Impact of thyroid disease on fertility and assisted conception

Journal Pre-proof Impact of thyroid disease on fertility and assisted conception David Unuane, MD, PhD, Brigitte Velkeniers, MD, PhD

PII:

S1521-690X(20)30005-1

DOI:

https://doi.org/10.1016/j.beem.2020.101378

Reference:

YBEEM 101378

To appear in:

Best Practice & Research Clinical Endocrinology & Metabolism

Please cite this article as: Unuane D, Velkeniers B, Impact of thyroid disease on fertility and assisted conception, Best Practice & Research Clinical Endocrinology & Metabolism, https://doi.org/10.1016/ j.beem.2020.101378. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 The Author(s). Published by Elsevier Ltd.

1

2

Impact of thyroid disease on fertility and assisted conception

3

4

David Unuane, MD, PhD a, Brigitte Velkeniers MD, PhD a

5

a

6

Belgium

Department of Endocrinology, Universitair Ziekenhuis Brussel, UZ Brussel, Vrije Universiteit Brussel,

7 8

Abstract

9

Thyroid auto-immunity (TAI) and/or thyroid dysfunction are prevalent in women of reproductive age and

10

have independently been associated with adverse fertility and pregnancy outcomes, in the case of

11

spontaneous conception or after assisted reproductive technology (ART). Thus, it seems reasonable to

12

screen for thyrotropin (TSH) and thyroid peroxidase autoantibodies (TPO-abs) in infertile women

13

attempting pregnancy. However, even if the relationship between fertility and thyroid dysfunction

14

and/or TAI persists when properly controlled for other variables, it remains challenging to claim

15

causation. Several studies with different designs (cross sectional, case –control, prospective and

16

retrospective cohort studies) have looked at the association between thyroid autoimmunity, thyroid

17

function and fertility. Heterogeneity among study results are related to small numbers of included

18

patients, poor study design, selection of causes of infertility and different assays used to measure TAI,

19

thyroid hormones and TSH reference values. Indeed, there is no consensus regarding the upper limit of

20

normal for TSH to define thyroid dysfunction and the cut-off levels for intervention. Furthermore, data

21

from interventional trials looking at the impact of levothyroxine treatment on fertility outcome in 1

22

randomised controlled studies are scarce. Despite the recent update of the guidelines by the American

23

Thyroid Association (ATA) for the Diagnosis and Management of Thyroid Disease during Pregnancy and

24

the postpartum, many questions remain unsettled in ART.

25 26 27 28

Key words: thyroid dysfunction, thyroid auto- immunity, female infertility, assisted reproductive technology

29

30

31

Introduction

32

Procreation is a fundamental evolutionary process to sustain life. Complex regulatory endocrine

33

and immune factors determine human fertility and implantation.

34

The international committee for monitoring assisted reproductive technology (ICMART) and the

35

world health organisation (WHO) revised glossary on ART terminology define infertility (clinical

36

definition) "as disease of the reproductive system defined by the failure to achieve a clinical

37

pregnancy after 12 months or more of regular unprotected sexual intercourse (1)".

38

The prevalence of infertility may vary considerably according to the ethnic background (2,3).

39

Overall it has been estimated that about one in six of all couples will face difficulties to conceive

40

(primary infertility) or to conceive the number of desired children (secondary infertility) (4,5).

41

The World Health Organization (WHO) task force on diagnosis and treatment of infertility

42

performed a study of 8500 infertile couples, in developed countries (6). Female related

43

infertility accounted for 37 % and combined male and female factors for 35% of the causes of 2

44

infertility. The most frequent causes of female infertility were: ovulatory disorders (25%),

45

endometriosis (15%), pelvic adhesions (12%), tubal blockage (11%), other tubal abnormalities

46

(11%) and hyperprolactinemia (7%). When confronted with fertility problems, couples often

47

consult fertility specialists for help. The field of fertility has exploded over the past two decades

48

and researchers have tried to identify treatable risk factors that contribute to infertility. Among

49

these risk factors are the presence of thyroid autoimmunity and thyroid dysfunction. Therefore,

50

screening for TSH and anti-thyroperoxidase antibodies (TPO-abs) is generally part of the initial

51

work-up. Further, different guidelines have been proposed to guide the practitioner in the

52

management of these thyroid disorders during pregnancy (7,8). Recently, the new guidelines by

53

the ATA have been released (7). These guidelines have to take into account the low level of

54

evidence and rated recommendations accordingly. Therefore, despite the efforts by the ATA,

55

the reader is sometimes left with unanswered questions. In the present narrative review we

56

aim to describe some new insights that link thyroid disease to infertility.

57 58

Thyroid disorders and female infertility

59 60

Epidemiological data have shown a high prevalence of thyroid disorders (dysfunction and

61

autoimmunity) in women of reproductive age (9). The incidence of hypothyroidism

62

rangesbetween 2% and 4% and is largely attributed to thyroid autoimmunuty (TAI) (10).

63

Subclinical thyroid anomalies, including subclinical hypothyroidism, hypothyroxinemia and / or

64

isolated TAI are much more frequent than clinical thyroid dysfunction. In women with TPO-abs,

65

the relative risk of female infertility is increased (RR= 2.25; 95% CI 1.02-5.12; p=0.045) (10). In 3

66

addition, women with recurrent miscarriages have a higher incidence of Tg- and/or TPO-abs,

67

amounting as high as 25% (11,12).

68

The importance of thyroid hormones has been highlighted ever since the discovery of TSH, its

69

receptor as well as thyroid hormone receptors (TR-α1 and TR-β1) on ovarian surface epithelium

70

and in oocytes of primordial, primary, and secondary follicles (13). Thyroid hormones seem to

71

participate in the complex regulation of ovarian function. In animal models thyroid hormones

72

synergize with FSH to exert direct stimulatory effects on granulosa cell functions, such as

73

morphologic differentiation, LH/HCG receptor formation, and induction of 3b-hydroxysteroid

74

dehydrogenase and aromatase (14). Thyroid hormones may also influence fertility aspects in an

75

indirect manner, altering gonadotropin releasing hormone (GnRH) and prolactin secretion, sex

76

hormone binding globulin (SHBG) levels and coagulation factors.

77

Considering the importance of thyroid hormones, even mild thyroid failure has been proposed

78

as one of the possible causes for adverse fertility and pregnancy outcomes.

79

80

1. Thyroid dysfunction and female infertility

81 82

1 (i) Subclinical and overt hyperthyroidism and female infertility

83 84

The extent to which hyperthyroidism is related to fertility issues is not well established due to

85

limited data in literature. However, in general, if not treated adequately, hyperthyroidism is

86

associated with early pregnancy loss (15). 4

87

Thyrotoxicosis increases sex hormone binding globulin (SHBG) and estradiol (E2) serum levels

88

compared to euthyroid women. The latter may result from an increase in SHBG or increased E2

89

and androgen production, combined with an increased conversion ratio to estrone and E2. In

90

addition, LH secretion is increased in Graves' disease patients compared to euthyroid patients

91

(16).

92 93

In a cross sectional study, Joshi et al reported primary or secondary infertility in 5.8 % of

94

hyperthyroid patients (17). Almost 10 years later a prospective study showed a prevalence of

95

supressed TSH (subclinical or overt hyperthyroidism) in 2.1% compared to 3 % in fertile controls.

96

Despite a comparable prevalence of hyperthyroidism, supressed TSH was more prevalent in

97

antibody-positive compared to the antibody-negative patients (10).

98

Menstrual disturbances are common in hyperthyroid women. Early data suggest menstrual

99

abnormalities in up to 65% compared to 17 % in healthy controls (17, 18, 19,20). More recent

100

data showed a considerably lower prevalence of menstrual abnormalities of about 22%

101

compared to 8 % in healthy controls (17). Secular trends in the diagnosis and treatment of

102

clinical hyperthyroidism probably account for these discrepant findings.

103

Hypomenorrhea, polymenorrhea, oligomenorrhea and hypermenorrhea are the most prevalent

104

menstrual abnormalities. Interestingly, according to a very early study looking at endometrial

105

biopsies, most hyperthyroid women seem to retain ovulatory cycles (21).

106 107 108

1 (ii) subclinical and overt hypothyroidism and female infertility

5

109 110

Overt hypothyroidism is associated with an increased risk of fertility problems and unfavourable

111

early and late pregnancy complications (22).

112

Hypothyroidism results in a number of hormonal changes. The rates of metabolic clearance of

113

both androstenedione and estrone are decreased, whereas peripheral aromatization is

114

increased. In addition, the plasma binding activity of SHBG is decreased. Consequently, plasma

115

concentrations of both total testosterone and E2 are decreased and their unbound fraction

116

increased. Hypothyroidism may also lead to a blunted LH response thereby stimulating TRH

117

secretion and increasing serum prolactin levels. As prolactin impairs pulsatile secretion of

118

gonadotrophin-releasing hormone (GnRH) this can lead to ovulatory dysfunction, including

119

insufficiency of the corpus luteum with low progesterone secretion in the luteal phase of the

120

cycle (17).

121

Therefore, clinical hypothyroidism leads to a number of ovulatory disturbances in women of

122

fertile age. There may be changes in cycle length as well as in the volume of menstrual bleeding

123

secondary to breakthrough bleeding following anovulation and/or disturbances in haemostatic

124

factors associated with hypothyroidism.

125

The prevalence of menstrual abnormalities reported is 25 % to 60 % in hypothyroid women

126

compared to 10% in euthyroid women. The predominant menstrual disturbance in hypothyroid

127

women described is oligomenorrhea (17).

128 129

The association of subclinical hypothyroidism (SCH) with infertility is not consistent in literature.

130

This can partly be explained by different cut-offs used to define the upper limit of normal of TSH

6

131

concentration and a lack of well-designed prospective studies. Using TRH tests to define SCH

132

Bohnet et al. linked 11% of infertility to subtle underfunctioning of the thyroid (23). Similarly,

133

again using TRH tests to define SCH, a higher prevalence of SCH (13.9% vs.3.9%) was also

134

reported in a retrospective study by Abalovich et al. (24) in 244 infertile women. In a cross-

135

sectional survey of 704 infertile women by Lincoln et al. (25), sixteen (2.3%) had elevated TSH

136

levels (normal, 0.45-4.09 mIU/mL). In our own prospective study we could not confirm an

137

increased prevalence of SCH among infertile women, although patients with fertility problems

138

had a higher mean TSH compared to fertile women (10).

139

In a large retrospective cross- sectional study in 11,254 women in Denmark, impaired fertility

140

was associated with subclinical hypothyroidism defined as TSH >3.7 mIU/L ( 26). A recent cross-

141

sectional study also showed higher normal TSH levels in women with unexplained infertility

142

compared to fertile controls (27).

143

Based on these data we conclude that TSH levels used in different studies to determine the

144

association of thyroid function with fertility problems varied considerably. In general,

145

association with adverse fertility outcomes seem to emerge at TSH levels above 4.0 mIU/L.

146 147 148 149

2. Thyroid autoimmunity and female infertility

150 151

TAI is the most frequent autoimmune disorder in women of childbearing age and increases the

152

risk of thyroid dysfunction. The prevalence of TAI is generally estimated at around 10%. (9, 28-

7

153

36) and has been shown to be more common in women seeking counselling for infertility (10,

154

11, 12). A meta-analysis pooling 4 studies showed that the presence of thyroid antibodies in

155

euthyroid patients is associated with unexplained subfertility (OR 1.5, 95% CI 1.1–2.0) (22).

156 157

Other causes of infertility have also been linked to TAI. Women with polycystic ovary syndrome

158

(PCOS) were found to have an increased prevalence of TAI. A possible explanation may be

159

polymorphisms of the PCOS-related gene for fibrillin 3, influencing the activity of TGF-β, a key

160

regulator of immune tolerance. Together with lower TGF-β, low vitamin D levels and the high

161

estrogen-to-progesterone ratio, these factors may contribute to autoimmunity (37).

162

As for endometriosis the data are more controversial, with inconsistent results of an increased

163

prevalence of TAI. (38). There is evidence that endometriosis is associated with a variety of

164

immunological changes, including antibodies to endometrial antigens.

165 166

In general, increased levels of TPO-abs have been defined as the most sensitive marker of TAI

167

and linked to the risk of (sub) clinical hypothyroidism (9). Therefore, most studies investigating

168

the prevalence of TAI in infertile women or any relationship between TAI and pregnancy

169

outcomes are based on the presence of TPO-abs alone and do not take into account

170

thyroglobulin autoantibodies (Tg-abs). In a cross-sectional study we investigated the prevalence

171

of both antibodies in 992 women consulting for infertility (39). Both TPO and Tg-Abs were

172

present in 8% of the cases. Isolated Tg-Abs were present in 5% of the cases compared to 4% of

173

isolated TPO-abs. Interestingly, women who had only T g-Abs positivity had a significant higher

8

174

median TSH compared to women without thyroid autoimmunity. The added value of measuring

175

Tg- antibodies in fertility work-up and ART outcome remains to be determined.

176 177

178 179

3. Thyroid autoimmunity and Assisted Reproductive Technology

180 181

TAI has been linked to adverse pregnancy outcomes with an increased risk of miscarriage and

182

preterm delivery in spontaneous pregnancy as well as in pregnancy after ART. Since the late

183

1990s several studies have been published on the impact of TAI on the outcome after ART. The

184

outcome of these studies however are conflicting, probably due to different study designs,

185

small numbers of enrolled patients with low absolute numbers of events analysed and the use

186

of surrogate endpoints for fertility outcome (Table 1). A study by Zhong et al. (40) comparing IVF

187

outcome in TAI positive and TAI negative women revealed that TAI positive women had a

188

significantly lower fertilization - (64.3% vs. 74.6%), implantation- (17.8% vs. 27.1%) and

189

pregnancy rate (33.3% vs. 46.7%) and a higher risk of abortion (26.9 vs. 11.8%) following IVF-ET

190

compared to their TAI negative counterparts. In two meta-analyses an increased miscarriage

191

rate was documented in TAI positive women with subfertility. In the first meta-analysis of four

192

prospective studies by Toulis, women with TAI were twice more likely to experience a

193

miscarriage after IVF pregnancies. (41) The actual difference in the number of miscarriages

194

between euthyroid women with TAI undergoing IVF and controls (the absolute risk) remained

195

small, resulting in a rather ‘silent’ effect on clinical pregnancy rate and delivery. 9

196

In a second meta-analysis of 12 studies, the risk of miscarriage was significantly higher among

197

women with TAI, resulting in a lower live birth rate (42) . However, the number of oocytes

198

retrieved fertilization- and pregnancy rates did not seem to be affected by TAI status. This meta-

199

analysis quoted a cautious interpretation of the data, because of heterogeneity among studies

200

included. This heterogeneity is due to the small samples of the cohort studies, the different

201

causes of infertility included, the different assays using varying cut-off levels for TSH, fT4 and

202

thyroperoxidase antibodies.

203

More recent studies on this subject could not confirm the association between TAI and their

204

detrimental role on fertility outcome. In a retrospective study by Lukaszuk et al., pregnancy

205

outcome between 114 TAI positive and 495 TAI negative infertile women was compared. No

206

significant difference in fertilization, implantation, pregnancy rates, live birth rates or higher risk

207

for miscarriage could be revealed (43). A prospective study by Sakar et al. also showed

208

comparable pregnancy and miscarriage rates between 49 TAI positive and 202 TAI negative

209

women after IVF. In this study however, delivery rates were not reported (44). According to Tan

210

et al. any previous association between TAI positive women and negative pregnancy outcome

211

may be confounded by a selection bias, in particular the selection of female patients who seek

212

treatment for fertility problems. To avoid this bias, the author selected couples with male factor

213

infertility and studied female partners with TAI but without known female infertility. They found

214

that pregnancy outcome was comparable among women with and without thyroid

215

autoimmunity after intra cytoplasmic sperm injection (45).

216

It seems important for both candidate couples and medical practitioners to predict the impact of

217

TAI on fertility outcome after a given number of IVF/ICSI treatment cycles. Therefore, our own 10

218

group looked into the impact of TAI on the probability of delivery after six IVF/ICSI cycles in

219

2352 women (333 with TAI and 2019 without TAI) (46). We provided two approaches often

220

used to estimate the effectiveness of ART. The first method calculates "crude" delivery rates by

221

dividing the number of women achieving live birth delivery up to a predetermined number of

222

cycles by the total number of women who started with ART treatment. To account for any effect

223

of "drop out" we also provided a second approach estimating live birth delivery using life-table

224

analysis ("expected" delivery rates). We found comparable live birth delivery inTAI positive and

225

TAI negative women for both crude (47% vs 47%) as well as expected (65% vs 76%) live birth

226

delivery rates. It has been suggested that a decreasing impact of TAI on fertility outcome after

227

ART could be explained by the increasing use of ICSI. In this respect, a recent meta-analysis

228

only including studies in which women were treated with ICSI, concluded that infertile women

229

with TAI undergoing ICSI had no increased risk of a first trimester miscarriage compared with

230

women without TAI (47). However, at present we lack data comparing IVF vs ICSI outcome in

231

TAI positive women to support that systematic use of ICSI should be favoured.

232

Since IVF/ICSI procedures may bypass some of the theoretical action points of an underlying

233

auto-immune process in comparison to intrauterine insemination (IUI) we also studied the

234

impact of TAI on fertility outcome after IUI treatment in 3143 women (187 with TAI and 2956

235

without TAI) (48). Again, we were unable to show any significant difference with respect to live

236

birth delivery-, pregnancy-, or miscarriage rate with odds ratio at 1.04 (95% CI 0.63 ; 1.69), 0.98

237

(95% CI 0.62 ; 1.55) and 0.74 (95% CI 0.23 ; 2.39) respectively between anti-TPO+ and anti-TPO-

238

women.

239

The underlying pathophysiological mechanisms linking possible negative impact of TAI on

240

pregnancy outcome after ART remains unsettled. The presence of TAI could represent a 11

241

peripheral marker of a general immune imbalance possibly leading to failure of fertilization,

242

implantation failure and sustained pregnancy. The mechanism by which TAI may operate

243

consists in the alteration of the endometrium receptivity that affects the foetal allograft.

244

Quantitative and qualitative changes in the profile of endometrial T cells with reduced secretion

245

of IL-4 and IL-10 along with hypersecretion of interferon-γ have been reported (49).

246

Hyperactivity and increased migration of cytotoxic natural killer cells may also alter the immune

247

and hormonal response of the uterus in women with TAI. Compared to women who have had

248

one or no miscarriage, women with multiple miscarriages seem to have an increased number of

249

CD5/20+ B cells with a higher prevalence in women with thyroid antibodies (50, 51). Polyclonal

250

B cell activation is more frequent in TAI and is associated with increased titers of non-organ

251

specific autoantibodies. Limited data report on the presence of anti-phospholipid (APL) and/or

252

anti-nuclear (ANA) antibodies in the population of infertile women with TAI (52). Lupus

253

anticoagulant and anti cardiolipin antibodies are significantly more prevalent in women with

254

infertility. Overall the impact of different associated autoimmune diseases on reproductive

255

health variables is largely overlooked and needs to be further investigated (53).

256

257

Alternativel,y thyroid antibodies could be directly causative since the presence of thyroid

258

antibodies may have an unfavourable effect on oocyte and embryo quality. Indeed, Monteleone

259

et al. also found a significantly lower oocyte fertilization and percentage of grade A embryos

260

when comparing infertile women undergoing IVF with thyroid autoimmunity to negative

261

controls (54). For the first time, this study documented the presence of thyroid antibodies in

12

262

ovarian follicular fluid .The authors hypothesized that thyroid antibodies may pass the "follicle-

263

blood" barrier during maturation of a secondary follicle. Cytotoxicity of these antibodies could

264

then damage the maturing oocyte and eventually reduce oocyte quality and fertilization

265

potential.

266

On the other hand, a positive TAI status increases the risk of developing (sub) clinical

267

hypothyroidism. Women with TAI are at increased risk of developing (sub) clinical

268

hypothyroidism during spontaneous pregnancy (55). In the case of assisted reproductive

269

technology (ART) and ovarian hyperstimulation, it has been reported that TSH levels increase

270

significantly above 2.5 mIU/L before pregnancy and even more so in the presence of TAI (56, 57,

271

58). Since thyroid hormones play an essential role in oocyte maturation and implantation, it has

272

been hypothesized that the decline in thyroid function induced by the stimulation protocol in

273

women with TAI may negatively influence pregnancy rate in ART .

274 275

4. SCH and outcome after ART

276

Recently, the reference range of TSH in fertility and pregnancy has become a matter of debate.

277

The issue is relevant because changes in the reference ranges of TSH can result in a considerably

278

larger proportion of patients being diagnosed and treated for SCH. TSH cut-offs should not

279

merely reflect a biochemical diagnosis, but ought to be linked with adverse fertility and

280

pregnancy outcomes.

281

SCH increases the risk of miscarriage compared to euthyroid women in spontaneous

282

pregnancies, although the risk increase in terms of absolute events is limited.

13

283

Recently Zhang et al conducted a meta-analysis of 9 cohort studies among women with SCH or

284

normal thyroid function before 20 weeks of pregnancy (59). In a sub group analysis including 3

285

studies , patients with SCH had a higher prevalence of miscarriage than euthyroid women (RR =

286

1.45, 95% CI1.07-1.96, P = 0.02) (95%CI 1.07±1.96, P = 0.02). However, the included studies used

287

different TSH cut-offs to define SCH.

288

In a prospective cohort study among healthy fertile women with a history of pregnancy loss,

289

subclinical hypothyroidism defined as TSH levels ≥ 2.5 -5 mIU/L , was not associated with

290

fecundity, pregnancy loss or live birth (60).

291

In assisted reproductive technology, the majority of studies suggests comparable pregnancy

292

outcomes whether using a TSH cut off level of < 2.5 mIU/L or < 5 mIU/L.

293

One study found higher pregnancy rates in women with TSH levels ≤ 2.5 mIU/L compared to

294

women with TSH levels > 2.5 mIU/L ( 22.3% vs 8.9%) after the 1st cycle of IVF/ICSI (61). However,

295

these findings could not be confirmed in subsequent studies. In a large retrospective cohort

296

study of first cycle IVF patients, implementing a stricter TSH cut off of 2. 5 mIU/L compared to

297

4.5 mIU/L, no differences in clinical pregnancy, delivery or miscarriages were documented (62).

298

Similarly, using the same cut off levels, Chai et al confirmed comparable live birth and

299

miscarriage rates in IVF/ICSI treated patients (63). Recently, a retrospective analysis in infertile

300

women investigated the effect of increments of 0.5 mIU/L at TSH levels ≤ 2.5 mIU/L on IVF

301

outcome (64). No effect was seen on implantation, live birth or miscarriage rates between

302

different TSH groups ≤ 2.5 mIU/L nor was there a difference when comparing TSH groups above

303

or below 2.5 mIU/ . In our own retrospective study, a TSH threshold of 2.5 mIU/L, compared

14

304

with a threshold of 5 mIU/L, did not result in statistically different cumulative live birth delivery

305

rates after 6 cycles of IVF/ICSI (46).

306

When considering IUI, in a study by Karmon et al. among euthyroid patients, preconception TSH

307

values in the high-normal range (between 2.5 and 4.9 mIU/L) were not associated with adverse

308

IUI outcomes in terms of live birth, clinical pregnancy orspontaneous abortion (65).

309

Using IUI, we confirmed these findings with no significant differences in live birth delivery-,

310

pregnancy- or miscarriage rate in subgroups according to TSH level (TSH≥2.5 mIU/l -5 mIU/L vs.

311

TSH <2.5 mIU/l) with an odds ratio at 1.05 (95% CI: 0.76; 1.47), 1.04 (95% CI: 0.77; 1.41)

312

and 0.95 (95% CI: 0.47; 1.94) respectively (48). A recent retrospective cohort study found that

313

among euthyroid women preconceptional TSH values between 2.5 and 4.9mIU/L did not have

314

a negative effect on IUI outcome (66).

315

In conclusion, the majority of studies confirm that the non-pregnant reference range can be

316

used for the assessment of SCH in women who plan to become pregnant by ART.

317 318

5. Levothyroxine treatment and ART

319 320

(i)

Overt hypothyroidism

321

Detailed information regarding the effect of hypothyroidism on controlled ovarian

322

hyperstimulation (COH) and IVF outcomes is limited, since no randomized controlled trials are

323

available. It is likely that we will never have these data since it would be unethical not to treat

324

patients with clinical overt hypothyroidism. Numerous detrimental effects of maternal overt

325

hypothyroidism on pregnancy outcomes and adverse neurocognitive effects of the offspring 15

326

have been documented in literature (22). Treatment of hypothyroidism with levothyroxine

327

usually restores a normal menstrual pattern, reverses hormonal alterations, and improves

328

fertility (16). However, some women with treated hypothyroidism still fail to conceive and seek

329

infertility treatment, including controlled ovarian hyperstimulation and/or IVF. Treatment of

330

overt hypothyroidism in pregnancy is mandatory and consists of levothyroxine therapy adjusted

331

to achieve normal trimester-specific serum levels of thyroid stimulating hormone (TSH).

332

The latest ATA guidelines recommend to target TSH concentration below 2.5 mIU/L, in patients

333

treated for clinical hypothyroidism. The absolute inability to mount thyroid hormones during

334

COH may indeed exert a detrimental influence on the oocyte and/or the endometrium.

335

However, the level of TSH to be reached before starting the ART procedure remains

336

controversial.

337 338 339

(ii)

Subclinical hypothyroidism

340

A meta-analysis of 3 RCT's showed some beneficial effect of levothyroxine treatment on

341

pregnancy after ART in women with subclinical hypothyroidism. (67). Although no benefit was

342

shown in regards to clinical pregnancy rates (pooled relative risk 1.75, 95% CI 0.90-3.38) there

343

was an increase in delivery rates (pooled relative risk 2.76, 95% CI 1.20-6.44). However, the

344

included studies did differ in respect to the upper limit of TSH to define SCH. This limit often

345

exceeded 4 mIU/L. Indeed, the larger part of the available data concludes that there is no

346

difference in ART outcome between euthyroid women with TSH levels < 2.5 mIU/L and those

347

with mild TSH elevations between 2.5-5.0 mIU/L (60-66). The use of appropriate reference

16

348

ranges should be based on local population and laboratory specific reference values.

349

Moreover,in pregnancy, ATA guidelines, advocate the use of TSH concentrations based on

350

pregnancy-specific reference ranges (7). If not available, the use of a fixed upper limit of 4.0

351

mIU/L, which usually corresponds to a reduction of 0.5 mIU/L compared with the nonpregnant

352

TSH reference range, is advised. These guidelines to diagnose and treat SCH, seem sound in the

353

case of ART, since adverse events were mainly documented in patients with TSH values above

354

4.0 mIU/L.

355 356

Lessons learned from Spontaneous pregnancies: do they apply to ART?

357 358

Four Randomized controlled trials have evaluated the effect of levothyroxine treatment in SCH

359

on different aspect of spontaneous pregnancy outcomes.. Negro et al. showed that treatment

360

with levothyroxine improved adverse pregnancy outcome in TPO-abs positive women, with

361

TSH levels exceeding 2.5 mIU/L (68). Nazapour et al. also showed a decrease in pre-term

362

delivery in TPO-abs positive women when treated with levothyroxine but only in patients with

363

TSH levels exceeding 4.0 mIU/L (69). The same authors also confirmed only beneficial effect of

364

LT4 therapy in reducing preterm delivery in TPO-abs negative women with TSH levels above 4

365

mIU/L.

366

Neither the CATS trial nor the recent study by Casey et al. were able to show a benefit of

367

levothyroxine treatment on neuropsychological development of the offspring of patients with

368

SCH (70, 71). One of the major concerns raised was the late treatment initiation in the 2°

17

369

trimester of pregnancy. Therefore, it remains unclear if earlier treatment could have had added

370

value for the outcomes of interest.

371

In the meta-analysis by Zhang et al, a subgroup analysis including two studies, compared the

372

influence of treatment without intervention in SCH on the risk of miscarriage. The results

373

indicated that compared with patients that underwent effective drug treatments, the risk of

374

miscarriage increased significantly in patients without intervention with an overall combined RR

375

of1.50 (95% CI 1.03–2.19, P = 0.04). However, the latter study needs to be interpreted with

376

caution because of potential selection bias, different diagnostic criteria for SCH and different

377

definitions used for pregnancy outcome. (59).

378

Given the inconsistencies and limitations of the available interventional trials, treatment with

379

levothyroxine seems appropriate in TPO-abs positive women when TSH levels rise above the

380

non-pregnancy specific reference range. To the best of our knowledge, evidence is weak to

381

initiate levothyroxine treatment in TPO-abs positive women with TSH levels between 2.5 and 4

382

mIU/L. Also, based on the recent study by Nazarpour in TPO-abs negative women, treatment

383

may be recommended only at TSH levels greater than 4 mIU/L (72). The current guidelines

384

suggest treating SCH with levothyroxine during gestation with the aim of keeping TSH <2.5

385

mIU/L. On the other hand, studies suggest that the elevation of TSH in pregnant women should

386

be the upper limit of the gestation-specific reference interval rather than 2.5 mIU/L and this

387

may also apply during pregnancy obtained after ART (7).

388 389

Levothyroxine treatment in TPO antibody positive euthyroid women.

390

18

391

At present, the benefit of levothyroxine treatment on pregnancy outcomes in euthyroid women

392

with TAI, both for spontaneous pregnancy and after ART, is doubtful.

393

In the case of ART, levothyroxine treatment does not seem to be beneficial in TAI positive

394

euthyroid women. Indeed, a RCT by Negro et al could not show any improvement in TAI positive

395

women undergoing IVF (73). These findings were supported by a recent RCT in China where

396

treatment with levothyroxine, compared with no levothyroxine treatment, did not reduce

397

miscarriage rates or increase live-birth rates among women who had normal thyroid function

398

and tested positive for thyroid autoantibodies undergoing in vitro fertilization and embryo

399

transfer (74). Recently, the results of the of TABLET trial, the largest study on this topic to date,

400

have been published. This multicentric, double blinded randomized controlled trial, investigated

401

the effect of levothyroxine 50 micrograms daily, in TPO-abs positive euthyroid women before

402

conception. Women with a medical history of (recurrent) miscarriage or receiving treatment for

403

infertility were included. Levothyroxine treatment did not result in a higher live birth- or

404

pregnancy rate compared with placebo, confirming the conclusions of previous RCT's (75).

405

Despite the low evidence, some guidelines recommend that treatment may be initiated for any

406

TSH elevation >2.5 mIU/L before ART, considering its potential benefits in comparison to its

407

minimal risk (7). However, some potential hazards have recently been highlighted pointing to

408

the association of maternal thyroid overtreatment during early pregnancy with deleterious

409

effects on offspring IQ and brain morphology in childhood (76). More recently, additional

410

concerns have been raised since levothyroxine treatment may increase the risk of pre- term

411

delivery, gestational diabetes and pre- eclampsia (77).

412 19

413

414

Clinical approach

415

It is recommended that all patients seeking medical advice for infertility should be screened for

416

underlying thyroid disease given the potential detrimental effect of thyroid dysfunction on

417

fertility outcome. A thorough medical history and physical examination should be the first step.

418

Apart from the more apparent symptoms such as weight changes, fatigue, irritability and

419

palpitations one should also pay attention to symptoms such as menstrual disturbances that

420

may point towards thyroid dysfunction. Laboratory assessment should include at least TSH and

421

TPO-abs.

422

In the case of overt thyroid dysfunction the appropriate therapy should promptly be initiated.

423

SCH is best treated when TSH values exceed the normal upper value defined by the population

424

specific non pregnant values, or in the absence of these reference ranges by a fixed cut-off of 4

425

mIU/L.

426

Given the uncertainties in literature concerning the harm/benefit ratio in patients with TSH

427

cutoff levels between 2.5 and 4.0 mIU/ , intervention in these cases should be restricted and

428

discussed with the patients in regard to the current best available evidence.

429

Indeed, at present, there is insufficient eveidnce to initiate levothyroxine treatment at pre-

430

conception TSH levels between 2.5 and 4.0 mIU/L, in particular in the case of ART.

431

In figure 1, we provide an algorithm for a clinical approach of thyroid disorders in infertile

432

women considering ART.

433

20

434

435

Summary

436

The interaction between thyroid disease and fertility is complex. Overt thyroid dysfunction

437

often leads to menstrual disturbances, fertility problems and pregnancy complications and

438

should therefore be treated accordingly. At present there is little evidence to advise

439

levothyroxine treatment at TSH levels between 2.5 and 4.0 mIU/L considering the possible

440

adverse effects of overtreatment in particular in patients with mild thyroid dysfunction. We

441

propose careful longitudinal follow up, especially in the presence of thyroid antibodies for

442

women undergoing an ART procedure. The cut-off value of 4 mIU/L for TSH emerges as the

443

intervention level for treatment of SCH in women with and without TAI in ART.

444

445 446

Practice points


plan to become pregnant by ART.

447 448







453

In general, association with adverse fertility outcomes seem to emerge at TSH levels above 4.0 mIU/L.

451 452

The use of appropriate reference ranges should be based on local population and laboratory specific reference values.

449 450

Non-pregnant reference range can be used for the assessment of SCH in women who




There is little evidence to advise levothyroxine treatment at TSH levels between 2.5 and 4.0 mIU/L. 21

454 455




Current scientific evidence considers the cut-off value of 4 mIU/L for TSH as a reasonable intervention level for treatment of SCH in women with and without TAI in ART.

456

457

458

Research agenda

459

We propose a prospective longitudinal study to:

460 461

•

during pregnancy in ART (and subsequent pregnancy)

462 463

•

•

468

Investigate in prospective cohort studies the association between autoimmunity (ANA and APL) and thyroid autoimmune disease in women

466 467

Evaluate if the presence of Tg-abs has an impact on pregnancy outcome both in the case of spontaneous pregnancy and in pregnancy after ART

464 465

Follow thyroid function in Tg-abs positive women versus Tg-abs negative women

•

Determine the optimal TSH reference ranges and the upper limit of TSH for treatment of TAI positive women, planned for ART.

469

22

470

23

471

References

472

1.

Infertility and Fertility Care. Hum Reprod 2009; 32: 1786–1801.

473

474

Zegers-Hochschild F, Adamson GD, Dyer S, Racowsky C et al. The International Glossary on

2.

Mascarenhas MN, Flexman SR, Boerma T et al. National, Regional, and Global Trands in

475

Infertility Prevalance since 1990: A sytematic Analysis of 277 Health surveys. PloS Med

476

2012; 9: e 1001356. Doi: 10.1371.

477

3.

Datta J, Palmer MJ, Tanton C et al. Prevalence of infertility and help seeking among 15.000 women and men. Hum Reprod 2016; 31: 2108-2118.

478

479

4.

Evers JLH. Female subfertility. Lancet 2002; 360:151-159.

480

5.

Thonneau P, Marchand S, Tallec A et al. Incidence and main causes of infertility in a resident population of three French regions. Hum Reprod 1991; 6: 811-816.

481

482

6.

Number 820, 1992, pp 1-111.

483

484

WHO Technical Report Series. 1992. Recent Advances in Medically Assisted Conception

*7.

Alexander EK, Elizabeth N , Pearce EN et al . Guidelines of the American Thyroid

485

Association for the Diagnosis and Management of Thyroid Disease during Pregnancy and

486

the Postpartum. Thyroid 2017; 27: 315-389.

487

8.

Subclinical hypothyroidism in the infertile female population: a guideline. Practice

488

Committee of the American Society for Reproductive Medicine. Fertil Steril 2015; 104:

489

545-555.

24

490

9.

Hollowell JG, Staehling NW, Flanders DW et al. Serum TSH, T(4), and thyroid antibodies in

491

the United States population (1988 to 1994): National Health and Nutrition Examination

492

Survey (NHANES III). J Clin Endocrinol Metab 2002; 87: 489–499.

493

10.

infertile women. Thyroid 2002; 12:997-1001.

494

495

Poppe K, Glinoer D, Van Steirteghem A et al. Thyroid dysfunction and autoimmunity in

11.

Iravani AT, Saeedi MM, Pakravesh J, Hamidi S & Abbasi M. Thyroid autoimmunity and

496

recurrent spontaneous abortion in Iran: a case-control study. Endocr Pract 2008; 14: 458-

497

464.

498

12.

outcome in women undergoing assisted reproduction. Hum Reprod 1999 14: 2886-2890.

499

500

13.

14.

15.

507

Millar LK, Wing DA, Leung AS et al. Low birth weight and preeclamsia, in pregnancies complicated by hyperthyroidism. Obstet Gynecol 1994; 84: 946-949.

505

506

Anasti JN, Flack MR, Froehlich J et al. A potential novel mechanism for precocious puberty in juvenile hypothyroidism. J Clin Endocrinol Metab 1995; 80: 276-279.

503

504

Aghajanova L, Lindeberg M, Carlsson IB et al. Receptors for thyroid-stimulating hormone and thyroid hormones in human ovarian tissue. Reprod Biomed Online 2009; 18: 337-347.

501

502

Kutteh WH, Schoolcraft WB, Scott RT Jr. Antithyroid antibodies do not affect pregnancy

16.

Krassas GE, Poppe K, Glinoer D. Thyroid function and human reproductive health. Endocr Rev 2010; 31: 702–755.

25

508

17.

precede thyroid dysfunction or goitre. J Postgrad Med 1993; 39: 137–141.

509

510

18.

19.

Benson RC, Dailey ME. The menstrual pattern in hyperthyroidism and subsequent posttherapy hypothyroidism. Surg Gynecol Obstet 1955; 100: 19–26.

513

514

Krassas GE, Pontikides N, Kaltsas T et al. Menstrual disturbances in thyrotoxicosis. Clin Endocrinol (Oxford) 1994; 40: 641–644.

511

512

Joshi JV, Bhandarkar SD, Chadha M et al. Menstrual irregularities and lactation failure may

20.

Tanaka T, Tamai H, Kuma K et al. Gonadotropin response to luteinizing hormone releasing

515

hormone in hyperthyroid patients with menstrual disturbances. Metabolism 1981; 30:

516

323–326.

517

21.

Endocrinol Metab 1952; 12: 846–855.

518

519

Sturgis SH, Lerman J & Stanbury JB. The menstrual pattern in thyroid disease. J Clin

*22.

Van den Boogaard E, Vissenberg R, Land J et al. Significance of (sub)clinical thyroid

520

dysfunction and thyroid autoimmunity before conception and in early pregnancy: a

521

systematic review. Hum Reprod Update 2011; 17:605–619.

522

23.

1981; 2: 1278.

523

524 525

Bohnet HG, Fiedler K & Leidenberger FA. Subclinical hypothyroidism and infertility. Lancet

24.

Ablovich M, Mitelberg L, Allami C et al. Subclinical hypothyroidism and thyroid autoimmunity in women with infertility. Gynocol endocrinol 2007; 23: 279-283.

26

526

25.

Med 1999; 44: 455-457.

527

528

Licoln SR, Ke RW, Kuttek Wf. Screening for hypothyroidism in infertile women. J Reprod

26.

Feldthusen AD, Pedersen PL, Larsen J et al. Impaired fertility associated with subclinical

529

hypothyroidism and thyroid autoimmunity: the Danish General Suburban Population

530

Study. J Pregnancy 2015; 132718. doi: 10.1155/2015/132718.

531

27.

Associated With Unexplained Infertility. J Clin Endocrinol Metab 2018; 103: 632–639.

532

533

Jokar TO, Lindsay T. Fourman LT et al. Higher TSH Levels Within the Normal Range Are

28.

Moreno-Reyes R, Glinoer D, Van Oyen H & Vandevijvere S. High prevalence of thyroid

534

disorders in pregnant women in a mildly iodine-deficient country: a population-based

535

study. J Clin Endocrinol Metab 2013 98: 3694-3701.

536

29.

and subsequent fetal death. Thyroid 2010. 20: 989-993.

537

538

Ashoor G, Maiz N, Rotas M et al. Maternal thyroid function at 11 to 13 weeks of gestation

30.

Benhadi N, Wiersinga WM, Reitsma JB et al. Ethnic differences in TSH but not in free T4

539

concentrations or TPO antibodies during pregnancy. Clin Endocrinol (Oxf) 2007; 66:765-

540

770.

541

31.

ethnicity. Clin Chem 2007. 53: 1658-1664.

542

543 544

La'ulu SL, Roberts WL. Second-trimester reference intervals for thyroid tests: the role of

32.

Abbassi-Ghanavati M, Casey BM, Spong CY et al. Pregnancy outcomes in women with thyroid peroxidase antibodies. Obstet Gynecol 2010; 116:381-386.

27

545

33.

Wasserman EE, Nelson K, Rose NR et al.. Maternal thyroid autoantibodies during the third

546

trimester and hearing deficits in children: an epidemiologic assessment. Am J Epidemiol

547

2008; 167: 701-710.

548

34.

Pearce EN, Oken E, Gillman MW et al. Association of first-trimester thyroid function test

549

values with thyroperoxidase antibody status, smoking, and multivitamin use. Endocr Pract

550

2008 14: 33-39.

551

35.

Moleti M, Lo Presti VP, Mattina F et al. Gestational thyroid function abnormalities in

552

conditions of mild iodine deficiency: early screening versus continuous monitoring of

553

maternal thyroid status. Eur J Endocrinol 2009; 160: 611-617.

554

36.

McElduff A, Morris J. Thyroid function tests and thyroid autoantibodies in an unselected

555

population of women undergoing first trimester screening for aneuploidy. Aust N Z J

556

Obstet Gynaecol 2008. 48: 478-480.

557

37.

polycystic ovary syndrome. Eur J Endocrinol 2015; 172:R9-21.

558

559 560

561

Gaberšček S, Zaletel K, Schwetz V et al. Mechanisms in endocrinology: thyroid and

38.

Eisenberg VH, Zolti M, Soriano D. Is there an association between autoimmunity and endometriosis? Autoimmun Rev 2012; 11:806-814.

*39. Unuane D, Velkeniers B, Anckaert E et al. Thyroglobulin autoantibodies: is there any

562

added value in the detection of thyroid autoimmunity in women consulting for fertility

563

treatment? Thyroid 2013. 23: 1022-1028.

28

564

40.

outcome following in vitro fertilization and embryo transfer. Int J Med Sci 2012; 9:121-125.

565

566

Zhong YP, Ying Y, Wu HT et al. Relationship between antithyroid antibody and pregnancy

*41. Toulis KA, Goulis DG, Venetis CA et al. Risk of spontaneous miscarriage in euthyroid

567

women with thyroid autoimmunity undergoing IVF: a meta-analysis. Eur J Endocrinol 2010;

568

162:643-652.

569

*42. Busnelli A, Paffoni A, Fedele L & Somigliana E. The impact of thyroid autoimmunity on

570

IVF/ICSI outcome: a systematic review and meta-analysis. Hum Reprod Update 2016;

571

22:793-794.

572

43.

Lukazuk K, Kunicki M, Kulwikowska P et al. The impact of the presence of antithyroid

573

antibodies on pregnancy outcome following intracytoplasmatic sperm injection‑ICSI and

574

embryo transfer in women with normal thyreotropine levels. J Endocrinol Invest 2015. 38:

575

1335-1343.

576

44.

outcomes of assisted reproduction? J Obstet Gynaecol 2016; 36: 213-217.

577

578

Sakar MN, Unal A, Atay AE, et al. Is there an effect of thyroid autoimmunity on the

45.

Tan S, Dieterle S, Pechlavanis S, et al. Thyroid autoantibodies per se do not impair

579

intracytoplasmic sperm injection outcome in euthyroid healthy women Eur J Endocrino

580

2014; 170: 495-500.

581

46.

Unuane D, Velkeniers B, Deridder S et al. Impact of thyroid autoimmunity on cumulative

582

delivery rates in in vitro fertilization/intracytoplasmic sperm injection patients. Fertil Steril

583

2016; 106:144-150. 29

584

47.

Poppe K, Autin C, Veltri F, et al. Thyroid autoimmunity and intracytoplasmic sperm

585

injection outcome: a systematic review and meta-analysis. J Clin Endocrinol Metab. 2018;

586

Mar 12. doi: 10.1210/jc.2017-02633.

587

48.

women on live birth rate after intrauterine insemination. Hum Reprod 2017; 32:915-922.

588

589

49.

Twig G, Shina A, Amital H & Shoenfeld Y. Pathogenesis of infertility and recurrent pregnancy loss in thyroid autoimmunity. J Autoimmun 2012; 38: 2-3.

590

591

Unuane D, Velkeniers, Bravenboer B et al. Impact of thyroid autoimmunity in euthyroid

50.

Roberts J, Jenkins C, Wilson R et al. Recurrent miscarriage is associated with increased

592

numbers of CD5/20 positive lymphocytes and an increased incidence of thyroid antibodies.

593

Eur J Endocrinol 1996. 134: 84–86.

594

51.

antibodies and pregnancy loss. Am J Reprod Immunol. 2001; 45:72-77.

595

596

52.

53.

601

Carp HJA, Selmi C, Shoenfeld Y. The autoimmune bases of infertility and pregnancy loss. J Autoimmun 2016; 38:266–274.

599

600

Deroux A, Dumestre-Perard C, Dunand-Faure C et al. Female Infertility and Serum Autoantibodies: a Systematic Review. 2016. Clin Rev Allergy Immunol 2016; 53: 78-86

597

598

Matalon ST, Blank M, Ornoy A & Shoenfeld YThe association between anti-thyroid

54.

Monteleone P, Parrini D, Faviana P, et al. Female Infertility Related to Thyroid Autoimmunity: The Ovarian Follicle Hypothesis. Am J Reprod Immunol 2012 66: 108-114.

30

602

55.

Medici M, De Rijke YB, Peeters R P et al. Maternal Early Pregnancy and Newborn Thyroid

603

Hormone Parameters: The Generation R Study. J Clin Endocrinol Metab 2012; 97: 646 –

604

652.

605

56.

ovarian hyperstimulation. J Clin Endocrinol Metab 2000; 85:545-548.

606

607

Muller A, Verhoeff A, Mantel M et al. Decrease of free thyroxine levels after controlled

57.

Poppe K, Glinoer D, Tournaye H et al. Impact of ovarian hyperstimulation on thyroid

608

function in women with – and without thyroid autoimmunity. J Clin Endocrinol Metab

609

2004; 89:3808-3812.

610

58.

on thyroid function. Thyroid 2008; 18:801-802.

611

612

Poppe K, Glinoer D, Tournaye H, et l. Impact of the ovarian hyperstimulmation syndrome

59.

Zhang Y, Wang H, Pan X et al. Patients with subclinical hypothyroidism before 20 weeks of

613

pregnancy have a higher risk of miscarriage: A systematic review and meta-analysis. PLoS

614

One 2017; 12: e0175708. doi: 10.1371

615

60.

Plowden TC, Schisterman EF, Sjaarda LA et al. Subclinical hypothyroidism and thyroid

616

autoimmunity are not associated with fecundity, pregnancy loss, or live birth. J Clin

617

Endocrinol Metab. 2016; 101:2358–2365.

618 619

61.

Fumarola A, Grani G, Romanzi D et al. Thyroid function in infertile patients undergoing assisted reproduction.Am J Reprod Immunol. 2013; 70: 336-41.

31

620

62.

Reh A, Grifo J , Danoff A . What is a normal thyroid-stimulating hormone (TSH) level?

621

Effects of stricter TSH thresholds on pregnancy outcomes after in vitro fertilization. Fertil

622

Steril 2010; 94, 2920-2922.

623

63.

Chai J, Yeung WY, Lee CY et al. Live birth rates following in vitro fertilization in women

624

with thyroid autoimmunity and/or subclinical hypothyroidism. Clin Endocrinol (Oxf) 2014;

625

80: 122-127.

626

64.

Green KA, Werner MD, Franasiak JM et al. Investigating the optimal preconception TSH

627

range for patients undergoing IVF when controlling for embryo quality. J Assist Reprod

628

Genet 2015; 32:1469-1476. doi: 10.1007/s10815-015-0549-4.

629

65.

Karmon AE, Batsis M, Chavarro JE & Souter I. Preconceptional thyroid-stimulating

630

hormone levels and outcomes of intrauterine insemination among euthyroid infertile

631

women. Fertil Steril 2015; 103; 258-263.

632

66.

Tuncay G, Karaer A, Coskun EI et al. The impact of thyroid-stimulating hormone levels in

633

euthyroid women on intrauterine insemination outcome. BMC Womens Health 2018; 18:

634

51. doi: 10.1186/s12905-018-0541-0.

635

*67. Velkeniers B, Van Meerhaeghe A, Poppe K et al. Levothyroxine treatment and pregnancy

636

outcome in women with subclinical hypothyroidism undergoing assisted reproduction

637

technologies: systematic review and meta-analysis of RCTs. Hum Reprod Update 2013;

638

19 :251-258.

32

639

68.

Negro R, Schwartz A, Gismondi R et al. Universal screening versus case finding for

640

detection and treatment of thyroid hormonal dysfunction during pregnancy. J Clin

641

Endocrinol Metab 2010; 95: 1699-1707.

642

*69. Nazarpour S, Ramezani Tehrani F, Simbar M et al. Effects of levothyroxine treatment on

643

pregnancy outcomes in pregnant women with autoimmune thyroid disease. Eur J

644

Endocrinol 2017; 176: 253-265.

645

*70. Casey BM, Thom EA, Peaceman AM et al. Treatment of subclinical hypothyroidism or hypothyroxinemia in pregnancy. N Engl J Med 2017; 376:815-25.

646

647

71.

cognitive function. N Engl J Med 2012; 366: 493-501.

648

649

Lazarus JH, Bestwick JP, Channon S et al Antenatal thyroid screening and childhood

*72. Nazarpour S, Ramezani Tehrani F, Simbar M et al Effects of Levothyroxine on Pregnant

650

Women With Subclinical Hypothyroidism, Negative for Thyroid Peroxidase Antibodies. J

651

Clin Endocrinol Metab 2018; 103: 926-935.

652

73.

Negro R, Mangieri T, Coppola L et al. Levothyroxine treatment in thyroid peroxidase

653

antibody-positive women undergoing assisted reproduction technologies: a prospective

654

study. Hum Reprod 2005. 20: 1529-1533.

655

74.

Wang H, Gao H, Chi H et al. Effect of levothyroxine on miscarriage among women with

656

normal thyroid function and thyroid autoimmunity undergoing in vitro fertilization and

657

embryo transfer: a randomized clinical trial. JAMA 2017; 318:2190–2198.

33

658

*75. Dhillon-Smith RK, Middleton LJ, Sunner KK, et al. Levothyroxine in Women with Thyroid Peroxidase Antibodies before Conception. N Engl J Med. 2019; 380: 1316-1325.

659

660

76.

Korevaar T, Muetzel R , Medici M , et al. Association of maternal thyroid function during

661

early pregnancy with offspring IQ and brain morphology in childhood: a population-based

662

prospective cohort study. Lancet diabetes Endocrinol 2016; 4: 35-43.

663

77.

Maraka S , Mwangi R , McCoy RG , et al; Thyroid hormone treatment among pregnant

664

women with subclinical hypothyroidism: US national assessment. BMJ 2017; 25;356:i6865.

665

doi: 10.1136/bmj.i6865.

666

34

Women of infertile couples

TSH, Thyroid antibodies

TSH> 4.0mIU/L

Start LT4 6 weeks prior to OH/ART

Follow-up thyroid function every 4 weeks

TSH 0.3-4.0mIU/L

Thy-abs-

Thy-abs +

TSH<0.3mIU/L

fT4 and fT3

fT3 and fT4

normal

increased

Follow-up thyroid function every 4 weeks

If TSH>4.0mUI/L

Check TSH after OH

Treat with ATD or surgery

Start LT4+followup every 4 weeks during pregnancy

Fig 1. Proposed algorithm for clinical approach of thyroid disorders in infertile women considering ART

Table 1. Main characteristics of studies on the association between TAI and IVF/ICSI outcome

Author

Year

Study design

Aim of study

Thyroid antibodies tested

Number TAI +

Number TAI -

Thyroid function status

Main conclusion

Kutteh et al.

1999

Retrospective cohort study

LBR; CPR; MR

TPOAbs; Tg Abs

143

730

Euthyroid

No effect on pregnancy outcome

Muller et al.

1999

Prospective cohort study

CPR; MR; OPR

TPOAbs

25

148

Euthyroid

No effect on pregnancy outcome

Poppe et al.

2003

Prospective cohort study

LBR; CPR; MR

TPOAbs

32

202

Euthyroid

Lower LBR; increased MR

Negro et al.

2005

Prospective cohort study

LBR; CPR; MR; NOR

TPOAbs

43

576

Euthyroid

Lower LBR; no effect on CPR

Negro et al.

2007

Retrospective cohort study

LBR; CPR; MR; NOR

TPOAbs

42

374

Euthyroid

No effect on pregnancy outcome

Kilic et al.

2008

Prospective cohort study

CPR; MR; NOR

TPOAbs; Tg Abs

23

31

Euthyroid

lower CPR

Zhong et al.

2012

Retrospective cohort study

CPR; MR; IR; NOR; FR

TPOAbs; Tg Abs

90

676

Not specified

lower CPR, FR,IR and higher MR

Karacan et al.

2013

Prospective cohort study

NOR; FR; IR; CPR; MR; OPR

TPOAbs; Tg Abs

34

219

Euthyroid

No effect on pregnancy outcome

Mintziori et al.

2014

Retrospective cohort study

LBR;NOR; CPR; MR

TPOAbs; Tg Abs

15

67

Euthyroid

No effect on pregnancy outcome

Tan et al.

2014

Retrospective cohort study

LBR; CPR; MR

TPOAbs; Tg Abs

110

725

Euthyroid

No effect on pregnancy outcome

Chai et al.

2014

Retrospective cohort study

LBR; CPR; MR

TPOAbs; Tg Abs

89

419

Euthyroid

No effect on pregnancy outcome

Lukasuk et al.

2015

Retrospective cohort study

LBR; CPR; MR; NOR; IR

TPOAbs

114

551

Euthyroid

No effect on pregnancy outcome

Litwicka et al.

2015

Prospective cohort study

LBR; CPR; MR, NOR

TPOAbs; Tg Abs

60

134

Euthyroid

Lower LBR; increased MR

Sakar et al

2016

Prospective cohort study

CPR; OPR; MR; NOR

Not specified

49

202

Not specified

No effect on pregnancy outcome

Unuane et al

2016

Retrospective cohort study

LBR; CPR; MR

TPOAbs

333

2019

Euthyroid

No effect on pregnancy outcome

LBR= Life birth rate; CPR= Clinical pregnancy rate; OPR= Ongoing pregnancy rate; MR= Miscarriage rate; IR= Implantation rate; NOR= Number of oocytes retrieved; FR= Fertilization rate