0021-972X/78/4704-0894$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1978 by The Endocrine Society

Vol. 47, No. 4 Printed in U.S.A.

Prolonged Hyperprolactinemia in Preterm Infants MAX PERLMAN, JOSEPH SCHENKER, MEIRA GLASSMAN, AND MENASHE BENDAVID Departments of Pediatrics, Obstetrics and Gynecology, and Pharmacology, Hadassah University Hospital and Hebrew University-Hadassah Medical School, Jerusalem, Israel ABSTRACT. Serum PRL levels were followed serially in full term (FT; 39-41 weeks) and preterm (PT; 30-32 weeks) infants, from birth to 12 and 20 postnatal weeks, respectively. Values were higher in FT infants than in PT infants on day 1 after birth (267 ± 20 vs. 156 ± 8 ng/ml) but were similar in both by the age of 2-4 weeks (69 ± 8 vs. 69 ± 6 ng/ml). Between the ages of 4-12 weeks, the serum PRL in FT infants fell to near adult levels (24 ± 2 ng/ml), but this fall was seen much later


TUDIES of pituitary gland and of serum PRL concentrations have established that the human fetal pituitary gland can synthesize, store, and release PRL (1, 2). The concentration of PRL in cord blood rises with increasing gestational age and levels exceed those of maternal blood at term (1-4). Serum levels of PRL in full term (FT) infants fall fairly rapidly in the first postnatal week, with a subsequent more gradual decline to near-adult levels at 6 weeks of age (1). The progressive rise of serum PRL values with gestation together with the postnatal fall in serum PRL have been noted to approximately parallel fetal and neonatal estrogen levels (2) and a causative relationship has been suggested. We have made more extended postnatal observations of PRL levels in FT and preterm (PT) infants which show that high serum PRL levels are maintained for at least 12 weeks in PT infants. These results suggest that stimulation of the anterior pituitary gland by gestational estrogens provides only a partial explanation for the high serum PRL values in early infancy. Received July 11, 1977. Address requests for reprints to: Dr. M. Perlman, Department of Pediatrics, Hadassah University Hospital, P.O. Box 499, Jerusalem, Israel.

in PT infants, between 12-20 weeks postnatally (23 ± 2 ng/ml). When values in FT and PT infants were compared at parallel postmenstrual ages in contradistinction to postnatal ages, a similar course of PRL was discernable in both groups. These data may provide indirect evidence for the establishment and maturation of inhibition of PRL secretion (i.e. PRL-inhibitory factor production) postnatally, between 44-52 weeks postmenstrually. (J Clin Endocrinol Metab 47: 894, 1978)

Materials and Methods The subjects were selected according to criteria of gestational age and absence of complicating factors. A total of 12 PT infants, 30-32 weeks gestational age, provided 45 samples between the postnatal ages of 1 day and 12 weeks (Fig. 1). For PRL determination, 0.6 ml blood was obtained from the subjects at the time of sampling for routine blood tests. Birth weight, length, and head circumference were between the 25th and 75th percentiles for gestational age, clinical assessment of gestational age was appropriate for calculated age, and the course was uncomplicated. Eight normal FT infants provided blood samples at 2-4 weeks of age and 6 others provided blood samples at the age of 12 weeks. Informed parental consent was obtained for blood sampling. Blood was taken from the antecubital vein in most instances. Special attention was paid to rapid sampling in an attempt to minimize possible effects of stress. After clotting, the serum was separated by centrifugation at 4 C and samples were stored at -20 C until assayed by a homologous RIA (5). "Cold" and "labeled" PRL were prepared in our laboratory (5) and standardized according to the Lewis hPRL preparation. Antihuman PRL was kindly provided by the NIAMDD, NIH, Bethesda, MD.

Results FT infants There was a sharp fall in serum PRL con894

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 05:34 For personal use only. No other uses without permission. . All rights reserved.



postconceptual age. At 20 weeks of age, PT infants had a mean serum PRL value of 23 ng/ml, which was identical to that of FT at 12 weeks when the postconceptual age was equivalent. • FT


100 -



4 6 8 10 12 U 16 18 20 POSTNATAL AGE (weeks)

FIG. 1. Serum PRL values in relation to postnatal age in FT and PT infants.

centrations from a mean of 267 ng/ml at 1 day to a mean of 69 ng/ml at the age of 2-4 weeks (Fig. 1). At the age of 12 weeks, FT had a mean PRL value of 24 ng/ml, which is similar to normal adult values in this laboratory (3-30 ng/ml). This value was significantly lower (P < 0.01) than that of PT of the same postnatal >age (60 ng/ml).

PT infants The mean serum PRL value of 156 ng/ml 1 day after birth was significantly (P < 0.01) • less than that of FT (266.7 ng/ml) of the same postnatal age (Fig. 1). Values decreased at 1.5, 2.5, and 4 weeks to a mean of 69 ng/ml, with a subsequent plateau to 12 weeks when the mean level was 60 ng/ml. This was significantly higher than the value in FT of the same >• postnatal age and was still well above adult values. On the other hand, the value at 12 weeks was not significantly different from that of FT infants of 2-4 weeks of age, a time at which both groups of infants were of similar

Our data confirm previous observations of the ability of the human neonate to maintain high serum PRL values after birth (1, 2) and of differences between FT and PT infants (1). In this study, we have shown that the prolonged hyperprolactinemia of infancy is even more marked in PT infants (>12 weeks) than in FT infants (4-6 weeks) (derived from combined data of Guyda and Friesen (1), Aubert et al. (6), and the present study). Aubert et al. (2) have recently summarized current knowledge of PRL secretion in the human neonate, including evidence that the fetal pituitary is the major if not the exclusive source of the hormone detected in cord blood. On the basis of observations in normal and anencephalic fetuses as well as an apparent relationship between serum PRL values and levels of estrogen, they concluded that the high levels of estrogens may explain the high concentration of PRL in the fetus in late gestation. However, the prolonged hyperprolactinemia for at least 12 weeks after birth in PT infants (demonstrated in the present study) seems to be unrelated to gestational estrogens. In PT infants, the initial fall in PRL levels is followed by levels similar to those seen in FT infants from the age of 4 days to the age of 4 weeks (Fig. 1). The plateau of serum PRL levels, however, is more prolonged in PT infants and the fall to near-adult levels occurs between 12-20 weeks postnatal age, in contrast to the period between 4-12 weeks in FT infants. These findings agree with the data of Guyda and Friesen in FT and PT infants aged 6 weeks (1). The differences between PT and FT infants seem to be comprehensible when age is expressed as postmenstrual weeks (postconceptual plus 2 weeks) rather than postnatal age (Fig. 2). The consistency of the relationship of postmenstrual age in contrast to postnatal age on serum PRL levels may

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 05:34 For personal use only. No other uses without permission. . All rights reserved.


896 300







- 6 - 4 - 2 0 2 4 6 8 10 12 WEEKS BEFORE ANO AFTER TERM ( week 0) FIG. 2. Serum PRL values in relation to weeks before and after term in FT and PT infants.

hold an additional clue to the control of the secretion of this hormone in the perinatal period. The high PRL values observed in early infancy, beyond the period of possible gestational estrogen influence, may be due to relatively unrestrained hypothalamic stimulation by either a putative PRL-releasing factor or TRH or, alternatively, to reduction or absence of hypothalamic inhibition by PRL-inhibitory factor (PIF). Normal PRL values in anencephalics (2, 3) militate against the former explanation and favor the idea of deficiency in PIF, insensitivity of the pituitary lactotrophs to the action of PIF, or both. The demonstration of increased serum PRL levels after TRH stimulation of fetal and neonatal monkeys (7, 8) and human anencephalics (2, 3) indicates that the lactotrophs are responsive to this humoral influence. However, the apparently minor influence of delivery stress on PRL secretion, evidenced by similar PRL values in cord blood and on day 1 and by the small rise in serum PRL levels in the early minutes after birth (9), suggests a relative independence of PRL secretion from the central nervous sys-

JCE&M Vol47

1978 No 4

tem in this age group. This idea is reinforced by the observation that reserpine administra- r tion to the near-term ovine fetus did not raise plasma PRL levels (10). This was interpreted to indicate lack of development of inhibitory pituitary control of PRL secretion in the mature fetus of this species. Neurological inhibition of a number of an- , terior pituitary hormones seems to develop late relative to stimulation, in infancy rather than in fetal life, and has been related to myelinization and cerebral cortical development (11). The fact that, in contrast to some other pituitary hormones, PRL secretion is A predominantly determined by hypothalamic inhibition is consistent with the course of relatively high serum PRL values in the fetus and neonate. Serum values of GH were also observed to be high in the human neonate (12), both in FT and PT infants. The delay in < the postnatal fall of serum GH levels in PT as compared to FT infants is analogous to that of PRL seen in the present study and may . suggest a common ontogenetic mechanism for PIF and somatostatin. We conclude that the relatively high levels

Prolonged hyperprolactinemia in preterm infants.

0021-972X/78/4704-0894$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1978 by The Endocrine Society Vol. 47, No. 4 Printed in U...
374KB Sizes 0 Downloads 0 Views