Development and Validation of a Patient-Reported Outcome Measurement for Symptom Assessment in Cirrhotic Ascites

Objectives: As current treatments of cirrhotic ascites are not associated with survival benefit, symptom relief is the major therapeutic end point. We developed a questionnaire (Ascites-Q; modified polycystic liver disease questionnaire) and assessed validity and responsiveness for symptom assessment in cirrhotic ascites. Methods: Ascites-Q was compared with Functional Assessment of Chronic Illness Therapy-Ascites Index (FACIT-AI; developed for malignant ascites) and Japanese Ascites Symptom Inventory-7 (ASI-7) in cirrhotics undergoing large-volume paracentesis. Convergent validity was defined as correlation >0.4 between ascites questionnaires and quality of life (QoL) visual analog scale. Responsiveness was assessed by comparing scores at baseline and 7 days after large-volume paracentesis. To test discriminative ability, we compared scores of patients with cirrhotic controls without ascites (n=24) and diuretic-sensitive ascites (n=46). Results: We included 90 patients with refractory cirrhotic ascites (61% male, mean age 59 years, Model of End-Stage Liver Disease (MELD) score 16, median paracentesis volume 4,100 ml). Higher symptoms scores were correlated with lower QoL (Ascites-Q: r=0.479, P<0.001, FACIT-AI: r=0.313, P=0.007; ASI-7: r=0.340, P=0.004), but only Ascites-Q showed convergent validity (r>0.4). Symptoms decreased after paracentesis (Ascites-Q: 57 to 34, FACIT-AI: 44 to 33, and ASI-7: 57 to 25, all P<0.001). Ascites-Q and ASI-7 discriminated between controls without ascites, diuretic-sensitive, and refractory ascites (Ascites-Q: 16 vs. 35 vs. 56 points, ASI-7: 2 vs. 25 vs. 61 points, all P<0.05), whereas FACIT-AI (39 vs. 40 vs. 52 points) could not (P=0.314). Ascites-Q was validated at 3 months in an independent cohort with ascites controlled with a pump. Conclusions: The Ascites-Q is the best ascites-specific outcome to evaluate symptom relief in cirrhotic ascites.