Ethnicity and the need for correction of biochemical and ultrasound markers of chromosomal anomalies in the first trimester: a study of Oriental, Asian and Afro-Caribbean populations

OBJECTIVES: To assess whether there is a need to correct first-trimester biochemical markers (free beta-hCG and pregnancy-associated plasma protein-A (PAPP-A)) or first-trimester fetal nuchal translucency thickness (NT) in different ethnic groups, when screening for Downs syndrome at 11-14 weeks of gestation. METHODS: Free beta-hCG, PAPP-A and fetal NT were measured at 11-14 weeks of gestation in a group of women presenting for first-trimester screening in two OSCAR centres. The group comprised 61 219 sets of data from Caucasian women (the reference group); 4835 sets of data from South Asian women; 3450 sets of data from Oriental women and 2727 sets of data from Afro-Caribbean women. The Oriental data set was supplemented with a further 480 cases collected in Hong Kong and the Afro-Caribbean data set was supplemented with 216 cases collected from Kings College. The difference in marker values between the reference group and the other ethnic groups was compared before and after weight correction for the biochemical markers using standard statistical techniques. A correction factor for ethnic origin was applied for all three markers and the screen-positive rate before and after correction was assessed for the various groups. RESULTS: After maternal weight correction, in Afro-Caribbean women, the median PAPP-A was increased by 55% and the free beta-hCG increased by 11%. In south Asian women, the PAPP-A was increased by 8% and the free beta-hCG decreased by 7.5%. In Oriental women, the PAPP-A was increased by 9% and the free beta-hCG by 6%. For delta NT in Afro-Caribbean women, the values were 0.064 mm lower on average than in Caucasian women and for south Asian women 0.045 mm lower. The difference of -0.012 for Oriental women was not significant. Before correcting for ethnic origin, these changes resulted in the screen-positive rates being lower in the Afro-Caribbean group (3.7% vs 5.6%), the south Asian group (4.3% vs 5.6%) and Oriental group (4.9% vs 5.6%). After correction, the screen-positive rates were largely similar in the four groups. CONCLUSION: Differences in median PAPP-A, free beta-hCG and, to a lesser extent, in NT exist in Afro-Caribbean, South Asian and Oriental women. In populations where the medians and delta NT reference ranges are established in predominantly Caucasian populations, some correction for ethnicity is appropriate and can redress differences in screen-positive rates between these different groups.