Influence of Atenolol on Within-Subject Variation of Thyroid Function Tests

Abstract

Atenolol is a commonly used beta-blocker that may affect serum thyroid hormone concentrations.' However its influence on the wellrecognized within-subject variation':' of the free levels of thyroxine (free T4), tri-iodothyronine (free T3) and thyroid stimulating hormone (TSH) has not been previously documented. Knowledge of this effect may be important when monitoring thyroid status in relevant patients, Twelve clinically euthyroid men (mean age 53 years, range 37-64; mean wt 80 kg, 61-95), who were receiving 100mg of atenolol daily for stable angina, and a closely matched group (n = 12; mean age 48 years, 35-61; mean wt 79 kg, 59-99) not receiving atenolol were studied. Serum levels of free T4 (Amerlex-M RIA, reference range 8·525·0 pmol/L), free T3 (Amerlex-M RIA, reference range 2·6-8,0 pmol/L) and TSH (Amerlite RIA, reference range 0·3-4'OmU/L) were measured in the cases and controls on four separate occasions (six-month intervals) over an 18-month period; inter-assay CVs 4-6%. None of the patients from either group suffered major illness, received additional medication, or showed a significant change in body weight during the study; serum creatinine remained < 120J1.mol/L. Compliance with atenolol was confirmed by direct questioning and occasional packet counts, The values of FT4, FT3 and TSH were compared between the groups by Student's t-test. Within-subject variance was estimated by analysis of variance and compared between the groups using the F-test;4 this analysis was not factored for analytical variance. The ratio of within-subject variance (including analytical variance) to total variance was also calculated, Although there were no significant differences in the mean levels of the three hormones between the groups, the within-subject variance of FT4 and FT 3 was significantly greater in the atenololtreated patients than in controls (see Table I), With the control group, the ratio of withinsubject variance to total variance for FT4, FT3 and TSH was 25, 45 and 25%, respectively, By contrast, with atenolol treatment the ratio of variances for FT4, FT 3 and TSH increased by a factor of 2,2'1, and 1,8, respectively. However, even after allowing for analytical variance, the within-subject variance for FT3 in this group was higher than that reported by Browning et al? This might have been due to differences in the characteristics of the sample population and to the smaller number of observations made in the present study,