Coronary artery disease (CAD) is a common comorbidity in patients with end-stage lung disease and a relative contraindication to lung transplantation.1,2 This is due to the substantial contribution of cardiovascular events to mortality in lung transplant recipients, potentially reflecting an accelerated atherosclerotic process previously documented in transplant recipients.3,4 This acceleration is at least in part driven by endothelial dysfunction and metabolic derangements caused by immune suppressive drugs, most notably the calcineurin inhibitors and corticosteroids.5,6
Despite these concerns, patients undergoing lung transplant with mild-to-moderate CAD have been reported to have similar outcomes to those without.7 For patients with more severe CAD documented prior to transplant, the revascularization options—percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG)—have respective advantages and disadvantages. PCI can be done pretransplant in a minimally invasive fashion and results in less operative complexity, but is felt to be less beneficial for diffuse disease (3 or more vessels requiring therapy) or disease of the left main coronary artery.8 Modern drug-eluting stents also require antiplatelet therapy that often delays listing for transplant.9 CABG has advantages in terms of definitive revascularization with no associated listing delay due to antiplatelet therapy, but increases the complexity of the transplant operation. Previous studies have demonstrated that acceptable short- and long-term outcomes are possible with both strategies.10,11
Our center has historically pursued a strategy of standardized surgical revascularization when lesions requiring therapy are documented during the transplant workup. We sought to determine the results of this strategy, and specifically whether lung transplant recipients with CAD who undergo or do not undergo CABG have comparable outcomes to those with no CAD. We hypothesized both CAD and CABG at the time of lung transplant would be associated with lower overall survival compared to those recipients with no CAD.
MATERIALS AND METHODS
We conducted a retrospective cohort study of all adult first-time double and single lung transplants performed in the University of Alberta Hospital Lung Transplant program between January 1, 2004 and December 31, 2013. We excluded patients transplanted beyond 2013, as we modified our approach to coronary disease during that period to include preoperative PCI based on contemporary literature.11 Recipient management protocols are available in the Supplemental Materials and Methods (SDC, http://links.lww.com/TP/B679). Data were sourced from our centers’ prospectively collected database for which patient consent was waived. This study was approved by the University of Alberta Health Research Ethics Board, Pro00075298.
The presence of CAD was determined by cardiac catheterization with coronary angiography. All candidates ≥ 50 years and those between 40 and 50 years with ≥ 1 risk factor for CAD (hypertension, diabetes mellitus, dyslipidemia, or smoking) undergo cardiac evaluation with left and right heart catheterization with coronary angiography. We defined 3 groups: candidates without CAD (NoCAD), candidates with CAD not requiring CABG (CAD-NoCABG), and candidates with CAD who underwent simultaneous CABG (CAD-CABG). Patients who did not require coronary angiography by the above criteria were designated NoCAD.
Coronary Artery Revascularization
During the study period, we treated CAD with either medical therapy alone or revascularization plus medical therapy, based on CAD severity. Our program’s approach to coronary revascularization over the study timeframe was to perform CABG at the time of lung transplantation. Revascularization was offered to candidates with severe (>70% luminal narrowing) single- or multi-vessel disease and preserved/mildly reduced ejection fraction. Candidates with moderate to severe reductions in ejection fraction were either considered ineligible for lung transplantation or offered combined heart-lung transplantation and not included for analysis in this study.
The primary outcome was time to death or retransplantation. Secondary outcomes included: 1-year survival; posttransplant development of grade 3 primary graft dysfunction (grade 3 PGD, defined as arterial oxygen [PaO2]/fraction of inspired oxygen [FiO2] ratio < 200 mm Hg) at 48 or 72 hours; duration of postoperative mechanical ventilation (time from operation end to permanent removal of support, including reintubation time); duration of intensive care unit (ICU) stay; duration of total hospitalization; and survival to hospital discharge.
Continuous data were compared between 2 groups using unpaired t-tests for normally distributed data and Wilcoxon rank sum for nonnormally distributed; for 3-group comparisons, ANOVA or Kruskal-Wallis tests were used. Categorical data were compared using Fisher exact χ2 tests for 2-group comparisons, and Pearson χ2 or trend tests for 3-group comparisons. Baseline data elements were compared using both 2-group comparisons (P < 0.05 is depicted in Tables 1 and 2 by a and b) and 3-way overall comparisons. Our primary survival analysis was via a multivariable proportional hazards regression model, adjusted for age, gender, and diagnostic indication for transplant. These variables were chosen via a priori designation of important covariates based on domain knowledge, and we restricted adjustment to these in the interest of model parsimony and to avoid overfitting. We tested additional variables post hoc—diabetes, recipient total pack years of smoking, donor age, total allograft ischemic time, and pack years of donor smoking—to ensure they were not sources of unmeasured confounding, and none were identified. Our secondary survival analyses included overall survival via Kaplan-Meier with log-rank testing across 3 groups and 1-year survival using χ2 testing. There was no missing data in the covariates or outcome in the proportional hazards model, so methods to handle missing data were not required. Statistical analyses were performed with JMP software version 12.1.0 (SAS Institute Inc., Cary, NC). A 2-tailed P value of <0.05 was considered statistically significant.
We performed a total of 352 adult transplant procedures over this period of time. After exclusion of 16 heart-lung and 3 retransplants, our remaining study cohort consisted of 333 double- and single-lung transplant recipients (Figure 1). Patients with CAD were more likely to be older, male, and have interstitial lung disease (ILD) (Table 1). The majority of procedures were double-lung transplants (317 patients, 95%), reflecting center preference. CAD was documented in 106 cases (32%). Of those, 24 patients underwent CABG (CAD-CABG) while 82 did not (CAD-NoCABG). The remaining 227 patients did not have CAD (NoCAD). The CAD-CABG group was more likely to have severe CAD, while the CAD-NoCABG group was more likely to have mild disease. All patients who underwent simultaneous CABG were supported intraoperatively using cardiopulmonary bypass (CPB). The majority of transplants procedures (317, 95%) were done with intraoperative CPB during the study timeframe, again reflecting center preference, with the remainder performed with no circulatory support (off-pump). There were no cases supported by intraoperative extracorporeal membrane oxygenation during the study period.
The 3-way CAD status was not associated with retransplant-free survival in our multivariable proportional hazards model after adjustment for age at transplant, gender, and indication for transplant (Wald χ2 test P = 0.34). No covariate violated the proportional hazards assumption (global P = 0.59). Survival differences were not observed between groups as assessed via pairwise hazard ratios: CAD-CABG versus NoCAD hazards ratio (HR) 1.44 (95% confidence interval [CI], 0.70-2.69, P = 0.30); CAD-CABG versus CAD-NoCABG HR 1.08 (95% CI, 0.53-2.05, P = 0.81); and CAD-NoCABG versus NoCAD HR 1.32 (95% CI, 0.86-2.03, P = 0.20).
Secondary analysis of survival by Kaplan-Meier and log-rank testing (Figure 2) showed that 3-way CAD status was associated with overall retransplant-free survival (P < 0.005). The median survival durations for each group were as follows: CAD-CABG: median 5.65 years (95% CI, 3.08-8.23); CAD-NoCABG 5.66 years (95% CI, 1.72-9.49); NoCAD 9.21 years (95% CI, 5.49-12.93). One-year survival was not different between groups by 3-way testing (CAD-CABG 74%, CAD-NoCABG 87%, NoCAD 90%; Pearson χ2 test P = 0.07), nor was survival to hospital discharge (P = 0.06, Table 2).
The development of grade 3 PGD at 48 or 72 hours was more common in the CAD-CABG and CAD groups (35% and 31%, respectively) compared with NoCAD group (15%, P = 0.04 and <0.01) (Table 2). Duration of posttransplant mechanical ventilation, ICU length of stay and hospital length of stay were all longer in the CAD-CABG and CAD groups compared with the NoCAD group. These parameters tended to be greater in the CAD-CABG compared with those in the CAD group, but not significantly so in any case.
The 3-way CAD status—CAD-CABG, CAD-NoCABG, and NoCAD—did not impact overall retransplant-free survival in our analysis. CAD and simultaneous CABG were associated with an increase in perioperative complexity.
The presented results are consistent with prior literature. Zanotti et al7 reported no survival association of mild-to-moderate CAD in their cohort, though the absence of severe CAD in their group may limit comparisons. Sherman et al10 also described no survival association in patients with CAD using a different protocol from ours, with a large number of patients having undergone preoperative PCI. Similarly, Castleberry et al11 noted no significant survival difference but did observe an increase in perioperative complexity in patients with the CABG group compared to the preoperative PCI group. Center preferences may affect outcomes and are difficult to account for in single-center studies, most notably in the presented study being our program’s preference for CBP over the study timeframe (Table 1) even in the non-CAD group, which may differ from other programs and potentially affect outcomes.12,13 Nonetheless, taken together, these experiences suggest that CABG can be safely performed at the time of lung transplantation in highly selected candidates, and that lung transplantation with more mild and moderate forms of CAD is reasonable.
We noted higher rates of severe PGD postoperatively in the CAD-CABG and CAD-NoCABG groups compared with the NoCAD cohort, and this is similar to the findings of Castleberry et al.11 This could plausibly be related at least 2 factors: worsening of existing PGD through myocardial stunning and diastolic dysfunction in the CABG group, resulting in greater left ventricular pressures postoperatively; and a greater burden of ILD as indication for transplant, both factors previously demonstrated to increase PGD risk.14
The discrepancy between the results of the adjusted proportional hazards model (primary) and the Kaplan-Meier analysis (secondary) indicates confounding by the factors for which we adjusted: age, gender, and underlying diagnosis. Patients with CAD were more likely to be male, older, and have ILD, the latter two being associated with worse outcome after lung transplant.2,3,15,16 As such, the results suggest that our concerns about worse survival in CAD patients were related to other group characteristics.
The order in which the CABG and the lung allograft implants are performed is a potential dilemma. At our institution, we have performed the CABG portion of the operation both before and after the lung allografts are implanted. Performing the coronary bypass before the lung transplant allows a longer reperfusion time for the heart prior to weaning from CPB while performing the coronary bypass after lung transplantation may have the advantage of avoiding any traction-related injuries to the bypass grafts. The respective advantages and disadvantages need to be weighed in context of institution-specific considerations, and we do not view one method as superior to the other.
This study has several limitations. Direct comparisons between groups are inherently problematic due to indication bias: patients with severe CAD are not equally likely to receive a transplant with or without CABG, and vice versa. A potential solution to this would be the use of propensity scores; however, the number of patients in the CAD-CABG group was too small to allow for this. Overall and group-to-group comparisons (Tables 1 and 2) as well the secondary survival analyses need to be interpreted with caution given multiple testing; these findings should be considered exploratory rather than confirmatory. Our study lacks outcome data on the development of postoperative cardiovascular events including stroke and myocardial infarction. Anecdotally, these are infrequent in all patient subgroups at our center due to aggressive pretransplant risk stratification and close lifelong clinical follow-up, but without a formal analysis, it is impossible to know their effect as a potential mediator of long-term survival. Our center does not consistently perform functional assessments for cardiac ischemia pretransplant, which may have further stratified physiologically relevant CAD in the moderate and severe groups. Finally, the small number of simultaneous CABG cases (n = 24) limits the study’s power with respect to this group, despite the fact that this is a relatively large cohort in the context of prior literature.
CAD—either requiring CABG at the time of lung transplant or not—did not impact overall retransplant-free survival after lung transplantation in our cohort. We feel CABG can be safely performed at the time of lung transplant at experienced centers willing to sustain the increase in perioperative complexity. Prospective studies comparing CAD interventions—CABG, PCI, or medical therapy—in eligible cases would be useful to better determine the optimal course of action in a complex patient population.
The authors acknowledge the University of Alberta Hospital and Alberta Health Services for use of resources.
1. Jones RM, Enfield KB, Mehrad B, et al. Prevalence of obstructive coronary artery disease in patients undergoing lung transplantation: case series and review of the literature.Catheter Cardiovasc Interv20148411–6
2. Weill D, Benden C, Corris PA, et al. A consensus document for the selection of lung transplant candidates: 2014—an update from the pulmonary transplantation council of the international society for heart and lung transplantation.J Heart Lung Transplant20153411–15
3. Yusen RD, Edwards LB, Dipchand AI, et al. The Registry of the International Society for Heart and Lung Transplantation_ Thirty-third Adult Lung and Heart–Lung Transplant Report—2016; Focus Theme_ Primary Diagnostic Indications for Transplant.HEALUN2016351170–1184
4. Ducloux D, Courivaud C, Bamoulid J, et al. Alloimmune responses and atherosclerotic disease after kidney transplantation.Transplantation2015991220–225
5. Lemieux I, Houde I, Pascot A, et al. Effects of prednisone withdrawal on the new metabolic triad in cyclosporine-treated kidney transplant patients.Kidney Int20026251839–1847
6. Oflaz H, Turkmen A, Kazancioglu R, et al. The effect of calcineurin inhibitors on endothelial function in renal transplant recipients.Clin Transplant2003173212–216
7. Zanotti G, Hartwig MG, Castleberry AW, et al. Preoperative mild-to-moderate coronary artery disease does not affect long-term outcomes of lung transplantation.Transplantation201497101079–1085
8. Serruys PW, Morice MC, Kappetein AP, et al.; SYNTAX InvestigatorsPercutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease.N Engl J Med200936010961–972
9. Bittl JA, Fihn SD, Fleisher LA, et al. 2016 ACC/AHA guideline focused update on duration of dual antiplatelet therapy in patients with coronary artery disease.J Thorac Cardiovasc Surg20161521243–1275
10. Sherman W, Rabkin DG, Ross D, et al. Lung transplantation and coronary artery disease.Ann Thorac Surg2011921303–308
11. Castleberry AW, Martin JT, Osho AA, et al. Coronary revascularization in lung transplant recipients with concomitant coronary artery disease.Am J Transplant201313112978–2988
12. Machuca TN, Collaud S, Mercier O, et al. Outcomes of intraoperative extracorporeal membrane oxygenation versus cardiopulmonary bypass for lung transplantation.J Thorac Cardiovasc Surg201514941152–1157
13. Ius F, Kuehn C, Tudorache I, et al. Lung transplantation on cardiopulmonary support: venoarterial extracorporeal membrane oxygenation outperformed cardiopulmonary bypass.J Thorac Cardiovasc Surg201214461510–1516
14. Chaikriangkrai K, Jyothula S, Jhun HY, et al. Impact of pre-operative coronary artery disease on cardiovascular events following lung transplantation.J Heart Lung Transplant2016351115–121
15. Wilson ME, Vakil AP, Kandel P, et al. Pretransplant frailty is associated with decreased survival after lung transplantation.J Heart Lung Transplant2016352173–178
16. Liu J, Jackson K, Weinkauf J, et al. Baseline lung allograft dysfunction is associated with impaired survival after double-lung transplantation.J Heart Lung Transplant2018377895–902