Maternal Hypothyroidism and Documented History of Pregnancy Loss: A Registry Based Cross-Sectional Study of 66,279 Pregnancies in North-Eastern Iran
Background Maternal hypothyroidism is a frequent endocrine disorder in pregnancy and has been linked to pregnancy loss, but most evidence comes from selected hospital populations with limited control for social and psychological factors. We examined whether a physician recorded diagnosis of hypothyroidism is associated with a documented history of pregnancy loss in a large primary care population. Methods Registry based cross-sectional study using the Sina Electronic Health Record System, which covers prenatal care in the primary health care network of Mashhad University of Medical Sciences (1 March 2017–20 September 2025). All pregnancies with a recorded diagnosis of hypothyroidism were enumerated (n = 33,079) and an approximately equal random sample of pregnancies without such a record was drawn from the same source population (n = 33,200; total N = 66,279). All included pregnancies ended in a delivery; the outcome was a documented history of pregnancy loss recorded in the maternal file. Because sampling was conditional on exposure and independent of the outcome, exposure specific prevalences, prevalence ratios and prevalence odds ratios are unbiased for the source population, whereas the pooled prevalence is not. Covariates were pre-specified on clinical and causal grounds and all were retained in the model. We report adjusted prevalence odds ratios (aPOR) from logistic regression and adjusted prevalence ratios (aPR) from modified Poisson regression with robust standard errors. Results A documented history of pregnancy loss was present in 3,578 of 33,079 women with recorded hypothyroidism (10.8%, 95% CI: 10.5–11.2) and in 2,145 of 33,200 women without (6.5%, 95% CI: 6.2–6.7): a prevalence difference of 4.4 percentage points (95% CI: 3.9–4.8) and a crude prevalence ratio of 1.67 (95% CI: 1.59–1.76). After adjustment, hypothyroidism remained associated with the outcome (aPOR = 1.61, 95% CI: 1.51–1.71; aPR = 1.50, 95% CI: 1.42–1.58). The strongest correlate was pregnancy number (aPOR = 12.22, 95% CI: 10.95–13.64 for a fourth or later pregnancy), followed by Iranian nationality (aPOR = 5.54, 95% CI: 4.29–7.16) and residence outside the metropolitan area (aPOR 1.41–2.23). Maternal age was positively associated with the outcome before adjustment but inversely associated after adjustment for pregnancy number (aPOR = 0.50, 95% CI: 0.41–0.61 for women older than 40 years). Depression (aPOR = 1.18), recorded obesity (aPOR = 1.20) and genitourinary disease (aPOR = 1.37) were also independently associated, whereas a positive domestic-violence screen, anxiety, diabetes, hypertension and a history of cancer were not. Conclusions Recorded maternal hypothyroidism was associated with a roughly 50% higher prevalence of documented pregnancy loss. Because the registry records neither thyroid function values, antibody status and levothyroxine treatment, nor the timing of the diagnosis relative to the loss, the analysis cannot establish temporality, distinguish overt from subclinical disease, or separate a biological effect from more complete documentation in women under closer follow up. The findings support the recording of thyroid function values and treatment in structured prenatal fields so that these questions can be addressed directly, and they should not be used to justify a particular screening or treatment policy.
Authors
- Mohammad Saeidi (ORCID: https://orcid.org/0000-0002-5987-9932)
- Amin Moradi (ORCID: https://orcid.org/0009-0009-2404-5215)
- Ehsan Mousa-Farkhani
- Shaghayegh Afzalnia
- Mehdi Norouzi
Institutions
- Baqiyatallah University of Medical Sciences (IR)
- Mashhad University of Medical Sciences (IR)
- Alborz University of Medical Sciences (IR)
Publication Details
- Journal
- Research Square
- Published
- 2026-11-20
- DOI
- https://doi.org/10.21203/rs.3.rs-11574546/v1
- Primary Topic
- Thyroid Disorders and Treatments
- Type
- preprint