Primary cirrhotic prophylaxis of bacterial peritonitis falls short

WASHINGTON – Patients with cirrhosis who received secondary prophylaxis for spontaneous bacterial peritonitis had better outcomes than patients who received primary prophylaxis in a review of more than 300 patients at 14 North American centers.

The mortality rate during follow-up of cirrhotic patients hospitalized while on primary prophylaxis against spontaneous bacterial peritonitis (SBP) was 19%, compared with a 9% death rate among cirrhotic patients hospitalized while on secondary prophylaxis, Jasmohan S. Bajaj, MD, said at the annual Digestive Disease Week^®.

Dr. Jasmohan Bajaj

Although the findings raised questions about the value of primary prophylaxis with an antibiotic in cirrhotic patients for preventing a first episode of SBP, secondary prophylaxis remains an important precaution.

“There is clear benefit from secondary prophylaxis; please use it. The data supporting it are robust,” said Dr. Bajaj, a hepatologist at Virginia Commonwealth University, Richmond. In contrast, the evidence supporting benefits from primary prophylaxis is weaker, he said. The findings were also counterintuitive, because patients who experience repeat episodes of SBP might be expected to fare worse than those hit by SBP just once.

Dr. Bajaj also acknowledged the substantial confounding that distinguishes patients with cirrhosis receiving primary or secondary prophylaxis, and the difficulty of fully adjusting for all this confounding by statistical analyses. “There is selective bias for secondary prevention, and there is no way to correct for this,” he explained. Patients who need secondary prophylaxis have “weathered the storm” of a first episode of SBP, which might have exerted selection pressure, and might have triggered important immunologic changes, Dr. Bajaj suggested.

The findings also raised concerns about the appropriateness of existing antibiotic prophylaxis for SBP. The patients included in the study all received similar regimens regardless of whether they were on primary or secondary prophylaxis. Three-quarters of primary prophylaxis patients received a fluoroquinolone, as did 81% on secondary prophylaxis. All other patients received trimethoprim-sulfamethoxazole. These regimens are aimed at preventing gram-negative infections; however, an increasing number of SBP episodes are caused by either gram-positive pathogens or strains of gram-negative bacteria or fungi resistant to standard antibiotics.

Clinicians “absolutely” need to rethink their approach to both primary and secondary prophylaxis, Dr. Bajaj said. “As fast as treatment evolves, bacteria evolve 20 times faster. We need to find ways to prevent infections without antibiotic prophylaxis, whatever that might be.”

The study used data collected prospectively from patients with cirrhosis at any of 12 U.S. and 2 Canadian centers that belonged to the North American Consortium for the Study of End-Stage Liver Disease. Among 2,731 cirrhotic patients admitted nonelectively, 492 (18%) were on antibiotic prophylaxis at the time of their admission, 305 for primary prophylaxis and 187 for secondary prophylaxis. Dr. Bajaj and his associates used both the baseline model for end-stage liver disease score and serum albumin level of each patient to focus on a group of 154 primary prophylaxis and 154 secondary prophylaxis patients who were similar by these two criteria. Despite this matching, the two subgroups showed statistically significant differences at the time of their index hospitalization for several important clinical measures.

The secondary prophylaxis patients were significantly more likely to have been hospitalized within the previous 6 months, significantly more likely to be on treatment for hepatic encephalopathy at the time of their index admission, and significantly less likely to have systemic inflammatory response syndrome on admission.

Also, at the time of admission, secondary prophylaxis patients were significantly more likely to have an infection of any type at a rate of 40%, compared with 24% among those on primary prophylaxis, as well as a significantly higher rate of SBP at 16%, compared with a 9% rate among the primary prophylaxis patients. During hospitalization, nosocomial SBP occurred significantly more often among the secondary prophylaxis patients at a rate of 6%, compared with a 0.5% rate among those on primary prophylaxis.

Despite these between-group differences, the average duration of hospitalization, and the average incidence of acute-on-chronic liver failure during follow-up out to 30 days post discharge was similar in the two subgroups. And the patients on secondary prophylaxis showed better outcomes by two important parameters: mortality during hospitalization and 30 days post discharge; and the incidence of ICU admission during hospitalization, which was significantly greater for primary prophylaxis patients at 31%, compared with 21% among the secondary prophylaxis patients, Dr. Bajaj reported.

Dr. Bajaj has been a consultant for Norgine and Salix Pharmaceuticals and has received research support from Grifols and Salix Pharmaceuticals.

mzoler@mdedge.com