Higher Maternal Thyroid-Stimulating Hormone Levels Tied to Depression and Anxiety in Pregnancy

A routine blood test that nearly every pregnant woman receives may carry far more psychological information than clinicians currently extract from it. New research from Türkiye suggests that when thyroid-stimulating hormone, or TSH, drifts above the 2.5 mIU/L threshold during pregnancy, women report measurably more symptoms of depression and anxiety, even when their thyroid function would otherwise be considered unremarkable. The study, published in Reproductive Sciences, adds a psychological dimension to one of the most debated questions in modern obstetric medicine: where exactly the line should be drawn between normal and abnormal thyroid activity in expectant mothers.

The thyroid is a small, butterfly-shaped gland at the base of the neck, but its influence during pregnancy is enormous. Its hormones help regulate fetal brain development, metabolic rate, and cardiovascular adaptation to gestation. In early pregnancy, the placental hormone human chorionic gonadotropin suppresses TSH, the pituitary signal that tells the thyroid to work, which is why pregnancy-specific reference ranges for TSH are lower than those used in the general population. When the gland cannot keep pace, TSH rises, a pattern often labeled subclinical hypothyroidism when free thyroxine remains normal. Whether mild elevations of this signal harm the mother’s mental state has remained contested, with previous studies producing a patchwork of conflicting results.

Researchers led by Havva Betul Bacak of the University of Health Sciences Gaziosmanpasa Training and Research Hospital in Istanbul designed a prospective observational cohort to address that question directly. Between November 2025 and January 2026, they recruited 208 pregnant women attending antenatal care at their institution. Before any clinical examination or blood draw, each participant completed a battery of well-validated self-report questionnaires: the Beck Depression Inventory, the Beck Anxiety Inventory, the Arizona Sexual Experiences Scale, and the 13-item Turkish Short-Form Quality of Life questionnaire. Completing the psychometric instruments before the medical evaluation was a deliberate design choice, intended to prevent a physician’s assessment or discussion of thyroid results from coloring the women’s self-reported emotional state.

After predefined exclusions, the analytic sample settled at 192 women. The investigators excluded anyone with autoimmune thyroid disease, current thyroid medication use, and other conditions that could confound thyroid physiology, and women whose TSH measured at or above 2.5 mIU/L were referred for endocrinology consultation as part of standard care. The final cohort split into 102 women with TSH below 2.5 mIU/L and 90 women at or above that cutoff. Critically, the two groups were statistically indistinguishable in age, gestational age, number of previous pregnancies, parity, and body mass index, which strengthens the interpretation that differences in psychological scores were not simply reflections of different demographic profiles.

The headline result was clear and consistent for mood. Median scores on the Beck Depression Inventory were 13.0 in the higher-TSH group versus 10.0 in the lower-TSH group, a difference that reached statistical significance with a p-value of 0.002. The Beck Anxiety Inventory told a similar story, with median scores of 11.0 versus 8.0 and a p-value of 0.009. While these are moderate symptom scores rather than diagnoses of major depressive disorder, the direction and consistency of the shift across two independent instruments is notable. Perinatal depression affects roughly one in seven women worldwide by some estimates, and anxiety disorders in pregnancy are similarly common, so any physiological marker that tracks with symptom burden is clinically interesting even if the effect is modest.

Equally interesting was what did not change. Scores on the sexual function scale and the quality-of-life instrument showed no statistically significant differences between the TSH groups, with p-values of 0.400 and 0.068 respectively. This dissociation is intriguing because mood, sexual function, and quality of life are typically intertwined. The authors suggest that quality-of-life changes may emerge at higher degrees of thyroid dysfunction, or that pregnancy itself dampens the measurable influence of mild thyroid signals on sexuality, since expectant mothers report sexual difficulties at high rates regardless of thyroid status. The mood findings, by contrast, appear to register the thyroid signal more sensitively.

Beyond the fixed 2.5 mIU/L cutoff, the researchers performed exploratory pairwise analyses using finer TSH strata. The first detectable separation in depression scores appeared between women below 2.5 mIU/L and those in the 2.5 to below 4.5 mIU/L band, with a p-value of 0.007, while comparisons between the 2.5 to below 4.5 and the 4.5-and-above groups were not significant for any of the four measures. In complementary analyses treating TSH as a continuous variable, higher maternal TSH was weakly but monotonically associated with higher depression and anxiety scores and lower quality of life, with no evidence of a nonlinear or threshold-like relationship. In other words, the data hint that the psychological gradient may begin at the lower end of the elevated range rather than accumulating only after TSH climbs far above it.

The authors are careful, and appropriately so, about what these findings cannot establish. The study did not measure free thyroxine or thyroid peroxidase antibodies, two variables that international guidelines consider essential for classifying thyroid dysfunction during pregnancy. It relied on a fixed pragmatic TSH cutoff rather than trimester-specific reference intervals, which recent individual-participant-data meta-analyses have shown vary substantially across populations and assay methods. The statistical comparisons were unadjusted, without multivariable models accounting for potential confounders, and the single-center design limits generalizability. The team explicitly frames the results as exploratory and hypothesis-generating, not as evidence of an independent or causal association between TSH and mental health. Reverse causation remains plausible too: women experiencing more depressive or anxious symptoms may have different neuroendocrine profiles for reasons unrelated to thyroid pathology.

Nevertheless, the study lands in an active and consequential scientific conversation. The American Thyroid Association, the American College of Obstetricians and Gynecologists, and other professional bodies have long disagreed about whether a 2.5 mIU/L TSH threshold should trigger treatment, and large trials of levothyroxine therapy for mild thyroid abnormalities in pregnancy have generally failed to show clear benefits for cognitive or obstetric outcomes. Prior research has linked antenatal thyroid measures to postpartum depression, and meta-analyses have connected overt hypothyroidism with female sexual dysfunction, but data on subclinical elevations and psychological well-being during gestation itself have been sparse and inconsistent. By collecting psychometric data prospectively, before clinical interaction, and by examining sexual function and quality of life alongside mood, the Turkish team has added a rarely explored angle to the debate.

The practical implications, for now, are more about research design than treatment. The monotonic, weak association between TSH and mood symptoms suggests that if a real psychological effect of mild thyroid underactivity in pregnancy exists, it is subtle and distributed across the upper range of normal rather than concentrated above any sharp threshold. That pattern argues for studies that measure free hormones and antibodies, apply trimester-specific reference intervals, adjust for psychosocial confounders, and follow women into the postpartum period, where the consequences of untreated perinatal mood disorders are best documented. In the meantime, the study offers clinicians a reminder that the familiar TSH value on a prenatal panel may be worth reading not only as a metabolic number but as a potential, if still unproven, window into the emotional landscape of pregnancy.

Subject of Research: Maternal thyroid-stimulating hormone levels and their association with depression, anxiety, sexual function, and quality of life during pregnancy.

Article Title: Association of Maternal Thyroid-stimulating Hormone Levels with Psychological Well-being, Sexual Function, and Quality of Life During Pregnancy

Article References: Bacak, H. B., Ketenci Gencer, F., Coskun, E. S., Irice, F., Aciyiyen, Y., Terkan, H., Celikoglu Berker, E., Deligozoglu, S., Kumbasar, S., & Salman, S. (2026). Association of Maternal Thyroid-stimulating Hormone Levels with Psychological Well-being, Sexual Function, and Quality of Life During Pregnancy. Reproductive Sciences. https://doi.org/10.1007/s43032-026-02206-4

Image Credits: AI Generated

DOI: 10.1007/s43032-026-02206-4

Keywords: pregnancy, thyroid-stimulating hormone, TSH, depression, anxiety, sexual function, quality of life, subclinical hypothyroidism, prenatal care, Reproductive Sciences, Association, Maternal