Identifying a targeted population at high risk for infections after liver transplantation in the MELD era

Sun H‐Y, Cacciarelli TV, Singh N. Identifying a targeted population at high risk for infections after liver transplantation in the MELD era.
Clin Transplant 2011: 25: 420–425. © 2010 John Wiley & Sons A/S. Abstract: Impact of model for end‐stage liver disease (MELD) scoring system on post‐transplant infections and associated risk factors are unknown. Infections <90 d post‐transplant were assessed in 277 consecutive liver transplant recipients from 1999 to 2008. “High‐risk” factors for infections were pre‐defined as MELD score >30, ICU stay >48 h prior to transplant, intraoperative transfusion ≥15 units, retransplantation, post‐transplant dialysis, or reoperation. Of the 240 recipients in the MELD era (2002–2008), 48.5% had any high‐risk factor. The OR for infection was 1.69, 2.00, 18.00, and 4.50 in recipients with any 1, 2, 3, and ≥4 high‐risk factors, respectively (χ² for trend, p < 0.001). In logistic regression model, recipient age (OR 1.12, p < 0.05) and any high‐risk factor (OR 2.42, p < 0.05) were associated with infections. Compared with 37 pre‐MELD recipients, the overall infections and mortality at 12 months did not differ in the two eras. In Cox regression model, recipient age (OR 1.09, p < 0.05) and any high‐risk factor (OR 2.42, p < 0.05) remained associated with infections. The overall frequency of infections did not increase in the MELD era. Pre‐defined risk factors accurately predicted the risk of infections in these patients.     

Dynamic changes in MELD score not only predict survival on the waiting list but also overall survival after liver transplantation

The predictive value of MELD score for post‐transplant survival has been under constant debate since its implementation in 2001. Aim of this study was to assess the impact of alterations in MELD score throughout waiting time (WT) on post‐transplant survival. A single‐centre retrospective analysis of 1125 consecutive patients listed for liver transplantation between 1997 and 2009 was performed. The impact of MELD score and dynamic changes in MELD score (DeltaMELD), as well as age, sex, year of listing and WT were evaluated on waiting list mortality and post‐transplant survival. In this cohort, 539 (60%) patients were transplanted, 223 (25%) died on list and 142 (15%) were removed from the waiting list during WT. One‐, three‐ and five‐year survival after liver transplantation were 83%, 78% and 76% respectively. DeltaMELD as a continuous variable proved to be the only significant risk factor for overall survival after liver transplantation (hazard ratio (HR): 1.06, 95% confidence interval (CI) 1.02–1.1, P = 0.013). The highest risk of post‐transplant death could be defined for patients with a DeltaMELD > 10 (HR: 4.87, 95% CI 2.09–11.35, P < 0.0001). In addition, DeltaMELD as well as MELD at listing showed a significant impact on waiting list mortality. DeltaMELD may provide an easy evaluation tool to identify patients on the liver transplant waiting list with a high mortality risk after transplantation in the current setting. Temporarily withholding and re‐evaluating these patients might improve overall outcome after liver transplantation.     

术后首次MELD评分及其衍生评分对肝衰竭患者肝移植术后早期生存率的预测价值

目的  探讨术后首次终末期肝病模型(MELD)评分及其衍生评分MELD联合血清钠(MELD-Na)评分、MELD联合血乳酸(MELD-Lac)评分对于肝衰竭患者肝移植术后早期生存率的预测能力。方法  回顾性分析135例肝衰竭肝移植受者的临床资料,根据术后28 d的生存情况分为早期生存组(110例)和早期死亡组(25例),比较两组患者的临床资料,采用受试者工作特征(ROC)曲线确定MELD评分、MELD-Na评分与MELD-Lac评分对肝衰竭患者肝移植术后早期生存率预测的最佳截取值,以评价不同评分预测肝衰竭患者肝移植术后早期生存率的能力。结果  两组患者术后首次MELD评分、MELD-Na评分、MELD-Lac评分比较,差异均有统计学意义(均为P < 0.05)。术后首次MELD评分、MELD-Na评分、MELD-Lac评分预测肝衰竭患者肝移植术后早期生存率的AUC分别为0.653 [95%可信区间(CI) 0.515~0.792]、0.648(95% CI 0.514~0.781)、0.809 (95% CI 0.718~0.900),最佳截取值分别为18.09、18.09、19.97,约登指数分别为0.398、0.380、0.525,灵敏度分别为0.680、0.680、0.840,特异度分别为0.720、0.700、0.690。MELD-Lac评分预测肝衰竭患者肝移植术后早期生存率的AUC大于MELD评分和MELD-Na评分,差异均有统计学意义(均为P < 0.05)。结论  术后首次MELD评分及MELD-Na评分对于肝衰竭患者肝移植术后早期生存率预测能力一般,而术后首次MELD-Lac评分是肝衰竭患者肝移植术后早期生存率更为可靠的预测指标。     

The introduction of MELD-based organ allocation impacts 3-month survival after liver transplantation by influencing pretransplant patient characteristics

Introduction of the model of end-stage liver disease (MELD) for organ allocation has changed the waiting-list management. Despite reports of unaffected survival after orthotopic liver transplantation (OLT) in the MELD era, survival rates have decreased in our center. The aim of this study was to identify factors contributing to reduced survival. Three-month survival, recipient and graft parameters of all 323 OLT between 2004 and 2008, which fall into a pre- ( N  = 220) and a post-MELD ( n  = 103) era, were analysed by Kaplan–Meier-, Mann–Whitney- and Fisher tests. After the introduction of MELD, mean scores at OLT increased (14.8 vs. 18.6, P  = 0.002). The main indications for OLT were not statistically different between eras. Post-MELD recipients were older (47.9 vs. 50.9 years, P  = 0.025), donors younger (NS), cold ischemia time shorter (696 vs. 635 min., P  = 0.001), and duration of surgery longer (218 vs. 245 min., P  = 0.001). Procedure time significantly correlated with MELD and international normalized ratio (INR). Three-month survival dropped (from 88.6% to 79.6%, P  = 0.03). Independent variables of survival were creatinine, urea and duration of surgery. Reduced 3-month survival was associated with longer surgery duration, higher creatinine and urea likely reflecting higher recipient morbidity. Survival probability should be incorporated into MELD-based graft allocation.     

MELD评分对肝移植手术患者预后的预测价值

目的探讨术前终末期肝病模型（model for end－stage liver disease,MELD）评分在肝移植治疗终末期肝病早期预后中的预测价值。方法回顾106例终末期肝病患者行肝移植治疗的临床资料,计算术前MELD评分,根据并发症、死亡检验ROC曲线中最佳曲线Youden指数最高时的MELD截断值进行分组,并对各组早期并发症发生率和生存率结果进行分析。结果本组106例肝移植患者中各种严重并发症发生率为29．25％,住院28d和术后3个月生存率分别为90．57％和89．62％;非并发症组、并发症组以及生存组、死亡组的MELD评分均值分别为12．00、21．19和13.28、28．27,其MELD分值差异有统计学意义（P〈0．01）：评价并发症的ROC曲线下面积为0．24±0．05（P〈0．01）,死亡检验ROC曲线下面积为0．87±0．06（P〈0．01）,死亡检验ROC曲线Youden指数最高时的MELD截断值分别为18.42和27．15;与MELD≤18．42组相比,18．42—27．15组和≥27．15组两组的并发症发生率、死亡率均显著增加（P〈0．01）。结论终末期肝病患者术前MELD评分分值越高,肝移植后早期严重并发症发生率和死亡率越高：MELD分值对行肝移植术的患者发生严重并发症的预测效果较差,但对死亡的预测效果较好：高分值MELD（≥27．15）是预测肝移植患者术后高并发症发生与死亡的较好指标。     

The association of pre-operative anaemia with survival after orthotopic liver transplantation

Anaemia is common in patients with end-stage liver disease. Pre-operative anaemia is associated with greater mortality after major surgery. We analysed the association of pre-operative anaemia (World Health Organization classification) with survival and complications after orthotopic liver transplantation using Cox and logistic regression models. We included patients undergoing their first orthotopic liver transplantation between 2004 and 2016. Out of 599 included patients, 455 (76%) were anaemic before transplantation. Pre-operative anaemia was not associated with the survival of 485/599 (81%) patients to 1 year after liver transplantation, OR (95%CI) 1.04 (0.64–1.68), p = 0.88. Pre-operative anaemia was associated with higher rates of intra-operative blood transfusions and acute postoperative kidney injury on multivariable analysis, OR (95%CI) 1.70 (0.82–2.59) and 1.72 (1.11–2.67), respectively, p < 0.001 for both. Postoperative renal replacement therapy was associated with pre-operative anaemia on univariate analysis, OR (95%CI) 1.87 (1.11–3.15), p = 0.018.     

Predictors of death on the waiting list for liver transplantation characterized by a long waiting time

The number of patients dying while on the liver transplantation (LT) waiting list (WL) has continued to increase in recent years as a result of severe shortage of organs. Therefore, it is important to evaluate the existing models that predict death on the WL and to determine the independent predictors of death. The study cohort comprised 152 adult patients listed for LT in our centre over a period of 2 years (January 2001 to January 2003). The 12-month survival rate has been calculated by Kaplan-Meier method. The survival analysis performed by Cox proportional hazard model has evaluated the three parameters which compose the model for end-stage liver disease (MELD) score. Forty-four patients (28.9%) died while listed for LT. The survival rate was 92% at 3 months, 80% at 6 months and 69% at 12 months. Median survival was not reached. MELD score was found to be an excellent predictor of death at 12 months on our WL--c-statistic (area under curve) 0.84. In our survival analysis, only international normalized (prothrombin) ratio (INR) and serum creatinine were identified as an independent predictors of death (P < 0.0001). A new simplified version of the MELD score, which does not include serum bilirubin, is proposed and its c-statistic as predictor for death on the WL at 12 months is 0.86, as good as the original MELD score, when evaluated on our list. There is a fourfold increase in mortality on our WL for LT between 3 and 12 months after the inclusion. A simplified version of the MELD score, using only serum creatinine and INR might be taken into account when predicting 12 months mortality on WL with longer waiting time, but it has to be confirmed by other prospective studies.     

The trials and tribulations of liver allocation

Allocation policies are necessary to ensure a fair distribution of a scarce resource. The goal of any liver transplant allocation policy is to achieve the best possible outcomes for the waiting list population, irrespective of the indication for transplant, whilst maximizing organ utilization. Organ allocation for liver transplantation has evolved from simple centre-based approaches driven by local issues, to complex, evidence-based algorithm prioritizing according to need. Despite the rapid evolution of allocation policies, there remain a number of challenges and new approaches are required to ensure transparency and equity on the decision-making process and the best possible outcomes for patients on the waiting list. New ways of modelling, together with novel outcome criteria, will be required to enable a dynamic adaptability of the allocation policies to the ever changing demographics of the donor population and the changing landscape of indications for transplantation.     

APACHEII模式联合MELD评分在预测活体肝移植围手术期预后中的作用

目的探讨APACHE(急性生理学和慢性健康评分)II模式联合MELD(终末期肝病模型)评分如何准确地评估活体肝移植围手术期预后。方法总结2006年6月至2009年5月在上海交通大学附属瑞金医院行活体肝移植术38例病人临床资料。结果围手术期存活组与死亡组病人的APACHE II分值分别为13.03±3.47和23.67±3.27;死亡风险度分别为(7.05±3.70)%和(25.07±9.34)%。两组病人的APACHE II分值、死亡风险度差异具统计学意义(P<0.001)。排除外科因素后APACHE II模式对预后的评估具有更好的准确性。MELD>25分与MELD<25分的病人预期病死率分别为(7.10±3.84)%和(15.11±11.93)%,差异具统计学意义(P<0.05)。APACHE II评分和MELD评分的接受者操作特征曲线(ROC)界值分别为20分和25分。结论应用APACHE模式对活体肝移植进行评估时应注意避免外科因素干扰;校正后APACHE II模式预测准确性更佳;APACHE II>20分或MELD>25分的病人预期病死率则显著增高。     

Model for end‐stage liver disease‐sodium and survival benefit in liver transplantation

There are currently no studies calculating the survival benefit of liver transplantation (LT) according to model for end‐stage liver disease‐sodium (MELD‐Na) and based on the competing risk (CR) method. We enrolled consecutive adult patients with chronic end‐stage liver disease entering the waiting list (WL) for primary LT (WL group = 337) and undergoing LT (LT group = 220) in the period 2006–2009. Two independent multivariable regressions (WL and LT models) were created to measure the prognostic power of MELD‐Na with respect to MELD. For the WL model, both Cox and CR multivariable analyses were performed. Estimates were finally included in a Markov model to calculate 3‐year survival benefit. WL Cox model: MELD‐Na (P < 0.0001) and MELD (P < 0.0001) significantly predicted survival. WL CR model: MELD‐Na (P = 0.0045) and MELD (P = 0.0109) significantly predicted survival. LT Cox model: MELD‐Na (P = 0.7608) and MELD score (P = 0.9413) had not correlation with survival. Benefit model: MELD and MELD‐Na had an overlapping significant impact on 3‐year survival benefit; CR method determined a significant decrease in 3‐year life expectancy (LE) estimations. MELD‐Na and MELD scores similarly predicted 3‐year LT survival benefit, but the gain in LE is significantly lower when a CR method is adopted.     

MELD as a metric for survival benefit of liver transplantation

Currently, there is debate among the liver transplant community regarding the most appropriate mechanism for organ allocation: urgency‐based (MELD) versus utility‐based (survival benefit). We hypothesize that MELD and survival benefit are closely associated, and therefore, our current MELD‐based allocation already reflects utility‐based allocation. We used generalized gamma parametric models to quantify survival benefit of LT across MELD categories among 74 196 adult liver‐only active candidates between 2006 and 2016 in the United States. We calculated time ratios (TR) of relative life expectancy with transplantation versus without and calculated expected life years gained after LT. LT extended life expectancy (TR > 1) for patients with MELD > 10. The highest MELD was associated with the longest relative life expectancy (TR = _1.051.20_1.37 for MELD 11‐15, _2.292.49_2.70 for MELD 16‐20, _5.305.72_6.16 for MELD 21‐25, _15.1216.35_17.67 for MELD 26‐30; _39.2643.21_47.55 for MELD 31‐34; _120.04128.25_137.02 for MELD 35‐40). As a result, candidates with the highest MELD gained the most life years after LT: 0.2, 1.5, 3.5, 5.8, 6.9, 7.2 years for MELD 11‐15, 16‐20, 21‐25, 26‐30, 31‐34, 35‐40, respectively. Therefore, prioritizing candidates by MELD remains a simple, effective strategy for prioritizing candidates with a higher transplant survival benefit over those with lower survival benefit.     

肝移植预后的影响因素和预测方法的临床进展

目的了解影响肝移植预后的因素以及可以预测肝移植预后方法的研究进展,为肝源的分配以及肝移植围术期的治疗提供指导和参考。方法检索PubMed、CNKI、万方等数据库中关于肝移植预后的影响因素以及预测其预后方法研究的相关文献并对此进行综述和分析总结。结果肝移植作为目前治疗终末期肝病的有效方法,影响肝移植预后的因素主要包括内环境的改变、全身炎症反应以及一般全身情况。在终末期肝病模型基础上结合血钠离子、乳酸、肌肉量以及网织红细胞计数和血红蛋白浓度建立的新预测模型提高了对肝移植预后的预测能力。结论结合影响肝移植预后的因素选择更有针对性的预测肝移植预后模型可能能更准确地预测肝移植患者预后,为围术期管理和治疗提供参考,使有限的肝源发挥最大的价值而挽救更多的生命。     

End-stage renal disease in liver transplants

Renal dysfunction is one of the most significant problems following orthotopic liver transplantation (OLTx). Since the major risk factor for delayed renal dysfunction following OLTx is presumed to be cyclosporine (CsA) nephrotoxicity, it has been suggested that CsA is the most probably cause of end-stage renal disease (ESRD) in this population of patients. To test this hypothesis the records of OLTx patients in our center who developed ESRD requiring dialysis were reviewed. There were 132 consecutive adult patients with end-stage liver disease (ESLD) who received 146 OLTxs between 1990 and 2000. Five patients (3.4%) developed ESRD requiring dialysis. Four of the five patients developed nephrotic range proteinuria prior to reaching ESRD. Renal biopsy in four patients showed focal segmental glomerulosclerosis, diabetic nephropathy, membranous nephropathy and cyclosporine toxicity. The underlying hepatic and metabolic disease may have played a role in the genesis of glomerular diseases in these OLTx patients. Perhaps if more renal biopsies are performed in OLTx patients with chronic renal failure, we might discover that, although CsA/tacrolimus therapy is a definite risk factor for post-transplantation chronic renal failure, other disease processes may also play a significant role.     

终末期肝病模型评分较高的良性终末期肝病患者肝移植疗效评价

目的 探讨终末期肝病模型(MELD)评分较高的良性终末期肝病患者的肝移植疗效.方法 回顾分析80例良性终末期肝病肝移植患者的资料,根据MELD评分的不同将患者分成两组,MELD评分≥30分的23例为高MELD评分组,MELD评分＜30分的57例为低MELD评分组.分别比较两组患者手术时间、术中无肝期、术中血液制品输入量、术后重症监护病房(ICU)治疗时间和受者1年存活率,同时比较死亡患者和存活患者的临床资料,寻找导致术后死亡的危险因素.结果 高MELD评分组的手术时间、术中血液制品输入量、ICU治疗时间以及术后3个月内的死亡率明显高于低MELD评分组,差异有统计学意义(P＜0.05),而术中无肝期和患者1年存活率,两组间的差异无统计学意义(P＞0.05).死亡者和存活者相比较,MELD评分的差异无统计学意义(P＞0.05),而术前机械通气、血清钠水平、持续性肝性脑病(重型)等方面的差异有统计学意义(P＜0.05).结论 对于良性终末期肝病患者,单纯依靠MELD评分不足以准确判断患者肝移植术后的生存状态,高MELD评分者也可获得较好的肝移植结果,术前严重的低钠血症、重度肝性脑病以及机械通气是除MELD评分以外影响患者术后生存状况的危险因素.     

MELD评分系统在肝移植中的应用和意义

目的 讨论终末期肝病模型（MELD）的产生与发展，评价对肝移植的影响。方法回顾性分析MELD在肝移植应用中的有关文献。结果MELD广泛应用于预测和评定终末期肝病的严重程度及患者等待肝移植期间死亡危险度，以决定器官分配的优先顺序。结论MELD为新的评分系统，可减少患者等待肝移植的时间，客观地、精确地预测终末期肝病患者的短期生存率和死亡危险度，是较为理想的器官分配评分系统。     

乳酸浓度和终末期肝病模型评分对肝移植术后早期死亡率预测准确性的比较

目的比较乳酸浓度与终末期肝病模型(model for end-stage liver disease, MELD)预测肝移植术后早期死亡率的准确性,为临床提供一种简便及时的预测工具。方法回顾性分析2017年于本院接受同种异体肝移植手术的121例患者的临床资料,男92例,女29例,年龄25～78岁,ASAⅢ或Ⅳ级。按术后30 d内是否存活分为两组:生存组和死亡组。收集术前和术毕乳酸浓度,计算术前和术毕MELD评分。比较两组术前乳酸浓度、术毕乳酸浓度、术前MELD评分和术毕MELD评分。采用受试者工作特征(receiver operating characteristic, ROC)曲线比较术前和术毕乳酸浓度和MELD评分对肝移植患者术后早期(术后30 d)死亡率的预测准确性。结果术后30 d内存活109例(90.0%),死亡12例(10.0%)。死亡组术前MELD评分、术前乳酸浓度、术毕MELD评分和术毕乳酸浓度均明显高于生存组(P0.05)。术前乳酸浓度的ROC曲线下面积(area under the curve, AUC)为0.78(95%CI 0.63～0.93),临界值为2.43 mmol/L;术前MELD评分的AUC为0.70(95%CI 0.53～0.87),临界值为24.50分,两者AUC差异无统计学意义。术毕乳酸浓度的AUC为0.85(95%CI 0.70～0.99),临界值为9.57 mmol/L;术毕MELD评分的AUC为0.74(95%CI 0.61～0.88),临界值为25.42分;术毕乳酸浓度的AUC明显高于术毕MELD评分(P0.05)。结论乳酸浓度,尤其是术毕乳酸浓度对于肝移植术后早期死亡率的预测能力优于MELD评分。     

Reduced Priority MELD Score for Hepatocellular Carcinoma Does Not Adversely Impact Candidate Survival Awaiting Liver Transplantation

The liver organ allocation policy of the United Network for Organ Sharing (UNOS) is based on the model for end-stage liver disease (MELD). The policy provides additional priority for candidates with hepatocellular carcinoma (HCC) who are awaiting deceased donor liver transplantation (DDLT). However, this priority was reduced on February 27, 2003 to a MELD of 20 for stage T1 and of 24 for stage T2 HCC. The aim of this study was to determine the impact of reduced priority on HCC candidate survival while on the waiting list. The UNOS database was reviewed for all HCC candidates listed after February 27, 2002, The HCC candidates were grouped into two time periods: MELD 1 (listed between February 27, 2002, and February 26, 2003) and MELD 2 (listed between February 27, 2003 and February 26, 2004). For the two time periods, the national DDLT incidence rates for HCC patients were 1.44 versus 1.53 DDLT per person-year (p = NS) and the waiting times were similar for the two periods (138.0 +/- 196.8 vs. 129.0 +/- 133.8 days; p = NS). Furthermore, the 3-, 6- and 12-month candidate, patient survival and dropout rates were also similar nationally. Regional differences in rates of DDLT for HCC were observed during both MELD periods. Consequently, the reduced MELD score for stage T1 and T2 HCC candidates awaiting DDLT has not had an impact nationally either on their survival on the waiting list or on their ability to obtain a liver transplant within a reasonable time frame. However, regional variations point to the need for reform in how organs are allocated for HCC at the regional level.     

血浆置换在重型肝炎肝移植术前应用的临床观察

目的观察和评价术前血浆置换对重型肝炎肝移植的作用。方法回顾性分析连续35例接受原位肝移植的重型肝炎病人的临床资料,并按术前是否行血浆置换分为术前血浆置换组(PE组)19例和术前无血浆置换组(对照组)16例。观察PE组血浆置换前后临床生化指标及内毒素水平的变化;比较两组手术时间、出血量、输血量和应用血制品情况;对两组病人术后恢复过程、并发症生存率进行比较。结果PE组患者血浆置换后,肝功能生化指标明显好转,内毒素水平及血氨较置换前显著下降,凝血酶原时间、凝血酶原国际标化比值较置换前明显好转(P<0.01)。PE组患者手术时间明显缩短,出血量和输血量少于对照组。血浆置换组病人ICU住院时间短、胃肠功能恢复早,与对照组相比差异有显著性。血浆置换组病人呼吸道感染发生率、胆道并发症少于对照组,但差异无显著性;术后半年生存率血浆置换组为78.9%,对照组为56.3%,统计学也无显著性差异(P=0.273)。而死亡病例终末期肝病模型(MELD)评分显著高于生存组病例(P<0.01)。结论重型肝炎患者肝移植术前行血浆置换治疗能改善术前肝功能、凝血功能和减少术中出血。但血浆置换并不能降低围手术期并发症和提高存活率。     

Serum cholestasis markers as predictors of early outcome after liver transplantation

BACKGROUND: Early cholestasis is not uncommon after liver transplantation and usually signifies graft dysfunction. The aim of this study was to determine if serum synthetic and cholestatic parameters measured at various time points after transplantation can predict early patient outcome, and graft function. METHODS: The charts of 92 patients who underwent 95 liver transplantations at Rabin Medical Center between 1991 and 2000 were reviewed. Findings on liver function tests and levels of serum bilirubin, alkaline phosphatase (ALP), and gamma glutamyl transpeptidase (GGT) on days 2, 10, 30, and 90 after transplantation were measured in order to predict early (6 months) patient outcome (mortality and sepsis) and initial poor functioning graft. Pearson correlation, chi(2) test, and Student's t-test were performed for univariate analysis, and logistic regression for multivariate analysis. RESULTS: Univariate analysis. Serum bilirubin >/=10 mg/dL and international normalized ratio (INR) >1.6 on days 10, 30, and 90, and high serum ALP and low albumin levels on days 30 and 90 were risk factors for 6-month mortality; serum bilirubin >/=10 mg/dL on days 10, 30, and 90, high serum ALP, high GGT, and low serum albumin, on days 30 and 90, and INR >/=1.6 on day 10 were risk factors for sepsis; high serum alanine aminotransferase, INR >1.6, and bilirubin >/=10 mg/dL on days 2 and 10 were risk factors for poor graft function. The 6-month mortality rate was significantly higher in patients with serum bilirubin >/=10 mg/dL on day 10 than in patients with values of <10 mg/dL (29.4% vs. 4.0%, p = 0.004). Patients who had sepsis had high mean serum ALP levels on day 30 than patients who did not (364.5 +/- 229.9 U/L vs. 70.8 +/- 125.6 U/L, p = 0.005). Multivariate analysis. Significant predictors of 6-month mortality were serum bilirubin >/=10 mg/dL [odds ratio (OR) 9.05, 95% confidence intervals (CI) 1.6-49.6] and INR >1.6 (OR 9.11, CI 1.5-54.8) on day 10; significant predictors were high serum ALP level on day 30 (OR 1.005, 1.001-1.01) and high GGT level on day 90 (OR 1.005, CI 1.001-1.01). None of the variables were able to predict initial poor graft functioning. CONCLUSIONS: Several serum cholestasis markers may serve as predictors of early outcome of liver transplantation. The strongest correlation was found between serum bilirubin >/=10 mg/dL on day 10 and early death, sepsis, and poor graft function. Early intervention in patients found to be at high risk may ameliorate the high morbidity and mortality associated with early cholestasis.