This cross-sectional study examined the relationship between central thyroid hormone regulation and chronic kidney disease (CKD) among patients with type 2 diabetes mellitus (T2DM) who had normal peripheral thyroid hormone concentrations. The analytic sample comprised 2,759 T2DM patients with normal serum free triiodothyronine (FT3) and free thyroxine (FT4). The primary objective was to determine whether alterations in central thyroid sensitivity—measured by several derived indices—were associated with eGFR-defined CKD.
Central thyroid sensitivity or resistance was assessed using multiple calculated indices reported in the source abstract: serum thyroid stimulating hormone (TSH) and three feedback/resistance indices: the Thyroid Feedback Quantile-based Index (TFQI), the thyrotrophic thyroxine resistance index (TT4RI), and the TSH Index (TSHI). The study also considered the FT3/FT4 ratio as an indicator related to peripheral conversion dynamics. These indices were used to quantify reduced central sensitivity or relative resistance to thyroid hormone signaling despite normal peripheral hormone levels.
CKD in this analysis was defined using estimated glomerular filtration rate (eGFR). The abstract specifies that associations were examined with eGFR-defined CKD, although detailed staging or cutoff values for CKD were not reported in the abstract. The primary outcome in all reported models was presence versus absence of CKD based on eGFR.
Associations between central thyroid indices and CKD were tested using multivariable logistic regression models. To explore the shape of the relationship, restricted cubic spline (RCS) analyses were applied, and the RCS tests yielded a statistically significant non-linear relationship (p < 0.001). To assess robustness and reduce confounding, the authors conducted propensity score-matched analyses; matched sample size reported in the abstract was 684 pairs.
The principal findings reported in the abstract are: higher TSH levels and greater central thyroid hormone resistance as reflected by elevated TFQI, TT4RI, and TSHI were independently associated with increased odds of CKD in multivariable models. Specifically, participants in the highest TSH quartile had a 1.50-fold increased risk of CKD compared with lower quartiles in multivariable regression (OR 1.50, 95% CI 1.08–2.07). Restricted cubic spline analysis confirmed a significant non-linear association between the central thyroid sensitivity markers and CKD risk (p < 0.001).
The association between higher TSH and CKD remained after propensity score matching; in the matched analysis of 684 pairs the odds ratio for CKD in the highest-TSH group was 1.66 (95% CI 1.21–2.27). This persistence of effect after matching was presented as evidence of robustness against measured confounding. No additional sensitivity or subgroup analyses are described in the abstract.
The authors conclude that in T2DM patients with normal peripheral thyroid hormones, elevated TSH and decreased central thyroid hormone sensitivity are independently associated with a higher risk of eGFR-defined CKD. They propose that these readily calculated indices (TFQI, TT4RI, TSHI, and TSH level) could serve as practical clinical biomarkers for early renal risk stratification among patients with T2DM.
The source is an abstract of a cross-sectional study; therefore, temporal or causal inferences cannot be drawn. The abstract does not provide details on covariates adjusted for in multivariable models, exact eGFR thresholds or CKD stages used, laboratory assay methods, or potential residual confounding. The abstract reports propensity score matching but does not list matching variables. These methodological specifics were not reported in the source abstract.
Among 2,759 T2DM patients with normal FT3 and FT4, higher TSH and markers of central thyroid hormone resistance (TFQI, TT4RI, TSHI) were independently associated with increased odds of eGFR-defined CKD. Restricted cubic spline analysis showed a significant non-linear relationship, and results were consistent after propensity score matching. The authors suggest these indices may be useful as clinical biomarkers for renal risk stratification, while the cross-sectional design and limited methodological detail in the abstract constrain causal interpretation and generalizability.