LOW TSH LEVELS IN NURSING-HOME RESIDENTS NOT TAKING THYROID-HORMONE

Citation
Pj. Drinka et al., LOW TSH LEVELS IN NURSING-HOME RESIDENTS NOT TAKING THYROID-HORMONE, Journal of the American Geriatrics Society, 44(5), 1996, pp. 573-577
Citations number
20
Categorie Soggetti
Geiatric & Gerontology","Geiatric & Gerontology
ISSN journal
00028614
Volume
44
Issue
5
Year of publication
1996
Pages
573 - 577
Database
ISI
SICI code
0002-8614(1996)44:5<573:LTLINR>2.0.ZU;2-0
Abstract
BACKGROUND: Many practitioners perform a thyroid stimulating hormone ( TSH) assay as a screening test in older patients. The introduction of sensitive TSH assays with lower normal limits has created a laboratory abnormality that is often of uncertain significance. Mechanisms inclu de autonomous overproduction of thyroid hormone, nonthyroidal illness including medication effects, and hypothalamic/pituitary dysfunction. OBJECTIVE: To characterize the clinical status and course of nursing h ome residents with low TSH and normal total T-4 levels in the absence of treatment with thyroid hormone. DESIGN: Retrospective chart review was performed to determine participants status at the time of the low TSH level, with additional recording of follow-up thyroid hormone leve ls, cardiac rhythm, and mortality. Mortality was compared with that of a control group matched for age and sex. SETTING: A nursing home for veterans and their spouses. MAIN RESULTS: Forty subjects with low TSH and initially normal total T-4 were identified. Only three subjects we re subsequently diagnosed as hyperthyroid. TSH levels of 18 subjects s ubsequently normalized, and four additional subjects had low total T-3 levels suggesting a nonthyroidal mechanism. Seven subjects died durin g the first 4 months of follow-up compared with three in a control gro up (P < .001). Nine of the 40 subjects had a history of or current atr ial fibrillation. No new atrial fibrillation was documented during 388 months of EKG follow-up. CONCLUSIONS: Low total T-3 levels, TSH norma lization, and excess mortality suggest that nonthyroidal illness plays a role in the pathogenesis of low TSH determinations in the nursing h ome. Autonomous production of thyroid hormone also plays a role. We be lieve that the term ''subclinical hyperthyroidism'' should be used onl y if the clinician believes that autonomous overproduction of thyroid hormone is the cause of a low TSH level. If subsequent research shows correctable adverse consequences associated with subclinical hyperthyr oidism from autonomous overproduction of thyroid hormone, a more aggre ssive diagnostic approach will be needed to define the mechanism of a low TSH level at the time of its discovery.