Outcomes after lung retransplantation for restrictive allograft syndrome have not improved over 2 decades Authors: Vikram Krishna, Allen A Razavi, Brian Lim, Zachary Brennan, Qiudong Chen, Laura DiChiacchio, Dominick Megna, Joanna Chikwe, Michael E Bowdish, Pedro Catarino Published: 2026 ABSTRACT AbstractObjectiveIn 2015, restrictive allograft syndrome (RAS) was first definitively associated with poorer outcomes after lung retransplantation compared with bronchiolitis obliterans syndrome (BOS). This study aimed to evaluate how strategies and outcomes of lung retransplantation for RAS have evolved over time.MethodsThe United Network for Organ Sharing database was queried to identify adult lung retransplants for RAS between 2005 and 2025. Primary transplants, multiorgan transplants, and retransplants with unknown indications were excluded. Patients were stratified into Era 1 (2005-2015) and Era 2 (2016-2025). The primary outcome was 5-year survival.ResultsA total of 293 patients were included, with 141 (48%) in Era 1 and 152 (52%) in Era 2. The proportion of retransplants for restrictive disease was similar across eras (19% vs 22%, P = .16). Compared with Era 1, patients from Era 2 had greater allocation scores, less preoperative ventilator dependence, and similar waitlist times. Retransplants in Era 2 were characterized by longer ischemia times and a significantly greater use of bilateral retransplantation (65% vs 41%, P < .001). Postoperatively, patients in Era 2 required more prolonged mechanical ventilation, inhaled nitric oxide, and extracorporeal membrane oxygenation support. Thirty- and 90-day mortality were similar between eras. On multivariable analysis, era of retransplantation was not associated with improved 5-year survival (adjusted hazard ratio, 0.98; 95% CI, 0.70-1.38, P = .93)ConclusionsSince 2015, lung retransplantation for RAS has increasingly favored bilateral procedures, accompanied by greater perioperative support. Despite these changes, long-term survival remains poor, underscoring the need for improved candidate selection and strategies to prevent recurrent chronic allograft injury.Key Words: lung, retransplantation, survival, restrictive allograft syndrome, chronic lung allograft dysfunction Long-term survival remains poor for RAS retransplant recipients. Central Message Despite an increase in bilateral procedures and greater perioperative support, retransplantation for restrictive allograft syndrome has not improved long-term survival. Perspective As retransplantation is increasingly considered for restrictive allograft syndrome, understanding its long-term value is essential. With our findings, we aim to help inform patient counseling and candidate selection as well as highlight the need to align technical feasibility with expected benefit when managing chronic allograft failure. Chronic lung allograft dysfunction (CLAD) remains the principal limitation to long-term survival after lung transplantation.1,2 Among its phenotypes, restrictive allograft syndrome (RAS) has emerged as a particularly severe form, characterized by progressive loss of lung volume, fibrotic remodeling, and markedly worse survival compared with bronchiolitis obliterans syndrome (BOS).1,3 Since its formal recognition as a distinct entity, RAS has been consistently associated with poor outcomes after both primary transplantation and retransplantation.3, 4, 5 Lung retransplantation represents the only definitive treatment option for advanced CLAD, and outcomes after retransplantation for RAS have historically been inferior to those for other indications.4,6 However, over the past 2 decades, substantial advances in surgical technique, organ preservation, allocation systems, and critical care have expanded the technical feasibility of lung transplantation in patients with increasing complex.7,8 Despite these advances, it remains unclear whether contemporary retransplantation strategies have translated into improved outcomes for patients with RAS. Previous studies have been limited by small sample sizes, short follow-up, or single-center design.9 Furthermore, few have evaluated long-term trends in outcomes alongside evolving perioperative practices at a national level.2 The purpose of our study was to examine temporal changes in practice patterns and survival outcomes after lung retransplantation for restrictive allograft dysfunction over a 20-year period using a national retrospective cohort. Given major advances in surgical technique, organ allocation, and perioperative support in the modern era, we hypothesized that long-term survival after lung retransplantation for RAS would improve over time. Methods Data Source This national retrospective cohort study was performed using the United Network of Organ Sharing (UNOS) national thoracic organ transplantation database for all adult lung transplants from January 2005 to March 2025. The database included information from all transplant recipients, wait-listed candidates, and donors in the United States, which was submitted by members of the Organ Procurement and Transplant Network. The study was exempt from review by the Cedars-Sinai Medical Center Institutional Review Board because all data were deidentified. Study Design We identified all adult patients (≥18 years) in the UNOS database who underwent first-time lung retransplantation during the study period. Primary transplants, those with multiple retransplants, combined multiorgan transplants, retransplants from unknown causes, patients younger than 18 years of age, and those with missing survival data were excluded. Retransplant indication for RAS was identified using the UNOS code “TCR_DGN,” which refers to the primary diagnosis at listing. On the basis of our cohort of all retransplants, codes 1552, 1556, and 1604 were used to identify lung retransplants for RAS. Because RAS was first definitively associated with poorer outcomes after retransplantation in 2015,4 patients were stratified into Era 1 (2005-2015) and Era 2 (2016-2025) on the basis of year of retransplantation. The primary outcome was 5-year survival (median follow-up, 8.0 [6.01-11.9] years). Secondary outcomes included perioperative support requirements and 90-day mortality. Statistical Analysis Baseline characteristics and postoperative outcomes between eras were compared using the χ2 test or Fisher exact test for categorical variables and Student t test or Wilcoxon rank-sum test for continuous variables. Continuous variables were reported using mean (±standard deviation) or median (quartile 1-quartile 3) on the basis of their distribution. Survival curves were constructed using the Kaplan-Meier method and stratified by era. Adjusted survival comparisons were performed using a multivariable Cox proportional hazards regression model, accounting for covariates selected on the basis of clinical relevance. The proportional hazards assumption was evaluated using Schoenfeld residual testing, and no violations were observed for retransplantation era. Analyses were performed using complete-case data, and observations missing any variables for a given model were excluded. All statistical analyses were performed using STATA 19.5 (StataCorp). Results Recipient and Donor Characteristics Between 2005 and 2025, a total of 1453 adult retransplantations were performed in the United States, with 753 (51.8%) in Era 1 and 700 (48.2%) in Era 2. Among all retransplants, 293 (20.2%) were done for RAS. In total, 141 (48.1%) patients were transplanted in Era 1 (2005-2025) and 152 (51.8%) were transplanted in Era 2 (2016-2025). The proportion of retransplants performed for restrictive disease did not differ significantly between eras (18.7% [141/753] vs 21.7% [152/700], P = .16, Figure 1). Baseline donor and recipient characteristics are listed in Table 1. The median recipient age of retransplants was similar between eras (Era 1: 54.0 [46.0-63.0] vs Era 2: 53.0 [37.5-62.0], P = .09). Lung allocation scores were lower at match for Era 1 recipients, compared with Era 2 (50.0 [41.6-72.3] vs 57.3 [42.3-85.8], P = .05). Race and body mass index did not differ significantly between the eras. Markers of end-organ function differed significantly over time. Recipients from Era 1 had greater serum creatinine (1.10 [0.88-1.30] vs 0.97 [0.70-1.22], P < .001), greater total bilirubin (0.50 [0.30-0.80] vs 0.35 [0.20-0.60], P < .001), and required more preoperative ventilator support (15.6% vs 7.9%, P = .04) at the time of transplant. Bar graph comparing the rate of retransplantation for restrictive allograft syndrome among all retransplants stratified by era (Era 1 [2005-2015] and Era 2 [2016-2025]). Table 1Baseline donor and recipient characteristics of retransplant recipients for restrictive allograft syndrome stratified by era (n = 293)Variable2005-2015 (n = 141)2016-2025 (n = 152)P valueDonor Donor sex, female58 (41.1%)60 (39.5%).77 Donor age, y31 (21-46)32.5 (24-44).79 Donor Po2/Fio2, mm Hg419 (294-490)435 (334-504).35 DCD donor2 (1.4%)7 (4.6%).11 Donor cause of death.01 Anoxia66 (46.8%)56 (36.8%) CNS tumor1 (0.7%)0 (0.0%) Head trauma22 (15.6%)50 (32.9%) Stroke49 (34.8%)43 (28.3%) Unknown/other3 (2.1%)3 (2.0%) EVLP use0 (0.0%)11 (7.2%).001Recipient Age54.0 (46.0-63.0)53.0 (37.5-62.0).09 Female sex42 (29.8%)61 (40.1%).06 Race/ethnicity.15 White114 (80.9%)112 (73.7%) Asian3 (2.1%)1 (0.7%) Black14 (9.9%)15 (9.9%) Hispanic10 (7.1%)23 (15.1%) Other0 (0.0%)1 (0.7%) BMI24.19 (21.29-27.84)24.46 (21.10-28.47).61 LAS at match50.0 (41.60-72.26)57.29 (42.25-85.84).05 Total bilirubin, mg/dL0.50 (0.30-0.80)0.35 (0.20-0.60)<.001 Creatinine, mg/dL1.10 (0.88-1.30)0.97 (0.70-1.22).002 ECMO support at transplant6 (4.3%)14 (9.2%).09 Ventilation at transplant22 (15.6%)12 (7.9%).04 Dialysis at transplant3 (2.1%)5 (3.3%).53 Functional status.29 Mild (80%-100%)2 (1.4%)4 (2.7%) Moderate (50%-70%)56 (39.7%)53 (35.3%) Severe (10%-40%)83 (58.9%)90 (60.0%) Unknown0 (0.0%)3 (2.0%) Transplant type<.001 Double lung58 (41.1%)99 (65.1%) Single lung83 (58.9%)53 (34.9%) Waitlist time, d27 (9-81)29 (12-64.5).52 Ischemic time, h4.70 (3.52-5.85)6.37 (4.77-8.35)<.001 Travel distance, nautical miles114 (12-351)202 (83.5-436).002 Primary transplant indication.041 Group A: obstructive disease10 (8.4%)21 (13.6%) Group B: pulmonary vascular disease4 (3.4%)7 (4.5%) Group C: cystic fibrosis and immunodeficiency10 (8.4%)27 (17.5%) Group D: restrictive lung disease95 (79.8%)99 (64.3%) Primary transplant type<.001 Double lung51 (42.9%)99 (64.3%) Single lung68 (57.1%)55 (35.7%) Interval between primary and retransplant, y1.65 (0.46-3.22)3.71 (1.84-6.32)<.001Values are expressed in % (n) or median (interquartile range). Bolded values indicate statistical significance. Po2, Partial pressure of oxygen; Fio2, fraction of inspired oxygen; DCD, donation after circulatory death; CNS, central nervous system; EVLP, ex vivo lung perfusion; BMI, body mass index; LAS, lung allocation score; ECMO, extracorporeal membrane oxygenation. Baseline donor and recipient characteristics of retransplant recipients for restrictive allograft syndrome stratified by era (n = 293) Values are expressed in % (n) or median (interquartile range). Bolded values indicate statistical significance. Po2, Partial pressure of oxygen; Fio2, fraction of inspired oxygen; DCD, donation after circulatory death; CNS, central nervous system; EVLP, ex vivo lung perfusion; BMI, body mass index; LAS, lung allocation score; ECMO, extracorporeal membrane oxygenation. Donor age and partial pressure of oxygen/inspired oxygen fraction did not differ across eras. The rate of donation after circulatory death donors between both eras was similar (Era 1: 1.4% vs Era 2: 4.6%, P = .11). The use of ex vivo lung perfusion was more prominent in Era 2 compared with Era 1 (7.2% vs 0.0%, P = .001). Era 2 was also associated with longer ischemia time (6.37 [4.77-8.35] hours vs 4.70 [3.52-5.85] hours, P < .001) and travel distances (202 [83.5-436] nautical miles vs 114 [12-351] nautical miles, P = .002). There was no difference in waitlist days over time (Era 1: 27 [9-81] vs Era 2: 29 [12-64.5] days, P = .51). There was a marked shift toward bilateral lung retransplantation over time as double-lung procedures increased significantly in Era 2 compared with Era 1 (65.1% vs 41.1%, P < .001). The distribution of primary transplant indications varied significantly, with a lower proportion of restrictive disease in Era 2 (64.3% vs 79.8%) and a relative increase in cystic fibrosis and obstructive lung disease (P = .04). There was also a greater proportion of bilateral transplants in Era 2 compared with Era 1 (64.3% vs 42.9%, P < .001). In addition, the interval between primary and retransplantation was significantly longer in the contemporary era (3.71 [1.84-6.32] years vs 1.65 [0.46-3.22] years, P < .001). Postoperative Outcomes Postoperative outcomes for retransplants are listed in Table 2. Recipients who underwent retransplant in Era 2 experienced significantly greater perioperative support requirements. Prolonged mechanical ventilation was more common in Era 2 compared with Era 1, including ventilation exceeding 48 hours (60.5% vs 44.7%, P = .007) and intubation beyond 72 hours (47.4% vs 9.9%, P < .001). Table 2Postoperative outcomes of recipients who underwent retransplant for restrictive allograft syndrome stratified by era (n = 293)VariableEra 1: 2005-2015 (n = 141)Era 2: 2016-2025 (n = 152)P valueVentilated >48 h63 (44.7%)92 (60.5%).007Intubated ≥72 h14 (9.9%)72 (47.4%)<.001Inhaled nitric oxide ≥72 h0 (0.0%)31 (20.4%)<.001Airway dehiscence1 (0.7%)4 (2.6%).20Posttransplant stroke2 (1.4%)6 (4.0%).18Posttransplant dialysis15 (10.6%)34 (22.7%).006ECMO at 72 h (n = 180)1 (3.3%)31 (20.7%).02Reintubation26 (18.4%)38 (25.3%).16Acute rejection10 (7.1%)8 (5.3%).5330-d mortality10 (7.1%)12 (7.9%).8090-d mortality20 (14.2%)25 (16.4%).59Cause of death.48 Cardiovascular6 (4.7%)6 (8.1%) Cerebrovascular10 (7.8%)2 (2.7%) Graft failure22 (17.1%)11 (14.9%) Hemorrhage1 (0.8%)1 (1.4%) Infection23 (17.8%)17 (23.0%) Malignancy17 (13.2%)4 (5.4%) Multiorgan failure8 (6.2%)6 (8.1%) Other2 (1.6%)3 (4.1%) Pulmonary25 (19.4%)17 (23.0%) Unknown15 (11.6%)7 (9.5%)Values are expressed in % (n). Bolded values indicate statistical significance. ECMO, Extracorporeal membrane oxygenation. Postoperative outcomes of recipients who underwent retransplant for restrictive allograft syndrome stratified by era (n = 293) Values are expressed in % (n). Bolded values indicate statistical significance. ECMO, Extracorporeal membrane oxygenation. Use of inhaled nitric oxide for more than 72 hours (20.4% vs 0%, P < .001) and rates of postoperative dialysis (22.7% vs 10.6%, P = .006) were also increased substantially in Era 2. Postoperative extracorporeal membrane oxygenation support use was also increased in Era 2 compared with Era 1 (20.7% vs 3.3%, P = .02). Rates of airway dehiscence, stroke, reintubation, and acute rejection were similar between eras. Despite increased perioperative support, 30-day mortality (Era 1: 7.1% vs Era 2: 7.9%, P = .80) and 90-day mortality (Era 1: 14.2% vs Era 2: 16.4%, P = .59) did not differ between groups. Causes of death after retransplantation were similar between eras. Graft failure, infection, and pulmonary causes accounted for most deaths in both periods, with no significant shift in cause-specific mortality over time. Survival Analysis Five-year survival for all retransplants improved in Era 2 compared with Era 1 (Era 1: 38.5%; 95% Cl, 35.0-42.1 vs Era 2: 44.7% 95% Cl 39.8-49.6, Figure 2). On multivariable analysis adjusting for recipient, donor, and perioperative factors, Era 2 was associated with improved survival, compared with Era 1 (adjusted hazard ratio [aHR], 0.85; 95% CI, 0.73-0.99, P = .04, Table E1). In contrast, 5-year survival after retransplantation for RAS remained poor across eras (Era 1: 32.6%; 95% CI, 25.0-40.5 vs Era 2: 32.3%; 95% CI, 22.5-42.4, Figure 3). On multivariable analysis, era of transplantation was not associated with improved survival (aHR, 0.98; 95% CI, 0.70-1.38, P = .93, Table 3). Greater allocation score (aHR, 1.01; 95% CI, 1.00-1.02, P = .02) and donor death from stroke (aHR, 1.57; 95% CI, 1.04-2.37, P = .03) were independently associated with increased mortality. Longer waitlist-time was also associated with worse survival (aHR 0.998; 95% CI, 0.996-0.999, P = .02). Kaplan-Meier survival curve depicting 5-year overall survival of all retransplants by era. The shaded area represents 95% CI. aHR, Adjusted hazard ratio. Kaplan-Meier survival curve depicting 5-year overall survival of retransplants for restrictive allograft syndrome stratified by era. The shaded area represents 95% CI. aHR, Adjusted hazard ratio. Table 3Multivariable Cox regression model for survival of recipients of retransplants for restrictive allograft syndrome from 2005 to 2025VariableHazard ratio (95% CI)P valueEra Era 1: 2005-2015Reference Era 2: 2016-20250.98 (0.70-1.38).93Donor Age, y1.00 (0.98-1.01).71 Po2/Fio2, mm Hg1.00 (0.99-1.00).75 Cause of death AnoxiaReference Head trauma1.29 (0.87-1.90).21 Stroke1.57 (1.04-2.37).03 Unknown/other1.52 (0.53-4.30).43Recipient Age, y1.01 (1.00-1.02).09 Male sex0.90 (0.66-1.24).52 Allocation score at transplant1.01 (1.00-1.02).02 Total bilirubin at transplant, mg/dL1.24 (0.97-1.57).09 Creatinine at transplant, mg/dL0.90 (0.61-1.33).60 ECMO support at transplant0.60 (0.30-1.19).14 Ventilation at transplant1.12 (0.70-1.79).65 Days on waitlist0.998 (0.996-0.999).02 Ischemia time, h0.99 (0.93-1.05).78Bolded values indicate statistical significance. Po2, Partial pressure of oxygen; Fio2, fraction of inspired oxygen; ECMO, extracorporeal membrane oxygenation. Multivariable Cox regression model for survival of recipients of retransplants for restrictive allograft syndrome from 2005 to 2025 Bolded values indicate statistical significance. Po2, Partial pressure of oxygen; Fio2, fraction of inspired oxygen; ECMO, extracorporeal membrane oxygenation. Discussion In this national analysis of lung retransplantation for restrictive chronic lung allograft dysfunction spanning 2 decades, we demonstrated that long-term outcomes have remained largely unchanged despite substantial evolution in transplant strategy and perioperative management. Although contemporary retransplant recipients exhibited greater allocation scores, underwent bilateral retransplantation more frequently, and required markedly greater perioperative support, early mortality and 5-year survival did not improve over time. Our findings suggest that retransplantation alone has not overcome the fundamental biological drivers of restrictive allograft failure. Importantly, when placed in the context of overall retransplantation outcomes, these findings suggest that the lack of improvement in RAS is not simply attributable to a failure to advance retransplantation more broadly. In our cohort, 5-year survival for all lung retransplants improved in the contemporary era, whereas outcomes for RAS remained unchanged. This divergence supports the hypothesis that RAS represents a distinct high-risk phenotype in which advances in surgical technique and perioperative care have not translated into improved long-term outcomes. Since RAS was first recognized as a distinct phenotype of chronic lung allograft dysfunction with particularly poor prognosis, transplant programs adapted their clinical approach to improve outcomes.10,11 However, reports continue to demonstrate inferior survival after retransplantation for RAS compared with BOS, which may have prompted changes in operative strategy, including a preference for bilateral retransplantation and broader acceptance of perioperative risk.2,3,7,12 Our data confirm that in the contemporary era, recipients who underwent retransplant more frequently experienced significantly longer ischemia times, required substantially greater postoperative mechanical and renal support, and more frequently underwent double-lung retransplantation. The increase in ischemia time may be partially explained by the greater proportion of bilateral retransplantation, which are inherently more complex and often associated with longer operative and procurement times. Temporal changes in retransplantation for RAS are further informed by differences in primary transplant characteristics. In the contemporary era, fewer patients had restrictive disease as the initial indication, with increases in cystic fibrosis and obstructive disease, reflecting evolving transplant populations. More patients also underwent bilateral transplantation and had longer intervals to retransplantation, suggesting a shift toward more chronic graft dysfunction. However, despite these differences, survival after retransplantation for RAS remained unchanged across eras. The findings suggest that variation in patient population and disease trajectory do not account for poor outcomes, supporting a central role for mechanisms intrinsic to RAS. The lack of survival improvement also implies that the drivers of failure in RAS cannot be addressed by surgical extent alone. Restrictive allograft dysfunction is characterized by diffuse fibrotic remodeling of lung parenchyma and pleura, often accompanied by immune-related injury and antibody-driven processes.1,4,13 These mechanisms may be systemic or host-mediated, which raises the possibility that retransplantation may simply reintroduce susceptible tissue into an unchanged immunological environment. In this context, retransplantation does not address the underlying biological disease. An important finding from this study is that the modern era of retransplantation for RAS was associated with a significantly increased use of advanced perioperative adjuncts. Era 2 recipients experienced greater rates of prolonged mechanical ventilation, inhaled nitric oxide use, dialysis, and extracorporeal membrane oxygenation. These patterns likely reflect both increased willingness to transplant high-risk candidates and broader availability of supportive technologies.2 Notably, the increased use of perioperative adjuncts occurred despite modest improvements in selected pretransplant markers of physiological reserve, including renal and hepatic function, and reduced reliance on preoperative ventilation support. Contemporary transplant practice also evolved with changes in donor use and organ preservation. In our cohort, donation after circulatory death use remained low and similar between eras, while EVLP was more frequently used in the contemporary era. Despite these advances, survival did not improve, suggesting that expanded donor strategies, like perioperative support, increase feasibility but do not alter the long-term trajectory of restrictive allograft failure. Importantly, the greater rates of perioperative support observed in the contemporary era should not be interpreted as evidence that the biological severity of RAS has worsened over time. Rather, these differences likely reflect changes in clinical practice, including a lower threshold for deployment of advanced supportive therapies and possible standardization of perioperative management. Despite the more frequent use of perioperative adjuncts, survival outcomes remained unchanged across eras. Adjunctive therapies may facilitate temporary physiological stabilization, but they do not appear to alter the underlying trajectory of restrictive allograft failure. The lack of survival improvement over the 2 decades also highlights the importance of the timing of retransplantation in RAS. Although median waitlist times were relatively short and similar between eras, longer waitlist time was associated with worse survival. This association should be interpreted in the context of disease trajectory rather than duration alone, as the effect size was modest. Waitlist time may function as a surrogate for underlying disease severity or clinical instability. Together, these findings suggest that by the time retransplantation is pursued, some patients may already have progressed beyond a threshold at which graft replacement provides durable benefit. Although current clinical diagnostic tools may not reliably distinguish patients with potentially reversible disease, early identification of restrictive dysfunction, before advanced fibrosis and pleural involvement develop, may be critical to improving outcomes.5,14, 15, 16 The study findings have important implications for candidate selection and patient counseling. Although retransplantation remains an important option for select patients with chronic allograft failure, our data suggest that outcomes for RAS should be taken into consideration independently from other indications for retransplantation. The persistent lack of survival benefit despite an increased perioperative burden underscores the need for individualized risk-benefit assessment and transparent discussion regarding expected outcomes. Instead of a broad application of retransplantation for RAS, a more selective approach may be warranted. Clinicians should consider phenotypic severity, decline trajectory, immunologic risk, and center experience. Our data do not argue against retransplantation for RAS but rather suggest that technical escalation alone may not be enough to address underlying biological disease. The increased resource intensity associated with retransplantation for RAS also warrants careful consideration. Lung transplantation operates within a framework of profound organ scarcity, and operations that consume increasing perioperative resources without demonstrable survival benefit require ongoing evaluation.1,2,12,17 Meaningful progress in the management of RAS will likely require a shift away from surgical escalation toward early intervention and disease modification. Improved phenotyping of CLAD, integration of immunologic and molecular biomarkers, and targeted antifibrotic/immunomodulatory therapies may provide additional opportunities for treating RAS rather than retransplantation alone. Lastly, preventive strategies focused on reducing the progression of RAS after primary transplantation may improve long-term outcomes. Our study has limitations inherent to its retrospective, registry-based design. Classification of restrictive allograft syndrome relied on registry coding and may be subject to temporal misclassification, particularly in the earlier era. The UNOS database also lacks granular data on immunologic, histopathologic, and peri-operative physiological variables, limiting assessment of biological mechanisms underlying graft failure. Although the database provides cause of death for recipients, it does not provide data on whether these patients redeveloped RAS after their second transplant. We also could not account for center-level variation in retransplant operative experience and perioperative management. Changes in allocation system and listing practices over time also may have influenced cohort composition. In addition, the focus on transplanted patients precludes assessment of waitlist mortality or delisting, which may further inform access disparities. Our study focuses only on RAS patients without addressing other indications for retransplantation, including BOS. Future studies incorporating detailed phenotypic comparisons are needed to further clarify whether these findings are specific to RAS. Lastly, our small sample of RAS patients in both eras limits statistical power. Over the past 2 decades, lung retransplantation for restrictive lung allograft dysfunction has been characterized by a progressive escalation in surgical strategies and expanded peri-operative support. However, long-term survival has remained unchanged. Our findings suggest that the limitations of retransplantation for RAS are rooted in disease biology. Future progress will depend on earlier risk stratification, more selective candidate assessment, and biologically targeted strategies rather than further escalation of surgery. Conflict of Interest Statement The authors reported no conflicts of interest. The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest. Footnotes A.A.R. is supported by grants from the National Institutes of Health for advanced heart disease research (T32HL116273). Appendix E1 Table E1Multivariable Cox regression model for survival of all retransplants from 2005 to 2025VariableHazard ratio (95% CI)P valueEra Era 1: 2005-2015Reference Era 2: 2016-20250.85 (0.73-0.99).04Donor Age, y1.00 (0.99-1.01).98 Po2/Fio2, mm Hg1.00 (1.00-1.00).999 Cause of death AnoxiaReference CNS tumor0.82 (0.38-1.77).62 Head trauma1.16 (0.96-1.39).13 Stroke1.27 (1.05-1.53).01 Unknown/other1.18 (0.66-2.10).58Recipient Age, y1.01 (1.01-1.02)<.001 Male sex0.99 (0.86-1.14).88 Allocation score at transplant1.01 (1.00-1.01).006 Total bilirubin at transplant, mg/dL1.10 (1.05-1.15)<.001 ECMO support at transplant1.16 (0.86-1.56).33 Ventilation at transplant1.12 (0.91-1.37).29 Days on waitlist1.00 (1.00-1.00).61 Ischemia time, h0.99 (0.96-1.02).70Bolded values indicate statistical significance. Po2, Partial pressure of oxygen; Fio2, fraction of inspired oxygen; CNS, central nervous system; ECMO, extracorporeal membrane oxygenation. Multivariable Cox regression model for survival of all retransplants from 2005 to 2025 Bolded values indicate statistical significance. Po2, Partial pressure of oxygen; Fio2, fraction of inspired oxygen; CNS, central nervous system; ECMO, extracorporeal membrane oxygenation. References