ABSTRACT
ObjectiveTo investigate the value of MELD 3.0， MELD， and MELD-Na scores in assessing the 90-day prognosis of patients with acute-on-chronic liver failure （ACLF） through a comparative study. MethodsA retrospective analysis was performed for the clinical data of 605 patients with ACLF who were treated in Tianjin Third Central Hospital， The Fifth Medical Center of Chinese PLA General Hospital， and Beijing YouAn Hospital from November 2012 to June 2019， and according to the 90-day follow-up results after admission， they were divided into survival group with 392 patients and death group with 213 patients. The receiver operating characteristic （ROC） curve， the area under the ROC curve （AUC）， net reclassification improvement （NRI）， integrated discrimination improvement （IDI）， and decision curve analysis （DCA） curve were used to investigate the value of MELD 3.0， MELD， and MELD-Na scores at baseline， day 3， week 1， and week 2 in predicting the prognosis of the disease. ResultsAt day 3 and week 1， MELD 3.0 score had an AUC of 0.775 and 0.808， respectively， with a better AUC than MELD score （P<0.05）. At day 3， week 1， and week 2， MELD 3.0 score showed an NRI of 0.125， 0.100， and 0.081， respectively， compared with MELD in predicting the prognosis of ACLF patients， as well as an NRI of 0.093， 0.140， and 0.204， respectively， compared with MELD-Na score in predicting prognosis. At baseline， day 3， week 1， and week 2， MELD 3.0 showed an IDI of 0.011， 0.025， 0.017， and 0.013， respectively， compared with MELD in predicting the prognosis of ACLF patients. At day 3 and week 2， MELD 3.0 showed an IDI of 0.027 and 0.038， respectively， compared with MELD-Na in predicting the prognosis of ACLF patients. All the above NRIs and IDIs were >0， indicating a positive improvement （all P<0.05）. DCA curves showed that MELD 3.0 was superior to MELD at day 3 and was significantly superior to MELD-Na at week 2. There was no significant difference in the ability of the three scores in predicting the prognosis of ACLF patients with different types， and there was also no significant difference in the ability of the three scores in predicting the prognosis of ACLF patients with the etiology of HBV infection， alcohol， or HBV infection combined with alcohol， while MELD 3.0 was superior to MELD for ACLF patients with other etiologies （P<0.05）. ConclusionMELD 3.0 score is better than MELD and MELD-Na scores in predicting the 90-day survival of patients with ACLF， but with limited superiority.
ABSTRACT
Objective To explore the predictive values of the initial model for end-stage liver disease (MELD) score, MELD combined with serum sodium (MELD-Na) score and MELD combined with serum lactic acid (MELD-Lac) score for early survival rate after liver transplantation in patients with liver failure. Methods Clinical data of 135 recipients undergoing liver transplantation for liver failure were retrospectively analyzed. All patients were divided into the early survival group (n=110) and early death group (n=25) according to the survival at postoperative 28 d. Clinical data were compared between two groups. The optimal cut-off values of MELD, MELD-Na and MELD-Lac scores for predicting early survival rate after liver transplantation in patients with liver failure were determined by the receiver operating characteristic (ROC) curve. The predictive values of different scores for early survival rate after liver transplantation in patients with liver failure were evaluated. Results Significant differences were observed in the initial MELD, MELD-Na and MELD-Lac scores after liver transplantation between two groups (all P < 0.05). For the initial MELD, MELD-Na and MELD-Lac scores in predicting early survival rate after liver transplantation in patients with liver failure, the AUC were 0.653 [95% confidence interval (CI) 0.515-0.792], 0.648 (95%CI 0.514-0.781) and 0.809 (95%CI 0.718-0.900), the optimal cut-off values were 18.09, 18.09 and 19.97, Youden's indexes were 0.398, 0.380 and 0.525, the sensitivity was 0.680, 0.680 and 0.840, and the specificity was 0.720, 0.700 and 0.690, respectively. The AUC of MELD-Lac score was higher than those of MELD and MELD-Na scores, and the differences were statistically significant (both P < 0.05). Conclusions Compared with the initial MELD and MELD-Na scores after liver transplantation, the initial MELD-Lac score is a more reliable index for predicting early survival rate after liver transplantation in patients with liver failure.
ABSTRACT
ObjectiveTo investigate the value of MELD combined with serum sodium concentration (MELD-Na) in predicting the short-term outcome of patients with HBV-related acute-on-chronic liver failure (ACLF) in China. MethodsPubMed, Embase, CNKI, VIP, and Wanfang Data were searched for related articles in English and Chinese published from January 2006 to March 2018. The second version of Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) was used to evaluate the quality of each article. The presence or absence of threshold effect was examined. According to the presence or absence of heterogeneity, a random effect model or a fixed effect model was used for pooling sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio (DOR). The receiver operating characteristic (ROC) curve was plotted and the area under the ROC curve (AUC) was calculated. The Deek’s funnel plot asymmetry test was used to investigate the presence or absence of publication bias. ResultsA total of eight studies were included, with a total of 1386 patients, among whom there were 1173 (84.6%) male patients. The mean age of the patients included in these studies ranged from 41.3 to 46.9 years. The tests for heterogeneity showed significant heterogeneity between studies and a random effect model was used for pooling. The sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and DOR were 0.67 (95% confidence interval ［CI］: 0.59-0.75), 0.77 (95%CI: 0.63-0.87), 2.89 (95%CI: 1.84-4.54), 0.43 (95%CI: 0.35-0.52), and 7.32 (95%CI: 4.33-12.40), respectively. The AUC was 0.75 (95CI: 0.72-0.79). The Deek’s funnel plot asymmetry test showed no significant publication bias. ConclusionMELD-Na score has a certain value in predicting the short-term outcome of patients with HBV-related ACLF in China.
ABSTRACT
Objective To study the value of preoperative MELD-Na score (Model for End-Stage Liver Disease-Sodium) in predicting complication severity grades after liver transplantation (LT) for severe hepatitis.Methods Patients who underwent LT for severe hepatitis between August 1,2004,and September 1,2014 were retrospectively studied.The Accordion severity grading system was used to classify the complication severity grades after LT.The grades were classified as grade 1 (mild),grade 2 (moderate),grade 3-5 (severe),and grade 6 (death).The area under the curve (AUC) was calculated by plotting the receiver operating characteristic curve (ROC) to evaluate the predictive accuracy of the MELD-Na score for the severe and mortality grades after LT.The correlation between the MELD-Na score with the complication severity grade after LT was studied by the Spearman correlation and by multivariate analysis.Results The incidences of postoperative complications for the 159 patients in this study were:grade 2 in 43 patients (27.0％,MELD-Na score 27.3 ±7.4),grade 3 in 41 patients (25.8％,MELD-Na score 32.7 ± 12.4),grade 4 in 31 patients (19.5％,MELD-Na score 34.3 ± 12.1),grade 5 in 9 patients (5.7％,MELD-Na score 30.7 ± 12.3),grade 6 in 35 patients (22％,MELD-Na score 37.1 ± 10.4).There was no grade 1 patient.The AUC of the MELD-Na score for the severe and death groups were 0.631 (P ＜ 0.05;95 ％ CI,0.533 ～ 0.728) and 0.670 (P ＜ 0.05;95 ％ CI,0.574 ～ 0.766) respectively.The MELD-Na score was significantly correlated with the Accordion severity grade (rho 0.297,P ＜ 0.01) on Spearman correlation analysis.Multivariate analysis showed that a MELD-Na score ≥25 was a risk factor of postoperative severe grade complication (P ＜ 0.05,OR =4.35),a MELD-Na score ≥35 was a risk factor of postoperative mortality (P ＜0.01,HR =4.72).Conclusion The MELD-Na score was significantly correlated with the Accordion severity grade,which efficaciously predicted the complication severity grades after liver transplantation.
ABSTRACT
Objective To evaluate the value of preoperative model for end-stage liver disease combined with serum sodium (MELD-Na) score for the prediction of the incidence of acute kidney injury (AKI) early after liver transplantation. Methods Clinical data of 315 recipients undergoing orthotopic liver transplantation by retrograde inferior vena caval perfusion were retrospectively analyzed. According to preoperative MELD-Na score, all patients were divided into group A (MELD-Na score≤10, n=115), group B (10<MELD-Na score≤20, n=118) and group C (MELD-Na score>20, n=82). Preoperative and intraoperative parameters of the recipients were statistically compared among three groups. Preoperative parameters included serum creatinine (Scr), blood urea nitrogen (BUN), albumin (Alb), total bilirubin (TB), prothrombin time-international normalized ratio (PT-INR), mean arterial pressure (MAP) and serum Na+,etc. Intraoperative parameters included operation time, vena caval occlusion time, hemorrhage volume, quantity of red blood cell infusion, quantity of plasma transfusion and total fluid infusion volume, etc. The incidence and staging of AKI early after liver transplantation in the recipients were statistically compared among three groups. Spearman's rank correlation analysis was performed to analyze the correlation between preoperative MELD-Na score and AKI staging. Results Preoperative BUN, Alb, TB, PT-INR, MAP and Na+ in the recipients significantly differed among three groups (all P<0.05). Intraoperative vena caval occlusion time, hemorrhage volume, quantity of red blood cell infusion and quantity of plasma transfusion significantly differed among three groups (all P<0.05). In 315 recipients undergoing liver transplantation, the incidence of AKI within postoperative 1 week was 64.8% (204/315), and 43% (49/115), 71% (84/118) and 87% (71/82) in group A, B and C. Statistical significance was identified among three groups (all P<0.05). Spearman's rank correlation analysis revealed that preoperative MELD-Na score was positively correlated with AKI staging (r=0.442, P=0.000). Conclusions MELD-Na score not only acts as a parameter evaluating preoperative patients' conditions, but also serves as a pivotal parameter predicting postoperative incidence of AKI.
ABSTRACT
Objective To evaluate the predictive effect of preoperative CTP score,MELD score and MELD-Na score on short-term prognosis (≤ 3 months) after liver transplantation,to analyze factors which correlated with survival,and to seek indicators that accurately predicted short-term outcomes.Methods The clinical data of 73 consecutive patients with end-stage liver diseases who underwent liver transplantation in a single center were retrospectively analyzed.The area under the ROC curve (AUC) was used to determine the predictive power.Correlated factors were analyzed by multivariate logistic regression.The statistical processing package used was SAS 9.1.3 software.Results 11 (15.1％) of 73 patients died within 3 months after liver transplantation.The areas under the ROC curve of the preoperative CTP score,MELD score and MELD-Na score for predicting short-term survival were 0.817,0.839 and 0.860 respectively.There was no significant difference among these 3 scoring systems.On univariate analysis,indicators significantly correlated with early mortality were preoperative serum sodium,serum urea,PT-INR,CTP score,MELD score and MELD-Na score.On logistic multiple regression,only MELD-Na score remained as a significant indicator (P =0.001,β =-2.496,OR =0.085,95％ CI:0.019 ～ 0.370).Conclusions The MELD-Na scoring system showed superior predictability of early mortality in patients who underwent liver transplantation.The preoperative MELD-Na score was an independent risk factor of short-term survival.The higher the MELD-Na score,the higher was the early mortality.