Abstract
INTRODUCTION: People with advanced HIV disease face high risks of severe illness and death. CD4 testing enables timely diagnosis and appropriate care, yet access remains limited in many settings. This review investigated the diagnostic accuracy of the WHO clinical staging for identifying advanced HIV disease. METHODS: We conducted a systematic review and meta-analysis of studies published between 1 January 1998 and 1 May 2024 that assessed both WHO clinical staging and CD4 counts in people living with HIV aged 5 years and older (PROSPERO: CRD42024558372). We pooled sensitivity and specificity estimates of WHO Stage 3/4 for detecting advanced HIV disease (CD4 <200 cells/μL) using bivariate random-effects meta-analysis. Risk of bias was assessed using QUADAS-2, and certainty of evidence was appraised using Grading of Recommendations, Assessment, Development, and Evaluations (GRADE). RESULTS: Of 15,194 studies screened, 335 relevant studies were identified, from which 25 were included in evidence synthesis and 21 in the meta-analysis. Most studies were from the WHO African (19/25) and South-East Asian (5/25) regions. Risk of bias was moderate to high in 88% of studies, primarily due to issues with clinical staging assessment. Pooled sensitivity and specificity of WHO Stage 3/4 were 60.7% (95% CI: 48.0%-72.1%) and 72.4% (95% CI: 61.4%-81.3%), respectively. Specificity was significantly higher outside the African region (p < 0.001). In a population of 100,000 people living with HIV with 30% advanced HIV disease prevalence, WHO staging would miss 11,700 true advanced HIV disease cases and misclassify 19,600. CONCLUSIONS: WHO clinical staging alone shows low accuracy for detecting advanced HIV disease, risking both missed diagnoses and overtreatment. CD4 testing remains essential for accurately identifying and managing advanced HIV disease.