Duration of extracorporeal membrane oxygenation support and survival outcomes: a retrospective analysis for lung transplant recipients with idiopathic pulmonary fibrosis

Introduction

Background

Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive, fibrotic interstitial lung disease characterized by irreversible deterioration. Patients with IPF eventually progress to respiratory failure and death, with a median survival of 2–5 years following diagnosis (1,2). Lung transplantation (LTx) is the only effective treatment option for these patients (3,4). In its advanced stage, IPF is often complicated by pulmonary hypertension (PH), leading to respiratory failure and hemodynamic instability. Due to these factors, extracorporeal life support (ECLS) is often necessary (5,6). With advances in technology and clinical experience, extracorporeal membrane oxygenation (ECMO) has been widely adopted for IPF and the perioperative period of LTx, replacing cardiopulmonary bypass (CPB) as a form of ECLS (7-10).

Rationale and knowledge gap

Although LTx prolongs survival and improves quality of life for IPF recipients, their overall survival (OS) remains significantly worse compared to recipients of other organ transplants, with a 5-year survival rate of approximately 54–61.2% (11-13). Previous studies have focused on the use of ECMO as bridge to LTx (14-16), its intraoperative application (17-21), and postoperative supportive therapies for primary graft dysfunction (PGD) (22-25). However, the relationship between the intraoperative and postoperative duration of ECMO and OS in LTx recipients remains unclear. This issue is particularly important for improving treatment outcomes and survival rates in patients with IPF, especially in the context of postoperative complications.

Objective

Given the anticipated differences in outcomes and treatment duration based on post-surgical indications for ECMO, we hypothesized that a prolonged duration of ECMO may lead to poorer OS. This study conducted a retrospective analysis to explore the association between ECMO support duration and OS in LTx recipients with IPF as the primary underlying disease. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-675/rc).

Methods

Study design and setting

We conducted a retrospective cohort study involving 284 LTx recipients with IPF at the Affiliated Wuxi People’s Hospital of Nanjing Medical University China, from January 2015 to December 2020, with follow-up until January 2022. The inclusion criteria were: (I) Age >18 years; (II) IPF patients; (III) comprehended and signed an informed consent form; (IV) LTx were conducted between 2015 and 2020. The exclusion criteria were: (I) retransplantation; (II) postoperative mortality within 72 h; (III) loss of follow up; (IV) relatively incomplete clinical and follow-up information (Figure 1).

Figure 1 CONSORT diagram. ECMO, extracorporeal membrane oxygenation; IPF, idiopathic pulmonary fibrosis; LTx, lung transplantation; VA, veno-arterial; VV, veno-venous.

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Medical Ethics Committee of The Affiliated Wuxi People’s Hospital of Nanjing Medical University (No. KY21061). Due to the retrospective nature of the study, informed consent was waived. The donation and transplantation of all organs adhered to the ethical principles outlined in the Declaration of Istanbul.

Clinicopathological data collection and definition

Perioperative variables for both donors and recipients were extracted from the electronic medical record system. Variables with a missing rate greater than 10% were excluded from the analysis. A total of 24 variables associated with OS in LTx recipients were selected based on existing evidence and literature. These included (I) recipient factors:sex, age, body mass index (BMI), tobacco use, hypertension, diabetes, central venous pressure (CVP), cardiac index (CI), N-terminal brain natriuretic peptide (NT-proBNP), forced vital capacity (FVC), lung allocation score (LAS) (26), mechanical ventilation (MV); (II) donor factors: sex, cold ischemia time (CIT); (III) intraoperative factors: surgical type, ECMO strategy, operation time, duration of ECMO time, blood transfusion volume; (IV) postoperative factors: intensive care unit (ICU) length of stay, mechanical ventilation time (MVT), PGD, acute kidney injury (AKI), survival time.

BMI was calculated by dividing weight in kilograms by the square of height in meters (kg/m²). Patients were categorized as underweight (BMI <18.5 kg/m²), normal weight (BMI 18.5–24.9 kg/m²), overweight (BMI 25.0–29.9 kg/m²), or obese (BMI ≥30.0 kg/m²) (27).

AKI was defined according to the Kidney Disease: Improving Global Outcomes (KDIGO) clinical practice guideline (28). Specifically, AKI was diagnosed within 7 days post-transplant based on either an increase in serum creatinine (SCr) of ≥0.3 mg/dL (26.5 µmol/L) within 48 hours, or a ≥1.5-fold rise from baseline. SCr measurements were obtained at eight time points: T0 (within 24 hours before LTx) and T1–T7 (daily after LTx).

PGD was defined and graded based on the 2016 consensus criteria of the International Society for Heart and Lung Transplantation (ISHLT) (29). Grade 3 PGD was diagnosed by the presence of diffuse pulmonary infiltrates on chest radiographs and a PaO₂/FiO₂ ratio <200 mmHg within 48 to 72 hours post-transplant. Radiographs were interpreted by two independent transplant surgeons blinded to clinical data. Patients placed on ECMO after transplantation and exhibiting radiographic infiltrates were also classified as having grade 3 PGD. In this study, all references to PGD refer to grade 3 PGD.

Variable selection was conducted carefully and is detailed in the Appendix 1.

ECMO support

The Affiliated Wuxi People’s Hospital of Nanjing Medical University adheres to the Guidelines for Extracorporeal Membrane Oxygenation Application in Perioperative Lung Transplantation (30) regarding intraoperative ECMO indications for lung transplant surgeries.

ECMO initiation

A multidisciplinary team, including anesthesiologists, pulmonologists, ECMO intensivists, and lung transplant surgeons, evaluates the necessity of ECMO. Intraoperative ECMO initiation is considered under the following conditions: (I) pulmonary artery systolic pressure (PAP) >50 mmHg on intraoperative echocardiography, or mean PAP >25 mmHg on right heart catheterization; (II) New York Heart Association (NYHA) cardiac function classification ≥ class III; (III) inability to achieve adequate oxygenation during single-lung ventilation despite other methods to improve ventilation/perfusion ratio; (IV) prophylactic intraoperative ECMO support for anticipated hemodynamic instability, hypercapnia, or hypoxemia during LTx; (V) difficulty weaning from CPB after LTx, necessitating veno-arterial (VA) ECMO assistance; (VI) severe PGD post-surgery unresponsive to conventional treatment, requiring veno-venous (VV) ECMO or VA-ECMO support; (VII) patients at high risk of PGD who received intraoperative ECMO, requiring continued ECMO support post-surgery to minimize ventilator-induced lung injury; (VIII) patients with primary PH prone to left heart dysfunction and PGD post-surgery, necessitating gradual reduction of VA-ECMO flow over 3–5 days to facilitate cardiac function recovery and alleviate pulmonary edema; (IX) acute rejection post-surgery with severe hypoxemia, where conventional respiratory support is insufficient; (X) early post-operative PGD requiring re-transplantation, for transitional support.

ECMO configuration and cannulation strategy

For patients presenting with isolated hypercapnia or hypoxemia and normal pulmonary artery pressure, VV-ECMO is the preferred configuration. In cases of moderate to severe PH or cardiac dysfunction, femoro-arterial VA-ECMO is directly selected. Should peripheral ECMO prove insufficient to maintain intraoperative oxygenation or circulation, central VA-ECMO is established intraoperatively, necessitating a clamshell incision. If a patient’s pre-operative support is VV-ECMO, but significant circulatory failure occurs during LTx, the ECMO mode can be switched to central VAV-ECMO, involving internal jugular vein and femoral vein drainage with ascending aorta perfusion. Post-operatively, this can be transitioned to peripheral VAV-ECMO (internal jugular vein and femoral vein drainage, femoral artery perfusion) or revert to VV-ECMO. For VA-ECMO, axillary artery-femoral vein cannulation is performed, with femoral vein cannulation achieved percutaneously via the Seldinger technique, while axillary artery cannulation involves surgical cut-down and end-to-side anastomosis. For VV-ECMO, internal jugular vein cannulation is performed and combined with femoral vein puncture to establish the circulatory circuit.

Hemodynamic and respiratory goals

During ECMO operation, patient oxygen saturation is maintained above 90%. A mixed gas, typically an air-oxygen 2:1 ratio at a 3 L/min flow, is utilized for membrane lung gas exchange instead of pure oxygen. Arterial blood gas is actively monitored, and mixed gas flow or ratio is adjusted to maintain blood oxygen and carbon dioxide partial pressures within normal ranges. To ensure adequate oxygen supply to vital organs, hemoglobin is maintained above 80 g/L and plasma colloid osmotic pressure at 15–20 mmHg. ECMO aims to reduce MV dependency, thereby allowing for lung-protective ventilation strategies post-ECMO to minimize barotrauma and volutrauma.

Coagulation management

ECMO circuits are heparin-coated, providing some anticoagulant effect; however, systemic heparinization is still required during insertion, typically with an initial dose of 100 U/kg. During ECMO circulation, activated clotting time (ACT) is maintained around 150 s, and activated partial thromboplastin time (APTT) at 50–70 s. Heparin dosage is adjusted based on the patient’s underlying conditions or surgical wound bleeding. Low-dose or no heparin anticoagulation may be adopted early post-operatively under high ECMO flow to reduce bleeding risk. Anticoagulation targets are ACT 130–150 s and APTT 50–60 s. For prolonged ECMO (>3 days), routine heparin sodium anticoagulation is recommended, maintaining ACT or APTT at 1.5 times baseline, with adjustments based on the patient’s specific bleeding and clotting status. In cases of suspected or confirmed heparin-induced thrombocytopenia, argatroban or bivalirudin may be considered for alternative anticoagulation.

ECMO weaning

ECMO decannulation is considered when the support level falls below 30% of total cardiopulmonary function. For VV-ECMO weaning, ECMO blood flow is gradually reduced to 2.5–3.0 L/min, followed by a progressive decrease in ECMO ventilation. Decannulation is considered if oxygenation remains satisfactory, with no carbon dioxide retention and significant imaging improvement. A temporary (6-hour) cessation of VV-ECMO gas delivery, demonstrating satisfactory patient oxygenation, aids in predicting successful weaning. For VA-ECMO weaning, in addition to lung function recovery, thorough assessment of cardiac function recovery is crucial. Indicators for cardiac recovery include hemodynamic stability achievable with low-dose vasoactive drugs, an intrinsic pulse pressure greater than 20 mmHg, and improved bedside ultrasound parameters such as cardiac stroke volume, ventricular size, aortic velocity time integral, and ejection fraction. Although definitive ultrasound criteria are not established, an aortic velocity time integral greater than 10 cm, left ventricular ejection fraction of 20–25%, and mitral lateral annulus systolic S-wave velocity greater than 6 cm/s may predict successful VA-ECMO weaning.

Post-decannulation management

Following removal of percutaneous venous catheters, direct effective compression for at least 30 minutes is applied, followed by a pressure dressing. For arterial catheters, whether placed percutaneously or via cut-down, removal should ideally be performed after surgical vascular repair or with a Proglide percutaneous vascular closure device. Post-decannulation, ultrasound assessment for lower limb thrombosis is recommended.

Statistical analysis

Continuous variables that were normally distributed were expressed as mean ± standard deviation (SD) and analyzed using the t-test. For continuous variables that did not follow a normal distribution, results were reported as median [interquartile range (IQR)] and examined using the Wilcoxon rank sum test. Categorical variables were presented as frequencies and percentages, with comparisons conducted using the Chi-squared test or Fisher’s exact test when the expected frequency was less than 5. Survival curves were estimated by Kaplan-Meier method, plotted with the 95% confidence intervals (CIs), and were compared between groups using log-rank test.

We established several time-dependent Cox proportional hazards regression models to analyze the association between OS and ECMO duration. The crude model (model 1) only included ECMO use time. Model 2 adjusted for donor factors based on model 1. In model 3, preoperative recipient factors were adjusted based on model 2. Model 4 further adjusted for intraoperative factors as variables based on model 3. Model 5 is regarded as the main model that included all the 13 variables. Additionally, we evaluated the possible non-linear trends between the duration of ECMO time and the mortality using restricted cubic splines with four knots at the 5th, 35th, 65th, 95^th percentiles of the distribution (31). The P value for non-linearity was calculated by testing the null hypothesis that the coefficient of the second spline is equal to 0. Then a two-piecewise linear regression model was constructed to calculate the turning point and with manually selected break points near local extrema of survival (32). Simultaneously conduct a stratified analysis of postoperative complications in patients. Data analysis was carried out using R version 4.3.1 and P<0.05 was considered significant for all other statistical tests.

Results

Clinical characteristics

During the study period, a total of 284 patients underwent LTx for IPF at the Affiliated Wuxi People’s Hospital of Nanjing Medical University. After applying the exclusion criteria, 8 patients were excluded: 4 for retransplantation, 2 for postoperative mortality within 72 hours, and 2 due to loss to follow-up. Consequently, a total of 276 recipients were included in the final analysis (Figure 1). Of the 276 included recipients, 120 died during the follow-up period (Death group), while 156 survived (Survival group). The baseline clinical and demographic characteristics of the study cohort are presented in Table 1.

Among a total of 276 IPF patients who underwent LTx, 231 (83.7%) were male, with a median age of 61 years (range: 54–66 years) at the time of transplantation. The distribution of patient weight status was as follows: 54 (19.6%) underweight, 135 (48.9%) normal weight, 67 (24.3%) overweight, and 20 (7.2%) obese. Among all 276 recipients, the median duration of ECMO support was 24 hours (range: 7.25–40.0 hours). A total of 80 patients (29.0%) received VA-ECMO, 149 (54.0%) received VV- ECMO, and 47 (17.0%) did not receive ECMO support. The median survival time for the entire cohort was 383 days (range: 34.3–625.5 days). Recipients in the mortality cohort had a significantly longer ECMO duration [28.5 (IQR: 15.8–64.0) vs. 22 (range: 5.0–29.0) hours] and a higher incidence of PGD and AKI (all P<0.05). Additional baseline characteristics are summarized in Table 1. A stratified analysis based on ECMO strategies is provided in Tables S1,S2.

Kaplan-Meier analysis by ECMO strategy

At 1 year post-LTx, the estimated survival rate was 74.6% (95% CI: 61.6–90.4%) among recipients without ECMO support, 71.3% (95% CI: 61.9–82.0%) among those receiving VV-ECMO, and 71.1% (95% CI: 59.0–85.7%) among those supported with VA-ECMO. Kaplan-Meier analysis revealed no statistically significant difference in cumulative survival between recipients with ECMO support and those without ECMO (P=0.22) (Figure 2A). After stratifying the ECMO group into VV-ECMO and VA-ECMO subgroups, no statistically significant difference in survival was observed (P=0.44) (Figure 2B).

Figure 2 Kaplan-Meier overall survival curves. (A) The overall survival in IPF recipients with ECMO compared with those without ECMO (P=0.22). (B) The overall survival in IPF recipients on VA-ECMO and VV-ECMO compared with those without ECMO (P=0.44). The shaded areas represent the 95% confidence bands. ECMO, extracorporeal membrane oxygenation; IPF, idiopathic pulmonary fibrosis; VA, veno-arterial; VV, veno-venous.

ECMO support duration associated with OS

In the crude model (Model 1), the mortality risk increased by 16% for each additional day of ECMO support [hazard ratio (HR) =1.16, 95% CI: 1.12–1.22, P<0.001]. The main model (Model 5) demonstrated a 13% increase in mortality risk for each additional day of ECMO support (HR =1.13, 95% CI: 1.07–1.20, P<0.001) (Table 2). After adjusting for all variables (Model 5), only the group without complications of AKI (HR =1.42, 95% CI: 0.98–2.05, P=0.47) and the group without PGD and AKI (HR =0.51, 95% CI: 0.08–3.19, P=0.06) lost statistical significance. Notably, among recipients without PGD, ECMO duration was strongly associated with an increased risk of mortality (HR =1.99, 95% CI: 1.10–3.61, P=0.02), reflecting a 90% higher mortality risk compared to those with PGD (HR =1.09, 95% CI: 1.02–1.16, P=0.01).

The continuous relationship between duration of ECMO time and the adjusted mortality risk revealed a nonlinear relationship (Figure 3A, P-overall ≤0.001; P-nonlinear =0.002). Across different ECMO strategy strata, the association was consistent in the VV-ECMO group (P-overall =0.004, P-non-linear =0.005), as shown in Figure 3B. However, no significant nonlinear association was observed in the VA-ECMO group (Figure 3C). Moreover, the two-piecewise linear regression model showed that the inflection points in the total cohort were identified at ECMO durations of 0.8 and 5.54 days (Figure 3D), while in the VV-ECMO cohort, the inflection points were identified at 1.12 and 4.67 days (Figure 3E).

Figure 3 The relationship between duration of ECMO time and HR. (A) Total; (B) VV; (C) VA. The two-piecewise linear regression between duration of ECMO time and HR. (D) Total with inflection point 0.8 and 5.54 days. (E) VV with inflection point duration 1.12 and 4.67 days. ECMO, extracorporeal membrane oxygenation; HR, hazard ratio; VA, veno-arterial; VV, veno-venous.

Therefore, based on the inflection points of ECMO usage duration estimated by the two-piecewise linear model, the total population and the VV-ECMO population were categorized accordingly. When the duration of ECMO usage is between 0.8 and 5.54 days, the adjusted main model (Model 5), which accounts for all covariates, indicates that for each additional day of ECMO use, the risk of mortality increases by 51%. Furthermore, when ECMO usage exceeds 5.54 days, each additional day of ECMO is associated with a 1.05-fold increase in the risk of death (Table S3).

Discussion

Key findings

This study showed that the type of ECMO used does not affect post-transplant OS in recipients, and recipients who did not receive ECMO did not demonstrate improved survival outcomes. Additionally, the mortality risk increased by approximately 13% for each additional day of ECMO use in recipients. After stratification, the mortality risk increased by 9% for recipients requiring VV-ECMO, 25% for those requiring VA-ECMO, 9% for recipients with PGD, and 99% for non-PGD recipients with each additional day of ECMO use. In non-PGD recipients, mortality risk significantly increases with prolonged ECMO use, a trend also observed, though to varying degrees, in other recipient subgroups. Restricted cubic spline analysis revealed a nonlinear relationship between ECMO duration and adjusted mortality risk. When ECMO duration was between the two inflection points (0.8 and 5.54 days), the HR increased rapidly. Conversely, when ECMO duration was <0.8 or >5.54 days, the risk decreased or plateaued.

Strengths and limitations

Compared to a previous study (21), this research further quantifies the relationship between ECMO duration and mortality risk, addressing a gap in the literature and providing robust data to support further research and the optimization of clinical practice guidelines.

This study not only concludes that prolonged ECMO use adversely affects recipient survival but also identifies precise critical values using a piecewise linear regression model, which have not been addressed in previous research (16,33,34). Additionally, we performed a detailed time analysis in days to preserve detailed information regarding temporal mortality changes, whereas earlier studies typically grouped duration into broad categories, potentially overlooking nuanced effects.

The limitations of this article include the following: first, this study is a retrospective cohort, single-center study design which may have sample selection bias and inherent limitations in the results. The absence of multicenter data may affect the reliability and external validity of the findings. Second, the relatively small sample size of ECMO-treated recipients may limit the interpretability and generalizability of the results. Larger-scale studies are necessary to gain a more comprehensive understanding of ECMO’s role in LTx. Due to sample size and data collection limitations, only 13 covariates were included, while other potentially influential factors were excluded, potentially confounding the assessment of survival in LTx recipients. Third, variables with severe missing data (≥10%) were screened out, and multiple chained imputation was applied, which may impact the accuracy of the final results. The use of multiple chained imputation may influence the accuracy and reliability of findings. Restricted by the sample size and data collection limitations, 13 covariates were included in this study. However, other potentially influential covariates were not considered, leading to confounding factors affecting the accurate assessment of survival in recipients. Although we documented the timing and methods of intraoperative and postoperative ECMO use, potential variations in ECMO application in response to clinical changes were not fully captured. This limitation may affect the accurate interpretation of ECMO’s effects in LTx. Fourth, this study focused exclusively on recipients with IPF as the primary diagnosis. This choice was made due to IPF’s high prevalence and significant impact on survival, as it is a chronic, progressive fibrosing interstitial pneumonia leading to severe lung function decline and respiratory failure, with median survival of 2–5 years after diagnosis (2). Moreover, the rising prevalence of IPF and the fact that LTx is currently the only effective intervention to prolong survival emphasize the relevance of this focus (1,3). Additionally, we were unable to perform a secondary analysis of waitlist survival stratified by ECMO duration or support modality. This limitation was primarily due to the retrospective study design and incomplete pre-transplant follow-up data for some patients, which made it difficult to accurately assess outcomes prior to transplantation. However, this exclusivity may limit the generalizability of our findings to patients with other underlying lung diseases.

Comparison with similar research and explanations of findings

The increasing adoption of ECLS, particularly ECMO, in LTx underscores the imperative for standardized approaches to optimize patient outcomes. This pressing need is reflected and addressed in recent expert consensus documents concerning perioperative ECLS management. The American Association for Thoracic Surgery (AATS) 2022 Expert Consensus Document (35) specifically highlights the critical role of mechanical circulatory support throughout the LTx journey, yet concurrently acknowledges significant practice variability and identifies key knowledge gaps regarding optimal utilization strategies and their impact on long-term survival. Building upon this foundational guidance, the ISHLT has provided detailed, phase-specific recommendations. Their consensus statement on intraoperative ECLS use offers critical insights into indications, configurations, and management during the transplant procedure itself (36), while a subsequent statement addresses the complexities of postoperative ECLS, including crucial aspects like weaning strategies and complication management (37). In light of these consensus frameworks, our current findings directly address several evidence gaps. While the AATS and ISHLT consensus documents provide essential principles for when and how to employ ECLS, they simultaneously emphasize the need for more granular data on the temporal aspects of support and their prognostic implications. This study uniquely quantifies the nonlinear relationship between ECMO duration and mortality risk in IPF LTx recipients, identifying clinically actionable thresholds at 0.8 and 5.54 days. The significantly elevated mortality risk observed per additional day of support, particularly pronounced in non-PGD patients (99% increase) and those on VA-ECMO (25% increase), provides compelling empirical evidence supporting the consensus emphasis on minimizing unnecessary ECLS duration. Furthermore, our stratification by support strategy (VV vs. VA) and complication status (PGD, AKI) offers nuanced insights that can refine the risk assessment and weaning protocols advocated in the ISHLT postoperative guidance. Consequently, these results contribute robust, disease-specific data to inform the practical implementation of the broad principles outlined in the AATS and ISHLT consensus statements, thereby facilitating more personalized ECLS management in IPF-LTx.

Currently, limited research has examined the duration of intraoperative and postoperative ECMO. One retrospective analysis (33), reported that recipients supported with ECMO for more than 48 hours had a 3.21-fold higher mortality risk compared to those supported for less than 10 hours. In contrast, when ECMO is used as a bridge to LTx, the median duration typically ranges from 7 to 12 days or longer (14-16), whereas intraoperative and postoperative ECMO support tends to be shorter, lasting from several hours to around 10 days (18,20,21). This reflects the complexity and variability of postoperative clinical trajectories in transplant recipients. Moreover, most prior studies have concentrated on the relationship between ECMO duration as a bridge to LTx and postoperative survival or bridging success (16,33,34). However, Myles Smith and colleagues (38), in a study not focused on post-lung transplant patients, analyzed the relationship between the duration of VA-ECMO and survival rates. They observed that survival initially improved with increasing VA-ECMO duration, followed by a sharp decline, and eventually plateaued. Peak survival was noted at 4 and 12 days, with variability in these thresholds likely due to differences in sample sizes and patient populations. The study attributed early post-ECMO mortality to treatment failure, which may also explain the declining hazard ratio observed in the early phase of our analysis.

Our stratified analysis revealed that each additional day on ECMO increased the mortality risk by 9% in patients with grade 3 PGD. Surprisingly, in patients without grade 3 PGD, the mortality risk associated with prolonged ECMO use nearly doubled. Theoretically, grade 3 PGD reflects more severe postoperative pathology; however, our findings showed a lower hazard ratio in these patients, which is counterintuitive and warrants further exploration. We propose several possible reasons for this observation. First, a report has shown that timely recognition of PGD and initiation of ECMO therapy after LTx can significantly improve survival rates, which suggests that although patients with grade 3 PGD may have more severe conditions (22), the application of ECMO may provide critical cardiopulmonary support. Additionally, aggressive medical interventions in these patients may improve survival, as supported by prior research indicating that preemptive ECMO use leads to better outcomes (21). Therefore, timely ECMO application combined with appropriate interventions may attenuate the adverse survival impact of PGD. Second, the inconsistencies in our findings may stem from a limited sample size and short follow-up duration, which can affect statistical power and result stability. Finally, our dataset captured only ECMO duration during and after surgery, without clearly defining its temporal relationship with PGD onset, potentially introducing residual confounding. However, it is crucial to interpret this association cautiously. The requirement for prolonged ECMO support likely serves as a surrogate marker for underlying patient severity, including the complexity of the surgical procedure, the presence and severity of PGD or other significant postoperative complications, and potentially pre-existing comorbidities that predispose to a more complicated recovery. These factors, rather than the ECMO duration, are the probable primary drivers of the increased mortality observed. While this limitation presents opportunities for future investigation, it also underscores the potential harms of prolonged ECMO use. This further underscores the importance of enhancing multi-system function monitoring and comprehensive management during ECMO use in the postoperative period of LTx. In conclusion, future research should elucidate the intricate interplay between ECMO and PGD to improve our understanding of ECMO’s effects on outcomes, thereby informing personalized treatment approaches.

In the Affiliated Wuxi People’s Hospital of Nanjing Medical University a majority of (83%) recipients received ECMO support in intraoperative and postoperative period The Kaplan-Meier curve results demonstrate that the survival rate of recipients receiving ECMO support does not differ significantly from those who did not, which aligns with most of the existing literature (17-20). However, this contrasts with the finding of Hoetzenecker et al. (21), who reported improved survival outcomes in recipients who received ECMO support during LTx. Similarly, no significant association was found between different ECMO strategies and overall mortality rates. However, patients receiving ECMO treatment exhibited higher CVP, reduced cardiac function, longer surgical durations, increased transfusion requirements, and a higher incidence of complications—findings that were more pronounced in the VA-ECMO group. This suggests that VA-ECMO patients often present with more complex clinical conditions. Other studies have similarly shown that recipients requiring ECMO support tend to have greater transfusion needs and are more prone to developing PGD compared to those not receiving ECMO (20,39). Although limited, existing report suggests a growing need for VA-ECMO as intraoperative support, particularly in recipients with right ventricular dysfunction and/or inadequate systemic perfusion, likely due to the increasing prevalence of IPF and PH (6). Our study demonstrates that, despite differences in the indications for VA and VV-ECMO, the use of VA ECMO can lead similar OS outcomes compared to VV ECMO. A previous study have indicated that VA-ECMO can maintain systemic blood pressure by directly pumping blood into the systemic circulation and provides a protective effect for graft re-perfusion by bypassing the pulmonary circulation (40), which indicates the importance of selecting an appropriate ECMO strategy based on strict indications to ensure recipients survival during LTx. Upon a detailed review of prior research cohorts, we found that the ECMO group consisted of recipients with a variety of underlying diseases, and the grouping was solely based on ECMO usage without correcting for baseline characteristic differences between these groups. In contrast, our study exclusively included recipients with IPF and applied a more rigorous grouping process that considered both ECMO usage and the specific type of ECMO strategy employed. This methodological approach enhances the novelty of our study and facilitates a more accurate evaluation of ECMO-related outcomes in the context of LTx.

Implications and actions needed

This study highlights the clinical utility of ECMO duration as a stratification tool for mortality risk and prognostic assessment in IPF-LTx recipients (41). The identification of inflection points at 0.8 and 5.54 days of ECMO use in this study provides clinically meaningful thresholds that can aid clinicians in risk stratification, prognostic assessment, and potentially optimize the timing of supportive treatment interventions and related discussions. While meeting patients’ life support needs, it is essential to actively explore more appropriate timing and strategies for ECMO weaning to avoid unnecessary prolonged use. In cases where extended ECMO use is unavoidable, proactive assessment of potential adverse outcomes for patients is warranted. Previous research has highlighted the importance of careful patient selection and the prevention of treatment-related complications in reducing early mortality rates (42). However, for LTx recipients, the postoperative course is complex, and it must be acknowledged that rescue interventions inherently carry a certain failure rate. Therefore, although a longer ECMO duration is associated with increased mortality, this more likely reflects the severity of the patient’s underlying condition and associated medical complexities, rather than being a direct cause of death. Future prospective studies are needed to validate these thresholds and clarify the complex interplay between ECMO duration, patient factors, and prognostic outcomes.

While prolonged ECMO duration serves as a robust marker of disease severity and complication burden, clinicians should prioritize mitigating underlying drivers of ECMO dependence rather than focusing solely on duration reduction. Prospective validation of these thresholds is essential to develop evidence-based protocols integrating ECMO duration with multiparameter risk assessment. Multicenter studies should further explore its synergistic value with biomarkers to optimize post-transplant management.

Conclusions

For IPF patients undergoing LTx, ECMO duration exhibits a strong nonlinear association with mortality. Prolonged support serves as a robust indicator of underlying disease severity and postoperative complications portending adverse outcomes. The identified inflection points (0.8 and 5.54 days) offer clinically actionable thresholds for mortality risk stratification and prognostic assessment. Prospective validation is required to confirm these temporal markers and elucidate relationships between ECMO duration, patient factors, and clinical endpoints.

Acknowledgments

We would like to thank Blessing Mugwambi for his help in polishing our paper.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-675/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-675/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-675/prf

Funding: This work was supported by Noncommunicable Chronic Diseases-National Science and Technology Major Project (No. 2024ZD0528600), Shaanxi Province Key Research and Development Program (No. 2024SF-YBXM-083), Shaanxi Province Postdoctoral Science Foundation (No. 31271000000012), National Natural Science Foundation of China (No. 82470107), Jiangsu Provincial Science and Technology Planning Major Research and Development Project (No. BE2022697), and Cohort and Clinical Research Program of Wuxi Medical Center, Nanjing Medical University (No. WMCC202301).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-675/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Medical Ethics Committee of The Affiliated Wuxi People’s Hospital of Nanjing Medical University (No. KY21061). Due to the retrospective nature of the study, informed consent was waived. The donation and transplantation of all organs adhered to the ethical principles outlined in the Declaration of Istanbul.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Raghu G, Chen SY, Yeh WS, et al. Idiopathic pulmonary fibrosis in US Medicare beneficiaries aged 65 years and older: incidence, prevalence, and survival, 2001-11. Lancet Respir Med 2014;2:566-72. [Crossref] [PubMed]
2. Raghu G, Collard HR, Egan JJ, et al. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med 2011;183:788-824. [Crossref] [PubMed]
3. Lederer DJ, Martinez FJ. Idiopathic Pulmonary Fibrosis. N Engl J Med 2018;378:1811-23. [Crossref] [PubMed]
4. Shenderov K, Collins SL, Powell JD, et al. Immune dysregulation as a driver of idiopathic pulmonary fibrosis. J Clin Invest 2021;131:e143226. [Crossref] [PubMed]
5. Rajagopal K, Bryant AJ, Sahay S, et al. Idiopathic pulmonary fibrosis and pulmonary hypertension: Heracles meets the Hydra. Br J Pharmacol 2021;178:172-86. [Crossref] [PubMed]
6. Bartolome S, Hoeper MM, Klepetko W. Advanced pulmonary arterial hypertension: mechanical support and lung transplantation. Eur Respir Rev 2017;26:170089. [Crossref] [PubMed]
7. Biscotti M, Yang J, Sonett J, et al. Comparison of extracorporeal membrane oxygenation versus cardiopulmonary bypass for lung transplantation. J Thorac Cardiovasc Surg 2014;148:2410-5. [Crossref] [PubMed]
8. Bermudez CA, Shiose A, Esper SA, et al. Outcomes of intraoperative venoarterial extracorporeal membrane oxygenation versus cardiopulmonary bypass during lung transplantation. Ann Thorac Surg 2014;98:1936-42; discussion 1942-3. [Crossref] [PubMed]
9. Machuca TN, Collaud S, Mercier O, et al. Outcomes of intraoperative extracorporeal membrane oxygenation versus cardiopulmonary bypass for lung transplantation. J Thorac Cardiovasc Surg 2015;149:1152-7. [Crossref] [PubMed]
10. McFadden PM, Greene CL. The evolution of intraoperative support in lung transplantation: Cardiopulmonary bypass to extracorporeal membrane oxygenation. J Thorac Cardiovasc Surg 2015;149:1158-60. [Crossref] [PubMed]
11. Valapour M, Lehr CJ, Schladt DP, et al. OPTN/SRTR 2021 Annual Data Report: Lung. Am J Transplant 2023;23:S379-442. [Crossref] [PubMed]
12. Valapour M, Lehr CJ, Skeans MA, et al. OPTN/SRTR 2020 Annual Data Report: Lung. Am J Transplant 2022;22:438-518. [Crossref] [PubMed]
13. Valapour M, Lehr CJ, Skeans MA, et al. OPTN/SRTR 2019 Annual Data Report: Lung. Am J Transplant 2021;21:441-520. [Crossref] [PubMed]
14. Xia Y, Ragalie W, Yang EH, et al. Venoarterial Versus Venovenous Extracorporeal Membrane Oxygenation As Bridge to Lung Transplantation. Ann Thorac Surg 2022;114:2080-6. [Crossref] [PubMed]
15. Hashimoto K, Hoetzenecker K, Yeung JC, et al. Intraoperative extracorporeal support during lung transplantation in patients bridged with venovenous extracorporeal membrane oxygenation. J Heart Lung Transplant 2018;37:1418-24. [Crossref] [PubMed]
16. Shou BL, Kalra A, Zhou AL, et al. Impact of Extracorporeal Membrane Oxygenation Bridging Duration on Lung Transplant Outcomes. Ann Thorac Surg 2024;118:496-503. [Crossref] [PubMed]
17. Zhang R, Xu Y, Sang L, et al. Factors associated with intraoperative extracorporeal membrane oxygenation support during lung transplantation. Respir Res 2020;21:85. [Crossref] [PubMed]
18. Scaravilli V, Morlacchi LC, Merrino A, et al. Intraoperative extracorporeal membrane oxygenation for lung transplantation in cystic fibrosis patients: Predictors and impact on outcome. J Cyst Fibros 2020;19:659-65. [Crossref] [PubMed]
19. Ius F, Sommer W, Tudorache I, et al. Five-year experience with intraoperative extracorporeal membrane oxygenation in lung transplantation: Indications and midterm results. J Heart Lung Transplant 2016;35:49-58. [Crossref] [PubMed]
20. Ius F, Aburahma K, Boethig D, et al. Long-term outcomes after intraoperative extracorporeal membrane oxygenation during lung transplantation. J Heart Lung Transplant 2020;39:915-25. [Crossref] [PubMed]
21. Hoetzenecker K, Schwarz S, Muckenhuber M, et al. Intraoperative extracorporeal membrane oxygenation and the possibility of postoperative prolongation improve survival in bilateral lung transplantation. J Thorac Cardiovasc Surg 2018;155:2193-2206.e3. [Crossref] [PubMed]
22. Van Raemdonck D, Hartwig MG, Hertz MI, et al. Report of the ISHLT Working Group on primary lung graft dysfunction Part IV: Prevention and treatment: A 2016 Consensus Group statement of the International Society for Heart and Lung Transplantation. J Heart Lung Transplant 2017;36:1121-36. [Crossref] [PubMed]
23. Takahashi T, Terada Y, Pasque MK, et al. Outcomes of Extracorporeal Membrane Oxygenation for Primary Graft Dysfunction After Lung Transplantation. Ann Thorac Surg 2023;115:1273-80. [Crossref] [PubMed]
24. Chen Q, Catarino P. Venoarterial or Venovenous Extracorporeal Membrane Oxygenation for Severe Primary Graft Dysfunction After Lung Transplant? Ann Thorac Surg 2023;115:1280-1. [Crossref] [PubMed]
25. Hunt ML, Cantu E. Primary graft dysfunction after lung transplantation. Curr Opin Organ Transplant 2023;28:180-6. [Crossref] [PubMed]
26. Egan TM, Murray S, Bustami RT, et al. Development of the new lung allocation system in the United States. Am J Transplant 2006;6:1212-27. [Crossref] [PubMed]
27. Rubino F, Cummings DE, Eckel RH, et al. Definition and diagnostic criteria of clinical obesity. Lancet Diabetes Endocrinol 2025;13:221-62. [Crossref] [PubMed]
28. Kellum JA, Lameire N, Aspelin P, et al. Kidney disease: Improving global outcomes (KDIGO) acute kidney injury work group. KDIGO clinical practice guideline for acute kidney injury. Kidney International Supplements 2012;2:1-138.
29. Snell GI, Yusen RD, Weill D, et al. Report of the ISHLT Working Group on Primary Lung Graft Dysfunction, part I: Definition and grading-A 2016 Consensus Group statement of the International Society for Heart and Lung Transplantation. J Heart Lung Transplant 2017;36:1097-103. [Crossref] [PubMed]
30. Branch of Organ Transplantation of Chinese Medical Association, National Quality Management and Control Center for Lung Transplantation. Guideline on the application of extracorporeal membrane oxygenation during the perioperative period of lung transplantation (2019 edition). Organ Transplantation 2019;10:402-9.
31. Cleophas TJ, Zwinderman AH. Spline Regression Modeling. In: Cleophas TJ, Zwinderman AH. editors. Regression Analysis in Medical Research: For Starters and 2nd Levelers. Springer International Publishing; 2018:267-77.
32. Muggeo VMR. Estimating regression models with unknown break-points. Statistics in Medicine 2003;22:3055-71. [Crossref] [PubMed]
33. Minqiang L, Xiaoshan L, Bo X, et al. A Retrospective Analysis for Risk Factors and Early Prognosis of Delayed Withdrawal Extracorporeal Membrane Oxygenation After Lung Transplantation. Transplantation 2021;105:867-75. [Crossref] [PubMed]
34. Camboni D, Philipp A, Lubnow M, et al. Support time-dependent outcome analysis for veno-venous extracorporeal membrane oxygenation. Eur J Cardiothorac Surg 2011;40:1341-6; discussion 1346-7. [Crossref] [PubMed]
35. Expert Consensus Panel. The American Association for Thoracic Surgery (AATS) 2022 Expert Consensus Document: The use of mechanical circulatory support in lung transplantation. J Thorac Cardiovasc Surg 2023;165:301-26. [Crossref] [PubMed]
36. Martin AK, Mercier O, Fritz AV, et al. ISHLT consensus statement on the perioperative use of ECLS in lung transplantation: Part II: Intraoperative considerations. J Heart Lung Transplant 2024; Epub ahead of print. [Crossref]
37. Martin AK, Mercier O, Bottiger B, et al. ISHLT consensus statement on the perioperative use of ECLS in lung transplantation: Part III: Postoperative considerations. J Heart Lung Transplant. 2025; Epub ahead of print. [Crossref]
38. Smith M, Vukomanovic A, Brodie D, et al. Duration of veno-arterial extracorporeal life support (VA ECMO) and outcome: an analysis of the Extracorporeal Life Support Organization (ELSO) registry. Crit Care 2017;21:45. [Crossref] [PubMed]
39. Loor G, Huddleston S, Hartwig M, et al. Effect of mode of intraoperative support on primary graft dysfunction after lung transplant. J Thorac Cardiovasc Surg 2022;164:1351-1361.e4. [Crossref] [PubMed]
40. Calabrese F, Pezzuto F, Fortarezza F, et al. Evaluation of Tissue Ischemia/Reperfusion Injury in Lung Recipients Supported by Intraoperative Extracorporeal Membrane Oxygenation: A Single-Center Pilot Study. Cells 2022;11:3681. [Crossref] [PubMed]
41. Wilkey BJ, Elliott T, Jones TE, et al. Survey of Extracorporeal Life Support Application in Lung Transplantation. Clin Transplant 2025;39:e70094. [Crossref] [PubMed]
42. Quintel M, Gattinoni L, Weber-Carstens S. The German ECMO inflation: when things other than health and care begin to rule medicine. Intensive Care Med 2016;42:1264-6. [Crossref] [PubMed]