Purpose: Serologic tests to detect invasive candidiasis generally have been unreliable. We prospectively evaluated the clinical utility of a new, promising commercial latex particle agglutination test (i.e., Cand-Tec, Ramco Laboratories, Inc., Houston, Texas). This assay detects Candida antigens in serum.
Patients and methods: We examined the reliability of Cand-Tec to diagnose invasive candidiasis in 142 consecutive in-patients intensively treated with chemo-radiation therapy. Serum samples were collected at admission and then weekly, until the patients' death or hospital discharge. Evaluation for clinical utility was done using various reference titers. Twenty-nine patients had invasive candidiasis whereas 113 patients did not have documented invasive candidal infections.
Results: At a titer of 1:8, the Cand-Tec test had sensitivity of 38%, specificity of 90%, positive predictive value of 50%, and negative predictive value of 85%. Weekly use of the Cand-Tec test did not improve early detection of invasive candidiasis, providing only a mean interval of 0.4 day from the first positive Cand-Tec result to a definitive diagnosis of invasive candidiasis by blood culture, tissue biopsy, or autopsy. In addition, surveillance cultures from the oropharynx or stool were not helpful in identifying those patients who would develop an invasive fungal infection.
Conclusion: In the context of current clinical management strategies for suspected fungal infection, the Candida antigen detection assay (Cand-Tec) is not a reliable method for diagnosis of deep candidiasis in neutropenic patients. Until better methods of early detection are available, patients at high risk for the development of invasive candidiasis should continue to receive empiric antifungal agent therapy.