Objectives: Despite advances in surgical technique and postoperative care, pediatric liver transplant outcomes remain highly variable, particularly in centers with limited access to deceased donor organs. We evaluated survival rates and identified key clinical, surgical, and laboratory parameters associated with posttransplant mortality in a high-volume single center performing primarily living donor liver transplants.
Materials and Methods: We retrospectively analyzed 98 pediatric liver transplants performed from January 2018 to February 2024 at the Istinye University Organ Transplantation Unit, Istanbul, Türkiye. All cases were ABO compatible. Assessed variables included recipient demographic data, indications, weight, preoperative and postoperative day 7 laboratory results, graft type, and complications. Survival was defined as time from liver transplant to death from any cause. Kaplan-Meier and Cox regression analyses were used to identify predictors of survival. P < .⁰⁵ was considered statistically significant.
Results: Median age at transplant was 17 months; 51.0% of patients weighed ≤10 kg. Living donor transplants accounted for 92.6% of cases. Overall crude survival was 74.5%. Cumulative survival rates at 6 months, 1 year, and 3 years were 78.5%, 77.4%, and 73.2%, respectively. Kaplan-Meier survival analysis showed that recipients with a body weight of ≤10 kg had a significantly higher mortality rate and shorter overall survival (P = .⁰¹²). Multivariate Cox regression identified 2 independent predictors of mortality: higher graft-to-recipient weight ratio (hazard ratio 1.498; P = .⁰³²) and platelet count decrease for each increment of 10 000 platelets/mm³ on postoperative day 7 (hazard ratio 1.046; P = .⁰⁰⁴).
Conclusions: This study highlights 3 modifiable or monitorable factors, that is, low transplant weight, high graft-to-recipient weight ratio, and early postoperative thrombocytopenia, as key predictors of reduced survival in pediatric liver transplant recipients. These findings emphasize the need for careful donor-recipient matching and early postoperative monitoring, particularly in centers that rely primarily on living donor transplants.

Key words : Body weight, Complications, Liver transplantation, Survival rate, Thrombocytopenia

Introduction
Liver transplant (LT) is widely accepted as an effective and reliable treatment for pediatric liver failure and a broad spectrum of metabolic disorders. During the past 3 decades, the increasing use of living donor LT (LDLT) and the establishment of well-defined donor safety limits have expanded access to transplantation, particularly in regions with limited access to deceased donor organs. This progress has contributed to an increased demand for LT centers in Türkiye. However, establishing and sustaining a high-quality pediatric LT program requires the coordinated integration of numerous complex components, such as optimal pretransplant management of the underlying disease and associated malnutrition, accurate timing of transplant in acute liver failure, strict adherence to ethical standards and donor safety in LDLT, and robust collaboration among surgical, anesthesiology, and intensive care teams through unified perioperative protocols. Furthermore, a well-trained, cohesive, and interdisciplinary team is essential to ensure seamless preoperative, intraoperative, and postoperative care. Consequently, the development of a new LT center that fully complies with international standards and ethical principles is a time-consuming and resource-intensive process. In this study, we aimed to evaluate the first 5-year survival outcomes and associated prognostic factors of a pediatric LT center that provides care to both national and international patients in accordance with the principles of the Declaration of Istanbul.¹ Due to the inability to include international patients on the national organ allocation list and the overall scarcity of deceased donor organs in the country, the program predominantly performs LDLTs. By analyzing these outcomes, we aimed to not only identify opportunities for further improvement in survival but also to provide practical insights for newly established pediatric LT programs.

Materials and Methods
From January 2018 to February 2024, 98 pediatric LTs were performed at the Istinye University Organ Transplantation Unit, Istanbul, Türkiye. Ethical board approval was obtained from the Istinye University Ethical Board on November 22, 2024 (approval No. 24.240). Our study conformed to the ethical guidelines of the 1975 Declaration of Helsinki. Written informed consent was obtained from patients or their guardians. All LT cases were ABO compatible. In living donor organ transplants, nonrelated donor transplant is not permitted for either domestic or international patients unless approved by the national organ transplantation ethics committee. We retrospectively evaluated the etiology of LT, age, weight, height, body mass index, and preoperative and postoperative day 7 laboratory results for alanine aminotransferase (ALT), aspartate aminotransferase (AST), total/direct bilirubin, γ-glutamyl transferase, alkaline phosphatase, hemoglobin, platelet count (PLT), prothrombin time, creatinine, and albumin. We also assessed type of liver graft, graft-to-recipient weight ratio (GRWR), biliary anastomosis type (duct-to-duct or hepaticojejunostomy), number of graft biliary ducts, and intraoperative biliary stent application. In addition, we evaluated length of stay in the pediatric intensive care unit and organ transplant unit, early allograft dysfunction (defined as bilirubin ≥10 mg/dL on postoperative day 7, international normalized ratio ≥1.6 on postoperative day 7, or ALT or AST >2000 U/L within the first 7 postoperative days), postoperative complications affecting gastrointestinal (perforation, gastroparesis, ileus, and bleeding), vascular (thrombosis, stenosis, bleeding), and biliary systems (stenosis, leak, cholangitis); infections (type, timing, treatment), rejection episodes (timing, treatment), immunosuppression regimens, survival status, and follow-up length.

Preoperative evaluation and surgical procedures
Computed tomographic volumetry was used for preoperative evaluation of ABO-compatible living donors. Donor hepatectomy type was determined based on recipient body weight, underlying disease, and donor vascular and biliary anatomy. All transplant procedures were performed by the same team of 4 surgeons with specialization in LT. When performed, deceased donor grafts are selected according to ABO compatibility and recipient body weight.

Postoperative follow-up and immunosuppression
Patients underwent routine laboratory testing and ultrasonographic evaluation every 12 hours during the first 48 hours after transplant and daily thereafter. Vital signs and drain output were monitored 4 times daily throughout hospitalization. After discharge, patients were followed weekly during the first month, monthly until the end of the first year, and every 3 months thereafter. Domestic patients were followed in person, whereas international patients were monitored using the same schedule with remote consultations when necessary. Immunosuppression consisted of methylprednisolone and tacrolimus in patients weighing <5 kg, and a triple regimen of methylprednisolone, tacrolimus, and mycophenolate mofetil was applied for patients weighing ≥5 kg, unless contraindicated. Methylprednisolone was discontinued at 3 months and mycophenolate mofetil at 6 months in the absence of rejection. Target tacrolimus trough levels were 8 to 10 ng/mL during the first 3 months, 5 to 8 ng/mL during the period from month 4 to month 12, and approximately 5 ng/mL thereafter. Cytomegalovirus prophylaxis with oral valganciclovir (15 mg/kg/day) was initiated on postoperative day 3 and continued for 6 months in cytomegalovirus mismatch cases and for 3 months in other patients.

Statistical analyses
We used the Kolmogorov-Smirnov test and the Levene test to evaluate assumptions of normal distribution and homogeneity of variances. We expressed categorical data as frequency (with percentage) and quantitative data as mean values (±SD) or median values (with either min-max or 25th-75th percentiles) where appropriate. Unless otherwise stated, we analyzed categorical data with the Pearson chi-square test. The primary endpoint was overall survival (OS), defined as the time from transplant to death from any cause. Kaplan-Meier survival analysis via the log-rank test was used to determine statistically significant associations between OS, body weight at time of transplant, and type of biliary anastomosis. Cumulative survival rates (at 6 months, 1 year, and 3 years) and mean survival durations with 95% CI were calculated. We conducted univariate Cox proportional hazards regression models to examine associations between demographic characteristics, clinical features, and biochemical measurements with OS. We used multivariate Cox proportional hazards regression models with a forward likelihood ratio procedure to identify independent predictors that most affected OS after adjustment for confounding factors. We included variables with a univariable P < .¹⁰ in the multivariable model. For each independent variable, we calculated hazard ratios (HR), 95% CIs, and Wald statistics. We used SPSS software (version 25; IBM) for all statistical analyses. P < .⁰⁵ was considered statistically significant.

Results
Our center has been operating as an active LT center for 6 years, with a LDLT rate of 92.6%. The average number of LTs per year is 90, with an average of 16 pediatric transplants. In the present study, data from 98 cases, with transplant ages ranging from 84 days to 18 years, were evaluated. Of the cases, 71 (72.4%) were male patients, and 27 (27.6%) were female patients. There were 50 recipients (51.0%) with a body weight of ≤10 kg at the time of transplant (Table 1). Hepaticojejunostomy was performed in 46 cases (46.9%); duct-to-duct anastomosis was performed in 52 cases (53.1%). The most common indication for LT was biliary atresia at 28.1%, followed by progressive familial intrahepatic cholestasis (Table 2). Laboratory differences on postoperative day 7 versus baseline values were evaluated. The levels of AST, bilirubin, and platelets showed significant differences (P < .⁰⁰¹), whereas values for ALT, GGT, hemoglobin, and international normalized ratio did not show significant differences (Table 3).

Survival rates
The overall crude survival rate among all cases was 74.5%, and cumulative survival rates at 6 months, 1 year, and 3 years were 0.785, 0.774, and 0.732, respectively (Figure 1). The expected mean survival time was calculated to be 57.7 months (95% CI, 51.1-64.3). Because of very low rate of retransplant, patient survival was assessed instead of graft survival. Median follow-up period was 23.5 months (minimum 0 days to maximum 77.3 months). Univariate Cox proportional hazards regression analysis of potential factors that might affect survival identified the following risk factors: recipient weight of ≤10 kg at the time of transplant, GRWR, decrease in platelet count for each increment of 10 000 platelets/mm³ on postoperative day 7 versus the preoperative measurement (delta PLT), and gastrointestinal complications (Table 4). When these 4 factors were analyzed with multivariate Cox proportional hazards regression, 2 factors were found to be significant: high GRWR and high delta PLT (Table 5). Kaplan-Meier survival analysis demonstrated that recipients with a body weight of ≤10 kg at the time of transplant had a significantly higher mortality rate and shorter OS versus recipients weighing >10 kg (P = .⁰¹²) (Table 6, Figure 2). In univariate Cox proportional hazards regression analysis, a body weight at transplant of ≤10 kg was associated with a significantly increased risk of mortality versus a body weight >10 kg (HR 2.898; 95% CI, 1.209-6.946; P = .⁰¹⁷) (Table 4). However, this association did not remain significant in the multivariable Cox model after adjustment for other covariates (Table 5). As the GRWR level increased, the mortality rate also increased significantly (HR 1.739; 95% CI, 1.268-2.384; P < .⁰⁰¹). In the next step, percentiles at 33.3 and 66.6 were calculated for GRWR, and cases were divided into tertiles. When the first tertile (GRWR <1.73) was taken as the reference, although the mortality rate increased in the second tertile (GRWR between 1.73 and 3.09), this association was not significant (HR 1.285; 95% CI, 0.392-4.213; P = .⁶⁷⁹). On the other hand, compared with the first tertile, the mortality rate in the third tertile (GRWR >3.09) was found to be significantly increased (HR 3.595; 95% CI, 1.293-9.994; P = .⁰¹⁴). Each decrease of 10 000 platelets/mm³ in platelet levels on postoperative day 7 versus the preoperative measurement significantly increased the mortality rate (HR 1.058; 95% CI, 1.027-1.090; P < .⁰⁰¹).

Discussion
Our cumulative survival rates at 6 months, 1 year, and 3 years were 78.5%, 77.4%, and 73.2%, respectively. These outcomes are lower than the 86% survival rate reported for primary pediatric LT by the European Liver Transplant Registry in the post-2010 era.² However, when the analysis was limited to patients who survived beyond the first posttransplant year, the 3-year survival rate in our cohort was comparable to registry data, reaching 96% versus 97% reported by the European Liver Transplant Registry.² Several center-specific and cohort-specific factors may explain the relatively lower OS observed in our study. First, our transplant program was established relatively recently, and early outcomes may reflect the learning curve inherent to newly established high-volume centers. Second, the extremely limited availability of deceased donor organs for domestic patients, together with legal and logistical barriers preventing international patients from being listed on the national waiting system, has resulted in a predominant reliance on LDLT. In our study, the finding that 84% of patient deaths occurred within the first 6 months after LT once again underscores the critical importance of this early postoperative period in patient follow-up. Although early mortality rates following LT in Europe were reported to decrease over time up to 2017, the same emphasis was reiterated in European data from 2018 to 2022, with an early mortality rate of 82% reported by Baumann and colleagues² and Junge and colleagues.³ The LDLT procedure is technically more complex versus deceased donor transplant and is often associated with longer postoperative hospitalization and an increased risk of perioperative complications, which may adversely affect early survival outcomes. Consistent with this observation, Junge and colleagues reported lower survival rates following LDLT versus deceased donor transplant in the European registry between 2018 and 2022.3 The authors hypothesized that this difference may be related to case selection, because living donor transplant is frequently offered to more severely ill patients without access to deceased donor organs due to organ shortages. In contrast, 2023 data for LT from the database managed by the Organ Procurement and Transplantation Network and Scientific Registry of Transplant Recipients demonstrated higher 5-year survival rates for living donor transplant recipients versus deceased donor transplant recipients in the United States (94.5% vs 90.3%), although the reasons underlying this advantage were not fully explored.⁴ Despite these regional differences, available evidence collectively supports the safety and feasibility of LDLT in pediatric patients. Retransplant is a well-recognized determinant of long-term survival in pediatric LT. In our cohort, the retransplant rate was relatively low, largely due to severe organ shortages, and the effect of retransplant on OS was therefore limited. In a study by Moosburner and colleagues, based on 30 years of experience, 1-year, 5-year, and 10-year survival rates following living donor transplant were reported as 87.9%, 87.9%, and 83.5%, respectively; however, a retransplant rate of 12% was also reported, which significantly influenced survival outcomes.⁵ Similarly, Junge and colleagues reported that among 2221 pediatric LTs performed between 2018 and 2022 in the European registry, 135 patients required retransplant, underscoring the substantial effect of graft failure on survival.³ High pretransplant disease severity was hypothesized to negatively affect survival in our cohort. Although patients in our study had relatively high PELD/MELD scores, multivariate analysis did not demonstrate a significant association between PELD/MELD scores and survival. Previous studies have reported conflicting results. For example, Raices and colleagues identified a PELD score >25 as a significant predictor of mortality.⁶ Another study focusing on biliary atresia demonstrated increased mortality in patients with PELD scores >20.7 In our cohort, the limited number of patients with lower PELD/MELD scores likely reduced the statistical power to detect a significant association. Low body weight and young age represent additional challenges in pediatric LTs. Patients weighing <10 kg constitute a high-risk group due to small vessel diameter, limited intra-abdominal space, and increased technical complexity. In our study, patients weighing <10 kg demonstrated lower survival rates. Similarly, Baumann and colleagues, in a report summarizing 30 years of transplant experience, demonstrated significantly higher early mortality and lower OS in patients who received transplant before 1 year of age.² Junge and colleagues also reported significantly lower 5-year graft survival in patients who received transplant before 1 year of age in the European registry between 2018 and 2022.3 Conversely, analysis of national data from Poland showed comparable survival rates between patients weighing <6 kg and patients weighing 6 to 10 kg, although hospitalization duration and complication rates were higher in the lower weight group.⁸ Data from the Society of Pediatric Liver Transplantation registry further demonstrated that patients weighing <5 kg had mortality rates comparable with the mortality rates of heavier patients, despite greater severity of clinical illness, receiving partial grafts more frequently, experiencing longer hospital stays, and having higher morbidity rates.⁹ The GRWR is another critical factor shown to influence outcomes in pediatric LTs. Although the adverse effect of excessively high GRWR is a well-recognized factor, a definitive safe upper limit has not been clearly established. Previous studies have suggested upper thresholds for GRWR ranging from 3% to 4%.¹⁰ In our cohort, higher GRWR values were independently associated with increased mortality in the multivariable Cox regression model when GRWR was analyzed as a continuous variable. When GRWR was further categorized to identify a clinically relevant threshold, a cut-off value of >3.09% was associated with significantly increased mortality in univariate analysis; however, this categorical threshold did not retain statistical significance in the multivariable model. Consistent with our findings, Avanaz and colleagues reported lower survival rates in patients with GRWR exceeding 4%.¹¹ In contrast, Zakaria and colleagues observed no significant differences in graft or patient survival for monosegmental or reduced left lateral segment grafts versus standard left lateral segment grafts with GRWR ≥4%.¹² Eguchi and colleagues, in a large cohort of over 10 000 LDLTs, suggested that the traditionally accepted upper GRWR limit of 4% could be extended to 5%.¹³ Similarly, Goldaracena and colleagues proposed that transplants with GRWR >4% may be feasible in settings of organ shortage when delayed abdominal wall closure was applied.¹⁴ Despite these discrepancies, there is general consensus that complication risk increases when GRWR exceeds approximately 3.5% to 4%, highlighting the importance of individualized risk-benefit assessment in such cases. Postoperative platelet dynamics also emerged as a significant predictor of survival in our cohort. We observed that for every incremental decrease of 10 000 platelets/mm³ in platelet count on postoperative day 7, mortality risk increased. Early thrombocytopenia in the first postoperative week following LT is a common and expected phenomenon, largely reflecting platelet sequestration during graft regeneration. Nevertheless, accumulating evidence indicates that profound thrombocytopenia is associated with a higher risk of postoperative complications. Although thrombocytopenia after LT has been associated with bleeding, impaired graft regeneration, and major complications, it remains unclear whether post-LT thrombocytopenia is a cause or a consequence of adverse outcomes.¹⁵ Lesurtel and colleagues demonstrated that a platelet count <60 000 platelets/mm³ on postoperative day 5 was associated with increased mortality and severe complications in adult whole organ LT recipients.¹⁶ Han and colleagues reported that intraoperative platelet transfusion improved liver regeneration without increasing complication rates, and higher intraoperative platelet levels were associated with better regenerative outcomes in patients who did not receive transfusions.¹⁷ Takahashi and colleagues further highlighted the association between thrombocytopenia and impaired graft regeneration and emphasized the need for alternative strategies to platelet transfusion due to its potential complications.¹⁵ In subsequent experimental studies, thrombopoietin-induced thrombocytosis was shown to improve graft regeneration and survival.¹⁸ Although routine platelet supplementation or pharmacological stimulation of platelet production is not currently recommended due to the risk of hepatic artery thrombosis in the early postoperative period, our findings underscore the prognostic relevance of postoperative platelet trends and the need for further investigation. This study was conducted at a relatively new but high-output single transplant center, offering a realistic perspective from a region affected by severe organ shortages and societal barriers to organ donation. Although our findings provide valuable insights into multifactorial determinants of survival, longer follow-up periods and multicenter studies are required to improve generalizability. Further research focusing on postoperative platelet kinetics and the mechanistic relationship with graft regeneration and survival may contribute to the development of evidence-based management strategies. Overall, our study adds meaningful data to the pediatric LT literature by comprehensively evaluating survival outcomes and associated risk factors in a challenging clinical setting.

Conclusions
Liver transplant is an established and life-saving treatment for pediatric liver failure and a wide spectrum of metabolic disorders; however, the establishment of a new transplant center and the achievement of optimal outcomes remain challenging. In this study, conducted at a recently established and predominantly living donor-dependent pediatric LT center, we evaluated survival outcomes and factors associated with mortality. Our findings indicated that a body weight of ≤10 kg at the time of transplant remains a significant risk factor for reduced survival. In addition, higher GRWR was independently associated with poorer outcomes, and a postoperative platelet count decrease of more than 10 000 platelets/mm³ on day 7 emerged as a negative prognostic marker. These results highlight the importance of careful perioperative risk stratification and close postoperative monitoring, particularly in high-risk pediatric recipients. Although longer follow-up and multicenter validation are required, our data provide clinically relevant insights from a newly established, living donor-based transplant program and may contribute to improvements in patient selection, perioperative management, and outcomes in similar settings.