Fungal Blood Cultures in Patients with AIDS
ABSTRACT & COMMENTARY
Synopsis: Fungal blood cultures did not assist in diagnosing fungal disease in patients with AIDS,and introduced confusion because of the isolation of filamentous moulds, all of which proved to be contaminants.
Source: Mess T, et al. Utility of fungal blood cultures for patients with AIDS. Clin Infect Dis 1997;25:1350-1353.
Mess and colleagues set out to define the clinical utility of fungal blood cultures for patients with AIDS. The charts of 322 patients seen over a two-year period were retrospectively reviewed for evidence of invasive fungal infection. In addition to routine blood cultures, a total of 1162 fungal blood cultures had been taken, and 62 fungal and/or routine blood cultures yielded fungi. There were 23 patients with invasive fungal infection, and fungaemia was diagnosed by routine blood cultures in 15 cases (7 with Cryptococcus neoformans, 3 with Candida albicans, 2 with Candida glabrata, and 1 each with Candida parapsilosis, Coccidioides immitis, and Histoplasma capsulatum). The other eight cases were diagnosed from specimens other than blood (3 bone marrow aspirate, 2 bronchoscopy, 2 antigen, and 1 skin biopsy) and included three cases of Coccidioides immitis, two of Cryptococcus neoformans, and one each of Aureobasidium spp., Paecilomyces spp., and Candida parapsilosis. Fungal blood cultures yielded the etiological agent from 12 of those also diagnosed by routine blood cultures and did not identify a single case not already detected by conventional blood cultures. Moreover, filamentous fungi were isolated exclusively from fungal blood cultures from a further 11 patients and were all considered contaminants since none of these patients had evidence of invasive fungal disease. Thus, all the invasive fungal infections were diagnosed by means other than fungal blood cultures, suggesting that their routine use of FBC should be re-evaluated.
COMMENT FROM J. PETER DONNELLY, PhD
Fungal blood cultures did not come out of this study well. Granted, Mess and colleagues were not enamored with them from the outset, considering them "labor intensive and expensive and representing only one of many tests that can result in the diagnosis of invasive fungal infection." Their dim view proved to be more than justified by the results since these cultures contributed nothing to the diagnosis of the 23 genuine cases and might have led to misdiagnosis in the 11 cases whose blood yielded contaminants. In performance jargon, the sensitivity of FBCs was only 60%, while the specificity was high at 96% and was more or less the same for routine blood cultures.
The broad use of the term "fungal infection," although commonplace and handy, is nonetheless misleading. It is deceiving, especially when considering fungemia since this is seldom a feature of diseases cause by the filamentous fungi that accounted for all the contaminants. The nature of cryptococcosis, histoplasmosis, and coccidiomycosis are also such that means other than blood cultures stand a better chance of helping with the diagnosis. Thus, it is only when attempting to diagnose candidosis that blood cultures are expected to have a reasonable chance to be of any real value and, even here, there appeared to have been no advantage in using routine dedicated blood cultures. So, when diagnosing invasive fungal infection, one is compelled to rely on means other than blood cultures.
However, this is only one aspect of the utility of fungal blood cultures. Another equally important reason for taking such cultures is to exclude the possibility of fungal disease, thereby reducing the pressures for continuing or even starting empirical therapy. Since fungal blood cultures are designed to allow better recovery of fungi of whatever complexion, failure to isolate any actually does provide us with a means of excluding fungemia with a very high degree of confidence since the negative predictive value will be very high (97% in this study). We do not know how often empirical therapy was started, but the associated costs might outweigh the $567 per patient incurred by the use of fungal blood cultures in the authors’ hospital. Were negative cultures to lead to early discontinuation of empirical therapy or even to witholding it altogether, the cost of taking a culture to exclude disease might actually prove less than the cost of empirical therapymaking such cultures both useful and cost-effective. Thus, before writing off fungal blood cultures entirely, it would be worthwhile attempting to evaluate the utility of fungal blood cultures in excluding fungal disease and ascertaining the effect, if any, on the institution of empirical therapy.
In the study by Mess and colleagues, fungal blood cultures contributed nothing to the diagnosis of genuine contaminants in:
a. two cases.
b. 40 cases.
c. 23 cases.
d. 12 cases.
e. 50 cases.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content