A meta-analytic evaluation of the polymerase chain reaction for the diagnosis of HIV infection in infants.

OBJECTIVE To evaluate the sensitivity and specificity of the polymerase chain reaction (PCR) for the diagnosis of infection with human immunodeficiency virus (HIV) in infants. DATA SOURCES We used studies published between 1988 and 1994 identified in a literature search of 17 databases, including MEDLINE. STUDY SELECTION Studies were included if DNA amplification by PCR was performed on peripheral blood mononuclear cells from infants or children. DATA EXTRACTION Two investigators independently extracted data. The study design was assessed independently by 2 investigators who were blinded to study results. DATA SYNTHESIS Thirty-two studies met the inclusion criteria and were analyzed. The median reported sensitivity was 91.6% (range, 31%-100%), and the median specificity was 100% (range, 50%-100%). A summary receiver operating characteristic curve based on all 32 studies indicated that PCR has a maximum joint sensitivity and specificity between 93.2% and 94.9%. Subgroup analysis indicated that the joint sensitivity and specificity was significantly (P = .04) higher in older infants (98.2%) than in neonates (aged < or = 30 days; 93.3%). For infants at low risk of perinatal transmission (probability of transmission, 8.3%), the positive predictive value for PCR is 55.8% in neonates and 83.2% in older infants. A negative PCR result reduces the probability of HIV infection to less than 3%. No studies met all criteria for study design. CONCLUSIONS Although PCR is one of the best available tests for diagnosis of HIV infection in neonates and infants, it is not definitive. Therefore, PCR should be interpreted with the aid of careful clinical follow-up examinations. The sensitivity and specificity of PCR in neonates is lower than in older infants, which results in a low positive predictive value; however, negative tests are informative. Delaying the use of PCR until after the neonatal period or repeating PCR on independent samples obtained 30 to 60 days later will reduce test errors.