Abstract:
Background: Hypothyroidism is closely associated with metabolic dysfunction, insulin resistance, and early cardiac maladaptation. Conventional clinical markers—including Thyroid-Stimulating Hormone (TSH), Free Thyroxine (FT4), blood pressure, Body Mass Index (BMI), and preserved Ejection Fraction (EF)—frequently fail to identify individuals at highest cardiometabolic risk. This limitation creates a critical gap in early risk stratification for diabetes and structural heart disease. To address this unmet need, we developed the Residual-Maladaptation Score (RMS), a proactive composite risk signal designed to capture subclinical cardiometabolic stress not detectable by traditional metrics.
Methods: In this single-center derivation study, 277 adults with established hypothyroidism were analyzed. RMS integrates three mechanistic domains: (1) cardiac maladaptation quantified by negative EF prediction residuals derived from multivariable modeling, (2) vascular load assessed by pulse pressure, and (3) thyroid regulatory discordance defined by TSH–FT4 mismatch. Components were standardized and combined into a continuous score. Associations with diabetes and valvular abnormalities were evaluated using logistic regression adjusted for conventional risk factors. Discriminatory performance was assessed using Receiver Operating Characteristic (ROC) analysis. Internal validity and reproducibility were examined using 5-fold cross-validation and 5,000-sample bootstrap resampling.
Results: RMS significantly outperformed conventional parameters (TSH, FT4, EF, BMI) in identifying cardiometabolic risk. Mean RMS values were higher in patients with diabetes compared with non-diabetic patients (0.14 vs −0.28; p = 0.001) and markedly elevated in those with valvular abnormalities (1.04 vs −0.07; p < 0.001). Conventional thyroid markers did not discriminate risk (TSH p = 0.306; FT4 p = 0.259). RMS independently predicted diabetes (OR 1.61 per unit increase; p = 0.002) and valvular abnormalities (OR 2.48; p < 0.001). Model accuracy was 67.5% for diabetes and 93.8% for valvular disease, with moderate discrimination for diabetes (AUC 0.66) and good discrimination for valvular abnormalities (AUC 0.75). Cross-validated and bootstrap-corrected AUCs (Diabetes-0.66 and Valvular abnormality-0.75 ) were consistent, confirming strong internal reproducibility.
Conclusion: RMS is a novel, internally validated, proactive risk signal that reveals clinically meaningful cardiometabolic maladaptation in hypothyroid patients despite normal conventional markers. RMS offers meaningful clinical advantages by earlier identification of high-risk individuals, results in proactive diabetes screening, timely cardiology referral, and enhanced surveillance for structural heart disease in Patients with Hypothyroidism.
