INTRODUCTION
Living-donor liver transplantation (LDLT) offers a significant survival advantage to patients with end-stage liver disease undergoing early transplantation, and has emerged as a viable alternative to deceased-donor liver transplantation [1]. However, despite advances in surgical techniques and perioperative management, the 5-year mortality rate for LDLT has remained high (13.0%), which is attributable to the high incidence of postoperative complications [2]. Enhanced recovery after surgery (ERAS) protocols have revolutionized perioperative management across various surgical specialties, including liver transplantation (LT), by integrating multimodal strategies, such as minimally invasive techniques, optimal pain control, and early mobilization, to accelerate recovery and minimize postoperative complications [3,4]. Central to the ERAS protocols in perioperative care is the concept of early extubation, which is defined as immediate tracheal extubation in the operating room (OR) or within 1 h of LT completion [5,6].
Recent studies have demonstrated that early extubation is associated with lower rates of pulmonary complications and shorter lengths of stay in the intensive care unit (ICU) and hospital. However, the long-term effect of immediate extubation in the OR on overall survival (OS) in patients undergoing LDLT remains unclear [7-10]. Addressing this knowledge gap is crucial because understanding the long-term benefits and risks of immediate extubation can refine perioperative care practices and improve patient outcomes. Therefore, this study aimed to investigate the association between immediate extubation in the OR and long-term OS of patients undergoing LDLT. Specifically, we aimed to determine whether immediate extubation is related to improved postoperative outcomes such as reduced complications and increased survival rates.
We hypothesized that immediate extubation is associated with improved postoperative outcomes and that its benefits can improve future perioperative care protocols for patients undergoing LDLT.
MATERIALS AND METHODS
Patients
This retrospective cohort study was approved by the Institutional Review Board of the Pusan National University Yangsan Hospital (approval no. 55-2024-005), which waived the requirement for informed consent. This study was performed in accordance with the Declaration of Helsinki (revised 2013) and the Strengthening the Reporting of Observational Studies in Epidemiology guidelines [11].
We reviewed electronic medical and anesthesia records to collect information on patients aged ≥ 18 years who underwent LDLT for end-stage liver failure between September 2010 and March 2021. Patients were excluded if they met any of the following criteria: a model for end-stage liver disease (MELD, a predictive score that measures the severity of liver failure in patients with chronic liver disease using international normalized ratio [INR], serum total bilirubin, and serum creatinine, and evaluates the short-term survival chance during the next three months) score ≥ 26, preoperative intubation, hepatic encephalopathy of grade III or higher at the time of surgery, a latest intraoperative arterial blood gas analysis showing an arterial partial pressure of oxygen (PaO2) to inspired oxygen fraction (FiO2) ratio of < 200 mmHg, preoperative pulmonary edema diagnosed with chest X-ray examination or chest computed tomography, or a preoperative diagnosis of hepatopulmonary syndrome.
Hepatic encephalopathy is categorized according to the West Haven criteria, ranging from grade I to IV [12]. Grade Ⅰ includes minor changes in awareness, euphoria or anxiety, shortened attention span, and difficulty with simple arithmetic tasks. Grade Ⅱ includes lethargy or apathy, time disorientation, evident personality changes, and inappropriate behavior. Grade Ⅲ includes somnolence to semi-stupor, responsiveness to stimuli, confusion, gross disorientation, and bizarre behavior. Grade Ⅳ encompasses coma and an inability to assess the mental state.
Patients were classified into two groups based on the location of extubation. Those extubated immediately in the OR were categorized as OREX, and those extubated in the ICU were categorized as NOREX.
Outcome measurements
The primary outcome of this study was the OS of patients who underwent LDLT. OS was defined as the period from the date of surgery to the date of death from any cause or to the date of the last follow-up visit. Secondary outcomes included the length of stay in the ICU and hospital, duration of mechanical ventilation, incidence of reintubation, and occurrence of composite complications within 30 days post-surgery. The duration of mechanical ventilation was defined as the time spent on mechanical ventilation in the ICU, starting immediately after the operation. A 30-day composite complication was defined according to the European Perioperative Clinical Outcome definitions and included one or more of the following, with multiple events counted only once [13]: all-cause death, bleeding control measures (surgery for bleeding control or vascular embolization due to acute postoperative bleeding), vascular complications (stenosis or thrombosis in the hepatic artery or portal vein, pseudoaneurysm in the inferior vena cava or hepatic artery, arteriovenous fistula, or celiac stenosis), biliary complications (stricture, bile leakage, obstruction, or infection), graft rejection (acute rejection, chronic rejection, or graft-versus-host disease), major adverse cardio-cerebrovascular events (acute myocardial infarction, acute coronary syndrome, heart failure, and ischemic or hemorrhagic stroke), respiratory complications (postoperative mechanical ventilation for more than 48 hours or reintubation for respiratory failure, pneumonia, or acute respiratory distress syndrome), and renal complications (a Kidney Disease Improving Global Outcomes stage ≥ 2 or a need for renal replacement therapy). Moreover, we examined the preoperative and intraoperative prognostic factors influencing the long-term survival of LDLT recipients.
Anesthetic technique
General anesthesia was administered in accordance with our hospital’s protocol for LDLT. Induction involved the use of intravenous anesthetics (propofol, thiopental, or etomidate) and neuromuscular blocking agents (rocuronium or cisatracurium) supplemented with an opioid (remifentanil). Anesthesia was maintained using inhaled anesthetic gases (desflurane or sevoflurane) or total intravenous anesthesia with propofol and remifentanil. Mechanical ventilation was regulated to maintain a tidal volume of 8 to 10 ml/kg and a respiratory rate of 10 to 14 breaths/min, using a 40-50% oxygen/air mixture. Monitoring included measuring the blood pressure in both the femoral and radial arteries, whereas hemodynamic variables were determined using a pulmonary artery catheter. At the end of the surgery, the decision to perform a tracheal extubation in the OR was jointly made by the attending anesthesiologist and the transplant surgeon, based on standardized extubation criteria. The criteria encompassed the following: the patient being fully awake, complete reversal from neuromuscular blockade, spontaneous breathing with a tidal volume ≥ 5 ml/kg, a respiratory rate 30 breaths/min, an SpO2 ≥ 95% with an FiO2 0.5, normothermia (≥ 35.5°C), stable vital signs, a satisfactory metabolic status (pH > 7.25), and an absence of ongoing surgical bleeding [5,14-16]. However, even if a patient’s condition was suitable for extubation, certain patients were not extubated in the OR at the surgeon’s discretion due to surgeon fatigue or understaffed ICUs. Post-surgery, all patients were transferred to the ICU. The ICU clinician determined the timing of weaning from mechanical ventilation for patients who were not extubated in the OR.
Statistical analyses
Continuous variables are expressed as mean ± standard deviation or median with interquartile range, whereas categorical variables are expressed as numbers and percentages. Intergroup differences were evaluated using Student’s t-test or the Wilcoxon rank-sum test for continuous variables and the chi-squared test or Fisher’s exact test for categorical variables, as appropriate. Logistic regression analysis was performed to evaluate the association between extubation in the OR and postoperative outcomes. In addition, propensity score matching was performed to reduce selection bias and control for confounding factors, including intraoperative factors with significantly different baseline characteristics between the two groups. Propensity scores were generated using logistic regression models, with the nearest-neighbor method applied for matching. Imbalances after matching were assessed using a standardized mean difference threshold of 0.2. Covariates for matching included the recipient’s sex; age; body mass index (BMI); etiology; Child-Pugh score; MELD score; MELD-Na (MELD incorporated with serum sodium); hepatic encephalopathy grade; ascites; INR; bilirubin, albumin, creatinine, and sodium levels; serum platelet (PLT) counts; acute normovolemic hemodilution; intraoperative continuous renal replacement therapy (CRRT); packed red blood cell (pRBC) transfusion; fresh frozen plasma (FFP) transfusion; PLT concentrate units; estimated blood loss; last intraoperative PaO2/FiO2 ratio; duration of anesthesia administration; and duration of surgery. These covariates were selected based on their potential impact on the outcomes of interest, as they are known to influence both the severity of liver disease and postoperative complications. By accounting for these variables, we aimed to achieve a more accurate comparison between the two groups.
Univariate and multivariate Cox proportional hazards models were used to identify the potential prognostic factors associated with OS. The selection criteria for the final multivariate regression model were based on biological plausibility, clinical significance, and statistical significance (P < 0.20).
Survival probability was estimated using the Kaplan-Meier method, and a log-rank test was used to compare the cumulative 5-year OS between the OREX and NOREX groups. A two-tailed P value < 0.05 was considered statistically significant. Data manipulations and statistical analyses were performed using the R statistical language (version 4.2.1; R Core Team, 2022) on Windows 10 x64-bit (build 17763).
RESULTS
Among 277 patients, 72 were excluded based on the previously stated exclusion criteria; finally, 205 patients were included in this study (Fig. 1). Among them, 98 (47.8%) were extubated in the OR, and 107 (52.2%) in the ICU. The baseline and preoperative patient characteristics are presented in Table 1. The follow-up period was 74.9 (40.6-102.4) months. Table 2 compares intraoperative characteristics between the two groups. The OREX group required fewer intraoperative transfusions of pRBC, FFP, and PLT than the NOREX group. In addition, the percentage of patients that experienced an estimated blood loss greater than 2,000 ml was lower in the OREX group than in the NOREX group (36.7% vs. 53.3%, P = 0.025).
The OREX group had shorter hospital and ICU stays and fewer 30-day composite and biliary complications than the NOREX group (Table 3). However, these results may have been influenced by various intraoperative factors. Therefore, a logistic regression analysis was performed for each complication, adjusted for age, sex, BMI, hepatorenal syndrome, intraoperative pRBC transfusion, intraoperative blood loss, and duration of surgery. Subsequently, we examined the association between extubation in the OR and each postoperative outcome (Supplementary Table 1). The logistic regression analysis revealed that extubation in the OR was significantly associated with a lower incidence of 30-day composite complications (odds ratio: 0.32, 95% confidence interval [CI]: 0.17-0.61, P < 0.001) (Supplementary Table 1). Furthermore, shorter ICU and postoperative hospital stays as well as a shorter duration of mechanical ventilation in the ICU, were significantly associated with extubation in the OR (odds ratio: 0.41, 95% CI = 0.21-0.78, P = 0.007; odds ratio: 0.17, 95% CI: 0.08-0.34, P < 0.001; odds ratio: 0.01, 95% CI: 0.00-0.02, P < 0.001, respectively) (Supplementary Table 1).
No statistically significant differences in reintubation rates or pulmonary complications were observed between the OREX and NOREX groups (Table 3). Furthermore, the logistic regression analysis demonstrated that these factors were not significantly associated with extubation in the OR (Supplementary Table 1).
Extubation in the OR was not an independent factor associated with OS in the univariate analysis (hazard ratio [HR]: 0.50, 95% CI: 0.24-1.05, P = 0.066). In the multivariate analysis, extubation in the OR did not show a statistically significant association with OS (HR: 0.79, 95% CI: 0.35-1.80, P = 0.580). The multivariate analysis included all significant variables identified in the univariate analysis, with the exception of intraoperative CRRT, which was administered to only one patient in this study. In the multivariate analysis, the preoperative PLT count (HR: 1.01, 95% CI: 1.00-1.01, P = 0.012), preoperative creatinine levels (HR: 2.15, 95% CI: 1.12-4.13, P = 0.021), and duration of surgery (HR: 1.20, 95% CI: 1.010-1.43, P = 0.037) were associated with a worse OS (Table 4, Fig. 2).
After propensity score matching, 54 patients in the OREX group were matched with 54 patients in the NOREX group. The overall postoperative outcomes following LDLT after matching were consistent with those observed before matching and are summarized in Table 3. After matching, the OREX group experienced fewer 30-day composite complications than the NOREX group (33.3 vs. 61.1%, P = 0.007). Moreover, after matching, OR extubation did not serve as an independent determinant of OS in the univariate (HR: 0.77, 95% CI: 0.25-2.38, P = 0.644) and multivariate analyses (HR: 1.00, 95% CI: 0.28-3.56, P = 0.988). These findings are consistent with the results obtained before matching.
Although the Kaplan-Meier curve revealed that the 5-year OS rates were higher in the OREX group (91.3%, 95% CI: 85.7-97.3) than that in the NOREX group (81.0%, 95% CI: 73.6-89.1), this difference was not statistically significant (P = 0.060) (Fig. 3).
DISCUSSION
This study analyzed the association between immediate extubation in the OR and long-term OS in patients undergoing LDLT and found that immediate extubation in the OR following LDLT did not improve OS compared with extubation in the ICU. Despite the lack of improvement in OS, the incidence of 30-day postoperative composite complications significantly reduced, and both ICU and hospital stays, which are important indicators of improved short-term recovery and resource utilization, shortened. However, these beneficial effects do not translate into a clinically significant decrease in all-cause mortality after LDLT.
An observational study involving 10,517 LT recipients indicated that prolonged intubation (> 96 h) was associated with increased mortality 1-year post-transplantation [17]. Similarly, a cohort study comprising 209 orthotopic LT recipients reported higher mortality rates at 1 month, 1 year, and 3 years post-transplantation in patients who experienced delayed extubation in the ICU [18]. While these studies focused primarily on short-term mortality and early extubation in the ICU, they provide a crucial context for our findings. In contrast, our study is the first to investigate the relationship between immediate extubation in the OR and long-term OS following LDLT, showing a trend toward an improved 5-year OS in the OR extubation group, although the P value was borderline (P = 0.060). This finding is significant because it expands the current understanding of extubation practices in LDLT, particularly highlighting the potential long-term benefits of immediate extubation in the OR, which has been underexplored in the literature. However, this study was conducted at a single center, and the limited sample size may have diminished the statistical power and affected the outcomes. Moreover, including patients who underwent early extubation in the ICU in the NOREX group may have concealed the true effect of extubation on survival following LDLT. Traditionally, early extubation involves both immediate extubation in the OR and extubation within a few hours of ICU admission [5,19,20]. Further research is warranted to compare postoperative outcomes between extubation in the OR and early extubation in the ICU, as well as to standardize the definition of early extubation. Extubation in the OR after LT presents additional challenges for anesthesiologists, despite the goal of minimizing complications and optimizing survival outcomes. This study provides valuable insights that could guide future clinical practices and protocols regarding extubation strategies for LDLT.
The ERAS protocol in LT aims to reduce short-term postoperative complications, with many studies highlighting the benefits of early extubation in decreasing postoperative morbidities [9,18,21-23]. A systematic review identified significantly lower reintubation and pulmonary complication rates in cohorts that underwent early extubation [22]. Additionally, compared with extubation in the ICU, early extubation has been associated with fewer non-pulmonary complications, including reduced renal and cardiovascular issues and shorter ICU and hospital stays [14,20,22,24]. This study also found a significant correlation between extubation in the OR and reduced composite complications, along with shorter ICU and hospital stays, which is consistent with previous research findings. Moreover, the duration of mechanical ventilation was shorter in the OR extubation group than that in the ICU extubation group. Ragonete et al. highlighted postoperative ventilation duration as a critical predictor of ICU mortality following LT [25]. In addition, Xu et al. [23] reported a lower incidence of composite mechanical ventilation-related adverse outcomes and a shorter ICU stay in the OR extubation group after LT. Therefore, reducing the duration of mechanical ventilation may reduce ICU morbidity after LT.
Identifying prognostic factors for survival after LDLT enhances patient selection and facilitates preoperative management. In this study, a higher preoperative PLT count was identified as an independent predictor of poor OS. Moreover, recent studies have associated increased preoperative PLT counts with an increased risk of hepatocellular carcinoma recurrence and reduced survival in various solid tumors [26-28]. PLTs, beyond their role in hemostasis and liver tissue regeneration, contribute to tumor growth and metastasis [26,29-31]. This interaction may increase tumor recurrence rates, thereby affecting OS. Further research is required to elucidate the comprehensive impact of preoperative PLT count on recurrence-free survival after LDLT.
Preoperative renal dysfunction is a well-known risk factor for the development of post-transplant chronic kidney disease and significantly affects the survival rates of LT recipients [32-36]. Identifying a sensitive and feasible biomarker of renal dysfunction is crucial for predicting long-term mortality in transplant recipients. This study revealed that high preoperative creatinine levels were predictors of poor OS. Creatinine, a readily available and cost-effective biomarker, correlates with the glomerular filtration rate (GFR). The Multinational Observational Study in Transplantation demonstrated that pre-LT serum creatinine levels independently predict the GFR 1-year post-LT, which in turn predicts the 5-year GFR following LT [32]. Early detection facilitates timely therapeutic interventions for renal protection and may improve OS after LT [37].
In this study, prolonged surgery was another risk factor for reduced survival in LDLT recipients, reflecting a complex surgical process and a potentially increased risk of complications. Filali et al. found that an operative time exceeding 4.5 h and surgical complications were independent risk factors for reduced survival in 1,630 patients who underwent LT [38]. Studies have also linked intraoperative pRBC transfusion with early mortality post-LDLT, suggesting that transfusing more than 8 units of pRBC independently predicts mortality [39,40]. In this study, the OREX group received fewer blood transfusions of pRBC, FFP, and PLT. However, neither blood transfusion nor blood loss was an independent risk factor for OS.
This study had a few limitations. First, the retrospective and observational design limited the ability to eliminate selection bias and control for confounding variables, despite using logistic regression analysis to mitigate these issues and compare postoperative complications between the groups. To reduce selection bias, we performed propensity score matching. The matched results were consistent with those obtained before matching, thus enhancing the reliability of our findings. However, the sample size became smaller after the matching. Second, the inclusion of patients who underwent early extubation in the ICU within the NOREX group may have interfered with the observed effects of immediate extubation in the OR. This overlap may have masked the true benefits of extubation in the OR on survival outcomes. Moreover, our study did not account for all the relevant perioperative factors that could influence survival after LDLT, such as the living donor’s age or graft size. These factors could significantly affect postoperative recovery and long-term survival, and their inclusion in future studies could provide a more comprehensive analysis.
Despite these limitations, the study has several strengths. To the best of our knowledge, this is the first study to investigate the effect of immediate extubation in the OR on long-term OS after LDLT. Previous studies focused on postoperative complications and early mortality. Furthermore, this is the first South Korean study to include the largest cohort of patients who underwent immediate extubation in the OR after LDLT.
In conclusion, immediate extubation in the OR did not significantly improve OS compared with extubation in the ICU after LDLT. Nonetheless, it was associated with reduced 30-day postoperative composite complications and shorter ICU and hospital stay, indicating significant short-term benefits. Large-scale prospective studies are required to accurately assess the effect of immediate extubation in the OR on long-term survival. However, the results of this study suggest that immediate extubation in the OR could enhance future perioperative care protocols for patients undergoing LDLT.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement
