Prevalence of cryptococcal antigenemia and performance of the lateral flow assay for detection of cryptococcal antigen.
Etolu, John Wilson
MetadataShow full item record
Introduction: Cryptococcal infection is a common opportunistic infection and is a leading cause of death among HIV-infected individuals in sub-Saharan Africa. Cryptococcal antigenemia is predictor of cryptococcal meningitis (CM) and death. Screening for cryptococcal antigen among severely immunosuppressed HIV- infected patients is potentially lifesaving; this has not been fully embraced in resource limited settings like Uganda, probably because of the challenges of the current diagnostic test for serum cryptococcal antigen. Objectives: In this study, we determined the prevalence of cryptococcal antigenemia among severely immunosuppressed hospitalized HIV-infected patients, as well as the correlation between the Lateral Flow Assay (LFA) and the standard Latex Agglutination (LA) in detecting cryptococcal antigen at Mulago national referral and teaching hospital. We also determined the factors associated with cryptococcal antigenemia and outcomes of patients with cryptococcal antigenemia as secondary objectives. This study will describe the new burden of cryptococcal antigenemia and strengthen evidence for the use of the LFA in resource limited settings like Uganda. Methods: We conducted a cross-sectional study where we consecutively enrolled 236 severely immunosuppressed HIV-infected patients admitted between June and August 20012 in Mulago hospital general medical wards. Eligible patients who were: adults ≥ 18years, with CD4 T cell count of ≤ 100 cells/µl, with no prior CM treatment and have given informed consent; were enrolled and screened for serum cryptococcal antigen (CrAg) using the LFA and LA tests. The correlation between the LFA and LA was determined. Patients with positive serum CrAg or signs of meningitis underwent lumbar puncture (LP) to screen for CM and were followed to determine their outcomes. Factors independently associated with cryptococcal antigenemia were also determined. Data analysis: Using STATA version 10.0, the prevalence of cryptococcal antigenemia was determined, and the factors independently associated with cryptococcal antigenemia were determined using multiple logistic regression model while correlation between the LFA and the LA was determined using the Kappa statistic. Results: Of the 236 patients enrolled in the study; the median CD4 T cell count was 27cells/µl (IQR: 12 to 54), the mean age was 34.8 years (SD 9.53), median body mass index (BMI) of 15.2 (IQR: 14.5 to 17.0). The prevalence of cryptococcal antigenemia was 52 (22%) of 236 patients, 35 patients of the 38 in whom LP was done had CM, 3 patients were negative for CM. The percent agreement between the LFA and the LA was 99% (Kappa statistic K=0.99, P<0.0001). Neck pains (adjusted odds ratio (AOR) 7.25), cough (AOR 0.3) were independently associated with cryptococcal antigenemia. Conclusion: This study demonstrated a high prevalence of cryptococcal antigenemia among severely immunosuppressed HIV-infected hospitalized patients in Mulago Hospital, the new POC test the LFA has excellent agreement with the standard of care test the LA test in detecting cryptococcal antigen. Neck pain and cough were independently associated with cryptococcal antigenemia in our study. Recommendation: The LFA can be used as a point of care test in resource limited settings. Screening of patients with severe immunosuppression for cryptococcal antigenemia should be done for hospitalized patients as part of their care and those who turn out positive should have LP done.