Thyroid hormones as predictors of short- and long-term mortality in very old hospitalized patients.

BACKGROUND

Although extensively investigated, the prognostic role of thyroid hormone abnormalities in older participants remains uncertain. We investigated the relationship between thyroid hormones and mortality during hospitalization and in a prolonged follow-up in frail older patients.

METHODS

A nonconcurrent cohort study was conducted by enrolling 450 participants hospitalized for an acute disease, who were classified into four groups (euthyroidism, hypothyroidism, hyperthyroidism, and low triiodothyronine [T3] syndrome), according to clinical and laboratory data. Multidimensional geriatric assessment variables were considered in order to identify short- and long-term predictors of death.

RESULTS

Participants were very old (mean age: 84 years) and frail, as indicated by severely impaired functional status, extensive comorbidity, high prevalence of dementia, and hospital mortality (8%). Prevalence of any thyroid dysfunction was 40.7%; 32% of participants had low T3 syndrome, which was associated with an excess hospital mortality risk (odds ratio: 2.7, 95% confidence interval [CI]: 1.1-6.5; p = .025), adjusted for demographic, clinical, functional, and laboratory data. Conversely, long-term mortality was unrelated to low T3 syndrome. In euthyroid participants, increasing levels of free thyroxine (FT4) were associated with a slightly greater mortality (hazard ratio, CI: 2.12, 0.99-4.54; p = .053) in adjusted Cox regression models.

CONCLUSIONS

This observational study on a cohort of very old, frail hospitalized patients gives support to the independent prognostic short-term, but not long-term, role of low T3 syndrome. Moreover, in older euthyroid participants, increasing levels of FT4 are a weak marker of poorer long-term survival. Thyroid hormones may help monitor changes in general health status and predict short- and long-term clinical outcomes in very old, frail patients.