From: Thyroid hormones and frailty in older adults: systematic review and dose–response meta-analysis
First author, Year | Country | Study Design | Total participants / setting | Age (mean ± SD) years | Thyroid hormone | Frailty definition | Adjusted confounding factors | Main results |
---|---|---|---|---|---|---|---|---|
Yeap [20], 2012 | Australia | Cross-sectional study | 3943 men/ Community | 75.2 ± 4.1 | TSH(0.4–4.0mIU/L), FT4(10–23 pmol/L) | Frailty phenotype criteria: Frail Nonfrail | Age, BMI, smoke status, diabetes, social support, impairment of seeing or hearing, testosterone and Insulin–like growth factor–I level | Nonsignificant association noted between TSH and frailty (p > 0.5). FT4 had the highest odds for frailty in two quartiles (Q3:Q1, OR = 1.32, 95% CI = 1.01–1.73; Q4:Q1, OR = 1.36, 95% CI = 1.04–1.79; p = 0.010). Neither subclinical hyperthyroidism (OR = 0.69, 95% CI = 0.21–2.31) nor subclinical hypothyroidism (OR = 1.17, 95% CI = 0.90–1.53) was significantly associated with frailty |
Virgini [19], 2015 | Switzerland | Prospective cohort study | 1455 men/ Community | 73.6 ± 5.8 | Subclinical hyperthyroidism, subclinical hypothyroidism, euthyroidism TSH(0.55–4.78mIU/L), FT4( 0.8–1.75 ng/dL) | Frailty phenotype criteria: Robust group Prefrail and frail group | age, race, BMI and clinicalcenter | Compared with those with euthyroid, men with subclinical hyperthyroidism had an increased likelihood of high frailty status (adjusted OR = 2.48, 95% CI = 1.15–5.34) |
Bertoli [21], 2017 | Italy | Observational study | 112/ 62 hospitalized, 50 outpatient | 79.1 ± 7.0 | TSH(0.35–4.5μIU/mL), FT4(0.8–1.75ng/dL), FT3(2.3–4.2 pg/mL) | Frailty score | NA | Frailty score was significantly correlated with FT3 (p < 0.0001), but not FT4 (p = 0.1974) |
Veronese [14], 2017 | Italy | 2571 cross-sectional, 1732 longitudinal | 3099 (1245 men, 1854 women)/ community | Men 73.2 ± 6.5 Women 74.7 ± 7.3 | TSH (0.3 and 4.2 mUI/L) Quintiles cutoffs for men: 0.7, 1.0, 1.3, and 2, while for women 0.8, 1.1, 1.5, and 2.5 mUI/L | Frailty phenotype criteria: Frail Nonfrail | Age, BMI, smoke status, alcohol drinker, education, monthly income, ADL, geriatric depression, MMSE scores, Charlson comorbidity score, eGFR, number of drugs | With the third quintile of serum TSH (Q3) as the reference group, the highest quintile (Q5) was associated with the highest frailty risk in men (OR = 1.55, 95% CI = 1.03–2.33) and in women (OR = 1.97, 95% CI = 1.59–2.45) |
Bano [8], 2018 | Netherlands | Prospective cohort study | 9,640/ NA | 64.9 ± 9.7 | TSH (0.40 to 4.0 mIU/L), FT4( 0.86 to 1.94 ng/dL) | Frailty index: 45-item | age, sex, cohort, smoking, alcohol, and education | TSH (p < 0.0003) and FT4 (p < 0.0001) with frailty at baseline |
Pasqualetti [22], 2018 | Italy | Longitudinal study | 619/ hospitalized | 83.8 ± 7.4 | TSH(0.4–4.0mIU/L), FT4(0.70–1.70ng/dL), FT3( 2.7–5.0 pg/mL) | MPI score: > 0.66 = frailty; 0.34–0.66 = Prefrail 0.34 = Robust | age, sex, MPI, FT3, LDH, Hb, CRP and albumin | MPI score was inversely and strongly correlated with FT3 (p < 0.001) and moderately and positively correlated with FT4 (p < 0.05) |
Arosio [23], 2020 | Italy | Cohort study | 593/community or nursing home | 80.1 ± 15.7 | TSH(0.27–4.2µIU/ml), FT4(0.9–1.7ng/dl), FT3(2.3–4.4 pg/ml) | Frailty index: 30 items | sex, age and study center | Correlation of frailty index with FT3 (ρ = − 0.281, p < 0.001), TSH (ρ = − 0.223, p = 0.003) was negative Correlation of frailty index with FT4 was positive (ρ = 0.189, p = 0.001) |
Xiu [24], 2020 | China | Cross-sectional study | 240 (T2DM)/ NA | 68.9 ± 6.9 | TSH(0.55–4.78mIU/mL), FT4(0.89–1.76ng/dL), FT3(2.3–4.2 pg/mL) | Frailty phenotype criteria: Frail Prefrail Robust | age, sex, 25(OH) D3, eGFR, FT3 | Logistic regression showed that low FT3 was significantly associated with an increased risk of frailty (OR = 4.53, 95% CI = 1.89–10.83; P = 0.001) |
Bhalla [25], 2021 | USA | Cross-sectional study | 150/ inpatients | 70.0 ± 6.2 | TSH(0.5-5µIU/ml), FT4(0.70–1.48ng/dL), FT3(1.50–4.20 pg/ml) | Frailty index: 30 items | age | Patients with lower TSH (0.31 ± 0.11 µIU/mL) had higher mean frailty index (0.25 ± 0.12), and patients with normal TSH (1.84 ± 0.84 µIU/mL) had lower mean frailty index (0.15 ± 0.07; p < 0.001) An association of FT3 levels with FI was inverse (p = 0.13), and it disappeared when age was adjusted for (p = 0.4) |
Liu [26], 2021 | China | Cross-sectional study | 146/ inpatients | 85.0 ± 8.2 | TSH(0.35–4.94mIU/L), FT4(9.01–19.05pmol/L), FT3(2.63–5.70pmol/L), T4(62.88–150.80 nmol/L), T3(0.88–2.44 nmol/L) | Frailty phenotype criteria: Frail Prefrail Robust | age, sex, BMI, smoking, and HbA1c | Frailty was significantly associated with serum TSH (OR = 1.258) and T3 (OR = 0.102) levels |