Predictors of tolvaptan short‐term response in patients with refractory ascites: A meta‐analysis

Background and Aim Tolvaptan represents an oral V2‐receptor antagonist, which has been suggested as a promising add‐on diuretic treatment for refractory ascites. The present meta‐analysis aims to accumulate current evidence and identify which clinical and laboratory factors are linked to short‐term response to tolvaptan therapy. Methods Medline, Scopus, Cochrane Central Register of Controlled Trials, Clinicaltrials.gov, and Google Scholar databases were searched from inception. All observational studies reporting the correlation of patients' characteristics with tolvaptan response were selected. Results Tolvaptan response was associated with significantly higher baseline body weight (mean difference: 4.59 kg, 95% confidence interval [CI]: [3.58, 5.61]), presence of hepatitis C (odds ratio: 1.59 95% CI: [1.18, 2.14]), lower blood urea nitrogen (BUN) (mean difference: −6.88 mg/dL, 95% CI: [−8.13, −5.63]), lower serum creatinine (mean difference: −0.17 mg/dL, 95% CI: [−0.30, −0.05]), lower C‐reactive protein (mean difference: −1.43 mg/dL, 95% CI: [−2.52, −0.35]), and higher sodium levels (mean difference: 1.00 mEq/L, 95% CI: [0.45, 1.55]). The outcomes of bodyweight, hepatitis C, BUN, and C‐reactive protein remain significant independently of response definition and risk of bias. Conclusions The present findings suggest bodyweight, BUN, C‐reactive protein, and hepatitis C as potential predictive factors of tolvaptan short‐term response in patients with refractory ascites. Future studies are needed to introduce cut‐off values and construct an optimal combined screening model.