maternal and fetal prolactin in pregnancy-induced hypertension

Arch Gynecol Obstet (1990) 247:73-81

Archives of

Gynecology and Obstetrics

© Springer-Verlag 1990

Maternal and fetal prolactin in pregnancy-induced hypertension

M. G. Marlettini 2, A. Cassani 1, A. M. Morselli-Labate 3, S. Crippa 1, A. Contarini 1, R. Miniero 4, L. Plat64, and C. Orlandi 1

1 II Clinica Ostetrica e Ginecologica, 2 Istituto di Patologia Medica e Metodologia Clinica I I e Medicina Del Lavoro, 3 Istituto di Clinica Mediea I, Universita' di Bologna, 4 Laboratorio Centralizzato Policlinieo S. Orsola, Bologna, Italy

Summary. In plasma from 35 women with pregnancy-induced hypertension (PIH) and 35 normal pregnant women both at 39 weeks of gestation, plasma prolactin levels were measured at 8.30 a.m. (PRL1) and 9.30 a.m. (PRL2) under basal conditions. At delivery umbilical cord blood samples were taken for measurement of fetal prolactin (PRLF). PRL1 and PRL2 were higher in women with PIH, but no significant relations were found betwen PRL1/ PRL2 and blood pressure. PRLF did not differ when infants of mothers with PIH and infants of normal pregnant women were compared, but PRLF had a significant direct independent relation with PRL2. The latter relation may be due to the increase in placental oestrogens during pregnancy, which stimulate both the maternal and fetal hypophyses and their prolactin secretion. PRLF did not show any relation with neonatal morbidity, but PRL1 showed a significant direct relation with the Apgar score at 5 rain.

Key words: Prolactin - Pregnancy - Hypertension

Introduction

Prolactin has been alleged to play a role in PIH. However plasma concentrations of prolactin in women with PIH have been found to be increased or unchanged in comparison with normal pregnant women (Jenkins et al. 1978, Kulseng-Haussen et al. 1979, Campbell et al. 1982). These conflicting results may be due to prolactin levels being affected by various factors such as mental and physical stress.

The role of prolactin as far as blood pressure is concerned has been the subject of further investigation, i.e. whether prolactin contributes directly to blood pressure or whether high prolactin levels and high blood pressure are both

Offprint requests to: Dr. M. G. Marlettini, Istituto di Patologia Medica e Metodologia Clinica I I e Medicina de1 Lavoro, Universita' di Bologna, Via Massarenti, 9, 1-40138 Bologna, Italy

74 M . G . Marlettini et al.

signs of an altered central dopaminergic activity as in primary hypertension (Stumpe et al. 1977).

As for fetal prolactin many studies favour the view of an autonomous production of prolactin from the fetal pituitary (Siler-Khodr et al. 1974). However the mechanism regulating its secretion and its physiological significance remains almost unknown, probably because of the limited means for investigation of the fetus in utero. Fetal prolactin levels increase progressively during pregnancy especially in the last few months reaching levels which are higher than maternal levels at the time of birth. The physiological significance of the increased levels of prolactin in the fetus and neonate are still uncertain. Moreover higher umbilical cord blood prolactin levels have been found in infants of preeclamptic mothers than in infants of normal mothers (Grosso et al. 1980).

The aim of this study was to evaluate both maternal plasma prolactin and umbilical cord plasma prolactin levels in women with PIH and normotensive gravid women near term. In addition we examined the relations between maternal and fetal prolactin levels and between both maternal and fetal prolactin levels and the condition of the infant at birth.

Methods

The study was carried out in 35 women with PIH and in 35 normal gravid women at the 39th week of gestation. Informed consent was obtained from all women. PIH was defined as the presence after the 20th week of gestation of a sitting blood pressure _>140/90 mmHg. Blood pressure was measured during two periods of 20 min which were 8 h apart (Blood Pressure Monitor Sentry).

No women had received drugs for at least 3 weeks before the study and were on an unrestricted sodium intake.

Body Mass Index (BMI) (weight/height 2) before pregnancy and at the time of prolactin measurement were evaluated. On the day before the study the women collected 24-h urine samples for measurement of proteins and sodium, potassium, calcium and aldosterone content. Creatinine concentration was also measured in order to allow calculation of 24-h endogenous creatinine clearance. On the day of the study the means of three blood pressure and heart rate recordings obtained at 2-min intervals were taken as values for blood pressure and heart rate in every subject. After an overnight fast an intravenous cannula was placed in an antecubital vein at 7.30 a.m. and the subjects rested in the left lateral recumbent position. Basal blood samples for prolactin evaluation were collected at 8.30 a.m. (PRL1) and 9.30 a.m. (PRL2). Blood samples were also taken for measurement of plasma renin activity (PRA), serum uric acid, serum proteins, serum sodium, serum potassium, serum calcium, plasma total oestriol (free and conjugated) and human placental lactogen.

At delivery cord blood samples were taken for measurement of urea, creatinine, glucose, proteins and albumin, sodium, potassium, calcium and magnesium. The Apgar scores and birthweight in relation to the 10th centile in our normal population (Parmeggiani et al. 1976) were recorded.

Sodium and potassium in serum and urine were measured by means of specific ion electrodes (Orion Research). Calcium and magnesium in serum and calcium in urine, serum uric acid, serum and urinary creatinine and serum proteins were measured with the CX4 Beckman Autoanalyzer. Proteins in urine were measured by means of a quantitative colorimetric method for total proteins in urine (Lancer Microprotein Rapid stat-Diagnostic Kit).

Plasma prolactin was measured by means of double antibody RIA (Kit Biodata on Kemptek 3.300) as already described in a previous paper (Marlettini et al. 1987). PRA, urinary aldosterone, total plasma oestriol and human placental lactogen were measured as already reported in the above mentioned paper (Marlettini et al. 1987).

Maternal fetal prolactin in pregnancy hypertension 75

Statistical analysis

The distributions of the continuous variables were expressed as means +_ Standard Deviation (SD). The statistical significance of the differences of these variables between hypertensive and normal pregnant women was evaluated by means of the Wilcoxon-Matched Pairs test (Siegel 1956). When discrete variables (primigravidae/multigravidae, presence of oedema, types of delivery) were considered, the McNemar test was applied.

Differences of PRL1, PRL2 and PRLF according to the presence of oedema and of PRLF according to birthweight, sex of newborn and type of delivery were examined by means of the Mann-Whitney U-test (Siegel 1956).

To identify sets of variables that correlated with maternal and fetal prolactin levels a stepwise multiple regression analysis was used. As criteria for the selection of variables, probability levels of 0.05 and 0.1 were chosen as tolerance limits.

All the variables used in the multiple regression analysis were first normalized by means of logarithmic and/or polynomial transformations. The contemporary non-significant (P > 0.1) results of the Kolmogorov-Smirnov (Siegel 195 0 , Skewness (Dixon 1983) and Kurtosis (Cramer 1946) tests were chosen as the statistical criteria for the acceptance of the normalizing procedures.

Even though Apgar score is "per se" a discrete ordinal variable, its distribution over a wide range of values enables this variable to satisfy all the criteria adopted for considering it normally distributed after transformation. Such a procedure allowed us to use it as an independent variable in the multiple regression procedure. All the statistical evaluations were performed on a personal Epson PC AX Computer by means of the SPSS/PC+ Statistical package (SPSS Inc. 1986).

Results

Table i shows the clinical data, Table 2 the biochemical data and Table 3 shows the week of delivery and gives details of the newborn. Table 4 shows the biochemical cord blood results and Table 5 the maternal and cord blood prolactin values.

The groups were similar as far as age, ratio of primigravidae to multigravidae and the incidence and amount of oedema were concerned (Table 1).

The women with PIH had significantly higher levels of heart rate (z = -2.2, P < 0.05), proteinuria (z = -2 .8 , P < 0.01) and of BMI before pregnancy (z = -2 .3 , P < 0.05) than the normal gravid women (Table 2).

Table 1. Clinical data about women with pregnancy-induced hypertension (H) and of normotensive pregnant women (N)

H N Pa

Age (years) 30.3 + 6.2 29.6 + 7.4 NS Ratio primigravidae/multigravidae 16/19 18/17 NS Oedema (N. patients):

Tibial 17 12 NS Generalized 2 0 NS

BMI before pregnancy (g/cm 2) 2,5 + 0.4 2.3 + 0.3 <0.05 BMI at the time of prolactin evaluation

(g/cm 2) 3.0 + 0.4 2.8 + 0.4 NS SBP (mmHg) 148.1 + 8.7 113.1 + 9.5 <0.001 DBP (mmHg) 99.0 _+ 7.4 70.0 _+ 6.8 <0.001 Heart rate (beats/rain) 90.4 _+ 11.8 82.4 + 12.6 <0.05

a significance of the difference between H and N. Data are expressed as means + SD

76 M.G. Maflettini et al.

Table 2. Biochemical values in women with pregnancy-induced hypertension (H) and normotensive pregnant women (N)

H N P~

Serum uric acid (mg/dl) 5.0 + 1.3 4.4 + 1.2 NS Serum creatinine (mg/dl) 0.85 + 0,19 0.76 + 0.15 NS Serum proteins (g/dl) 5.7 + 0,5 5.7 + 0.4 NS Serum sodium (mEq/1) 137,8 _+ 2.5 137.4 + 2,8 NS Serum potassium (mEq/1) 4.1 _+_ 0.3 4.0 + 0.3 NS Serum calcium (mg/dl) 8.6 + 0.5 8.5 + 0.5 NS Proteinuria (g/l) 0.27 + 0,83 0.06 + 0.09 <0.01 Urinary sodium (mEq/24 h) 117.3 + 53.6 100.3 + 51.2 NS Urinary potassium (mEq/24 h) 43.5 _+ 20.3 35.2 + 15.1 NS Urinary calcium (mEq/24 h) 4.2 _+ 2.7 4.4 + 2.4 NS PRA (ng/ml/h) 4.4 + 2.7 5.6 + 5.2 NS Urinary aldosterone (rag/24 h) 55.2 + 28.9 74.1 + 89.5 NS Serum oestriol (ng/ml) 158.4 _+ 82.2 169.2 + 81.9 NS Humal placental lactogen (~g/ml) 6.5 + 2.22 7.3 + 1.71 NS

significance of the difference between H and N. Data are expressed as means + SD

Table 3. Week of delivery and neonatal data in women with pregnancy-induced hypertension (H) and normotensive pregnant women (N)

H N pa

Week of delivery 39.5 + 2.1 40.3 + 1.0 <0.01 Neonatal weight (kg) 2,995.1 + 678 3,492.3 + 442 <0.01 % Neonatal weight b 108.5 + 18.3 119.7 + 14.5 <0.05 Apgar score:

at I rain 8.02 _+ 1.07 8.08 + 1.04 NS at 5 min 9.28 + 1.25 9.68 + 0.71 NS

significance of the difference between H and N. Data are expressed as means + SD; b related to the 10 th percentile in our normal population at the same gestational age

Table 4. Biochemical values in cord blood

H N pa

Cordal urea (mg/dl) Cordal creatinine (mg/dl) Cordal glucose (mg/dl) Cordal proteins (g/all) Cordal albumin (g/dl) Cordal sodium (mEq/1) Cordal potassium (mEq/1) Cordal calcium (mg/dl) Cordal magnesium (mg/dl)

0.29 _+ 0,07 0.21 + 0.10 0.96 _+ 0.18 0.86 + 0.24 0.84 + 0.2 0.86 + 0.2

5.6 ___ 0.6 5.8 + 0.8 3.27 _ 0,54 3,29 + 0.61

137.4 + 4,1 137.5 + 3.4 4.81 _+ 0.51 4,77 + 0.54 10.1 -2-- 0.81 10.5 __+ 0.83 2.08 __+_ 0.25 2.08 +__ 0.27

<0.01 <0.05 NS NS NS NS NS < 0.05 NS

significance of the difference between H and N. Data are expressed as means +_- SD


Table 5. Maternal and cord blood prolactin values in women with PIH (H) and normotensive pregnant women (N)

H N pa

PRL1 (ng/ml) 264.3 _+ 143.3 205.1 _+ 86.7 0.054 PRL2 (ng/ml) 250.7 +_ 133.9 189.1 + 76.7 <0.05 PRLF (ng/ml) 248.1 + 80.8 270.4 +__ 157.9 NS

a significance of the difference between H and N. Data are expressed as means + SD

The week of delivery was earlier in hypertensive than in normotensive pregnant women (z = -2 .9 , P < 0.01) (Table 3). The incidence of lower segment cesarean section was similar in hypertensive and normotensive pregnant women (18 hypertensive and 12 normotensive pregnant women had lower segment cesarean section) and this was also the case for the sex of infants (16 males in patients with PIH and 19 in normal gravid women). Neonatal weight both absolute (z = -3.1, P < 0.01) and related to gestational age (z = -2 .2 , P < 0.05) was lower in women with PIH than in normal gravid women (Table 3). Infants of hypertensive mothers had significantly higher levels of serum creatinine (z = -1.99, P < 0.05) and urea (z = -4.08, P < 0.001) and significantly lower levels of serum calcium (z = -2.33, P < 0.05) than infants of normotensive mothers (Table 4).

Patients with PIH had higher levels of PRL1 than normal gravid women, the difference being near the level of significance (z = -1.92, P < 0.054), and significantly higher values of PRL2 (z = -2.13, P < 0.05) (Table 5). PRLF did not differ significantly between hypertensive and normotensive mothers (Table 5). Within each group (hypertensive and normotensive gravid women) PRL1 and PRL2 were similar (Table 5). In normal pregnant women PRLF was significantly higher than PRL1 (z = -2.37, P < 0.05) and PRL2 (z = -2.98, P < 0.01), while in hypertensive pregnant women PRLF did not differ significantly from PRL1 and PRL2 (Table 5).

PRL1 and PRL2 in both groups of women were not related to the presence of maternal oedema. PRLF in hypertensive and normotensive mothers was not related to the type of delivery, birthweight and sex of the infants.

In order to select sets of variables correlating with PRL1 and PRL2, multiple regression was carried out with PRL1 and PRL2 as dependent variables and BMI before pregnancy and at the time of prolactin evaluation, systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate, serum creatinine, serum uric acid, serum proteins, sodium, potassium and calcium in serum and in a 24-h urine specimen, PRA, urinary aldosterone, plasma total oestriol, human placental lactogen as independent variables.

PRL1 had a significant direct relation to serum potassium (t = 2.12, P < 0.05) and urinary potassium (t = 2.57, P < 0.05). PRL2 showed a significant direct relation ,Mth serum potassium (t -- 2.12, P < 0.05) and urinary potassium (t = 3.08, P < 0.01) and a significant inverse relation with urinary aldosterone (t = -2.43, P < 0.05).

78 M . G . Marlettini et al.

By means of multiple regression PRL1 and PRL2 were also related to neonatal weight, birthweight centile and the Apgar score at 1 and 5 min. PRL1 had a significant direct relation to the Apgar score at 5 min (t = 2.45, P < 0.05).

A multiple regression was also caried out with PRLF as the dependent variable and week of delivery, neonatal weight, birthweight centile, Apgar score at 1 and 5 rain, cord blood urea, creatinine, glucose, proteins and albumin, sodium, potassium, calcium, magnesium, PRL1 and PRL2 as the independent variables.

PRLF showed no relation with neonatal characteristics, but did show a significant direct relation with PRL2 (t -- 2.18, P < 0.05).

Discussion

Prolactin is mainly under inhibitory tuberoinfundibular dopaminergic control, but a stimulatory serotoninergic pathway seems to play a certain rote in prolactin release.

Many other factors such as mental and physical stress, endorphins, plasma osmolality, sodium homeostasis and oestrogens influence prolactin release.

In order to reduce the effect of mental stress we performed two measure- ments of prolactin in basal conditions I h apart.

In normotensive mothers there was a fall from PRL1 to PRL2 which however did not reach statistical significance, while in hypertensive mothers the values of PRL1 and PRL2 were similar. This is at variance with a previous study (Marlettini et at. 1987) which showed a significant fall from PRL1 to PRL2 in women with PIH studied at the 37th week of gestation. In the latter study a greater amount of mental stress was probably present in hypertensive mothers.

There are conflictir~g views about the effect of prolactin on arteriolar resistance. While an in vitro study showed potentiation of arteriolar smooth muscle response to pressor factors such as norepinephrine and angiotensin II by given doses of prolactin (Manku et al. 1973), other experimental studies have shown a hypotensive effect of prolactin (Bryant et al. 1973).

Previous studies have shown increased basal prolactin in men with essential hypertension (Stumpe et al. 1977) and an association between higher serum prolactin levels and increased peripheral sympathetic tone; both elevated prolactin and norepinephrine levels reverted to normal with dopaminergic agonists (Kollock et al. 1980).

Another study (Sowers et al. 1982) has shown higher mean prolactin values over a 24-h recumbent period as well as higher prolactin and norepinephrine levels after isometric exercise and upright posture (which markedly affect the adrenergic nervous system) in patients with primary hypertension than in normotensive controls. All these studies suggest a decreased central dopaminergic activity in primary hypertension causing an increased peripheral sympathetic output and governing the response of prolactin and catechotamines to posture and stress.

As seen in a previous study (Marlettini et al. 1987), PRL1 and PRL2 as well as heart rate were higher in women with PIH than in normal pregnant women.


A direct contribution of prolactin to blood pressure in man is unlikely, as it is also shown by the analyses of multiple regression which showed no relation between PRL1/PRL2 and blood pressure. Instead, as suggested in the above mentioned study (Marlettini et al. 1987) in which we also found significant increase in prolactin values with upright posture in women with PIH in comparison with normotensive pregnant women, the greater levels of PRL1 and particularly of PRL2 in women with PIH may be expression of reduced central dopaminergic activity which may also be responsible for an increased peripheral sympathetic tone. The higher heart rate in women with PIH, which is an index of the peripheral sympathetic tone, seems to support this hypothesis.

Prolactin seems to reduce renal excretion of water, sodium and potassium in man and other vertebrates (Horrobin et al. 1971). Our study however did not show any relations between prolactin levels and sodium in serum and urine. Instead, significant direct relations were present between PRL1/PRL2 and serum and urinary potassium, while PRL2 also showed a significant inverse relation with urinary aldosterone. The absence of any relation between PRL1/ PRL2 and sodium in serum and urine might be due to the fact that sodium excretion in pregnancy is regulated by a delicate balance between progesterone and aldosterone and probably vasodilator prostaglandins (Weber et al. 1982). These factors may have masked an effect of prolactin on sodium excretion.

However we have no definite explanation for the significant direct relations between PRL1/PRL2 and serum and urinary potassium while, as a tentative explanation, the significant inverse relation between PRL2 and urinary aldosterone, might be due to a feedback mechanism owing to the sodium retaining effect of both hormones.

The fetal hypophysis seems to be the source of fetal prolactin (Siler-Khodr et al. 1977) and both the hypophyseal and the plasma concentrations of fetal prolactin dramatically increase after the 20th week of gestation (Aubert et al. 1975). The activity of prolactin during fetal life has yet to be clarified. Several hypotheses have been put forward. Prolactin may regulate the development of the suprarenal cortex and the metabolism of the steroid hormones, it may act like growth hormone in stimulating fetal growth and maturation and may stimulate lung maturation and surfactant synthesis (Winters et al. 1975, Quirk et al. 1982).

In our study PRLF was significantly higher than PRL1 and PRL2 in normal pregnant women, while no difference was present in women with PIH. These results may be due to the higher levels of PRL1 and PRL2 in women with PIH.

Moreover with multiple regression PRLF showed a significant direct relation with PRL2. This relation between PRLF and PRL2 may be due to the fact that the increase in prolactin during fetal life is linked to the increase in placental oestrogens during pregnancy, which stimulate both the maternal and fetal hypophyses and their prolactin secretion (Shutt et al. 1974). The importance of placental oestrogens for fetal prolactin secretion is also demonstrated by the progressive fall in plasma oestrogens at birth, which is associated with a simultaneous drop in serum prolactin in the first few days of life.

At variance with other studies which showed higher prolactin levels in infants of preeclamptic mothers than in infants of normotensive mothers

80 M, G. Marlettini et al.

(Grosso et al. 1980), PRLF was similar in infants of hypertensive and normotensive mothers, despite the fact that the infants of hypertensive mothers had higher levels of serum urea and creatinine and lower levels of serum calcium than the infants of normotensive mothers, which may suggest a certain degree of fetal hypoxia in the infants of hypertensive mothers. Unlike other studies (Grosso et al. 1980, Poonai et al. 1975), we could not show any relation between PRLF levels and birthweight or Apgar scores at 1 and 5 min. In addition, in agreement with another study (Yaginuma 1981), no relation was found between PRLF and both type of delivery and sex of the infants. Thus no relation was present between PRLF and fetal morbidity, while a relation between the incidence of respiratory distress and low cord blood prolactin levels, as seen in other studies (Grosso et al. 1980), could not be shown because of the absence of respiratory distress in our infants. However, it would seem to us that a relation between fetal prolactin and neonatal morbidity is unlikely.

But a direct relation was present between maternal prolactin and the Apgar score at 5 min. Lower neonatal morbidity thus seems to be associated with increased maternal prolactin levels.

In conclusion in PIH higher values of maternal prolactin seem to reflect a reduced central dopaminergic activity and increased peripheral sympathetic tone (Stumpe et al. 1977). A certain relation was seen between maternal and cord blood prolactin levels which may be due to the increase in placental oestrogens during pregnancy, which stimulate both the maternal and fetal hypophyses and their prolactin secretion. Finally no relation was seen between fetal prolactin and neonatal morbidity, but there may be a relation between maternal prolactin levels and neonatal morbidity.

References

1. Aubert ML, Grumbach MM, Kaplan SL (1975) The ontogenesis of human fetal hormones. J Clin Invest 56:155-166

2. Bryant EE, Douglas BH, Ashburn AD (1973) Circulatory changes following prolactin adminis- tration. Am J Obstet Gynecol 115:53-57

3. Campbell DM, MacGillivray I (1982) Water electrolyte and hormone changes in preeclampsia and normal primigravid pregnancies. In: Samour M, Symonds EM, Zuspan FP, El Tomi N (eds) Pregnancy hypertension. Ain Shams University Press, Cairo, p 95

4. Cramer H (1946) Mathematical methods of statistics. Princeton University Press, Princeton 5. Dixon WJ (1983) BMDP statistical software. Revised printing. University of California Press,

Berkeley, CA 6. Grosso DS, McDonald CP, Thomasson JE, Christian CD (1980) Relationship of newborn

serum prolactin levels to the respiratory distress syndrome and maternal hypertension. Am J Obstet Gynecol 137:569-574

7. Horrobin DF, Burstyn PG, Lloyd IJ, Durkin N, Lipton A, Muiruri KL (1971) Action of prolactin on human renal function. Lancet II:352-354

8. Jenkins DM, Perru LA (1978) Plasma prolactin in pregnancy-induced hypertension. Br J Obst Gynaecol 85:755-757

9. Kollock R, Kobayashi K, Dequattro V (1980) Dopaminergic control of sympathetic tone and blood pressure: evidence in primary hypertension. Hypertension 2:390-394

10. Kulseng-Haussen S, Torjesen PA, Haussen KF (1979) Serum prolactin and thyroid stimulating hormone levels following thyreotropin releasing hormone stimulation in preeclamptic patients. Acta Obstet Gynecol Scand 58:245-248


11. Manku MS, Nasser BA, Horrobin DF (1973) Effect of prolactin on the response of rat aortic and arteriolar smooth-muscle preparations to noradrenaline and angiotensin, Lancet II:991- 994

12. Marlettini MG, Cassani A, Morselli Labate AM, Rusticali AG, Crippa S, Trabatti M, Miniero R, Plate' L, Orlandi C (1987) Role of prolactin in pregnancy hypertension. Ctin Exp Hypert A 9:t099-1119

13. Parmeggiani A, Bottiglioni F, Guerresi F, Montesano L, Silvi E (1976) L'accrescimento umano intrauterino: II Analisi statistica e proposte di strumento diagnostico. Statistica 2:281-2%

14. Poonai APV, Tang K, Poonai PV (1975) Relation between glucose, insulin and growth hormone in the fetus during labor and at delivery. Obstet Gyneco145:155-158

15. Quirk JG jr, MacDonald PC, Johnston JM (1982) Role of fetal pituitary prolactin in fetal lung maturation. Sere Perinatol 6:238-249

16. Shutt DA, Smith ID, Shearman RP (1974) Oestrone, 17-beta-oestradiol and oestriol levels in human fetal plasma during gestation and at term. J Endocrinol 60:333-343

17. Siegel S (1956) Nonparametric statistics for the behavioral sciences. McGraw-Hill, New York, NY

18. Siler-Khodr TM, Morgenstern LL, Greenwood FC (1974) Hormone synthesis and release from human fetal adenohypophysis in vitro. J Clin Endocrinol Metab 39:891-899

19. Sowers JR, Nyby M, Jasberg K (1982) Dopaminergic control of prolactin and blood pressure: altered control in essential hypertension. Hypertension 4:431-438

20. SPSS Inc (1986) SPSSx User's Guide. 2nd edn. McGraw-Hill, New York, NY 21. Stumpe KO, Higuchi M, Kollock R, Kruck F, Vetter H (1977) Hyperprotactinaemia and

antihypertensive effect of bromocriptine in essential hypertension. Lancet II:211-214 22. Weber PC, Siess PC, Sherer B, Held E, Witzgall H, Lorenz R (1982) Arachidonic acid

metabolites in hypertension and arteriosclerosis. Klin Wochenschr 60:479-488 23, Winters AJ, Colston C, MacDonald PC, Parker JC (1975) Fetal plasma prolactin levels. J Clin

Endocrinol Metab 41:626-634 24. Yaginuma T (1981) Possible circadian periodicity of fetal prolactin secretion late in gestation.

Acta Endocrino198:106-111

Received May 18, 1989/Accepted November 29, 1989

maternal and fetal prolactin in pregnancy-induced hypertension

Documents