Copyright: 2019 This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.
No recent study has focused on clinical features of subclinical hypothyroidism (SCH), especially in older patients. TSH measurement has remarkably evolved these last 20 years and thus reconsideration is needed. In our prospective multicenter study (2012-2014) including 807 subjects aged <60 years (<60y) and 531 subjects ≥60 years (≥60y), we have monitored 11 hypothyroidism-related clinical signs (hCS) together with TSH, FT4, FT3 and anti-thyroperoxidase antibodies values. hCS expression has been compared in patients with SCH vs euthyroidism in each age group. The number of hCS above 60y of age were found to be more elevated in the euthyroid population (1.9 vs 1.6, p<0.01) than in the SCH population (2.3 vs 2.6, p=0.41) while increase in hCS is limited to SCH subjects in the <60y group (p<0.01). The percentage of subjects with at least 3 signs increased with SCH in the <60y group (42.6% vs 25.0%, p<0.01) but not ≥60y (34.4% vs 33.9%, p=0.96). In older individuals, only three hCS could be related to both SCH and a decreased T3/T4-ratio (0.26 vs 0.27, p<0.01), suggesting either a reduced activity of TSH, or an adaptive response with aging. While hCS are clearly associated with SCH in patients <60y, they are not so informative in older subjects. TSH measurements carried out on the basis of hCS need to be interpreted with caution in aged patients. A reassessment of the TSH reference range in older patients is clearly needed and should be associated to more appropriate monitoring of thyroid dysfunction
Subclinical hypothyroidism (SCH) is defined by elevated TSH (Thyroid-Stimulating Hormone) values and normal levels of free thyroxin (FT4). Among thyroid diseases, this disorder remains elusive and may affect about 1-11% of the worldwide population (
hCS are composed of a remarkable wide array, including signs of hypometabolism, cardiovascular, neuromuscular and mucocutaneous incidents. They have been often combined to establish clinical scores to help diagnosing thyroid deficiency: Wayne’s index (1959), Billewicz’s score (1969) or Zulewski’s score (1997) (Supplementary Table 1) (
In this study, we have explored the expression of 11 of the 12 hCS included in the Zulewski’score (
Figure 1. Flowchart of the recruitment, selection and exclusion process of the cohort.
Recruitment of the prospective multicenter study was performed in the Departments of Endocrinology and Geriatrics at the Lyon-Sud Hospital (France), and in the Departments of Endocrinology and Nuclear Medicine at the Chambery Hospital (France). All men and women who consulted in these four hospital departments involved in the study were invited to participate. No announcement has been made. All patients for whom a TSH assay was prescribed by current care physicians were invited to participate, irrespective of any history of thyroid dysfunction, and gave informed consent. Patients with differentiated thyroid carcinoma, under thyroxine treatment or missing information were excluded. Hyperthyroid patients (TSH <0.17 mIU/L) were also excluded as well as patients with elevated TSH (>4.05 mIU/L) and low FT4 (<11.5 pmol/L) defined as overt hypothyroid. We used the upper limit of 4.05mIU/L for TSH as recommended by the assay manufacturer (IRMA, Beckman Coulter) and in accordance with the National “Haute Autorité de Santé” (HAS) standard set (
Samples were obtained at the time of inclusion (no follow-up). TSH was estimated (IRMA TSH, Villepinte, France) as well as FT4 (RIA, Beckman Coulter, Villepinte, France), FT3 (RIA, Beckman Coulter, Villepinte, France) and TPOAbs (RIA, ThermoFisher Scientific B.R. A.H.M.S, Asnières, France). Age, sex, thyroid-interfering medication (amiodarone, psychotropic and interferon) and the absence/presence of hCS as described by Zulewski and colleagues (
The diagnostic performance of each hCS was quantified using sensitivities, specificities, likelihood ratios (LR) within a 95% confidence interval (CI). Logistic regression models were used for quantifying the effect of the hCS and predicting the probability of SCH according to the combinations of signs. Empirical ROC curves were built using predicted probabilities. Diagnostic performance was quantified by estimation of the area under the ROC curves (AUROCs) with its 95% CI. AUROCs were compared using the non-parametric method of Delong. Comparisons of means and of percentages were carried out using respectively ANOVA and chi-square test. Analyses were carried out using the statistical software SPSS version 21 and R version 3.2.1. For all the comparisons a p<0.05 was considered for the statistical significance.
Table 1 Mean number of overall clinical signs.
|Groups & Classes||<60 years||≥60 years|
The average age of the cohort was of 43±11 years for the <60y group and of 71±9 years for the ≥60y population. Females were more represented in the younger participants. 12.9% of the subjects were under medication in the <60y group and 16.0% in the ≥60y (p=0.13) (Supplementary Table 2) and the inclusion of such patients in the study did not modify the results of the study (data not shown). TSH levels were higher (p<0.05) and FT3 levels lower (p<0.01) in older individuals.
The proportion of SCH was 9.7% (Supplementary Table 3). A higher incidence was observed in the ≥60y (11.7%) compared to <60y individuals (8.4%) (p=0.06). SCH subjects were more likely to have TPOAbs than euthyroid individuals (36.9% vs 14.7%, p<0.01). Auto-immunity was similar in both age groups (3.0% vs 3.5% for ≥60y and <60y respectively, p=0.10). Furthermore, the percentage of subjects without TPOAbs and with a TSH betwin 4.1-10.0 mIU/L was higher in seniors (7.9% vs 4.6%, p<0.05).
The mean TSH level was significantly higher in ≥60y (2.35±1.73) than in the younger group (2.15±1.66) (p<0.05). For <60y individuals in the range of 4.1-10.0mIU/L, TSH was higher in the presence of autoimmunity (Fig. 2A). FT3 values were reduced only in euthyroid older persons (Fig. 2B). FT4 levels were similar between age groups but significantly decreased when TSH increased (Fig. 2C). As a consequence, the FT3/FT4 ratio decreased in euthyroid older subjects with normal TSH and increased with elevated TSH values in both age groups (Fig. 2D).
The number of hCS was significantly higher in patients with higher TSH levels (p<0.01). However, this was true only for the population of younger but not for older subjects (
Figure 3 pictures graphically that patients with no hCS represented the largest proportion individuals in the euthyroid group, while patients with three or more hCS were more frequent in the SCH group (p<0.05) (Fig. 3A, left side). However, when considering patients aged ≥60 years, the number of hCS did not differ significantly with TSH levels (Fig. 3C, left side). In both age groups, hCS seemed more frequent in presence of anti-TPO antibodies (Table 1 and Fig. 3, right side). However, this finding was not statistically significant.
Also, estimating the AUROCs showed no impact of age on the increase of hSC (p=0.28): 68% (CI 95%, 0.61-0.75) and 62.4% (CI 95%, 0.55-0.70) for <60y and ≥60y groups respectively (Supplementary Fig. 1) as well as between males and females (Supplementary Fig. 2).
In the <60y subjects, weight increase (OR 1.75; CI 95%, 1.02-3.03, p=0.04) and coarse skin (OR 6.64; CI 95%, 1.93-22.81, p=0.003) were associated with SCH while in older subjects, SCH was associated with periorbital puffiness (OR 2.42; CI 95%, 1.01-5.84) and also coarse skin (OR 7.3; CI 95%, 2.48-21.85) (p<0.001). Of interest, the younger group displayed up to 4 hCS with high specificity (≥95%) (97%, periorbital puffiness; 99%, slow movements; 99%, coarse skin; 95%, diminished sweating) whereas only 2 of them could be highlighted in the ≥60y (98%, slow movements; 97%, coarse skin). Sensitivities were low for both groups (<50%) (Supplementary Table 4). Height hCS had good positive LR in the <60y group but only 3 in the >60y group. Three hCS showed LR >2 in both age groups: periorbital puffiness, slow movement and coarse skin. Most of negative LR was near the value of 1. Only weight increase, coarse skin and paraesthesia in subjects <60y, plus coarse skin in older individuals were found significant (Supplementary Table 5).
We also observed that older subjects presented more frequently diminished sweating and hoarseness of voice (p<0.05), dry skin, constipation, impairment of hearing and periorbital puffiness (p<0.01) in the euthyroid population (Fig. 4A), while only impaired hearing was more represented (p<0.05) in the SCH population. Cold skin was also less frequent in the older compared to younger subjects (p<0.05) (Fig. 4B). Regarding TSH level, the prevalence of 8 hCS was increased in younger SCH subjects compared to euthyroid individuals: paraesthesia, cold skin, constipation (p<0.05), hoarseness of voice, coarse skin, weight increase, slow movements and periorbital puffiness (p<0.01) (Supplementary Fig. 3). In contrast, in older individuals, only 3 hCS were more frequent when TSH was in the subclinical range (p<0.05, slow movements and periorbital puffiness; p<0.01, coarse skin). In these subjects, we also observed that weight increase was less frequent in the absence of autoimmunity (p<0.05) and that constipation (p<0.05) and periorbital puffiness (p<0.01) were more frequent with autoimmunity. Slow movements were more frequent in younger SCH subjects with TPOAbs (p<0.05) (data not shown).
In addition, to determine which signs could be determined as nonspecific symptoms, we performed an F-test based on normal (NACB recommendations) and hypothyroid populations of our cohort. We found that 4 signs showed low Relative Risk Values: paraesthesia (1.42), constipation (1.36), dry skin (1.23) and cold skin (1.18). If we base our analysis on the 7 remaining clinical signs, main results were not affected (data not shown).
Finally, we found about 10% more females than males in the <60 group compared to the ≥60y group. To control this point, we compared the expression of clinical signs between males and females and found that it does not modify our conclusion, but probably minimize it.
In this study, we showed that clinical features of SCH decreased with age i.e. ≥60y and observed no satisfactory correlation to usual biological parameters of thyroid testing in the elderly.
Six hCS were more represented in older than in younger euthyroid subjects (p<0.01): periorbital puffiness, impairment of hearing, constipation, dry skin and diminished sweating and hoarseness of voice but these signs are also described in physiological aging (
Earlier studies proposed combined evaluation of hCS (
The SCH prevalence of 9.7% is in good agreement with the 9.2-9.5% reported so far (
A similar proportion of the <60y and ≥60y participants were under medication, suggesting that elevated TSH was not due to polypharmacy in the elderly (
Our results clearly indicate that the clinical evaluation of SCH is insufficiently reliable for patients aged 60 years or older. The clinical evidence of thyroid dysfunction is barely detectable and only abnormality is a marginally elevated TSH. As a result, the SCH diagnosis should be made with caution in older people. Accordingly, treatment with levothyroxine in older persons with SCH may fail to provide no symptomatic benefits (
Among the 11 evaluated hCS, periorbital puffiness, slow movements and mostly coarse skin appeared rather specific for SCH in older patients. However, the low sensitivity of hCS and the altered levels of TSH with aging make the diagnosis of SCH difficult in older individuals. Defining an age-related reference range and/or developing assays able to measure bioactive TSH is needed to improve SCH detection in the elderly.
We warmly thank all collaborators from the Departments of Endocrinology and Geriatrics at the Lyon-Sud Hospital, and from the Departments of Endocrinology and Nuclear Medicine from the Chambery Hospital for their personal motivation in recruiting subjects, collecting blood samples and biological measurements.
The Colorado thyroid disease prevalence study. ,
Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). ,
Changes in the prevalence of hypothyroidism: the HUNT Study in Norway. ,
Prevalence, clinical and biochemical profile of subclinical hypothyroidism in normal population in Mumbai. ,
Prevalence of unknown thyroid disorders in a Sardinian cohort. ,
Prevalence and clinical characteristics of subclinical hypothyroidism in an opportunistic sample in the population of Castile-León (Spain). ,
Thyroid Stimulating Hormone Reference Range and Prevalence of Thyroid Dysfunction in the Korean Population: Korea National Health and Nutrition Examination Survey 2013 to 2015. ,
Diagnosis and Management of Subclinical Hypothyroidism in Elderly Adults: A Review of the Literature. ,
Alterations in thyroid function tests in aged hospitalized patients: prevalence, aetiology and clinical outcome. ,
How significant is the TSH level in the circulation? Immuno-Anal Biol Spéc,
Reference ranges for Thyroid-Stimulating Hormone and Free Thyroxine in older men: Results from the Health In Men Study. ,
Prevalence of thyroid dysfunction and autoimmunity in the older population and implications of age-specific reference ranges. ,
Thyroid Hormone Therapy for Older Adults with Subclinical Hypothyroidism.,
Clinical scoring scales in thyroidology: A compendium. ,
Estimation of tissue hypothyroidism by a new clinical score: evaluation of patients with various grades of hypothyroidism and controls. ,
2013 ETA Guideline: Management of Subclinical Hypothyroidism. ,
Hypothyroid Symptoms Fail to Predict Thyroid Insufficiency in Old People: A Population-Based Case-Control Study. ,
Does age play a role in clinical presentation of hypothyroidism?,
Development of a short version of the Thyroid-Related Patient-Reported Outcome ThyPRO. ,
Hypothyroidism in the elderly: diagnosis and management. ,
Thyroid disease in the elderly. Part 2. Predictability of subclinical hypothyroidism. ,
TSH - Clinical aspects of its use in determining thyroid disease in the elderly. How does it impact the practice of medicine in aging?,
Recommandations professionnelles. Hypothyroïdies frustes chez l’adulte: diagnostic et prise en charge. ;
Age-specific distribution of serum thyrotropin and antithyroid antibodies in the US population: implications for the prevalence of subclinical hypothyroidism. ,
Age-related changes in thyroid function: a longitudinal study of a community-based cohort. ,
Age-specific TSH reference ranges have minimal impact on the diagnosis of thyroid dysfunction. ,
Laboratory medicine practice guidelines. Laboratory support for the diagnosis and monitoring of thyroid disease. ,
Subclinical hypothyroidism: an update for primary care physicians. ,
The anatomic basis of midfacial aging. In:
Current concepts in age-related hearing loss: epidemiology and mechanistic pathways. ,
Insights into the pathophysiology and mechanisms of constipation, irritable bowel syndrome, and diverticulosis in older people. ,
Cunliffe Scientific Awards. Characteristics and pathomechanisms of endogenously aged skin. ,
The aging voice. ,
Clinical characteristics and outcomes of myxedema coma: Analysis of a national inpatient database in Japan. ,
Symptomatologie de l’hypothyroïdie. In:
Functional thyroid pathology in the elderly. ,
Extreme longevity is associated with increased serum thyrotropin. ,
Longitudinal changes in thyroid function in the oldest old and survival: the Cardiovascular Health Study All-Stars Study. ,
Age-related modifications in the regulation of the hypothalamic-pituitary-thyroid axis. ,
Oxidative stress and the ageing endocrine system. ,
Thyrotropin isoforms: implications for thyrotropin analysis and clinical practice. ,
TSH enhancement of FT4 to FT3 conversion is age dependent. ,
High prevalence of autoantibodies among Danish centenarians. ,
Autoimmunity in the Elderly: Insights from Basic Science and Clinics - A Mini-Review. .
Variability among TSH measurements can be reduced by combining a glycoengineered calibrator to epitope-defined immunoassays. ,
New immunoassays of bioactive TSH improve detection of thyroid disorders. ,