In vitro susceptibility of Cryptococcus neoformans isolates to five antifungal drugs using a colorimetric system and the reference microbroth method.

Minimum inhibitory concentrations (MICs) of amphotericin B, 5-flucytosine, fluconazole, itraconazole and ketoconazole were determined against 42 clinical isolates of Cryptococcus neoformans var. neoformans using the Alamar YeastOne colorimetric method and the NCCLS reference microdilution method. No strains with resistance to amphotericin B, itraconazole or ketoconazole were detected with either method. Using the reference method, the MICs of fluconazole were >/= 64 mg/L, whereas using the colorimetric method all MICs were >/=16 mg/L. The MIC values of 5-flucytosine were also higher using the reference method (8-16 mg/L for 32% of isolates) compared with the colorimetric method. The percentage of agreement between the methods, using a difference of two dilutions, was 70.7% for itraconazole, 73.2% for amphotericin B, 80% for fluconazole, 88% for 5-flucytosine and 95% for ketoconazole. Overall, we conclude that for fluconazole and 5-flucytosine, in a low but not insignificant number of isolates, results with the two methods are discordant, some isolates being found sensitive with the colorimetric test, but resistant with the reference method.

[An evaluation of blood cultures obtained via a central venous catheter]. / Valoración de los cultivos de sangre obtenidos mediante catéter venoso central.

BACKGROUND: The aim of this study was to determine the use of blood samples obtained through central venous catheters (CVC) in the diagnosis of bacteremias in patients admitted to an intensive care unit (ICU). METHODS: A prospective randomized study was carried out in which 125 blood samples obtained by CVC were compared with 125 samples simultaneously obtained by venipuncture. RESULTS: In 105 sample pairs (84%) the results of both samples were in agreement (Kappa index 0.61). The sensitivity of the technique in the diagnosis of bacteremia was 76.5%, specificity 86.8% positive predictive value 68.4% and negative predictive value 90.8%. No differences were found in the efficacy of the technique carried out by venous punction or that of the samples obtained through catheters inserted in subclavian, femoral and antecubital veins (p = 0.99). CONCLUSIONS: Blood cultures obtained through central venous catheters constitute a technical alternative for the diagnosis of bacteremia in cases in which venipuncture is difficult or of doubtful asepsis.