| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Endocrinology and Diabetes Division, VA Greater Los Angeles Healthcare System, Departments of Medicine and Physiology, David Geffen School of Medicine at University of California, Los Angeles, Los Angeles, California 90073
Address all correspondence and requests for reprints to: Gregory A. Brent, Endocrinology and Diabetes Division, VA Greater Los Angeles Healthcare System, 11301 Wilshire Boulevard, Los Angeles, California 90073. E-mail: gbrent{at}ucla.edu.
It has long been recognized that maternal thyroid hormone excess or deficiency can influence the outcome for mother and fetus at all stages of pregnancy, as well as interfere with ovulation and fertility (1, 2). Maternal hypothyroidism is the most common disorder of thyroid function in pregnancy and has been associated with fetal loss, pregnancy-induced hypertension, preterm delivery, placental abruption, and reduced intellectual function in the offspring (3, 4, 5, 6). These adverse outcomes have been associated with both overt hypothyroidism (elevated serum TSH concentration and reduced free T4 concentration), found in about 0.2% of pregnancies, as well as subclinical hypothyroidism (elevated serum TSH concentration and normal serum free T4 concentration), found in about 2.3% of pregnancies (3, 4, 5, 6). Maternal overt hyperthyroidism (suppressed serum TSH and elevated serum free T4 and T3 concentrations) is less common, affecting approximately two of 1000 pregnancies (2). The adverse events for mother and fetus include fetal loss, fetal growth restriction, preeclampsia, and preterm delivery (1, 2, 7). Mild or subclinical hyperthyroidism (suppressed serum TSH alone) is found in about 1.7% of pregnancies and has not been specifically associated with adverse pregnancy outcomes (8). The severity and duration of maternal thyroid disease necessary to produce these abnormalities, however, and the value and timing of therapeutic intervention remain controversial.
Screening is defined as testing for a disease when there are no signs or symptoms, with the goal of improving health outcomes. In view of the potential for serious adverse events associated with maternal thyroid disease, and the apparent benefits of treatment, many have recommended routine thyroid function screening in pregnancy (9). The group of physicians performing most prenatal care, obstetrician-gynecologists, however, have not advocated routine screening for thyroid disease in pregnancy (2, 10). Professional associations of endocrinologists have taken varying views, with evidenced-based review panels concluding that there is not sufficient evidence to mandate routine screening (11), and other expert panels advocating screening (9). Groups that gathered as part of two multidisciplinary conferences that included endocrinologists and obstetricians, as well as other experts, concluded that the evidence is insufficient for routine thyroid screening in pregnancy (12, 13). These groups endorsed randomized prospective trials of treatment of subclinical hypothyroidism to inform the screening debate (12, 13).
An alternative to universal thyroid function screening in pregnancy, advocated by a number of groups, is a case-finding approach in which women at increased risk for thyroid disease are identified and tested (11, 12, 13). The American College of Obstetrics and Gynecology endorses case finding, recommending thyroid function testing in women with a history of thyroid disease or with symptoms (10). A central question in the screening debate has been the fraction of pregnant women with thyroid disease who would be successfully identified as a result of a case-finding strategy. The study by Vaidya et al. (14), in this issue, directly addresses this very important question. In this study, 1560 consecutive pregnant women seen for their first prenatal visit had thyroid function tests, and most also had thyroid peroxidase (TPO) antibodies measured. Information was gathered to determine the presence of a number of identified risk factors for thyroid disease. Based on these thyroid disease risk factors, women were divided into a low-risk group, which included about 75% of the women, and a high-risk group, which included about 25%. Forty women (2.6%) were found to have an elevated serum TSH, and most of these women, 70%, were in the high-risk group. A "fully" suppressed serum TSH (defined as <0.03 mIU/liter) was found in 1.9% of women. The risk factors most strongly associated with abnormal thyroid function were personal history of thyroid disease, other autoimmune disease, and family history of thyroid disease.
Can the findings from Vaidya et al. (14) be generalized to other populations of pregnant women? The authors point out that the population was predominantly Caucasian, from a focused geographic area, with a similar profile of iodine intake (14). Despite sampling from a relatively narrow population, the fraction of women with thyroid function test abnormalities and the nature of the abnormalities were very close to those reported in a much larger series from North America with over 25,000 pregnant women (3). Importantly, the risk of thyroid disease found among first- and second-degree relatives in the study by Vaidya et al. (14) is very close to that found in a European study in nonpregnant women, and there was a similar strength of association of family history with thyroid disease (15). This concordance in findings suggests that the study population is similar to those from other reported series.
Based on the findings of Vaidya et al. (14), most pregnant women with thyroid disease would be identified by case finding, although at least 30% would not. Is it enough to diagnose approximately two thirds of pregnant women with thyroid disease? Are there factors that should be considered that would identify the one third of pregnant women with thyroid dysfunction, and was there any difference in the type of thyroid dysfunction in this group? The risk factors used by Vaidya et al. have been confirmed by many other studies, and it is unlikely that additional readily available factors would improve sensitivity (15). The women in the low-risk group had less severe hypothyroidism and were less likely to have a TSH level greater than 10 mIU/liter than those in the high-risk group, but even mild elevations in maternal serum TSH are associated with adverse outcomes in pregnancy (3, 4, 5, 6). The approach by Vaidya et al. demonstrates the maximum number of women that could be identified by case finding but likely overestimates the yield of the case-finding approach in clinical practice. All women in the study were tested, and the appropriate risk factor data were gathered. Although a personal history of thyroid disease or autoimmune disease would likely be obtained in most prenatal visits, family history was a factor for over 80% of women in the high-risk group, and this information may not be routinely collected in practice. This same group reported the results of a clinical practice audit in which only 20% of high-risk women were tested in pregnancy, despite guidelines to specifically identify and test these women (16).
The vast majority of pregnant women with thyroid disease have subclinical hypothyroidism (2, 3), and any consideration of thyroid function test screening in pregnancy depends on the clinical significance of this condition. Several studies have shown that adverse events are found in pregnant women with overt and subclinical hypothyroidism but not in women with hypothyroidism adequately replaced with L-T4 (4, 6). Although this implies that L-T4 treatment prevents the adverse events associated with maternal hypothyroidism, this had not been directly tested. Findings from the recent prospective randomized controlled treatment study of almost 1000 pregnant women by Negro et al. (5) demonstrated the benefit of early intervention with L-T4 in women with positive TPO antibodies and normal-range serum TSH levels. L-T4 treatment reduced the rate of miscarriage and preterm delivery to the levels in women without thyroid disease. These findings provide evidence for the value of identification and treatment of subclinical hypothyroidism in pregnancy.
Several recent studies indicate deficits in knowledge about thyroid disease in pregnancy among obstetrician-gynecologists and endocrinologists, but these deficits are even more marked among primary care physicians (17, 18). About 50% of obstetrician-gynecologists are satisfied with their knowledge of management of thyroid disease, although a significant fraction of those surveyed had low confidence in their training in this area (17). Despite the absence of recommendations for universal thyroid test screening, the clinical practice of endocrinologists and obstetrician-gynecologists may differ. A study of prenatal care among clinical practices representing 85% of deliveries in Maine reported that almost 50% of doctors routinely tested at least 95% of pregnant women for thyroid disease in the first trimester (19). Thyroid function test screening was much more common among obstetricians, 56%, compared with family practitioners, 8%.
There are a number of barriers before routine thyroid function screening can be recommended in pregnancy (12, 13). Issues to be resolved include the thyroid test(s) to be used, timing of the determination, threshold to characterize as an abnormality, appropriate intervention, and monitoring. Most would advocate a serum TSH assay with trimester-specific normal ranges, based on the ability to detect thyroid hormone deficiency or excess, and its the best-characterized test with respect to guidelines for intervention (2, 20, 21). Addition of a TPO measurement may also be considered due to related adverse pregnancy events, as well as identifying those women at risk for postpartum thyroiditis. Women with isolated low serum free T4 concentration and normal serum TSH have been reported in a number of series and were found in 7.8% of the entire cohort reported by Vaidya et al. (14), with equal frequency in the low- and high-risk groups. Some have reported reduced intellectual function in the offspring of women with an isolated low free T4 (22), and these women would not be identified if only a serum TSH is measured. Free T4 assays generally result in lower values in pregnancy, and the influence of pregnancy also varies among different assay kits (21). An additional factor to include in a screening decision is the risk/benefit of identifying and treating thyroid disease. Although L-T4 treatment should have few adverse effects, this must be factored into any consideration. Reduction of maternal hypothyroidism has focused not only on better diagnosis, but also prevention, by ensuring adequate iodine intake. Recent recommendations for North America include iodine supplementation in pregnancy (23).
There are divergent views on the optimal approach to identify thyroid disease in pregnant women, although all share in the desire to identify and treat thyroid disorders to improve pregnancy outcomes for mothers and infants. The association of pregnancy loss and preterm delivery with maternal hypothyroidism and response to treatment provide proximal end points that are more easily measured than the very long-term studies needed to demonstrate reversal of the associated intellectual deficit in children. A prospective randomized study currently in progress and led by Dr. John Lazarus, Wales Controlled Antenatal Thyroid Screening Study (CATS), is designed to directly test the value of screening for thyroid disease and treating women with serum TSH elevations. It is sufficiently powered to determine conclusively the benefit of early identification and intervention. Until these data are available, however, there is evidence that supports the benefit of intervention with L-T4 therapy to reduce pregnancy loss and preterm delivery when mild maternal hypothyroidism is identified (5). Vaidya et al. (14) demonstrate that at least one third of pregnant women with thyroid disease will be missed by a high-risk case-finding approach. With results such as those reported in this issue, clinicians of all types will join with the obstetricians in Maine and many others and will begin to test pregnant women routinely for thyroid function. Larger prospective intervention trials, as are being performed, are invaluable and will significantly inform the approach to thyroid disease screening. Any mandated screening program must carry very clear guidelines for timing and intervention and must be developed in partnership with obstetrician-gynecologists, who provide care for most pregnant women. When the potential adverse outcomes are so significant and the tools to diagnose and intervene are easily accessible, however, leaving maternal thyroid disease undiagnosed, even in one third of pregnant women, is no longer acceptable.
Footnotes
Abbreviation: TPO, Thyroid peroxidase.
Received November 8, 2006.
Accepted November 10, 2006.
References
This article has been cited by other articles:
 |
|
 |
|
  R. Stricker, M Echenard, R Eberhart, M-C Chevailler, V Perez, F A Quinn, and R. Stricker Evaluation of maternal thyroid function during pregnancy: the importance of using gestational age-specific reference intervals Eur. J. Endocrinol., October 1, 2007; 157(4): 509 - 514. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
