The objective of the present study was to investigate whether increased beta-human chorionic gonadotrophin (beta HCG) plasma concentrations in an unselected population of nulliparas could predict the occurrence of complicated pregnancy-induced hypertension (PIH). The design was that of a prospective population study. It was conducted at the obstetric departments of Amiens University Hospital and Creil General Hospital on 434 consecutive nulliparas with singleton pregnancies after natural fertilization who accepted the systematic offer of trisomy 21 screening but for whom this disorder was finally estimated. Measurement of plasma concentration of beta HCG (ELISA method) was carried out between 14 and 20 weeks (mean: 17 weeks) of amenorrhea, and measurement of blood pressure and proteinuria (> 300 mg/24 h or Albustix +2) during the first, second and third term and 2-3 months after the delivery, as well as measurement of birth weight for determination of small for gestational age (SGA) babies, 37 women developed PIH, 10 without other complication, 16 with proteinuria (5 of which with SGA babies) and 11 with SGA babies. Furthermore 2 patients presented abruptio placentae without PIH. 395 women did not develop PIH including 389 normotensive women and 6 chronic hypertensive patients without superimposed toxemia. Only 1 was diabetic. None had chronic renal disease. Mean (+/- SD) levels of beta HCG were higher in PIH than in controls: 46,805 +/- 19,068 versus 23,479 +/- 13,463 IU. A pathologic threshold was chosen as the mean for the whole population + 1 SD: 25,613 + 15,479 = 41,082 IU. Elevated levels (above this value) were significantly associated with isolated PIH or PIH complicated with proteinuria and/or with SGA babies. The positive predictive value of this criterion was respectively 11, 15 and 12% for each of these complications. The relative risk (and 95% confidence limit) of women with elevated beta HCG for each of these complications was 20 (6-79), 11 (4-43) and 22 (7-93). Elevated plasma beta HCG found around 17 weeks of amenorrhea predicts PIH complicated with either proteinuria or SGA babies with a positive predictive value comparable to that of the best and earliest test proposed up to now to select nulliparas at high risk of preeclampsia, namely the abnormalities of the Doppler waveforms of the uterine arteries. Since this test is simpler to perform, it represents the most convenient method to screen a population of nulliparas for evaluation of the benefits of low-dose aspirin.

We carried out a retrospective study on 96 patients whose uterine waveforms were abnormal, in order to determine if a laterally located placenta further increased this risk. Our population was separated into 2 groups (75 women with a laterally located placenta, 21 with a centrally located placenta). We have found a significant difference between ipsi placental uterine blood flow velocity and contralateral placental blood flow velocity when the placenta is located laterally, where as the 2 arteries are similar when the placenta has a central location. The following factors were used to compare the two groups: the patient's previous obstetric record and the incidence of hypertensive disorders and fetal distress. In the group with laterally located placentas, the incidence of complications in previous pregnancies was higher, hypertension was more frequently serious or presented complications, fetal distress and cesarean sections were more frequent as was IUGR. All cases of intrauterine death were found in the group with laterally located placentas. In determining the predictive value of the "laterally located placenta" test, we found in the centrally located placenta group that hypertensive disorders were often less serious and less frequently complicated by albuminuria, that the outcome of pregnancy disorders was less serious for the fetus, and that the neonate was less hypotrophic. No incidence of intrauterine death was recorded for the centrally located group. Placental laterality significantly increases both the frequency and the seriousness of fetomaternal hypertensive disorders.

A defect of placenta maturation has been described in hypertension of pregnancy. Plasma beta chorionic gonadotropins (betaHCG) of placental origin rise at the onset of pregnancy and reach a peak between 9 and 10 weeks of amenorrhoea. As we were making systematic assays between 14 and 20 weeks in a trisomy detection programme, we looked for differences in plasma betaHCG levels between women with pregnancy-induced arterial hypertension and pregnant women with normal blood pressure. We also studied the predictive value of such assays. Pregnancy-induced hypertension was found in 6 women in a population of 89 nulliparas and in 1 2 women in a population of 163 multiparas. BetaHCG levels were significantly higher in women who later developed hypertension among both nulliparas (52 833 +/- 19 538 IU vs 24 499 +/- 16 485 IU) and multiparas (50 558 +/- 23 597 IU vs 20 911 +/- 11 677 IU). In nulliparas, taking 43 000 IU as threshold of pathology we found that the predictive value of betaHCG was higher than that of other tests which had gone through controlled studies (sensitivity 67 percent, specificity 91.6 percent, positive predictive value 36 percent, negative predictive value 97.4 percent, relative risk 5.4). In multiparas, taking 38 000 as threshold and combining this marker with obstetrical history it was possible to predict the occurrence of hypertension more precisely than with other markers which had gone through controlled studies (sensitivity 66.7 percent, specificity 98 percent, positive predictive value 61.4 percent, negative predictive value 97.3 percent, relative risk 8.4).