sepsis, and fatal stroke after emergency ECMO) and in the BLTx-ventilation groups (two while on mechanical ventilation (38 and 75 days) for invasive aspergillosis and one from bacterial meningitis on day 145 after mechanical ventilation for 82 days.
Complications of Protocol ECMO Therapy
Major complications potentially related to ECMO included a patient with aortic dissection type B and cholestasis 6 days after BLTx. After endovascular repair with stents, he required prolonged respiratory weaning and was discharged on day 70. He died on the 229th day of gastrointestinal bleeding. Type B dissection, presumably having led to liver failure, was considered a complication of arterial ECMO cannulation. A second patient required several redo thoracotomies for bleeding, resulting in long-term mechanical ventilation. The third patient, after 21 days on ECMO, developed lower limb ischemia after removal of the cannulas requiring embolectomy and fasciotomy.
Our results suggest that one important problem causing early PGD after BLTx in patients with end-stage PAH is, rather than elevated pulmonary pressures with increased shear stress or RV failure, LV dysfunction with predominantly diastolic failure, and elevated filling pressures. Hemodynamic and echocardiography findings suggest that the LV of patients with long-standing PAH is small, stiff, and dysfunctional and as such, probably not capable of handling the normal or elevated cardiac output immediately after transplantation. Recovery of the LV may take a few days, and bridging this period with awake v/a ECMO seems to be a successful strategy leading to a postoperative 90-day survival of 100% in our series. These data are important because lung transplantation for PAH has been associated with a 3.5-fold increased risk to develop severe PGD10,11 and a subsequent high post-operative mortality rate of 24% at 3 months,22 suggesting specific mechanisms predisposing for PGD in recipients with severe PAH. According to a recent study by Diamond and colleagues,10 there are six independent clinical risk factors for grade 3 PGD that influence significantly 90-day and 1-year mortality after transplantation: donor smoking, single lung transplantation, use of CPB, recipient overweight, PAH, and sarcoidosis. According to our explanation in Results, donor smoking and recipient overweight are the only two clinical risk factors predefined by Diamond and colleagues’ exhaustive study that could independently influence our study. Neither donor smoking history nor recipient body mass index show, however, significant differences between the three patient groups, effectively ruling out important known confounders for PGD development. Although we have to acknowledge the possibility that further factors we have not yet well defined also play roles, we therefore assume LV dysfunction being a specific mechanism playing an important role in the development of severe PGD in patients with PAH. Until 2010, in our program, most patients with PAH were listed for combined HLTx to circumnavigate the problems associated with early PGD.23 However, combined HLTx itself is associated with important shortcomings. Two organs have to be allocated toward one recipient, and access to suitable donor hearts has become increasingly difficult. Complexity of the operation, an additional risk of early dysfunction of the heart, plays roles in increased perioperative morbidity and mortality after combined HLTx as compared to lung transplantation only, with overall 1-year survivals of 68% versus 81%, respectively.22
The Vienna group reported the intraoperative use of v/a ECMO with extension of support for a median of 8 hr after the operation in a series of 17 BLTx recipients with various forms of PH, resulting in a 6-month survival of 88%.15 Their general postoperative strategy was prolonged mechanical ventilation after early ECMO weaning. Interestingly, they report four patients had “temporary left ventricular failure” at the time of weaning from the respirator, causing one early death. Given that both deaths in the study and the temporary left ventricular failures occurred in 11 patients with underlying idiopathic PAH, we postulate ECMO support may have been weaned too early in their study, whereas the response to complex medical therapy alone for LV dysfunction after transplantation by the same and other groups24,25 may be highly variable and of questionable reproducibility.
Respirator weaning is commonly associated with stress, high blood pressures, and elevated cardiac output, potentially resulting in LV failure and cardiogenic pulmonary edema in PH patients, hitherto leading to prolonged sedation and mechanical ventilation. This phenomenon has been described in the past,26 but was considered a reperfusion injury, attributed to PH and mechanical shear stress, caused by the hypertrophied RV. Our data show that the underlying mechanism for this syndrome may rather not be RV, but LV dysfunction. The importance of LV dysfunction is also appreciated on comparison of the HLTx and BLTx-ventilation groups, where combined HLTx results in lower PGD grades and shorter ventilation times. We thus tried to overcome both, LV dysfunction and prolonged mechanical ventilation, by ECMO support and early extubation. Awake ECMO is currently the state-of-the-art approach to avoid mechanical ventilation,21,27 and our group has shown recently that it is superior to mechanical ventilation in patients requiring bridging to lung transplantation. We show that patients with severe PH receiving BLTx can be extubated after a median of 3 days while remaining supported by ECMO (median, 8 days). In analogy to pretransplant awake ECMO bridging, the postoperative time span can also be successfully bridged with the awake ECMO concept, presumably until LV remodeling and adaptation to increased preload occurs.
Several lines of evidence support our hypothesis. The Amsterdam group showed that in patients with chronic thromboembolic PH that reduction of LV free wall mass, muscular atrophy and myocyte shrinkage is reversible after pulmonary endarterectomy.28 A comparable remodeling process has also been observed in patients with severe mitral valve stenosis after successful mitral valvuloplasty.29 In patients undergoing lung transplantation for pulmonary emphysema, LVEDD, LV mass, and LVEF may increase by 25% within 3 months after lung transplantation.30 In the herein described study, the timing of ECMO weaning was determined by the results of echocardiography and invasive hemodynamics, in particular, stable LA pressure below 10 mm Hg on ECMO flow reduction. In the first 5 postoperative days, ECMO flows have not been reduced to allow for safe extubation while using a restrictive fluid management to maintain LA pressure less than 10 mm Hg. Later, daily reductions in ECMO flow proved useful to decide whether the LV has yet adapted (failure was indicated by increasing LA pressures and increasing LVEDD) or was capable to handle a normal cardiac output (when ECMO flow reduction no longer resulted in LA and LVEDD increases).
There are several limitations to our study, most importantly the relatively small sample size and the lack of a parallel control group. Because patients with PAH do not undergo lung transplantation as frequent as in past years, and collection of a solid sample size thus may take a long while, we decided to compare our results with outcomes of historic BLTx and HLTx patients to provide a possibility of comparison. Given the historic survival data, the 100% postoperative survival rate in our cohort is compelling and warrants further use of the proposed management strategy. One additional limitation might be the unknown effect of preoperative ECMO (4/23 patients) on LV dysfunction, potentially adding on the effect of postoperative ECMO. In addition to that, preoperative ECMO has been associated with impaired outcomes after transplantation, a possibility that we deem unlikely considering our favorable experience with awake ECMO bridging in the past.20 A shortcoming of our study is the lack of attempts to adjust the interval of initial postoperative ECMO support to severity of LV dysfunction or other preoperative parameters. Thus, potentially, a few patients with still sufficient cardiac function, where subjected to ECMO support longer than necessary, putting them at the risks of ECMO-associated complications. Future work on this topic is warranted.
In conclusion, we consider transient dysfunction of the LV after BLTx for severe PH to be one of the mechanisms causing respiratory failure during weaning from mechanical ventilation. To overcome this problem, we propose awake v/a ECMO support until recovery of LV function.
MATERIALS AND METHODS
All patients in the BLTx-ECMO group and six patients in the BLTx-ventilation group received sequential BLTx on v/a ECMO (see below). Seven BLTx-ventilation and all HLTx patients were transplanted on standard CPB.
Hemodynamic Monitoring in the BLTx-ECMO Group
All patients received a Swan-Ganz catheter (Edwards Lifesciences, Irvine, CA) with continuous cardiac output monitoring. An LA catheter (Maquet Cardiopulmonary AG, Hirrlingen, Germany) was implanted in the last 17 patients for LA pressure monitoring targeting at less than 10 mm Hg during ECMO weaning.
BLTx-ECMO Group Management
On surgical preparation of the lung hilum, 60 IU/kg heparin were administered. Venoarterial ECMO was established percutaneously with a FemTrak femoral venous cannula (Edwards Lifesciences, Irvine, CA) and a NovaPort arterial femoral cannula (NovaLung GmbH, Hechingen, Germany). In one 7-year-old patient, ECMO was implanted by open groin surgery. An additional 6F introducer sheath (4F in the 7-year-old; Cordis Corporation, Miami, FL) was inserted into the femoral artery and connected to the arterial branch of the ECMO for antegrade leg perfusion. The ECMO system consisted of a Maquet Rotaflow centrifugal pump and PLS Set with the Quadrox polymethylpentene membrane oxygenator (Maquet Cardiopulmonary AG, Hirrlingen, Germany). After initiation of ECMO, the flow was adjusted to approximately 80% of predicted cardiac output, usually around 4 L per min. In BLTx-ventilation group patients transplanted using ECMO, ECMO was explanted intraoperatively. In the BLTx-ECMO group, per protocol (Figure 1A) flow remained unchanged for 5 postoperative days. Arterial oxygenation was monitored from right radial arterial catheters because femoral artery catheters might be misleading for admixture of well-oxygenated blood from the ECMO. Beyond 24 hr after surgery, the activated clotting time was maintained between 160 and 180 sec. Early extubation after transplantation was aimed for under ECMO support. Beyond day 5, ECMO weaning was initiated while controlling for LV function during intermittent reduction of ECMO blood flow to 0.6 L/m2 body surface area. When there were neither echocardiographic signs of LV dysfunction nor LA pressure increases over 10 mm Hg during reduction of blood flow, the ECMO flow was reduced in 0.5 L per min steps until 0.6 L/m2 body surface area was reached and then the ECMO was explanted.
Transthoracic or transesophageal echocardiography were performed in all patients before transplantation, on postoperative days 2 and 5, immediately before and after ECMO explantation, and at discharge, in 2-dimensional, m-mode, pw-mode, cw-mode, and color Doppler technique. Measurements of the left ventricular end-diastolic and end-systolic diameters, of the right ventricular end-diastolic and end-systolic diameters, of the LV and RV muscle thickness as well as calculations of the LV and RV ejection fractions and fractional shortening were performed. During ECMO support all parameters were retrieved under both normal and reduced ECMO blood flows. Baseline and discharge echocardiographies were completed by examination of mitral, aortic, and tricuspid valves including their flow parameters and estimation of valvular pressure gradients. Standard parasternal and apical projections were applied using an Envisor C, Philips, system with a 5-MHz transthoracic transducer (21422A; Agilent Technologies, Böblingen, Germany) or a multiplane transesophageal transducer (T6210; Agilent Technologies). The baseline and discharge examination was performed with an iE33 XMatrix, Philips, system and X5-1, Philips, transthoracic transducer.
All patients received triple immunosuppression with cyclosporine, mycophenolate mofetil, and methylprednisolone. No antibody induction therapy was used in this cohort.
Data are expressed as mean ± standard deviation. Analyses were performed with SPSS statistical package (SPSS21.0; IBM Corporation, Armonk, NY) or GraphPad Prism 6 (GraphPad Software, San Diego, CA). The independent sample t test was used to analyze the changes from baseline data. P less than 0.05 was considered statistically significant. Differences regarding PGD levels (levels 2 and 3 vs. lower levels) between treatment groups were analyzed with two-sided chi-square tests (or Fisher’s exact test, if appropriate); the significance level was set to 5%. Because the tests were explorative, there was no adjustment for multiple testing. Kaplan-Meier-survival curves were calculated.
1. Humbert M, Sitbon O, Chaouat A, et al. Survival in patients with idiopathic, familial, and anorexigen-associated pulmonary arterial hypertension in the modern management era. Circulation
2010; 122: 156.
2. Hoeper MM, Schwarze M, Ehlerding S, et al. Long-term treatment of primary pulmonary hypertension with aerosolized iloprost, a prostacyclin analogue. N Engl J Med
2000; 342: 1866.
3. McLaughlin VV, Shillington A, Rich S. Survival in primary pulmonary hypertension: the impact of epoprostenol therapy. Circulation
2002; 106: 1477.
4. Kemp K, Savale L, O’Callaghan DS, et al. Usefulness of first-line combination therapy with epoprostenol and bosentan in pulmonary arterial hypertension: an observational study. J Heart Lung Transplant
2012; 31: 150.
5. Galie N, Hoeper MM, Humbert M, et al. Guidelines for the diagnosis and treatment of pulmonary hypertension: the Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS), endorsed by the International Society of Heart and Lung Transplantation (ISHLT). Eur Heart J
2009; 30: 2493.
6. Taylor DO, Edwards LB, Aurora P, et al. Registry of the International Society for Heart and Lung Transplantation: twenty-fifth official adult heart transplant report—2008. J Heart Lung Transplant
2008; 27: 943.
7. Whyte RI, Robbins RC, Altinger J, et al. Heart-lung transplantation for primary pulmonary hypertension. Ann Thorac Surg
1999; 67: 937.
8. Topilsky M, Heller I. Prostacyclin in primary pulmonary hypertension: a bridge to heart-lung transplantation. Isr J Med Sci
1993; 29: 656.
9. Franke U, Wiebe K, Harringer W, et al. Ten years experience with lung and heart-lung transplantation in primary and secondary pulmonary hypertension. Eur J Cardiothorac Surg
2000; 18: 447.
10. Diamond JM, Lee JC, Kawut SM, et al. Clinical risk factors for primary graft dysfunction after lung transplantation. Am J Respir Crit Care Med
2013; 187: 527.
11. Kuntz CL, Hadjiliadis D, Ahya VN, et al. Risk factors for early primary graft dysfunction after lung transplantation: a registry study. Clin Transplant
2009; 23: 819.
12. Barr ML, Kawut SM, Whelan TP, et al. Report of the ISHLT Working Group on Primary Lung Graft Dysfunction part IV: recipient-related risk factors and markers. J Heart Lung Transplant
2005; 24: 1468.
13. Halldorsson AO, Kronon M, Allen BS, et al. Controlled reperfusion prevents pulmonary injury after 24 hours of lung preservation. Ann Thorac Surg
1998; 66: 877.
14. Bittner HB, Binner C, Lehmann S, Kuntze T, Rastan A, Mohr FW. Replacing cardiopulmonary bypass with extracorporeal membrane oxygenation in lung transplantation operations. Eur J Cardiothorac Surg
2007; 31: 462.
15. Pereszlenyi A, Lang G, Steltzer H, et al. Bilateral lung transplantation with intra- and postoperatively prolonged ECMO support in patients with pulmonary hypertension. Eur J Cardiothorac Surg
2002; 21: 858.
16. Ius F, Kuehn C, Tudorache I, et al. Lung transplantation on cardiopulmonary support: venoarterial extracorporeal membrane oxygenation outperformed cardiopulmonary bypass. J Thorac Cardiovasc Surg
2012; 144: 1510.
17. Glassman LR, Keenan RJ, Fabrizio MC, et al. Extracorporeal membrane oxygenation as an adjunct treatment for primary graft failure in adult lung transplant recipients. J Thorac Cardiovasc Surg
1995; 110: 723.
18. Meyers BF, Sundt TM 3rd, Henry S, et al. Selective use of extracorporeal membrane oxygenation is warranted after lung transplantation. J Thorac Cardiovasc Surg
2000; 120: 20.
19. Nguyen DQ, Kulick DM, Bolman RM 3rd, Dunitz JM, Hertz MI, Park SJ. Temporary ECMO support following lung and heart-lung transplantation. J Heart Lung Transplant
2000; 19: 313.
20. Fuehner T, Kuehn C, Hadem J, et al. Extracorporeal membrane oxygenation in awake patients as bridge to lung transplantation. Am J Respir Crit Care Med
2012; 185: 763.
21. Olsson KM, Simon A, Strueber M, et al. Extracorporeal membrane oxygenation in nonintubated patients as bridge to lung transplantation. Am J Transplant
2010; 10: 2173.
22. Christie JD, Edwards LB, Aurora P, et al. The Registry of the International Society for Heart and Lung Transplantation: Twenty-sixth Official Adult Lung and Heart-Lung Transplantation Report—2009. J Heart Lung Transplant
2009; 28: 1031.
23. Franke UF, Wahlers T, Wittwer T, et al. Heart-lung transplantation is the method of choice in the treatment of patients with end-stage pulmonary hypertension. Transplant Proc
2002; 34: 2181.
24. Birsan T, Kranz A, Mares P, et al. Transient left ventricular failure following bilateral lung transplantation for pulmonary hypertension. J Heart Lung Transplant
1999; 18: 304.
25. Verbelen T, Van Cromphaut S, Rega F, Meyns B. Acute left ventricular failure after bilateral lung transplantation for idiopathic pulmonary arterial hypertension. J Thorac Cardiovasc Surg
2013; 145: e7.
26. Okada Y, Hoshikawa Y, Ejima Y, et al. Beta-blocker prevented repeated pulmonary hypertension episodes after bilateral lung transplantation in a patient with primary pulmonary hypertension. J Thorac Cardiovasc Surg
2004; 128: 793.
27. Dalton HJ. Extracorporeal life support: moving at the speed of light. Respiratory care
2011; 56: 1445.
28. Hardziyenka M, Campian ME, Reesink HJ, et al. Right ventricular failure following chronic pressure overload is associated with reduction in left ventricular mass evidence for atrophic remodeling. J Am Coll Cardiol
2011; 57: 921.
29. Tischler MD, Sutton MS, Bittl JA, Parker JD. Effects of percutaneous mitral valvuloplasty on left ventricular mass and volume. Am J Cardiol
1991; 68: 940.
30. Rensing BJ, McDougall JC, Breen JF, Vigneswaran WT, McGregor CG, Rumberger JA. Right and left ventricular remodeling after orthotopic single lung transplantation for end-stage emphysema. J Heart Lung Transplant
1997; 16: 926.