| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Medicine, Medical Biochemistry (R.J.), Blood Transfusion Services (C.D.), and Radiology (H.A.), University of Wales College of Medicine, Cardiff, United Kingdom CF83 2WW
Address all correspondence and requests for reprints to: Dr. L. D. K. E. Premawardhana, Department of Medicine, Caerphilly Miner’s Hospital, St. Martin’s Road, Caerphilly, Mid Glamorgan, United Kingdom CF83 2WW.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The natural history of antithyroid antibodies in subjects with or without autoimmune thyroid disease is unclear. Some reports suggest no change (8, 9, 10), and some find a decrease (11, 12) in the prevalence of antibodies during long term follow-up. A recent population survey in randomly selected individuals reports an increase in prevalence (7). Hitherto, there have been no reports of the variability of TPOAb levels in subjects with PPTD during long-term follow-up.
Thyroid ultrasound hypoechogenicity correlates well with lymphocytic infiltration of the thyroid gland and thyroid dysfunction in women with PPTD (13). The value of this technique in the diagnosis of established autoimmune destruction in patients with PPTD has been demonstrated (3).
TPOAb and thyroid ultrasound hypoechogenicity are surrogate markers of autoimmune thyroid destruction. The prevalence and behavior of such abnormalities during long term follow-up may provide further information on the natural history of postpartum thyroiditis. As far as we are aware, this approach has not been undertaken by previous investigators. However, a previous study demonstrated the usefulness of microsomal antibody titers at delivery in the prediction of early hypothyroidism requiring T4 replacement therapy, but concluded that antibody titers were not useful in the prediction of late thyroid dysfunction (14). We, therefore, describe the prevalence of thyroid dysfunction, the levels and changes in thyroid peroxidase antibodies, and the prevalence and progression of abnormalities in thyroid U/S morphology during long term follow-up.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The Caerphilly postpartum thyroiditis project has been described previously (4). Briefly, 1248 pregnant women were screened at 16 weeks gestation, and 145 TPOAb-positive subjects and 229 TPOAb-negative age-matched subjects were recruited. At follow-up, 98 TPOAb-positive women [48 with PPTD (group 1) and 50 without PPTD (group 2)] and 70 TPOAb-negative control women (group 3) remained in the study. The median (range) length of follow-up was 77 (61–134) months for group 1, 81 (63–139) months for group 2, and 77 (66–140) months for group 3. Although a follow-up rate of 68% for TPOAb-positive subjects (groups 1 and 2) was satisfactory, a rate of 31% (over a median follow-up period of 77 months; range, 66–140) for TPOAb-negative subjects (group 3) was less so. The subjects in this study were contacted individually by letter and telephone. The low response rate in group 3 was due to a combination of factors, including residence outside the area, unrelated illness, and problems with regard to attendance at clinics. However, a comparison of subjects who were available for follow-up with those who did not attend failed to show any significant differences with regard to age, parity at time of index pregnancy, and socio-demographic status. The local ethics committee granted approval for the study.
Follow-up visit
Each subject was seen at a special follow-up clinic by an endocrine physician (J.H.L., L.D.K.E.P., and F.A.). The examining physicians were not aware of the TPO antibody or thyroid status of the subjects postpartum. Blood was taken for free T3 (FT3), free T4 (FT4), TSH, and thyroid antibody analysis. U/S scans were performed by a radiologist (H.A.), who was unaware of the past and present clinical, biochemical. or immunological status of the subject, based on a previously designed protocol, and results were recorded as previously described (3). A cohort of 16 subjects (from group 1) who had developed permanent hypothyroidism at 9 months after delivery and who were receiving T4 replacement were invited to discontinue their medication and were seen 6 weeks later to assess clinically and biochemically the need for continuation of therapy. One further subject from group 2 who was given T4 replacement therapy by her family physician after the first postpartum year had ended was similarly assessed.
Assays
TPOAb. TPOAb were measured by enzyme-linked immunosorbent assay (15) standardized against NIBSC 66/387, the antithyroid microsome serum (National Institute for Biological Standards and Control, Holly Hill, London, UK; normal, <19.4 kIU/L). The intra- and interassay variation for TPOAb was 4.9% (at a mean of 155 kIU/L) and 7.6% (at a mean of 149 kIU/L), respectively. The levels quoted are the mean of quadruplicate measurements.
FT4, FT3, and TSH. Serum levels of FT4, FT3, and TSH were measured on an automated immunoassay analyzer, the ACS-180 Plus (Chiron Corp., Halstead, UK). The FT4 and FT3 were competitive labeled antibody assays using an acridinium ester as label and paramagnetic particles as solid phase. TSH was a two-site immunochemiluminometric assay. Normal values were as follows: FT4, 9.8–23 pmol/L; FT3, 4–6.8 pmol/L; and TSH, 0.5–5.2 mU/L. The between-batch precisions of the assays were as follows: FT4: mean, 13.6 pmol/L; coefficient of variation (CV), 4%; FT3: mean, 5.27 pmol/L; CV, 4.85%; and TSH: mean, 4.89 mU/L; CV, 7.56%.
Human leukocyte antigen (HLA) typing
HLA analysis for class II antigens was available in 152 subjects (group 1, n = 43; group 2, n = 46; group 3, n = 63) as previously described (16).
Thyroid ultrasonography
The thyroid ultrasound scan methods have been described previously (3). The thyroid parenchyma was regarded as normal (grade 0) when it was uniformly hyperechoic relative to muscle. Mild thyroiditis was diagnosed when there were small and sparse focal areas of hypoechogenicity and intervening normally echogenic thyroid parenchyma (grade 1). Moderate thyroiditis was diagnosed when there were more widespread focally hypoechogenic areas within the thyroid but with preservation of some normally echogenic gland (grade 2). Advanced thyroiditis was diagnosed when the thyroid was diffusely hypoechogenic or when hypoechogenic areas were interspersed with areas of abnormally high echogenicity to produce a mixed echopattern (grade 3). For the purposes of this study, the presence of solitary nodules or cysts was disregarded in terms of classification of the background thyroid parenchyma. Subjects had between 1–12 thyroid U/S scans during the first postpartum year and an additional scan at the time of the follow-up visit. The highest U/S scan grade (i.e. the scan showing the most severe thyroiditis) during the postpartum period was compared with the U/S scan grade at follow-up.
Sampling of blood and U/S scans
Sampling of blood. At the time of the follow-up visit, 48 group 1, 50 group 2, and 70 group 3 subjects had blood taken for TPOAb analysis and thyroid function tests. The results of peak postpartum TPOAb levels (the highest levels during the postpartum period) were available for comparison with those obtained at follow-up in 44 group 1 and 44 group 2 subjects.
U/S scans. Thyroid U/S scans were performed at the follow-up visit in 46 of 48 group 1, 48 of 50 group 2 (i.e. 94 of 98 TPOAb-positive), and 66 of 70 group 3 (TPOAb-negative) subjects. Scans at the time of the index illness were available for comparison with follow-up scans on 41 of 98 TPOAb-positive and 27 of 70 TPOAb-negative subjects.
Statistical methods
Data were analyzed using the SPSS, Inc.
(SPSS, Inc., Chicago, IL), statistical package. Groups of
variables were compared using Student’s t test after log
transformation of the data where appropriate. Data tables were analyzed
using the 
2 test.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The median ages of the subjects were as follows: group 1, 35.5 yr; group 2, 35.0 yr; and group 3, 34.0 yr, respectively. The median number of pregnancies was not statistically different in the three groups.
Pregnancies after index pregnancy
The incidence of further pregnancies after the index pregnancy in each of the groups studied was the same, with no statistically significant difference.
Thyroid function at follow-up
Related to early postpartum status in TPOAb-positive (groups 1 and
2) and TPOAb-negative (group 3) subjects. Of the subjects who were
antibody positive, 22 of 48 group 1 subjects (46%), 2 of 50 (4%)
group 2 subjects, and 1 of 70 (1.4%) group 3 subjects had abnormal
thyroid function (defined as a raised TSH level with or without low
FT4/FT3 levels; Table 1
). There was a significant difference in
the prevalence of thyroid dysfunction between the TPOAb-positive and
TPOAb-negative groups and within the former group between group 1 and
group 2 subjects (P < 0.001). At the time of
follow-up, TSH was significantly higher and FT4
levels were lower in group 1 than in group 2 subjects
(P < 0.004), reflecting the higher prevalence of
thyroid dysfunction in this group.
|
Related to stopping replacement therapy in subjects with early permanent hypothyroidism. Of the 16 group 1 subjects who developed early permanent hypothyroidism requiring T4 replacement therapy in the first postpartum year (Hypo E), 9 subjects had a hypothyroid and 7 had a biphasic form of PPTD (therefore, 100% had hypothyroidism as a component of PPTD). Twelve group 1 subjects developed hypothyroidism upon stopping replacement therapy for 6 weeks, requiring reinstitution of T4. This suggested that their early hypothyroidism was permanent. The only subject in group 2 who was receiving replacement therapy was taking 50 µg T4/day, remained euthyroid upon stopping therapy, and did not require further therapy. She probably had experienced a transient form of hypothyroidism after the first postpartum year. Four subjects in the PPTD group declined to stop therapy, as they had developed symptoms of hypothyroidism when treatment was stopped for one of several reasons during the period of follow-up (suggesting that they, too, were permanently hypothyroid). However, their thyroid tests were consistent with adequate replacement.
Related to peak postpartum TSH levels. Peak postpartum TSH (reflecting the severity of thyroid dysfunction) was significantly different between the group of subjects who had thyroid dysfunction at follow-up and the group of subjects whose thyroid function was normal (geometric mean ± 1 SD, 30.85 ± 3.97 vs. 14.63 ± 2.73 mU/L; P < 0.01). Peak postpartum TSH was greater than 20 mU/L in a higher percentage of subjects who had thyroid dysfunction at follow-up (70%) than in those whose thyroid function was normal (16.5%). However, in the former group there was no difference in peak postpartum TSH levels in subjects who developed early thyroid dysfunction compared to those who developed late thyroid dysfunction.
In subjects with normal thyroid function at the end of the first postpartum year. Six of 32 (19%) group 1 subjects who had normal thyroid function at the end of the first postpartum year had abnormal thyroid function at follow-up compared to only 2 of 50 (4%) in group 2 (P < 0.03).
TPOAb levels
TPOAb levels during the postpartum period and at follow-up.
TPOAb levels were measured monthly during the immediate postpartum
period in all three groups of subjects. There was a significant
difference in the peak postpartum levels of TPOAb (defined as the
highest level of TPOAb obtained during monthly sampling in the first
year postpartum) between group 1 vs. group 2 subjects
(P < 0.007). There was also a significant difference
in follow-up TPO antibody levels between group 1 and group 2 subjects
(median periods of 77 and 81 months, respectively, after
delivery; P < 0.03; Table 2).
|
). Thyroid ultrasonography
Thyroid U/S hypoechogenicity at follow-up. A significantly
higher number of TPOAb-positive subjects (60 of 94, 64%) had abnormal
thyroid ultrasound scans at follow-up compared to TPOAb-negative
subjects (9 of 66, 13.6%; P < 0.02; Fig. 1). Similarly, within the TPOAb-positive
group, a significantly higher number of group 1 subjects (35 of 46,
76%) had thyroid U/S abnormalities compared to group 2 subjects (25 of
48, 52%; P < 0.006).
|
). HLA analysis
No frequencies of the HLA-DR and HLA-DW antigens tested were significantly different when the three groups were compared with each other and each group, in turn, was compared with the control group. A single exception to the above was the finding of a low frequency of HLA-DR10 in the group of subjects who developed late hypothyroidism (P < 0.05). There was also an increase in the frequency of HLA-DR11 and HLA-DR12 in group 1, which, however, was not statistically significant.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
TPOAb is a marker for the development of PPTD when measured in early pregnancy (19). The higher peak postpartum TPOAb level in group 1 subjects was a marker of the severity of the postpartum immune rebound phenomenon and resultant immune-mediated thyroid destruction. A persistent elevation of TPOAb occurred in both groups of TPOAb-positive subjects. The natural history of antithyroid antibodies in subjects with and without autoimmune thyroiditis is unclear. Several reports suggest no change in prevalence/titers (8, 9, 10), and some find a decrease (11, 12) during long term follow-up. The effect of T4 replacement therapy was variable (9, 10, 12). A recent study of randomly selected individuals in the north of England showed an increase in prevalence of antithyroid antibodies from 9% to 26% in women and from 2% to 9% in men during a 20-yr follow-up (7). The current study in TPOAb-positive postpartum women demonstrated a persistent elevation in TPOAb levels after an initial postpartum rise. Although there was a tendency toward a reduction in levels, this did not reach statistical significance. A clear cut-off level of TPOAb predictive of long-term outcome was not evident in this study.
In our subjects, destructive thyroiditis, as evidenced by U/S hypoechogenicity, was maximal at the time of postpartum thyroiditis, with gradual improvement to a lesser grade over the period of follow-up. However, persistence of hypoechogenicity in the majority of group 1 subjects (significantly different prevalence when compared to group 2) would indicate an on-going process of destructive thyroiditis. The value of thyroid ultrasound hypoechogenicity as a surrogate marker of autoimmune thyroiditis has been demonstrated (20, 21). However, its use in the follow-up of PPTD is not well documented (3, 22). We are unable on the basis of this study to comment on its value as a predictor of long term thyroid dysfunction in view of the small numbers available for study, but such data as there are suggest no clear-cut predictive pattern.
Studies from Canada (23, 24) and Scandinavia (25) reported the association of PPTD and HLA-DR3, -4, and -5. Further evidence of HLA associations were demonstrated in Japanese subjects with hypothyroid PPTD and permanent hypothyroidism after PPTD (HLA-DR9) (6). More recently, Kologlu et al. described an association of a combination of HLA-A1,B8,DR3 and HLA-B8,DR3 in women who had PPTD (26). We have shown a weak association between a reduced frequency of HLA-DR15 and DQ6 and an increased frequency of HLA-DR5 and DQ7 and PPTD (16). However, in our previous follow-up study, we were not able to demonstrate a link between HLA and PPTD (5). The current follow-up study in a larger group of subjects confirmed this lack of association between HLA antigens and long-term outcome after PPTD. The clinical significance of the low frequency of HLA-DR10 in subjects with late thyroid dysfunction is unclear. The number of subjects was too small for any meaningful comparisons.
Finally, the current study makes it clear that the effects of PPTD are not confined to the immediate postpartum period, and it underlines the need for long term surveillance of TPOAb-positive subjects who develop PPTD (a relative risk of developing long-term thyroid dysfunction of 32 compared to TPOAb-negative subjects). The early development of permanent hypothyroidism in one third of them and the tendency for a further cohort to develop thyroid dysfunction over several years (a relative risk of 12.9 for subjects who developed PPTD, but were euthyroid during the first postpartum year), make regular follow-up necessary. There is also the risk of developing PPTD after subsequent pregnancies (27). Long term follow-up of TPOAb-positive subjects who do not develop PPTD (a relative risk of 2.8 of developing thyroid dysfunction compared to TPOAb-negative subjects) does not seem to be as important, although they too may develop PPTD after subsequent pregnancies.
|
Acknowledgments |
|---|
Received March 3, 1999.
Revised July 13, 1999.
Revised August 30, 1999.
Accepted September 9, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. Dosiou, G. D Sanders, S. S Araki, and L. M Crapo Screening pregnant women for autoimmune thyroid disease: a cost-effectiveness analysis. Eur. J. Endocrinol., June 1, 2008; 158(6): 841 - 851. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Subsection Reports J. Clin. Endocrinol. Metab., August 1, 2007; 92(8_suppl): s8 - s47. [Full Text] [PDF]
|
|
|
|
|
|
M. Abalovich, N. Amino, L. A. Barbour, R. H. Cobin, L. J. De Groot, D. Glinoer, S. J. Mandel, and A. Stagnaro-Green Management of Thyroid Dysfunction during Pregnancy and Postpartum: An Endocrine Society Clinical Practice Guideline J. Clin. Endocrinol. Metab., August 1, 2007; 92(8_suppl): s1 - s47. [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]
|
|
|
|
 |
|
 B. M. Casey and K. J. Leveno Thyroid disease in pregnancy. Obstet. Gynecol., November 1, 2006; 108(5): 1283 - 1292. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L D K E Premawardhana and J H Lazarus Management of thyroid disorders. Postgrad. Med. J., September 1, 2006; 82(971): 552 - 558. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F Azizi The occurrence of permanent thyroid failure in patients with subclinical postpartum thyroiditis Eur. J. Endocrinol., September 1, 2005; 153(3): 367 - 371. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J H Lazarus and L D K E Premawardhana BEST PRACTICE NO 184 Screening for thyroid disease in pregnancy J. Clin. Pathol., May 1, 2005; 58(5): 449 - 452. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. N. Pearce, A. P. Farwell, and L. E. Braverman Thyroiditis N. Engl. J. Med., June 26, 2003; 348(26): 2646 - 2655. [Full Text] [PDF]
|
|
|
|
 |
|
 O.E. Okosieme, A.B. Parkes, B. McCullough, D. Doukidis, B.P. Morgan, C.J. Richards, and J.H. Lazarus Complement activation in postpartum thyroiditis QJM, March 1, 2002; 95(3): 173 - 179. [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 |
