Risk of Adverse Pregnancy Outcomes in Young Women with Thyroid Cancer: A Systematic Review and Meta-Analysis

Simple Summary This meta-analysis of 22 articles investigated whether thyroidectomy or radioactive iodine treatment (RAIT) in patients with differentiated thyroid cancer was associated with an increase in adverse pregnancy outcomes, such as miscarriage, preterm delivery, and congenital malformations. The results of this meta-analysis suggest that thyroid cancer treatment, including RAIT, is not associated with an increased risk of adverse pregnancy outcomes, including miscarriage, preterm labor, and congenital anomalies. Abstract This meta-analysis investigated whether thyroidectomy or radioactive iodine treatment (RAIT) in patients with differentiated thyroid cancer (DTC) was associated with an increase in adverse pregnancy outcomes, such as miscarriage, preterm delivery, and congenital malformations. A total of 22 articles (5 case-control and 17 case series studies) from 1262 studies identified through a literature search in the PubMed and EMBASE databases from inception up to 13 September 2021 were included. In patients with DTC who underwent thyroidectomy, the event rates for miscarriage, preterm labor, and congenital anomalies were 0.07 (95% confidence interval [CI], 0.05–0.11; 17 studies), 0.07 (95% CI, 0.05–0.09; 14 studies), and 0.03 (95% CI, 0.02–0.06; 17 studies), respectively. These results are similar to those previously reported in the general population. The risk of miscarriage or abortion was increased in patients with DTC when compared with controls without DTC (odds ratio [OR], 1.80; 95% CI, 1.28–2.53; I2 = 33%; 3 studies), while the OR values for preterm labor and the presence of congenital anomalies were 1.22 (95% CI, 0.90–1.66; I2 = 62%; five studies) and 0.73 (95% CI, 0.39–1.38; I2 = 0%; two studies) respectively, which showed no statistical significance. A subgroup analysis of patients with DTC according to RAIT revealed that the risk of miscarriage, preterm labor, or congenital anomalies was not increased in the RAIT group when compared with patients without RAIT. The results of this meta-analysis suggest that thyroid cancer treatment, including RAIT, is not associated with an increased risk of adverse pregnancy outcomes, including miscarriage, preterm labor, and congenital anomalies.


Introduction
According to recent cancer statistics, approximately 75% of differentiated thyroid cancer (DTC) occurs in women, with the highest incidence found in those aged 50-59 years in the United States [1]. DTC is one of the most common cancers affecting women aged 15-39 years, and recent studies have shown an increase in the incidence of DTC in this The present study was registered in the "International Platform of Registered Systematic Review and Meta-Analysis Protocols" in 2022 (INPLASY202240075) and was conducted according to PRISMA guidelines.
A literature search was conducted according to the protocol recommended by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (Table S1). Two investigators (S.M. and Y.J.P.) refined data extraction tables prior to data extraction. These two investigators searched citation databases, including PubMed and EMBASE (from inception until 13 September 2021), and extracted data independently using the predefined tables for data extraction. Discrepancies were resolved by discussion with a third investigator (K.H.Y.). Search terms included combinations of the following: ("Pregnancy"), ("Infertility"), ("Birth"), ("abortion"), ("miscarriage"), ("preterm") OR ("ovary") AND ("thyroid cancer") in the title or abstract.

Study Selection
Studies with the following characteristics were included: (1) population: pregnant women aged ≥20 years; (2) intervention: total thyroidectomy with/without RAIT or subtotal thyroidectomy; (3) comparators: pregnant women without thyroid cancer in casecontrol studies (there were no comparators in case series studies); (4) outcomes: miscarriage Cancers 2022, 14, 2382 3 of 16 or abortion, preterm delivery, and congenital malformations; and (5) study design: casecontrol or case series designs using a registry of patients with thyroid cancer.
We excluded studies with the following characteristics: (1) articles on animal studies or in vivo experiments; (2) articles that included only abstracts; (3) non-original articles, including expert opinions or reviews; and (4) studies with insufficient information on adverse pregnancy outcomes.

Quality Assessment
The Newcastle-Ottawa Quality Assessment Scale was used to assess the methodological quality of case-control studies [23]. Based on eight items, a maximum of nine points were awarded to each study, categorized into three broad perspectives: selection, comparability, and exposure. Studies with a score of 7 or higher were defined as having a low risk of bias [24]; case series study designs were considered to have a high risk of bias owing to the lack of control data. Any discrepancies were resolved through a discussion with a third investigator (K.H.Y.).

Data Analyses and Statistical Methods
The event rate of studies was estimated based on the incidence of adverse pregnancy outcomes in patients with thyroid cancer. The pooled event rate was calculated for each adverse pregnancy outcome using a random-effects model. Odds ratios (ORs) and 95% confidence interval (CIs) were computed for each study using the Mantel-Haenszel method. Pooled ORs were calculated for each adverse pregnancy outcome based on thyroid cancer treatment using a random-effects model.
The heterogeneity among the studies was tested using Higgins' I 2 statistic, where I 2 ≥ 50% indicated heterogeneity. Publication bias was tested using Egger's test and a funnel plot. In addition, to evaluate the effect of RAIT on adverse pregnancy outcomes, we conducted a subgroup analysis of studies that included patients who received RAIT. All statistical analyses and graphical presentations were conducted using the Comprehensive Meta-Analysis software version 3 (Biostat Inc., Englewood, NJ, USA).

Risk of Bias Assessment
The Newcastle-Ottawa Quality Assessment Scale for case-control studies revealed that four [25,27,30,42] out of five case-control studies had a low or moderate risk. (Table S2). One study [31] had a high risk of bias. Five studies classified as case series studies [14,20,29,35,36] were included in subgroup analysis, which compared the risk of adverse pregnant outcomes in patients with DTC according to RAIT. Two studies [14,20] had a low or moderate risk, and three studies [29,35,36] had a high risk of bias.

Risk of Bias Assessment
The Newcastle-Ottawa Quality Assessment Scale for case-control studies revealed that four [25,27,30,42] out of five case-control studies had a low or moderate risk. (Table  S2). One study [31] had a high risk of bias. Five studies classified as case series studies [14,20,29,35,36] were included in subgroup analysis, which compared the risk of adverse pregnant outcomes in patients with DTC according to RAIT. Two studies [14,20] had a low or moderate risk, and three studies [29,35,36] had a high risk of bias.
The event rate for preterm labor among cases of pregnancy in the random-effects model was 0.07 (95% CI, 0.05-0.09; I 2 = 82.0%) ( Figure 3A). Five case-control studies [25,27,30,31,42] were included to compare the risk of preterm labor associated with thyroid cancer treatment. The OR for preterm labor was 1.22 (95% CI, 0.90-1.66; I 2 = 62%) in patients with thyroid cancer when compared with those without thyroid cancer ( Figure 3B), which was not significantly different. Publication bias was not detected (Egger's test: p = 0.56).
The event rate for preterm labor among cases of pregnancy in the random-effects model was 0.07 (95% CI, 0.05-0.09; I 2 = 82.0%) ( Figure 3A). Five case-control studies [25,27,30,31,42] were included to compare the risk of preterm labor associated with thyroid cancer treatment. The OR for preterm labor was 1.22 (95% CI, 0.90-1.66; I 2 = 62%) in patients with thyroid cancer when compared with those without thyroid cancer ( Figure  3B), which was not significantly different. Publication bias was not detected (Egger's test: p = 0.56).
The subgroup analysis of studies that included patients with an interval of 1 year or more between conception and RAIT revealed that the risk of miscarriage or abortion, preterm labor, and congenital anomalies did not differ between patients who were treated with RAIT and those who were not ( Figure 6). Cancers 2022, 14, x FOR PEER REVIEW 10 of 17 Figure 5. Effect of RAIT on adverse pregnancy outcomes. (A) Miscarriage or abortion, (B) preterm labor, and (C) congenital anomalies. * The study was classified as a case series design because one arm data of patients with thyroid cancer was used in the study. Studies referenced: [14,20,22,26,[28][29][30][31][32][33][34][35][36][38][39][40][41].
The subgroup analysis of studies that included patients with an interval of 1 year or more between conception and RAIT revealed that the risk of miscarriage or abortion, preterm labor, and congenital anomalies did not differ between patients who were treated with RAIT and those who were not ( Figure 6). Figure 5. Effect of RAIT on adverse pregnancy outcomes. (A) Miscarriage or abortion, (B) preterm labor, and (C) congenital anomalies. * The study was classified as a case series design because one arm data of patients with thyroid cancer was used in the study. Studies referenced: [14,20,22,26,[28][29][30][31][32][33][34][35][36][38][39][40][41].

Discussion
In this meta-analysis, the risk of adverse pregnancy outcomes, including miscarriage, preterm delivery, and congenital anomalies, did not differ between pregnant women with or without thyroid cancer. In the subgroup analysis, RAIT did not increase the risk of adverse pregnancy outcomes in patients with DTC treated with RAIT when compared with those who did not receive RAIT.
In the treatment of DTC, thyroidectomy and RAIT are applied as standard treatments [5]. Total thyroidectomy can lead to postoperative hypothyroidism. In addition, postoperative hypothyroidism reportedly occurs in approximately 30% of patients even after subtotal thyroidectomy [43]. Considering that many patients with DTC undergo thyroid hormone suppression therapy, these patients may experience various thyroid functional statuses (euthyroid, subclinical/overt, hyperthyroid, or hypothyroid) according to the individual thyroid-stimulating hormone target or compliance with levothyroxine [44].
Based on studies emphasizing that subclinical hyperthyroidism is not associated with maternal or neonatal complications, the American Thyroid Association recommends that patients with thyroid cancer maintain the same thyroid-stimulating hormone goal before and during pregnancy [5]. Nevertheless, epidemiological studies on the effects of Figure 6. The risk of adverse pregnancy outcomes in patients with an interval of 1 year or more between conception and RAIT. Effect of RAIT on adverse pregnancy outcomes. (A) Miscarriage or abortion, (B) preterm labor, and (C) congenital anomalies. * The study was classified as a case series design because one arm data of patients with thyroid cancer was used in the study. Studies referenced: [14,20,22,26,[28][29][30][31][32][33][34][35][36][38][39][40][41].

Discussion
In this meta-analysis, the risk of adverse pregnancy outcomes, including miscarriage, preterm delivery, and congenital anomalies, did not differ between pregnant women with or without thyroid cancer. In the subgroup analysis, RAIT did not increase the risk of adverse pregnancy outcomes in patients with DTC treated with RAIT when compared with those who did not receive RAIT.
In the treatment of DTC, thyroidectomy and RAIT are applied as standard treatments [5]. Total thyroidectomy can lead to postoperative hypothyroidism. In addition, postoperative hypothyroidism reportedly occurs in approximately 30% of patients even after subtotal thyroidectomy [43]. Considering that many patients with DTC undergo thyroid hormone suppression therapy, these patients may experience various thyroid functional statuses (euthyroid, subclinical/overt, hyperthyroid, or hypothyroid) according to the individual thyroid-stimulating hormone target or compliance with levothyroxine [44].
Based on studies emphasizing that subclinical hyperthyroidism is not associated with maternal or neonatal complications, the American Thyroid Association recommends that patients with thyroid cancer maintain the same thyroid-stimulating hormone goal before and during pregnancy [5]. Nevertheless, epidemiological studies on the effects of thyroid dysfunction caused by thyroid hormone suppression therapy or thyroidectomy in terms of adverse pregnancy outcomes are lacking [45].
This meta-analysis with case-control studies demonstrated an increased risk of miscarriage or abortion, and this result may be biased because of the small number of studies and considering the potentially significant publication bias. In addition, although this meta-analysis showed that thyroid cancer treatment did not increase the risk of preterm labor, significant heterogeneity was noted among the included studies. Two studies showed a higher risk of miscarriage and preterm labor in patients with DTC [25,30]. Blackburn et al. reported a higher incidence of miscarriage and preterm labor in patients with DTC.
However, the hazard ratio was not significant after adjusting for comorbidities [25]. Garsi et al. also reported that patients with DTC had a significantly higher risk of mis-carriage and preterm labor after receiving treatment for DTC than before treatment [30]. Considering the advanced age after treatment compared with that before treatment, the higher incidence of adverse pregnancy outcomes after DTC treatment may be the effect of advanced maternal age [30]. This meta-analysis provides data on the event rates in patients with DTC.
The event rate for miscarriage was 0.07 in patients with DTC, which is similar to those in the general population from national representative data (0.01-0.18) [46][47][48]. Four European case series studies [22,33,39,41] and one Indian case series study [34], which reported the obstetric history of women with DTC, showed that the prevalence of miscarriage at least once in their lifetime was similar to that in the general population in the EPIC study [49] or general Indian population [50]. The events rates for preterm labor were 0.07 in patients with DTC, which were similar to those in the general population (0.06-0.23) [51][52][53][54].
The event rates for congenital anomalies were 0.03 in patients with DTC, which were similar to those in the general population (0.01-0.03) [51,52,55,56]. Nonetheless, a large population-based study conducted by Kim et al. revealed a higher risk of congenital anomalies in women with DTC compared with that in the general population from the Korean National Health Insurance Service. The study by Kim et al. included more pregnant women aged >35 years than the study on the general population (34% vs. 15.9%), which could have resulted in a higher risk [20,56]. Although we could not perform subgroup analysis according to thyroid functional status, the present study provides substantial evidence that thyroid cancer treatment does not increase the risk of adverse pregnancy outcomes compared to women without DTC.
RAIT is known to be able to affect gonadal tissues [16,17,42]. In men, an association between RAIT and a transient reduction in sperm count, elevated follicle-stimulating hormone (FSH) levels, and testicular damage have been reported [16,57]. A recent longitudinal prospective study revealed a statistically significant increase in the number of chromosomal abnormalities in sperm at 3 and 13 months after RAIT with 100 mCi [58]. Therefore, contraception is usually recommended at least for 3 months in men after RAIT [5]. In addition, high radioactive iodine (RAI) activities of 500-800 mCi increased the risk of sustained elevation of FSH [5].
Therefore, the American Thyroid Association (ATA) recommends sperm banking for men who need cumulative RAI activities greater than 400 mCi [5,59]. Proper hydration, frequent urination, and avoidance of constipation may also be helpful in reducing radiation exposure to the gonads [60]. In women, RAIT has been reported to be associated with oligomenorrhea, transient secondary amenorrhea, and premature menopause [5].
About 12-31% of menstrual irregularities and 8-16% of amenorrhea [17] or a significant decrement of anti-Müllerian hormone (AMH) [18] in the first year after RAIT have been reported. Although, many previous epidemiologic studies have not found conclusive evidence for decreased fertility in these women [18,30,34,35,61], there is significant heterogeneity between studies. Research reported that RAIT was associated with delayed childbearing and reduced birthrates in a specific population of advanced age (>35 years) [19]. Therefore, in women over 35 years of age with low-risk DTC, RAIT should be carefully considered when planning pregnancy [19,62].
These women should be informed and counseled about the potential deleterious effects on fertility and fertility [63]. AMH measurement is suggested as a good option to estimate ovarian reserve for fertility patients in RAIT decision-making process, although it cannot fully estimate the risk of infertility [62]. As suggested by the American Society of Clinical Oncology, interventions for preserving fertility, including oocyte cryopreservation, may be useful particularly in women with a limited ovarian reserve [63,64], although its evidence in women with RAIT remains lacking. Further studies are warranted.
The ATA recommends that reproductive-age women receiving RAIT should undergo negative screening evaluation for pregnancy and should avoid pregnancy for 6-12 months after receiving RAI [5]. Despite these recommendations, RAI may inadvertently be administered to pregnant women because of a clinician's negligence or false-negative pregnancy test results [65]. The effects of inadvertent exposure on embryos and fetuses vary depending on the pregnancy stage and absorbed RAI dose [65]. Exposure to RAI during the very early stage of pregnancy may result in cellular damage and embryo death although it is unlikely to induce congenital anomalies in the surviving embryos [66].
At 3-7 weeks after conception, exposure to RAI can lead to congenital anomalies, such as microcephaly, cleft palate, and genital deformities [66]. Considering that the thyroid gland is formed by 10-12 weeks of gestation, exposure to RAI after 10 weeks of gestation can result in fetal thyroid ablation [67][68][69]. Exposure after 8 weeks of gestation can impair the central nervous system. In particular, mental retardation has been frequently reported with exposure at 8-25 weeks after conception [70,71]. Additionally, exposure to RAI can increase the risk of some cancers, such as leukemia, skin cancer, lung cancer, breast cancer, and thyroid cancer [71,72].
When inadvertent exposure occurs, potassium iodide can be helpful in reducing fetal exposure to RAI within 12 h of RAI administration [65,73]. However, data on therapeutic abortion are limited [74]. During pregnancy, congenital anomalies should be closely monitored. Levothyroxine supplementation should be considered to maintain maternal thyroid hormone levels at the high end of the normal range. For neonates, thyroid function should be evaluated, and levothyroxine supplementation should be initiated to prevent any neurological impairment [65].
This meta-analysis provides data on the event rates in patients who received RAIT. The event rates in patients receiving RAIT were 0.09, 0.08, and 0.04 for miscarriage, preterm labor, and congenital anomalies, respectively, which were similar to those observed in the general population [46][47][48][51][52][53][54][55][56]. In addition, this meta-analysis with case-control studies showed that RAIT did not increase the risk of miscarriage, preterm labor, and congenital anomalies without significant heterogeneity among the included studies compared with those with DTC who did not receive RAIT.
The strengths of this study include the collection of evidence through a rigorous systematic review and meta-analysis. However, the present study has certain limitations. We could not adjust for the complications of DTC treatment, including hyperparathyroidism, the stage of DTC, and recurrence, because corresponding data were unavailable. In addition, a subgroup analysis according to total thyroidectomy or hemilobectomy, thyroid functional status, and RAIT dosage was not conducted due to the lack of data.

Conclusions
The meta-analysis results suggest that thyroid cancer treatment is not associated with an increased risk of adverse pregnancy outcomes. In particular, RAIT after thyroidectomy was not found to increase the risk of adverse pregnancy outcomes in patients with DTC compared with those with DTC who did not receive RAIT.

Data Availability Statement:
The data presented in this study are available upon reasonable request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.