Abstract
Determination of Candida albicans mannan levels by an enzyme-linked immunosorbent assay was compared with dry weight measurement and viable counting as a means to estimate growth of the fungus in broth cultures. Levels of extracted mannan showed good correlation with fungal biomass (dry weight) (r = 0·984) and, although biomass and mannan levels correlated with viable counts, the correlation was not as strong (r = 0·954-0·959). The technique offered an accurate alternative to viable counts in the brain and kidney when mannan estimation was used to follow the distribution of C. albicans in an experimental model of systemic candidosis in mice. Mannan levels were less useful in following the course of the infection in the liver and spleen where accumulation of free and serum-bound mannan was a problem. Mannan estimation would therefore seem to offer an accurate alternative to viable counts for determining the quantity of Candida both in vitro and in certain infected organs of experimental animals, where measurement of fungal biomass is not feasible and viable counts can be grossly inaccurate.