Werner & Ingbar's The Thyroid: A Fundamental & Clinical Text, 9th Edition

27.Sporadic and Postpartum Thyroiditis

John H. Lazarus

Sporadic thyroiditis, also known as silent thyroiditis, and postpartum thyroiditis are important causes of transient thyrotoxicosis as well as transient hypothyroidism. Pathologically, both conditions are characterized by chronic lymphocytic infiltration, and the term “destructive thyroiditis” is commonly used to describe the inflammation resulting in thyroid dysfunction (1). The thyrotoxicosis is caused by rapidly progressive tissue injury, followed by the release of large amounts of thyroid hormone into the circulation. While thyroid pain and tenderness are characteristic features of subacute thyroiditis, they are rare in sporadic or postpartum thyroiditis. The clinical hallmarks of destruction-induced thyrotoxicosis are abrupt onset of thyrotoxic symptoms, sometimes accompanied by the development of a goiter or enlargement of a preexisting goiter. Laboratory features, in addition to those that are diagnostic of thyrotoxicosis, include a low thyroid radioactive iodine uptake and high serum thyroglobulin (Tg) concentrations. Both these conditions are associated with the presence of thyroid antibodies, usually against thyroid peroxidase, and hypothyroidism, due to continuing immune activity, often follows the thyrotoxic phase. Because the destructive process is focal or self-limited, recovery to euthyroidism usually occurs. However, both sporadic and postpartum thyroiditis may evolve into permanent hypothyroidism either at the end of the acute episode or after years of slowly declining thyroid function.

SPORADIC (SILENT) THYROIDITIS

Sporadic (painless) thyroiditis associated with transient thyrotoxicosis was identified with increasing frequency some 25 years ago but now seems to be waning in frequency. The terms “silent thyroiditis” and “painless thyroiditis” (1,2,3,4,5,6) are most often used to describe this syndrome, although other terms have been used (7). Silent thyroiditis was initially described as a painless form of subacute thyroiditis (a disorder dominated by thyroid pain and tenderness) because of its similar clinical course (5,6); thyroid biopsy has shown it to be a form of lymphocytic thyroiditis (8,9,10), similar to, although usually less extensive than, that found in chronic autoimmune thyroiditis and indistinguishable from postpartum thyroiditis. The term “sporadic thyroiditis” is used to define this form of lymphocytic thyroiditis seen outside of the postpartum period. The condition follows a self-limited course of a few weeks to several months, and transient hypothyroidism often occurs during recovery. Multiple episodes may occur in the same person (10).

INCIDENCE

Silent thyroiditis has been reported in the United States, Europe, Canada, India, and Japan. Its recognition increased in the 1970s such that in some areas (e.g., Wisconsin) it accounted for as many as 23% of all cases of thyrotoxicosis (10). This high relative frequency of silent thyroiditis as a cause of thyrotoxicosis has not been reported in other parts of the United States and Europe, and may have been due in part to ingestion of ground beef contaminated with thyroid tissue, a form of iatrogenic thyrotoxicosis (see later in the chapter). Silent thyroiditis accounted for 6% of 100 consecutively referred thyrotoxic patients in Toronto but no cases in a similar population in Cardiff (11), and comprised less than 5% of all cases of thyrotoxicosis in Philadelphia (12), New York (13), and coastal Virginia (6) but 15% in Texas (6). The condition was a rare finding on the East and West Coasts of the United States, as well as in Europe and Argentina, but common around the Great Lakes in the United States and Canada (14). An occasional cluster of cases has been reported, for example, in association with subacute thyroiditis in Connecticut (15). The variable incidence may be due to variation in ascertainment rate as well as the fact that the transient and painless nature of the illness and its symptoms may be attributed to a “flu-like” illness by those unfamiliar with the disease. Some asymptomatic patients are also found to have silent thyroiditis by routine testing. Two epidemics of thyrotoxicosis in the Midwest believed to be due to silent thyroiditis proved to be due to the contamination of ground beef with thyroid tissue (16,17).

The affected patients are usually white or Asian. Women predominate in a ratio of 1.5 to 2:1, which is a much lower ratio than reported for almost all other types of thyroid disease, for which the ratios range from 3 to 10:1. Most patients are in the third through sixth decade of life, but patients as young as 5 years and as old as 93 years have been reported (10).

ETIOLOGY

Multiple reports describing only lymphocytic thyroiditis in biopsy specimens (8,9,10) disprove suggestions that the condition is a silent form of subacute thyroiditis. A search for antibodies to influenza viruses A and B, parainfluenza viruses types 1, 2, and 3, adenovirus, respiratory syncytial virus, mumps virus, measles virus, and coxsackie viruses types 1 through 6 in 18 patients with silent thyroiditis revealed only 1 patient with a significant rise in antibody titer during the course of the disease (10). No positive cultures for viruses or bacteria or electron microscopic evidence of viral inclusion bodies have been reported. A history of an illness such as an upper respiratory infection is unusual, although infection with rubella virus was implicated in one case (18), and an unidentified antecedent infection or exposure to antigen causing a transient increase in serum IgM concentration was reported in another (19). A seasonal cluster in summer and late autumn, the simultaneous occurrence of subacute thyroiditis and silent thyroiditis in a wife and husband (20), and the occurrence, in a short period of time, in five nursery school coworkers (21) suggest an infectious agent. However, viral serologic studies of acute and convalescent serum in the last group were negative.

Although most patients have a partial or complete remission, follow-up study for 1 to 10 years has shown the persistence of or later development of thyroid autoantibodies, thyroid enlargement, or permanent hypothyroidism in about half of the patients.

A substantial percentage of patients who have silent thyroiditis have personal or family histories of autoimmune thyroid disease. Histocompatibility antigen (HLA) haplotype studies show an increased frequency of HLA-DR3 and DR5 in patients with silent thyroiditis (22,23). Intrathyroidal T-cell phenotypes are similar in silent thyroiditis and chronic lymphocytic thyroiditis (24). Taken together, these facts suggest that silent thyroiditis is an early and unusual presentation of chronic lymphocytic thyroiditis, with an unknown factor causing the onset or exacerbation of the destructive process. Variant or atypical forms characterized by thyroid pain or tenderness similar to subacute thyroiditis are recognized only when biopsies are done (25,26).

Several factors have been suggested to be the initiating event in silent thyroiditis (Table 27.1). An increase in serum interleukin-6, an inflammatory mediator, has been found in patients with amiodarone-induced destructive thyroiditis with thyrotoxicosis (27). Thyroid iodine content has been found to be normal or slightly decreased during the thyrotoxic phase of silent thyroiditis (48), and urinary iodide excretion is increased during this phase but is decreased during recovery (10). This increase in urinary iodide reflects release of iodide and iodinated compounds from the damaged thyroid. Iodine-induced thyrotoxicosis and silent thyroiditis are sometimes indistinguishable, but the few patients with the former who have had thyroid biopsies have not had lymphocytic thyroiditis. Excessive iodine intake has been proposed as an aggravating or inciting factor because of its effects on lymphocytic thyroiditis in animals and on human thyroid disease. Other possible associated initiating factors include drugs, such as lithium (29), and immune disorders. Interestingly, in patients with hepatitis B treated with interferon-γ, no thyroid dysfunction or thyroid autoimmunity was observed (49). In contrast, patients with chronic active hepatitis C treated with interferon-α may develop thyrotoxicosis secondary to silent thyroiditis (29). Simple palpation has been reported to cause a multifocal granulomatous folliculitis without serum thyroid hormone concentration abnormalities (46). The thyrotoxicosis and thyroiditis noted in patients recovering from parathyroid surgery for parathyroid adenoma may be due to trauma to the thyroid gland during neck exploration for parathyroid surgery (47). Thyrotropin (TSH) receptor–stimulating antibodies (50,51) and TSH receptor–blocking antibodies may occur in patients with silent thyroiditis (52). Furthermore, there is evidence that antithyroid peroxidase antibodies are associated with the development of the hypothyroid phase (53) and also that both types of TSH-receptor antibodies are associated with transient hypothyroidism in this disease (54).

The presence of these antibodies may be related to more severe immunological damage in the thyroid resulting in hypothyroidism.

TABLE 27.1. ETIOLOGICAL AND/OR INITIATING FACTORS IN SPORADIC THYROIDITIS


Infection

Rarely confirmed

Drugs

Amiodarone (27)

 Lithium (28,29,30,30a)

Interleukin-2 (31,32)

 Interferon alpha (33,34,35,36,37)

 Granulocyte colony-stimulating factor (38)

Leuprolide acetate (39)

Association with Other

Rheumatoid arthritis (40)

 Autoimmune Diseases

Systemic lupus erythematosus (41)

Graves' disease (42,43)

 Systemic sclerosis (44)

 Thymoma (45)

Local Factors

Palpation thyroiditis (46)

Following parathyroid surgery (47)


PATHOLOGY

Thyroid tissue (Fig. 27.1) obtained by needle biopsy or surgery during the thyrotoxic phase of silent thyroiditis shows focal or diffuse lymphocytic infiltration (9,10,24) consistent with lymphocytic thyroiditis. Some thyroid follicles are disrupted or collapsed and contain intrafollicular macrophages, while adjacent follicles may be normal. Fibrosis is usually minimal but may be extensive. Lymphoid follicles are found in about half of patients, whereas Hürthle cells are sparse or absent. A few multinucleated giant cells within and around disrupted thyroid follicles may be seen in the biopsy specimens of some patients with painless thyroiditis. The pathologic changes of silent thyroiditis compared to those of chronic lymphocytic thyroiditis cannot be distinguished when unidentified biopsy specimens from patients with either disease are compared (10). While there is more extensive follicular disruption in silent thyroiditis, the pathological features are usually more extensive in chronic lymphocytic thyroiditis. Fine-needle aspiration biopsies are similar in the two conditions, the specimens containing lymphoid cells, thyroid follicular cells with Hürthle-cell changes, and a few multinucleated giant cells (31).

FIGURE 27.1. Histopathalogic findings in silent thyroiditis. A: Focal lymphocytic infiltration with partially or completely collapsed thyroid follicles. The thyroid follicular cells appear normal. B: Extensive lymphocytic infiltration with collapsed thyroid follicles and follicles containing mononuclear cells (Hematoxylin and eosin, 100× magnification).

Thyroid tissue obtained during the hypothyroid or early recovery phase of silent thyroiditis usually shows mild lymphocytic thyroiditis and regenerating thyroid follicles that contain little colloid. Thyroid biopsies done after 6 months to 3 years showed persistent mild lymphocytic thyroiditis in some patients and were normal in others (24).

CLINICAL FEATURES

The onset and severity of silent thyroiditis are variable. In one series, about 8% of patients were asymptomatic and were found by routine thyroid function testing (10). Thyrotoxicosis is usually mild, resulting in little if any disability. The symptoms and signs are those of thyrotoxicosis in general (including nervousness, fatigue, and tachycardia, all occurring in more than 80) (10) but may include atrial fibrillation, diffuse myalgia, and periodic paralysis (8).

Exophthalmos, localized myxedema, and thyroid acropathy do not occur, although lid lag and lid retraction (signs of increased sympathetic tone) are frequently present. The thyroid is enlarged in 50% to 60% of patients (6), is usually symmetrical, and is rarely more than 2 to 3 times normal. It is usually described as being firmer than normal, and the surface may be slightly irregular. The disease has been found in single nodules and in ectopic thyroid tissue (10). Thyroid pain or tenderness is rare and when present is mild. Fever has been described (55).

COURSE

A schematic representation of the course of silent thyroiditis is shown in Figure 27.2. Most patients present in the thyrotoxic phase, but a few are in the euthyroid or hypothyroid phase. Most patients have had symptoms of thy rotoxicosis for 1 to 4 weeks when the disease is discovered, and these symptoms persist for about 1 to 5 weeks after discovery. About 40% then have a hypothyroid phase, which usually lasts between 4 to 16 weeks but occasionally persists indefinitely.

FIGURE 27.2. Schematic representation of the changes in serum thyroxine (T4), triiodothyronine (T3), and thyrotropin (TSH) concentrations, and 24-hour thyroid 131I uptake during the course of silent thyroiditis and subacute thyroiditis. The hatched areas represent the normal ranges. The serum T4 and T3lines separate during the early part of the euthyroid phase to show that only about 30% of patients subsequently develop hypothyroidism. In the recovery phase, the serum T3 line separates to show that about 15% of patients have transient elevations in serum T3 concentrations during this phase. To convert serum T4 values to nmol/L, multiply by 12.87; to convert serum T3 values to nmol/L, multiply by 0.015. (From Woolf PD. Transient painless thyroiditis with hyperthyroidism: a variant of lymphocytic thyroiditis? Endocr Rev 1980;1: 411, and The Endocrine Society, modified with permission.)

LABORATORY FINDINGS

Thyroid function tests vary substantially, depending on the stage of the disease (6,10,52), and relate directly to the pathologic state of the thyroid gland.

THYROTOXIC PHASE

With inflammation and damage to the thyroid follicles, Tg proteolysis is activated, and its products leak into the circulation, resulting in high serum total and free T4 and T3 concentrations. The T4 to T3 ratio is higher than in Graves' thyrotoxicosis. Serum TSH is low and does not increase in response to thyrotropin-releasing hormone (TRH). Other iodinated materials, such as Tg and iodinated albumin, are found in the serum in increased quantities; the concentrations of the former may be as high as 400 ng/mL. As a result of both thyroid follicular cell damage and decreased TSH secretion, the thyroid follicular cells are unable to transport iodine, so that the thyroid radioiodine uptake falls to low values. Serum inorganic iodide concentrations are slightly increased and urinary iodide excretion is increased two- to fivefold because of the leakage of iodide from the damaged thyroid (6). Anti–thyroid peroxidase (anti-TPO) antibodies are high in about 60% and anti-Tg antibodies in 25% of patients, but TSH receptor antibodies are rarely found.

The erythrocyte sedimentation rate, total and differential white blood cell counts, and serum protein concentrations are high in about half of patients. Serum interleukin-12 concentrations are high, suggesting that a T helper 1 (Th1) type immune response occurs in this condition (56).

HYPOTHYROID AND RECOVERY PHASES

As the thyrotoxic phase abates, the serum T4 and T3 concentrations fall into the normal range and either remain there or, after 1 to 6 weeks in about 40% of patients, fall to subnormal concentrations (Fig. 27.2). In the interval when the serum T4 and T3 concentrations are normal, serum TSH concentrations and thyroid 131I radioiodine uptake values remain low. After 2 to 4 weeks, if the patient remains euthyroid, or toward the end of the hypothyroid phase, these latter two tests become normal or elevated. Within another 1 to 3 weeks, serum T4 and T3 concentrations return to normal if hypothyroidism had occurred earlier. The hypothyroid interval lasts 4 to 10 weeks, occasionally longer. Permanent hypothyroidism occurs in less than 5% of patients. Antithyroid antibody titers usually reach their highest levels during the hypothyroid phase, then decrease and disappear in about half of patients (10). In the other half, the antibody titers fall but remain measurable for many years. The intrathyroidal iodine content is decreased by 50% to 70% 1 to 3 months after the onset of thyroiditis and still decreased by 20% to 30% at 10 to 12 months (48). Serum Tg concentrations gradually fall during the recovery phase but are often still slightly elevated one to two years later, indicating persistent thyroid inflammation (48). Urinary iodide excretion decreases markedly during the recovery phase because the thyroid is relatively depleted of its iodine and thyroid hormone stores (10).

DIAGNOSIS

The main features are a mild to moderate degree of thyrotoxicosis with corresponding elevation in serum T4 and T3 concentrations, a low thyroid 123I uptake, slight thyroid enlargement (in 50% to 60% of patients), no thyroid tenderness, and no history of therapy with thyroid hormone. A fine-needle or core-needle biopsy of the thyroid gland showing lymphocytic thyroiditis confirms the diagnosis, but is rarely needed. As many patients are thought to have Graves' thyrotoxicosis, TSH-receptor antibody measurement will further differentiate the two conditions in most but possibly not every case, because a few patients with silent thyroiditis have those antibodies in their serum (52,57). A normal red blood cell zinc concentration has been claimed also to differentiate accurately silent thyroiditis from Graves' disease (58). Although a low thyroid radioiodine uptake value is most useful in identifying silent thyroiditis, it is important to rule out iodine contamination due to any cause resulting in the suppressed uptake.

Iodine-induced thyrotoxicosis can easily be confused with silent thyroiditis. It has been recognized in patients with numerous preexisting thyroid diseases, such as a solitary nodule or nontoxic diffuse or multinodular goiter, as well as in patients without any history of thyroid disease. The thyroid 123I uptake in such patients is usually low. Rare diseases that cause thyrotoxicosis with low thyroid radioiodine uptake values are struma ovarii (if the struma is a thyroid adenoma), differentiated carcinoma of the thyroid, and thyrotoxicosis factitia. The most difficult entity to recognize is thyrotoxicosis factitia. Surreptitious intake of thyroid hormone should always be suspected in a patient with thyrotoxicosis with low radioiodine uptake and no goiter. It may be differentiated from silent thyroiditis by history, if of long duration; patient behavior, often histrionic or denial; the absence of thyroid antibodies; and low serum Tg concentrations.

Ultrasonography usually reveals a slight increase in thyroid volume with decreased echogenicity. This study may be useful in diagnosis and further monitoring of these patients (4).

TREATMENT

Patients with silent thyroiditis may receive inappropriate treatment, such as an antithyroid drug, because of failure to recognize the disorder. Most often, the thyrotoxicosis is mild and not particularly bothersome, so patients need only be counseled about the disease and reassured that their symptoms will subside in a few weeks. They should be reminded that a short period of hypothyroidism may occur during recovery and that about 10% will have recurrent episodes in the future. When symptoms of thyrotoxicosis are troublesome, the patient may be given β-adrenergic antagonist, sedative, or tranquilizer therapy. Therapy with thionamide antithyroid drugs is inappropriate because increased thyroid hormone biosynthesis is not the cause of thyrotoxicosis in painless thyroiditis. Indeed, propylthiouracil administration does not change the course or severity of thyrotoxicosis in patients with silent thyroiditis (59). The few patients in whom thyrotoxicosis is more disabling may benefit from antiinflammatory therapy before the illness runs it natural course. Prednisone reduces the inflammatory process, causing decreases in both thyroid size and serum T4 and T3 concentrations, which may decline to the normal range within 7 to 10 days (59). The optimal starting dose of prednisone or duration of therapy has not been established, but therapy similar to that used successfully for many years in subacute thyroiditis works well. Initial therapy with 40 to 60 mg/day in single or divided doses and gradual reduction of the daily dose by 7.5 to 15 mg/week for a 4-week course is usually adequate.

Ipodate therapy, which also rapidly reduces serum T3 concentrations is also an effective therapy in extremely severe cases of thyrotoxicosis (60). In rare patients who have been disabled from recurring episodes of silent thyroiditis, subtotal thyroidectomy has been beneficial (9,59). Ablation of the thyroid with 131I could also be used in these instances, but one must wait until the thyroid 131I uptake has recovered before this therapy can be given.

When hypothyroidism develops in silent thyroiditis, it is usually mild, and the patient can be advised that it will disappear within 4 to 10 weeks. If the symptoms of hypothyroidism are bothersome, sufficient T4 should be given to relieve the hypothyroid symptoms but not so much as to decrease serum TSH to below normal, since continued TSH secretion may hasten recovery. The therapy can then be slowly withdrawn. In the occasional patient who develops permanent hypothyroidism, the serum T4 concentration falls and the serum TSH concentration rises when T4 therapy is withdrawn. If hypothyroidism lasts for longer than 6 months, it is probably permanent, and no further attempts at withdrawal need be undertaken. All patients should be followed at 1- to 2-year intervals for evidence of goiter or hypothyroidism, because up to half of patients who have an episode of silent thyroiditis eventually develop permanent thyroid disease (61).

POSTPARTUM THYROIDITIS

In 1948 H.E.W. Roberton, a general practitioner in New Zealand, described the occurrence of lassitude and other symptoms of hypothyroidism during the postpartum period (62). These complaints were treated successfully with thyroid extract. The syndrome remained generally unrecognized until the 1970s, when reports from Japan (63) and Canada (64) rediscovered the existence of postpartum thyroiditis (PPT) and recognized the immune nature of the condition (see reviews 65,66,67,68). Postpartum thyroiditis is essentially sporadic thyroiditis in the postpartum period. Several types of thyroid dysfunction that may arise following delivery (Fig. 27.3). The term “postpartum thyroiditis” relates to destructive thyroiditis occurring during the first 12 months postpartum and not to Graves' disease, although the two conditions may occur concurrently.

FIGURE 27.3. Development of thyroid dysfunction in the postpartum period. (From Amino N, Tada H, Hidaka Y. Thyroid disease after pregnancy: post-partum thyroiditis. In: Wass JAH, Shalet, SM, eds. Oxford textbook of endocrinology and diabetes. Oxford:Oxford University Press, 2002:527–532, adapted with permission.)

INCIDENCE

A variable incidence (from 3% to 17%) has been reported worldwide (70) because of wide variations in the number of women studied, the frequency of thyroid assessment, diagnostic criteria employed, and differences in hormone assay methodology. However, there is a general consensus that the disease occurs in 5% to 9% of unselected postpartum women (71,72,73,74,75). Recent studies have confirmed that the incidence and general characteristics of PPT are similar in different areas, e.g., Brazil (75), Iran (76), and Turkey (77). Women with type 1 diabetes have a threefold incidence of PPT compared to nondiabetic women (78). PPT is also more likely to occur in a woman who has had a previous episode (vide infra).

ETIOLOGY

There is abundant evidence that PPT is an immunologically related disease (79). Biopsy of the thyroid shows lymphocytic infiltration similar to that seen in Hashimoto's thyroiditis (80). Immunogenetic studies have demonstrated associations with haplotypes consistent with autoimmune thyroid disease. A higher incidence of HLA-A26, -BW46, and -BW67, together with a significantly lower frequency of HLA-BW62 and -CW7, as well as an increased incidence of HLA-A1 and -B8 in women with this condition, has been noted (81,82). HLA class II antigen associations with autoimmune thyroid disease have shown an association between HLA-DR3, -DR4, and -DR5 and PPT (83,84,85,86). The class III area of the major histocompatibility complex contains the coding for several proteins that are important in the pathogenesis of the autoimmune diseases (viz. tumor necrosis factor, heat shock protein 70, complement). The frequency of the three complement proteins Bf, C4A, and C4B is significantly different in PPT compared with normal (86). Taken together, the immunogenetic findings suggest that PPT may well be similar to Hashimoto's disease.

Gestation and the postpartum period are characterized by fluctuations in the immune response, and PPT is accompanied by a significant elevation of circulating thyroid antibodies. This rapid perturbation of the immune system in the postpartum year contrasts with the relatively stable situation in chronic Hashimoto's thyroiditis (87). Although the antibody response is dramatic, its precise role in the immunopathogenesis of the condition remains to be determined; probably the antibody titer is merely a marker of disease, and the immunological damage is mediated by lymphocyte and complement associated mechanisms. Studies on antibody functional affinity (88) and IgG subclass distribution (89) of the TPO antibodies in PPT suggest that, as in other autoimmune diseases, activation of the complement system may have a role in pathogenesis. Quantitative analysis of the classical complement pathway activation has shown that not only is there activation of the complement system by thyroid-directed autoantibodies (90), but complement activation is related to the extent of the thyroiditis (42) and correlates with the severity of the thyroid dysfunction (91).

During pregnancy, maternal immune reactions are regulated to prevent rejection of the fetal allograft such that the cytokine profile is a T helper 2 (Th2) pattern, which switches back to the Th1 state postpartum (92).

Early postpartum (within 3 months) changes in T-cell subsets have been described similar to those seen in Hashimoto's disease (93). The peripheral T-lymphocyte subset ratio (CD4/CD8) has been shown to be higher in TPO antibody–positive women who developed postpartum thyroid dysfunction compared with similar antibody-positive women who did not (94). Study of lymphocyte populations during and after pregnancy has shown a generalized activation of immune activity at 36-weeks gestation in TPO antibody–positive women who went on to develop postpartum thyroid dysfunction, compared with those who did not; furthermore, the former group had lower serum cortisol concentrations predelivery (95).

Thus, both humoral and cellular immune reactions are involved in the development of postpartum thyroid disease, but the influence of the immunological changes in late gestation and the early postpartum period, which suggest a failure of continuing immunosuppression in late pregnancy, requires further clarification. Recent interest has focused on fetal microchimerism (defined as the presence of fetal cells in maternal tissues during and after pregnancy) as a potential immunomodulatory factor in the development of PPT (96); loss of placental immune suppression postpartum may allow activation of intrathyroidal fetal immune cells. Subsequent immune stimulation would initiate or exacerbate autoimmune thyroid disease, including PPT.

CLINICAL SPECTRUM

PPT with thyroid dysfunction (i.e., postpartum thyroid dysfunction, PPTD) is characterized by an episode of transient thyrotoxicosis followed by transient hypothyroidism. The former presents at about 14 weeks postpartum, followed by transient hypothyroidism at a median of 19 weeks (97). Very occasionally the hypothyroid state is seen first. Not all women manifest both thyroid states, and the thyrotoxic episode may escape detection as it may be of short duration. PPTD is almost always associated with the presence of antithyroid antibodies, usually anti-TPO antibodies that rise in titer after 6-weeks postpartum. Anti-Tg antibodies occur in about 15% and is the sole antibody in less than 5% of cases. However, postpartum thyroid dysfunction has been described in small numbers of women who do not have circulating thyroid antibodies (65,98). There is no evidence that variations in ambient iodine concentrations cause the condition (99).

The clinical course of PPTD is illustrated in Figure 27.4, which also shows the postpartum rise in anti-TPO titer associated with this condition. The thyroid dysfunction that occurs in up to 50% of the antibody-positive women comprises 19% with thyrotoxicosis alone, 49% hypothyroidism alone, and the remaining 32% with both (i.e., biphasic). Although the clinical manifestations of the thyrotoxicosis state are not usually severe, lack of energy and irritability are particularly prominent even in thyroid antibody–positive women who do not develop thyroid dysfunction. In contrast, the symptomatology of the hypothyroid phase may be profound. Many classic hypothyroid symptoms occur before the onset of overt hyperthyroidism and persist even after recovery is seen in hormone levels. The most frequent symptoms are lack of energy, aches and pains, poor memory, dry skin, and cold intolerance. Postpartum thyroiditis can also occur in women receiving T4 therapy before pregnancy (100). Early accounts of noted an anecdotal association of depression with thyroid dysfunction in the postpartum period. Quantitative evaluation of depressive symptomatology shows an increase of mild to moderate depression in thyroid-antibody–positive women even when they remain euthyroid during the postpartum period, and this may present as early as 6-weeks postpartum (101,101a), although other smaller studies did not confirm these findings (102,103).

FIGURE 27.4. Emergence of transient thyrotoxicosis and hypothyroidism postpartum. The patient was found to have antithyroid peroxidase antibodies at 16-weeks gestation and was studied weekly postpartum for 36 weeks.

DIAGNOSIS

Thyrotoxicosis in the postpartum period may be due to a recurrence or the development of new Graves' disease or to the thyrotoxicosis phase of PPT. Despite the strong association between thyroid autoantibodies and the development of PPT (68), only some 50% of TPO Ab-positive women develop PPTD. Nevertheless, inspection of Table 27.2 shows that the presence of TPO-Ab measured at various times during gestation or postpartum has a useful predictive value for thyroid dysfunction (104,105,106,107,108,109,110). The occurrence of PPTD in TPO-Ab-negative women in most studies may relate to ethnic variability in the predilection for PPTD, the sensitivity of TPO-Ab assay used and the timing of TPO-Ab sampling (effects of immune modulation of pregnancy and the postpartum period may affect detection rates of TPO-Ab). A variety of screening strategies was used with screening done at variable times (antenatal, delivery, and postpartum). Screening for microsomal antibodies was the method used in earlier studies, but TPO-Ab was used in later studies. Sensitivity of TPO-Ab measurement in previous studies was relatively low compared with the present, although specificity was higher.

TABLE 27.2. ANALYSIS OF PREDICTIVE VALUE OF THYROID ANTIBODIES IN RELATION TO OCCURRENCE OF POSTPARTUM THYROID DYSFUNCTIONa


Ab

Ab + ve Subjects

Sampling Time–pp (months)

Sensitivity

Specificity

PPV

Reference


Micro

44 (7)

2–5

0.77

0.95

0.52

104

Micro

62 (1)

3

0.89

0.92

0.4

105

Micro

38 (3)

3

0.87

0.97

0.53

106

Micro

41 (*)

Delivery

0.45

0.95

0.63

85

Micro

54 (11)

2–4

0.76

0.91

0.65

85

Micro

69 (8)

5–7

0.86

0.9

0.68

85

Micro

15 (6)

Trimester 3

0.71

0.92

0.52

107

TPO

31 (5)

Trimesters
1/3

0.67

0.93

0.31

108

TPO

55 (31)

6

0.64

0.95

0.64

103

Micro

40 (46)

6

0.46

0.98

0.78

103

Micro Tg

50 (65)

Early pregnancy

1/12pp 0.14

1/12pp 0.88

1/12pp 0.07

 3/12pp 0.37

3/12pp 0.90

3/12pp 0.21

98

Micro

55 (*)

Delivery

 0.73

109

TPO

66 (*)

Trimester 1

-

-

0.55

99

TPO

308 (0)

Ealy pregnancy

1.00

0.62

0.48

110


a Note the variable times at which antibody measurements were performed.

Key: Ab, thyroid antibody; micro, microsomal antibody; pp, postpartum; PPV, positive predictive value; Tg, thyroglobulin antibody; TPO, thyroid peroxidase antibody; (), number of PPTD subjects with negative antibodies; (*), unknown or not included.


As PPTD is a destructive process, thyroidal radioiodine uptake will be very low at early and late times after isotope administration. TSH receptor antibodies are not seen in PPTD unless there is coexisting Graves' disease. Thyrotoxicosis due to PPT is diagnosed by a low serum TSH together with an elevated free T4 or T3, with either set of criteria occurring on more that one occasion. Thyroid hormone concentrations should be normal in the postpartum period. Antibodies to T4 and T3 may cause confusion in diagnosis, but they are infrequent (68).

Postpartum hypothyroidism may be defined as either a serum TSH concentration >3.6 mU/L (i.e., lower than the usual upper limit of the normal reference range) together with a free T4 < 0.6 ngldl (8 pmol/L) or a TSH 10 mU/L or more on one or more occasions. Thyroid ultrasonography has demonstrated diffuse or multifocal echogenicity, reflecting the abnormal thyroid morphology and consistent with the known lymphocytic infiltration of the thyroid (68). When anti-TPO antibodies are borderline during pregnancy, structural and echogenic changes might be of predictive value for the development of PPTD (75). The destructive nature of the thyroiditis is also shown by the increase in urinary iodine excretion in the thyrotoxic phase of the syndrome (68), as well as by an early rise in serum Tg (68). Despite active thyroiditis, there is no rise in interleukin-6 or C-reactive protein in PPTD (110a), but the inflammation may be responsible for a decrease in total transforming growth factor (TGF)-β1 and a concurrent increase in active TGF-β1 (111).

MANAGEMENT

The thyrotoxic phase is relatively asymptomatic, although tachycardia and palpitations may be noted. These will respond to β-adrenoreceptor blocker therapy. Thionamide antithyroid drugs are contraindicated because of the destructive nature of the condition. While the thyrotoxicosis of PPTD always resolves, several long-term studies of the hypothyroid phase have documented persistence of hypothyroidism in 20% to 30% of cases (65). If postpartum hypothyroidism is diagnosed, the patient will respond to an initial dose of 100 µg of T4 and should be subsequently managed as for primary hypothyroidism. It is impossible at this stage to determine which patients will have transient hypothyroidism (and therefore require only short-term T4 treatment) or develop permanent disease. A pragmatic strategy is to treat all postpartum hypothyroid women for 1 year, then to stop T4 and reevaluate thyroid function (112).

Women who have developed transient postpartum hypothyroidism and who have recovered spontaneously or have stopped T4 therapy should have thyroid function checked annually, as a 50% rate of permanent hypothyroidism rate has been noted after 7 years' follow-up. In contrast, thyroid antibody–positive women who remain euthyroid during the first postpartum year require less frequent assessment; their rate of hypothyroidism at 7 years is only 5% (113). Recurrence of transient PPTD was observed in up to 30% to 40% in an early series (114), but in a larger number of women studied there was a 70% chance of recurrent PPTD after a first attack and a 25% risk even if the woman was only anti-TPO-positive without thyroid dysfunction during the first postpartum period. From the foregoing discussion, it is apparent that PPTD occurs in 50% of the 10% of women who may be found to be anti-TPO–positive in early pregnancy. In addition, 30% of all anti-TPO–positive women postpartum will develop depressive symptomatology (independent of thyroid function) compared with about 20% of antibody-negative women. The incidence and morbidity of the thyroid dysfunction and depressive symptomatology suggest that screening for PPTD may be justified. For example, a strategy involving measurement of thyroid antibodies in early gestation at the time of “booking” may alert the clinician to the risk of postpartum problems and encourage early treatment (115).

REFERENCES

1. Dorfman SG, Cooperman MT, Nelson RS, et al. Painless thyroiditis and transient hyperthyroidism without goiter. Ann Intern Med 1977;86:24–28.

2. Gordon M. “Silent” thyroiditis with symptomatic hyperthyroidism in an elderly patient. J Am Geriatr Soc 1978;26:375–377.

3. Hofeldt FD, Weled BJ, Brown JE, et al. “Silent thyroiditis” versus thyrotoxicosis factitia. Minn Med 1976;59:380–382.

4. Miyakawa M, Tsushima T, Onoda N, et al. Thyroid ultrasonography related to clinical and laboratory findings in patients with silent thyroiditis. J Endocrinol Invest 1992;15:289–295.

5. Papapetrou PD, Jackson IMD. Thyrotoxicosis due to “silent” thyroiditis. Lancet 1975;1:361–363.

6. Woolf PD. Transient painless thyroiditis with hyperthyroidism: a variant of lymphocytic thyroiditis? Endocr Rev 1980;4:411–420.

7. Lazarus JH. Silent thyroiditis and subacute thyroiditis. In: Werner & Ingbar's, The thyroid. 7th ed. Philadelphia: JB Lippincott, 1996:577–591.

8. Gluck FB, Nusynowitz MD, Plymate S. Chronic lymphocytic thyroiditis, thyrotoxicosis, and low radioactive iodine uptake. Report of four cases. N Engl J Med 1975;293:624–628.

9. Gorman CA, Duick DS, Woolner LB, et al. Transient hyperthyroidism in patients with lymphocytic thyroiditis. Mayo Clin Proc 1978;53:359–365.

10. Nikolai TF, Brosseau J, Kettrick MA, et al. Lymphocytic thyroiditis with spontaneously resolving hyperthyroidism (silent thyroiditis). Arch Intern Med 1980;140:478–482.

11. Williams I, Ankrett VO, Lazarus JH, et al. Aetiology of hyperthyroidism in Canada and Wales. J Epidemiol Community Health 1983;37:245–248.

12. Schorr AB, Miller JL, Shtasel P, et al. Low incidence of painless thyroiditis in the Philadelphia area. Clin Nucl Med 1986;11: 379–380.

13. Vitug AC, Goldman JM. Silent (painless) thyroiditis: evidence of a geographic variation in frequency. Arch Intern Med 1985;145:473–475.

14. Schneeberg NG. Silent thyroiditis. Arch Intern Med 1983;143: 2214.

15. Dulipsingh L, Ikram Z, Malchoff CD, et al. A cluster of cases of subacute and silent thyroiditis in the northern Connecticut, Greater Hartford area. Conn Med 1998;62:395–397.

16. Hedberg CW, Fishbein DB, Janssen RS, et al. An outbreak of thyrotoxicosis caused by the consumption of bovine thyroid gland in ground beef. N Engl J Med 1987;316:993–998.

17. Kinney JS, Hurwitz ES, Fishbein DB, et al. Community outbreak of thyrotoxicosis: epidemiology, immunogenetic characteristics, and long-term outcome. Am J Med 1988;84:10–18.

18. Nakamura S, Kosaka J, Sugimoto M, et al. Silent thyroiditis following rubella. Endocrinol Jpn 1990;37:79–85.

19. Masuno M, Kosaka J, Makamura S. Silent thyroiditis in an eleven-year-old girl, associated with transient increase in serum IgM and thyroid hormone. Endocrinol Jpn 1991;38:219–222.

20. Morriston J, Caplan RH. Typical and atypical (“silent”) subacute thyroiditis in a wife and husband. Arch Intern Med 1978;138:45–48.

21. Ogura T, Hirakawa S, Suzuki S, et al. Five patients with painless thyroiditis simultaneously developed in a nursery school. Endocrinol Jpn 1980;35:225–230.

22. Farid NR, Hawe BS, Walfish PG. Increased frequency of HLA-DR3 and 5 in the syndromes of painless thyroiditis with transient thyrotoxicosis: evidence for an autoimmune aetiology. Clin Endocrinol (Oxf) 1983;19:699–704.

23. Fein HG, Metz S. Nikolai TF, et al. HLA antigens in thyroiditis: differences between silent and postpartum lymphocytic forms, and comparison with subacute and goitrous autoimmune thyroiditis. In: Walfish PG, Wall J, Volpe R, eds. Autoimmunity and the thyroid. Orlando, FL: Academic Press, 1985:373–375.

24. Mizukami Y, Michigishi T, Hashimoto T, et al. Silent thyroiditis: a histologic and immunohistochemical study. Hum Pathol 1988;19:423–431.

25. Ishihara T, Mori T, Waseda N, et al. Histological, clinical and laboratory findings of acute exacerbation of Hashimoto's thyroiditis—comparison with those of subacute granulomatous thyroiditis. Endocrinol Jpn 1987;34:831–841.

26. Shigemasa C, Ueta Y, Mitani Y, et al. Chronic thyroiditis with painful tender thyroid enlargement and transient thyrotoxicosis. J Clin Endocrinol Metab 1990;70:385–390.

27. Bartalena L, Grasso L, Brogioni S, et al. Serum interleukin-6 in amiodarone-induced thyrotoxicosis. J Clin Endocrinol Metab 1994;78:423–427.

28. Chow CC, Lee S, Shek CC, et al. Lithium-associated transient thyrotoxicosis in 4 Chinese women with autoimmune thyroiditis. Aust N Z J Psychiatry 1993;27:246–253.

29. Miller KK, Daniels GH. Association between lithium use and thyrotoxicosis caused by silent thyroiditis. Clin Endocrinol (Oxf) 2001;55:501–508.

30. Dang AH, Hershman JM. Lithium-associated thyroiditis. Endocr Pract 2002;8:232–236.

30a. Shimizu M, Hirokawa M, Manabe T, et al. Lithium associated autoimmune thyroiditis. J Clin Pathol 1997;50: 172–174.

31. Vassilopoulou-Sellin R, Sella A, Dexeus FH, et al. Acute thyroid dysfunction (thyroiditis) after therapy with interleukin-2. Horm Metab Res 1992;24:434–438.

32. Reid I, Sharpe I, McDevitt J, et al. Thyroid dysfunction can predict response to immunotherapy with interleukin-2 and interferon-2 alpha. Br J Cancer 1991;64:915–918.

33. Lisker-Melman M, Di Biscelglie AM, Usala SJ et al. Development of thyroid disease during therapy of chronic viral hepatitis with interferon alfa. Gastroenterology 1992;102:2155–2160.

34. Kamikubo K, Takami R, Suwa T, et al. Silent thyroiditis developed during alpha-interferon therapy. Am J Med Sci 1993;306:174–176.

35. Kodama T, Katabami S, Kamijo K, et al. Development of transient thyroid disease and reaction during treatment of chronic hepatitis-C with interferon. J Gastroenterol 1994;29:289–292.

36. Roti E, Minelli R, Guiberti T, et al. Multiple changes in thyroid function in patients with chronic active HCV hepatitis treated with recombinant interferon-alpha. Am J Med 1996;101:482–487.

37. Mazziotti G, Sorvillo F, Stomaiuolo G, et al. Temporal relationship between the appearance of thyroid autoantibodies and development of destructive thyroiditis in patients undergoing treatment with two different type-1 interferons for HCV-related chronic hepatitis: a prospective study. J Endocrinol Invest 2002;25:624–630.

38. Hoekman K, Von Blomberg-Van der Flier BME, Wagstaff J, et al. Reversible thyroid dysfunction during treatment with GM-CSF. Lancet 1991;338: 541–542.

39. Kasayama S, Miyake S, Samejima Y. Transient thyrotoxicosis and hypothyroidism following administration of the GnRH agonist leuprolide acetate. Endocr J. 2000;47:783–785.

40. Sakata S, Nagai K, Shibata T, et al. A case of rheumatoid arthritis associated with silent thyroiditis. J Endocrinol Invest 1992;15:377–380.

41. Magaro M, Zoli A, Altomonte L, et al. The association of silent thyroiditis with active systemic lupus erythematosus. Clin Exp Rheumatol 1992;10:67–70.

42. Iitaka M, Ishii J, Ishikawa N, et al. A case of Graves' disease with false hyperthyrotropinemia who developed silent thyroiditis. Endocrinol Jpn 1991;38:667–671.

43. Ho SC, Eng PH, Fok AC, et al. Thyrotoxicosis due to the simultaneous occurrence of silent thyroiditis and Graves' disease. Thyroid 1999;9:1127–1132.

44. Yamamoto M, Fuwa Y, Chimori K, et al. A case of progressive systemic sclerosis (PSS) with silent thyroiditis and anti-bovine thyrotropin antibodies. Endocrinol Jpn 1991;38:265–270.

45. Murao S, Yoshinouchi T, Sato M, et al. Silent thyroiditis after excision of a thymoma. Intern Med 1998;37:604–605.

46. Carney AG, Moore SB, Northcutt RC, et al. Palpation thyroiditis (multifocal granulomatous folliculitis). Am J Clin Pathol 1975;64:639–647.

47. Walfish PG, Caplan D, Rosen IB. Postparathyroidectomy transient thyrotoxicosis. J Clin Endocrinol Metab 1992;75:224–227.

48. Smallridge RC, De Keyser FM, Van Herle AJ, et al. Thyroid iodine content and serum thyroglobulin: clues to the natural history of destruction-induced thyroiditis. J Clin Endocrinol Metab 1986;62:1213–1219.

48a. Wong V, Fu AX, George J, et al. Thyrotoxicosis induced by alpha-interferon therapy in chronic viral hepatitis. Clin Endocrinol (Oxf) 2002;56:793–798.

49. Kung AWC, Jones BM, Lai CL. Effects of interferon-y therapy on thyroid function, T-lymphocyte subpopulations and induction of autoantibodies. J Clin Endocrinol Metab 1990;71: 1230–1234.

50. Mitani Y, Shigesmasa C, Kouchi T, et al. Detection of thyroid stimulating antibody in patients with inflammatory thyrotoxicosis. Horm Res 1992;37:196–201.

51. Nakamura S, Sugimoto M, Kosaka J, et al. Silent thyroiditis with thyroid-stimulation-blocking antibodies (TSBAb). Japanese Journal of Medicine 1990;29:623–627.

52. Sarlis NJ, Brucker-Davis F, Swift JP, et al. Graves' disease following thyrotoxic painless thyroiditis. Analysis of antibody activities against the thyrotropin receptor in two cases. Thyroid 1997;7:829–836.

53. Yamamoto M, Sakurada T, Yoshida K, et al. Thyroid function and antimicrosomal antibody during the course of silent thyroiditis. Endocrinol Jpn 1987;34:357–363.

54. Nikolai TF, Coombs GJ, McKenzie AK, et al Treatment of lymphocytic thyroiditis with spontaneously resolving hyperthyroidism (silent thyroiditis). Arch Intern Med 1982;142:2281–2283.

55. Baytner-Zamir H, Zelikovsky L, Jarchowsky J. Prolonged fever in painless subacute thyroiditis diagnosis by gallium scan. Harefuah 1996;130;453–455, 504.

56. Hidaka Y, Okumura M, Fukata S, et al. Increased serum concentration of interleukin-12 in patients with silent thyroiditis and Graves' disease. Thyroid 1999;2:149–153.

57. Ho SC, Eng PH, Fok AC, et al. Thyrotoxicosis due to the simultaneous occurrence of silent thyroiditis and Graves' disease. Thyroid 1999;9:1127–1132.

58. Sayama N, Yoshida K, Mori K, et al. Measurement of red blood cell zinc concentration with Zn-test kit: discrimination between hyperthyroid Graves' disease and transient thyrotoxicosis. Endocr J 1998;45:767–772.

59. Nikolai TF, Coombs GJ, McKenzie AK, et al Treatment of lymphocytic thyroiditis with spontaneously resolving hyperthyroidism (silent thyroiditis). Arch Intern Med 1982;142:2281–2283.

60. Arem R, Munipalli B. Ipodate therapy in patients with severe destruction-induced thyrotoxicosis. Arch Intern Med. 1996;156:1752–1757.

61. Nikolai TF, Coombs GJ, McKenzie AK. Lymphocytic thyroiditis with spontaneously resolving hyperthyroidism and subacute thyroiditis: long-term follow-up. Arch Intern Med 1981;141: 1455–1458.

62. Roberton HEW. Lassitude, coldness, and hair changes following pregnancy, and their response to treatment with thyroid extract. BMJ 1948;2:93–94.

63. Amino N, Miyai KJ, Onishi T, et al. Transient hypothyroidism after delivery in autoimmune thyroiditis. J Clin Endocinol Metab 1976;42:296–301.

64. Ginsberg J, Walfish PG. Postpartum transient thyrotoxicosis with painless thyroiditis. Lancet 1977;1:1125–1128.

65. Muller AF, Drexhage HA, Berghout A. Postpartum thyroiditis and autoimmune thyroiditis in women of childbearing age: recent insights and consequences for antenatal and postnatal care. Endocr Rev 2001;22:605–630.

66. Roti E, Uberti E. Post-partum thyroiditis—a clinical update. Eur J Endocrinol 2002;146:275–279.

67. Stagnaro-Green A. Clinical review 152: postpartum thyroiditis. J Clin Endocrinol Metab 2002;87:4042–4047.

68. Lazarus JH, Premawardhana LDKE, Parkes AB. Postpartum thyroiditis. Autoimmunity 2002;35:169–173.

69. Amino N, Tada H, Hidaka Y. Thyroid disease after pregnancy: post-partum thyroiditis. In: Wass JAH, Shalet, SM, eds. Oxford textbook of endocrinology and diabetes. Oxford: Oxford University Press, 2002:527–532.

70. Lazarus JH, Premawardhana LDKE, Parkes AB. Postpartum thyroiditis. In: Weetman AP, ed. Immunology & medicine series: endocrine autoimmunity and associated conditions. London: Kluwer Academic Publishers, 1998;27:83–97.

71. Nikolai TF, Turney SL, Roberts RC. Postpartum lymphocytic thyroiditis. Arch Intern Med 1987;47;221–224.

72. Fung HYM, Kologlu M, Collison K, et al. Postpartum thyroid dysfunction in mid Glamorgan. BMJ 1998;296:241–244.

73. Gerstein HC. How common is postpartum thyroiditis? Arch Intern Med 1990;150:1397–1400.

74. Walfish PG, Meyerson J, Provias JP, et al. Prevalence and characteristics of past-partum thyroid dysfunction: results of a survey from Toronto, Canada. J Endocrinol Invest 1992;15:265–272.

75. Barca MF, Knobel M, Tominori E, et al. Prevalence and characteristics of postpartum dysfunction in São Paulo, Brazil. Clin Endocrinol (Oxf) 2000;53:21–31.

76. Shahbazian HB, Sarvghadi F, Azizi F. Prevalence and characteristics of postpartum thyroid dysfunction in Tehran. Eur J Endocrinol 2001;145:397–401.

77. Bagis T, Gokcel A, Saygili ES. Autoimmune thyroid disease in pregnancy and postpartum period: relationship to spontaneous abortion. Thyroid 2001;11:1049–1053.

78. Gallas PRJ, Stolk, RP, Bakker K, et al. Thyroid dysfunction during pregnancy and in the first postpartum year in women with diabetes mellitus type 1. Eur J Endocrinol 2002;147:443–451.

79. Weetman AP, McGregor AM. Autoimmune thyroid disease: further developments in our understanding. Endocr Rev 1994;15:788–830.

80. Mizukami Y, Michigishi T, Nonomura A, et al. Postpartum thyroiditis—a clinical, histologic and immunologic study of 15 cases. Am J Clin Pathol 1993;100:200–205.

81. Tachi J, Amino N, Tamaki H, et al. Long term follow-up and HLA association in patients with postpartum hypothyroidism. J Clin Endocrinol Metab 1988;66:480–484.

82. Kologlu M, Fung H, Darke C, et al. Postpartum thyroid dysfunction and HLA status. Eur J Clin Invest 1990;20:56–60.

83. Farid NR, Haw BS, Walfish PG. Increased frequency of HLA-DR3 and 5 in the syndromes of painless thyroiditis with transient thyrotoxicosis: evidence for an autoimmune etiology. Clin Endocrinol (Oxf) 1983;19:699–704.

84. Jansson R, Safwenberg J, Dahlberg PA. Influence of the HLA-DR4 antigen and iodine status on the development of autoimmune postpartum thyroiditis. J Clin Endocrinol Metab 1985;60:168–173.

85. Vargas Mt, Briones-Urbina R, Gladman D, et al. Antithyroid microsomal autoantibodies and HLA-DR5 are associated with postpartum thyroid dysfunction: evidence supporting an autoimmune pathogenesis. J Clin Endocrinol Metab 1988;67:327–333.

86. Parkes AB, Darke C, Othman S, et al. MHC class II and complement polymorphism in postpartum thyroiditis. Eur J Endocrinol 1996;134:449–453.

87. Davies TF. The thyroid immunology of the postpartum period. Thyroid 1999;9:675–684.

88. Weetman AP, Fung HYM, Richards CJ, et al. IgG subclass distribution and relative functional affinity of thyroid microsomal antibodies in postpartum thyroiditis. Eur J Clin Invest 1990;20:133–136.

89. Briones-Urbina R, Parkes AB, Bogner U, et al. Increase in antimicrosomal antibody related IgG1, IgG2 and IgG3 and titres of antithyroid peroxidase antibodies, but not antibody dependent cell mediated cytotoxicity in post-partum thyroiditis with transient hyperthyroidism. J Endocrinol Invest 1990;13:879–876.

90. Parkes AB, Othman S, Hall R, et al. The role of complement in the pathogenesis of postpartum thyroiditis. J Clin Endocrinol Metab 1994;79:395–400.

91. Parkes AB, Adams H, Othman S, et al. The role of complement in the pathogenesis of postpartum thyroiditis: ultrasound echogenicity and the degree of complement induced thyroid damage. Thyroid 1996;6:169–174.

92. Wegmann TG, Lin H, Guilbert L, et al. Bidirectional cytokine interactions in the maternal-fetal relationship: is successful pregnancy a TH2 phenomenon? Immunol Today 1993;14:353–356.

93. Watanabe M, Iwatani Y, Kaneda T, et al. Changes in T, B, and NK lymphocyte subsets during and after normal pregnancy. J Reprod Immunol 1997;37:368–377.

94. Stagnaro-Green A, Roman H, Cobin RH, et al. Prospective study of lymphocyte-initiated immunosuppression in normal pregnancy: evidence of a T-cell etiology for postpartum thyroid dysfunction. J Clin Endocrinol Metab 1992;74:645–653.

95. Kokandi AA, Parkes AB, Premawardhana LDKE, et al. Association of postpartum thyroid dysfunction with antepartum hormonal and immunological changes. J Clin Endocrinol Metab 2003;88:1126–1132.

96. Ando T, Davies TF. Postpartum autoimmune thyroid disease: The potential role of fetal microchimerism. J Clin Endocrinol Metab 2003;88:2965–2971.

97. Lazarus JH, Hall R, Othman S, et al. The clinical spectrum of postpartum thyroid disease. QJM 1996;89:429–435.

98. Sakaihara M, Yamada H, Kato EH, et al. Postpartum thyroid dysfunction in women with normal thyroid function during pregnancy. Clin Endocrinol (Oxf) 2000;53:487–492.

99. Nohr SB, Jorgensen A, Pedersen KM, et al. Postpartum thyroid dysfunction in pregnant thyroid peroxidase antibody-positive women living in an area with mild to moderate iodine deficiency: is iodine supplementation safe? J Clin Endocrinol Metab 2000;85:3191–3198.

100. Caixas A, Albareda M, Garcia-Patterson A, et al. Postpartum thyroiditis in women with hypothyroidism antedating pregnancy? J Clin Endocrinol Metab 1999;84:4000–4005.

101. Harris B, Othman S, Davies J, et al. Association between postpartum thyroid dysfunction and thyroid antibodies and depression. BMJ 1992;305:152–156.

101a. Kuijpens JL, Vader HL, Drexhage HA, et al. Thyroid peroxidase antibodies during gestation are a marker for subsequent depression postpartum. Eur J Endocrinol 2001;145:579–584.

102. Lucas A, Pizarro E, Granada ML, et al. Postpartum thyroid dysfunction and postpartum depression: are they two linked disorders? Clin Endocrinol (Oxf) 2001;55:809–814.

103. Kent GN, Stuckey AG, Allen JR, et al. Postpartum thyroid dysfunction; clinical assessment and relationship to psychiatric affective morbidity. Clin Endocrinol (Oxf) 1999;51:429–438.

104. Jansson R, Bernander S, Karlsson A, et al. Autoimmune thyroid dysfunction in the postpartum period. J Clin Endocrinol Metab 1984;58:681–687.

105. Amino N, Mori H, Iwatani Y, et al. High prevalence of transient postpartum thyrotoxicosis and hypothyroidism. N Engl J Med 1982;306:849–852.

106. Lervang HH, Pryds O, Ostergaard Kristensen HP. Thyroid dysfunction after delivery: incidence and clinical course. Acta Med Scand 1987;222:369–374.

107. Pop VJM, de Rooy HA, Vader HL, et al. Microsomal antibodies during gestation in relation to postpartum thyroid dysfunction and depression. Acta Endocrinol (Copenh) 1993;129:26–30.

108. Kuijpens JL, De Haan-Meulman M, Vader HL, et al. Cell-Mediated immunity and postpartum thyroid dysfunction: a possibility for the prediction of disease. J Clin Endocrinol Metab 1998;83:1959–1966.

109. Solomon BL, Fein HG, Smallridge RC. Usefulness of antimicrosomal antibody titres in the diagnosis of postpartum thyroiditis. J Fam Pract 1993;36:177–182.

110. Premawardhana LD, Parkes AB, John R, et al. Thyroid peroxidase antibodies in early pregnancy: utility for prediction of postpartum thyroid dysfunction and implications for screening. Thyroid 2004;14:610–615.

110a. Pearce EN, Bogazzi F, Martino E, et al. The prevalence of elevated serum C-reactive protein levels in inflammatory and noninflammatory thyroid disease. Thyroid 2003;13:643–648.

111. Oliveri A, De Angelis S, Vaccari V, et al. Postpartum thyroiditis is associated with fluctuations in transforming growth factor-β serum levels. J Clin Endocrinol Metab 2003;88:1280–1284.

112. Owen PJD, Lazarus JH. The treatment of post-partum thyroid disease. J Endocrinol Invest 2003;26:290–291.

113. Premawardhana LD, Parkes AB, Ammari F. Postpartum thyroiditis and long-term thyroid status: prognostic influence of thyroid peroxidase antibodies and ultrasound echogenicity. J Clin Endocrinol Metab 2000;85:71–75.

114. Dahlberg PA, Jansson R. Different aetiologies in post-partum thyroiditis. Acta Endocrinol (Copenh) 1983;104:195–200.

115. Lazarus JH, Ammari F, Oretti R, et al. Clinical aspects of recurrent postpartum thyroiditis. Br J Gen Pract 1997;47:305–308.