REGULAR CONTENT
Final ID
544
Type
Original Scientific Research-Poster Only
Authors
J Ronald1, Q Wang2, P Suhocki1, M Hall1, W Pabon-Ramos1, D Sopko1, T Smith1, C Kim1
Institutions
1Duke University Medical Center, Durham, NC, 2Duke Univ / Huazhong University of Science and Technology, Durham, NC
Purpose
The albumin-bilirubin (ALBI) grade is a recently developed discriminatory tool for assessing liver dysfunction in patients with cirrhosis and hepatocellular carcinoma that equals or outperforms Child-Pugh score for predicting survival. The Model for End-Stage Liver Disease (MELD) score is also a measure of liver dysfunction that was originally designed to predict survival after transjugular intrahepatic portosystemic shunt (TIPS) creation. This study compared the ALBI and MELD scores for predicting survival after TIPS.
Materials & Methods
In this retrospective study, pre-procedure ALBI and MELD score components were available for 198 patients (58 females, mean age=55 years) undergoing TIPS from 2005 to 2012. The medical record was reviewed to determine clinical outcomes. Survival after TIPS creation was estimated using the Kaplan-Meier method. Discriminatory ability was assessed using the C-index derived from Cox proportional hazards models.
Results
TIPS were created in cirrhotic patients with portal hypertension for management of ascites (n=128), variceal hemorrhage (n=62), or both (n=8). The most common cirrhosis etiologies included alcoholic (n=96), chronic viral hepatitis (n=74), and non-alcoholic fatty liver disease (n=22). There were 5 patients with an ALBI grade of 1, 76 with an ALBI grade of 2, and 117 with an ALBI grade of 3. The mean MELD score was 14. ALBI score, ALBI grade, and MELD were each significant predictors of 60 day mortality from hepatic failure and overall survival (all p<0.05). Based on the C-index, the MELD score was a better predictor of both 60 day and overall survival (C-index=0.74 and 0.67) than either ALBI score (0.70 and 0.63) or ALBI grade (0.63 and 0.60). When both ALBI and MELD score were directly compared in a multivariate Cox proportional hazards model, MELD score was a statistically significant predictor of 60 day and overall mortality (p=0.0001 and p=0.0004) whereas ALBI score was not (p=0.19 and p=0.14).
Conclusions
The ALBI score is a significant predictor of mortality after TIPS creation. However, the MELD score remains the superior predictor of mortality after TIPS.
Final ID
544
Type
Original Scientific Research-Poster Only
Authors
J Ronald1, Q Wang2, P Suhocki1, M Hall1, W Pabon-Ramos1, D Sopko1, T Smith1, C Kim1
Institutions
1Duke University Medical Center, Durham, NC, 2Duke Univ / Huazhong University of Science and Technology, Durham, NC
Purpose
The albumin-bilirubin (ALBI) grade is a recently developed discriminatory tool for assessing liver dysfunction in patients with cirrhosis and hepatocellular carcinoma that equals or outperforms Child-Pugh score for predicting survival. The Model for End-Stage Liver Disease (MELD) score is also a measure of liver dysfunction that was originally designed to predict survival after transjugular intrahepatic portosystemic shunt (TIPS) creation. This study compared the ALBI and MELD scores for predicting survival after TIPS.
Materials & Methods
In this retrospective study, pre-procedure ALBI and MELD score components were available for 198 patients (58 females, mean age=55 years) undergoing TIPS from 2005 to 2012. The medical record was reviewed to determine clinical outcomes. Survival after TIPS creation was estimated using the Kaplan-Meier method. Discriminatory ability was assessed using the C-index derived from Cox proportional hazards models.
Results
TIPS were created in cirrhotic patients with portal hypertension for management of ascites (n=128), variceal hemorrhage (n=62), or both (n=8). The most common cirrhosis etiologies included alcoholic (n=96), chronic viral hepatitis (n=74), and non-alcoholic fatty liver disease (n=22). There were 5 patients with an ALBI grade of 1, 76 with an ALBI grade of 2, and 117 with an ALBI grade of 3. The mean MELD score was 14. ALBI score, ALBI grade, and MELD were each significant predictors of 60 day mortality from hepatic failure and overall survival (all p<0.05). Based on the C-index, the MELD score was a better predictor of both 60 day and overall survival (C-index=0.74 and 0.67) than either ALBI score (0.70 and 0.63) or ALBI grade (0.63 and 0.60). When both ALBI and MELD score were directly compared in a multivariate Cox proportional hazards model, MELD score was a statistically significant predictor of 60 day and overall mortality (p=0.0001 and p=0.0004) whereas ALBI score was not (p=0.19 and p=0.14).
Conclusions
The ALBI score is a significant predictor of mortality after TIPS creation. However, the MELD score remains the superior predictor of mortality after TIPS.