Determination of the optimal model for end-stage liver disease score in patients with small hepatocellular carcinoma undergoing loco-regional therapy

A combination of models for end-stage liver disease and cirrhosis-related complications to predict the prognosis of liver cirrhosis

BACKGROUND

The Child-Pugh score, the model for end-stage liver disease (MELD) score, and the occurrence of cirrhosis-related complications are independent prognostic predictors used in the assessment of chronic liver diseases.

OBJECTIVES

The objectives of this study were to determine the best prognostic scoring system, and to create a combined method to predict the prognosis of liver cirrhosis more accurately.

METHODS

We retrospectively reviewed 435 cirrhotic patients from January 2009 to June 2010 and evaluated their short- and medium-term survival. Child-Pugh, MELD and its advanced scoring systems were computed for each patient. The sensitivity and specificity of these scoring systems were analyzed and their validity was assessed using concordance (c)-statistics in predicting the prognosis of cirrhotic patients.

RESULTS

Overall, 107 patients died within 6 months and 150 patients died within 1 year. The clinical and biochemical characteristics, cirrhosis-related complications, and the scores were significantly different among the survivors and patients who died. The largest area under the receiver operating characteristic curve was 0.741 for the integrated MELD (iMELD) at 6 months and 0.713 for iMELD at 12 months, indicating that iMELD was the best scoring system tested. Given this result, we created a new scoring system that combined iMELD and an index of cirrhosis-related complications, called iMELD-C. This novel system had c indexes of 0.758 for the 6-month survival and 0.746 for the 1-year survival.

CONCLUSIONS

The iMELD-C score is a better predictor of both short- and medium-term survival in patients with cirrhosis.

The MELD score in patients awaiting liver transplant: strengths and weaknesses

Adoption of the Model for End-stage Liver Disease (MELD) to select and prioritize patients for liver transplantation represented a turning point in organ allocation. Prioritization of transplant recipients switched from time accrued on the waiting list to the principle of "sickest first". The MELD score incorporates three simple laboratory parameters (serum creatinine and bilirubin, and INR for prothrombin time) and stratifies patients according to their disease severity in an objective and continuous ranking scale. Concordance statistics have demonstrated its high accuracy in stratifying patients according to their risk of dying in the short-term (three months). Further validations of MELD as a predictor of survival at various temporal end-points have been obtained in independent patient cohorts with a broad spectrum of chronic liver disease. The MELD-based liver graft allocation policy has led to a reduction in waitlist new registrations and mortality, shorter waiting times, and an increase in transplants, without altering overall graft and patient survival rates after transplantation. MELD limitations are related either to the inter-laboratory variability of the parameters included in the score, or to the inability of the formula to predict mortality accurately in specific settings. For some conditions, such as hepatocellular carcinoma, widely accepted MELD corrections have been devised. For others, such as persistent ascites and hyponatremia, attempts to improve MELD's predicting power are currently underway, but await definite validation.

Allocation priority in non-urgent liver transplantation: An overview of proposed scoring systems

Given the lack of donors, a correct organ allocation system for candidates to liver transplantation is essential to increase graft and patient survival. The most used organ allocation tools are Child-Turcotte-Pugh and model for end-stage liver disease. It is generally accepted that model for end-stage liver disease score is superior to the Child-Turcotte-Pugh classification in predicting the short-term survival of cirrhotic patients awaiting liver transplantation. Since 2002, model for end-stage liver disease is widely used for liver allocation. In recent years, to overcome limitations of the consolidated scores, some adjustments to the original model for end-stage liver disease formula and new scoring systems have been proposed. Published data suggest that integrating serum sodium and model for end-stage liver disease may improve the score prognostic accuracy but further studies are necessary to confirm this issue. The updated model for end-stage liver disease, obtained through a revision of traditional model for end-stage liver disease parameters and tested in a large cohort of patients, is of great interest at the moment. In conclusion, several scoring systems have been described for organ allocation, but today, none is definitely able to overcome the limitations of the Child-Turcotte-Pugh and model for end-stage liver disease systems.

Preliminary results of liver transplantation for hepatocellular carcinoma among allocation organ policy strategies, neoadjuvant treatments, and intention-to-treat analysis

We retrospectively evaluated the impact of our strategy for patients with hepatocellular carcinoma (HCC) according to an intention-to-treat analysis and drop-out probability. We evaluated only patients within the Milan criteria. We analyzed the outcomes of neoadjuvant strategies for HCC, organ allocation policy, and systematic application of strategies to increase the deceased donor pool as the current tendency to expand transplantability criteria for those patients. Kaplan-Meier survival probability rates at 1, 3, and 5 years according to an intention-to-treat analysis were 87.02%, 74.53%, and 65.93% for transplanted patients (n=108), and 50%, 14.29%, and 14.29% for the excluded or waiting list group (n=13), respectively (P< .0001). Drop-out risk at 3, 6, and 12 months was 2.40%, 8.59%, and 16.54%, respectively. During the same period, the mortality probability rates at 3, 6, and 12 months among patients without HCC awaiting orthotopic liver transplantation (OLT) were 3.60%, 9.50%, and 18.34%, respectively. Drop-out rate was lower among patients treated before OLT (P< .0001). On the basis of the neoadjuvant treatment results to reduce drop-out risk, we suggest avoiding the high priority for the HCC cohort, particularly within the first 6 months from entrance on the waiting list, because this approach can reduce the chances of patients with end-stage liver disease (ESLD) alone.

Selecting an optimal prognostic system for liver cirrhosis: the model for end-stage liver disease and beyond

In comparison with the Child-Turcotte-Pugh (CTP) system, recent studies suggested that the model for end-stage liver disease (MELD) may more accurately predict the survival for patients with cirrhosis. In the US, the liver allocation system was changed in 2002 from a status-based algorithm utilizing CTP scores to one using continuous MELD severity scores as a reference system in prioritizing adult patients on the waiting list. Direct evidence that demonstrates the benefits of MELD is the fact that the mortality rates of transplant candidates on the waiting list have remarkably decreased after the implementation of the MELD. The MELD score is closely associated with the degree of portal hypertension as reflected by the hepatic venous pressure gradient. Hyponatraemia occurs as a result of advanced cirrhosis, and a serum sodium (Na) level <126 mEq/L at the time of listing for transplantation is a strong independent predictor of mortality. Several MELD-derived prognostic models that incorporate serum Na into calculation have been proposed in the hopes of further improving the MELD's prognostic accuracy. Additionally, serum parameters such as creatinine and international normalized ratio are subject to interlaboratory variations and may need unifying standartisations. Patients with refractory complications of cirrhosis may need a priority MELD score to prioritize them on the waiting list. Appropriate modifications and the fine-tuning of the MELD based on well-designed prospective studies are necessary in solving the current controversial issues.

Outcome of patients with hepatocellular carcinoma listed for liver transplantation within the Eurotransplant allocation system

Although hepatocellular carcinoma (HCC) has become a recognized indication for liver transplantation, the rules governing priority and access to the waiting list are not well defined. Patient- and tumor-related variables were evaluated in 226 patients listed primarily for HCC in Belgium, a region where the allocation system is patient-driven, priority being given to sicker patients, based on the Child-Turcotte-Pugh (CTP) score. Intention-to-treat and posttransplantation survival rates at 4 years were 56.5 and 66%, respectively, and overall HCC recurrence rate was 10%. The most significant predictors of failure to receive a transplant in due time were baseline CTP score equal to or above 9 (relative risk [RR] 4.1; confidence interval [CI]: 1.7-9.9) and alpha fetoprotein above 100 ng/mL (RR 3.0; CI: 1.2-7.1). Independent predictors of posttransplantation mortality were age equal to or above 50 years (RR 2.5; CI: 1.0-3.7) and United Network for Organ Sharing pathological tumor nodule metastasis above the Milan criteria (RR 2.1; CI: 1.0-5.9). Predictors of recurrence (10%) were alpha fetoprotein above 100 ng/mL (RR 3.2; CI:1.1-10) and vascular involvement of the tumor on the explant (RR 3.6; CI: 1.1-11.3). Assessing the value of the pretransplantation staging by imaging compared to explant pathology revealed 34% accuracy, absence of carcinoma in 8.3%, overstaging in 36.2%, and understaging in 10.4%. Allocation rules for HCC should consider not only tumor characteristics but also the degree of liver impairment. Patients older than 50 years with a stage above the Milan criteria at transplantation have a poorer prognosis after transplantation.

Validation of the HCC-MELD for dropout probability in patients with small hepatocellular carcinoma undergoing locoregional therapy

BACKGROUND

The model for end-stage liver disease (MELD) is used in prioritizing cirrhotic patients awaiting liver transplantation. Patients with small hepatocellular carcinoma (HCC) are eligible candidates. An HCC-MELD equation was recently proposed to predict the dropout rate of HCC patients on the waiting list. This study aimed to validate the accuracy of this equation.

METHODS

We investigated 390 patients with small HCC who were candidates for liver transplantation and underwent locoregional therapy.

RESULTS

The estimated probability of dropout according to the equation was 8.2% for T1 stage and 13.5% for T2 stage HCC (p < 0.0001). The actual disease progression rate at three months was 2.1% for T1 and 3.0% for T2 stage HCC. At six months, the progression rate was 5.3% for T1 stage and 6.8% for T2 stage. The area under receiver operating characteristic curve of the HCC-MELD equation was 0.81 at three months and 0.80 at six months. Patients undergoing radiofrequency ablation (RFA) had significantly lower dropout rates compared with other treatment groups according to the equation (p = 0.0007). The actual tumor progression rate was also the lowest for the RFA group at both three and six months.

CONCLUSION

The HCC-MELD equation is a feasible predictive model for patients with small HCC undergoing locoregional therapy.

Intention-to-treat analysis of liver transplantation in selected, aggressively treated HCC patients exceeding the Milan criteria

This prospective study analyzed the dropout probability and intention-to-treat survival rates of patients with hepatocellular carcinoma (HCC) selected and treated according to our policy before liver transplantation (LT), with particular attention to those exceeding the Milan criteria. Exclusion criteria for LT were macroscopic vascular invasion, metastases, and poorly differentiated disease at percutaneous biopsy. A specific multi-modal adjuvant algorithm was used to treat HCC before LT. A total of 100 HCC patients were listed for LT: 40 exceeded the Milan criteria in terms of nodule size and number (MILAN OUT) either at listing or in list, while 60 patients continued to meet the criteria (MILAN IN). The Milan criteria did not prove to be a significant predictor of dropout probability or survival rates using Cox's analysis. Cumulative dropout probability at 6 and 12 months was 0% and 4% for MILAN OUT, and 6% and 11% for MILAN IN. The intention-to-treat survival rates at 1 and 3 years were 95% and 85% in MILAN OUT, and 84% and 69% in MILAN IN. None of the 68 transplanted patients had recurrent HCC after a median 16-month follow-up (0-69 months). In conclusion, LT may be effective for selected, aggressively-treated HCC patients exceeding the Milan criteria.

Limitation of the model for end-stage liver disease for outcome prediction in patients with cirrhosis-related complications

The model for end-stage liver disease (MELD) has been used to prioritize cirrhotic patients awaiting liver transplantation. Bleeding esophageal varices, spontaneous bacterial peritonitis and hepatic encephalopathy are major complications of cirrhosis and traditional indications for liver transplantation evaluation. However, these complications are not included in the MELD and it is not clear if these complications correlate with MELD score in terms of outcome prediction. This study aimed to investigate the feasibility of cirrhosis-related complication as a prognostic predictor in 290 cirrhotic patients. The MELD score and outcome were compared between patients with and without cirrhosis-related complications. There was no significant difference of the MELD score between patients with (n = 67) and without (n = 223) complications (11.6 +/- 2.9 vs. 12.2 +/- 3.2, p = 0.184). The area under the receiver operating characteristic curve was 0.687 for MELD vs. 0.604 for complications (p = 0.174) at six months, and the area was 0.641 for MELD vs. 0.611 for complications (p = 0.522) at 12 months. A high MELD score and presence of complications had a similar profile of predictive accuracy and both were significant predictors of mortality at six and 12 months in multivariate logistic regression analysis. Patients with cirrhosis-related complications at presentation had a decreased survival compared with those without complications (p < 0.0001). In conclusion, the occurrence of cirrhosis-related complications is a predictor of poor prognosis. While early transplantation referral is recommended, these patients do not necessarily have a higher MELD score and could be down-staged in the MELD era.

A systematic review of the performance of the model for end-stage liver disease (MELD) in the setting of liver transplantation

The Model for End-Stage Liver Disease (MELD) score is now used for allocation in liver transplantation (LT) waiting lists, replacing the Child-Turcotte-Pugh (CTP) score. However, there is debate as whether it is superior to CTP score to predict mortality in patients with cirrhosis on the LT waiting list and after LT. We reviewed studies comparing the accuracy of MELD vs. CTP score in transplantation settings. We found that in studies of the LT waiting list (12,532 patients with cirrhosis), only 4 of 11 showed MELD to be superior to CTP in predicting short-term (3-month) mortality. In addition, 2 of 3 studies (n = 1,679) evaluating the changes in MELD score (DeltaMELD) showed that DeltaMELD had better prediction for mortality than the baseline MELD score. The impact of MELD on post-LT mortality was assessed in 15 studies (20,456 patients); only 6 (9,522 patients) evaluated the discriminative ability of MELD score using the concordance (c) statistic (the MELD score had always a c-statistic < 0.70). In 11 studies (19,311 patients), high MELD score indicated poor post-LT mortality for cutoff values of 24-40 points. In re-LT patients, 2 of 4 studies evaluated the discriminative ability of MELD score on post-LT mortality. Finally, several studies have shown that the predictive ability of MELD score increases by adding clinical variables (hepatic encephalopathy, ascites) or laboratory (sodium) parameters. On the basis of the current literature, MELD score does not perform better than the CTP score for patients with cirrhosis on the waiting list and cannot predict post-LT mortality.

Model for end-stage liver disease and organ allocation in liver transplantation: where are we and where should we go?

The Child-Turcotte-Pugh (CTP) score has been used for decades to measure the severity of chronic liver disease. Recent studies have shown that the model for end-stage liver disease (MELD) more accurately predicts the short- and mid-term survival for patients with cirrhosis compared to the CTP system. MELD, which has 3 parameters (serum bilirubin, creatinine, and prothrombin time) that need logarithmic transformation, has the advantage of a wide-range continuous scale and is more objective and less variable. The liver allocation system has changed from a status-based algorithm using the CTP score, to one using a continuous MELD severity score as a reference system to prioritize adult patients on the waiting list since 2002 in the USA. However, there are potential limitations of MELD. An intrinsic defect is that some important parameters, such as hepatic encephalopathy and ascites, which are common adverse complications in cirrhosis, are not included in MELD. It has been suggested to incorporate a low serum sodium level into the prognostic model to enhance the predictive ability. Moreover, the change of MELD over time may provide updated information for patients on the transplant waiting list. In summary, although there was encouraging evidence supporting the prognostic advantage of MELD, the optimal role of MELD in the setting of outcome assessment for cirrhotic patients needs more study. Appropriate modifications and fine tuning of MELD are necessary for determining the ranking status of patients on the waiting list, to avoid a futile transplantation and improve overall patient survival.

Different model for end-stage liver disease score block distributions may have a variable ability for outcome prediction

BACKGROUND

The model for end-stage liver disease (MELD) scoring system has become the prevailing criteria for organ allocation in liver transplantation. However, it is not clear if the predictive accuracy of MELD is equally homogeneous in different distribution of MELD score blocks.

METHODS

We investigated 472 cirrhotic patients (mean MELD, 14.3+/-5.5), and compared the predictive accuracy of MELD and the corresponding Child-Turcotte-Pugh (CTP) scores in patients with low (<16), intermediate (10-20) and high (>14) MELD score range by using c-statistic for area under the receiver operating characteristic curve (AUC) at different time frames.

RESULTS

The MELD scores well correlated with CTP scores at baseline (rho=0.492, P<0.001). Overall, MELD was significantly better than the CTP system to predict the risk of mortality. However, in stratified analysis there were no significant differences between MELD and CTP for the c-statistic in patients with low and intermediate range MELD scores at 3-, 6-, 9-, and 12-month (p values all > 0 1). Among patients with high MELD scores, MELD was consistently more accurate than the CTP system in predicting the mortality at 3- (AUC, 0.715 vs. 0.543, P=0.020), 6- (0.705 vs. 0.536, P=0.003), 9- (0.737 vs. 0.507, P<0.001) and 12-month (0.716 vs. 0.526, P<0.001), respectively.

CONCLUSIONS

MELD has a better performance only in a subset of patients with higher MELD scores. The outcome in patients with lower range MELD scores cannot be reliably predicted solely with their MELD scores, and alternative prognostic markers should be used in conjunction to enhance the predictive accuracy.

Proposal of a modified Child-Turcotte-Pugh scoring system and comparison with the model for end-stage liver disease for outcome prediction in patients with cirrhosis

The model for end-stage liver disease (MELD) has a better predictive accuracy for survival than the Child-Turcotte-Pugh (CTP) system and has been the primary reference for organ allocation in liver transplantation. The CTP system, with a score range of 5-15, has a ceiling effect that may compromise its predictive power. In this study, we proposed a refined CTP scoring method and investigated its predictive ability. An additional point was given to patients with serum albumin < 2.3 g/dL, bilirubin > 8 mg/dL or prothrombin time prolongation > 11 seconds. The modified CTP system, containing class D, was compared to the MELD and original CTP system in 436 patients. There was a significant correlation between the MELD and modified CTP score (rho = 0.59, P< 0.001). Using mortality as the endpoint, the area under receiver operating characteristic curve for modified CTP system was 0.895 compared with 0.872 for MELD (P = 0.450) and 0.809 for original CTP system (P < 0.001) at 3 months; the area was 0.890, 0.837 and 0.756, respectively (P = 0.051 and < 0.001, respectively) at 6 months. The risk ratio per unit increase for the modified CTP score was 2.7 and 3.08 at 3 and 6 months respectively (P < 0.001). In conclusion, the modified CTP system can be proposed as an alternative prognostic model for cirrhotic patients. By extending the score range according to the influence of the laboratory-derived variables, the modified CTP system has a better performance than the original system and is as efficient as the MELD for outcome prediction.

Evaluation of the increase in model for end-stage liver disease (DeltaMELD) score over time as a prognostic predictor in patients with advanced cirrhosis: risk factor analysis and comparison with initial MELD and Child-Turcotte-Pugh score

BACKGROUND/AIMS

The model for end-stage liver disease (MELD) has been used to prioritize cirrhotic patients awaiting liver transplantation. The change in MELD score over time (DeltaMELD) may have additional prognostic value. We investigated the ability of DeltaMELD to predict the outcome of advanced cirrhosis and prospectively assessed the factors associated with increasing DeltaMELD.

METHODS

Risk factors were determined in 58 prospectively followed-up patients. The predictive power of DeltaMELD, initial MELD and Child-Turcotte-Pugh (CTP) score was compared by using c-statistic in 351 patients.

RESULTS

Ascites (P=0.020) and hepatic encephalopathy (P=0.023) were significantly associated with increasing MELD score at 3 months. The area under receiver operating characteristic (ROC) curve for DeltaMELD/month was 0.779 compared with 0.718 for MELD (P=0.130) and 0.528 for CTP score (P<0.001) at 6 months; the area was 0.822, 0.744 and 0.528, respectively (P=0.018 and <0.001, respectively) at 12 months. DeltaMELD/month >2.5 was the only significant prognostic predictor at 6 (odds ratio: 9.8, P<0.001) and 12 months (odds ratio: 16.3, P<0.001) in multivariate logistic analysis.

CONCLUSIONS

Increasing MELD score is associated with the onset of ascites and encephalopathy. DeltaMELD is superior to initial MELD and CTP scores to predict intermediate term outcome in patients with advanced cirrhosis.