CONTEXT: Objective measurement of prenatal alcohol exposure (PAE) is essential for identifying children at risk for adverse outcomes, including fetal alcohol spectrum disorders. Biomarkers have been advocated for use in universal screening programs, but their validity has not been comprehensively evaluated. OBJECTIVE: To systematically review the validity of objective measures of PAE. DATA SOURCES: Thirteen electronic databases and supplementary sources were searched for studies published between January 1990 and October 2015. STUDY SELECTION: Eligible studies were those that evaluated the diagnostic accuracy of objective measures of PAE. DATA EXTRACTION: Three reviewers independently verified study inclusion, quality assessments, and extracted data. RESULTS: Twelve studies met inclusion criteria. Test performance varied widely across studies of maternal blood (4 studies; sensitivity 0%-100%, specificity 79%-100%), maternal hair (2 studies; sensitivity 19%-87%, specificity 56%-86%) maternal urine (2 studies; sensitivity 5%-15%, specificity 97%-100%), and biomarker test batteries (3 studies; sensitivity 22%-50%, specificity 56%-97%). Tests of the total concentration of 4 fatty acid ethyl esters (in meconium: 2 studies; in placenta: 1 study) demonstrated high sensitivity (82%-100%); however, specificity was variable (13%-98%). LIMITATIONS: Risk of bias was high due to self-report reference standards and selective outcome reporting. CONCLUSIONS: Current evidence is insufficient to support the use of objective measures of prenatal alcohol exposure in practice. Biomarkers in meconium and placenta tissue may be the most promising candidates for further large-scale population-based research.