In conclusion, the recommendations provided in this document represent a distillation of various experiences, as well as data obtained from published studies. It is hoped that this compilation will provide a basis upon which future developments and advances can be made in the therapeutic approach to infectious complications of peritoneal dialysis.

In conclusion, the recommendations provided in this document represent a distillation of various experiences, as well as data obtained from published studies. It is hoped that this compilation will provide a basis upon which future developments and advances can be made in the therapeutic approach to infectious complications of peritoneal dialysis.

D uring the past 15 years, CAPD has come into its own as a kidney replacement therapy. From the beginning, infectious peritonitis has always been a nagging problem resulting in trials of multiple pre ventative and therapeutic modalities. Giant strides have been made in the reduction of peritonitis episodes, mainly by technologic improvement, but there remains the need for effective, nontoxic, inexpensive antimicrobial agents to treat those that acquire infectious complications. Almost simultaneous with the development of CAPD, there has been an explosion of new antimicrobial agents available to the clinician. Because of this, many trials of a variety of antimicrobial agents have been used in CAPD-associated peritonitis in an attempt to find a "better bullet." In this issue of PDI, Millikin et al. (page 56-66) have undertaken a rather monumental task to critically summarize the clinical trials of antimicrobial peritonitis therapy. They rightfully point out that few trials are prospective, frequently involve small numbers of patients and have often been noncomparative. Modes of drug delivery, therapy duration and reports of toxicity and efficacy have been erratic and/or diverse. Some may be utilizing "pooled data," and this can be fraught with pitfalls. Nevertheless, based on the types of reports in the literature, this method for review purposes is probably acceptable. Perhaps this critique will stimulate more scientifically controlled comparative studies containing larger numbers of patients. From the authors' review, it seems clear that for Grampositive peritonitis vancomycin delivered either i.v. or i.p. is effective and its pharmacokinetics are very desirable. It is relatively nontoxic, although some products have caused chemical peritonitis when administered primarily as a large loading dose. Of concern are reports of some Grampositive organ isms, mainly Enteracacci and coagulase-negative Staphylacacci, that are resistant to vancomycin. Further, other glycopeptides under investigation do not appear to have any advantage over vancomycin. Clearly, other compounds need to be screened and developed in the likely event that resistant species will become more widespread. Optimal management of Gram-negative rod and fungal peritonitis remains to be established. Most of the current recommendations are based on what seems logical rather than on good data. It is unlikely that anyone center will encounter sufficient numbers of cases of these entities to perform appropriate studies; multicenter cooperative investigations will be necessary to provide appropriate answers. The authors conclude that ceftazidime and vancomycin would be the "apparent optimal" empiric regimen for peritonitis. While this seems logical, there is insufficient data on Gram-negative peritonitis. This would be a less toxic regimen, but certainly more expensive. Finally, the quinolones appear to be promising, based on published data and future analysis that will occur in this journal. There will probably be no magic antimicrobial bullet curing all infectious peritonitis in the near future. But further advances perhaps in care of exit sites and/or the development of catheter materials or methods that prevent adherence of microbial agents to foreign bodies may further reduce the incidence of peritonitis. E. Dale Everett

1Division of Nephrology, Hennepin County Medical Center, University of Minnesota, Minneapolis, Minnesota, 2Department of Medicine, University of Missouri, Columbia, Missouri, 3Division of Pediatric Nephrology, UCLA Center for Health Sciences, Los Angeles, California, 4Kidney Disease Program, University of Louisville, Louisville, Kentucky, 5Department of Microbiology, Toronto Western Hospital, Toronto, Ontario, Canada, 6Division of Infectious Diseases, Hennepin County Medical Center, University of Minnesota, Minneapolis, Minnesota, 7Division of Nephrology, Department of Medicine, Manchester RoyalInfirmary, Manchester, U.K., and BDivision of Pharmacy Practice, School of Pharmacy,

In conclusion, the recommendations provided in this document represent a distillation of various experiences. as well as data obtained from published studies. It is hoped that this compilation will provide a basis upon which future developments and advances in the therapeutic approach to CAPD patients with peritonitis can be made. Paramount among these will be the development of more efficient and rapid diagnostic techniques, as well as new and superior peritonitis regimens utilizing precisely derived dosing recommendations that minimize toxicity and yet provide a high degree of efficacy. In addition, new insights into the clinical approach related to catheter management should be developed, particularly as they pertain to exit and tunnel infections, as well as the optimal interval before catheter reinsertion can be safely performed.