Article Text


Evaluation of CHROMagar candida for rapid identification and Etest for antifungal susceptibility testing in a district general hospital laboratory
  1. J E Ambler,
  2. M Kerawala,
  3. A Yaneza,
  4. Y J Drabu
  1. Department of Microbiology, North Middlesex Hospital NHS Trust, London N18 1QX, UK

    Statistics from

    The incidence and clinical importance of fungal infections in immunocompromised patients is increasing, and the isolation of multiple yeast species from clinical specimens is not uncommon. Candida albicans remains the most frequently isolated yeast species; but others, inherently or potentially resistant to amphotericin B and azole compounds (C krusei, C parapsilosis, C lusitaniae, C tropicalis, and C glabrata), are also being reported.1

    Unlike sabouraud dextrose agar (SDA) (LabM, Bury, UK), CHROMagar candida (Mast Diagnostics, Bootle, UK), a chromogenic, differential culture medium, can detect mixed populations and facilitate rapid, accurate identification of C albicans.2

    The procedure for antifungal susceptibility testing by the broth based NCCLS reference method M27-A is time consuming.3 The Etest (AB BIODISK, Solna, Sweden), a simple alternative agar based quantitative diffusion method, can be readily incorporated into a clinical laboratory routine.4

    Over a period of 10 weeks, 31 yeast isolates were recovered on SDA, from 22 patients in intensive care, HIV, and Oncology/Haematology units of the North Middlesex Hospital NHS Trust. Colony pigmentation, following subculture on to CHROMagar candida, was assessed after 48 hours incubation at 37°C. Two specimens yielded mixed populations. The identification of the isolates was confirmed using API 20C AUX (Bio Mérieux, Basingstoke, UK).

    Twenty five isolates, yielding green colonies, were identified as C albicans (germ tube positive). Six isolates, yielding non-green colonies and germ tube negative, were identified as C glabrata (four), C parapsilosis (one), and C norvegensis (one; unidentified by API 20C AUX, but confirmed by Centraalbureau voor Schimmelcultures, Delft).

    Amphotericin B, fluconazole, and itraconazole MICs were determined by our laboratory (laboratory 1) using the Etest on RPMI 1640 containing 2% glucose and MOPS buffer (RPMI) and casitone agar (Cambridge Diagnostic Services Ltd, Cambridge, UK). The performance of commercially bought agar (RPMI and casitone) was within the recommended specifications. MIC endpoints were read after 24, 48, and 72 hours incubation at 37°C, and adjusted to the next upper twofold value if MICs evaluated by Etest were in between dilutions. Etest reproducibility was examined by another laboratory (laboratory 2; AB BIODISK, Solna, Sweden) and MICs correlated with semi-automated spectrophotometric readings of microdilution broth cultures by a third laboratory (laboratory 3; Janssen Research Laboratory, Beerse, Belgium).5 The quality control strains used by laboratory 1 were C albicans ATCC 90028 and C glabrata ATCC 90030; laboratories 2 and 3 used C parapsilosis ATCC 22019 and C krusei ATCC 6258 as controls. The results obtained with these strains were within the expected limits; the exception was itraconazole, which was higher by one dilution in laboratory 1 and lower by one dilution in laboratory 3. The MIC ranges (mg/litre) defined by NCCLS3 and recommended by AB BIO DISK (shown in parenthesis) are as follows: C parapsilosis ATCC 22019 were 0.25 to 1 (2), 2 to 8 (16), and 0.06 to 0.25; C krusei ATCC 6258 were 0.5 to 2, 16 to 64 (≥ 256), and 0.12 to 0.5; C albicans ATCC 90028 were 0.25 to 2, 0.25 to l, and ≤ 0.125 to 0.5; and C glabrata ATCC 90030 were 0.5 to 2, 8 to ≥ 32, and 0.5 to ≥ 32 for amphotericin B, fluconazol, and itraconazole, respectively.

    Interlaboratory reproducibility (results within two doubling dilution difference) of Etest MICs, determined after 48 hours on RPMI and casitone agar, was 100% and 97% (amphotericin B), 97% and 90% (fluconazole), and 90% and 87% (itraconazole), respectively. The correlation between the Etest MICs using RPMI and casitone agar at 48 hours and the spectrophotometric method was 94% and 77% (amphotericin B), 100% and 65% (fluconazole), and 32% and 13% (itraconazole), respectively. Etest MICs on RPMI were more reproducible and correlated better with the spectrophotometric method than those on casitone agar (table 1). Etest MICs for itraconazole were greater than those obtained by the spectrophotometric method. Trailing endpoints (fig 1) did not hinder the interpretation of the fluconazole Etests, and isolates exhibiting itraconazole resistance by the Etest were not all confirmed by the spectrophotometric method. The itraconazole discrepancy between Etests and broth spectrophotometry is high, and further work is needed to obtain a better correlation. The correct selection of MIC end points is crucial, and might have had an impact on the results.

    The rapid identification and accurate susceptibility testing of yeast would help to modify treatment and influence clinical outcome. We conclude that CHROMagar candida and Etest are worth considering in a district general hospital setting.

    Table 1

    MIC ranges for yeast isolates as determined by Etest and spectrophotometric broth microdilution methods

    Figure 1

    Fluconazole (FL): change of morphology at endpoint. MIC, 48 mg/litre; itraconazole (IT): sharp end point. MIC, 0.5 mg/litre.


    We would like to thank Å Karlsson (laboratory 2) and F Odds (laboratory 3) for their valuable comments and for performing the susceptibility testing.


    View Abstract

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.