Background and aims:
Spontaneous bacterial peritonitis (SBP)is the most common bacterial infection in patients with decompensated cirrhosis and is associated with high mortality and morbidity. Despite advances in medical care, 1-year mortality after the first episode of SBP ranges from 31% to 93%. SBP prophylaxis is given to patients either as primary prophylaxis (PP) to prevent the initial episode or secondary prophylaxis (SP) to prevent recurrence. The evidence for this practice is based on outdated data published more than two decades ago. With the emergence of multidrug-resistant organisms (MDRO), routine SBP prophylaxis practices need to be re-evaluated. We aimed to investigate the impact of PP and SP on survival and the development of MDRO.
Method: Data was collected retrospectively for patients admitted with cirrhosis-related ascites from three tertiary centres in Australia between January 2022 and December 2023, with follow-up until death or December 2024.
Results: A total of 379 patients were included, 61% being male, a median (IQR) age of 62 (19) years, MELD-Nascoreof21(9) and follow-up of 1.2(1.9) years. Alcohol was the predominant aetiology (59%) for cirrhosis, and 73 (19.2%) had MDRO at baseline. On index admission,13 (3.4%) patients were already oSP,21(5.5%) were on PP, and 51(13.5%) patientswerediagnosedwithSBP.Postindexadmission,32(8.4%) and 48(21%) patients were commenced on PP and SP, respectively. During the follow-up, 52 (13.7%) patients had an episode of SBP. The mean SBP-free survival was similar between those on PP and those meeting criteria but not commenced on PP [2.9v2.7 years, c2=3.7, p=0.2], and with or without SP [2.2 v 2.0 years, c2=1.1, p=0.3]. Overall survival was similar between those not on SP [median (95%CI): 1.7 (1.22.1) years] and those on SP [median (95%CI):1.8 (1.5–2.2) years], c2= 0.5, p =0.5. Overall, the risk of developing MDRO culture-positive SBP during follow-up was significantly higher for those on either PP or SP [27.5% v 9.4%, c2=18.1, p<0.001]. The risk was only higher for those who had MDRO positive status at baseline [64.7%v 5.4%, c2=29.6, p< 0.001] but not for those with MDRO −ve status at baseline [17.5% v 10.3%, c2=2.5, p=0.1].
Conclusion: This study shows that in a real-world setting, the use of PP and SP for SBP does not have a survival benefit. In fact, those who have MDRO at baseline have worse survival outcomes if started on prophylaxis. Future randomised controlled trials are required to justify PP and SP for SBP.