Utility of Fetal Karyotype in the Evaluation of Phenotypically Abnormal Stillbirths

Abstract

The objectives of this study are to test the hypothesis that stillbirths without aneuploidy-associated phenotypes have a low incidence of karyotypic abnormalities, similar among those with and without other anatomic defects. We employed a uniform postmortem protocol to examine fetuses and placentas in 962 consecutive stillbirths measuring ≥20 weeks in clinically determined gestational age submitted to the Women and Infants Hospital Division of Perinatal Pathology from 1990 through 2005. Classification of anatomic (macroscopic) abnormalities was based on a priori criteria. Anatomic fetal abnormalities were noted in 387 cases. Conventional karyotype analysis was successfully performed on 346 fetal tissue samples, 114 in anatomically normal and 232 in anatomically abnormal fetuses. The distribution of karyotypic abnormalities among cases with and without anatomic abnormalities was compared. Of the 962 stillbirths, 40% (387) had malformations. Tissue culture for karyotype analysis was attempted in 412 cases from both groups and failed in 66 cases (16%). At the 450 to 500-band resolution level, 60 of the remaining 346 karyotypes were abnormal. Of the 232 malformation cases with successful karyotyping, 59 had phenotypic attributes indicative of aneuploidy, all of which had later karyotype confirmation. Of the remaining 173 anomalous fetuses with karyotype analysis, only 1 demonstrated a karyotypic abnormality. All 114 karyotypes performed in stillbirths without anatomic abnormalities were normal. Among ≥20-week stillbirths, aneuploid karyotypes are uncommon except in fetuses with suspect phenotypes. The 95% probability estimates of karyotype abnormality in the phenotypically abnormal and normal stillbirths, 5.5% and 5.6%, respectively, do not differ. These data do not have sufficient power to detect a small difference in rates of karyotypic abnormalities between the 2 groups of ≥20-week stillbirths. However, this series indicates that this technology is uninformative among stillborn fetuses that lack aneuploidy phenotypes.