National hospital survey of anaerobic culture and susceptibility testing methods: results and recommendations for improvement.

The methods for performing anaerobic bacterial isolation and identification continue to change and improve. Anaerobic susceptibility testing has become controversial, and method-dependent variability has been noted. To assess the status of clinical anaerobic bacteriology in the United States, we surveyed, by means of a questionnaire, 120 hospitals, selected at random, with bed capacities of 200 to 1,000, and we received responses from 88 (73%). All hospitals performed cultures for anaerobes. The media and methods used for transport, initial processing, incubation, and identification varies between the different regions in the United States. Thirty percent of laboratories did not perform susceptibility studies, 16% used a reference laboratory, and 54% performed them in house. For half the laboratories, susceptibility testing was performed on isolates depending on the source; in this case, blood cultures were tested by 97% of the laboratories, serious infections were tested by 60%, sterile body sites were tested by 73%, pure cultures were tested by 47%, and tests were done by physician request by 39%. For laboratories doing testing, the broth disk method, no longer sanctioned by the National Committee for Clinical Laboratory Standards, was used most often (56%), followed by microdilution (33%), beta-lactamase testing (25%), macrotube dilution (2%), and agar dilution (2%). The antimicrobial agents tested were as follows: penicillin-ampicillin, 94%; clindamycin, 94%, metronidazole, 90%; chloramphenicol, 80%; cefoxitin, 76%; tetracyclines, 51%; and erythromycin, 45%. All other agents were tested by less than or equal to 25% of laboratories; the methods used could be improved to make the results more timely and consequently more clinically relevant.