乌鸦传媒

Proton pump inhibitors (PPIs), when used for prior gastrointestinal bleeding (GIB), were associated with better survival for veterans with cirrhosis, a retrospective study found.

Among more than 76,000 such patients, PPI use was associated with a lower risk of all-cause mortality for those hospitalized for GIB (adjusted HR 0.88, 95% CI 0.84-0.91), but not for all other patients (aHR 0.99, 95% CI 0.97-1.02), reported Nadim Mahmud, MD, MS, of the University of Pennsylvania in Philadelphia, and colleagues.

For every additional 320 mg of omeprazole (Prilosec) equivalent-months of cumulative PPI use, mortality risk was 7% greater for those not hospitalized for GIB (HR 1.07, 95% CI 1.06-1.08), the authors wrote in .

PPI use was also significantly associated with decompensation (HR 1.64, 95% CI 1.61-1.68) and severe infection (HR 1.21, 95% CI 1.18-1.24) in a probability weighted model.

An exploratory analysis showed PPIs linked to a greater risk of liver-related mortality (HR 1.23, 95% CI 1.19-1.28) and a lower risk of non-liver related mortality (HR 0.88, 95% CI 0.85-0.91).

"These results suggest that while there is a signal towards adverse liver-related outcomes with PPI exposure, when viewed in the broader context of overall survival PPIs may be beneficial so long as they are prescribed for appropriate indications such as [peptic ulcer disease]-related bleeding," the researchers noted. "Clinical practice efforts to deprescribe PPIs in patients with cirrhosis should recognize that those with a history of hospitalized GIB may experience benefit with continued PPI therapy."

Gastric acid suppression can lead to intestinal changes in the microbiota causing bacterial overgrowth and infection, such as spontaneous bacterial peritonitis (SBP), Mahmud's group noted.

Long-term use has also been associated with bone disease, dementia, and kidney disease, and PPIs are generally considered to be overused in the U.S., noted David E. Bernstein, MD, of Northwell Health in Manhasset, New York, who was not involved in the study.

"This study should alert physicians who care for patients with cirrhosis to be cognizant of the negative effects of these medications and to only prescribe PPIs for well-established and accepted indications and to review the indication for PPIs with the patient at every visit," Bernstein told Medpage Today.

For their study, Mahmud and colleagues examined national Veterans Health Administration data on 76,251 patients (median age 62-64) with compensated cirrhosis. Among them, 23,628 were receiving a PPI at baseline, 28,016 were new PPI initiators at follow-up, and 24,607 had no PPI exposure. Exclusion criteria included decompensation, liver transplantation, a recent diagnosis of hepatocellular carcinoma, limited follow-up, and fewer than two annual outpatient visits.

Inverse probability treatment weighting and all-cause mortality models adjusted for time-varying covariates of hospitalized GIB, statin exposure, antiplatelet exposure, and cardiovascular comorbidities, among others.

More patients on PPIs at baseline were white, compared with those without any PPI exposure (64% vs 60%). And they had a higher BMI (29 vs 28), with more cardiovascular and metabolic comorbidities, as well as alcohol-related liver disease (32% vs 26%) and metabolic associated fatty liver disease (27% vs 23%).

The most frequently prescribed PPIs included: omeprazole (77%), pantoprazole (Protonix, 22%), and lansoprazole (Prevacid, 0.6%). Most patients were on daily doses of 20 mg (63%) or 40 mg (32%) of omeprazole equivalents, with a few on 80 mg (5%).

Notably, 11,998 were hospitalized for GIB. Over an average follow-up of 49 months, 28,628 all-cause deaths occurred. Of those, 11,731 were liver-related deaths.

The authors acknowledged several limitations to the data. Residual confounding could have led to bias in statistical point estimates, and findings may not be generalizable to non-veteran cohorts.

Comment