A number of recent population studies suggest that pregnancies in women with thyroid gland disorders may have an increased rate of pregnancy complications, including preterm delivery, placental abruption, and abnormal neuropsychologic development in offspring.1–6 Overt and subclinical hypothyroidism are the two disorders most commonly associated with adverse outcomes, and together, they are thought to represent a continuum of autoimmune thyroiditis. This often is preceded by the detection of serum autoantibodies directed against thyroid cell constituents. Although many types of antithyroid antibodies have been described, the most common is any of a group of antibodies directed against various parts of the thyroid peroxidase molecule. The prevalence of thyroid peroxidase antibodies is increased almost 10-fold in women compared with men, it increases with age, and it has been reported in 5–10% of pregnant women.2,7 Although most pregnant women with abnormally elevated serum levels of thyroid peroxidase antibodies are clinically and chemically euthyroid, they may be at increased risk for spontaneous abortion and preterm delivery.8–14 Importantly, postpartum thyroiditis in such women is common, and a substantive proportion develops overt hypothyroidism during ensuing years.
Because of the link of thyroid autoantibodies with continuing progression to incipient and then overt thyroid failure, it seems reasonable that such antibodies might be a surrogate marker for thyroid-related adverse pregnancy outcomes. Indeed, this possibility has led some to suggest that routine prenatal screening for antithyroid antibodies, coupled with antepartum thyroxine supplementation for affected women, might even prevent some of these adverse perinatal occurrences.13,15–17 With this background, we designed the present study to further elucidate any association of thyroid peroxidase antibodies with an increased incidence of adverse pregnancy outcomes to include hypertensive disorders, preterm premature rupture of membranes, preterm delivery, and placental abruption. To accomplish this, we measured serum antithyroid peroxidase antibody concentrations in a previously described cohort of more than 17,000 women who presented for prenatal care during the first half of pregnancy and in whom we already had determined serum thyrotropin (TSH) and thyroxine (T4) concentrations.
MATERIALS AND METHODS
This study constitutes another in a series of investigations performed to elucidate thyroid physiology and pathophysiology in pregnant women. Excess serum obtained for routine prenatal rubella antibody testing was delivered to our laboratory from November 1, 2000, to April 14, 2003. For the initial study, and as described in detail in previous reports,1,18 serum TSH concentrations were measured and samples with abnormal values were reflexively assayed for free T4 concentration. Women with both abnormal TSH and free T4 levels were referred to an Obstetrical Complications Clinic for further evaluation and they were excluded from the study. Excess serum from samples was stored at −80°C for future analyses such as the one now reported.
Thus, for the present study, and as also previously detailed,2 serum samples were retrieved for those women who had been screened during the first 20 weeks of gestation and who had been delivered of a singleton newborn weighing 500 g or more. These serum aliquots were analyzed for thyroid peroxidase antibody concentrations using a chemiluminescent immunoassay (Immulite 2000 Analyzer; Diagnostic Products Corporation, Los Angeles, CA). The analytical sensitivity of this thyroid peroxidase assay was 5.0 international units/mL and its coefficient of variation was 9.8% within run and 11.3% between runs. Thyroid peroxidase antibody levels greater than 50 international units/mL were considered abnormal and these women were considered to be thyroid peroxidase-antibody positive.2,3
To further subcategorize the thyroid peroxidase antibody-positive and -negative cohorts, previously determined serum levels of TSH and free T4 were used. Gestational age-specific thresholds for the 2.5th and 97.5th percentiles were used to establish the normal range of serum TSH as 0.08–3.0 milliunits/L and free T4 as 0.86–1.9 ng/dL.2,19 Those women with a TSH level 3.0 milliunits/L or greater and a normal range free T4 were identified as having subclinical hypothyroidism. Women whose TSH level was within the normal range and whose serum free T4 was less than 0.86 ng/dL were identified to have isolated maternal hypothyroxinemia.
Perinatal outcomes in the cohort of thyroid peroxidase antibody-positive women were compared with the outcomes in the cohort of women who were thyroid peroxidase antibody-negative. Similarly, such comparisons were made for antibody-positive and -negative groups of women subcategorized into those with subclinical hypothyroidism compared with those who were chemically euthyroid as defined by normal serum concentrations of TSH and free T4. Obstetric and neonatal outcomes were ascertained by accessing the computerized perinatal database into which data are routinely entered for all women who are delivered at Parkland Hospital. Neonatal outcome data are abstracted from discharge records and entered into a separate database. Details and validity of these databases have been previously described.1 For the current study, results from serum thyroid hormone and antibody determinations were stored electronically and linked to the perinatal and neonatal databases. The protocol for this study has been approved by the University of Texas Southwestern Medical Center Institutional Review Board committee and found to be minimal risk research.
Pearson's χ2 and Student's t test were used for univariable two-group comparisons. Spearman correlations were used to evaluate the association between continuous measures. Logistic regression was applied to examine the significance for selected pregnancy outcomes adjusted for maternal age, race, parity, and weight. The Hosmer-Lemeshow statistic was used to examine the goodness of fit for the logistic regression model. Statistical computations were performed using SAS 9 (SAS Institute, Cary, NC). Two-sided P values <.05 were judged statistically significant.
During the study period, there were 17,298 women who met the criteria for registration during the first half of pregnancy, who underwent thyroid screening, and who were delivered of a singleton neonate. Their mean gestational age at screening was 11.9 weeks and ranged from 6 to 20 weeks of gestation. Outcome results were excluded in the 79 women for whom serum was unavailable for thyroid peroxidase antibody testing. In the remaining cohort, 1,012 (6%) women were considered to be thyroid peroxidase antibody-positive with a level greater than 50 international units/mL. Incidence of thyroid peroxidase antibody-positive status according to thyroid function categories is shown in Figure 1. Although the incidence of 5% (95% confidence interval [CI] 2–8%) in women with isolated maternal hypothyroxinemia was similar to the 4% incidence (95% CI 4–5%) in euthyroid women, it was, as expected, remarkably higher at 31% (95% CI 28–36%) in women with subclinical hypothyroidism. For reference purposes, also shown in Figure 1, is the 61% incidence (95% CI 52–70%) of positive thyroid peroxidase antibodies in a group of women with overt hypothyroidism whose outcomes are not included in this analysis.
Maternal characteristics of thyroid peroxidase antibody-positive and -negative women are compared in Table 1. Thyroid peroxidase antibody-positive women were older, heavier, and more often parous than thyroid peroxidase antibody-negative women. The incidence of thyroid peroxidase antibody was 2.6% in African-American (95% CI 1.9–3.4%), 6.1% in Hispanic (95% CI 5.8–6.6%) and 8.4% in white women (95% CI 5.8–11.8%). Thus, the racial composition of the thyroid peroxidase antibody-positive women is significantly different than those of the cohort with negative antibodies (Table 1).
Pregnancy outcomes in the thyroid peroxidase antibody-positive and -negative groups of women are compared in Table 2. The incidences of hypertensive disorders, diabetes, preterm birth, preterm premature rupture of membranes, and cesarean delivery rate were not significantly different either before or after adjustment for demographic differences. The one exception was the incidence of placental abruption, which was more than threefold higher in women who were thyroid peroxidase antibody-positive (odds ratio 3.4, 95% CI 1.7–6.7). Of the 10 women who were thyroid peroxidase antibody-positive and who had an abruption, their individual serum antibody levels ranged from 53 to 392 international units/mL. Seven of these same 10 women were euthyroid with a normal TSH and free T4 level. Neonatal outcomes for thyroid peroxidase-positive and -negative groups of women are shown in Table 3. The incidences of adverse outcomes were not significantly different between the two groups.
To negate any possible effects of thyroid dysfunction on pregnancy outcomes in women with thyroid peroxidase antibodies, we evaluated pertinent pregnancy outcomes in women who were euthyroid as defined by normal TSH and free T4 level. Specifically, we excluded those women with either subclinical hypothyroidism or isolated maternal hypothyroxinemia. Euthyroid women who were thyroid peroxidase antibody-positive had a fourfold increased incidence of placental abruption compared with similarly euthyroid women in whom thyroid peroxidase antibodies were negative (odds ratio 4.0, 95% CI 1.9–8.6). None of the other pregnancy outcomes, including preterm birth, was found to be associated with thyroid peroxidase antibody-positive status in euthyroid women.
Finally, because our previous observations showed an association between subclinical hypothyroidism and preterm delivery and placental abruption,1 we analyzed the data to determine if thyroid peroxidase antibody status in women with subclinical hypothyroidism would be informative. These results (Fig. 2) indicate that women with subclinical hypothyroidism who are also thyroid peroxidase antibody-positive were not at a further increased risk for adverse outcomes.
There are three interesting findings from this investigation evaluating a link between thyroid peroxidase antibodies and adverse pregnancy outcomes. First, pregnant women with thyroid peroxidase antibody levels of more than 50 international units/mL, whom we termed thyroid peroxidase antibody-positive, have a more than threefold increased risk of placental abruption when compared with women identified to be thyroid peroxidase antibody-negative. Second, we found no association between thyroid peroxidase antibody status and other adverse perinatal outcomes to include hypertensive disorders, fetal growth restriction, preterm prematurely ruptured membranes, and preterm delivery. Third, we found that there was not a greater predictive value for these adverse outcomes when results of thyroid peroxidase antibody concentrations were combined with those of serum TSH and free T4 levels in a cohort of women with subclinical hypothyroidism.
The finding of increased placental abruption in the group of thyroid peroxidase antibody-positive women comports with our previously reported finding of a threefold increased incidence of placental abruption in women with subclinical hypothyroidism.1 There are at least three other population studies that do not report such an association with placental abruption.4,18,20 In two of these studies, placental abruption outcomes were not specifically described. However, in the study of nearly 11,000 pregnant women enrolled in the First and Second Trimester Evaluation of Risk (FaSTER) trial, there was no increased rate of placental abruption in women with significant levels of antithyroid antibodies.4 It is noteworthy that the reported placental abruption incidence of 0.9% in these women is seemingly high and is similar to that of 1.0% that we now describe for thyroid peroxidase antibody-positive women. It is possible that these reported variances in the incidence of abruption may be at least partially explicable by the older population described in the FaSTER cohort.21
With the exception of placental abruption, there were no differences in a number of other adverse perinatal outcomes between the cohorts. Specifically, we found no increased incidences in the rates of preterm delivery, low-birth-weight neonates, or neonatal intensive care unit admissions between these two groups. By contrast, in a study of 4,500 Italian women, Negro et al14 reported a threefold increased preterm delivery rate in a group of thyroid peroxidase antibody-positive women compared with those found to be antibody-negative (22% compared with 8%, P<.01). Our findings are consistent with the FaSTER trial4 in which women with antithyroid antibodies were not found to have excessive preterm births. This was despite their finding of significantly increased incidence of preterm prematurely ruptured membranes in the antibody-positive group (3–4% compared with 1%, P<.001). Finally, a Finnish study of more than 5,800 women did not report a difference in preterm birth.20 They did, however, describe an increased perinatal mortality rate in women positive for thyroid peroxidase antibodies (2.4% compared with 0.8%, P<.05). It seems important to emphasize that four of the seven perinatal deaths in the thyroid peroxidase antibody-positive group were born less than 28 weeks of gestation. In summary, our findings are consistent with much of the literature of a lack of association with adverse pregnancy outcome and thyroid peroxidase antibody status.
Our third finding is that when thyroid peroxidase antibody measurements were considered along with the other thyroid-related analytes, there were no further clinical benefits. We have reported that women with subclinical hypothyroidism were found to have a twofold increased incidence of preterm delivery before 34 weeks of gestation as well as a threefold increased incidence of placental abruption when compared with euthyroid women.1 Because of the current finding of a fourfold increased rate of placental abruption in thyroid peroxidase antibody-positive euthyroid women, we analyzed thyroid peroxidase antibody status in the cohort of women with subclinical hypothyroidism. Surprisingly, consideration of thyroid peroxidase antibody status did not offer any added benefit to identify a subgroup of women with subclinical hypothyroidism at greater risk for adverse pregnancy outcomes previously described.
The findings now presented can be interpreted from both a biologic as well as a clinical viewpoint. From the biologic perspective, observations of increased placental abruption in thyroid peroxidase antibody-positive women may provide insight into the pathophysiology of abruption. Abnormally high levels of thyroid autoantibodies are a marker for maternal autoimmune or inflammatory disorders, which contribute to increased adverse perinatal outcomes. There is support for such a mechanism from other investigators who have described an association between thyroid antibodies and preterm prematurely ruptured membranes as well as preterm birth.4,14 We have also previously hypothesized that thyroid hormones are essential for placental implantation and development.1 Antithyroid antibodies are possibly markers for thyroid hormone deficiency that may be causally linked to placental abruption.
From the clinical viewpoint, findings now presented might lead some to suggest an application for universal prenatal antithyroid antibody screening. For screening to be beneficial, an intervention to mitigate adverse outcomes would be necessary. To date, only one small randomized trial14 has demonstrated the use of T4 replacement in women with thyroid peroxidase antibodies to reduce preterm birth. Importantly, untreated thyroid peroxidase antibody-positive women in the study had a remarkably higher preterm birth rate than is reported in this and other studies.20 Furthermore, to show a salutary effect of T4 supplementation to prevent abruption would likely require randomization of a massive number of pregnant women.
We remain of the view that, until sufficiently powered randomized trials demonstrate a positive effect, either by identifying a preventable pregnancy risk factor or by improving pregnancy outcomes through thyroid hormone supplementation, universal prenatal screening for thyroid-related analytes is not justified. One such study currently in progress is being conducted by the Maternal-Fetal Medicine Units Network of the Eunice Kennedy Shriver National Institute of Child Health and Human Development. At this time, only targeted prenatal screening for thyroid dysfunction is recommended by the American College of Obstetricians and Gynecologists.22
1. 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.
2. Casey BM, Dashe JS, Spong CY, McIntire DD, Leveno KJ, Cunningham GF. Perinatal significance of isolated maternal hypothyroxinemia identified in the first half of pregnancy. Obstet Gynecol 2007;109:1129–35.
4. Cleary-Goldman J, Malone FD, Lambert-Messerlian G, Sullivan L, Canick J, Porter TF, et al. Maternal thyroid hypofunction and pregnancy outcome. Obstet Gynecol 2008;112:85–92.
5. 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.
6. Pop VJ, Kuijpens JL, van Baar AL, Verkerk G, van Son MM, de Vijlder JJ, et al. Low maternal free thyroxine concentrations during early pregnancy are associated with impaired psychomotor development in infancy. Clin Endocrinol (Oxf) 1999;50:149–55.
7. Pop VJ, Brouwers EP, Vader HL, Vulsma T, van Baar AL, deVijlder JJ. Maternal hypothyroxinaemia during early pregnancy and subsequent child development: a 3-year follow-up study. Clin Endocrinol (Oxf) 2003;59:282–8.
8. Kuijpens JL, De Hann-Meulman M, Vader HL, Pop VJ, Wiersinga WM, Drexhage HA. Cell-mediated immunity and postpartum thyroid dysfunction: a possibility for the prediction of disease? J Clin Endocrinol Metab 1998;83:1959–66.
9. Stagnaro-Green A, Roman SH, Cobin RH, el-Harazy E, Alvarez-Marfany M, Davies TF. Detection of at-risk pregnancy by means of highly sensitive assays for thyroid autoantibodies. JAMA 1990;264:1422–5.
10. Stagnaro-Green A, Glinoer D. Thyroid autoimmunity and the risk of miscarriage. Best Pract Res Clin Endocrinol Metab 2004;18:167–81.
11. Stagnaro-Green A, Chen X, Bogden JD, Davies TF, Scholl TO. The thyroid and pregnancy: a novel risk factor for very preterm delivery. Thyroid 2005;15:351–7.
12. Glinoer D, Riahi M, Grun JP, Kinthaert J. Risk of subclinical hypothyroidism in pregnant women with asymptomatic autoimmune thyroid disorders. J Clin Endocrinol Metab 1994;79:197–204.
13. Wilson R, Ling H, MacLean MA, Mooney J, Kinnane D, McKillop JH, et al. Thyroid antibody titer and avidity in patients with recurrent miscarriage. Fertil Steril 1999;71:558–61.
14. Negro R, Formoso G, Mangieri T, Pezzarossa A, Dazzi D, Hassan H. Levothyroxine treatment in euthyroid pregnant women with autoimmune thyroid disease: effects on obstetrical complications. J Clin Endocrinol Metab 2006;91:2587–91.
15. Sinclair D. Clinical and laboratory aspects of thyroid autoantibodies. Ann Clin Biochem 2006;43:173–83.
16. Glinoer D. The systematic screening and management of hypothyroidism and hyperthyroidism during pregnancy. Trends Endocrinol Metab 1998;9:403–11.
17. Poppe K, Glinoer D. Thyroid autoimmunity and hypothyroidism before and during pregnancy. Hum Reprod Update 2003;9:149–61.
18. Negro R, Schwartz A, Gismondi R, Tinelli A, Mangieri T, Stagnaro-Green A. Universal screening versus case finding for detection and treatment of thyroid hormonal dysfunction during pregnancy. J Clin Endocrinol Metab 2010;95:1699–707.
19. Dashe JS, Casey BM, Wells CE, McIntire DD, Byrd EW, Leveno KJ, et al. Thyroid-stimulating hormone in singleton and twin pregnancy: importance of gestational age-specific reference ranges. Obstet Gynecol 2005;106:753–7.
20. Männistö T, Vääräsmäki M, Pouta A, Hartikainen AL, Ruokonen A, Surcel HM, et al. Perinatal outcome of children born to mothers with thyroid dysfunction or antibodies: a prospective population-based cohort study. J Clin Endocrinol Metab 2009;94:772–9.
21. Cleary-Goldman J, Malone FD, Vidaver J, Ball RH, Nyberg DA, Comstock CH, et al. Impact of maternal age on obstetric outcome. Obstet Gynecol 2005;105:983–90.
© 2010 by The American College of Obstetricians and Gynecologists.
22. Thyroid disease in pregnancy. ACOG Practice Bulletin No. 37. American College of Obstetrics and Gynecology. Int J Gynaecol Obstet 2002;79:171–80.