NDT Advance Access originally published online on December 21, 2007
Nephrology Dialysis Transplantation 2008 23(5):1769-1770; doi:10.1093/ndt/gfm902
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Occurrence of peritonitis in APD versus CAPD: methodologic problems
Correspondence and offprint requests to: E-mail: nessims{at}smh.toronto.on.caSir,
We read with interest the article by Rabindranath et al. ‘Automated vs continuous ambulatory peritoneal dialysis: a systematic review of randomized controlled trials’ [1], especially with reference to the occurrence of peritonitis on continuous ambulatory peritoneal dialysis (CAPD) versus automated peritoneal dialysis (APD). The authors report both the relative risk and the rate ratio for peritonitis based on three randomized controlled trials involving a total of 139 patients.
While the review found a significant reduction in peritonitis rates with APD relative to CAPD, no reduction in the relative risk was found. The authors relate this apparent discrepancy to the fact that some patients may have had more than one episode of peritonitis. It is important to note that comparing peritonitis rates is far more appropriate than determining relative risk, because a peritonitis rate in any given individual incorporates information on the number of episodes of peritonitis as well as the duration of follow-up on peritoneal dialysis (PD). In contrast, a relative risk simply determines if one group is at a higher risk of developing peritonitis than another group, irrespective of the number of episodes per patient or the follow-up time. For example, if one patient has three peritonitis episodes within 6 months of initiating PD and another has one peritonitis episode 2 years after starting PD, a relative risk analysis would consider these patients to be equivalent with respect to the outcome because the occurrence of peritonitis is registered only as a ‘yes/no event’ (despite the fact that most clinicians would agree that these patients are quite different from one another). Based on these concerns, we would argue that relative risk is not a meaningful way to assess the association of any variable with peritonitis.
With regard to the authors’ conclusion that APD is associated with a lower peritonitis rate than CAPD, we note that this conclusion is based on only two randomized controlled trials. Of these trials, the study by Bro et al. [2] reported only three total episodes of peritonitis (one on APD, two on CAPD), likely related to the relatively short (6 month) follow-up time in this study. Since so few peritonitis episodes occurred in this trial, the results of the ‘review’ of the effect CAPD versus APD on peritonitis are almost entirely driven by the findings of the other trial by De Fijter et al. from 1994, showing a lower peritonitis rate with APD. The conclusions of the review with respect to peritonitis should therefore be interpreted with caution. While we commend the authors for trying to synthesize the existing literature on CAPD versus APD into a systematic review, the paucity of data available from randomized trials with respect to peritonitis limits the value of this exercise.
Conflict of interest statement. Dr. Bargman has received speaker honoraria from Baxter, and Dr. Nessim received educational fellowship funding from Baxter in 2006.
1 Division of Nephrology, St. Michael's Hospital, 30 Bond Street, 8CC Toronto, ON M5B 1W8 2 Division of Nephrology, Toronto General Hospital, University Health Network , 200 Elizabeth Street, 8NU-840 Toronto, ON M5G 2C4, Canada
References
- Rabindranath KS, Adams J, Ali TZ, et al. Automated vs continuous ambulatory peritoneal dialysis: a systematic review of randomized controlled trials. Nephrol Dial Transplant (2007) 22:2991–2998.
[Abstract/Free Full Text] - Bro S, Bjorner JB, Tofte-Jensen P, et al. A prospective, randomized and multicenter study comparing APD and CAPD treatment. Perit Dial Int (1999) 19:526–533.
[Abstract/Free Full Text] - De Fijter CW, Oe LP, Nauta JJ, et al. Clinical efficacy and morbidity associated with continuous cyclic compared with continuous ambulatory peritoneal dialysis. Ann Int Med (1994) 120:264–271.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||