Sensitivity and specificity of ultrasound for the detection of neural tube and ventral wall defects in a high-risk population.
OBJECTIVE: To test the hypothesis that with experienced operators, sonography alone is sensitive and specific for diagnosing open neural tube and ventral wall defects in a population at risk for these disorders. METHODS: Consecutive women at risk for a fetus with an open neural tube defect because of elevated maternal serum alpha-fetoprotein (MSAFP) or family history were evaluated prospectively using prenatal sonography. Amniocentesis was not done routinely but was done for specific indications, such as limited visualization by sonography, markedly elevated MSAFP, or other risk factors for fetal chromosomal disorders. Complete postnatal outcome information was obtained, and the diagnostic accuracy of sonography was compared with that of amniocentesis plus sonography. RESULTS: Sonography alone was 97% (66 of 68) (95% confidence interval [CI] 0.898, 0.996) sensitive and 100% (n = 2189) (CI 0.998, 1.0) specific in diagnosing open neural tube defect, and was 100% sensitive (n = 17) (CI 0.805, 1.0) and specific (n = 2240) (CI 0.998, 1.0) in diagnosing ventral wall defect. In two cases of neural tube defect, there were other suspicious findings on sonography, and amniocentesis was performed for confirmation. CONCLUSION: When used by experienced operators, prenatal sonography is sensitive and specific for the diagnoses of neural tube and ventral wall defects in a targeted at-risk population.