Use of beta-blockers is associated with improved 30 day survival in patients with spontaneous bacterial peritonitis

Introduction: Use of beta-blockers after spontaneous bacterial peritonitis (SBP) leads to increased long-term mortality. However, the effect on survival of the infection is unclear. In a murine peritonitis model, inhibition of the neuroadrenergic system even increased survival. Thus, we studied the role of beta-blockers for short-term survival in patients with SBP.

Methods: We determined transplant-free 30-day survival in cirrhotic patients who developed SBP between March 2012 and April 2014 in our department. Interleukin 8 (IL-8) was measured in in ascites supernatant by ELISA. Ascites cells were stimulated in vitro with 100 ng/ml Lipopolysacchairde (LPS) with or without pre-treatment with 5 µg/ml propranolol and IL-8 production was assessed by ELISA.

Results: The study population comprised 55 patients with SBP (32 (58%) male, median age 59 years). Cirrhosis was due to alcohol in 36 (66%) patients, to viral hepatitis in 11 (20%) patients. Seven (13%) patients had hepatocellular carcinoma (HCC). Median MELD score was 21, and 42 (76%) patients were classified as Child-Pugh stage C. SBP was nosocomial in 35 (64%) patients. 34 (62%) and 21 (38%) patients were without and with beta-blockers at time of SBP, respectively: 14 (67%) with propranolol, the rest with metoprolol, bisoprolol or nebivolol. Clinical and laboratory data did not differ between patients with and without beta-blockers.

30-day survival was 58% (n= 32): 16/21 (76%) in patients on beta-blockers versus 16/34 (41%) in patients without (p = 0.049). Poor survival was associated with the presence of HCC (1/7 (14%) versus 31/48 (65%); p = 0.017) and MELD score > 22 (6/25 (24%) versus 26/30 (87%), p < 0.001)

A forward conditional Cox logistic regression analysis confirmed MELD score ≤ 22 (p < 0.001) and use of beta-blockers (p = 0.049) as prognostic factors of 30-day survival.

During SBP, patients on beta-blockers had higher fractions of mononuclear cells among ascites leucocytes than patients without (median 31% versus 19%; p = 0.036). IL-8 levels in the ascites were lower in patients without (1289 pg/ml) than with beta-blockers (470 pg/ml; p = 0.29) in a subgroup of 19 patients were ascites was available. In-vitro, an increase in IL-8 production after stimulation of ascites cells with LPS could be attenuated by pre-treatment with propranolol (n= 7; p = 0.033).

Conclusion: Intake of beta-blockers in patients with SBP was associated with a slightly better 30-day survival, which might be due to attenuated IL-8 secretion and a more balanced peritoneal inflammation. Increased mortality reported for beta-blockers after SBP is probably not related to the infection itself.

Corresponding author: Lutz, Philipp

E-Mail: [e-mail]