Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 18 min ago
Peer-reviewedCC BY 4.0Frontiers in Endocrinology

Higher TSH within the normal range is independently associated with reduced IVF clinical pregnancy odds: a K-means cluster analysis challenging the binary cutoff paradigm

  • Mohamed M. Amin1
  • SI. Hassanein1,2
  1. 1German University in Cairo
  2. 2Biochemistry Department, Faculty of Pharmacy and Biotechnology, German University in Cairo
Read full textPDFDOICite

Abstract

Background: The impact of thyroid-stimulating hormone (TSH) variation within the normal reference range on IVF outcomes remains controversial, with prior studies yielding conflicting results.

Methods: We conducted a retrospective cohort study of 878 euthyroid women (TSH 0.25–4.0 µIU/mL) undergoing fresh IVF/ICSI cycles at Al-Hayat Hospital, Iraq (2022–2024). Patients were classified into four groups using k-means clustering of serum TSH values: Cluster 1 (2.9–4.0 µIU/mL), Cluster 2 (2.2–2.8 µIU/mL), Cluster 3 (1.4–2.0 µIU/mL), and Cluster 4 (0.25–1.3 µIU/mL). Clinical, hormonal, and embryological parameters were compared across clusters, and stepwise logistic regression was performed to adjust for confounders.

Results: Clinical pregnancy was achieved in 474/878 women (54%). As expected, women with pregnancy were younger (30.1 ± 5.1 vs. 31.2 ± 5.5 years, p = 0.003) and had lower mean TSH (1.95 ± 0.85 vs. 2.08 ± 0.96 µIU/mL, p = 0.030). Ovarian reserve and response markers were higher in the pregnancy group, including antral follicle count (AFC) (11.3 ± 4.2 vs. 10.4 ± 4.2, p = 0.001), preovulatory follicle count (PFC) (11.9 ± 4.2 vs. 11.0 ± 4.2, p = 0.003), and oocyte yield (10.4 ± 5.0 vs. 9.6 ± 4.8, p = 0.010). Oocyte maturation (80.5 ± 15.5% vs. 76.9 ± 17.1%, p = 0.003) and fertilization rates (88.0 ± 14.0% vs. 84.2 ± 16.9%, p = 0.003) were also significantly higher, while women without pregnancy required more ampoules (41.2 ± 18.8 vs. 36.6 ± 16.5, p < 0.001). Stepwise regression identified age (OR = 0.97, p = 0.059), ampoules (OR = 0.99, p = 0.024), oocyte yield (OR = 1.04, p = 0.014), maturation rate (OR = 1.10, p = 0.002), fertilization rate (OR = 1.06, p = 0.015), and TSH clusters as independent predictors. Cluster 3 (TSH 1.4–2.0 µIU/mL) showed the highest pregnancy rate (58.2%), while Cluster 1 (TSH 2.9–4.0 µIU/mL) had the lowest (42.7%) and was independently associated with reduced odds (OR = 0.53, 95% CI 0.35–0.80, p = 0.0024). Subgroup analysis revealed that unexplained and ovarian factor infertility were most sensitive to thyroid variation, whereas male factor infertility remained stable and PCO cases diverged.

Conclusions: K-means clustering revealed that women with higher-normal TSH (Cluster 1: 2.9–4.0 µIU/mL) had independently reduced odds of clinical pregnancy (OR = 0.53, 95% CI 0.35–0.80, p = 0.002) compared with Cluster 3 (1.4–2.0 µIU/mL), challenging the adequacy of a single binary cutoff and supporting individualized, data-driven thyroid assessment in IVF.

Related

  1. Integrated Clinical, Biological, and Echocardiographic Factors Independently Associated With In-Hospital Mortality in Infective Endocarditis: A Single-Centre Cohort Study From Morocco
    Tariq El Karimy, Hiba Rizkou, Joumana Elmasrioui +1
    CureusPeer-reviewedCC BYFull textGeneral medicineEurope PMC
  2. Clinical Characteristics Associated with Biochemical Pregnancy Occurrence Among IVF Patients with Tubal Disorders: A Single-Centre Retrospective Cohort Study in Indonesia
    Ilonna Putri Pertiwi, Hanom Husni Syam, Eppy Darmadi Achmad
    International Journal of Women's HealthPeer-reviewedCC BY-NC 4.0Full textObstetrics, gynaecology & paediatricsPMC
  3. High genetic predisposition and vertebral fracture burden in pregnancy and lactation-associated osteoporosis: clinical insights from a Taiwanese cohort.
    Wei-Yih Chiu, Yu-An Chen, Shyang-Rong Shih +6
    Frontiers in endocrinologyPeer-reviewedCC BYFull textSurgery & anaesthesiaEurope PMCCrossref
  4. Chronic erythema nodosum leprosum in Hansen’s disease: a challenging clinical management
    Michele Spinicci, Anna Barbiero, Tommaso Manciulli +9
    Transactions of the Royal Society of Tropical Medicine and HygienePeer-reviewedCC BY 4.0Full textInfectious diseasePMC
  5. Serum T50 is not independently associated with coronary artery calcification in early chronic kidney disease
    Clinical kidney journalPeer-reviewedCC BY-NCFull textGeneral medicineEurope PMC

Cite

BibTeXRIS
BibTeX
@article{amin2026higher,
  title = {{Higher TSH within the normal range is independently associated with reduced IVF clinical pregnancy odds: a K-means cluster analysis challenging the binary cutoff paradigm}},
  author = {Mohamed M. Amin and SI. Hassanein},
  journal = {Frontiers in Endocrinology},
  year = {2026},
  month = oct,
  doi = {10.3389/fendo.2026.1856685},
  url = {https://www.frontiersin.org/articles/10.3389/fendo.2026.1856685/full},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Higher TSH within the normal range is independently associated with reduced IVF clinical pregnancy odds: a K-means cluster analysis challenging the binary cutoff paradigm
AU  - Mohamed M. Amin
AU  - SI. Hassanein
PY  - 2026
DA  - 2026/10/23
JO  - Frontiers in Endocrinology
DO  - 10.3389/fendo.2026.1856685
UR  - https://www.frontiersin.org/articles/10.3389/fendo.2026.1856685/full
AB  - Background: The impact of thyroid-stimulating hormone (TSH) variation within the normal reference range on IVF outcomes remains controversial, with prior studies yielding conflicting results.  Methods: We conducted a retrospective cohort study of 878 euthyroid women (TSH 0.25–4.0 µIU/mL) undergoing fresh IVF/ICSI cycles at Al-Hayat Hospital, Iraq (2022–2024). Patients were classified into four groups using k-means clustering of serum TSH values: Cluster 1 (2.9–4.0 µIU/mL), Cluster 2 (2.2–2.8 µIU/mL), Cluster 3 (1.4–2.0 µIU/mL), and Cluster 4 (0.25–1.3 µIU/mL). Clinical, hormonal, and embryological parameters were compared across clusters, and stepwise logistic regression was performed to adjust for confounders.  Results: Clinical pregnancy was achieved in 474/878 women (54%). As expected, women with pregnancy were younger (30.1 ± 5.1 vs. 31.2 ± 5.5 years, p = 0.003) and had lower mean TSH (1.95 ± 0.85 vs. 2.08 ± 0.96 µIU/mL, p = 0.030). Ovarian reserve and response markers were higher in the pregnancy group, including antral follicle count (AFC) (11.3 ± 4.2 vs. 10.4 ± 4.2, p = 0.001), preovulatory follicle count (PFC) (11.9 ± 4.2 vs. 11.0 ± 4.2, p = 0.003), and oocyte yield (10.4 ± 5.0 vs. 9.6 ± 4.8, p = 0.010). Oocyte maturation (80.5 ± 15.5% vs. 76.9 ± 17.1%, p = 0.003) and fertilization rates (88.0 ± 14.0% vs. 84.2 ± 16.9%, p = 0.003) were also significantly higher, while women without pregnancy required more ampoules (41.2 ± 18.8 vs. 36.6 ± 16.5, p < 0.001). Stepwise regression identified age (OR = 0.97, p = 0.059), ampoules (OR = 0.99, p = 0.024), oocyte yield (OR = 1.04, p = 0.014), maturation rate (OR = 1.10, p = 0.002), fertilization rate (OR = 1.06, p = 0.015), and TSH clusters as independent predictors. Cluster 3 (TSH 1.4–2.0 µIU/mL) showed the highest pregnancy rate (58.2%), while Cluster 1 (TSH 2.9–4.0 µIU/mL) had the lowest (42.7%) and was independently associated with reduced odds (OR = 0.53, 95% CI 0.35–0.80, p = 0.0024). Subgroup analysis revealed that unexplained and ovarian factor infertility were most sensitive to thyroid variation, whereas male factor infertility remained stable and PCO cases diverged.  Conclusions: K-means clustering revealed that women with higher-normal TSH (Cluster 1: 2.9–4.0 µIU/mL) had independently reduced odds of clinical pregnancy (OR = 0.53, 95% CI 0.35–0.80, p = 0.002) compared with Cluster 3 (1.4–2.0 µIU/mL), challenging the adequacy of a single binary cutoff and supporting individualized, data-driven thyroid assessment in IVF.
SN  - 1664-2392
ER  -