2.1 Patient Cohort and Study Design

We retrospectively reviewed 1645 LTx performed at Hannover Medical School, Germany, between January 2005 and December 2021. Of these, 421 transplantations were performed in 405 pediatric recipients, defined as under 18 years of age. Combined thoracic organ transplantations and liver re-transplantations within the same hospital stay were excluded from further analysis.

2.2 Study Endpoints

Primary study endpoints were postoperative complications, such as pleural effusions, as well as vascular and biliary complications requiring intervention or surgical revision within the first 30 days after LTx. In addition, the need for retransplantation and patient survival within the first year following LTx were selected as endpoints.

2.3 Definition of Variables

For subgroup analysis, pediatric recipients were divided into three age groups: Recipients under 3 years of age were defined as infants and toddlers, 3- to 12-year-olds were defined as children and patients older than 12 years of age were defined as adolescents.

The body mass index (BMI) was calculated using the common formula (= height {m}/weight {kg}²). The body surface area (BSA) was calculated using the Mosteller formula for recipients and pediatric donors (= (height {cm} × weight {kg}/3600)^1/2) and the Dubois formula for donors older than 18 years of age (= 0.007184 × height {cm}^0.725 × weight {kg}^0.425). The mismatch ratios were reported as donor-recipient quotients. To analyze whether the transplant era influenced the incidence and/or type of postoperative complications, recipients were divided into three groups according to the date of LTx (2005–2010, 2010–2015, and 2015–2021). The presence of portal hypertension was described as a permanent increase in portal venous pressure above 5 mmHg. Invasive pressure measurements were not performed as standard, so that indirect signs (e.g., esophageal or fundus varices, ascites, liver cirrhosis, cavernous portal vein transformation) were usually used for diagnosis. We considered a pleural effusion to be a sonographically diagnosed effusion that required intervention (thoracentesis or chest tube placement). We counted insufficiency, stenosis, or thrombosis of the arterial or venous anastomosis that occurred within the first 30 days after transplantation and required intervention as vascular complications. Postoperative bleeding that could not be attributed to anastomotic insufficiency but had a different source of bleeding or was of diffuse origin was listed separately. Biliary complications include insufficiency or stenosis of the biliary anastomosis, a peripheral bile leak, or necrosis of the bile duct within the first 30 days after transplantation, which required intervention.

2.4 Statistical Methods

Data were collected and anonymously analyzed using SPSS statistical software (version 27.0.1.0; SPSS Inc.; IBM Corporation). The statistical significance level was set at a p-value of < 0.050. Statistically significant values are marked bold in the tables. Metric data was tested for normal distribution using the Kolmogorov–Smirnov test. In case of normal distribution, mean was compared using the Student's t test, otherwise the Mann–Whitney U test was applied. For the statistical comparison of nominal or ordinal data between groups the Chi² test or the Fisher's exact test were used. Univariate binary logistic regression analysis was performed to quantify the effect of the variables on nominal study endpoints. To identify independent risk factors a multivariate binary logistic regression analysis was performed, by a stepwise forward selection of purposeful selected variables with a p-value of < 0.200 in the univariate regression analysis and with missing values of < 10%.

2.5 Ethics

The legal guardians of pediatric patients provided informed consent for the use of their data for scientific purposes at the time of hospital admission, which represents the general policy of our institution. The ethical committee of the Hannover Medical School stated that no further approval for retrospective data collection is needed. Patient records and patient data were anonymized and de-identified prior to analysis.

3.1 Recipients and Donor Characteristics and Matching

During the analyzed period, 1645 LTx were performed at our institution. Of these, 421 transplantations in 405 pediatric recipients were included in our analysis.

51.8% of the pediatric recipients were male with a median age of 2.97 (0.05–17.97) years at transplant. Due to the heterogeneity of the pediatric patient collective, underlying diseases and thus perioperative complications varied significantly depending on recipient age: Infants and toddlers represented the largest age group with 212 cases (50.4%), followed by children (n = 151; 35.9%) and adolescents (n = 58; 13.8%). While disease of the biliary tract, especially biliary atresia, were the most common underlying hepatic disease in infants and toddlers (rate of 59.3% in infants vs. 25.8% in children vs. 8.8% in adolescents, p < 0.001), metabolic diseases were the predominant underlying disease in the group of adolescents (rate of 10.5% in infants vs. 27.2% in children vs. 42.1% in adolescents, p < 0.001).

In 315 (74.8%) cases the recipients had developed a liver cirrhosis at the time of transplant leading to ascites in 253 (60.1%) and portal hypertension in 223 (53.8%) cases. The manifestation of an encephalopathy at the time of transplant was less frequent with 44 (10.5%) cases. In 34.7% of cases (n = 146), recipients underwent prior Kasai operation or a previous LTx (n = 36; 8.6%). In 89 cases (21.1%) the pediatric recipients received previous biliary interventions (such as stenting of the bile duct or percutaneous transhepatic cholangiodrainage), a correlation with patients undergoing retransplantation was observed: Of 36 patients with previous liver transplantation, 21 (58.3%) had a previous intervention (such as stenting of the bile duct or percutaneous transhepatic cholangiodrainage), while only 68 (18.1%) of 376 first-time transplant recipients underwent a prior intervention (p < 0.001).

Recipient age was associated with the incidence of postoperative complications: Vascular complications occurred significantly more frequently in the postoperative course of infants and toddlers compared to other recipient age groups (28.2% in infants vs. 19.5% in children vs. 12.3% in adolescents, p = 0.018). Accordingly, infants and toddlers showed a trend towards an increased rate of re-transplantations in the same hospital stay (13.5% in infants vs. 8.1% in children vs. 8.8% in adolescents, p = 0.232). Pleural effusions were the most common complication with 208 cases (49.4%) and occurred more often in the group of children (43.4% in infants vs. 59.1% in children vs. 52.6% in adolescents, p = 0.012). Also, biliary complications were particularly frequent in the group of children (17.7% in infants vs. 24.8% in children vs. 17.5% in adolescents, p = 0.219).

Donors were of male gender in 48.7% (n = 205) of cases with a median age of 27 (0–60) years. Livers from pediatric donors defined as under the age of 18 years were transplanted in 140 (33.3%) cases. Living donation occurred in 102 (24.2%) cases, with the mother (n = 43; 10.2%), the father (n = 39; 9.3%) or other first grade relatives as donors (n = 19; 4.5%). The median BMI of the donors was 22.49 kg/m² (11.11–35.0 kg/m²). Donor age or BMI did not show an effect on the onset of perioperative complications within the first 30 days after transplantation. In almost one-third of the included cases (n = 119; 28.3%) the liver was allocated via high urgency status.

Table 1 gives an overview of biometrical data of recipients, donors, and donor-recipient matching.

3.2 Transplantation and Surgical Details

In 2.9% (n = 12) of cases a combined liver-kidney transplantation was performed. A split-liver transplantation was performed in 67.5% (n = 284) of cases, most frequently via engraftment of a left lateral segment (n = 216; 51.3%). The median cold ischemia time was 540 (117–1025) minutes, and the median portal venous clamping time was 63 (30–138) minutes. Both the arterial and venous anastomoses were performed prior to reperfusion of the transplant liver. A primary aortal anastomosis was performed in 15.2% (n = 64) of cases. A venous interposition was required in 13 cases (3.1%) for the reconstruction of portal vein and in 8 cases (1.9%) of the inferior vena cava, respectively. For biliary reconstruction, a biliodigestive anastomosis was chosen in 66.0% (n = 278) of cases, whereas an end-to-end anastomosis was performed in 28.3% (n = 119) of cases. The latter was associated with an increased incidence of biliary complications (34.5% of recipients with end-to-end anastomosis of the bile duct developed biliary complications vs. 14.5% without an end-end reconstruction of the bile duct, p < 0.001). In 67.2% (n = 283) of cases, a temporary abdominal closure was required most frequently due to intestinal edema (n = 59; 14.0%) and impaired perfusion of the liver (n = 43; 10.2%). In 154 cases (36.6%) a two-staged procedure for biliary reconstruction was performed. For this purpose, we place a stent, which is inserted transdermally into the bile duct, fixed by ligation at the distal end of the recipient's bile duct. During the second procedure, the ligated bile duct is then resected for biliary reconstruction (to avoid the usage of the previously ligated part of the bile duct for anastomosis). In recipients receiving split liver transplantation with very delicate and often short bile ducts, we place sterile gauze and an additional suction drain near the bile duct. Following transplantation, a ciclosporin-based immunosuppressive therapy was induced in 50.1% (n = 211) of cases especially in the early transplant era and a tacrolimus-based therapy in 45.8% (n = 193) of cases.

Table 2 gives an overview of surgical details and transplantation procedure.

3.3 Short-Term Outcome and Perioperative Complications

The median intensive care unit (ICU) stay was 7 (0–105) days, and the median hospital stay was 36 (3–176) days. The most frequent postoperative complication was pleural effusions in 49.4% (n = 208) of cases, requiring thoracentesis in 20.7% (n = 87) and the placement of a chest drain in 40.3% (n = 170) of cases, respectively. Fifty patients (11.9%) required both postoperative thoracentesis and the placement of a chest tube. To differentiate reactive pleural effusion from pleural effusions due to massive water retention or volume shifts, respectively, right-sided pleural effusions (n = 185, 88.9% of pleural effusions) requiring intervention were analyzed separately from bilateral (n = 19, 9.1% of pleural effusions). We observed a correlation between pleural effusion requiring intervention and venous stenosis within the first 30 days after transplantation (9.1% of pleural effusions occur in association with venous stenosis vs. 90.9% of pleural effusions occurring independently, p = 0.030).

The second most common postoperative complication was vascular complications observed in 22.6% (n = 95) of cases within the first 30 days after transplantation: Of these, 44 recipients (10.5%) experienced arterial complications due to thrombosis (n = 34; 8.1%), stenosis (n = 7; 1.7%), or insufficiency (n = 6; 1.4%). In 58 patients (13.8%) portal or central venous complications such as thrombosis (n = 28; 6.7%), stenosis (n = 27; 6.4%), or insufficiency (n = 7; 1.7) occurred. Eight (1.9%) recipients experienced both arterial and portal or central venous complications. In accordance with our clinical standard, vascular stenoses were firstly diagnosed by routinely performed sonography and then, if highly suspected, confirmed by angiography or CT scan and treated immediately if they were clinically relevant. Thrombosis of hepatic artery and portal vein were immediately surgically revised by reopening of the anastomosis, extensive flushing with heparin saline solution, removal of any remaining thrombi using a Fogarty catheter and subsequent new anastomosis. In some cases, the new anastomosis of hepatic artery had to be performed as aortal anastomosis. Interventional angiological revisions in cases of thrombosis of the portal vein or hepatic artery were relatively rare and tended to be a last chance treatment strategy for very early complications. In case of portal vein thrombosis, the applied heparin doses were increased. In both cases the frequency of ultrasound-assisted monitoring of graft perfusion and laboratory controls were increased. Early biliary complications were observed in 84 (20.0%) cases: mostly due to insufficiency of the biliary anastomosis (n = 34; 8.1%) or stenosis of the biliary anastomosis (n = 27; 7.6%), followed by peripheral bile leakage (n = 24; 5.7%) and necrosis of the bile duct (n = 7; 1.7%). Hemorrhage in absence of an insufficiency of the vascular anastomoses occurred in 16.2% (n = 68) of cases. Only 12 (2.9%) patients developed a wound infection requiring intervention in the postoperative course. Histopathologically confirmed graft rejection within the first 30 days after transplantation occurred in 22.6% of cases (n = 95). A re-transplantation within the first 30 days was required in 9.5% (n = 40) of cases mostly due to vascular complications (82.5% of all retransplantations had prior vascular complications vs. 17.5% without prior vascular complications, p < 0.001). Twelve patients (2.8%) died within the first 30 days following operation: Mainly due to multiorgan failure with consecutive cardiovascular failure (n = 6) and due to cerebral edema (n = 5). One year after transplantation 48 (11.4%) patients had needed retransplantation and 20 (4.7%) patients were deceased.

The allocation via high urgency status did not affect vascular or biliary complications, pleural effusions, or the need for retransplantation (p > 0.05). However, recipients with high urgency allocation showed a significantly higher 1-year mortality rate (9.8% with high urgency allocation status vs. 3.1% without high urgency status, p = 0.006).

The transplant era did not affect rates of vascular or biliary complications. Pleural effusions occurred significantly less frequently in the late era (37.5% of all pleural effusions occurred in the early era vs. 34.6% in the intermediate era vs. 27.9% in the late era, p = 0.045). The need for retransplantation within the first year after transplantation also decreased significantly in the late transplant era (18.4% in the early era vs. 9.9% in the intermediate era vs. 6.7% in the late era, p = 0.008). The mortality within the first year after transplantation decreased considerably but not significantly (7.6% in the early era vs. 5.9% in the intermediate era vs.1.5% in the late era, p = 0.070). Table 3 summarizes the complications in the postoperative course, graft function, and patient survival.

3.4 Identification of Risk Factors for the Onset of Pleural Effusions

The results of the univariate analysis for the onset of pleural effusion are shown in Table 4. Multivariate analysis identified a recipient age of 3–12 years (OR: 1.849; CI 95%: 1.170–2.923; p = 0.008) and donor age (OR: 1.019, CI 95%: 1.004–1.034, p = 0.010) as independent risk factors for the postoperative onset of pleural effusion requiring intervention. Prior liver transplantation (OR: 0.407; CI 95%: 0.189–0.879; p = 0.022) and biliary diseases (in most cases biliary atresia) as underlying disease represents a protective factor for the onset of pleural effusion in the postoperative course (OR: 0.565; CI 95%: 0.367–0.872; p = 0.010).

3.5 Identification of Risk Factors for Other Surgical Complications

The results of the univariate analysis for the onset of vascular complications are shown in Table 5. Multivariate regression shows that prior LTx (OR: 2.564; CI 95%: 1.091–6.025; p = 0.031), primary aortal anastomosis (OR: 2.654; CI 95%: 1.390–5.067; p = 0.003) and central venous interposition grafts (OR: 6.601; CI 95%: 1.403–31.050; p = 0.017) are independent risk factors for the postoperative development of vascular complications requiring intervention. Recipient liver weight on the other hand represents a protective factor for the onset of vascular complications (OR: 0.999; CI 95%: 0.998–1.000; p = 0.006). The result of univariate regression of biliary complications requiring intervention within the first 30 days after transplantations are shown in Table 6. Multivariate analysis revealed an end-end anastomosis of the bile duct as independent risk factor for the onset of postoperative biliary complications (OR: 2.039; CI 95%: 1.156–3.596; p = 0.014). Previous intervention such as stenting of the bile duct or percutaneous transhepatic cholangiodrainage (OR: 0.465; CI 95%: 0.229–0.946; p = 0.035) and days until anastomosis of the bile duct (OR: 0.722; CI 95%: 0.537–0.970; p = 0.030) represent protective factors regarding the development of biliary complications.

3.6 Identification of Risk Factors for Retransplantation and Patient Death Within the First Year After Transplantation

The results of the uni- and multivariate analysis regarding retransplantation and patient survival within the first year after transplantation are shown in Tables S1 and S2 which are provided in the supplements. Multivariate analysis revealed arterial complications (OR: 27.877; CI 95%: 9.956–78.055; p < 0.001) and venous complications (OR: 4.786; CI 95%: 1.665–13.755, p = 0.004) as independent risk factors for the need for retransplantation within the first year after LTx. Donors of male gender (OR: 0.328; CI 95%: 0.126–0.852; p = 0.022) and days until anastomosis of the bile duct (OR: 0.522; CI 95%: 0.313–0.870; p = 0.013) on the other hand, were found to be independent protective factors regarding retransplantation. Independent risk factors for patient death within the first year after transplantations were body surface area of the donor (OR: 4.249; CI 95%: 1.020–17.701; p = 0.047), tumor as underlying disease (OR: 4.608; CI 95%: 1.235–17.197; p = 0.023), prior encephalopathy (OR: 5.933; CI 95%: 1.686–20.871; p = 0.006), previous LTx (OR: 7.499; CI 95%: 1.960–28.689; p = 0.003) and arterial complications within the first 30 days after transplantation (OR: 4.539; CI 95%: 1.235–16.689; p = 0.023).

4.1 Pleural Effusions

The most common postoperative complications were pleural effusions in 49.4% of all included cases being in line with the reported rates of 32%–47% of pleural effusions after LTx in adults requiring chest drainage in up to 60.5% leading to significantly reduced short-term survival [5-7]. Accordingly, a correlation between the onset of pleural effusions after LTx and pulmonary infections and the subsequent development of a systemic inflammatory response syndrome is described [8]. Only one study by Mack et al. focused on pleural effusion following pediatric LTx and reported an incidence of 25% in a patient collective of 151 transplantations in total and showed a correlation between persistent pleural effusion and allograft rejection [11]. We could not observe similar findings in our patient collective. We identified a recipient age of 3–12 years as a risk factor for the development of pleural effusions. The reason for this could be the heterogeneity of underlying diseases in this patient subgroup with prolonged times on the waiting list and thus compromised preoperative physical condition and progressive cirrhotic remodeling of the liver leading to severe portal hypertension, the development of a hepatopulmonary syndrome and perioperative volume shifts, consequently. Accordingly, it is well established for adult recipients that chronic liver failure, concomitant portal hypertension, and the presence of ascites in the preoperative course are associated with reduced lung compliance, pulmonary dysfunction, and increased respiratory complications leading to increased morbidity and mortality [12-14]. To put this into perspective, however, it should be noted that 88.9% of all pleural effusions requiring intervention occurred on the right side in our patient collective, so that we must assume a reactive genesis here and a correlation with systemic diseases (such as cardiac or renal insufficiency or portal hypertension) contributed to only a small proportion of pleural effusions. Multivariate analysis further revealed donor age as independent risk factor for the onset of pleural effusions. A possible explanation might be that particularly for children in critical condition due to very advanced or acute underlying diseases, organs from less ideal donors need to be accepted. The critical condition then leads to the need for preoperative intensive care treatment and prolonged intense care unit stay in the postoperative course with corresponding volume shifts. Multivariate regression analyses further revealed biliary diseases (in most cases biliary atresia) as an indication for LTx as a protective factor against the postoperative development of pleural effusions, which, to our knowledge, has not been described in the literature yet. Although clinically standardized in the adult setting, there is no literature regarding the benefit of prophylactic intraoperative chest tube placements in pediatric recipients. In the adult setting, however, lower complication rates have been described for intraoperative chest drain placement under sterile conditions and general anesthesia compared to postoperative placement in the intensive care unit [6]. Postoperative thoracentesis or chest tubes leading to major complications such as hemothorax, pneumothorax, or injury of liver or spleen are described in up to 6% of pediatric patients [15, 16]. It can therefore be concluded that the high incidence of pleural effusions in pediatric recipients requires thoracentesis or placement of chest tube placement in many cases and can therefore influence the course of many patients decisively. Whether a recommendation toward prophylactic intraoperative chest drain placement under general anesthesia and sterile conditions can be inferred from this requires further large multicenter studies or meta-analyses.

4.2 Vascular Complications

Vascular complications, including thrombosis, stenosis, and insufficiencies of the arterial or (portal-) venous anastomoses, were observed in 22.6% of cases. In general, the rate of vascular complications following LTx ranges from 7% to 15% in the current literature and is significantly higher in pediatric patients than in adults [9]. Moreover, an increased incidence of vascular complications after living donation and use of split grafts has been reported in the past, both of which apply to our patient collective with a relevant rate [17, 18]. Of note, most authors list insufficiencies of vascular anastomoses as separate (hemorrhagic) complications hindering statistical comparison of reported rates between different publications, especially when summarizing types of complications.

Thrombosis of the hepatic artery is often described as the most common vascular complication with an incidence of 4%–15%, which is in line with an incidence of 8.1% in our patient collective [3, 19, 20]. The associated mortality rates vary between 42.8% and 100%, depending on the time of occurrence and applicable allocation principles such as high urgency allocation [17, 19, 21]. We could not observe similarly high mortality rates following hepatic artery thrombosis (15% in our patient collective). Multivariate analysis of our cohort revealed prior liver transplantation as independent risk factor for the onset of vascular complications requiring intervention. A possible explanation could be that the probability of vascular complications increases with each reoperation due to adhesions or the need for aortal anastomosis, as described by Sanada et al. [19] based on first, second, and third liver transplants. Some other authors have also stated reoperations as a risk factor for the development of hepatic artery thrombosis [22, 23].

Our multivariate analyses further revealed primary aortal anastomosis and the need for venous interposition as independent risk factor for the development of vascular complications most likely due to the more demanding surgical technique of vascular reconstruction and more anatomical variations as already reported by others [19]. Other risk factors described in the literature, such as split grafts or living donation, could not be confirmed by our analysis. Our analysis showed that the recipient liver weight represents a protective factor regarding vascular complications, which might be because the vessels of the recipients are correspondingly larger making vascular reconstruction more feasible. Kim et al. showed that the donor-recipient weight ratio represents a risk factor for the development of hepatic artery thrombosis following LTx in children under 25 kg [24]. Apart from the above-mentioned recipient liver weight, we did not make similar observations to Kim et al., according to whom donor-recipient mismatch represents a risk factor for the onset of perioperative complications in our patient collective, indicating a precise and rigorous donor selection in our center.

4.3 Biliary Complications

The rate of overall biliary complications in our analysis was 20.0%, mainly caused by stenosis and insufficiencies, which is in line with the reported range of 13.9%–20.0%—when including late-onset biliary complications, the incidence increases up to 47.4% [25-29]. Several risk factors (e.g., recipient age, prior Kasai operation, operation time, hepatic artery thrombosis or end-end reconstruction of the bile duct) have been described in the literature [26-28, 30-32]. Our multivariate regression revealed the reconstruction of the bile duct via end-end anastomosis as only independent risk factor. Despite the advantages of an end-to-end anastomosis of the bile duct, such as lower risk of ascending cholangitis, surgical feasibility and better endoscopic access, the superiority of biliodigestive anastomoses has been described in numerous publications, most notable reason being lower rates of stenosis [26-28, 30, 31]. In line with Laurence et al., we were not able to confirm living donation and the use of split grafts as risk factors for biliary complications, opposing previous publication [27, 29]. Our multivariate analysis showed that previous intervention such as PTCD or stenting of the bile duct are a protective factor against biliary complications. Our data reveal that—if feasible—preoperative intervention with the aim of optimizing bile drainage should be considered to prevent postoperative biliary complications. To the best of our knowledge, interventions such as stenting of the bile duct or PTCD placement are only described in the literature as complication management after liver transplantation [31, 33]. The protective effect shown in our multivariable analysis has not yet been described in the context of (pediatric) liver transplantation.

In our collective, a two-stage procedure with a planned second operation for biliary reconstruction was chosen in more than one-third of cases (n = 154; 36.6%) with planned reoperation for bile duct anastomosis after a median of 2 days. Multivariate analysis revealed the days between both operations as independent protective factor. According to the current state of knowledge, there is one prospective study of an adult patient collective that also recommends a two-staged procedure for high-risk patients showing promising results regarding the two-stage procedure [34]. In pediatric liver transplant setting, a second-stage procedure for biliary reconstruction is only described as a bailout procedure in a case report of a seven-year-old patient [35]. Our analysis indicates that a two-stage procedure for reconstruction of the bile duct can be considered preventive with regard to biliary complications in our patient collective. Whether this can be applied to all patient groups without exception requires further subgroup and multicenter analyses on larger patient populations. It would be advisable to choose this approach for recipients with a particularly high risk of vascular or biliary complications (e.g., due to a highly demanding vascular or biliary reconstruction in case of size-mismatch or anatomical peculiarities) or for recipients in whom primary abdominal closure is not possible due to other causes (e.g., intestinal swelling, impaired graft perfusion).

4.4 Retransplantation and Patient Survival

We observed retransplantation rates of 9.5% within the first 30 days and 11.4% within the first year after LTx, which is comparable to an incidence of 11.0% described in the SPLIT database [36]. Our multivariable analysis revealed arterial and portal or central venous complications as independent risk factors, whereas male gender of the donor and the days until biliary reconstruction could be identified as protective factors for retransplantation being in line with the current literature [17, 37-39]. The 1-year mortality rate in our patient collective was 4.8%, which is relatively low compared to other survival rates reported in the ELTR database, most likely due to the more recent observation period [40]. Our multivariate regression revealed the donor BSA, tumor as underlying disease, encephalopathy, prior liver transplantation, and arterial complications as independent risk factors for patient death, which apart from donor BSA has been described in previous publications [41-44]. Pleural effusions, the most common complication, did not affect graft- or patient survival in our analysis. Contrary to the results shown by de Ville de Goyet et al. [45], we could not observe a significant decrease of patient survival depending on the transplant era, most likely due to a significantly shorter observation period.

The limitations of our study lie in its monocentric retrospective nature, the lack of long-term results past 1 year, the lack of preoperative laboratory values, and consequently the missing PELD score and the heterogeneity of the patient collective due to different age groups (0–17 years) and various underlying diseases. To put this into perspective, it must be added that the correlation of the PELD score with the postoperative outcome is very variable in different studies [46-48]. One reason for the variable observations could be that the PELD score, originally developed to predict 90-day waiting list mortality, is highly dependent on country-specific allocation procedures, the proportion of living donors, HU organ allocation, and the local variation in organ shortage and is therefore a non-specific predictor of post-transplant outcome.