Clinical Endocrinology (1991) 35,4146

Prevalence of thyroid deficiency in pregnant women R. 2. Klein", J. E. Haddowt, J. D. Faix$, R. S. Browns, R. J. HermosT, A. Puikkinent and M. t. Mitcheiill 'Department of Maternal and Child Health, Dartmouth Medical School, Hanover, NH tFoundation for Blood Research, Scarborough, Maine, $Department of Pathology, Beth Israel Hospital, Harvard Medical School, Boston, Mass., §Department of Pediatrics, University of Massachusetts Medical School, Worcester, Mass., TState Laboratory Institute, Massachusetts Department of Public Health, Boston, Mass, USA (Received 26 November 1990; returned for revision 3 January 1991; finally revised 28 January 1991; accepted 14 February 1991)

Summary OBJECTIVE The present study was designed to determine the current prevalence of gestational hypothyroidism, since maternal thyroxine deficiency is associated wlth poor obstetric outcomes and mental retardation in the surviving offspring. DESIGN TSH concentrationswere measuredin the sera of women at 15-18 weeks of gestation. Those sera with TSH concentrations above 6 mull and the two sera closest in order with TSH concentrations below 6 mull were further anaiysed for 14, Fr4, TBG, and antithyroid antibodies. Study criteria for hypothyroidismwere sera with elevated concentrations of TSH plus both a free 14 concentratlon and a total 1 4 concentration and /or T4lTBG ratio more than two standard deviations below the mean for the control pregnant women. PATIENTS The sera were from 2000 consecutivewomen in Maine being tested for alpha-fetoprotein concentration a1 15-18 weeks of gestation. RESULTS TSH concentrationsabove 6 mull were found in the sera of 49 women, 2.5% of the pregnant women. Six women with elevated TSH concentratlons (range 64-54 mU/i) had both a F14 concentration and a T4/TBG ratio and/or a 14 concentration more than two standard deviations below the respective control means, meeting the study criteria for thyroid deflclency, and thus giving a prevalence of 0.3%. The remaining 43 women with elevated TSH concentrations were classified as having compensated thyroid disease although some may have been hypothyroid. Fifty-eight per cent of women with TSH concentrationsabove 6 mull and 90% of the women with

Correspondence: R. Z . Klein, Department of Maternal and Child Health, Dartmouth Medical School, Hanover, NH 03756,USA.

elevated TSH concentrations and at least one thyroxine index more than two standard deviationsbelow the control means had positive titres of antithyroid antibodies as opposed to 11YOof the controls. CONCLUSIONS Although it Is not known what severity of maternal thyrold deficiency is necessary to cause fetal brain damage, the present data indicate a sufficiently high prevalence of thyroid dysfunction to demand investlgatlon of the mental development of the offspring of women wlth thyroid dysfunction and of the effect of replacement therapy.

The conventional wisdom that maternal thyroxine concentration plays no part in fetal brain development is now challenged by evidence from several sources. Maternal hypothyroidism has been associated with mental retardation in the living euthyroid offspring as well as increased fetal and neonatal losses (Parkin & Greene, 1943; Hodges et al., 1952; Greenman et al., 1962 Echt & Doss, 1963; Jones & Man, 1969; Anderson & Beales, 1970; Lachelin, 1970; Goldsmith et al., 1973; Man & Serunian, 1976; Montoro et al., 1981; Francis & Riley, 1987; Matsuura, 1989). Pharoah reported that the intelligence of the euthyroid offspring of women with endemic hypothyroidism correlated with maternal thyroid hormone levels during pregnancy (Pharoah et al., 1984). Conversely, when the fetus is hypothyroid but the mother is euthyroid, brain function is preserved in utero, as demonstrated by long-term follow-up studies of infants with permanent congenital hypothyroidism diagnosed and treated adequately shortly after birth (New England Congenital Hypothyroidism Collaborative, 1985, 1990; Klein, 1989). Studies by Morreale de Escobar's group have shown that the maternal thyroxine traversing the placenta coupled with increased fetal brain 5'-deiodinase activity is adequate to maintain normal triiodothyronine concentration in the brain of the hypothyroid rat fetus (Obregon et al., 1984; Morreale de Escobar et al., 1985, 1988; Ruiz de Ona et al., 1988). When the mother is hypothyroid, fetal brain development could be damaged by lack of available thyroxine before fetal thyroid function begins or after that if the fetal thyroid were also affected by maternal antibodies such as thyrotrophin binding inhibitory immunoglobulins. Brain damage could also result if the mother's thyroid insufficiency led to a nonspecific inability to support the fetus optimally. It is not known whether damage to the fetal brain in association with maternal hypothyroidism is a threshold 41

42

Clinical Endocrinology (1991) 35

R. 2.Klein et a / .

effect or whether it occurs on a continuum. It is also not known how often various degrees of thyroid deficiency occur in pregnancy. This study was designed to establish the current prevalence of thyroid deficiency among pregnant women in order to evaluate the extent to which that condition could contribute to the overall incidence of mental retardation. Our ultimate purpose is to determine the severity of maternal hypothyroidism required to produce poor obstetric outcomes and mental retardation in the offspring and the efficacy of maternal thyroxine replacement therapy in preventing the adverse effects of maternal hypothyroidism.

140 -

120 c

s

loo8060 -

40 20 0-

Material and methods

The Foundation for Blood Research routinely tests the sera from all the patients in the practices of 144 obstetricians and midwives in Maine for alpha-fetoprotein screening. This cooperative network was recruited for a 1978 research study from the larger obstetrical practices in the state. These practices currently take care of over 60% of the pregnant women in the state. Aliquots of the frozen residues of 2000 sera obtained from two cohorts of consecutively examined women were assayed for thyroid stimulating hormone (TSH) concentrations. The sera were obtained during the 15th to 18th week of gestation. Gestational duration was determined in 40% of the pregnancies by ultrasonography and calculated from the date of the last menstrual period in the rest. The aliquots were coded so that the donors of the sera could not be linked with the laboratory results. TSH was assayed using a commercially available kit (Nichols Institute, San Juan Capistrano, California). Sera whose TSH concentration was 6 mU/l or more and control sera were subjected to further analyses. The control sera were those with TSH concentrations below 6 mU/1 that were in the test sequence immediately before and after the sera with TSH concentrations of 6 mU/1 or over. Total thyroxine (T4) and thyroxine binding globulin (TBG) concentrations were measured using the Amerlite enzyme immunoassay system (Amersham Corp., Arlington Heights, Illinois). Free thyroxine (FT4) concentrations were measured by the Kineticount two-step, nonanalog radioimmunoassay (Vitek Systems Inc., Hazelwood, Mo.). Antibodies to thyroid peroxidase (antimicrosomal antibodies) were measured by a commercially available kit (Ames, Elkhart, Indiana). Haemagglutination at a dilution of 1:lOO or more was considered a positive result. Student's t-test and analysis of variance were used to compare group means and forward stepwise regression analyses were used in studying correlations among age, gestational duration, T4, FT4, TSH, TBG, and antithyroid antibody concentrations. x2 analysis was used to compare

I

II60 8OE

0

2

4

6

8

I012

--

1 4 16 20 40 60

TSH ( m U/Il

Flg. 1. Distribution of TSH concentrations in sera of 2000 women at 15-18 weeks of gestation.

percentages of positive antibodies. Statistical analyses were performed using the StatView Program (Brainpower, Inc., Calabasas, California) on a MacIntosh computer. A P value of less than 0.05 was considered statistically significant. The minimal study criteria for thyroxine deficiency in those with TSH concentrations above 6 mU/1 were a FT4 concentration in association with a T4/TBG ratio and/or T4 concentration more than two standard deviations below the respective control means. Women with similar TSH concentrations but without these changes in thyroxine indices were considered to have compensated thyroid disease. This study was reviewed and approved by the Committee for the Protection of Human Subjects of Dartmouth Medical School and the Dartmouth-Hitchcock Medical Center. Results

Figure 1 represents the distribution of individual values of TSH. The abnormal distribution with skewing to the right results from the inclusion of women with hypothyroidism and with compensated thyroid disease. The geometric mean of the TSH concentrations of the 2000 women was 2.09 m u / 1. The median was 2.2 mu/]. TSH concentrations of 6.0 mU/l or more were present in 2.5% of the sera. The mean TSH concentration for the remaining 1951 sera with concentrations less than 6 mU/1 was 2.3 1.1 (SD) with a normal distribution of individual values. There was a statistically significant positive correlation of maternal age and TSH concentration among all 2000 sera, r =0.05, P < 0.03, which disappeared when the women with TSH concentrations equal to or above 6 mU/1 were removed from the analysis. Table 1 gives data for the 49 sera that were analysed further because TSH concentrations were above 6.0 mU/1

Maternal hypothyroidism

Clinical Endocrinology (1991) 35

43

Table 1 Mean thyroid related

measurements and age and gestational duration

Age (years)

T4 Gest. TSH (weeks) (mU/I) (nmol/l)

Control pregnant women TSH < 6.0 (n =99) Mean 26.9 16.8 2.34 145 SEM k0.54 k0.06 k2.8

FT4 (pmol/l)

13.4 k0.28

TBG T4/TBG Antibody (mg/l) (nmol/mg) (% pos.)

35.6 k0.68

4.2 kO.08

Pregnant women with elevated TSH concentrations TSH > 6 (n =49) Mean 29.1 16% 10.02 I37 11.5 37-2 3.7 SEM k0.78 k0.12 k5.1 kO.44 k1.03 k0.12 A P=0.019 P =0.0002 P=0403

11

58 P

Prevalence of thyroid deficiency in pregnant women.

The present study was designed to determine the current prevalence of gestational hypothyroidism, since maternal thyroxine deficiency is associated wi...
516KB Sizes 0 Downloads 0 Views