Intraperitoneal teicoplanin in CAPD peritonitis

Sir: We have noted with some concern that Schinella and colleagues (1) find that teicoplanin is ineffective in an intermittent intraperitoneal schedule for the treatment of CAPD peritonitis. This is at variance with our experience of more than 2 years' use in such patients. During that time we have reported our pharmacokinetic data (2) in detail, as well as results of a pilot therapeutic study in 12 patients (3) and an interim report of a randomized trial of teicoplanin compared with vancomycin (4). A total of 86 infections have now been treated in the latter trial: bacteriologi cal cure rates were 75% with teicoplanin and 86% with vancomycin (not significantly different). The treatment schedule with teicoplanin was as follows: patients received 400 mg i.v. on entry, and for 1 week 40 mg teicoplanin was added to each 2 L bag ( 4 bags are used every 24 h). Thereafter, the dosage was tapered: during the second week, 40 mg teicoplanin was added to alternate bags, and in the third week 40 mg teicoplanin was added to the overnight dwell bag only. Our success rate using this regime has been very similar to that observed with vancomycin. The majority of the small numbers of therapeutic failures have been due to Staphylacaccus aureu8, where tissue involvement (tunnel infections) has been a clinical feature. We are now exploring the possibility of reducing the total dosage of teicoplanin. Obvious means of doing this are to dispense with the initial i. v. dose, and to shorten the overall treatment period from 21 to 14 days. The consequence of therapeutic failure in this type of patient are potentially serious, as they involve the possible necessity of replacing the peritoneal catheter and the risk of losing the dialysing capacity of the peritoneal membrane. Therefore, we feel it is important to err on the side of overdosing rather than underdosing. In this context we would recommend that teicoplanin be added to every bag during the first week of treatment, rather than to alternate bags as suggested by Schinella et at. (1).