Objectives: To examine whether in women who are delivered of an extremely small for gestational age infant, raised levels of second trimester maternal serum alpha-fetoprotein (MSAFP) or human chorionic gonadotrophin (MShCG) levels are related to the presence of placental pathology detected at birth.
Design: Retrospective cross-sectional study.
Setting: Department of Obstetrics and Gynaecology, Antenatal Diagnosis Unit, Groningen University Hospital, The Netherlands.
Participants: Eighty-four women who were delivered of an extremely SGA infant (< 2.3rd centile) in whom the MSAFP and the MShCG levels were known and placental pathology reports were available (study group), and 8692 women in whom the MSAFP and MShCG levels were known and the pregnancy outcome was normal (control group). Pregnancies with congenital anomalies were excluded. Analyte levels were expressed in multiples of the median (MoM) for gestational age. Statistical analysis between groups was performed by ANOVA, after logarithmic transformation of the MoMs, to normalise their distribution.
Main outcome measures: 1. The means of the MSAFP and MShCG concentrations in the study group with and without placental lesions were compared with those in the control population. 2. The means of the MSAFP and MShCG levels in the study group with placental lesions were compared with those in the study group without placental lesions.
Results: 1. Comparison of study groups with controls: in the study group without placental lesions, the mean log MSAFP MoM (0.062), as well as the mean log MShCG MoM (-0.033), was not significantly different (P = 0.11 and P = 0.68, respectively) from the mean analyte levels in the control population (0.002 and 0.006, respectively). The mean logs of these analytes in the study group with placental lesions (0.162 and 0.129, respectively) were significantly higher compared with the MSAFP and MShCG levels in the control population (P < 0.001 for both analytes). 2. Comparison of study groups with each other: the mean log of the MSAFP level of 0.162 in the group with placental lesions was significantly different from the mean of 0.062 of the study group without placental lesions (P < 0.025). The higher mean log MShCG MoM of 0.129 in the group with placental lesions was significantly different from the mean log MShCG MoM of -0.033 in the study group without placental lesions (P < 0.025).
Conclusions: Raised levels of second trimester MSAFP and MShCG in women who are subsequently delivered of an extremely small for gestational age infant are related to the presence of pathological changes in the placenta, detectable at birth. It is speculated that these placental pathological changes, which frequently accompany small for gestational age pregnancies, have their origin in the second trimester, when the normal physiological changes of the placenta occur.