Exit-site infection and subsequent incident risk of peritoneal dialysis-associated peritonitis: a retrospective cohort study with meta-analysis

Background To date, evidence to support the risk estimates of peritoneal dialysis (PD)-associated peritonitis after the occurrence of exit-site infection (ESI) is limited. We aimed to investigate the association between ESI and the subsequent risk of PD-associated peritonitis. Methods We conducted a multicenter, retrospective cohort study of adult incident PD patients in Thailand between January 2006 and September 2022. We compared matched pairs (1:1) of individuals who experienced the first episode of ESI with their counterparts who did not experience any ESIs during follow-up. The subsequent risk of PD-associated peritonitis was assessed using multivariable Cox proportional hazards models with shared frailty correction to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Landmark analysis corresponding to the designed landmark time after ESI was conducted (within 1 month, 1–3 months, 3–6 months, 6–9 months, and 9–12 months). Secondary analyses explored causative organisms and ESI severity. An ancillary meta-analysis of all available evidence was summarized using a random-effect model. Results Among 5,368 incident PD cases in the cohort, a total of 610 patients were eligible, including 305 incident cases of the first episode of ESI and 305 corresponding matched index cases. The study cohort had a mean age of 59.3 years; most patients were male (56.7%), and most received continuous ambulatory PD (89.0%). Regarding adjustment for sociodemographic and clinical characteristics, PD practices, and laboratory profiles, the risk of subsequent PD-associated peritonitis was significantly increased among individuals who had experienced an ESI (HR, 5.30; 95% CI, 3.75–7.50). Following landmark analysis, the risk of PD-associated peritonitis increased at peak approximately 18-fold (within 1 month) and gradually decreased to 2-fold over the designed 9–12 months. Secondary analyses revealed consistent findings across the causative organism and severity strata of ESIs (HRs ranging from 5.17 to 5.98). An ancillary meta-analytic synthesis supported the main findings [3 included cohort studies; n = 1775; HR, 3.37 (95% CI, 1.35–8.41)]. Conclusion Regarding retrospective cohort analysis with ancillary meta-analytical synthesis, individuals experienced with ESI are at risk of developing PD-associated peritonitis. Effective strategies to prevent PD-associated peritonitis following an ESI episode are warranted to promote patient health and sustainable PD programs.