The issue with ethnic difference in peritoneal dialysis outcomes is controversial but plausible, and so this study is an important one.

This is an observational study derived from a population-based registry in Malaysia. Specifically, the manuscript mentioned the National Renal Registry database (page 4) and we assume that as the source of data in this study. The population refers to 1986 (mostly prevalent) peritoneal dialysis patients from 15 adult and 5 pediatric centres. The study objective is to look for risk factors of peritonitis and the authors highlight the uniqueness of studying multi-racial Asian population. Multivariate analysis showed that race, severe obesity (body mass index ≥ 35), Staphylococcus aureus nasal carriage, hypoalbuminaemia (first serum albumin < 30 g/L) and distance from home to dialysis centres were independently predictive of peritonitis. Remarkably, Staphylococcus aureus peritonitis rate was significantly lower among Chinese subjects.

Can we be sure that there isn't chance for selection bias? In other words, could those who are at higher risk of peritonitis preferentially treated with peritoneal dialysis (and thus overrepresentation) in races other than Chinese? First, the income status (Table 2) showed a more economically advantageous group among Chinese patients. Could there be a tendency for the Chinese to receive better care? Second, was there chance for language barrier to hinder the peritoneal dialysis training and treatment education in the minority group? Third, and of prime interest, was there difference in centre size (say, larger peritoneal dialysis centres for Chinese patients)? Centre effect needs to be taken into consideration because dialysis centres with less than 40 peritoneal dialysis patients had been shown to be an independent factor associated with the risk for the first peritonitis in Brazil (Martin LC et al. Clin J Am Soc Nephrol 2011). In a previous analysis (Fang W et al. Nephrol Dial Transplant 2008), comparison was made between 256 Canadian and 240 Chinese incident peritoneal dialysis; a remarkably (and statistically significant) lower peritonitis rate was demonstrated among Chinese dialysis patients (one episode every 60.6 versus one in every 36.1 patient-months). The reported peritonitis rate is pretty close to that reported among Chinese patients in your study, with one episode every 51.8 patient-months. Interestingly, another study by Blake PG et al (Am J Kidney Dis 2000) showed a significantly better self-reported compliance with continuous ambulatory peritoneal dialysis among Asians than Hispanics and black. That is another important point of note.