Controlled Study of Streptococcal Endocarditis— A Problem of Feasibility

To the Editor.—  In back-to-back articles in The Journal, Karchmer et al (241:1801, 1979) and Malacoff et al (241:1807, 1979) called for a controlled, prospective, multicenter study to determine the optimal therapy for endocarditis caused by penicillinsusceptible streptococci. This recommendation is based on the fact that although careful retrospective studies and small prospective studies of various antibiotic regimens have been published, these regimens have not been compared by investigators using identical clinical and microbiological criteria in similar patient populations. Thus, questions remain about the relative safety and efficacy of the different regimens. At first glance, it seems logical to call for a prospective, multicenter trial where these variables can be controlled for each of the different treatment regimens and the superiority of one treatment regimen can be conclusively demonstrated. However, such a trial will probably never be done for the following logistical reasons.The case fatality of patients with penicillin-susceptible streptococcal