Dear Editor,

In the context of clinically suspected non-thyroidal illness(NTI) the advice to retest patients with raised levels of thyroid stimulating hormone(TSH)(1) may extend even to those in whom TSH levels are in the range 20-32.4 mIU/L(2). In one study, over a period averaging 88 days(Standard Error ie SE=34), seven such subjects, with mean baseline TSH of 32.4 mIU/L(SE=3.6), experienced a spontaneous fall in that parameter to levels averaging 12.3 mIU/L(SE=3.7)as a result of recovery from NTI. In two of those subjects, both of whom tested negative for antimicrosomal antibodies, TSH fell to 2.1 mIU/L, and 2.3 mIU/L, respectively. In the other five, all of whom were antibody positive, TSH levels fell to 9.9, 10.1, 11.5, 21.7, and 28.5 mIU/L,respectively. In order to check whether or not patients tested for TSH have significant but clinically unsuspected NTI it might even be advisable to test for the acute phase marker, C reactive protein(CRP). This was the strategy followed in one study so as to check if patients with suspected hypoferritinaemia might have spurious "normalisation" of their serum ferrtin levels attributable to an acute phase response. In that study 10.2% and 44% of patients in primary care and in secondary care, respectively, had elevated CRP levels(3). Where necessary, TSH levels also need to be interpreted in the context of advanced age, given the recent analysis of the age-specific distribution of TSH levels in 14,376 subjects who were not only disease-free, but also tested negative for thyroid antibodies. In that analysis the 95% confidence limits for TSH yielded the range 6.17-10.85 mIU/L in patients aged 80 or more(4). One interpretation of these findings was that the age- related shift in the reference range "could either facilitate or be a consequence of healthy aging"(4). This view resonated with the findings in a cohort of 599 subjects 37 of whom had overt hypothyroidism, the latter characterised by the association of subnormal levels of free thyroxine and TSH levels in the range 4.86-33.0 mIU/L. The enrollment age was 85, and the mean follow up was 3.7 years(standard deviation 1.4). Over that period, in spite of non-treatment, these overtly hypothyroid participants showed a significant trend towards the lowest mortality rates when compared with the rest of the cohort(Cox regression, P for trend=0.03). Furthermore, increasing baseline levels of TSH were not associated with an accelerated increase in phsical disability, cognitive decline, or depressive symptoms(5). Accordingly, in primary care best practice is to take account of NTI, using clinical criteria, and, possibly, CRP as well, and to take cognisance of advanced age.

Oscar M Jolobe
Retired Geriatrician

References

(1) Smellie WSA., Vanderpump MPJ., Fraser WD., Bowley R., Shaw N Best Practice in primary care pathology; review 11 Journal of Clinical Pathology 2008:61:410-18

(2)Spencer C ., Elgen A., Shen D et al Specificity of sensitive assays of thyrotropin(TSH) used to screen for thyroid disease in hoispitalised patients Clinical Chemistry 1987:33:1391-6

(3) O'Broin S., Kelleher B., Balfe A., mCMahon C Evaluation of serum transferrin receptor assays in a centralised iron screening service Clinical and Laboratory Haematology 2005:27:190-4

(4) Surks MI., Hollowell JG Age-specific distribution of serum thyrotropin and antithyroid antibodies in the U.S. population: Implications for the prevalence of subclinical hypothyroidism Journal of Clinical Endocrinology and Metabolism 2007:92:4575-82

(5) Gussekloo J., van Exel E., de Craen AJM et al Throid status, disability and cognitive function, and survival in old age Journal of the American Medical association 2004:292:2591-9