Je. Obrien et al., RACE-ETHNICITY-SPECIFIC VARIATION IN MULTIPLE-MARKER BIOCHEMICAL SCREENING - ALPHA-FETOPROTEIN, HCG, AND ESTRIOL, Obstetrics and gynecology, 89(3), 1997, pp. 355-358
Objective: To identify any race-ethnicity-specific differences in seru
m alpha-fetoprotein (AFP), hCG, and unconjugated estriol (E3) levels i
n women between 14 and 21 weeks' gestation. Methods: Data from the 3-y
ear period 1992-1994 were analyzed from 208,257 women who had AFP scre
ening, of whom 155,142 also had hCG and 62,121 also had E3 screened, b
etween 14 and 21 weeks' gestation. Subjects were categorized into four
groups: white, black, Asian, and Hispanic. Results: There was a consi
stent pattern of analyte differences across gestational ages. Levels f
or AFP were generally higher in Asian and black women than in Hispanic
and white women (median AFP at 16 weeks-31.2, 30.9, 27.4, 27.3, respe
ctively), and levels of hCG and E3 were highest in Asians (hCG at 16 w
eeks-34.7, 30.3, 28.2, 26.8, respectively). Weight correction for AFP,
hCG, and E3 levels did not compensate for the ethnic differences. Con
clusions: Because hCG and E3 demonstrate the same general pattern of d
ifferences as AFP among ethnic groups, averaging values for all ethnic
groups tends inappropriately to lower calculated Down syndrome risks
for black and Asian women. Additionally, the slopes of the curves are
not parallel, such that separate data bases are preferable to multipli
cative correction factors. Separate data bases should be used in labor
atories with volume sufficient to permit the establishment of race-eth
nicity-specific regressions. Use of separate data bases should result
in more accurate screening. Copyright (C) 1997 by The American College
of Obstetricians and Gynecologists.