Visual determination of MIC end points for azole antifungal agents can be complicated by the trailing growth phenomenon. To determine the incidence of trailing growth, we performed testing of in vitro susceptibility to fluconazole and itraconazole using the National Committee for Clinical Laboratory Standards broth microdilution M27-A reference procedure and 944 bloodstream isolates of seven Candida spp., obtained through active population-based surveillance between 1998 and 2000. Of 429 C. albicans isolates, 78 (18.2%) showed trailing growth at 48 h in tests with fluconazole, and 70 (16.3%) showed trailing in tests with itraconazole. Of 118 C. tropicalis isolates, 70 (59.3%) showed trailing growth in tests with fluconazole, and 35 (29.7%) showed trailing in tests with itraconazole. Trailing growth was not observed with any of the other five Candida spp. tested (C. dubliniensis, C. glabrata, C. krusei, C. lusitaniae, and C. parapsilosis). To confirm whether or not isolates that showed trailing growth in fluconazole and/or itraconazole were resistant in vitro to these agents, all isolates that showed trailing growth were retested by the sterol quantitation method, which measures cellular ergosterol content rather than growth inhibition after exposure to azoles. By this method, none of the trailing isolates was resistant in vitro to fluconazole or itraconazole. For both agents, a 24-h visual end point or a spectrophotometric end point of 50% reduction in growth relative to the growth control after 24 or 48 h of incubation correlated most closely with the result of sterol quantitation. Our results indicate that MIC results determined by either of these end point rules may be more predictive of in vivo outcome for isolates that give unclear visual end points at 48 h due to trailing growth.