ABSTRACT Backgrounds This study aimed to investigate the incidence and potential risk factors for hepatic encephalopathy (HE) recurrence following the initiation of rifaximin therapy in patients with liver cirrhosis. Methods We conducted a multicenter retrospective cohort study of cirrhotic patients who initiated rifaximin between December 2016 and December 2023. Patients with a prior episode of overt HE who received rifaximin for secondary prophylaxis were included. Clinical data, laboratory parameters, and concomitant medications were extracted from medical records. Factors associated with HE recurrence after rifaximin initiation were analyzed using univariate and multivariate analyses. Results A total of 145 patients were included (median age, 70 years; male, 52.4%). During a median observation period of 26.4 months (interquartile range: 11.4–48.9), 52 patients (35.9%) experienced HE recurrence. In multivariate analysis, the baseline aMAP risk scores the only independent predictor of HE recurrence ( p = 0.041). Receiver operating characteristic (ROC) analysis identified a cutoff value of 70.843 for the aMAP risk score to predict HE recurrence. Patients with an aMAP risk score below 70.843 had significantly lower cumulative recurrence rates than those with scores ≥ 70.843 ( p = 0.012). Conclusions Despite good tolerability, more than one‐third of the patients experienced HE recurrence after rifaximin initiation. The aMAP risk score was useful for predicting the risk of recurrence, suggesting its potential clinical utility in the treatment of HE.