Procedure-related PD-peritonitis following gastroscopy and colonoscopy: A 15-year binational data-linkage study from Australia and New Zealand

Abstract

Aim

To describe the risk of post-procedural peritoneal dialysis (PD) peritonitis episodes following gastroscopy and colonoscopy over a 15-year period in Australia and New Zealand.

Background

The risk of post-procedural PD peritonitis is well recognised following colonoscopy and current ISPD guidelines recommend prophylactic antibiotics. Current guidelines do not recommend this prior to gastroscopy alone due to low level evidence.

Methods

Data linkage between hospital admission datasets and ANZDATA was performed in all jurisdictions in Australia and New Zealand, to identify all incident and prevalent patients on PD who underwent either gastroscopy or colonoscopy. ANZDATA reporting of peritonitis was interrogated to describe the risk of peritonitis occurring within seven days of either type of endoscopic procedure. The risk of permanent dialysis modality change at 30 days was estimated using competing risk analysis, as well as a description of the microbiology of these episodes.

Results

In total, 4433 admissions for gastroscopy (2447) colonoscopy (1191) or both (795) were identified amongst 3674 unique patients. In total there were 106 episodes of peritonitis, 51 of which occurred after gastroscopy alone. The event rate of post-procedure peritonitis was similar following gastroscopy (2.1%) and colonoscopy (2.4%) alone. Risk of peritonitis was similar when considering gastroscopy with and without intervention: 1.7% (95% CI 1.1–2.3) and 2.2% (95% CI 1.3–3.1), respectively. A similar trend was also present for colonoscopy with and without intervention: 2.8% (95%CI 1.0–4.7) and 2.7% (95%CI 1.5–4.0), respectively. The most important predictor of dialysis modality change after either procedure was having a polymicrobial infection (sHR 78.98, 95%CI 35.3–176.7). Regardless of procedure type, organisms of enteric origin along with coagulase negative staphylococcus were the most commonly identified.

Conclusions

There was a similar risk of peritonitis following either gastroscopy or colonoscopy over the time period. Further studies evaluating the effect and role of prophylactic antibiotics prior to gastroscopy alone are required.