Reciprocal Changes in Serum Concentrations of Triiodothyronine and Reverse Triiodothyronine between Summer and Winter in Normal Adult Men

Transcription

1 Endocrinol. Japon. 1980, 27 (4), Reciprocal Changes in Serum Concentrations of Triiodothyronine and Reverse Triiodothyronine between Summer and Winter in Normal Adult Men NORIMICHI KOONO Department of Internal Medicine, Hokkaido Central Hospital for Social Health Insurance, Nakanoshima, 1-8, Sapporo, Japan Abstract Serum concentrations of thyroxine (T4), 3, 5, 3'-triiodothyronine (T3), 3, 3', 5'-triiodothyronine (reverse T3, rt3), thyronine-binding globulin (TBG), thyroid stimulating hormone (TSH) and cortisol in summer and in winter were compared in carefully selected normal adult men (aged yr); 107 people from Sapporo (58 in summer and 49 in winter) and 15 hospital workers. A modest but significantly higher serum T3 concentration was recorded in winter than in summer, while T4, TSH, TBG and cortisol levels were similar in both seasons. Serum rt3 concentration in the Sapporo subjects was slightly higher in summer than in winter but difference was not statistically significant. However, a paired analysis of rt3 concentration in the hospital workers had significantly higher levels in summer than in winter, showed a significant rise in the rt3 to T3 ratio in summer compared to winter. These results were not likely to be due to changes in TBG, cortisol and hematocrit values. The present findings and our previous report on a similar TSH response to thyrotropin-releasing hormone in summer and in winter, indicate that the changes in thyroid hormone metabolism rather than in thyroidal secretion would, at least partly, be responsible for the reciprocal alteration in T3 and rt3 concentrations. Although a seasonal variation in serum thyroid hormone concentration in normal adult has been demonstrated by several authors (Osiba, 1957; Watanabe et al., 1963; Du Ruisseau, 1965; Rastogi and Sawhney, 1976; Nagata et al., 1976; Smals et al., 1977), a concomitant change in serum thyroid stimulating hormone (TSH) level has not been shown (Rastogi and Sawhney, 1976; Nagata et al., 1976; Hedstrand and Wide, 1973). Recent studies at Received December 13, our laboratory also failed to find winterrelated changes in the basal concentration of TSH or in the TSH response to thyrotropinreleasing hormone (TRH) desite an elevation in serum triiodothyronine (T3) concentration (Konno, 1978). These results cast some doubt on whether the pituitary function in secreting TSH might contribute to the preferential secretion of T3 from the thyroid gland during winter. Since it is well established that the peripheral conversion of thyroxine (T4) is a predominant source of T3 (Braverman et al., 1970; Sterling et al., 1970; Pittman et al., 1971), enhancement of this mechanism might be involved in the generation of serum T3 during winter. Supplementary to previous work, the present paper reports a comparative study of serum 3,3',5'-triiodothyronine (reverse T3, rt3), T4 and T3 concentration in summer and winter in carefully selected normal adult men and provides the evidence that serum rt3 and T3 levels vary reciprocally between the two seasons.

2 472 KONNO Endocrinol. Japon. August 1980 Materials and Methods Results 107 normal adult men (aged yr) living in Sapporo in northern Japan, were included in the present study. The subjects visited the hospital for health examination and their healthiness was evaluated by physical, biochemical and roentgenologic examinations. Blood specimens were obtained between 0830 and 0900 hr after an overnight fast and sera obtained were stored at -20 Ž until used. Of the 107 subjects, 58 were used for summer studies (June, 19, 1978 August 2, 1978) and 49 for winter (Dec. 19, 1978 Feb. 2, 1979). In addition to the subjects above, 15 adult men working in the hospital (aged yr) were also examined, from whom blood was drawn both in summer and in winter under the conditions described above. No special restrictions were made on food intake or physical activity and no attempt was made to avoid cold or heat stress. Measurement of serum T4, T3, rt3, Thyroninebinding globulin (TBG), TSH and cortisol by radioimmunoassay was done simultaneously in both summer and winter samples. Serum T4 and T3 were measured using a commercial test kit (Dainabott RI Laboratories). Serum rt3 was measured by the method described by Takagi et al. (Takagi et al., 1978). Serum TBG was measured using a test kit (Riagnost TBG, Behringwerke). Serum TSH was measured by double antibody radioimmunoassay (htsh-kit, Daiichi RI Laboratories, Tokyo and Calbiochem, San Diego). Serum cortisol was measured using a test kit (Daiichi RI Laboratories). The intraassay coefficients of variation for normal control serum were: T4: 3.3%, T3: 4.0%, rt3: 2.6%, TBG: 2.2%, TSH: 6.5%, cortisol 8.4%. The room temperature was measured at 0900 hr on the occasion of each sample collection in the hospital. The ambient temperature report was obtained from the official weather station in Sapporo. For statistical analysis, Student's t-test was used. In order to determine not simply whether two independent winter and summer samples were drawn from different groups, but whether either group is higher than the other, the Kolmogorov-Smirnov one-tailed test was applied (Siegels, 1956); X2= 4D2(n1n2/ n1 {2n), df=2, where n1 and n2 are sample sizes and D=max. bsn1(x1)-sn2(x2) b. The summary of results obtained in summer and in. winter is shown in Table 1. The age, body surface, hematocrit values and serum cholesterol concentrations were identical in both groups. Serum T4 concentration in the Sapporo subjects was 9.1 }1.9 ficant difference. Similar T4 values in both seasons were also seen in the hospital workers. The serum T3 concentration in the Sapporo subjects was 128 }17.5 ng/dl in summer and 138 }17.5 ng/dl in winter, the difference being statistically significant (p<0.01). The same trends was seen in the hospital workers, in which paired analysis was employed; 121 }11.2 ng/dl in summer and 130 }19.4 ng/dl (p<.0.02) in winter. As for serum rt3, the summer value was slightly higher than that for winter in the Sapporo subjects, but the difference was statistically not significant. However, a paired analysis of serum rt3 concentration in the hospital workers showed a significantly higher value in summer than that in winter; 26.8 }5.4 ng/dl vs 24.1 }3.5 ng/dl (p<0.005). Neither serum TSH, TBG nor cortisol concentration showed any seasonal variation in either the Sapporo subjects or the hospital workers. When the cumulative frequency distribution of summer and winter serum T3 concentration values in the Sapporo subjects were plotted on normal probability paper, a linearity was seen for both groups. The Kolmogorov-Smirnov test for goodness of fits was applied to the data; "d-max" was (at a T3 of ng/dl), and the calculated value of X2=6.43, which was greater than the critical value of 5.99 at a=0.05 (f=2) The room temperature measured at the time of the blood drawing ranged from 22 to 30 Ž in summer and 14 to 22 Ž in winter. The mean daily temperature on

3 T3 AND rt3 IN SUMMER AND IN WINTER Table 1. Summary of the results obtained in summer and in winter in normal adult men. Mean }S.D. Range. Number of subjects with TSH value less than threshold of detection (0.625ƒÊU/ml). Analysis by paired t-test. the day of sample collection ranged from 18.1 to 28.2 Ž in summer and from-10.1 to+3.0 Ž in winter. The correlation coefficient between serum T3 level and the mean daily ambient temperature was-0.23 (p<0.02), while the relationship between the T3 value and the room temperature was The individual values for serum T3, rt3 concentration and rt3/t3 ratios for hospital workers are shown in Fig. 2. In addition to significant seasonal changes in serum T3 and rt3 concentrations, there was a significant fall in the rt3/t3 ratio from a mean summer value of }0.047 to a mean winter value of }0.030 (p<0.005) Discussion The present study showed that the serum T3 concentration was significantly highter in winter than in summer, confiirming the findings of Nagata et al. (Nagata et al., 1976) as well as our own previous Fig. 1. Cumulative frequency distribution of serum T3 concentration in 107 normal adult men (58 in summer and 49 in winter) plotted on normal probability paper.

4 KONNO Endocrinol. Japon. August 1980 Fig. 2. Comparison of summer and winter individual values for serum T3 and rt3 and of the rt3 to Ta ratios in 15 normal adult men. results (Konno, 1978). There was an associated decrease in serum rt3 concentration and in the ratio of rt3 to T3 in winter compared with those in summer, although the winter-summer difference in serum rt3 was small as ascertained by paired analysis. These changes in serum T3 and rt3 concentrations are not secondary to homocon.centration or an elevation in serum thyroninebinding protein, because the hematocrit value and serum TBG concentration did not vary seasonally. Effects due to sample storage also are unlikely, since a recent study by Oddie et al. showed that T4, T3 and rt3 concentrations in cord serum decreased progressively with long term storage (Oddie et al., 1979), in contrast to the present findings of reciprocal changes in T3 and rt3 Although the winter-summer difference in the serum concentration of T3 was small i.e. 10 ng/dl difference between the means or 1.08 in the winter/summer ratio, the difference is significant for the following reasons; 1) the carefully selected subjects for summer and winter groups were matched for ages and body surfaces and the sample collection was performed at the same time. 2) The T3 assay was simultaneously carried out for the winter and summer samples, the intra-assay CV being not large enough to account for the difference. 3) The elevation in serum T3 concentration was observed in both the Sapporo subjects and in the hospital workers. 4) Both summer and winter samples showed a normal distribution and it was shown that each sample was drawn from a stochastically different population as analyzed by Kolmogorov-Smirnovs one-tailed test. Although the ambient temperature varied seasonally, it is difficult to quantify how much the people are affected by the changes in surrounding temperature. We have found a slight but significant inverse relation ship between serum T3 concentration and the ambient temperature but not with the room temperature. Similar results were reported by Osiba, Du Ruisseau and Smals et al. who have shown an inverse relationship between the ambient temperature and serum thyroid hormone concentration (Osiba, 1957; Du Ruisseau, 1965; Smals et al., 1977). Thus the ambient temberature may, at least partly, be responsible for a change in the serum T3 concentration. We are still not certain whether it is appropriate to conclude that serum T3 is elevated during winter or this hormone is reduced during summer, considering that the T3 concentration measured did not far exceed the physiological range. Previously we have found that the

5 T3 AND rt3 IN SUMMER AND IN WINTER seasonal change in serum T3 conentration was not accompanied by a change in the concentration of basal TSH or in the TSH response to TRH (Konno, 1978). It is there fore more likely that the present results reflect changes in thyroid hormone metabolism rather than alterations in thyroidal secretion. This view is supported by the following observations T4 did not vary seasonally, but serum T3 changed in inverse proportion to serum rt3, indicating that the alteration in peripheral monodeiodination of T4 to T3 is involved, as has been pointed out in other studies. Reciprocal changes in serum rt3 and T3 concentrations have been found in a variety of clinical situations such as caloric deprivation (Vagenahis et al., 1975), systemic illness (Chopra et al., 1975a), the newborn state (Chopra et al., 1975c), propranolol (Verhoeven et al., 1977) or propylthiouracil (Westgren et al., 1977). The decreasein serum T3 in these situations is regarded as due to an impairment of the peripheral monodeiodination of T4 (Chopra et al., 1978). The present findings on increased rt3 and decreased T3 levels in summer relative to those in winter qualitatively resemble, though less markedly, the above situations. The effect of adrenal steroid can be easily excluded by the similarity in the serum cortisol concentrations in both seasons. Since studies in man indicate that only a small fraction of circulating rt3 originates from thyroidal secretion (Chopra, 1976; Gavin et al., 1977; Burman et al., 1977), our results seem to indicate that, in addition to a diminution in the production of T3 from T4, there is an increase in the conversion of T4 to rt3 during summer. Furthermore, it is tempting to speculate that the diversion of T4, from conversion to highly potent T3, to the generation of poorly calorigenic rt3 would be a normal host adaptation to environmental temperature, serving to reduce basal energy requirements during the summer season. This interpretation is compatible with the current concept that peripheral monodeiodination of T4 to T3 and rt3 is not a random process but reflects two distinct routes of metabolism of T4 (Chopra et al., 1978). However, further studies of the kinetics of peripheral metabolism and production rates of T3, rt3 and T4 are required to clarify whether the occurrence of elevation of T3 and or reduction of rt3 during winter is a consequence of the stimulation of an activating pathway of T4 monodeiodination or of a diminution of an inactivating pathway or both. Acknowledgement The author is grateful to Professor T. Hiroshige (Department of Physiology, Hokkaido University School of Medicine) for his advice and critical review of this manuscript. The expert technical assistance of Mr. K. Hagiwara, Mr. R. Minami and Mr. T. Kojima is gratefully acknowledged. References Braverman, L. E., S. H. Ingbar and K. Sterling (1970). J. Clin. Invest. 49, 855. Burman, K. D., R. C. Dimond, F. D. Wright, J. J. Earll, J. Bruton and L. Wartofsky (1977). J. Clin. Endocrinol. Metab. 44, 660. Chopra, I. J., U. Chopra, S. R. Smith, M. Reza and D. H. Solomon (1975a). Ibid. 41, Chopra, I. J., J. Sack and D. A. Fisher (1975b). J. Clin. Invest. 55, Chopra, I. J., D. E. Williams, J. Orgiazzi and D. H. D. H. Solomon (1975c). J. Clin. Endocrinol. Metab. 41, 911. Chopra, I. J.(1976). J. Clin. Invest. 58, 32. Chopra, I. J., D. H. Solomon, U. Chopra, S. Wu, D. A. Fisher and Y. Nakamura (1978). Recent Prog. Horm. Res. 34, 521. DuRuisseau, J. P.(1965). J. Clin. Endocrinol. Metab. 25, Gavin, L., J. Castle, F. McMahon, P. Martin, M. Hammond and R. R. Cavalieri (1977). Ibid. 44, 733. Hedstrand, H. and L. Wide (1973). Br. Med. J. 17, 420. Konno, N.(1978). Endocrinol. Japon. 25, 635. Nagata, H., T. Izumiyama, K. Kamata, S. Kono, Y. Yukimura, T. Tawata, T. Aizawa and T. Yamada (1976). J. Clin. Endocrinol. Metab. 43, 1153.

6 KONNO Endocrinol. Japon. August 1980 Oddie, T. H., A. H. Klein, T. P. Foley and D. A. Fisher (1979). Clin. Chem. 25, Osiba, S.(1957). Jpn. J. Physiol. 7, 355. Pittman, C. S., J. B. Chambers Jr. and V. H. Read (1971). J. Clin. Invest. 50, Rastogi, G. K. and R. C. Sawhney (1976). Metabolism 25, 903. Siegels, S. Nonparametric statistics for the behavioral sciences, New York, McGraw-Hill, p. 127 (1956). Smals, A. G. H., H. A. Ross and P. W. C. Kloppenborg (1977). J. Clin. Endocrinol. Metab. 44, 998. Sterling, K., M. A. Brenner and E. S. Newman (1970). Science, 169, Takagi, A., Y. Isozaki, K. Kurata and S. Nagataki (1978). Jpn. J. Nucl. Med. 15, 275. Vagenakis, A. G., A. Burger, G. I. Portnay, M. Rudolph, J. T. O'Brien, F. Azizi, R. A. Arky, P. Nicod, S. H. Ingbar and L. E. Braverman (1975). J. Clin. Endocrinol. Metab. 41, 191. Verhoeven, R. P., T. J. Visser, R. Docter, G. Henneman and M. A. D. H. Schlekamp (1977). Ibid. 44, Watanabe, G., M. Uematsu and K. Horii (1963). Ibid. 23, 383. Westgren, U., A. Melander, E. WAhlin and J. Lindgren (1977). Acta Endocrinol. 85, 345.