November 22, 2021


Prematurity Awareness Month

By Mikayla Welch

Thyroid Autoantibodies and Adverse Pregnancy Outcomes: A Review

Editor’s Note: November is Prematurity Awareness Month, which provides an opportunity to raise awareness about conditions that may lead to prematurity, particularly preventable ones. The study summarized below was published in the British Medical Journal by Thangaratinam et al in 2011; it is titled, “Association between thyroid autoantibodies and miscarriage and preterm birth: Meta-analysis of evidence.” Their findings of a significant association between thyroid autoantibodies and both preterm birth and miscarriage should compel medical professionals to test women for this risk as a routine part of prenatal care to minimize the chances of poor pregnancy outcomes. The study was summarized by Mikayla Welch while on the FACTS elective in fertility awareness.


Compared to the general population of women, thyroid autoantibodies have been found in higher ranges in women with subfertility and recurrent pregnancy loss. The presence of thyroid autoantibodies is 10-31% in women with subfertility and 17-33% in women with a history of recurrent pregnancy loss. Despite having euthyroid function, the presence of thyroid peroxidase antibodies is associated with miscarriage, preterm birth, and adverse neurodevelopmental conditions in children. While the exact mechanism for these findings remains uncertain, it is thought to be due to a systemic autoimmune state or to subclinical hypothyroidism.

Miscarriage is defined as the loss of pregnancy prior to 20 weeks gestation and affects approximately 1 in 5 women who conceive. Additionally, preterm birth—delivery of a baby between 24 and 37 weeks of gestation—affects 6-10% of pregnancies. Due to the prevalence of these conditions and their potential relationship with thyroid health, the study by Thangaratinam et al aimed to review the association between thyroid autoantibodies and adverse pregnancy outcomes through systematic review and meta-analysis. The authors also assessed the role of levothyroxine treatment on pregnancy outcomes.



Studies were found by searching keywords on Medline, Embase, the Cochrane Library, and SCISEARCH. Additionally, the reference lists of relevant articles were searched for publications not identified in the databases. Articles were included in the review if they consisted of women with normal thyroid function, positive thyroid autoantibodies, and included adverse maternal or fetal outcomes. Studies were excluded from the review if women had overt hypo- or hyperthyroidism. 

Using these methods, 389 citations were identified. The Newcastle-Ottawa scale was used to assess the methodological quality and internal validity of the selected studies. Randomized trials were evaluated for quality and scored on Jadad’s scale. Ultimately, 30 primary articles with 36 studies (31 for miscarriage, 5 for preterm birth) were included in the systematic review.


Of the 31 studies evaluating miscarriage and thyroid autoantibodies, 28 (90%) found a positive association. Additionally, the odds of miscarriage with thyroid autoantibodies were increased with a history of recurrent miscarriages and subfertility. Similarly, all 5 studies (12,566 women) that looked at the relationship between thyroid autoantibodies and preterm birth found a positive association. While the study notes that increasing maternal age is an “independent” risk factor for miscarriage, the authors found no significant difference in age between women who were and were not positive for thyroid autoantibodies.

As mentioned above, one hypothesis for the association between adverse maternal and fetal outcomes and thyroid autoantibodies is underlying relative hypothyroidism. To examine this hypothesis, the authors did a meta-analysis of thyroid stimulating hormone (TSH) levels in six studies. The authors found a statistically significant difference in TSH levels with a weighted mean difference of 0.51 mlU/L higher in the thyroid autoantibody group. 

Levothyroxine therapy

The authors reviewed two randomized studies (187 women) that examined pregnancy outcomes with levothyroxine therapy. While both studies involved women with normal thyroid function and positive thyroid autoantibodies, one used levothyroxine at a dose of 1 μg/kg/day and the second utilized a titrated dose with a mean levothyroxine dose of 49.7 μg/day. The authors found a 52% relative risk reduction in miscarriages with levothyroxine therapy. One of the studies assessed the effect of levothyroxine on preterm births and found a 69% relative risk reduction with treatment. 


From this systematic review and meta-analysis, the authors conclude there is a strong association between thyroid autoantibodies and both preterm birth and miscarriage in women with normal thyroid function. Specifically, the odds of experiencing miscarriage tripled and the likelihood of preterm birth doubled in women who had thyroid autoantibodies. The authors also conclude that levothyroxine therapy may be of benefit for women in these clinical scenarios; however, there needs to be a large, placebo controlled, randomized trial to assess the benefit and any long-term risks. 

Due to the relationships outlined in this review, testing thyroid function and autoantibodies should be considered in every basic hormonal profile, including at the initial prenatal visit. Testing TSH, free thyroxine (free T4), and thyroid peroxidase (TPO) antibodies involves minimal cost to the patient and may be beneficial in preventing miscarriage and preterm birth. 

Clinical Applications of Fertility Awareness-Based Methods

Although the above summary concludes the review of the 2011 article [i] published in the British Medical Journal, other resources exist for these patients that deserve further discussion. Charting the female cycle with fertility awareness-based methods (FABMs) offers medical professionals a useful diagnostic tool for thyroid pathology. Evidence of thyroid disease can be apparent in a patient’s cycle through a number of signs. Thyroid disease may cause a dry cycle, or a lack of cervical mucus progression to fertile-type mucus. It can cause a lower-than-expected basal body temperature, which would be apparent in sympto-thermal charting methods. Furthermore, a short luteal phase, indicating inadequate levels or ratio of progesterone and estradiol, could be caused by thyroid dysfunction. Finally, thyroid disease should be included in the differential diagnosis for anovulation, which is partly due to disruption of follicular development and can be identified through charting.


[1] Thangaratinam S, Tan A, Knox E, Kilby M, Franklyn J, Coomarasamy A. Association between thyroid autoantibodies and miscarriage and preterm birth: Meta-analysis of evidence. British Medical Journal. 2011;342:d2616.

About the Author

Mikayla Welch

Mikayla Welch wrote this review as a fourth-year medical student at Kansas City University College of Osteopathic Medicine in Joplin, MO, where she lives with her husband, Ben. She is originally from Jefferson City, MO, and earned her undergraduate degree from Columbia College in Columbia, MO. She is pursuing residency training in internal medicine with hopes of completing an endocrinology fellowship in the future. She hopes to use information about fertility awareness-based methods to help her future patients.

Pin It on Pinterest