Galactomannan detection for the diagnosis of invasive aspergillosis in immunocompromized patients. A Cochrane Review of Diagnostic Test Accuracy

Background: Invasive aspergillosis (IA) is the most common life-threatening opportunistic invasive mycosis in immunocompromized patients. The main issues in the diagnosis of IA are the following: a test needs to be not too invasive or not too big a burden for the already weakened patient; and a tool is needed to guide therapy. The serum ELISA seems to have potential for both requirements, we therefore wanted to know whether the Platelia ELISA is sufficiently accurate to diagnose IA and to guide antifungal therapy. Objectives: To obtain summary estimates of the diagnostic accuracy of galactomannan detection in serum for the diagnosis of invasive aspergillosis. Search strategy: MEDLINE, EMBASE and Web of Science were searched with both Medical Headings and text words for both Aspergillosis and the sandwich ELISA. Furthermore, we tracked references. Selection criteria: We included studies assessing the diagnostic accuracy of galactomannan detection for the early diagnosis of IA, either prospective or retrospective and either case-control or cohort designs. Patients with neutropenia or patients whose neutrophils are functionally compromized were included. The index test was the Platelia© Aspergillus sandwich ELISA, the reference standard was a composite reference standard: EORTC/MSG criteria. Data collection and analysis: Data collection was done by six reviewers, divided into three pairs of a methodologist and a microbiologist. Data collection and quality assessment was done independently, through a piloted form. Disagreements were solved by discussion. Results: Seven studies reported results for cut-off value 0.5 ODI. Overall sensitivity was 79% (95% CI 64% to 93%) and overall specificity was 82% (71% to 92%). Twelve studies reported the results for cut-off value of 1.0 ODI, overall sensitivity was 71% (61% to 81%) and overall specificity was 90% (87% to 94%). Seventeen studies reported the results for cut-off value 1.5 ODI, sensitivity was 62% (45% to 79%) and specificity was 95% (92% to 98%). Authors’ conclusions: If we use the test at cut-off value 0.5 in a population of 100 patients with a disease prevalence of 8%, that will mean that 2 patients who have IA, will be missed (sensitivity 79%, 21% false negative rate). And 17 patients will be treated unnecessarily (specificity of 82%, 18% false negative rate). If we use the test at cut-off value 1.5 in the same population, that will mean that 3 IA patients will be missed (sensitivity 62%, 38% false negative rate) and 5 patients will be treated unnecessarily (specificity of 95%, 5% false negative rate). Proefschrift Mariska.indd 118 12-05-2008 18:18:35 119 Galactomannan detection for the diagnosis of invasive aspergillosis