Use of ethnic-specific medians for Hispanic patients reduces ethnic disparities in multiple marker screening.

OBJECTIVE: To estimate whether midtrimester maternal serum analyte concentrations differ between Caucasian and Hispanic women and whether using ethnic-specific medians affects quad screen performance.
METHOD: Caucasian and Hispanic patients with singletons who underwent maternal serum screening in our laboratory were identified. Alfa-fetoprotein (AFP), estriol, human chorionic gonadotrophin (hCG), and inhibin-A medians were derived separately for Caucasians, Hispanics, and for the composite group. Using composite medians, intergroup mean multiples of the medians (MoMs) for each analyte were compared. Using ethnic-specific medians, new MoMs were calculated and utilized in a risk estimation algorithm.
RESULTS: A total of 5478 Caucasian and 2246 Hispanic pregnancies were evaluated. Intergroup MoMs were significantly different for all analytes. AFP, hCG, and inhibin-A were lower in Hispanics, while estriol was higher (P < 0.0001). Using composite medians, the screen-positive rate (SPR) for trisomy 21 was 5.39% in Caucasians and 3.29% in Hispanics. Ethnic-specific medians reduced this disparity: 4.76% in Caucasians and 4.05% in Hispanics. The SPR for neural tube defects with composite medians was 1.44% for Caucasians and 0.89% for Hispanics; with ethnic-specific medians, the SPR was 1.42% for Caucasians and 1.07% for Hispanics.
CONCLUSION: Serum analyte concentrations differ between Caucasian and Hispanic gravidas. Use of ethnic-specific medians reduces the disparity in SPR for trisomy 21 and neural tube defects.