Background and Aims: Poor patient-reported outcomes (PROs) are common in cirrhosis, including poor sleep and health-related quality of life (HRQOL). HE is a major driver of poor PROs. Many clinicians initiate lactulose therapy to address poor PROs. PRO-triggered therapy, however, has not been studied till date. Methods: We conducted a 28-day randomized trial of crystalline lactulose therapy (20 g BID) compared with no HE-directed therapy in 52 patients with cirrhosis, portal hypertension, no prior HE, and high Work Productivity and Activity Impairment scores (0–10) attributed to cirrhosis. The primary outcome was change in global HRQOL measured with Short Form-8 Health Survey. Secondary outcomes included change in Animal Naming Test score, Work Productivity and Activity Impairment, and sleep quality (scored “very bad” to “very good”). Approach and Results: Overall, 52 patients underwent randomization; 3 subjects withdrew from the crystalline lactulose arm (1 before medication initiation, 1 due to an unrelated condition, and 1 due to high baseline bowel movements). The average age was 60 years, the average Model for Endstage Liver Disease—Sodium score was 10.5, and 56% of the patients had ascites. Baseline bowel movements were 2.3/day, with 35% of the patients having Bristol Stool Scale >4. At 28 days, there was no improvement in HRQOL: patients receiving crystalline lactulose had an 8.1-point (95% CI: 3.7–12.4) increase compared with 6.6 (95% CI: 2.3–10.8) in the control group (p = 0.6). Lactulose was associated with a significantly (p = 0.002) increased Animal Naming Test score (3.7, 95% CI: 2.1–5.4) versus the control group (0.2, 95% CI: −1.7, 1.4). Lactulose users reported more good sleep (92% vs. 52%, p = 0.001) and lower activity impairment (3.0 vs. 4.8, p = 0.02). Conclusions: Lactulose improves sleep and activity impairment in patients with poor PROs. We did not observe any improvement in global HRQOL after 28 days using the Short Form-8 Health Survey instrument.

Overall, 74 patients completed the trial, aged 56.6 6 11.5 years, 54% male, 41% with ascites, 38% with encephalopathy, and model for end-stage liver disease-sodium score 11.2 6 4.9. Many patients were receiving other cramp therapies at baseline. The baseline VAS for cramps was 4.2 6 3.4, the EQ-5D was 0.80 6 0.10, and 43% rated sleep as poor. At trial completion, the respective values for the pickle juice and control arms were 22.25 6 3.61 points on the VAS for cramps, compared with control tap water (20.36 6 2.87), P 5 0.03; a proportion of cramp-days with VAS-cramps <5 were 46% vs 35% (P 5 0.2); and the change in sleep quality was not different (P 5 0.1). The end-of-trial EQ-5D was 0.78 6 0.10 vs 0.80 6 0.10 (P 5 0.3). No differences in weight change were observed for those with and without ascites.