Etiology and mortality of spontaneous bacterial peritonitis in liver transplant recipients: A cohort study

Abstract

Spontaneous bacterial peritonitis (SBP) in liver transplantation (LT) recipients who progress to cirrhosis has received little attention. We investigated the adequacy of empirical treatment with third-generation cephalosporins for SBP in this population and the impact of transplantation on the evolution of the infection. We performed a cohort study with 138 SBP episodes: 19 in LT patients and 119 in non-LT patients. The etiology of SBP was identified for 73.7% of the episodes in LT patients and for 38.7% of the episodes in non-LT patients (P 5 0.004). The main microorganisms in recipients were Escherichia coli (35.7%) and Streptococcus pneumoniae (21.4%). The etiologies did not differ in non-LT patients. The cephalosporin sensitivity was similar in the 2 groups (85.7% versus 78.4%, P 5 0.7). LT recipients developed renal failure (57.9% versus 25.2%, P 5 0.004) and encephalopathy (42.1% versus 22%, P 5 0.08) more often than non-LT patients, and the mortality rates during episodes (52.6% versus 13.4%, P < 0.001) and at 6 months (70.6% versus 34.7%, P 5 0.005) were higher. According to a multivariate analysis, the mortality-associated risk factors at diagnosis were a Model for End-Stage Liver Disease (MELD) score > 18 odds ratio (OR) 5 6.1 and being an LT recipient (OR 5 4.45). At 6 months, the risk factors for mortality were a MELD score > 18 (OR 5 3.08), being an LT recipient (OR 5 3.47), a known etiology (OR 5 2.08), and the presence of hepatocellular carcinoma (OR 5 3.73). Liver Transpl 20:856-863, 2014. V C 2014 AASLD.Received January 14, 2014; accepted April 3, 2014.The treatment of choice for end-stage cirrhosis is liver transplantation (LT). Nonetheless, the recurrence of the primary disease, the development of liver cirrhosis (especially in patients with hepatitis C virus, in whom reinfection almost always occurs), and associated complications are common causes of late graft loss and have an impact on morbidity and mortality. [1][2][3][4] It is well recognized that spontaneous bacterial peritonitis (SBP) is associated with a high mortality rate (approximately 20%) in patients with cirrhosis, 5-7 and the first SBP episode predicts a significant decrease in survival, even in the short term, with 1-year survival rates ranging from 28.5% to 93%. [8][9][10][11][12][13] Nonantipseudomonal third-generation cephalosporins (3CP) are the treatment of choice for SBP, 5,6,10 but several recent studies have reported that bacterial resistance to 3CP has reached a disturbing level of 43%. 6,14 Nosocomial and health care-related infections are often associated with multidrug-resistant microorganisms such as Enterobacteriaceae producing extended-spectrum beta-

Outcomes of spontaneous bacterial peritonitis in liver transplant recipients with allograft failure

“…Although a history of treated pre‐first LT SBP or infection does not appear to be associated with post‐LT mortality , the impact of SBP on candidacy for re‐LT is not known and was not explored in the study by Perez‐Cameo et al. . In our study, only 10 patients who developed SBP post‐LT were listed for re‐LT, but 5 died before making it to transplant.…”

“…The majority of patients (85%) died within a very short time (median time to death 32.5 days) after the onset of the first episode of SBP. A recent study published by Perez-Cameo et al (10) also showed a higher mortality rate in post-LT recipients with SBP in comparison to non-LT recipients (52.6% vs. 13.4%, P ≤ 0.001); independent risk factors for mortality at the time of SBP diagnosis included MELD score >18 and being an LT recipient. Re-LT accounts for 5-22% of all LTs, and it is widely accepted that it is associated with decreased survival compared to a first LT (11)(12)(13)(14).…”

“…A recent study published by Perez‐Cameo et al. also showed a higher mortality rate in post‐LT recipients with SBP in comparison to non‐LT recipients (52.6% vs. 13.4%, P ≤ 0.001); independent risk factors for mortality at the time of SBP diagnosis included MELD score >18 and being an LT recipient.…”

“…Re-LT accounts for 5-22% of all LTs, and it is widely accepted that it is associated with decreased survival compared to a first LT (11)(12)(13)(14). Although a history of treated pre-first LT SBP or infection does not appear to be associated with post-LT mortality (7,15), the impact of SBP on candidacy for re-LT is not known and was not explored in the study by Perez-Cameo et al (10). In our study, only 10 patients who developed SBP post-LT were listed for re-LT, but 5 died before making it to transplant.…”

Bacterascites confers poor patient prognosis beyond MELD prediction

“…Given the accepted paradigm that an episode of SBP increases mortality beyond that which the MELD score predicts, patients were grouped into MELD categories ≤ 9, 10-19, 20-29, 30-39, and ≥ 30 and 3-month mortality in patients with SBP and BA were compared with published data. [13][14][15][16] Mortality was substantially higher across all MELD groupings for both SBP and BA when compared with the predicted mortality calculated by the MELD score alone (Table 4).…”