Objectives: Liver transplant is a life-saving procedure for a variety of end-stage liver diseases. Cardiovascular disorders are among the leading cause of death among patients with end-stage liver disease and those undergoing liver transplant procedures. MELD-XI score is a newly developed score for mortality prediction in patients with end-stage liver failure. In this study, we investigated the relationships among MELD-XI score, total in-hospital mortality, and coronary artery disease severity and extent among patients with end-stage liver failure who were awaiting transplant.
Materials and Methods: We retrospectively reviewed medical records of 121 patients with end-stage liver failure on transplant wait list. Study patients had undergone coronary angiography as part of pretransplant cardiac evaluation. We determined prevalence of coronary artery disease and Gensini score (which indicates extent of coronary artery disease) using coronary angiography and reviewed MELD-XI score and in-hospital mortality rates. We compared MELD-XI score and Gensini score in deceased and surviving patients and correlated both scores with mortality and with each other.
Results: Of 121 patients, 79 (65.3%) were men; mean age of the study population was 59.6 ±10.2 years. Twenty-eight patients (23.1%) had coronary artery disease, and 13 (10.7%) had severe coronary artery disease on coronary angiography. Twenty-three patients (19%) died while on the transplant wait list. Gensini score and MELD-XI scores were significantly higher in those who died (P < .05). MELD-XI score, but not Gensini score, was a significant independent predictor of death among patients awaiting liver transplant (hazard ratio = 1.35; 95% confidence interval, 1.04-1.78; P < .05).
Conclusions: MELD-XI score independently predicted in-hospital death among patients scheduled to undergo liver transplant. These patients also had increased prevalence and extent of coronary artery disease.
Key words : Coronary angiography, Gensini score, In-hospital mortality, Liver transplant candidate, Model for End-Stage Liver Disease
Liver transplant is a life-saving procedure for patients with end-stage liver disease.1 Liver transplant candidates are at increased risk of coronary artery disease (CAD) due to increased prevalence of CAD risk factors.2,3 Among these, age older than 50 years, male sex, hypertension, hyperlipidemia, diabetes mellitus, and obesity are the most prevalent ones. Presence of more than 1 risk factor (other than age) places liver transplant candidates at a moderate to severe risk of CAD.3-5 In addition, nonalcoholic hepatitis puts patients at increased risk of CAD, with severe CAD being seen in approximately 23% of these patients.6 Cardiovascular complications, particularly due to CAD, are a major cause of morbidity and mortality in patients with end-stage liver disease awaiting and undergoing liver transplant.7,8 Therefore, many patients on the liver transplant wait list undergo coronary angiography to determine the Gensini score, a validated tool to determine extent of CAD.9
The Model for End-Stage Liver Disease (MELD) score, calculated from the logarithmic conversions of serum total bilirubin and creatinine, was developed to predict clinical outcomes in patients with liver disease10,11 and is considered to be a good indicator of the severity of hepatic disorders in patients on liver transplant wait lists.12 MELD-XI is a modification of the original MELD; this tool excludes international normalized ratio and is for use in patients on anticoagulants.13 In addition to prediction of mortality in patients with liver disease, both MELD and MELD-XI scores have been shown to predict mortality in various cardiac conditions.11,12-23 Both MELD and MELD-XI scores reflect dysfunction of 2 critical organs, namely, liver and kidneys. Increased serum bilirubin has been associated with hepatocellular ischemia and hypoxia resulting from reduced cardiac output and/or increased central and, in turn, hepatic pressure.24,25 Similarly, the kidney is affected by low cardiac output and increased renal venous congestion caused by the so-called forward and backward heart failure, respectively.26 Therefore, MELD-XI score is closely related to cardiac condition and failure.27-29 Because CAD is the leading cause of cardiac dysfunction in patients with end-stage liver failure awaiting transplant, MELD-XI score may also reflect underlying CAD severity in those patients. Here, we aimed to evaluate MELD-XI score and Gensini score (used to assess CAD extent) to investigate correlations between these 2 scores in patients with liver failure awaiting transplant. We also evaluated the prognostic significances of both scores in the same population.
Materials and Methods
This retrospective study was approved by the Başkent University Institutional Review Board (Project No: KA19/309). The study included 121 in-hospital patients with end-stage/fulminant liver failure scheduled to undergo living-donor liver transplant at the Başkent University Faculty of Medicine Department of General Surgery (Ankara, Turkey). Demographic and clinical characteristics of patients were obtained from medical records.
All patients underwent cardiac examination and coronary angiography as part of pretransplant cardiac evaluations at the Başkent University Faculty of Medicine Department of Cardiology between January 2011 and January 2019. All patients were examined and tested noninvasively by the cardiology department and underwent coronary angiography per the cardiologist’s discretion on the basis of history, physical examination, and noninvasive cardiac testing. On coronary angiography, the rate of CAD (%) and the rate of severe CAD (%) were determined, with severe CAD defined as ≥ 50% stenosis in left main coronary artery or ≥ 70% stenosis in any major epicardial coronary artery or their major branches with a diameter of greater than 1.5 mm. The extent of CAD was evaluated by Gensini score on coronary angiography as described elsewhere.9
MELD-XI score was calculated from logarithmic conversions of serum total bilirubin and creatinine as follows17: 5.11 × ln(serum bilirubin in mg/dL) + 11.76 × ln(serum creatinine in mg/dL) + 9.44. In-hospital all-cause mortality rate was calculated, and mortality causes were categorized as cardiac, encephalitic, infectious, hemorrhagic, and other.
Distribution of quantitative data was tested with the Kolmogorov-Smirnov test. Descriptive statistics were reported as median (interquartile range) or number (percentage). Quantitative data were compared with Mann-Whitney U test and qualitative data with the chi-square test. Correlation analysis was done with Spearman correlation analysis. MELD-XI score and Gensini score were compared in deceased versus surviving patients, and both scores were correlated with each other. The significant predictors of in-hospital all-cause mortality were initially tested with univariate analysis using all available variables. All univariate predictors of mortality with P ≤ .2 were used in a binary logistic regression model to determine the independent predictors of in-hospital all-cause mortality. A receiver operating characteristic curve was drawn to determine the predictive power of MELD-XI score for in-hospital all-cause mortality.
Of 121 study patients, 79 (65.3%) were men and 42 (34.6%) were women. The mean age of the study population was 59.6 ± 10.2 years. Causes of liver failure were secondary to hepatitis C or B virus infection in 40 patients (33.1%), veno-occlusive disease in 3 patients (2.5%), alcoholic liver disease in 12 patients (9.9%), nonalcoholic steatohepatitis in 3 patients (2.5%), primary biliary cirrhosis in 8 patients (6.5%), autoimmune hepatitis in 5 patients (4.1%), cryptogenic in 47 patients (38.8%), Wilson disease in 2 patients (1.7%), and right-sided heart failure in 1 patient (0.8%). Patient demographic and clinical data are presented in Table 1. Of total patients, 23 (19.0%) died in the hospital. Death was caused by cardiac disorders (sudden cardiac death, myocardial infarction/ischemia, or heart failure) in 4 patients (17.4%), infectious diseases in 5 patients (21.7%), hepatic encephalopathy in 12 patients (52.2%), and gastrointestinal system or other site bleeding in 2 patients (8.7%).
The prevalences of CAD and severe CAD were higher in deceased patients, although only CAD prevalence reached statistical significance (P < .05) (Table 1). The median MELD-XI score of deceased patients was significantly higher than that shown in survivors: score of 12.70 (interquartile range [IQR] of 2.98) versus 11.00 (IQR of 2.25; P < .05). Similarly, the median Gensini score was also significantly higher than that shown in survivors: score of 4 (IQR of 26) versus 0 (IQR of 2; P < .05) (Table 1). However, MELD-XI and Gensini scores were not significantly correlated with each other (P > .05).
On univariate analysis, in-hospital mortality was significantly correlated with age, MELD-XI score, hemoglobin level, creatinine level, and severe CAD (P < .05 for all correlations) but not Gensini score (P > .05). On multivariate analysis, MELD-XI score, but not Gensini score, was an independent predictor of in-hospital mortality (hazard ratio = 1.35; 95% confidence interval, 1.04-1.78; P < .05). A receiver operating characteristic analysis showed that a MELD-XI score greater than 11.5 had a sensitivity of 69.6% and a specificity of 64.3% for the prediction of in-hospital mortality (area under the concentration curve = 0.645; P = .031) (Figure 1).
This study has some important results. First, among patients with end-stage liver failure due to various causes awaiting transplant, both Gensini score and MELD-XI scores were higher in patients who died before transplant than in those who survived. Second, although they did not reach statistical significance, the rates of CAD and severe CAD were greater among deceased patients. Third, Gensini score and MELD-XI score were not interrelated, possibly suggesting different mechanisms in predicting prognosis. Fourth, MELD-XI score but not Gensini score was a significant predictor of in-hospital mortality Cardiovascular complications are one of the leading causes of morbidity and mortality among patients undergoing liver transplant30 and warrant meticulous investigations.31-33 Among cardiovascular disorders, CAD is one of the most notable determinants of increased mortality and morbidity.34,35 Patients with end-stage liver failure are also at increased risk of death during their waiting period for liver transplant.30,36 Therefore, it is imperative to test and search for CAD and treat it accordingly in this population.
We found that patients who died before transplant had increased prevalence of cardiovascular risk factors, namely, diabetes mellitus, hypertension, smoking, family history of premature CAD, and hyperlipidemia; however, only hypertension and smoking were significantly higher among deceased versus surviving patients. This suggests that an accumulation of risk factors for CAD may explain the significantly higher CAD rate and Gensini score (an indicator of extent of CAD) in deceased patients. However, Gensini score was not a significant predictor of in-hospital death. This was not surprising, as patients with end-stage liver failure awaiting transplant have higher comorbidity rates. Many of these patients die because of other conditions such as sepsis, infections, bleeding, or hepatic encephalopathy. However, increased extent of CAD may nevertheless have contributed to worse prognosis in patients who died before transplant.
In contrast, MELD-XI score was a significant predictor of in-hospital death. Furthermore, Gensini score and MELD-XI scores were not interrelated, possibly suggesting different mechanisms in predicting prognosis. Collectively, although CAD prevalence and extent may increase in patients with end-stage liver failure, these may not reflect increased mortality but rather other operational factors. As noted above, MELD-XI score has been previously tested in patients with end-stage liver failure.13 Later, it was also tested in critically ill patients17 who had different cardiac disorders, including left ventricular assist device implantation,18 infective endocarditis,20 heart failure,22 and Fontan operation.23
MELD-XI score is believed to reflect dysfunction of 2 critical organs, namely, the liver and kidneys. Increased serum bilirubin has been associated with hepatocellular ischemia and hypoxia resulting from reduced cardiac output and/or increased central and hepatic pressure. Similarly, kidneys are affected by low cardiac output and increased renal venous congestion (by forward and backward heart failure, respectively). Therefore, MELD-XI score is closely related to cardiac condition and failure. Here, we verified the prognostic importance of MELD-XI score. In our investigation to determine whether CAD presence and extent (Gensini score) were related to MELD-XI score and mortality, we found Gensini score to be correlated with neither MELD-XI score nor mortality, although it was significantly higher in patients who died before transplant. This suggests that CAD extent is increased among patients with end-stage liver failure because of possible clustering of risk factors. It is plausible that increased MELD-XI in this population was due to noncardiac factors, such as liver failure, hepatorenal syndrome, infectious or septic complications, or a combination. Furthermore, we observed normal median left ventricular ejection fraction and tricuspid annular plane systolic excursion results for both deceased and surviving patient groups, suggesting that increased MELD-XI score was not related to reduced right or left ventricular systolic dysfunction. This means that the cardiac contribution to increased MELD-XI score may be limited in patients with end-stage liver failure, which is characterized by gravely deteriorated hepatic and renal function and poor prognosis.30,36
Our study had several limitations. First was its retrospective nature and relatively small sample size. Second, we did not investigate cardiac mortality as few patients died from cardiac causes.
Although CAD prevalence and extent were increased in in-hospital patients with end-stage liver failure who died before transplant, CAD extent was related to neither MELD-XI score nor mortality; thus, CAD extent alone does not appear to influence survival in patients with end-stage liver failure. MELD-XI score, on the other hand, independently predicted mortality, suggesting that renal and hepatic derangements related or unrelated to cardiac conditions may be more important predictors of mortality in patients with end-stage liver failure awaiting transplant. Randomized controlled studies with larger sample sizes are needed for future studies. Also, the relationship between Gensini score and MELD-XI score and their predictive powers for perioperative mortality should be studied in patients who undergo liver transplant.
Volume : 18
Issue : 1
Pages : 88 - 92
DOI : 10.6002/ect.TOND-TDTD2019.P32
From the 1Department of Cardiology and the 2Department of General Surgery,
Başkent University Faculty of Medicine, Ankara, Turkey
Acknowledgements: The authors have no conflicts of interest to declare. This study was supported by the Başkent University Research Fund.
Corresponding author: Suzan Keskin, Başkent University Faculty of Medicine Department of Cardiology, Yukarı Bahçelievler Mehallesi Mareşal Fevzi Çakmak Caddesi No:45 Çankaya, Ankara, Turkey
Phone: +90 539 950 27 36
Table 1. Demographic, Clinical, and Laboratory Parameters and Gensini and MELD-XI Scores in Deceased and Surviving Patient Groups
Figure 1. Receiver Operating Characteristic Curve of MELD-XI Score for In- Hospital Mortality Prediction