TABLE 2.

Meaningful test methods and conditions for antifungal susceptibility testing

Organism	Drug	Methods	Interpretation	References
Candida spp.	Fluconazole	M27-A and methods that give concordant results	S, <8 μg/ml; S-DD, >8 μg/ml and ≤32 μg/ml; R, >32μg/ml; data are best supported for mucosal disease, but the general principles also appear to hold for invasive disease; adequate drug dose is critical as the MIC rises; C. krusei is assumed intrinsically resistant and should not be tested	32, 108, 135, 182
	Itraconazole	M27-A and methods that give concordant results	S, ≤0.125 μg/ml; S-DD; >0.125 μg/ml and ≤0.5 μg/ml; R, >0.5 μg/ml, data apply only to mucosal disease; ensuring adequate absorption by use of the solution is often critical to drug efficacy	135, 182
	Ketoconazole	M27-A-like methods	No specific breakpoints proposed, but aggregate data suggest that isolates with an MIC of >0.125 μ/ml by M27-A are less likely to respond	27, 123, 128, 191, 205
	Flucytosine	M27-A-like methods	S, ≤4 μg/ml; I, >4 μg/ml and ≤16 μg/ml; R, >16 μg/ml; these breakpoints are based largely on historical data and animal models but appear rational based on the available clinical and pharmacokinetic data	7, 70, 89, 135
	Amphotericin B	M27-A-like methods that use AM3 broth, Etest, or MFC methods	Significant controversy and difficulty; agar-based methods appear to have the most potential, in general, and isolates with amphotericin B MICs of >0.5 μg/ml are probably resistant; specific breakpoints have NOT been proposed, and intralaboratory variation makes use of reference isolates with known resistance mandotory; fortunately, resistance appears uncommon among the four most common species (106)	34, 137, 180, 221
C. neoformans	Fluconazole	Modified M27-A (YNB broth, higher inoculum)	No specific breakpoints proposed; Data are limited but consistent: MICs of ≥2–4 μg/ml by this method appear to predict a greater chance of failure; failure definitely appears more likely as the MIC reaches 16 μg/ml	1, 93; Witt et al., abstr. a
	Flucytosine	M27-A-like methods	No specific breakpoints proposed, but breakpoints similar to those for Candida spp. (see above) would appear rational
	Amphotericin B	M27-A-like methods that use AM3 broth, Etest	Significant controversy and difficulty; agar-based methods have the most potential in general. Measurement of fungicidal concentrations may have value; no specific breakpoints proposed—testing should incorporate reference isolates with known resistance.	117, 185, 186
Mold fungi	Any	Unknown	M38-like methods correlate for itraconazole and A. fumigatus; correlations for amphotericin B are ill-defined	See text

^aM. D. Witt, L. A. Mortara, R. A. Larsen, R. J. Lewis, J. E. Edwards, and M. Ghannoun, Abstr. IXth Int. Coaf. AIDS, abstr. WS-B12-4, 1993.