| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Obstetrics and Gynecology (S.B.N., A.J.) and Endocrinology and Medicine (K.M.P., P.L.), Aalborg Hospital, DK 9000 Aalborg, Denmark
Address all correspondence and requests for reprints to: Dr. Susanne B. Nøhr, Department of Obstetrics and Gynecology, Aalborg Hospital, DK 9000 Aalborg, Denmark. E-mail: snohr{at}dadlnet.dk
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
PPTD developed in 55% of the participants. In 67% of the cases abnormal TSH was accompanied by abnormalities in thyroid hormones, whereas 33% had abnormal serum TSH only. There was no statistically significant difference in the frequency of PPTD in the three groups: +/+ group, 59% (95% confidence interval, 36-79%); +/- group, 60% (36-81%); and -/- group, 46% (26–67%). There were also no differences in the severity of the PPTD, as evaluated by duration and grade of deviation of TSH and thyroid hormones from normality. The occurrence, severity, and type of PPTD predominantly depended on the TPO-Ab level: TPO-Ab below 200 U/L at screening, 35% developed PPTD; TPO-Ab of 200–900 U/L, 54%; and TPO-Ab above 900 U/L, 75% developed PPTD. Women with low levels of antibodies predominantly remained euthyroid or had hyperthyroidism only, whereas women with high antibody levels had hyperthyroidism followed by hypothyroidism or hypothyroidism only. We conclude that iodine supplementation (150 µg) during pregnancy and the postpartum period to TPO-Ab-positive women living in an area with mild to moderate iodine deficiency did not induce or worsen PPTD. The study confirmed that screening for TPO-Ab in early pregnancy can predict women at high risk for development of PPTD.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
One possible side-effect of iodine is aggravation of thyroid (autoimmune) dysfunction, which has been demonstrated in experimental animals and in patients with thyroid disorders (6, 7, 8). Subclinical autoimmune thyroid disease with measurable antibodies against thyroid peroxidase (TPO-Ab) in serum is common in pregnant women, and many of these develop postpartum thyroid dysfunction (PPTD). A study performed in an iodine-sufficient population showed that iodine given to women with TPO-Ab in the postpartum period might aggravate thyroid dysfunction if they develop postpartum thyroiditis (9). In a previous intervention study we found that iodine supplementation (200 µg) given to pregnant women living in an area of moderate iodine deficiency may ameliorate some of the pregnancy-associated variations in maternal thyroid function (10). However, this study aroused suspicion that iodine supplementation could induce PPTD in women who were TPO-Ab positive (10).
The aim of the present study was to evaluate whether iodine supplementation during pregnancy and the postpartum period to TPO-Ab-positive women living in an area in Denmark with mild to moderate iodine deficiency imposes any danger of worsening PPTD.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
)Screening for TPO-Ab in serum was performed in early pregnancy (median, 11th gestational week) in a consecutive cohort of 1284 healthy pregnant Danish women (age, 18–35 yr) referred to Aalborg Hospital for an obstetrical routine ultrasound investigation. TPO-Ab was found in 117, corresponding to a prevalence of 9.1%.
|
Basic characteristics of the remaining cohort (n = 66) within the
three groups are shown in Table 1. There
were no statistically significant differences among the groups at the
start of the study. The study had a power of 90% for detecting a
doubling of the frequency of PPTD.
|
).
At inclusion, all participants stopped their previous intake of
supplements. All thyroid function tests were performed after closure of the study, and the occurrence of postpartum thyroid dysfunction was not evaluated during the study period, except in one woman. She presented with symptoms of hypothyroidism and had PPTD with a biphasic course. She recovered within the study period, and substitution with thyroid hormones was not necessary. The supplement code was not broken, nor was supplement ceased.
Iodine supplement
The vitamin tablets (Livol Super) were provided by Dansk Droge A/S (Copenhagen, Denmark) and contained the recommended daily allowances of various vitamins and minerals, including 50 µg selenium. The iodine content was 150 µg (which is the traditional iodine content of such tablets in Denmark), except in the batches where iodine was omitted.
Biochemical evaluation
TPO-Ab at screening was measured by routine biochemical procedures. Women with TPO-Ab of 100 U/mL or more were considered TPO-Ab positive and met the criteria for inclusion. In the randomized study, serum for biochemical evaluation of thyroid function and antibody level was collected at inclusion, at the 35th week of gestation, and at 3, 5, 7, and 9 months postpartum. Compliance was evaluated by 24-h urinary iodine measurements at time of inclusion, during pregnancy (35th week), and during the postpartum period (7 months postpartum). Blood samples were centrifuged shortly after sampling, and serum and urine were stored at -20 C until analysis. In all assays samples from one individual were analyzed in the same assay.
Total T4 (reference range, 60–140 nmol/L) and total T3 (1.2–2.7 nmol/L) were measured by RIAs (Amerlex-M, Ortho-Clinical Diagnostics, Amersham Pharmacia Biotech, Aylesbury, UK). TSH (detection limit, 0.008 mU/L; reference range, 0.40–4.0 mU/L), thyroglobulin (Tg; detection limit, 0.2 µg/L), and free T4 (FT4; reference range, 10–25 pmol/L) were measured by immunoluminometric assays, and free T3 (FT3; reference range, 3.4–7.1 pmol/L) was measured by a luminescence immunoassay (Lumitest, Brahms Diagnostica GmbH, Berlin, Germany). At screening, TPO-Ab were measured initially by luminescence immunoassay (detection limit, 100 U/mL; Lumitest anti-TPO, Henning Berlin GmbH & Co., Berlin, Germany) and later by a radioimmunoprecipitation assay (detection limit, 5.5 U/mL; Dynotest anti-TPOn, Brahms Diagnostica). For analyses of samples from the randomized study, the Dynotest Anti-TPO assay was used. Tg antibodies (Tg-Ab; detection limit, 5.5 U/mL) were measured by radioimmunoprecipitation assays (Dynotest anti-Tgn, Brahms Diagnostica). Although Tg-Ab may influence Tg values (11), the Tg data presented here were not adjusted for interference. In women with a hyperthyroid episode, TSH receptor autoantibodies (TRAb) were measured by RRA (detection limit, 2.4 U/L; reference range, <9 U/L; TRAk-Assay, Brahms Diagnostica).
Iodine in urine was measured in duplicate by the Ce/As method after alkaline ashing, using a modification of the method of Wilson and van Zyl (12) as previously described (13). The detection limit was 2 µg/L, and the recovery of added 127I and 125I was greater than 95% and not corrected for.
Urinary iodine excretion in the two groups (+/+ and +/- groups) that received 150 µg iodine during pregnancy was (median) 187 µg/24 h (105 µg/L; n = 41) in the 35th gestational week, whereas it was 71 µg/24 h (53 µg/L) in the -/- group. In the +/+ group that received iodine supplementation postpartum as well, it was (median) 124 µg/24 h (75 µg/L; n = 21) 7 months postpartum, whereas it was 65 µg/24 h (43 µg/L) in the women not receiving iodine (+/- and -/- groups). The relatively low urinary iodine excretion 7 months postpartum could be due to rebuilding of thyroid hormone stores after previous loss of hormone due to thyroiditis (14, 15), or it could be explained by loss of iodine in breast milk in mothers still nursing the child. Finally, a fall in compliance of some mothers when they resumed work (usually about 5–6 month postpartum) has to be considered.
Outcome measurements
PPTD was defined as abnormal TSH in the postpartum period. It was classified as subclinical if only TSH was abnormal and clinical if also thyroid hormones were abnormal (subclinical hypothyroidism: TSH, >4.0 mU/L and normal thyroid hormones; clinical hypothyroidism: TSH, >4.0 mU/L; and FT4, <10 pmol/L; or TT4, <60 nmol/L; subclinical hyperthyroidism: TSH, <0.4 and normal thyroid hormones; clinical hyperthyroidism: TSH, <0.4 mU/L; and FT4, >25 pmol/L; or TT4, >140 nmol/L; or FT3, >7.1 pmol/L; and/or TT3, >2.7 nmol/L). The severity of the PPTD was estimated by a biochemical dysfunction score, taking both duration and grade of deviation of thyroid function from normality into account. At each time of evaluation, a subclinical abnormality gave a score of 1, and a clinical abnormality gave a score 2, or 3 if all measured hormones were abnormal. A biochemical hypothyroid score of 3 was given if TSH was elevated and TT4, FT4, and TT3 were below the reference ranges. Free T3 was not used for evaluation of hypothyroidism because the FT3 assay suffered from lack of precision in the low range of measurements. This biochemical grading taking a low T3 into account correlated to a clinical score of hypothyroidism in a study by Zulewski et al. (16). In hyperthyroidism FT4 was increased in all episodes given a score of 2, except in one woman who had a late hyperthyroid episode with positive TRAb. She had a predominant increase in T3.
Three women with one or two missing values were evaluated as having scored zero, as all of the available TSH values were normal. One woman had subclinical hyperthyroidism at 3 and 5 months postpartum, but no further tests were available. For the score measurement it was assumed that no subsequent hypothyroid phase took place.
Late outcome was addressed by a questionnaire and a thyroid function blood test.
Statistical methods
Nonparametric statistical analyses were applied, because
antibody measurement (TPO and Tg) and TSH values (when dysfunction was
present) could not be transformed into a normal distribution. The
Statistical Package of Social Sciences (SPSS, Inc.,
Chicago, IL) was used to calculate descriptive data, to test
differences between groups (
2 test,
Kruskal-Wallis test, and Mann-Whitney test), to test differences within
groups (Wilcoxon signed rank test), for correlation’s (Spearman test),
and for multivariate analysis [logistic regression (full model)]. A
5% level of statistical significance was used.
Ethical aspects
The protocol was reviewed and approved by the local ethical committees. All participants gave informed written consent.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Iodine supplementation and PPTD
There was no statistically significant difference in the frequency
of PPTD in the three groups (by 2 test,
P = 0.56): +/+ group, 59% (95% confidence interval,
36–79%), +/- group, 60% (36–81%); and -/- group, 46%
(26–67%), although a small trend was observed. One of the six women
leaving the study had evidence of PPTD (abnormal TSH 3 months
postpartum), and one had normal TSH 3 months postpartum (no further
test available). Neither these data nor intention to treat analysis
including all six censored women altered the outcome.
The types of PPTD in the three groups are depicted in Fig. 2. There were no differences among the
groups. Neither were there any differences in the severity of the PPTD,
as evaluated by the duration and grade of deviation of TSH and thyroid
hormones from normality by the biochemical dysfunction score. Scores
from the women in relation to treatment groups are depicted in Fig. 3a.
|
|
The occurrence and severity of PPTD predominantly depended on the
TPO-Ab level. The well known pattern of a decline in TPO-Ab during
pregnancy and a postpartum surge was observed. There were no
statistically significant differences in TPO-Ab among the treatment
groups at any time of evaluation, nor were there any differences in the
absolute or fractional increment in TPO-Ab (maximum levels postpartum
in relation to levels at screening). The TPO-Ab values at screening
correlated to the postpartum surge in TPO-Ab (Spearman’s r =
0.87; P < 0.001). The TPO-Ab level at screening was
also a good predictor of the PPTD risk, as shown in Fig. 4, and it correlated to the severity of
PPTD, as evaluated by the biochemical dysfunction score (Fig. 3b).
|
, showing
the variations in thyroid function, TPO and Tg autoantibodies (85% of
the women had detectable Tg-Ab in the postpartum period), and serum Tg
in relation to the four types of postpartum thyroid function
(euthyroidism, hypothyroidism only, hyperthyroidism only, and biphasic
course). Parallel variations were observed in TPO-Ab and Tg-Ab, and the
variations in antibody quantity were essentially the same in all four
types, but at different levels. Antibody levels were high in women with
a biphasic course of PPTD and with hypothyroidism only and were much
lower in the hyperthyroidism only and euthyroid groups. The indication
of a less severe abnormality in women with hyperthyroidism only was
further substantiated by the variations in thyroid hormones and Tg.
Only 20% had thyroid hormone abnormalities, e.g. clinical
PPTD, as opposed to 75% in the hypothyroidism only group and almost
90% in women with a biphasic course. The debut and duration of
deviations were similar in the hyperthyroidism only and the biphasic
groups, whereas the hypothyroidism only group manifested later, as
evaluated by both TSH and Tg variations. One woman had an unusual
biphasic course, with hypothyroidism (5 month postpartum) followed by
hyperthyroidism. She was transiently TRAb positive during the
postpartum period and had clinical hyperthyroidism at the end of the
randomized study. She later reported that she had been followed by her
general practitioner because of fatigue until 16 months postpartum,
when thyroid hormone levels normalized and she became TRAb negative.
None of the other women with a hyperthyroid episode was TRAb
positive.
|
Long-term consequences
Although abnormalities in thyroid function mostly were transient
and tended to normalize 9 months postpartum (Fig. 5), 14 women
(21%) still had abnormal serum TSH at the end of the randomized study
(9 months postpartum). Twelve had hypothyroidism (9 subclinical and 3
clinical), and 2 had hyperthyroidism (1 subclinical and 1
clinical).
Follow-up 3 yr postpartum (median) of the initial cohort (n = 72) showed that four women later had a registered diagnose of thyroid disease. From the +/- group, the woman with symptoms of PPTD during the study period had transient hypothyroidism 1.5 yr later, one woman developed euthyroid goiter, and one developed permanent hyperthyroidism due to Graves’ disease (she left the study after pregnancy and therefore was excluded). From the -/- group, one woman had hyperthyroidism due to toxic adenoma.
Of the remaining 68 women, 67 completed a questionnaire and had a test of thyroid function. Four women had either subclinical (n = 2) or clinical (n = 2) hypothyroidism. These 4 women were all subclinically hypothyroid at the end of the randomized study. Thus, a total of 8 women (11%) had evidence of thyroid disease at follow-up.
Iodine supplementation and thyroid function in late pregnancy
Measurements of TSH, thyroid hormones, and thyroid
antibodies in week 35 of pregnancy in women receiving iodine (+/+ and
+/- groups) and women not receiving iodine (-/- group) are shown in
Table 2. There were no statistically
significant differences when comparing absolute values at 35 weeks
gestation. Iodine supplementation modified changes in TSH and Tg during
pregnancy. In the group without iodine there was a small, but
systematic, increase in TSH during pregnancy, whereas the TSH
alterations during pregnancy in the group with iodine showed a more
diverse pattern, with no significant difference between early and late
pregnancy. The small increase in Tg during pregnancy reverted to a
considerable fall after iodine supplementation.
FT4 declined during pregnancy, and iodine
supplementation made no significant difference. Antibody levels were
lower in late pregnancy, with no differences between groups.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
However, in the current study we found no significant increase in the prevalence, severity, or duration of PPTD when 150 µg iodine were given to TPO-Ab-positive women during pregnancy only or during pregnancy and the postpartum period. Iodine supplementation induced minor late pregnancy differences in thyroid function similar to those found in previous studies of pregnant women in areas of mild to moderate iodine deficiency (10, 18). These variations probably had little clinical significance.
Two randomized studies have previously focused on iodine supplementation and its impact on the prevalence and severity of PPTD (9, 19). Jansson et al. hypothesized, based on their observation of low iodine excretion during the hypothyroid phase of PPTD (14), that iodine deficiency might worsen this phase of PPTD. In a subsequent study performed in Sweden, an iodine-sufficient area, they randomized 58 TPO-positive women to receive no supplement, 0.1 mg L-T4, or 150 µg iodine/day during the postpartum period (9). Contrary to their hypothesis, they found an aggravation of the hypothyroid phase of PPTD in the iodine group. The frequency of PPTD was not different from those in the control and L-T4 groups. They recommended that iodine should not be given during the postpartum period.
Reinhard and colleagues addressed the efficacy and safety of iodine given during the postpartum period in an area of mild iodine deficiency (Germany) (19). They randomized 70 women with no previous thyroid disease and no iodine supplementation during pregnancy to receive either 50 or 250 µg iodine/day in the postpartum period (8 months). The study did not include a control group. Thyroid dysfunction developed in 6 women in the group receiving 50 µg iodine and in 5 women in the group receiving 250 µg. Seven had subclinical and 4 had manifest thyroid dysfunction, and 8 of these were either TPO-Ab or Tg-Ab positive. They concluded, on the basis of a prevalence of manifest thyroid dysfunction of 5.7% (4 of 70), that iodine supplementation, up to 250 µg/day, in the postpartum period was safe.
The present study supports that iodine supplementation during pregnancy and postpartum is safe, even in women with circulating TPO-Ab. We did not confirm that women given iodine supplementation during the postpartum period had a more severe hypothyroid phase of PPTD, as found by Kämpe et al. (9). On the contrary, we found that the hypothyroid phase tended to be less severe in women receiving iodine postpartum [significantly higher T4 (P = 0.03) and lower TSH (P = 0.04) in the +/+ group compared to the -/- group, with no significant difference in free T4]. As our study environment is more iodine deficient, the difference in findings could be a matter of substrate availability for hormone synthesis during the period of thyroid regeneration. Several studies have demonstrated high urinary losses of iodine during the previous phase of thyroiditis (14, 15).
The prevalence of TPO-Ab-positive women in early pregnancy (9.1%) and the overall prevalence of PPTD (55%) in TPO-positive women found in the current study are in accordance with general accepted findings (20, 21, 22). In Denmark, Lervang et al. (23) and Rasmussen et al. (24) previously reported population-based prevalences of PPTD of 3.9 and 3.3, respectively. In general, the prevalence seems to be slightly lower in areas with mild iodine deficiency, such as Denmark (23, 24), than in iodine-sufficient areas, such as the U.S., Canada, Japan, and Sweden (25, 26, 27, 28). However, the influence of dietary iodine intake on the prevalence of PPTD is not clear, and there are considerable differences in selections of study populations and follow-up (22).
The association between TPO-Ab level and the frequency and severity of PPTD found in this study is evident from other studies as well (25, 29, 30). In women with hyperthyroidism only antibody levels were the same as those in euthyroid women, and thyroid dysfunction was mild. Lervang et al. (23) and subsequently other researchers (31, 32, 33) reported lower TPO-Ab levels or lower frequency of individuals with detectable TPO-Ab when only a hyperthyroid phase was present. Kuijpens et al. suggested that PPTD in some women with hyperthyroidism only had a different nonautoimmune pathogenesis (33). Our data support that hyperthyroidism only and biphasic PPTD are different degrees of the same abnormality, with low levels of autoimmunity and partial thyroid destruction in the hyperthyroidism only group and more severe affection in the biphasic group, with hyperthyroidism followed by hypothyroidism.
Although symptoms of PPTD may be modest, this is a group of women with increased risk of more permanent thyroid disease, especially hypothyroidism (34), as also evident from the present study. Furthermore, maternal TPO-Ab during pregnancy is associated with postpartum depressive symptoms and impaired child development (31, 35, 36, 37), and new data suggest that low FT4 in the mother in the first trimester of pregnancy is associated with impaired psychomotor development in infancy (38) and a lower IQ in the child when evaluated at school age (39). As a consequence, implementation of routine screening for TPO-Ab in early pregnancy is a topical issue and a matter of current international discussion (20, 40, 41, 42).
In conclusion, iodine supplementation (150 µg) during pregnancy and the postpartum period to TPO-Ab-positive women living in an area with mild to moderate iodine deficiency did not induce or worsen PPTD. This study supports that screening for TPO-Ab in early pregnancy can predict women at high risk of developing PPTD, but further studies, including cost-benefit analyses, are needed before consensus about this issue can be made.
|
Acknowledgments |
|---|
Received December 15, 1999.
Revised February 25, 2000.
Accepted March 30, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. B. Zimmermann Iodine Deficiency Endocr. Rev., June 1, 2009; 30(4): 376 - 408. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B Zimmermann Iodine deficiency in pregnancy and the effects of maternal iodine supplementation on the offspring: a review Am. J. Clinical Nutrition, February 1, 2009; 89(2): 668S - 672S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Negro, G. Greco, T. Mangieri, A. Pezzarossa, D. Dazzi, and H. Hassan The Influence of Selenium Supplementation on Postpartum Thyroid Status in Pregnant Women with Thyroid Peroxidase Autoantibodies J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1263 - 1268. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. F. Muller, H. A. Drexhage, and A. Berghout Postpartum Thyroiditis and Autoimmune Thyroiditis in Women of Childbearing Age: Recent Insights and Consequences for Antenatal and Postnatal Care Endocr. Rev., October 1, 2001; 22(5): 605 - 630. [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 |
