Hypoalbuminemia is a predictor of mortality and rebleeding in peptic ulcer bleeding under proton pump inhibitor use

Background/purpose: Peptic ulcer bleeding remains a deadly disease, and a simple indicator of long-term outcomes is crucial. This study validated whether hypoalbuminemia and its related factors in patients with peptic ulcer bleeding can indicate long-term mortality and rebleeding under proton pump inhibitor use. Methods: The prospective cohort study enrolled 426 patients with peptic ulcer bleeding who had high risk stigmata at endoscopy and had received endoscopic hemostasis. They were divided into 79 patients in the hypoalbuminemia group (Hypo-AG, serum albumin <28 g/L), 135 in the marginal hypoalbuminemia group (Margin-AG, serum albumin 28–34.9 g/L), and 212 in the normal albuminemia group (Normal-AG, serum albumin ≥35 g/L). Each subject received 72-h of intravenous infusion and then the oral form of proton pump inhibitors and were monitored for 84 days to assess all-cause mortality and recurrent bleeding. Results: The primary outcome of all-cause mortality rates were increased in a stepwise fashion in a trend from Normal-AG, Margin-AG, to Hypo-AG (0–28th day: 1.9%, 2.2%, 12.8%, p < 0.001; 29th–84th day: 2.5%, 8.0%, 10.6%, p < 0.01). The secondary outcome of recurrent bleeding rates were also increased in the same fashion (0–28th day: 6.4%, 15.4%, 24.6%, p < 0.001; 29th–84th day: 0%, 3.0%, 4.2%, p = 0.01). Abnormal albuminemia was <30 g/L related to hemoglobin levels <70 g/L, nosocomial bleeding, cirrhosis, age ≥70 years, shock, and ulcer size ≥1.0 cm independently (p < 0.05). Conclusion: Hypoalbuminemia in patients with peptic ulcer bleeding can be an alarm indicator of all-cause mortality and recurrent bleeding in a long-term follow-up situation under proton pump inhibitor use (NCT01591083).