Objectives: Little is known about recipient characteristics and outcomes of patients with pulmonary arterial hypertension undergoing lung transplant, particularly those with systemic sclerosis-associated disease. Here, we describe the characteristics and outcomes of patients with pulmonary arterial hypertension undergoing lung transplant, focusing on systemic sclerosis-associated disease.
Materials and Methods: This retrospective study included pulmonary arterial hypertension patients undergoing lung or heart-lung transplant between July 1992 and August 2013 at a single center.
Results: We evaluated 51 patients with pulmonary arterial hypertension (37.3% males; mean age of 45.3 ± 11.9 y). Of 51 patients, 9 (17.6%) had systemic sclerosis-associated pulmonary arterial hypertension. Pulmonary arterial hypertension patients without systemic sclerosis-associated disease had higher mean pulmonary arterial pressure (P = .002), higher pulmonary vascular resistance (P = .008), and were more likely to have severe right ventricular systolic dysfunction (P = .006) than individuals with the disease. Mean hospital stay posttransplant was similar in the 2 groups (42.2 ± 43.3 vs 43.1 ± 19.4 d; P = .20). Higher pretransplant creatinine clearance (P = .0005), forced vital capacity (P = .01), and absence of vasopressor/inotrope use (P = .03) were associated with shorter hospital stay. Mortality for pulmonary arterial hypertension patients with versus without systemic sclerosis-associated disease was 0% versus 13% at 1 year, 29% versus 24% at 2 years, and 86% versus 53% at 5 years. Female sex (hazard ratio, 0.23; 95% confidence interval, 0.08-0.68) and less severe tricuspid regurgitation (hazard ratio, 0.31; 95% confidence interval, 0.11-0.89) were independently associated with long-term survival.
Conclusions: Pulmonary arterial hypertension patients with versus without systemic sclerosis-associated disease have comparable short-term and 2-year outcomes after lung or heart-lung transplant. Female sex and less severe tricuspid regurgitation were independently associated with better long-term survival. These outcomes did not vary when adjusted for the year of transplant.
Key words : Lung disease, Outcomes, Scleroderma, Survival
Pulmonary hypertension is a condition defined as a resting mean pulmonary artery pressure of ≥ 25 mm Hg. Pulmonary arterial hypertension (PAH) is characterized by pulmonary hypertension associated with a pulmonary artery occlusion pressure of ≤ 15 mm Hg and a pulmonary vascular resistance of > 3 Wood Units.1 Over the past 20 years, there has been remarkable progress in the treatment of PAH, and there are now 10 Food and Drug Administration-approved pharmacologic agents for this disease.2 Despite maximal medical therapy, many patients still develop progressive right heart failure,3 and lung transplant remains the only therapeutic option.2
Systemic sclerosis (SSc) is a chronic autoimmune disease that leads to life-threatening visceral involvement. In the lung, SSc usually manifests as interstitial lung disease and/or PAH. In the setting of SSc, the prevalence of PAH ranges from 5% to 12% and is a leading cause of death. Historically, the median survival of systemic sclerosis-associated PAH (SSc-PAH) was 12 months; however, this has now improved to 3 to 4.9 years,4 largely due to the approval of several PAH-specific therapies.5-8 Nevertheless, in some patients, SSc-PAH continues to progress, leaving lung transplant as the only viable therapeutic option. Similarly, in idiopathic PAH, survival has improved (current 5-year survival of 61%-65%)9; but for some patients, lung transplant remains necessary. Overall, the proportion of patients with PAH who receive transplants is low (at around 3%),10,11 and data are scarce on the characteristics and outcomes of this cohort, particularly the subgroup of patients with SSc-PAH.12,13
Interestingly, short-term outcomes after lung transplant for PAH patients are worse than for patients with diseases like chronic obstructive pulmonary disease, idiopathic pulmonary fibrosis, and cystic fibrosis.10 In fact, PAH patients who undergo lung transplant have higher 3-month mortality than other groups, due to early complications such as primary graft dysfunction.10,14 Nonetheless, those patients who survive more than 1 year had better 5- and 10-year survival (75% and 50%, respectively) than patients transplanted for chronic obstructive pulmonary disease (62% and 25%, respectively).10 The optimal time for listing PAH patients is challenging as some patients may improve with PAH-specific therapies, whereas others will deteriorate and become “too sick” for transplant.
Very few preoperative factors have been associated with 1-year mortality after lung transplant. These include older recipient age, lower cardiac output, lower forced vital capacity, need for high supplemental oxygen at rest, and a higher bilirubin level.10 A recent study proposed modifications to the lung allocation score for PAH patients, to better prioritize patients on transplant wait lists.15 The authors suggested adding 6-minute walk distance and right atrial pressure to the lung allocation score model.15 Limited investigations have tested the ability of peritransplant variables to predict survival after lung transplant in patients with PAH. Data are even more limited in the subgroup of patients with SSc-PAH because transplant is frequently denied due to concerns for poor outcomes. In fact, several transplant centers do not offer lung transplants to patients with SSc and comorbidities like gastroesophageal reflux disease, renal dysfunction, and skin fibrosis.15,16 Nevertheless, in carefully selected patients without significant extrapulmonary disease, survival rates are similar to individuals who undergo lung transplant for other lung diseases.12,16
The aim of the present study was to investigate the characteristics of patients with PAH who underwent lung transplant, examine their outcomes, and identify risk factors that predict mortality after transplant with a particular focus on SSc-PAH. We hypothesized that certain variables collected in the pre- and peritransplant period could predict short- and long-term outcomes in patients undergoing lung or heart and lung transplant for PAH.
Materials and Methods
This single-center retrospective study was approved by the Cleveland Clinic Institutional Review Board (study number 13-1164). Written informed consent was waived. We included patients with PAH17 who underwent lung or heart-lung transplant between July 1992 and August 2013. Patients were identified from the Cleveland Clinic Lung Transplantation Registry, which contains data on a large number of clinical, functional, spirometric, echocardiographic, and hemodynamic variables. Medical records of all PAH patients were reviewed by 2 investigators (SG and OM) to confirm transplant diagnosis. All patients with systemic sclerosis satisfied the American College of Rheumatology classification criteria for this disease18 and were further categorized as having either the limited cutaneous or the diffuse form.19
Patients with PAH or SSc-PAH were evaluated in accordance with the established international guidelines for selection of candidates for lung transplant.20 Patients with systemic sclerosis were not considered transplant candidates if they had renal impairment (defined as creatinine clearance < 50 mL/min), nonhealing or open skin wounds, severe sclerosis of the chest wall, and/or severe gastroesophageal reflux disease refractory to medical therapy. After transplant, patients were followed weekly for the first month, biweekly for the next 2 months, monthly for 6 months, and every 3 months thereafter.
Baseline data included age, sex, race, body mass index, World Health Organization functional class, duration of pulmonary hypertension, use of PAH-specific medications, serum sodium, total bilirubin, renal function (creatinine clearance by the Modification of Diet in Renal Disease equation21), and results of Doppler echocardiography, spirometry, and 6-minute walk test.22 We assessed the severity of tricuspid regurgitation and right ventricular (RV) dysfunction using echocardiography. Right heart catheterization was performed following standard protocols and using supplemental oxygen to keep the pulse oximetry saturation ≥ 90%. Mean pulmonary artery pressure, pulmonary capillary wedge pressure, pulmonary vascular resistance, transpulmonary gradient, and diastolic pulmonary gradient were recorded. Cardiac output was measured using both Fick method and thermodilution. Presence or absence of interstitial lung disease was documented based on computed tomography and spirometric evaluation.
We also documented donor and recipient cytomegalovirus status, pretransplant need for extracorporeal membrane oxygenation (ECMO), mechanical ventilation, and vasopressors/inotropes. We recorded type of transplant (single lung, double lung, or heart and lung transplant), surgical technique (sternotomy or thoracotomy), and intraoperative hypotension. Primary graft dysfunction was graded using the International Society of Heart and Lung Transplantation (ISHLT) working group consensus statement23 and acute renal failure defined as a rise of serum creatinine of ≥ 0.3 mg/dL or ≥ 50% increase from baseline within 72 hours after transplant. Reintubation was defined as the need to repeat endotracheal intubation within 48 hours of the first extubation after transplant. We recorded the need for ECMO after transplant, the duration of endotracheal intubation, the intensive care unit length of stay (ICU LOS), hospital length of stay (LOS), and discharge destination.
We assessed outcomes after discharge including the Karnofsky performance status24 at each outpatient visit, the number of hospital admissions, episodes of acute cellular rejection (defined per the ISHLT working formulation) within the first 12 months after transplant, and survival. Alive versus dead status was ascertained by review of the patient’s electronic medical record and querying the US Social Security Death Index.
Patient characteristics were described using means and standard deviations or medians and interquartile ranges for all continuous variables. Categorical variables are presented as counts and percentages. All analyses were 2-tailed and were performed at a significance level of 0.05. We focused on 2 important outcomes, including hospital LOS and long-term mortality. Univariate survival analyses (for mortality) and log-transformed linear regression (for total hospital LOS) were performed. Highly correlated variables were eliminated. Survival analyses were performed for patients with non SSc-PAH and patients with SSc-PAH using Kaplan-Meier methodology. Multivariable survival Cox model or linear regression analyses, using a stepwise approach, were performed on independent variables whose P values were < .1 in the univariate analyses. Sensitivity analyses were performed, adjusting for year of transplant as a continuous or categorical (before and after May 2005) variable. We used SAS 9.3 software (SAS Institute, Cary, NC, USA) for all analyses.
Fifty-one patients met our study criteria and were included in the analyses, comprising 42 patients with non-SSc-PAH (82.4%) and 9 patients with SSc-PAH (17.6%). Characteristics of the patients are shown in Table 1. Of note, 32 patients with a primary or secondary diagnosis of systemic sclerosis were evaluated for lung transplant during the study period, of which 9 patients with SSc-PAH were transplanted. Most of the patients in the SSc-PAH group had limited (8 patients, 88.9%) instead of diffuse systemic sclerosis. Among the non-SSc-PAH patients, 31 patients had idiopathic PAH, 9 had congenital heart disease-associated PAH, and 2 patients had connective tissue disease-associated PAH (systemic lupus erythematosus and rheumatoid arthritis).
At the time of transplant, most non-SSc-PAH patients were in World Health Organization functional class IV (87.5%), whereas most SSc-PAH patients were in World Health Organization functional class III. Compared with patients with SSc-PAH, patients with non-SSc-PAH had higher estimated RV systolic pressure (97 ± 22 vs 75 ± 21 mm Hg; P = .01) and a higher proportion had severe RV systolic dysfunction (83.3% vs 33.3% patients; P = .006). In addition, patients with non-SSc-PAH had worse pulmonary hemodynamics by right heart catheterization (mean pulmonary arterial pressure of 63.0 ± 19.3 vs 42.4 ± 12.3 mm Hg, P = .002; and pulmonary vascular resistance of 14.9 ± 13.6 vs 5.2 ± 2.5 Wood unit, P = .008) than patient with SSc-PAH (Table 1).
Outcomes (comparison between patient groups)
Most non-SSc-PAH patients underwent double lung or heart-lung transplant (39 patients, 92.8%). This percent was lower in SSc-PAH patients (7 patients, 77.8%). The mean duration of intubation was comparable in the 2 groups (7.7 ± 5.6 vs 8.9 ± 5.7 days; P = .55) (Table 2). More SSc-PAH patients required reintubation than non-SSc-PAH patients (87.5% vs 40.5%; P = .01). The mean ICU LOS in the non-SSc-PAH group was comparable to the SSc-PAH group (19.7 ± 20.4 vs 20.3 ± 13.2 days; P = .39), as was the mean hospital LOS (42.2 ± 43.3 vs 43.1 ± 19.4 days; P = .20). Patients in the non-SSc-PAH group had more readmissions in the first 12 months after transplant than patients with SSc-PAH (1.6 ± 1.5 vs 0.6 ± 1.1 episodes; P = .04). We found no significant differences in the incidence of acute renal failure (61.9% vs 62.5%; P = .97), need for ECMO posttransplant (14.3% vs 14.3%; P = 1.00), and primary graft dysfunction scores at 0, 24, 48, and 72 hours between the 2 groups.
Pretransplant characteristics and hospital length of stay for all patients
Using univariate linear regression models, we identified 10 pretransplant variables associated with hospital LOS in PAH patients (Table 3). Multivariate analyses showed that better pretransplant renal function, as evidenced by higher creatinine clearance (hazard ratio [HR] of -0.01; 95% confidence interval [CI], -0.02 to -0.01), higher pretransplant forced vital capacity (HR of -0.24; 95% CI, -0.42 to -0.06), and the lack of need for pretransplant vasopressors/inotropes (HR of -0.53; 95% CI, -1.02 to -0.04), was associated with shorter hospital LOS.
Long-term mortality in all patients
Univariate survival analyses identified 6 predictors of long-term mortality in PAH patients (Table 4). Final multivariate analyses showed that female sex (HR of 0.23; 95% CI, 0.08-0.68) and mild or moderate tricuspid regurgitation compared with severe tricuspid regurgitation (HR of 0.31; 95% CI, 0.11-0.89) were associated with better long-term survival. Similar results were observed when the multivariate analysis was performed for the non-SSc-PAH subgroup (Table 5). The mortality for the SSc-PAH group compared with the non-SSc-PAH group was 0% versus 13.2% at 1 year, 28.6% versus 23.7% at 2 years, and 85.7% versus 52.6% at 5 years (log-rank test P = .91) (Figure 1). There were no differences in survival based on the cause of PAH or the type of transplant (single lung transplant vs bilateral lung transplant vs heart-lung transplant) (Table 4). Importantly, there was no difference in survival based on the year of transplant (Table 4).
The year of transplant, as a continuous or binary (before and after May 2005) variable, had no significant impact on long-term mortality in the entire group of patients or in the subgroup with non-SSc-PAH. Similarly, year of transplant had no significant impact in the survival model. Hence, male sex and severe tricuspid regurgitation remained significant predictors of long-term mortality both in the entire cohort and in the non-SSc-PAH subgroup. When we adjusted the model that predicted LOS by year of transplant, both creatinine clearance and FVC remained significant predictors of LOS. Meanwhile, the need for vasopressors did not reach statistical significance (P = .07). These latter results were noted in all study patients and in the subgroup with non-SSc-PAH.
Our study showed that, at the time of transplant, patients with non-SSc-PAH have worse RV function and pulmonary hemodynamics than patients with SSc-PAH. We also found that, despite worse pretransplant pulmonary hemodynamics and RV function, patients with non-SSc-PAH had similar length of ICU stay, hospital stay, and short-term and 2-year mortality versus those patients with SSc-PAH. In PAH patients undergoing transplant, better renal function, higher pretransplant vital capacity, and absence of pretransplant vasopressor/inotrope use were associated with shorter hospital LOS. Furthermore, female sex and less severe tricuspid regurgitation provided a survival benefit. We included patients over a period of 21 years, and, despite several changes in the treatment of PAH over the years, the year of transplant did not affect outcomes in our study.
The Registry to Evaluate Early and Long-Term PAH Disease Management compared medically treated patients with connective tissue disease-associated PAH with patients with idiopathic PAH.25 In this study, patients with connective tissue disease-associated PAH, including those with SSc-PAH, had better hemodynamic measurements on right heart catheterization and were less likely to have RV systolic dysfunction, findings that coincide with our results.
Bilateral lung transplant is the procedure of choice for patients with PAH who are refractory to medical treatment.2 This is largely because PAH and RV dysfunction increase the risk of perioperative complications.26 Accordingly, in our study, the large majority of PAH patients underwent bilateral lung or heart-lung transplant. There is no consensus regarding the preferred transplant strategy for patients with SSc-PAH. Of the 29 patients with systemic sclerosis described by Schachna and associates,12 62% underwent single lung transplant, whereas Saggar and associates13 preferentially performed bilateral lung transplant in their 14 patients with this disease. In our study, of the 9 patients with SSc-PAH, 77.8% underwent bilateral lung transplant. This reflects the overall trend toward performing bilateral lung transplant in PAH patients despite cause.
We found no significant differences in the incidence of acute renal failure and need for ECMO posttransplant between the 2 groups. Furthermore, ICU and hospital LOS were similar for patients with SSc-PAH and non-SSc-PAH. However, a higher percent of patients with SSc-PAH required reintubation. This is somewhat surprising in view of the better pretransplant hemodynamics and RV function. We found that a better renal function, higher pretransplant vital capacity, and absence of pretransplant vasopressor/inotrope need were associated with shorter hospital LOS. Previous studies have shown that the presence of pretransplant pulmonary hypertension increases the likelihood of posttransplant renal dysfunction.27,28 Pre- and intraoperative use of vasopressors and abnormal renal function are known predictors of postoperative morbidity, especially among patients with pulmonary hypertension.28 Better lung function parameters are recognized to be associated with improved postoperative outcomes, likely due to improved clearance of secretions and shorter time on mechanical ventilator support.29 The association between higher pretransplant vital capacity and shorter LOS may reflect a better pretransplant muscular strength, which could be associated with shorter convalescence. Notably, non-SSc-PAH patients had more readmissions within the first year posttransplant than SSc-PAH subjects. The reasons for this latter finding are unclear.
There are limited data on the long-term outcomes of patients with PAH undergoing lung and heart-lung transplant. Most single-center and registry reports show higher early mortality in patients with PAH undergoing lung transplant than with other lung diseases.30,31 The ISHLT registry shows that patients with PAH have the highest unadjusted 3-month mortality (23%).10 However, among patients surviving at least 1 year, PAH subjects had better long-term survival (10.0 y) than individuals transplanted for chronic obstructive pulmonary disease or idiopathic pulmonary fibrosis (6.8 years for both).10 In our study, the mortality was 13.2%, 23.7%, and 52.6% at 1, 2, and 5 years in the non-SSc-PAH group, which is similar to the overall lung transplant mortality reported in the literature.10 Khan and associates found that short- and intermediate-term survival after lung transplant are similar in SSc compared with idiopathic PAH and interstitial lung disease patients.4 Our study showed similar short- and long-term survival, in carefully selected SSc-PAH patients.4 Although there was no statistical difference in mortality between the patients with SSc-PAH and non-SSc-PAH, the 5-year mortality appears higher in the SSc-PAH group (86% vs 53%). The lack of statistical difference may have resulted from the lack of power given the small number of patients with SSc-PAH.
Schaffer and associates32 reported that, in patients with PAH, a higher cardiac index and forced vital capacity, double lung transplant, or heart-lung transplant predicted survival, whereas male sex, worse functional status (New York Heart Association class IV), need for life support, creatinine clearance < 50 mL/min, and being listed for heart and lung transplant predicted mortality.32 Similarly, in our study, male sex was associated with higher mortality. Studies of medically treated PAH patients showed that male sex independently predicts worse outcomes.33-35 Interestingly, this relation appears to persist even after transplant, as seen in our study. Male sex has been associated with worse RV function in PAH patients, which may explain the worse outcomes.36
In our study, the presence of mild/moderate tricuspid regurgitation conveyed a survival advantage compared with severe tricuspid regurgitation. In medically treated patients with PAH, severity of tricuspid regurgitation predicts mortality and the need for transplant36; furthermore, improvement in the severity of tricuspid regurgitation with treatment is associated with a survival benefit.37 Interestingly, in our study, of the 25 patients with moderate tricuspid regurgitation, 6 (24%) underwent double lung transplant with concomitant valve repair, whereas none of the patients with mild tricuspid regurgitation underwent concomitant valve repair with double lung transplant. Combining tricuspid valve repair with double lung transplant in patients with severe pulmonary hypertension and tricuspid regurgitation has been shown to decrease primary graft dysfunction and improve pulmonary function with improved outcomes.38 This might explain the possible survival benefit that was observed in patients with mild/moderate tricuspid regurgitation compared with other groups.
There are limitations to the study, including its single center origin and retrospective nature. We had a relatively small sample size; however, PAH is an uncommon disease and transplant has become less frequent with the approval of effective medical treatments. Furthermore, only a few patients with SSc-PAH undergo transplant; hence, there are limited data on the characteristics and outcomes in this small but important population. In our study, we presented a relatively large sample size of patients with SSc-PAH31 and gathered a fairly robust data set, accounting for several variables associated with outcomes in patients with PAH. Interestingly, the posttransplant LOS and mortality were not affected by the year of transplant or time of lung allocation score implementation. Despite these limitations, our study adds important information to the scarce evidence regarding pretransplant characteristics and transplant outcomes in this population.
Patients with SSc-PAH have comparable short-term and 2-year outcomes versus patients with non-SSc-PAH. In PAH patients undergoing transplant, better renal function, higher forced vital capacity, and absence of pretransplant vasopressor/inotropic use were associated with a shorter hospital length of stay; meanwhile, female sex and milder tricuspid regurgitation were associated with better long-term survival.
Volume : 15
Issue : 6
Pages : 676 - 684
DOI : 10.6002/ect.2016.0209
From the the 1Department of Pulmonary, Allergy and Critical Care Medicine,
Respiratory Institute, Cleveland Clinic, Cleveland, Ohio, USA; the 2Pulmonary
and Critical Care, Southside Regional Medical Center, Petersburg, Virginia, USA;
and the 3Respiratory Institute Biostatistics Core, Quantitative Health Sciences,
Cleveland Clinic, Cleveland, Ohio, USA; 4Quality Improvement, Gold Coast Health
Plan, Camarillo, CA, USA
Acknowledgements: All authors have participated in the design of the study, data collection, interpretation of the results and critical revision of the manuscript for important intellectual content, and final approval of the manuscript submitted. A. R. Tonelli is supported by CTSA KL2 (Grant no. TR000440) from the National Center for Research Resources, a component of the National Institutes of Health Roadmap for Medical Research. Omar A. Minai is a member of the Scientific Advisory Boards of Actelion, Gilead, and Bayer and a member of the Speakers Bureau for Actelion and Gilead. The other authors have no conflicts of interest to declare.
Corresponding author: Adriano R. Tonelli, 9500 Euclid Avenue A-9, Cleveland, Ohio 44195, USA
Phone: +1 2164440812
Table 1. Patient Characteristics
Table 2. Outcomes in Study Patients
Table 3. Univariate and Multivariate Analyses for Hospital Length of Stay in All Study Patients (n = 51)
Table 4. Univariate and Multivariate Survival Analyses for Long-Term Mortality for All Study Patients (n = 51)
Table 5. Univariate and Multivariate Survival Analyses for Long-Term Mortality for Pulmonary Arterial Hypertension Patients Without Systemic Sclerosis (n = 42)
Figure 1. Kaplan-Meier Survival Curve for Study Patients