- Email: [email protected]
Effect of maternal administration of thyrotropin releasing hormone on the preterm fetal pituitary-thyroid axis
Effect of maternal administration of thyrotropin releasing hormone on the preterm fetal pituitary-thyroid axis F e r n a n d o M o y a , MD, Patricia M e n a , MD, A r n a l d o Foradori, MD, Mario Becerra, MD, AIvaro Inzunza, MD, a n d A l f r e d o G e r m a i n , MD From the Division of NeonataI-Perinatal Medicine, Departments of Pediatrics and Obstetrics and Gynecology, Universityof Texas Southwestern Medical Center at Dallas, the Hospital Sotero del Rio, Puente Alto, Chile, and the Departments of Nuclear Medicine and of Obstetrics and Gynecology, Universidad Catolica de Chile, Santiago, Chile We e v a l u a t e d the response of preterm fetuses to maternal intravenous injection of 400 ~g of thyrotropin releasing hormone (TRH) b e t w e e n 30 minutes and 5 hours before delivery (n = 12). An additional seven mothers received saline solution and served as control subjects. There were no statistically significant differences in gestational age, birth weight, or A p g a r scores between groups. At delivery, concentrations of maternal thyrotropin were e l e v a t e d in the TRH group compared with the control group (12.0 _+ 1.6 vs 5.6 • 0.5 mU/L; p <0.005); however, maternal triiodothyronine (1'3) values remained unchanged. Significant elevations of fetal thyrotropin and T3 were observed after maternal administration of TRH c o m p a r e d with control subjects (45.8 • 7.7 vs 8.4 • 0.9 mU/L [p <0.002] and 1.3 _+ 0.07 vs 0.7 _+ 0.04 nmol/L or 87 + 5 vs 49 9 3 n g / d l (p <0.001], respectively). Fetal thyroxine (T4) and prolactin values were also e l e v a t e d after exposure to TRH (135 • 5 vs 86 • 10 nmol/L or 10.5 • 0.4 vs 6.7 _+ 0.8 #g/dl [p <0.001] and 212 • 31 vs 105 • 28 ~g/L (p <0.05], respectively). Two hours after birth, a significant increase in T3 but not T4levels was observed in both groups of infants. These data indicate that fetal exposure to a single dose of TRH via maternal administration of this hormone results in marked stimulation of the preterm fetal pituitary-thyroid axis, as in the fetus at term, and that this treatment does not inhibit the early postnatal surge of Tz, (J PEDIATR1991;119:966-71)
Thyrotropin releasing hormone crosses the placenta readily in laboratory animals and human beings.~4 Once in the fetal circulation, TRH stimulates the release of thyrotropin and prolactin by the fetal pituitary gland. We have reported Presented in part at the Southern Society for Pediatric Research Meeting, New Orleans, La., Jan. 17-19, 1990 (published in abstract form: Clin Res 1990;38:48A and Pediatr Res 1990; 27:81A). Submitted for publication May 6, 1991; accepted June 21, 1991. Reprint requests: Fernando R. Moya, MD, Associate Professor of Pediatrics, Obstetrics, and Gynecology,Department of Pediatrics and Obstetrics and Gynecology,University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Blvd., Dallas, TX 75235-9063. 9/23/31953
966
that maternal administration of 400 ~g of TRH 2 hours before cesarean section at term results in elevations of cord blood levels of thyrotropin and triiodothyronine.4 A higher dose of TRH (600/zg) achieved similar results. In that study the postnatal surge of thyrotropin and T3 in term infants RDS T3 T4 TRH
Respiratory distress syndrome Triiodothyronine Thyroxine Thyrotropin releasing hormone
exposed to TRH before delivery was similar to that observed in control infants. Recently, other effects of TRH in the fetus have received considerable attention. Fetal exposure to TRH seems to ac-
Volume 119 Number 6
Effect o f prenatal T R H on fetal pituitary-thyroid axis
celerate fetal lung maturation, especially when used in conjunction with glucocorticoids,s-7 The mechanisms for this interaction are still unclear. However, effects of TRH on surfactant secretion, alveolar protein leak, ontogeny of /3-adrenergic receptors, and fetal breathing have been described. 5-m Because of the potential effect of T R H on prevention of respiratory distress syndrome, its use in combination with glucocorticoids has been suggestedJ 1, z2 If T R H is used to prevent RDS, fetuses would be exposed to this hormone during the preterm period rather than at term. The fetal pituitary-thyroid axis undergoes substantial maturation during the second half of gestation. Fetal serum levels of free T3 and thyroxine increase steadily from mid gestation to term, whereas fetal serum concentrations of thyrotropin tend to reach a plateau or to decrease slightly after 30 weeks of gestation.13 The response to TRH has been examined in a small number of premature neonates. 14, 15 Their mean increment of thyrotropin was relatively close to that of term newborn infants.16 However, these studies were done at variable postnatal ages when significant changes in the pituitary-thyroid axis would have already occurred. Furthermore, whether in utero exposure to TRH interferes with the postnatal changes of thyrotropin and thyroid hormones in more immature neonates is not known. We therefore decided to conduct this study. METHODS Pregnant women admitted to the Hospital Sotero del Rio, Puente Alto, Chile, in premature labor or about to undergo cesarean section before 34 weeks of gestation were eligible for the study. After giving consent, women were injected intravenously with either 1 ml of isotonic saline solution, the diluent in the T R H preparation (control subjects, n = 7), or 400 #g of TRH (Thypinone), between 30 minutes and 5 hours before delivery (n = 12). The intravenous injection was given slowly for several minutes while maternal vital signs, uterine contractility, and fetal heart rate were being monitored continuously. The dose of TRH, 400 #g, was based on our previous study showing no further stimulation of the fetal pituitary-thyroid axis with higher doses. 4 Hormonal determinations, Blood samples were collected from all women at the time of delivery, and from mixed umbilical cord blood because no significant arteriovenous umbilical cord differences have been noted for TSH, T3, and T4.17 Blood samples were also obtained at 2 hours after birth from all infants. The serum was separated quickly and frozen at - 2 0 ~ C until analysis. All serum samples were measured at least in duplicate in the same assay. Determinations of thyrotropin, total T3, total T4, and prolactin were done by radioimmunoassay as previously described. 4 Statistical analysis, Statistical analysis was performed with paired and unpaired Student t tests. A p value of <0.05
967
Table I. Obstetric and neonatal data
Gestational age (wk) Range Birth weight (kg) Range Presence of labor Delivery by cesarean section Apgar score at 5 rain <5 Mechanical ventilation
Control group ( n = 7)
TRH (n=t2)
29.1 _+ 0.9 26-34 1.36 _+ 0.33 1.07-2.07 7 3 1 2
30.4 _+ 0.8 26-34 1.61 + 0.14" 0.99-2.48 11 6 0 3
*p= 0.13. was considered significant. All values correspond to mean _+ SEM. This study was approved by the human investigation committees of the participating institutions. RESULTS Nineteen pregnant women and their newborn infants were studied. There was one twin pregnancy in the TRH group. No significant side effects were observed during or after the slow intravenous injection of TRH. Table I shows a comparison of several obstetric and neonatal variables in the two groups. All women in the control group, and 11 of 12 women in the TRH group, were in preterm labor. No significant differences were observed in gestational age, delivery by cesarean section, number of infants with low 5-minute Apgar score, or number of infants requiring mechanical ventilation. Maternal response to TRH. Deliveries were not elective; thus women given TRH received the injection between 30 minutes and 5 hours before delivery. Maternal thyrotropin concentration in the group injected with TRH was significantly higher than in control subjects ( 1 2 . 0 _ 1.6 vs 5.6 _+ 0.5 mU/L; p <0.005), whereas no differences were observed in maternal T3 after the administration of TRH (3.4 + 0.I vs 3.1 -+ 0.2 nmol/L or 224 _+ 9 vs 206 +-+_16 ng/dl). Fetal response to TRH (Table II). Among the control group, the mean fetal thyrotropin value was significantly higher than the maternal values (8.4 + 0.9 vs 5.6 +_ 0.5 mU/L; p <0.05). On the contrary, the mean fetal T3 value was much lower than the maternal concentrations (0.7 _+ 0.04 vs 3.1 + 0.2 nmol/L or 49 + 3 vs 206 _+ 16 ng/dl; p <0.01). Significant elevations of thyrotropin, T3, T4, and prolactin concentrations were observed after prenatal exposure to TRH between 30 minutes and 5 hours before delivery. As a group, preterm fetuses exposed to TRH bad approximately a sixfold increase in thyrotropin values and a 75% increase in T3 values. However, if only those fetuses delivered 2 hours after maternal TRH admin-
968
Moya et al,
The Journal of Pediatrics December 1991
CORD BLOOD TSH ( mUlL )
12o{ I
100~ I
8o! I [
I
40! L
2oh
[- .......................................................................................................
i ....................................................................................................
0 ~
0
l
l
L
!
l[ ~
1 2 INTERVAL B E T W E E N
J
A
J-
L
3 4 5 TRH AND BIRTH ( hours )
6
Fig. t. Individual cord-blood values of thyrotropin (TSH) ([]) from fetuses exposed to TRH versus interval between maternal TRH injection and birth. Mean values (+__2 SD) for control fetuses are plotted for comparison.
T a b l e II. Thyrotropin, T3, T4, and prolactin in cord blood and 2 hours after birth Control values
ThyrotroPin (mU/L) T3 (nmol/L [ng/dl]) T4 (nmol/L [~zg/dl]) Prolactin (~zg/L)
TRH
Cord
2 Hr
Cord
2 Hr
8.4 +_ 0.9 0.7 +_ 0.04 (49 +_ 3) 86 +_ 10 (6.7 _+ 0.8) 105 _+ 28
-1.2 +_ 0,1 (80 _+ 10)t 92 .+__5 (7.2 +- 0.4) --
45.8 _+ 7.7* 1.3 _+ 0,07 (87 _+ 5)* 135 _+ 5 (10.5 +_ 0.4)* 212 +_ 31w
-2.0 _ 0.2 (134 _+ 14)~: 122 _+ 5 (9.9 +_ 0.5) --
*p <0.002 by unpaired t test versus control cord values. tp = 0.01 by paired t test between cord and 2-hour values, ~p <0.005 by paired t test between cord and 2-hour values. w <0.05 by unpaired t test versus control cord values.
istration are considered (n = 5; Figs. 1 and 2), their thyrotropin level was elevated eightfold (66.5 +- 13.1 m U / L ) and the T3 level doubled (I.5 _+ 0.06 n m o l / L or 101 +_ 4 ng/dl). The influence of the interval between maternal T R H administration and birth on fetal serum thyrotropin and T3 concentrations is shown in Figs. 1 and 2. All fetal serum thyrotropin values and 12 of 13 fetal T3 values were elevated above control values after T R H stimulation up to 5 hours before delivery. Significant increases of serum concentrations of T3 were observed 2 hours after birth in all but one control infant and in all neonates in the T R H group (Table II), No postnatal changes of serum T4 were observed 2 hours after birth in either the control group or the T R H group,
DISCUSSION Pregnancy has a marked influence on the maternal pituitary-thyroid axis. Throughout gestation, total serum T3 and T4 concentrations increase, but after 20 weeks of gestation the maternal thyrotropin concentration remains unchanged.~3 The thyrotropin response to T R H does not seem to be modified by pregnancy, but the prolactin response is relatively lower, t8 In our previous study, T R H was administered to pregnant women at term whose babies were delivered by elective cesarean section without labor. 4 Significant increases of thyrotropin and prolactin were observed, but no changes in serum T3 levels occurred up to 2 hours after the T R H injection, In the present study, we evaluated the response to T R H in a group of women with preterm
Volume 119 Number 6
Effect of prenatal T R H on fetal pituitary-thyroid axis
969
CORD BLOOD T ( nrnol/L )
3
2 -
0 1.5 ~r
<>
<> 1~
r
0
0
0
0
0.5
O
--
0
I
~ _ _ ~
1
I
I
2
]
3
I
I _ _ L
4
I
l
5
6
INTERVAL B E T W E E N TRH AND BIRTH ( hours ) Fig. 2. Individual cord-blood values of T3 (0) from fetuses exposed to TRH versus interval between maternal TRH injection and birth. Mean values (_+2 SD) for control fetuses are plotted for comparison.
pregnancies, of whom all but one were in labor. A significant increase in the maternal thyrotropin level was also seen after TRH administration. Comparison of the magnitude of the maternal thyrotropin response between preterm and term pregnancies is not possible in this study because we determined thyrotropin values only at delivery, which occurred at a variable interval of time after TRH administration. In accordance with previous observations by us and other investigators, no changes in maternal serum T3 concentrations were seen after TRH administration. 4' 19 It is possible that the maternal T3 response to thyrotropin is decreased during the second half of pregnancy and during labor, when baseline T3 values have risen above those of nonpregnant women. 19, 20 The placenta is impermeable to thyrotropin; during the second and third trimesters of gestation, the fetal thyrotropin concentration is higher than the maternal values, as confirmed in this study. 2~ Cord blood values of thyrotropin in these preterm fetuses were somewhat higher than those which we reported previously for term fetuses (8.4 _+ 0.9 vs 4.8 _~_-1.0 mU/L). 4 Others also have reported higher fetal thyrotropin values at about 30 weeks of gestation than at term. 2l Furthermore, higher cord blood thyrotropin concentrations in infants born vaginally, in comparison with concentrations in those delivered by cesarean section, have been noted, z2 Thus the stimulus of labor and vaginal delivery might have contributed to a slightly higher cord blood
level of thyrotropin in our control group. In contrast, the fetal T3 values were much lower than the maternal T3 values among the control group. The human placenta is also relatively impermeable to T3 and T4, and maternal-fetal gradients of these hormones have been documented repeatedly.4, 2l, 23 The lower fetal levels of thyroid hormone can be partially explained by active conversion of T4 tO reverse T3, and of T3 to 3,3'-diiodothyronine by the placenta. 13,24 During intrauterine life there is a progressive rise in the fetal T4 level from about mid gestation, followed by an increase in the fetal T3 level a few weeks later. 21 Our findings are in general agreement with the reported increases of T3 and T4 levels during fetal development. 23, 25 Maternal administration of TRH before delivery resulted in elevations of serum thyrotropin, T3, T4, and prolactin levels in these preterm fetuses. The cord blood changes of thyrotropin and T3 observed 2 hours after maternal administration of TRH are relatively similar to those we reported in term fetuses exposed to TRH for that same interval. 4 The postnatal thyrotropin response to TRH has also been shown to be of the same magnitude in preterm and term infants.14, 16 Healthy preterm infants have postnatal changes in thyrotropin, T3, and T4 levels that are quantitatively smaller than those observed in term infants. 21, 26, 27 Furthermore, among sick premature infants with RDS the early rise of the T 4 level may not be observed; however, T3 elevations do oc-
970
Moya et al.
cur. 27 Even though the cord blood and 2-hour values of T3 were lower among control infants, the relative change in the T3 value postnatally was similar in both groups of preterm infants. No significant changes in the T4 value were seen in either group at 2 hours of age. However, most of the postnatal changes in T4 values in term and preterm infants are observed around 24 hours after birth. 15, 21, z7 Although in this study we did not determine the postnatal changes in thyrotropin values, it appears that prenatal exposure to a single dose of TRH does not interfere with normal postnatal changes in the pituitary-thyroid axis in preterm or term infants.4 However, repeated administration of TRH is associated with decreases in the thyrotropin response and increased concentrations of thyroid hormones in human beingsY, 29 Recent studies of human subjects suggest that the combined use of corticosteroids and TRH is more beneficial in preventing and ameliorating RDS and its consequences than the use of corticosteroids alone) 2, 3o Although combination hormonal therapy may seem an attractive new therapeutic modality, several questions remain. Perhaps one of the most important is the significanceof transient elevations of fetal thyroid hormone levels brought about by TRH stimulation during crucial periods of brain development. The in utero increases of thyroid hormones as a result of maternal TRH administration are in the range of the physiologic changes observed after birth in preterm and term neonates.21 This fact and the lack of long-term adverse effects among infants exposed to intraamniotic injection of a large amount of T4 suggest that prenatal use of TRH to enhance fetal lung maturation is relatively safe) 1 We thank Dr. B. J. Green (Abbott Laboratories, North Chicago, Ill.) for supplying the TRH preparation, and the nurses and midwives of the Hospital Sotero del Rio, Puente Alto, Chile, for their assistance during the study. REFERENCES
1. Roti E, Gnudi A, Braverman L, et al. Human cord blood concentrations of thyrotropin, thyroglobulin, and iodothyronines after maternal administration of thyrotropin-releasing hormone. J Clin Endocr Metab 198l;53:813-7. 2. MelmedS, Harada A, Murata Y, et al. Fetal response to thyrotropin-releasing hormone after thyroid hormone administration to the rhesus monkey: lack of pituitary suppression. Endocrinology 1979;105:334-41. 3. Devaskar U, Nitta K, Szewczyk K, Sadig F, deMello D. TranspiacentaI stimulation of functional and morphologicfetal rabbit lung maturation: effect of thyrotropin-releasinghormone. Am J Obstet Gyneeol 1987;157:460-4. 4. Moya F, Mena P, Heusser F, et al. Response of the maternal, fetal, and neonatal pituitary-thyroid axis to thyrotropinreleasing hormone. Pediatr Res 1986;20:982-6. 5. Rooney S, Marino P, Gobran L, Gross I, Warshaw J. Thyrotropin-releasinghormoneincreasesthe amount of surfactant in lung lavage from fetal rabbits. Pediatr Res 1979;13:623-5.
The Journal of Pediatrics December 1991
6. Ikegami M, Jobe AH, Petenazzo A, Seidner SR, Berry DB, Ruffini L. Effects of maternal treatment with corticosteroids, T3, TRH, and their combinationson lung function of ventilated preterm rabbits with and without surfactant treatments. Am Roy Respir Dis 1987;136:892-8. 7. Warburton D, Parton L, BuckleyS, Cosico L, Enns G, Saluna T. Combined effects of corticosteroid, thyroid hormones, and ~-agonist on surfactant, pulmonary mechanics,and/~-receptor binding in fetal lamb lung. Pediatr Res 1988;24:166-70. 8. Oulton M, Rasmusson MG, Yoon RY, Fraser M. Gestationdependent effects of the combinedtreatment of glacocorticoids and thyrotropin-releasing hormone on surfactant production by fetal rabbit lung. Am J Obstet Gynecol 1989;160:961-7. 9. Tabor B, Ikegami M, Jobe AH, Yamada T, Oetomo SB. Dose response of thyrotropin-releasinghormone on pulmonary maturation in corticosteroid-treatedpreterm rabbits. Am J Obstet Gynecol 1990;163:669-76. 10. Umans JG, Umans HR, Szeto HH. Effects of thyrotropin-releasing hormone in the fetal lamb. Am J Obstet Gynecol 1986;155:1266-71. 11. Moya FR, Gross I. Prevention of respiratory distress syndrome. Semin Perinatol 1988;12:348-58. 12. Morales WJ, O'Brien WF, Angel JL, Knuppel RA, Sawai S. Fetal lung maturation: the combineduse of corticosteroids and thyrotropin-releasing hormone. Obstet Gynecol 1989;73: 111-6. 13. Fisher DA. Maternal-fetal thyroid function in pregnancy. Clin PerinatoI I983;i0:615-26. 14. Jacobsen BB, Andersen H, Dige-Petersen H, Hummer L. Pituitary-thyroid responsiveness to thyrotropin-releasing hormone in preterm and small-for-gestationalage newborns.Acta Paediatr Scand 1977;66:541-8. 15. Cuestas RA. Thyroid function in healthy premature infants. J PEDIATR1978;92:963-7. 16. Jacobsen BB, Andersen H, Dige-Petersen H, Hummer L. Thyrotropin responseto thyrotropin-releasinghormone in full term, euthyroid and hypothyroid newborns. Aeta Paediatr Scand 1975;65:433-8. 17. Penny R, Simms ME, Campbell WG, Spencer CA, Nicoloff JT. Thyroid indices in arterial and venous cord blood: significantly greater levels of reverse triiodothyronine in venous blood than in arterial blood. Metabolism 1986;35:645-8. 18. Yl~korkalaO, Kivinen S, Reinila M. Serial prolactin and thyretropin responses to thyrotropin-releasing hormone throughout normal human pregnancy. J Clin Endocrinol Metab 1979; 48:288-92. 19. Miyamoto J. Prolactin and thyrotropin responses to thyrotropin-releasing hormone during the peripartal period. Obstet Gynecol 1984;63:639-44. 20. Harada A, Hershman JM, Reed AW, et al. Comparison of thyroid stimulators and thyroid hormone concentrationsin the sera of pregnant women. J Clin Endocrinol Metab 1979;48: 793-7. 21. Fisher DA, Klein AH. Thyroid developmentand disorders of thyroid function in the newborn.N Engl J Med 1981;304:70212. 22. Lao TT, Panesar NS. Neonatal thyrotropin and mode of delivery. Br J Obstet Gynecol 1989;96:1224-7. 23. Fisher DA, Dussault JH, Sack J, Chopra IJ. Ontogenesis of hypothalamic-pituitary-thyroid function and metabolism in man, sheep, and rat. Recent Prog Horm Res 1977;33:59-116. 24. Roti E, Fang S, Green K, Emerson C, Braverman L. Human placenta is an active site of thyroxine and 3,3' ,5-triiodothyro-
Volume 119 Number 6
25.
26.
27.
28.
Effect o f prenatal T R H on fetal pituitary-thyroid axis
nine tyrosyl ring deiodination. J Clin Endocrinol Metab 198 I; 53:498-501. Thorpe-BeestonJG, Nicolaides KH, Felton CV, Butler J, McGregor AM. Maturation of the secretion of thyroid hormone and thyroid-stimulating hormone in the fetus. N Engl J Med 1991;324:532-6. Abbassi V, Merchant K, Abramson D. Postnatal triiodothyronine concentrations in healthy preterm infants and in infants with respiratory distress syndrome. Pediatr Res 1977;11: 802-4. Klein AH, Foley B, Kenney FM, Fisher DA. Thyroid hormone and thyrotropin responses to parturition in premature infants with and without the respiratory distress syndrome. Pediatrics 1979;63:380-5. Snyder PJ, Utiger RD. Repetitive administration of thyrotro-
971
pin-releasing hormone results in small elevations of serum thyroid hormones and in marked inhibition of thyrotropin response. J Clin Invest 1973;52:2305-12. 29. Staub JJ, Girard J, Mueller-Brand J, et al. Blunting of TSH response after repeated oral administration of TRH in normal and hypothyroid subjects. J Clin Endocrinol Metab 1978;46: 260-6. 30. Ballard RA, Ballard PL, Creasy R, Gross I, Padbury JP. Prenatal thyrotropin-releasing hormone plus corticosteroid decreases chronic lung disease in very low birth weight infants [Abstract]. Clin Res 1990;38:192A. 31. Barkai G, Zarfin Y, Ben-Harari M, Reichman B, Sack J, Mashiaeh S. In utero thyroxine therapy for the induction of fetal lung maturity: long-term effects. J Perinat Med 1988; 16:145-8.
Clinical and laboratory observations Cerebrospinal fluid examination in symptom-free infants with risk factors for infection Susan Fielkow, MD, Susan Reuter, RN, a n d Samuel P. G o t o f f , MD From the Department of Pediatrics, Rush Presbyterian-St. Luke's Medical Center, Chicago, Illinois
Cerebrospinal fluid analysis is frequently performed on newborn infants as part of the evaluation for suspected bacterial infection. It is important to identify newborn infants with meningitis, but the number with meningitis is less than 3% of those evaluated, and lumbar puncture is associated with some risk, as well as costJ' a In addition, there is a high rate of unsuccessful or traumatic lumbar punctures.i~3 Increased attack rates of early-onset group B streptococcal infections4 and early-onset infections in general5 have been clearly related to prolonged rupture of membranes and chorioamnionitis. Attempts to identify infants with infec-
tions as early as possible has led to the practice of a "sepsis workup" in symptom-free infants, based on the presence of maternal risk factors such as chorioamnionitis. Although bacteremia has been documented in symptom-free infants,6, 7 which may justify a blood culture, we questioned the benefit of a CSF examination in addition to blood culture in symptom-free neonates and performed a retrospective chart review to answer the question.
See related article, p. 973.
I Presented at the meeting of the Midwest Society for Pediatric Research, Chicago, IlL, Nov. 2, 1990. Submitted for publication April 29, 1991; accepted June 17, 1991. Reprint requests: Samuel P. Gotoff, MD, Department of Pediatrics, Rush Presbyterian-St. Luke's Medical Center, 1653 W. Congress Pkwy., Chicago, IL 60612. 9/24/31835
CSF
Cerebrospinalfluid
METHODS The clinical and laboratory records accumulated during the 10-year period from 1978 to 1987 on all newborn infants from whom cerebrospinal fluid culture specimens were taken within the first week of life were reviewed. The infants were either inborn or transferred to Rush Presbyterian-St.
Thyrotropin releasing hormone crosses the placenta readily in laboratory animals and human beings.~4 Once in the fetal circulation, TRH stimulates the release of thyrotropin and prolactin by the fetal pituitary gland. We have reported Presented in part at the Southern Society for Pediatric Research Meeting, New Orleans, La., Jan. 17-19, 1990 (published in abstract form: Clin Res 1990;38:48A and Pediatr Res 1990; 27:81A). Submitted for publication May 6, 1991; accepted June 21, 1991. Reprint requests: Fernando R. Moya, MD, Associate Professor of Pediatrics, Obstetrics, and Gynecology,Department of Pediatrics and Obstetrics and Gynecology,University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Blvd., Dallas, TX 75235-9063. 9/23/31953
966
that maternal administration of 400 ~g of TRH 2 hours before cesarean section at term results in elevations of cord blood levels of thyrotropin and triiodothyronine.4 A higher dose of TRH (600/zg) achieved similar results. In that study the postnatal surge of thyrotropin and T3 in term infants RDS T3 T4 TRH
Respiratory distress syndrome Triiodothyronine Thyroxine Thyrotropin releasing hormone
exposed to TRH before delivery was similar to that observed in control infants. Recently, other effects of TRH in the fetus have received considerable attention. Fetal exposure to TRH seems to ac-
Volume 119 Number 6
Effect o f prenatal T R H on fetal pituitary-thyroid axis
celerate fetal lung maturation, especially when used in conjunction with glucocorticoids,s-7 The mechanisms for this interaction are still unclear. However, effects of TRH on surfactant secretion, alveolar protein leak, ontogeny of /3-adrenergic receptors, and fetal breathing have been described. 5-m Because of the potential effect of T R H on prevention of respiratory distress syndrome, its use in combination with glucocorticoids has been suggestedJ 1, z2 If T R H is used to prevent RDS, fetuses would be exposed to this hormone during the preterm period rather than at term. The fetal pituitary-thyroid axis undergoes substantial maturation during the second half of gestation. Fetal serum levels of free T3 and thyroxine increase steadily from mid gestation to term, whereas fetal serum concentrations of thyrotropin tend to reach a plateau or to decrease slightly after 30 weeks of gestation.13 The response to TRH has been examined in a small number of premature neonates. 14, 15 Their mean increment of thyrotropin was relatively close to that of term newborn infants.16 However, these studies were done at variable postnatal ages when significant changes in the pituitary-thyroid axis would have already occurred. Furthermore, whether in utero exposure to TRH interferes with the postnatal changes of thyrotropin and thyroid hormones in more immature neonates is not known. We therefore decided to conduct this study. METHODS Pregnant women admitted to the Hospital Sotero del Rio, Puente Alto, Chile, in premature labor or about to undergo cesarean section before 34 weeks of gestation were eligible for the study. After giving consent, women were injected intravenously with either 1 ml of isotonic saline solution, the diluent in the T R H preparation (control subjects, n = 7), or 400 #g of TRH (Thypinone), between 30 minutes and 5 hours before delivery (n = 12). The intravenous injection was given slowly for several minutes while maternal vital signs, uterine contractility, and fetal heart rate were being monitored continuously. The dose of TRH, 400 #g, was based on our previous study showing no further stimulation of the fetal pituitary-thyroid axis with higher doses. 4 Hormonal determinations, Blood samples were collected from all women at the time of delivery, and from mixed umbilical cord blood because no significant arteriovenous umbilical cord differences have been noted for TSH, T3, and T4.17 Blood samples were also obtained at 2 hours after birth from all infants. The serum was separated quickly and frozen at - 2 0 ~ C until analysis. All serum samples were measured at least in duplicate in the same assay. Determinations of thyrotropin, total T3, total T4, and prolactin were done by radioimmunoassay as previously described. 4 Statistical analysis, Statistical analysis was performed with paired and unpaired Student t tests. A p value of <0.05
967
Table I. Obstetric and neonatal data
Gestational age (wk) Range Birth weight (kg) Range Presence of labor Delivery by cesarean section Apgar score at 5 rain <5 Mechanical ventilation
Control group ( n = 7)
TRH (n=t2)
29.1 _+ 0.9 26-34 1.36 _+ 0.33 1.07-2.07 7 3 1 2
30.4 _+ 0.8 26-34 1.61 + 0.14" 0.99-2.48 11 6 0 3
*p= 0.13. was considered significant. All values correspond to mean _+ SEM. This study was approved by the human investigation committees of the participating institutions. RESULTS Nineteen pregnant women and their newborn infants were studied. There was one twin pregnancy in the TRH group. No significant side effects were observed during or after the slow intravenous injection of TRH. Table I shows a comparison of several obstetric and neonatal variables in the two groups. All women in the control group, and 11 of 12 women in the TRH group, were in preterm labor. No significant differences were observed in gestational age, delivery by cesarean section, number of infants with low 5-minute Apgar score, or number of infants requiring mechanical ventilation. Maternal response to TRH. Deliveries were not elective; thus women given TRH received the injection between 30 minutes and 5 hours before delivery. Maternal thyrotropin concentration in the group injected with TRH was significantly higher than in control subjects ( 1 2 . 0 _ 1.6 vs 5.6 _+ 0.5 mU/L; p <0.005), whereas no differences were observed in maternal T3 after the administration of TRH (3.4 + 0.I vs 3.1 -+ 0.2 nmol/L or 224 _+ 9 vs 206 +-+_16 ng/dl). Fetal response to TRH (Table II). Among the control group, the mean fetal thyrotropin value was significantly higher than the maternal values (8.4 + 0.9 vs 5.6 +_ 0.5 mU/L; p <0.05). On the contrary, the mean fetal T3 value was much lower than the maternal concentrations (0.7 _+ 0.04 vs 3.1 + 0.2 nmol/L or 49 + 3 vs 206 _+ 16 ng/dl; p <0.01). Significant elevations of thyrotropin, T3, T4, and prolactin concentrations were observed after prenatal exposure to TRH between 30 minutes and 5 hours before delivery. As a group, preterm fetuses exposed to TRH bad approximately a sixfold increase in thyrotropin values and a 75% increase in T3 values. However, if only those fetuses delivered 2 hours after maternal TRH admin-
968
Moya et al,
The Journal of Pediatrics December 1991
CORD BLOOD TSH ( mUlL )
12o{ I
100~ I
8o! I [
I
40! L
2oh
[- .......................................................................................................
i ....................................................................................................
0 ~
0
l
l
L
!
l[ ~
1 2 INTERVAL B E T W E E N
J
A
J-
L
3 4 5 TRH AND BIRTH ( hours )
6
Fig. t. Individual cord-blood values of thyrotropin (TSH) ([]) from fetuses exposed to TRH versus interval between maternal TRH injection and birth. Mean values (+__2 SD) for control fetuses are plotted for comparison.
T a b l e II. Thyrotropin, T3, T4, and prolactin in cord blood and 2 hours after birth Control values
ThyrotroPin (mU/L) T3 (nmol/L [ng/dl]) T4 (nmol/L [~zg/dl]) Prolactin (~zg/L)
TRH
Cord
2 Hr
Cord
2 Hr
8.4 +_ 0.9 0.7 +_ 0.04 (49 +_ 3) 86 +_ 10 (6.7 _+ 0.8) 105 _+ 28
-1.2 +_ 0,1 (80 _+ 10)t 92 .+__5 (7.2 +- 0.4) --
45.8 _+ 7.7* 1.3 _+ 0,07 (87 _+ 5)* 135 _+ 5 (10.5 +_ 0.4)* 212 +_ 31w
-2.0 _ 0.2 (134 _+ 14)~: 122 _+ 5 (9.9 +_ 0.5) --
*p <0.002 by unpaired t test versus control cord values. tp = 0.01 by paired t test between cord and 2-hour values, ~p <0.005 by paired t test between cord and 2-hour values. w <0.05 by unpaired t test versus control cord values.
istration are considered (n = 5; Figs. 1 and 2), their thyrotropin level was elevated eightfold (66.5 +- 13.1 m U / L ) and the T3 level doubled (I.5 _+ 0.06 n m o l / L or 101 +_ 4 ng/dl). The influence of the interval between maternal T R H administration and birth on fetal serum thyrotropin and T3 concentrations is shown in Figs. 1 and 2. All fetal serum thyrotropin values and 12 of 13 fetal T3 values were elevated above control values after T R H stimulation up to 5 hours before delivery. Significant increases of serum concentrations of T3 were observed 2 hours after birth in all but one control infant and in all neonates in the T R H group (Table II), No postnatal changes of serum T4 were observed 2 hours after birth in either the control group or the T R H group,
DISCUSSION Pregnancy has a marked influence on the maternal pituitary-thyroid axis. Throughout gestation, total serum T3 and T4 concentrations increase, but after 20 weeks of gestation the maternal thyrotropin concentration remains unchanged.~3 The thyrotropin response to T R H does not seem to be modified by pregnancy, but the prolactin response is relatively lower, t8 In our previous study, T R H was administered to pregnant women at term whose babies were delivered by elective cesarean section without labor. 4 Significant increases of thyrotropin and prolactin were observed, but no changes in serum T3 levels occurred up to 2 hours after the T R H injection, In the present study, we evaluated the response to T R H in a group of women with preterm
Volume 119 Number 6
Effect of prenatal T R H on fetal pituitary-thyroid axis
969
CORD BLOOD T ( nrnol/L )
3
2 -
0 1.5 ~r
<>
<> 1~
r
0
0
0
0
0.5
O
--
0
I
~ _ _ ~
1
I
I
2
]
3
I
I _ _ L
4
I
l
5
6
INTERVAL B E T W E E N TRH AND BIRTH ( hours ) Fig. 2. Individual cord-blood values of T3 (0) from fetuses exposed to TRH versus interval between maternal TRH injection and birth. Mean values (_+2 SD) for control fetuses are plotted for comparison.
pregnancies, of whom all but one were in labor. A significant increase in the maternal thyrotropin level was also seen after TRH administration. Comparison of the magnitude of the maternal thyrotropin response between preterm and term pregnancies is not possible in this study because we determined thyrotropin values only at delivery, which occurred at a variable interval of time after TRH administration. In accordance with previous observations by us and other investigators, no changes in maternal serum T3 concentrations were seen after TRH administration. 4' 19 It is possible that the maternal T3 response to thyrotropin is decreased during the second half of pregnancy and during labor, when baseline T3 values have risen above those of nonpregnant women. 19, 20 The placenta is impermeable to thyrotropin; during the second and third trimesters of gestation, the fetal thyrotropin concentration is higher than the maternal values, as confirmed in this study. 2~ Cord blood values of thyrotropin in these preterm fetuses were somewhat higher than those which we reported previously for term fetuses (8.4 _+ 0.9 vs 4.8 _~_-1.0 mU/L). 4 Others also have reported higher fetal thyrotropin values at about 30 weeks of gestation than at term. 2l Furthermore, higher cord blood thyrotropin concentrations in infants born vaginally, in comparison with concentrations in those delivered by cesarean section, have been noted, z2 Thus the stimulus of labor and vaginal delivery might have contributed to a slightly higher cord blood
level of thyrotropin in our control group. In contrast, the fetal T3 values were much lower than the maternal T3 values among the control group. The human placenta is also relatively impermeable to T3 and T4, and maternal-fetal gradients of these hormones have been documented repeatedly.4, 2l, 23 The lower fetal levels of thyroid hormone can be partially explained by active conversion of T4 tO reverse T3, and of T3 to 3,3'-diiodothyronine by the placenta. 13,24 During intrauterine life there is a progressive rise in the fetal T4 level from about mid gestation, followed by an increase in the fetal T3 level a few weeks later. 21 Our findings are in general agreement with the reported increases of T3 and T4 levels during fetal development. 23, 25 Maternal administration of TRH before delivery resulted in elevations of serum thyrotropin, T3, T4, and prolactin levels in these preterm fetuses. The cord blood changes of thyrotropin and T3 observed 2 hours after maternal administration of TRH are relatively similar to those we reported in term fetuses exposed to TRH for that same interval. 4 The postnatal thyrotropin response to TRH has also been shown to be of the same magnitude in preterm and term infants.14, 16 Healthy preterm infants have postnatal changes in thyrotropin, T3, and T4 levels that are quantitatively smaller than those observed in term infants. 21, 26, 27 Furthermore, among sick premature infants with RDS the early rise of the T 4 level may not be observed; however, T3 elevations do oc-
970
Moya et al.
cur. 27 Even though the cord blood and 2-hour values of T3 were lower among control infants, the relative change in the T3 value postnatally was similar in both groups of preterm infants. No significant changes in the T4 value were seen in either group at 2 hours of age. However, most of the postnatal changes in T4 values in term and preterm infants are observed around 24 hours after birth. 15, 21, z7 Although in this study we did not determine the postnatal changes in thyrotropin values, it appears that prenatal exposure to a single dose of TRH does not interfere with normal postnatal changes in the pituitary-thyroid axis in preterm or term infants.4 However, repeated administration of TRH is associated with decreases in the thyrotropin response and increased concentrations of thyroid hormones in human beingsY, 29 Recent studies of human subjects suggest that the combined use of corticosteroids and TRH is more beneficial in preventing and ameliorating RDS and its consequences than the use of corticosteroids alone) 2, 3o Although combination hormonal therapy may seem an attractive new therapeutic modality, several questions remain. Perhaps one of the most important is the significanceof transient elevations of fetal thyroid hormone levels brought about by TRH stimulation during crucial periods of brain development. The in utero increases of thyroid hormones as a result of maternal TRH administration are in the range of the physiologic changes observed after birth in preterm and term neonates.21 This fact and the lack of long-term adverse effects among infants exposed to intraamniotic injection of a large amount of T4 suggest that prenatal use of TRH to enhance fetal lung maturation is relatively safe) 1 We thank Dr. B. J. Green (Abbott Laboratories, North Chicago, Ill.) for supplying the TRH preparation, and the nurses and midwives of the Hospital Sotero del Rio, Puente Alto, Chile, for their assistance during the study. REFERENCES
1. Roti E, Gnudi A, Braverman L, et al. Human cord blood concentrations of thyrotropin, thyroglobulin, and iodothyronines after maternal administration of thyrotropin-releasing hormone. J Clin Endocr Metab 198l;53:813-7. 2. MelmedS, Harada A, Murata Y, et al. Fetal response to thyrotropin-releasing hormone after thyroid hormone administration to the rhesus monkey: lack of pituitary suppression. Endocrinology 1979;105:334-41. 3. Devaskar U, Nitta K, Szewczyk K, Sadig F, deMello D. TranspiacentaI stimulation of functional and morphologicfetal rabbit lung maturation: effect of thyrotropin-releasinghormone. Am J Obstet Gyneeol 1987;157:460-4. 4. Moya F, Mena P, Heusser F, et al. Response of the maternal, fetal, and neonatal pituitary-thyroid axis to thyrotropinreleasing hormone. Pediatr Res 1986;20:982-6. 5. Rooney S, Marino P, Gobran L, Gross I, Warshaw J. Thyrotropin-releasinghormoneincreasesthe amount of surfactant in lung lavage from fetal rabbits. Pediatr Res 1979;13:623-5.
The Journal of Pediatrics December 1991
6. Ikegami M, Jobe AH, Petenazzo A, Seidner SR, Berry DB, Ruffini L. Effects of maternal treatment with corticosteroids, T3, TRH, and their combinationson lung function of ventilated preterm rabbits with and without surfactant treatments. Am Roy Respir Dis 1987;136:892-8. 7. Warburton D, Parton L, BuckleyS, Cosico L, Enns G, Saluna T. Combined effects of corticosteroid, thyroid hormones, and ~-agonist on surfactant, pulmonary mechanics,and/~-receptor binding in fetal lamb lung. Pediatr Res 1988;24:166-70. 8. Oulton M, Rasmusson MG, Yoon RY, Fraser M. Gestationdependent effects of the combinedtreatment of glacocorticoids and thyrotropin-releasing hormone on surfactant production by fetal rabbit lung. Am J Obstet Gynecol 1989;160:961-7. 9. Tabor B, Ikegami M, Jobe AH, Yamada T, Oetomo SB. Dose response of thyrotropin-releasinghormone on pulmonary maturation in corticosteroid-treatedpreterm rabbits. Am J Obstet Gynecol 1990;163:669-76. 10. Umans JG, Umans HR, Szeto HH. Effects of thyrotropin-releasing hormone in the fetal lamb. Am J Obstet Gynecol 1986;155:1266-71. 11. Moya FR, Gross I. Prevention of respiratory distress syndrome. Semin Perinatol 1988;12:348-58. 12. Morales WJ, O'Brien WF, Angel JL, Knuppel RA, Sawai S. Fetal lung maturation: the combineduse of corticosteroids and thyrotropin-releasing hormone. Obstet Gynecol 1989;73: 111-6. 13. Fisher DA. Maternal-fetal thyroid function in pregnancy. Clin PerinatoI I983;i0:615-26. 14. Jacobsen BB, Andersen H, Dige-Petersen H, Hummer L. Pituitary-thyroid responsiveness to thyrotropin-releasing hormone in preterm and small-for-gestationalage newborns.Acta Paediatr Scand 1977;66:541-8. 15. Cuestas RA. Thyroid function in healthy premature infants. J PEDIATR1978;92:963-7. 16. Jacobsen BB, Andersen H, Dige-Petersen H, Hummer L. Thyrotropin responseto thyrotropin-releasinghormone in full term, euthyroid and hypothyroid newborns. Aeta Paediatr Scand 1975;65:433-8. 17. Penny R, Simms ME, Campbell WG, Spencer CA, Nicoloff JT. Thyroid indices in arterial and venous cord blood: significantly greater levels of reverse triiodothyronine in venous blood than in arterial blood. Metabolism 1986;35:645-8. 18. Yl~korkalaO, Kivinen S, Reinila M. Serial prolactin and thyretropin responses to thyrotropin-releasing hormone throughout normal human pregnancy. J Clin Endocrinol Metab 1979; 48:288-92. 19. Miyamoto J. Prolactin and thyrotropin responses to thyrotropin-releasing hormone during the peripartal period. Obstet Gynecol 1984;63:639-44. 20. Harada A, Hershman JM, Reed AW, et al. Comparison of thyroid stimulators and thyroid hormone concentrationsin the sera of pregnant women. J Clin Endocrinol Metab 1979;48: 793-7. 21. Fisher DA, Klein AH. Thyroid developmentand disorders of thyroid function in the newborn.N Engl J Med 1981;304:70212. 22. Lao TT, Panesar NS. Neonatal thyrotropin and mode of delivery. Br J Obstet Gynecol 1989;96:1224-7. 23. Fisher DA, Dussault JH, Sack J, Chopra IJ. Ontogenesis of hypothalamic-pituitary-thyroid function and metabolism in man, sheep, and rat. Recent Prog Horm Res 1977;33:59-116. 24. Roti E, Fang S, Green K, Emerson C, Braverman L. Human placenta is an active site of thyroxine and 3,3' ,5-triiodothyro-
Volume 119 Number 6
25.
26.
27.
28.
Effect o f prenatal T R H on fetal pituitary-thyroid axis
nine tyrosyl ring deiodination. J Clin Endocrinol Metab 198 I; 53:498-501. Thorpe-BeestonJG, Nicolaides KH, Felton CV, Butler J, McGregor AM. Maturation of the secretion of thyroid hormone and thyroid-stimulating hormone in the fetus. N Engl J Med 1991;324:532-6. Abbassi V, Merchant K, Abramson D. Postnatal triiodothyronine concentrations in healthy preterm infants and in infants with respiratory distress syndrome. Pediatr Res 1977;11: 802-4. Klein AH, Foley B, Kenney FM, Fisher DA. Thyroid hormone and thyrotropin responses to parturition in premature infants with and without the respiratory distress syndrome. Pediatrics 1979;63:380-5. Snyder PJ, Utiger RD. Repetitive administration of thyrotro-
971
pin-releasing hormone results in small elevations of serum thyroid hormones and in marked inhibition of thyrotropin response. J Clin Invest 1973;52:2305-12. 29. Staub JJ, Girard J, Mueller-Brand J, et al. Blunting of TSH response after repeated oral administration of TRH in normal and hypothyroid subjects. J Clin Endocrinol Metab 1978;46: 260-6. 30. Ballard RA, Ballard PL, Creasy R, Gross I, Padbury JP. Prenatal thyrotropin-releasing hormone plus corticosteroid decreases chronic lung disease in very low birth weight infants [Abstract]. Clin Res 1990;38:192A. 31. Barkai G, Zarfin Y, Ben-Harari M, Reichman B, Sack J, Mashiaeh S. In utero thyroxine therapy for the induction of fetal lung maturity: long-term effects. J Perinat Med 1988; 16:145-8.
Clinical and laboratory observations Cerebrospinal fluid examination in symptom-free infants with risk factors for infection Susan Fielkow, MD, Susan Reuter, RN, a n d Samuel P. G o t o f f , MD From the Department of Pediatrics, Rush Presbyterian-St. Luke's Medical Center, Chicago, Illinois
Cerebrospinal fluid analysis is frequently performed on newborn infants as part of the evaluation for suspected bacterial infection. It is important to identify newborn infants with meningitis, but the number with meningitis is less than 3% of those evaluated, and lumbar puncture is associated with some risk, as well as costJ' a In addition, there is a high rate of unsuccessful or traumatic lumbar punctures.i~3 Increased attack rates of early-onset group B streptococcal infections4 and early-onset infections in general5 have been clearly related to prolonged rupture of membranes and chorioamnionitis. Attempts to identify infants with infec-
tions as early as possible has led to the practice of a "sepsis workup" in symptom-free infants, based on the presence of maternal risk factors such as chorioamnionitis. Although bacteremia has been documented in symptom-free infants,6, 7 which may justify a blood culture, we questioned the benefit of a CSF examination in addition to blood culture in symptom-free neonates and performed a retrospective chart review to answer the question.
See related article, p. 973.
I Presented at the meeting of the Midwest Society for Pediatric Research, Chicago, IlL, Nov. 2, 1990. Submitted for publication April 29, 1991; accepted June 17, 1991. Reprint requests: Samuel P. Gotoff, MD, Department of Pediatrics, Rush Presbyterian-St. Luke's Medical Center, 1653 W. Congress Pkwy., Chicago, IL 60612. 9/24/31835
CSF
Cerebrospinalfluid
METHODS The clinical and laboratory records accumulated during the 10-year period from 1978 to 1987 on all newborn infants from whom cerebrospinal fluid culture specimens were taken within the first week of life were reviewed. The infants were either inborn or transferred to Rush Presbyterian-St.