Rifaximin Prophylaxis Fails for Bacterial Peritonitis in Cirrhotic Patients

Rifaximin prophylaxis was no better than placebo at preventing deaths and other negative outcomes from spontaneous bacterial peritonitis (SBP) in patients with severe cirrhosis, according to interim results of the phase III ProPILA-Rifax trial.

In a modified intention-to-treat analysis, transplant-free survival after 12 months among French patients who took 550 mg of rifaximin twice daily was 58% compared with 65% for those who took a placebo (log-rank, P=0.39), Thierry Thévenot, MD, PhD, of the Centre Hospitalier Régional et Universitaire in Besançon, France, reported at the European Association for the Study of the Liver (EASL) annual conference in Milan.

“Bacterial infections in cirrhotic patients represent the most important precipitating event for acute decompensation and account for significant mortality,” Thévenot told attendees. “One of the most common infections observed worldwide is spontaneous bacterial peritonitis.” Although antibiotic prophylaxis to prevent a first episode of SBP is recommended in very high-risk patients, it remains a topic of controversy, he noted.

After Thévenot and colleagues excluded 33 cases of suspected severe alcoholic hepatitis from the cohort, there was still no significant difference in transplant-free survival at 12 months between rifaximin and placebo (log-rank, P=0.61)

First-event cumulative incidence of hepatic encephalopathy, gastrointestinal (GI) bleeding, and hepatorenal syndrome were similar between the two groups (Gray’s test, P=0.22) and when analyzed as individual parameters, Thévenot told attendees.

The probability of transplant-free survival was not significantly different in the treatment and placebo groups at both 3 months (log-rank, P=0.29) and 6 months (P=0.44) as well.

Researchers also looked at probability of survival according to Model for End-Stage Liver Disease (MELD) score, Child-Pugh score, sex, and age. Again, there were no significant differences in treatment effects from rifaximin according to these subgroups.

“Is it time to say primary prophylaxis is dead?” Jasmohan Bajaj, MD, of the Virginia Commonwealth University and Richmond VA Medical Center, commented during a Q&A session. Bajaj was not associated with the study.

Bajaj added that his recent study conducted at VA centers found high rates of Escherichia coli resistance to ciprofloxacin among patients admitted for a first SBP episode who had been prescribed primary prophylaxis. As a result, “the VA currently nationally has banned the practice of primary prophylaxis in patients with SBP,” he said.

Guidelines from the American Association for the Study of the Liver (AASLD) recommend primary prophylaxis for SBP in patients with cirrhosis and upper GI bleeding. In patients without gastrointestinal hemorrhage and without a prior episode of SBP, antibiotic prophylaxis may be considered in selected patients at very high risk of SBP. However, evidence for primary or secondary prophylaxis with antibiotics other than norfloxacin is weak. Also, there are not only mounting concerns about the development of antibiotic resistance but also about the safety profile of fluoroquinolones. EASL guidelines are similar, and also recommend norfloxacin as a first-line agent for prophylaxis.

“Rifaximin may be an attractive alternative to norfloxacin because it is an oral broad-spectrum bactericidal antibiotic with a low GI absorption,” Thévenot explained. “The risk of emergent rifaximin resistance is lower than many other antibiotics.”

Bajaj also wanted to know why the ProPILA-Rifax trial used mortality as a primary endpoint, rather than time to first SBP. “Mortality is a very difficult endpoint to meet with so many other competing risk factors in between,” he said. “It would have been cleaner for us to conceptualize if the first SBP episode would have been your endpoint.”

Thévenot replied that mortality was chosen because it was a hard endpoint, but acknowledged that there could be many precipitating factors of mortality in patients with severe cirrhosis during the 1-year follow-up period.

Researchers also analyzed the effects of rifaximin on microbiome richness and diversity in five of the participating centers. Rifaximin significantly reduced microbiome richness after 3, 6, and 12 months of treatment when compared with placebo, but richness was restored at 18 months. There were no differences in microbiome composition or diversity between the rifaximin and placebo groups.

ProPILA-Rifax was a double-blind randomized trial conducted at 17 centers in France. Patients were enrolled between July 2018 and June 2022. Key inclusion criteria were cirrhosis with grade 2 or 3 ascites, along with impaired renal function or severe liver impairment. Main exclusion criteria were patients with previous SBP, any bacterial infection or antibiotic use within 7 days, GI bleeding within 7 days, grade 3 hepatic encephalopathy, being waitlisted for liver transplant with an estimated wait time of 3 months or less, receipt of transjugular intrahepatic portosystemic shunt, or hepatocellular carcinoma outside Milan criteria.

The mean age of participants was 60 years and 74% were male. The cause of cirrhosis among participants was primarily from alcohol (72%), followed by metabolic dysfunction-associated steatohepatitis (MASH, 18%), and viral hepatitis (3%). Severe alcoholic hepatitis was suspected in about 22% and hepatocellular carcinoma in 5.3%. The mean Child-Pugh score was 10.3, mean MELD score was 17.8, and mean total protein in ascites was 10 g/L.

The modified intention-to-treat analysis included 72 participants in the rifaximin group and 80 in the placebo group. The primary survival endpoint censored patients who received a liver transplant as they are no longer thought to be at risk of cirrhosis-related death.

There were few safety concerns during the study period. Only one patient in the placebo group had a treatment-related serious adverse event. In the rifaximin group, two patients developed pruritus and one patient had an unexpected death at home. There were no cases of Clostridium difficile colitis in the rifaximin group.

“Our conclusions should be taken cautiously,” Thévenot emphasized. “Our data are still very preliminary and will be analyzed further, based on patient compliance with treatment.