Introduction: Given that beta-blockers reduce the incidence of bacterial translocation in cirrhotic rats, the aim of this study was to compare the long-term incidence of spontaneous bacterial peritonitis in cirrhotic patients submitted to pharmacologic versus endoscopic treatment to prevent variceal rebleeding.
Patients and methods: Two hundred and thirty patients with variceal hemorrhage were included in two previous randomized trials performed to compare the efficacy of medication (nadolol plus isosorbide mononitrate, n=115) versus endoscopic treatment (n=115) with sclerotherapy or ligation for the prevention of rebleeding.
Results: The mean follow-up was 23+/-1.4 months. The characteristics of the patients and the number of patients on long-term prophylaxis with norfloxacin were similar in both groups. The incidence of spontaneous bacterial peritonitis was lower in the medication group (9 versus 14.7%, P=NS). The probability of spontaneous bacterial peritonitis was also lower in the medication group (6 versus 12% at 1 year, 22 versus 36% at 5 years; P=0.08), due to a significantly lower probability of community-acquired spontaneous bacterial peritonitis in this group (1 versus 10% at 1 year, 18 versus 32% at 5 years; P=0.02). Patients with no hemodynamic response to therapy had a significantly higher probability to develop community-acquired spontaneous bacterial peritonitis during follow-up than hemodynamic responders (P<0.03). Long-term probability of developing community-acquired spontaneous bacterial peritonitis is lower in patients submitted to pharmacologic treatment for preventing variceal rebleeding than in those submitted to endoscopic treatment.
Conclusion: Long-term pharmacologic prophylaxis of variceal rebleeding contributes to the prevention of community-acquired spontaneous bacterial peritonitis.