RESUMEN
BACKGROUND: In peritoneal dialysis-related peritonitis (peritonitis), delayed antibiotic therapy is associated with adverse outcomes. Identifying barriers to timely treatment may improve outcomes. AIM: To determine the impact of radiological investigations on treatment delay and predictors of hospitalisation and length of stay (LOS). METHODS: Retrospective review of patients with presumed peritonitis in Western Australia. RESULTS: In 153 episodes of peritonitis, 79 (51.6%) resulted in admission with a median LOS of 3 days (Q1, Q3: 1, 6). In a multivariable model, significant predictors of admission were abnormal exit-site (odds ration (OR) 5.7; 95% confidence interval (CI): 1.4, 23.6; p = 0.02), failure to detect a cloudy bag (OR 11.9; 95%CI: 3.2, 44.7; p < 0.001), female sex (OR 3.3; 95% CI: 1.4, 9.7; p = 0.027), radiological imaging within 24 h (OR 8.8; 95% CI: 2.2, 34.8; p = 0.002) and contact with ambulant care facility (OR 0.32, 95% CI: 0.11, 0.98; p = 0.04). Imaging within 24 h of presentation occurred in 41 (27%) episodes of peritonitis, mostly plain X-rays (91%), of which 83% were clinically irrelevant. Imaging performed within 24 h of presentation increased the median time to antibiotic treatment (2.9 h (Q1, Q3: 1.6, 6.4) vs 2.0 h (Q1, Q3: 1, 3.8; p = 0.046)). Imaging performed prior to administering antibiotics significantly increased the median time to treatment (4.7 h (Q1, Q3: 2.9, 25) vs 1.5 h (Q1, Q3: 0.75, 2.5; p < 0.001)) in those where imaging followed antibiotic treatment. CONCLUSIONS: Half of all presentations with peritonitis result in hospital admission. Radiological imaging was associated with an increased risk of hospitalisation, potentially contributes to treatment delay, and was mostly clinically unnecessary. When required, imaging should follow antibiotic therapy.