Article type
Year
Abstract
Background: Cryptococcal meningitis (CM) is a common HIV-associated fungal infection and has a case fatality rate of 35–65% in resource-limited settings (RLS). Early diagnosis of CM is crucial to patient survival. Lumbar puncture (LP) with culture and CrAg LA assay of cerebrospinal fluid (CSF) is the standard diagnostic approach but is often not possible in RLS.
Objectives: To inform new World Health Organization (WHO) guidelines on diagnosis, prevention and management of cryptococcal infection in adults and children with HIV, we conducted a systematic review to evaluate serum CrAg testing for CM diagnosis.
Methods: Using methods recommended by the Cochrane Diagnostic Test Accuracy Working Group, we searched a range of databases through July 2011. Studies compared CrAg LA testing in serum versus CSF culture testing or India ink staining in patients with HIV and suspected CM. Two authors screened them independently.
Results: We initially identified 2337 references, assessed full text of 53 and identified 12 studies, which included 2309 patients in Brazil, Democratic Republic of Congo, India, Italy, Rwanda, Thailand and the United States. We also received unpublished data from two sources. Pooled sensitivity was 97.0% (95% CI 95.1–98.3%). Pooled specificity (seven studies) was 98.6% (95% CI 98.0–99.1%). Our appraisal of study quality with the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tool raised no major concerns. Our GRADE analysis found the evidence to be of moderate quality.
Conclusions: While fungal culture remains the ‘gold standard’ for diagnosing CM, serum CrAg LA is an accurate and precise means of testing for it. If rapid results are assured, WHO recommends serum CrAg testing when LP is not immediately available or is contraindicated. Patients testing positive must still be referred for LP as soon as possible so that CSF opening pressure can be measured and the diagnosis confirmed.
Objectives: To inform new World Health Organization (WHO) guidelines on diagnosis, prevention and management of cryptococcal infection in adults and children with HIV, we conducted a systematic review to evaluate serum CrAg testing for CM diagnosis.
Methods: Using methods recommended by the Cochrane Diagnostic Test Accuracy Working Group, we searched a range of databases through July 2011. Studies compared CrAg LA testing in serum versus CSF culture testing or India ink staining in patients with HIV and suspected CM. Two authors screened them independently.
Results: We initially identified 2337 references, assessed full text of 53 and identified 12 studies, which included 2309 patients in Brazil, Democratic Republic of Congo, India, Italy, Rwanda, Thailand and the United States. We also received unpublished data from two sources. Pooled sensitivity was 97.0% (95% CI 95.1–98.3%). Pooled specificity (seven studies) was 98.6% (95% CI 98.0–99.1%). Our appraisal of study quality with the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tool raised no major concerns. Our GRADE analysis found the evidence to be of moderate quality.
Conclusions: While fungal culture remains the ‘gold standard’ for diagnosing CM, serum CrAg LA is an accurate and precise means of testing for it. If rapid results are assured, WHO recommends serum CrAg testing when LP is not immediately available or is contraindicated. Patients testing positive must still be referred for LP as soon as possible so that CSF opening pressure can be measured and the diagnosis confirmed.