Sufficient provision of thyroid hormone in the first trimester of pregnancy is essential for normal fetal brain development.1–3 There is growing evidence, however, suggesting that maternal thyroid hormone levels remain important until term.4–6 In the debate on benefits of screening for hypothyroidism in pregnancy,7,8 the question on the optimal free T4 (fT4) level remains unanswered,9 because reports on thyroid function in normal pregnancy are scarce.10,11
Preeclamptic patients are at particular risk. Several articles describe an association between preeclampsia and maternal thyroid dysfunction12–16 and low birth weight;17 some authors even suggested maternal thyroid function abnormalities to be a causal factor.18–21
Free T4 is generally found to be lower in umbilical cord samples from neonates born from preeclamptic pregnancies than in infants from normotensive pregnancies, but reports remain inconclusive.13,14,17,22–26 A low fT4 can be a consequence of lower maternal thyroid hormone levels; however, it can also be caused by impaired transfer of T4 due to placental insufficiency or fetal disease due to fetal growth restriction (FGR) and fetal acidosis.
The aim of the present study was to assess whether and to what extent thyroid function is impaired in women with severe and early hypertensive disorders in pregnancy, whether autoimmunity is involved, and to what extent neonatal thyroid function is affected.
PATIENTS AND METHODS
This prospective cohort study was performed between April 1, 2001, and June 1, 2003, in a subset of women with severe hypertensive disorders of pregnancy, participating in the Preeclampsia Eclampsia Trial of Amsterdam.27 This randomized clinical intervention trial in women with severe hypertensive disorders of pregnancy was carried out in 2 tertiary care perinatal centers in Amsterdam from September 1, 2000 to June 1, 2003. In our study of thyroid function, only the subset of women and neonates admitted in 1 of the centers (Academic Medical Center) were included. Because blood collection for thyroid hormone parameters at admission preceded the study, intervention data from both treatment arms were joined for our further analyses.
The study protocol was approved by the institutional review board of the Academic Medical Center. After informed consent, women were included upon admission if they were in the 24th to 34th week of a singleton pregnancy, with either pregnancy-induced hypertension in combination with fetal growth restriction (diastolic blood pressure > 90 mm Hg and estimated fetal weight < 5th centile or abdominal circumference < 10th percentile) or severe preeclampsia (diastolic blood pressure > 110 mm Hg and proteinuria > 0.3 g/L) or hemolysis, elevated liver enzymes, and low platelet count (HELLP) syndrome (lactate dehydrogenase (LDH) > 600 U/L, aspartate aminotransferase > 70 U/L, platelet count < 100 × 109/L). After informed consent, patients were randomly allocated to a temporizing management strategy with or without plasma volume expansion. Randomization was preformed by use of a designated palmtop computer with random number generation software. In all patients, obstetric management aimed to improve fetal prognosis through increasing gestational age at birth. Pregnancy was prolonged until deterioration of fetal or maternal condition necessitated delivery.28 In the absence of normal reference values for thyroid hormone in pregnancy, apart from comparing thyroid function results with our local reference ranges for the specific assays, they were also compared with earlier described results in a prospectively followed-up group of 10 healthy women who became pregnant by artificial insemination because of male infertility.10 Baseline, nonpregnant thyroid function was also known in these women. None of these pregnant women showed any sign of hypertensive disorders.
At admission, maternal blood samples were collected for determination of T4, T3, thyroid-stimulating hormone (TSH), freeT4(fT4), thyroxine-binding globulin (TBG), reverse-T3 (rT3) and thyroid peroxidase antibodies (TPO ab). Normal values of fT4 in pregnancies are not known.29,30 Therefore, we defined low maternal thyroxin levels by the lower limit of our laboratory (ie, nonpregnant) reference range: fT4 of < 9 pM.
A second sample was taken if fT4 was < 9 pM or TSH was < 0.4 or > 4 mU/L. If fT4 or TSH in the second sample were also outside these limits, the patient was referred to an endocrinologist for further evaluation.
At birth, an umbilical cord blood sample was collected for determination of T4, T3, TSH, fT4, TBG, and rT3. If the amount of cord blood after determination of arterial pH was limited and thyroid hormone analysis was incomplete, preference was given to determining fT4, TSH, T4, and T3. We compared results with data previously collected in our hospital using the same assays in a group of 114 premature infants aged younger than 30 weeks gestational age (mean gestational age 28 1/7 weeks ± 8 days), of whom 90% were appropriate for gestational age.31
Three months after term date, at a scheduled visit, a maternal postpartum blood sample was obtained for determination of T4, T3, TSH, fT4, TBG, rT3 and TPO antibodies. Values were compared with reference values of the laboratory. If 1 of these determinations was outside the reference range, the patient was further evaluated by an endocrinologist for thyroid disorders.
Blood was centrifuged immediately and stored at –20°C until assay. Levels of T4, T3 and rT3 were measured by in-house radioimmunoassay methods.32 Detection limits were 5.0, 0.3, and 0.03 nM, respectively; intra-assay coefficients of variation were 2–4%, 3–4% and 4–5%, respectively; and interassay coefficients of variation were 3–6%, 7–8% and 5–9%, respectively. Free T4 was measured by time-resolved fluoroimmunoassay (Delfia fT4, Wallac Oy, Turku, Finland). Detection limit was 2 pM, intra-assay coefficient of variation was 4–6%, and interassay coefficient of variation was 5–8%. TSH was also measured by time-resolved fluoroimmunoassay (Delfia hTSH Ultra, Wallac Oy, Turku, Finland). Detection limit was 0.01 mU/L, intra-assay coefficient of variation was 1–2%, and interassay coefficient of variation was 3–4%. Anti-TPO antibodies were measured by chemiluminescence immunoassay (LUMI-test anti-TPO, BRAHMS, Berlin, Germany). Detection limit was 30 kU/L, intra-assay and interassay coefficients of variation were 3–7% and 8–12%, respectively. Thyroxin binding globulin was determined by commercial radioimmunoassay (Eiken Chemical Co, Tokyo, Japan). Detection limit was 30 nM, intra-assay and interassay coefficients of variation were 2–4% and 4–6%, respectively.
We used 3 measures of severity of placental insufficiency. Ultrasound Doppler examination of a free loop of umbilical artery was performed twice weekly, and the most recent pulsatility index, calculated from the flow velocity profile, was recorded. Second, we used birth weight ratio as a measure for dysmaturity. The birth weight ratio is the observed birth weight divided by the expected weight at the corresponding gestational age according to the customized antenatal growth chart.33 By definition, an infant with the appropriate weight for gestational age is to have a birth weight ratio of 1, whereas a birth weight ratio less than 0.88 corresponds to a birth weight less than the tenth percentile (SGA). The net weight of the placenta was measured after birth after removal of cord and membranes; centile values were calculated by means of Dutch reference curves, stratified for parity and gender of the neonate.34
Data were analyzed with the statistical program SPSS 10.0.7 for Windows (SPSS Inc., Chicago, IL). Patient characteristics as well as hormone values were checked for normal distribution as indicated by 1-sample Kolmogorov-Smirnov test. All analyses were done for the whole group and within subgroups according to low fT4 or admission diagnosis. Groups were compared using the Student t test and χ2. One-way analysis of variance was used to compare groups according to admission diagnosis. Linear regression analysis and Pearson correlation were used in correlations of continuous variables. Multivariate regression analysis was performed by the enter model, with the significant factors of the univariate analyses combined with gestational age and maternal fT4. The statistical power to detect a clinically significant difference in maternal fT4 level between our study group and the comparison group of 1.5 pM was calculated as 80% (effect size = 0.84; n = 10 compared with n = 80; α = 0.05).
In the study period, 80 women were included. Table 1 shows the demographic and obstetric data at admission. There were 69 liveborn babies and 11 stillbirths. Median gestational age at birth was 30 6/7 weeks (range 26 1/7–36 6/7 weeks), birth weights ranged from 525 g to 2,310 g (median 1,100) and birth weight ratio ranged from 0.36 to 0.88 (median 0.63); all neonates but 1 were SGA. Median umbilical cord artery pH was 7.20, ranging from 6.94 to 7.44.
Table 2 shows mean maternal thyroid hormone levels at admission. There were no statistically significant differences between the study group and the comparison group, although thrombin time4 (TT4), fT4, T3, and TBG were lower than in the comparison group, in combination with a higher TSH. However, in 26 patients (33%) fT4 was less than 9 pM, their mean fT4 was 8.0 (± 0.66) pM, ranging from 6.8 to 8.9 pM. In the comparison group, 2 patients (20%) had fT4 levels less than 9 pM. On reassessment 1 to 3 weeks later, a mean fT4 of 10.5 (± 2.8) and a TSH of 3.9 (± 2.8) were found. Only 1 patient developed a specific thyroid disorder (see postpartum section). When we compared clinical characteristics on admission (systolic and diastolic blood pressure, admission diagnosis, ultrasound Doppler pulsatility index of the umbilical artery, highest level of proteinuria, lowest platelet count during observation, gestational age at delivery, interval admission-delivery, birth weight, birth weight ratio, Apgar score at 5 minutes, placenta weight centile) of women with low fT4 to women with normal fT4 values, no statistically significant differences were found. Levels of TPO antibodies were elevated in 7 patients at admission, but there was no significant relationship between the presence of TPO antibodies and fT4 below 9 pM (P = .22, χ2 = 1.5, df = 1). In univariate regression analyses, T3 and T4 levels had significant positive correlations with TBG levels, which is expected, because T3 and T4 are predominantly bound to TBG. Concentrations of this binding protein, subsequently, were significantly lower in women with a higher quantity of proteinuria (Pearson r = –0.33, P = .006), suggesting that lower T4 and T3 concentrations are due to loss of binding protein.
Postpartum maternal blood samples, taken 14 to 27 (mean 22) weeks after delivery were obtained in all patients but 1, who was lost to follow-up. Table 3 lists the results of maternal thyroid hormone levels at 3 months postterm date. TPO antibodies were elevated in 8 patients (10%). There were 2 patients (2.5%) with abnormal thyroid hormone parameters. One patient, with a low fT4 level during hospitalization, was diagnosed with Graves’ hyperthyroidism. The second patient had suppressed TSH (0.08 mU/L), elevated TT3 (2.75 nM), and normal T4 (150 nM) and fT4 (15.7 pM), suggesting T3 hyperthyroidism. All other patients with a prior low fT4 now had normal thyroid function.
Umbilical cord blood thyroid hormones were assessed in 46 (67%) liveborn neonates. Results are shown in Table 4. Because gestational age in the study group was 3 weeks older than in the comparison group, higher concentrations of T4, fT4, T3, and TBG were anticipated in the study group. Free T4, however, was significantly lower in this group than in cord blood of the comparison group. There was no correlation between maternal and cord blood fT4, as shown in Figure 1 (Pearson r = 0.17, P = .27). Contrary to normal pregnancies, in these growth restricted neonates, there was no relation between fT4 levels and gestational age (Figure 2; Pearson r = 0.18, P = .22). On further univariate testing of perinatal factors, only umbilical artery pH (Pearson r = 0.48, P = .001) and gender (Pearson r = –0.33, P = .03) were significant determinants, and gestational age, maternal fT4, Doppler pulsatility index, birth weight ratio, placenta weight centile, and treatment by plasma volume expansion were not. Multivariate linear regression analysis showed that umbilical cord fT4 was significantly dependent on umbilical artery pH and gender and was only slightly influenced by gestational age and maternal fT4.
This observational study of maternal and neonatal thyroid function was carried out in women referred to a tertiary care center because of early and severe hypertensive disorders of pregnancy. Although we did not find statistically significant differences with thyroid hormone levels of a group of 10 healthy women, TT4, fT4, and T3 were somewhat lower, consistent with the literature.29 Moreover, 33% of patients had fT4 concentrations below the lower limit of our local reference range of 9 pM. These women had no identifiable maternal disorder or specific clinical course. In contrast to some authors,14–16,23 we found total T4 and T3 to be of limited clinical value in assessing thyroid function in preeclampsia, because they reflect low TBG due to proteinuria. The observed fT4 levels spontaneously changed to normal at reassessment during pregnancy and more so 3 months after term at the scheduled postpartum visit.
In the light of the present discussion on the necessity for screening thyroid function in pregnancy, it is therefore pivotal that the normal lower limit of fT4 levels is identified, especially because no pathophysiologic pathway has been determined to explain the observed fT4 values in our study group. Notably, 2 subjects (20%) in the comparison group of healthy women had a third trimester fT4 level below 9 pM.
We anticipated a high prevalence of thyroid disorders and thyroid autoimmunity in the women in our study. However, 3 months postterm, specific thyroid abnormalities were diagnosed in only 2 women (2.5%), which is the normal prevalence of postpartum thyroid disease in the nonpreeclamptic population.35
In this study, umbilical fT4 levels in the neonates were lower than in the comparison group, a finding in concordance with literature.13,14,17,22,26 This low umbilical fT4 was not related to maternal fT4, therefore it is not likely to result from decreased maternal supply of fT4 or impaired transfer. According to our data, these low fT4 levels are due to prenatal acidosis as a result of utero placental insufficiency.
In the present study we were not able to investigate the duration of low fT4 levels in utero. In a cordocentesis study, high fetal TSH and low fT4 levels were found to be correlated to PO2 levels in FGR fetuses without signs of fetal distress, suggesting slowly advancing chronic hypothyroxinemia.36 The present study confirms the general concern about adequate fT4 supply in FGR fetuses. It raises the question of whether low fT4 is just a derivative of intrauterine malnutrition. It could well be an independent cause of impaired brain development and the observed impairment of neuropsychological development in infants who were born growth restricted.37,38
These data stress the importance of an adequate follow-up of growth-restricted preterm infants, because they are at risk of hypothyroxinemia, and follow-up of preterm neonates shows a high prevalence of developmental disorders, especially after low postnatal thyroid hormone levels.39
In summary, we have demonstrated that transient hypothyroxinemia is common in women with severe hypertensive disorders, but it is not associated with an increased incidence of thyroid disorders. The neonates show low fT4 and TSH levels, unrelated to maternal fT4 levels. These lower fT4 levels are most likely caused by fetal acidosis. Follow-up is in progress and will reveal whether developmental outcome is associated with low fT4 levels of mother, low fT4 levels of the neonate, or merely with the deleterious consequences of fetal growth restriction itself.
1. Haddow JE, Palomaki GE, Allan WC, Williams JR, Knight GJ, Gagnon J et al. Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child. N Engl J Med 1999;341:549–55.
2. Calvo RM, Jauniaux E, Gulbis B, Asuncion M, Gervy C, Contempre B, et al. Fetal tissues are exposed to biologically relevant free thyroxine concentrations during early phases of development. J Clin Endocrinol Metab 2002;87:1768–77.
3. Pop VJ, Brouwers EP, Vader HL, Vulsma T, van Baar AL, de Vijlder JJ. Maternal hypothyroxinaemia during early pregnancy and subsequent child development: a 3-year follow-up study. Clin Endocrinol (Oxf) 2003;59:282–8.
4. Morreale de Escobar G, Obregon MJ, Escobar del Rey F. Is neuropsychological development related to maternal hypothyroidism or to maternal hypothyroxinemia? J Clin Endocrinol Metab 2000;85:3975–87.
5. Blazer S, Moreh-Waterman Y, Miller-Lotan R, Tamir A, Hochberg Z. Maternal hypothyroidism may affect fetal growth and neonatal thyroid function. Obstet Gynecol 2003;102: 232–41.
6. Smallridge RC, Ladenson PW. Hypothyroidism in pregnancy: consequences to neonatal health. J Clin Endocrinol Metab 2001;86:2349–53.
7. Spong CY. Subclinical hypothyroidism: should all pregnant women be screened? Obstet Gynecol 2005;105:235–6.
8. Casey BM, Dashe JS, Wells CE, McIntire DD, Byrd W, Leveno KJ et al. Subclinical hypothyroidism and pregnancy outcomes. Obstet Gynecol 2005;105:239–45.
9. Pop VJ, van Baar AL, Vulsma T. Should all pregnant women be screened for hypothyroidism? Lancet 1999;354:1224–5.
10. Berghout A, Endert E, Ross A, Hogerzeil HV, Smits NJ, Wiersinga WM. Thyroid function and thyroid size in normal pregnant women living in an iodine replete area. Clin Endocrinol (Oxf) 1994;41:375–9.
11. Glinoer D. The regulation of thyroid function in pregnancy: pathways of endocrine adaptation from physiology to pathology. Endocr Rev 1997;18:404–33.
12. Osathanondh R, Tulchinsky D, Chopra IJ. Total and free thyroxine and triiodothyronine in normal and complicated pregnancy. J Clin Endocrinol Metab 1976;42:98–104.
13. Fetter WP, Waals-Van de Wal CM, Van Eyck J, Samson G, Bongers-Schokking JJ. Thyroid hormone concentrations in preterm infants born to pre-eclamptic women with placental insufficiency. Acta Paediatr 1998;87:186–90.
14. Lao TT, Chin RK, Swaminathan R, Lam YM. Maternal thyroid hormones and outcome of pre-eclamptic pregnancies. Br J Obstet Gynaecol 1990;97:71–4.
15. Basbug M, Aygen E, Tayyar M, Tutus A, Kaya E, Oktem O. Correlation between maternal thyroid function tests and endothelin in preeclampsia-eclampsia. Obstet Gynecol 1999; 94: 551–5.
16. Qublan HS, Al Kaisi IJ, Hindawi IM, Hiasat MS, Awamleh I, Hamaideh AH, et al. Severe pre-eclampsia and maternal thyroid function. J Obstet Gynaecol 2003;23:244–6.
17. Kaya E, Sahin Y, Ozkececi Z, Pasaoglu H. Relation between birth weight and thyroid function in preeclampsia-eclampsia. Gynecol Obstet Invest 1994;37:30–3.
18. Davis LE, Leveno KJ, Cunningham FG. Hypothyroidism complicating pregnancy. Obstet Gynecol 1988;72:108–12.
19. Leung AS, Millar LK, Koonings PP, Montoro M, Mestman JH. Perinatal outcome in hypothyroid pregnancies. Obstet Gynecol 1993;81:349–53.
20. Millar LK, Wing DA, Leung AS, Koonings PP, Montoro MN, Mestman JH. Low birth weight and preeclampsia in pregnancies complicated by hyperthyroidism. Obstet Gynecol 1994;84:946–9.
21. Mecacci F, Parretti E, Cioni R, Lucchetti R, Magrini A, La Torre P. et al. Thyroid autoimmunity and its association with non-organ-specific antibodies and subclinical alterations of thyroid function in women with a history of pregnancy loss or preeclampsia. J Reprod Immunol 2000;46:39–50.
22. Narin N, Kurtoglu S, Basbug M, Caksen H, Kafali M, Durak AC, et al. Thyroid function tests in the newborn infants of preeclamptic women. J Pediatr Endocrinol Metab 1999;12:69–73.
23. Martin CR, Van Marter LJ, Allred EN, Leviton A. Growth-restricted premature infants are at increased risk for low thyroxine. Early Hum Dev 2001;64:119–28.
24. Gemer O, Shenhav S, Segal S, Tur-Kaspa I. Thyroid hormone levels in cord blood of infants with acidemia at birth. Eur J Obstet Gynecol Reprod Biol 2000;93:53–5.
25. Pereira DN, Procianoy RS. Effect of perinatal asphyxia on thyroid-stimulating hormone and thyroid hormone levels. Acta Paediatr 2003;92:339–45.
26. Belet N, Imdat H, Yanik F, Kucukoduk S. Thyroid function tests in preterm infants born to preeclamptic mothers with placental insufficiency. J Pediatr Endocrinol Metab 2003;16:1131–5.
27. Ganzevoort W, Rep A, Bonsel GJ, De Vries JI, Wolf H. A randomized trial of plasma volume expansion in hypertensive disorders of pregnancy: influence on the pulsatility indices of the fetal umbilical artery and middle cerebral artery. Am J Obstet Gynecol 2005;192:233–9.
28. Schiff E, Friedman SA, Sibai BM. Conservative management of severe preeclampsia remote from term. Obstet Gynecol 1994;84:626–30.
29. Glinoer D. The regulation of thyroid function during normal pregnancy: importance of the iodine nutrition status. Best Pract Res Clin Endocrinol Metab 2004;18:133–52.
30. Pop VJ, Vulsma T. Maternal hypothyroxinaemia during (early) gestation. Lancet 2005;365:1604–6.
31. Van Wassenaer AG, Kok JH, Dekker FW, de Vijlder JJ. Thyroid function in very preterm infants: influences of gestational age and disease. Pediatr Res 1997;42:604–9.
32. Wiersinga WM, Chopra IJ. Radioimmunoassay of thyroxine (T4), 3,5,3′-triiodothyronine (T3), 3,3′,5′-triiodothyronine (reverse T3, rT3), and 3,3′-diiodothyronine (T2). Methods Enzymol 1982;84:272–303.
33. Gardosi J, Chang A, Kalyan B, Sahota D, Symonds EM. Customised antenatal growth charts. Lancet 1992;339:283–7.
34. Arts NFTh.De pathologische zwangerschap. In: Galjaard H, Kloosterman GJ, Treffers PE, editors. De voortplanting van de mens. Leerboek voor obstetrie en gynaecologie. 7th ed. Weesp (the Netherlands): Centen; 1985. p. 340–1.
35. Pearce EN, Farwell AP, Braverman LE. Thyroiditis [published erratum appears in N Engl J Med 2003;349:620]. N Engl J Med 2003;348:2646–55.
36. Thorpe-Beeston JG, Nicolaides KH, Snijders RJ, Felton CV, McGregor AM. Thyroid function in small for gestational age fetuses. Obstet Gynecol 1991;77:701–6.
37. Cheng SW, Chou HC, Tsou KI, Fang LJ, Tsao PN. Delivery before 32 weeks of gestation for maternal pre-eclampsia: neonatal outcome and 2-year developmental outcome. Early Hum Dev 2004;76:39–46.
38. Many A, Fattal A, Leitner Y, Kupferminc MJ, Harel S, Jaffa A. Neurodevelopmental and cognitive assessment of children born growth restricted to mothers with and without preeclampsia. Hypertens Pregnancy 2003;22:25–9.
39. Reuss ML, Paneth N, Pinto-Martin JA, Lorenz JM, Susser M. The relation of transient hypothyroxinemia in preterm infants to neurologic development at two years of age. N Engl J Med 1996;334:821–7.