False-positive cerebrospinal fluid cryptococcus antigen in Libman-Sacks endocarditis

False positive detection of serum cryptococcal antigens due to insufficient sample dilution: A case series

At present, with the development of technology, the detection of cryptococcal antigen (CRAG) plays an increasingly important role in the diagnosis of cryptococcosis. However, the three major CRAG detection technologies, latex agglutination test (LA), lateral flow assay (LFA) and Enzyme-linked Immunosorbent Assay, have certain limitations. Although these techniques do not often lead to false-positive results, once this result occurs in a particular group of patients (such as human immunodeficiency virus patients), it might lead to severe consequences.

Sensitivity of Cerebrospinal Fluid Cytology for the Diagnosis of Cryptococcal Infections: A 21-Year Single-Institution Retrospective Review

Objectives

Cryptococcal meningoencephalitis is the most common fungal infection of the central nervous system diagnosed by cerebrospinal fluid cytology (CSF) studies. Existing literature suggests that routine CSF cytomorphologic evaluations are exquisitely specific; however, less is known about their sensitivity.

Methods

An electronic record review of the cytopathology and microbiology files was conducted for the 21-year interval from January 1, 1995, through December 31, 2015.

Results

In 21 years, 12,584 CSF samples were processed in the laboratory. Of these, 24 (0.2%) were reported positive for cryptococcal organisms by light microscopy, and 129 CSF fungal cultures were positive for Cryptococcus species. All cotested specimens with positive cytology results were positive on culture (15 specimens, 100% specificity). Twenty-four samples with positive culture results were negative by CSF cytology (sensitivity 39%).

Conclusions

When culture is used as a gold standard, CSF cytology is 100% specific and 39% sensitive, with a positive predictive value of 100% and a negative predictive value of 99.8%.

Low Cryptococcus Antigen Titers as Determined by Lateral Flow Assay Should Be Interpreted Cautiously in Patients without Prior Diagnosis of Cryptococcal Infection

Detection of Cryptococcus antigen (CrAg) is invaluable for establishing cryptococcal disease. Multiple different methods for CrAg detection are available, including a lateral flow assay (LFA). Despite excellent performance of the CrAg LFA, we have observed multiple cases of low-titer (≤1:5) positive CrAg LFA results in patients for whom cryptococcosis was ultimately excluded. To investigate the accuracy of low-titer positive CrAg LFA results, we performed chart reviews for all patients with positive CrAg LFA results between June 2014 and December 2016. During this period, serum and/or cerebrospinal fluid (CSF) samples from 3,969 patients were tested with the CrAg LFA, and 55 patients (1.5%) tested positive. Thirty-eight of those patients lacked a history of cryptococcal disease and were the focus of this study. Fungal culture or histopathology confirmed Cryptococcus infection for 20 patients (52.6%), and CrAg LFA titers in serum and CSF samples ranged from 1:5 to ≥1:2,560. For the 18 patients (47.4%) without culture or histopathological confirmation, the CrAg LFA results were considered true-positive results for 5 patients (titer range, 1:10 to ≥1:2,560), due to clinical improvement with targeted therapy and decreasing CrAg LFA titers. The remaining 13 patients had CrAg LFA titers of 1:2 (n = 11) or 1:5 (n = 2) and were ultimately diagnosed with an alternative condition (n = 11) or began therapy for possible cryptococcosis without improvement (n = 2), leading to an overall CrAg LFA false-positive rate of 34%. We recommend careful clinical correlation prior to establishing a diagnosis of cryptococcal infection for patients with first-time positive CrAg LFA titers of 1:2.