Is Skeletal Responsiveness to Thyroid Hormone Altered in Primary Osteoporosis or Following Estrogen Replacement Therapy?

Authors


Abstract

Hyperthyroidism is characterized by increased bone turnover and resorptive activity. Similar changes in remodeling are seen in osteoporosis. To study the pathogenetic role of thyroid hormone in osteoporosis, we measured concentrations of free and total thyroid hormones and investigated the sensitivity of the skeleton toward thyroid hormones in 14 osteoporotic, 16 estrogen-treated, and 15 normal postmenopausal women with comparable thyroid status. Triiodothyronine (T3, 60 μg/day for 7 days) was administered to the three groups. The skeletal response was assessed by monitoring bone alkaline phosphatase (BAP), osteocalcin (BGP), and pyridinium cross-linked telopeptide domain of type I collagen (ICTP) in serum and urinary excretion of hydroxyproline (OHP), pyridinoline (PYR), and deoxypyridinoline (DPR) at days 0, 8, 15, and 57. Women on estrogen replacement therapy exhibited lower bone turnover than the normal postmenopausal women. Markers of bone formation were reduced by 19–43% and markers of resorption by 22–48%. The osteoporotic women displayed lower bone mass at the lumbar spine and the distal forearm (p < 0.01–0.001), but the levels of biochemical markers of bone formation and resorption were comparable to values obtained in the normal postmenopausal women. T3 stimulation caused significant increases (p values ranging between 0.05–0.001) in all three groups of the resorptive markers: ICTP (47%, 47%, 45%), OHP (29%, 30%, 33%), PYR (43%, 27%, 51%), and DPR (42%, 24%, 59%). Of the formative markers, only BGP increased significantly (32%, 40%, 47%) (p < 0.001). At day 57, however, all three formative markers increased compared with day 15 (p < 0.05–0.001). No significant differences in bone markers were demonstrated between groups. In the osteoporotic group, as the only group, serum calcium increased (p < 0.05) and serum PTH fell (p < 0.05). In conclusion, osteoporosis and estrogen substitution are not characterized by altered concentrations of thyroid hormones or responsiveness to thyroid hormones at the level of individual bone cells; however, altered responses pertaining to PTH and calcium were detected.

INTRODUCTION

The etiology and pathogenesis of primary osteoporosis are not known. Osteoporosis is defined by low bone mass compared with age- and gender-matched normal individuals. However, the low bone mass could be the result of either low peak bone mass or increased loss of bone after attainment of peak bone mass.

For more than 100 years, thyroid hormones have been known to have profound influence on bone.1 Histomorphometric studies of trabecular bone in patients suffering from hyperthyroidism have revealed increased activation frequency, negative bone balance, and, as a consequence, decreased trabecular bone volume.2–13 In cortical bone, cortical width was normal but porosity was greatly increased.3,10,13–16 These bone histomorphometric findings in hyperthyroidism are in accordance with calcium kinetic studies6,9,17–19 and the increases in biochemical markers of bone turnover reported.6,20–24 The biochemical markers showed highly significant positive correlations to thyroid hormone concentration both as total,21,25,26 and as free hormones.20,24,27 These changes described in hyperthyroidism are similar to those observed in postmenopausal osteoporosis and are known to be prevented by estrogen replacement therapy. Increased levels of thyroid hormone or increased sensitivity to thyroid hormones could over years result in increased age- or menopause-related bone loss or, in the younger years, decreased peak bone mass.

Estrogen replacement therapy after menopause has been shown by Schoutens et al.28 to decrease serum free triiodothyronine (FT3) levels. This may, however, also relate to progesterone, since this compound, in contrast to estrogen, decreases serum total triiodothyronine (TT3) and free thyroxine (FT4) (FT3 was not measured).29

The aim of the present study was therefore to investigate concentrations of thyroid hormones in the circulation and the skeletal sensitivity to thyroid hormones in postmenopausal osteoporotic women and women on estrogen replacement therapy in comparison with normal postmenopausal women.

MATERIALS AND METHODS

Materials

Fourteen normal postmenopausal women, 16 normal postmenopausal women on estrogen replacement therapy, and 14 women with osteoporosis participated in this study. Primary osteoporosis was defined by the presence of at least one low-energy vertebral fracture. Exclusion criteria were thyroid disease, parathyroid disease, diabetes mellitus, heart disease, hypertension, cancer, liver disease, or any other disease or medication (apart from osteoporosis and estrogen replacement therapy according to inclusion criteria) that could influence bone or mineral metabolism.

The women were divided into three groups (Table 1). Group 1, the osteoporotic group, included 14 postmenopausal women with spinal osteoporosis. The mean age was 63.0 years (range 57–70). The mean years since menopause was 15.4 years (range 2–24). Group 2, the estrogen substituted group, comprised 16 normal postmenopausal women, who were taking estrogen (n = 2) or a combination of estrogen and progesterone (n = 14) replacement. The mean age was 58.7 years (range 55–65). The mean years since menopause was 10.1 years (range 4–14). Group 3, the late postmenopausal group, consisted of 14 normal untreated postmenopausal women. The mean age was 61.9 years (range 57–68). The mean years since menopause was 13.6 years (range 7–30). The study was performed in accordance with the Helsinki II declaration and approved by the local ethical committee.

Table Table 1. Descriptive Data, Estrogen Status, and Energy Expenditure in Postmenopausal Osteoporotic Women, Postmenopausal Women on Estrogen Replacement Therapy and Normal Postmenopausal Women
original image

Design

All participants received oral triiodothyronine, (T3) 20 μg three times per day for 7 days. Biochemical markers of bone and mineral metabolism were measured on days −14, 1, 8, 15, and 57. Day 1 is the day the stimulation was initiated. Days −14 and 1 were analyzed separately, and afterward the mean of the two values was used as prestimulation value = day 0.

Methods

Biochemistry: Blood samples were drawn after an overnight fast and urine collected during a 24-h period on a gelatine-restricted diet. Serum and urine samples for determination of biochemical bone markers were frozen at −80°C until analysis. All other analyses were performed without delay.

Calcium, phosphate, and creatinine were measured in serum and urine according to standard laboratory methods. Serum parathyroid hormone (PTH) (1–84) was measured by a radioimmunoassay (Allegro Intact PTH, IRMA from Nichol's Institute, San Juan, Capistrano, CA, U.S.A.). The assay measures only the active intact PTH and not degradation products from cleavages of PTH.30,31 The intra-assay coefficient of variation (CV) was 2.6%, and the interassay CV was 5.8%.

Serum 25-hydroxyvitamin D (D2 + D3) was measured by a radioimmunoassay (Incstar Kit 60160; Incstar Corp., Stillwater, MN, U.S.A.) after acetonitrile extraction. Intra- and interassay CVs were 8 and 15%, respectively. Serum 1,25-dihydroxyvitamin D was extracted by acetonitrile, purified through a C18-OH reverse phase column and finally measured by a radioimmunoassay (Nichols-Kit 40-6040). Intra- and interassay CVs were 6.5 and 13.2%.

Thyroxine (T4) and T3 were measured by radioimmunoassay and equilibrium dialysis.32 Thyroid stimulating hormone (TSH) levels were measured by an immunoradiometric assay (RIA-GNOST, Behring Werke AB, Marburg, Germany). The intra- and interassay CVs for total T4 were 5 and 11.5%, for total T3, 5 and 12%; for free T4, 11.5 and 14%; and for free T3, 10 and 13.1%, respectively.

Serum cross-linked carboxy-terminal telopeptide of type I collagen (S-ICTP) was measured by an equilibrium radioimmunoassay. Intra- and interassay CVs were 5 and 6%, respectively.33 Urinary hydroxyproline (U-OHP) was measured spectrophotometrically with p-dimethylaminobenzal-dehyde substrate (Organon Tecnica, B.V. Boxtel, The Netherlands) Intra- and interassay CVs were 10 and 12%, respectively. Urinary pyridinoline (U-PYR) and deoxypyridirioline (U-DPR) were acid hydrolyzed and measured by fluorimetry after reversed phase high performance liquid chromatography (HPLC).34 Intra- and interassay CVs were both 10%. To compensate for sampling errors OHP, PYR, and DPR excretions were expressed as ratios relative to creatinine. Serum alkaline phosphatase (S-AP) was measured spectrophotometrically using p-nitrophenylphosphate as substrate according to recommendations from the Scandinavian Committee on Enzymes.35 Intra- and interassay CVs were 1.8 and 3.0%, respectively. Serum bone alkaline phosphatase (S-BAP) was measured spectrophotometrically in the supernatant after lectin precipitation.36 The intra- and interassay CVs were 8 and 25%, respectively. Serum osteocalcin (serum bone gla protein; S-BGP) was determined by a radioimmunoassay using rabbit antiserum against bovine bone-gla-protein.37 The intraassay CV was 5% and the interassay CV was 10%.

Indirect caloriometry: Energy expenditure (EE) was assessed in the fasting postabsorbtive state by indirect calorimetry for 30 minutes on day 1 and on day 8. A computerized, open circuit system was employed to measure gas exchange across a 25-l canopy (Deltatrac, Datex Instrumentarium, Inc., Helsinki, Finland). The monitor determines carbon dioxide production and oxygen consumption by multiplying dry air flow through the canopy with alterations in gas concentration over the canopy.38

Bone densitometry: The measurements of bone mineral concent (BMC) and density (BMD) were performed, using a Hologic (Waltham, MA, U.S.A.) dual-energy X-ray absorptiometry (DXA) scanner. The CVs were 0.4–5.0%.

Statistics: All results are given as means of the groups ± SD. After testing for normal distribution differences in groups, means were tested using unpaired two-tailed t-tests. Means of the same group at different days were tested using a paired two-tailed t-test. Calculations were performed using The Microsoft Excel Spreadsheet (Seattle, WA, U.S.A.), version 4.0 for Apple Macintosh.

RESULTS

Descriptive data

The osteoporotic women and the late postmenopausal women were age matched, but the estrogen-substituted women were significantly younger than the late postmenopausal group (p < 0.01) (Table 1). Mean menopausal age was comparable between the normal women and the osteoporotic and the estrogen-substituted women, respectively.

Heights were similar in the estrogen-substituted and the late postmenopausal group. The osteoporotic women were smaller than the late postmenopausal women, 158.8 versus 167.1 cm (p < 0.01). Body weights were comparable in the normal women and the osteoporotic and the estrogen-substituted women.

The osteoporotic women and the late postmenopausal women had similar levels of serum follicle stimulating hormone (S-FSH), serum luteinizing hormone (S-LH), and serum estradiol. The estrogen substituted group had significantly higher s-estradiol (p < 0.05) and correspondingly lower S-FSH levels (p < 0.01) than the late postmenopausal group.

No significant differences were observed in systolic and diabolic blood pressure and heart rate between the groups. Furthermore, no differences in energy expenditure was demonstrated between the groups.

Baseline data

In the following bone mineral measurements, calcium homeostasis and bone markers are compared between the osteoporotic and the estrogen-substituted groups and the normal postmenopausal group to detect significant physiological and pathophysiological differences.

The osteoporotic versus late postmenopausal women: The osteoporotic group displayed significantly lower bone mass at all three locations investigated (Table 2). BMC and BMD of the lumbar spine were 26 and 20% lower, respectively (p < 0.001 and p < 0.01). The Z score was −1.68 compared with −0.29 in the late postmenopausal group (p < 0.001). In the hip region, BMC and BMD of the trochanter were 18 and 15% lower in the osteoporotic group, respectively (p < 0.01 and p < 0.05). BMC and BMD of the femoral neck were 16 and 12% lower in the osteoporotic group compared with the late postmenopausal group, respectively (p < 0.01). BMC and BMD of Ward's triangle in the osteoporotic group were also lower than in the late postmenopausal group, 23 and 23%, respectively (p < 0.01). At the distal forearm, BMC and BMD of the proximal one-third were 10 and 11% lower, respectively (p < 0.05), of the mid-part 18 and 13% lower (p < 0.05), and of the ultradistal part 18 and 24% lower (p < 0.05 and p < 0.01). BMC and BMD of the distal forearm displayed a 16 and 14% reduction, respectively, in the osteoporotic group compared with the normal values in the late postmenopausal group (p < 0.05).

Table Table 2. BMC and BMD, Calcium Homeostasis, and Biochemical Markers of Bone Turnover in Postmenopausal Osteoporotic Women, Postmenopausal Women on Estrogen Replacement Therapy, and Normal Postmenopausal Women
original image

Serum levels of calcium, phosphate, and PTH and renal excretion of calcium and phosphate were not different between the two groups (Table 2). S-osteocalcin was significantly higher in the late postmenopausal than in the osteoporotic group (p < 0.05). S-total alkaline phosphatase tended to be higher in the osteoporotic group (p = 0.07). Peripheral thyroid hormones and TSH (Table 3) showed no differences between the two groups.

Table Table 3. Thyroid Parameters in Postmenopausal Osteoporotic Women, Postmenopausal Women on Estrogen Replacement Therapy and Normal Postmenopausal Women
original image

Estrogen substituted versus late postmenopausal women: BMC and BMD of the lumbar spine was 16 and 18% higher in the estrogen substituted group than in the normal postmenopausal group, respectively, (p < 0.05 and p < 0.01) (Table 2). These differences were also reflected in the Z score of the estrogen substituted group, 0.86 versus −0.29 in the normal postmenopausal group (p < 0.05). BMC and BMD of the hip were 12 and 14% higher in the estrogen substituted group compared with the normal postmenopausal group (p < 0.05 and p < 0.001, respectively). These higher values were observed in all regions: femoral neck (16 and 15%, p < 0.01), Ward's triangle (33 and 28%, p < 0.01), and for BMD, trochanter (12%, p < 0.01) and the intertrochanteric region (17%, p < 0.001). At the distal forearm, BMC and BMD of the proximal one-third were 10 and 13% higher, respectively, in the estrogen substituted group than in the normal postmenopausal group (p < 0.05 and p < 0.01). At the middle part, only the BMD was higher, 11% (p < 0.05), whereas at the ultradistal part, only a 17% higher value of BMC reached significance (p < 0.01). Both BMC and BMD of the distal forearm were higher, 11 and 11%, respectively, in the estrogen substituted group compared with the normal postmenopausal group (p < 0.05 and p < 0.01).

S-phosphate was significantly lower in the estrogen substituted group, (1.10 mmol/l) compared with the normal postmenopausal group (1.27 mmol/l) (p < 0.01) (Table 2). Serum calcium and PTH and renal excretion of calcium and phosphate were similar in the two groups. The markers of bone formation were lower in the estrogen-substituted group. S-osteocalcin was 45% (p < 0.001), S-total alkaline phosphatase 19% (p < 0.05), and S-bone alkaline phosphatase 36% lower than in the normal postmenopausal group (p < 0.001). Also, the markers of bone resorption revealed lower values, 26% in urinary excretion of hydroxyproline (p < 0.001), 39% in U-PYR (p < 0.01), 48% in U-DPR (p < 0.01), and 22% in S-ICTP (p < 0.01).

The free thyroid hormones were not different between the two groups; however, S-total thyroxine was significantly higher in the estrogen substituted group, 137 versus 120 nmol/l in the normal postmenopausal group (p < 0.05) (Table 3). S-total T3 showed the same tendency (p = 0.09). S-TSH was not different.

Results of stimulation

Thyroid status: Seven days of stimulation with T3 (60 μg/day), induced a 3-fold increase in S-total triiodothyronine (S-TT3) at day 8 in all three groups (p < 0.001) (Fig. 1). At days 15 and 57, S-TT3 had returned to prestimulation levels. S-free T3 (S-FT3) was increased to levels 3–3.5 above prestimulation levels (p < 0.001) at day 8. At day 15, the estrogen substituted group demonstrated a 9% reduction in S-FT3 (p < 0.01). At day 57, S-FT3 had returned to baseline levels in all three groups. S-total thyroxine (S-TT4) was decreased at day 8 by 23–27% (p < 0.001) in the three groups and at day 15 by 15–25% (p < 0.001). At day 57, all groups displayed values of S-TT4 comparable to prestimulation values. S-free thyroxine (S-FT4) followed the same pattern with a reduction of 23–30% (p < 0.001) in the three groups at day 8, and of 20–25% at day 15 (p < 0.05–0.001). At day 57, S-FT4 in the three groups had returned to baseline. S-thyroid stimulating hormone (S-TSH) was at day 8 reduced to 2–3% of prestimulation values in all groups (p < 0.001). At days 15 and 57, S-TSH had returned to baseline levels in all three groups.

Figure FIG. 1.

Changes in thyroid hormones, TSH, and energy expenditure in osteoporotic, estrogen-substituted, and normal postmenopausal women after T3 stimulation for 7 days.

Physiological effects: The participants lost weight during the week of thyroid hormone excess. The losses were 315–580 g or 0.5–0.9% in the three groups (p < 0.05–0.01); the change did not reach significance in the osteoporotic group. At day 15, body weights were as before stimulation in all groups.

The heart rate was increased in all three groups by 16–18% (p < 0.001). At days 15 and 57, the heart rate had returned to baseline. Systolic blood pressure was reduced in the osteoporotic groups at day 8 by 7% (p < 0.01). At days 15 and 57, systolic blood pressure was still reduced by 5% (p < 0.05) and 8% (p < 0.05), respectively, in the osteoporotic group. Diastolic blood pressure was unaltered at day 8, but at day 15 the estrogen-substituted group revealed a reduction in diastolic blood pressure of 6% (p < 0.01). At day 57, all three groups had diastolic blood pressure not different from baseline levels.

EE increased in the three groups by 11–14% (p < 0.01–0.001) (Fig. 1).

Calcium homeostasis: S-calcium increased significantly in the osteoporotic group upon stimulation with thyroid hormone to 2.35 from 2.31 mmol/l before stimulation (p < 0.05). However, the estrogen-substituted group revealed a decrease in S-calcium from 2.31 to 2.28 mmol/l (p < 0.05). In all groups, S-calcium had returned to prestimulation levels at day 15. At day 57, the decrease in the estrogen-substituted women again reached significance (p < 0.05), and the other groups exhibited unchanged values.

S-phosphate decreased at day 8 in the estrogen-substituted group by 8% (p < 0.01). At days 15 and 57, S-phosphate was unchanged compared with prestimulation values in the three groups.

The 24-h excretion of calcium did not reveal any significant changes at any day in any of the groups. However, the 24-h excretion of phosphate increased in all groups by 27–35% (p < 0.05–0.001). At days 15 and 57, U-phosphate had returned to values not significantly different from prestimulation values.

S-PTH decreased by 13% upon thyroid hormone stimulation in the osteoporotic group (p < 0.05). At day 15, all groups had S-PTH not different from prestimulation values. At day 57, the estrogen-substituted group displayed an 8% reduction compared with the prestimulation level (p < 0.05), and the two other groups had unchanged values.

Biochemical bone markers: S-osteocalcin (S-BGP), the sensitive marker of bone formation, increased in the three groups by 32–47% (p < 0.001) (Fig. 2). At day 15, the late postmenopausal groups had values not different from day 0; however, the osteoporotic and the estrogen-substituted groups revealed a reduction of 16% (p < 0.05) and 15% (p < 0.001), respectively. If the values at day 57 were compared with day 15, all the groups had 18–21% higher values (p < 0.05–0.001). S-total alkaline phosphatase (S-TAP) did not change significantly in any of the groups. However, as for S-BGP at day 57, S-TAP was increased by 5% in the osteoporotic group compared with day 15 (p < 0.05). s-BAP did not change at day 8, but at day 15, S-BAP was reduced by 8% in the osteoporotic (p < 0.05) and by 10% in the normal postmenopausal (p < 0.01) women, respectively. At day 57, all groups had unchanged values. However, comparisons between day 57 and day 15 again revealed increases in the osteoporotic and the late postmenopausal groups of 8 and 11% (p < 0.05).

Figure FIG. 2.

Changes in biochemical markers of bone formation in osteoporotic, estrogen-substituted, and normal postmenopausal women after T3 stimulation for 7 days.

The bone resorption marker S-ICTP increased in the three groups following thyroid hormone stimulation by 45–47% (p < 0.001) (Fig. 3). At day 15, S-ICTP was still increased by 14–16% (p < 0.05–0.01) in the normal and the estrogen-substituted groups. At day 57, all groups had values as before stimulation. The 24-h U-OHP increased by 29–33% (p < 0.05–0.001). At days 15 and 57, U-OHP had returned to prestimulation levels. The 24-h U-PYR increased by 27–51% (p < 0.05). At day 15, the three groups had values not different from prestimulation levels. Renal excretion of deoxypyridinoline (U-DPR) increased by 24–59% in the three groups (p < 0.05). At day 15, U-DPR had returned to baseline levels.

Figure FIG. 3.

Changes in biochemical markers of bone resorption in osteoporotic, estrogen-substituted, and normal postmenopausal women after T3 stimulation for 7 days.

Comparisons of the changes between groups: The osteoporotic versus normal postmenopausal normal women. The relative decrease in S-PTH in the osteoporotic group of 13% at day 8 was significantly different from the relative increase in the late postmenopausal normal group of 4% (p < 0.05). The estrogen-substituted versus normal postmenopausal women. The relative decrease in diastolic blood pressure at day 15 in the estrogen-substituted group of 6% was significantly different from the relative increase of 2% in the late postmenopausal group (p < 0.05).

The relative decrease in S-calcium of 1% in the estrogen-substituted group at day 8 was different from the relative increase of 1% in the normal group (p < 0.01). At day 15, the continuously elevated S-calcium in the estrogen-substituted group and decreased S-calcium in the normal group was still significantly different (p < 0.05). The relative reduction of 7% in S-PTH at day 57 in the estrogen-substituted group was significantly (p < 0.05) different from the relative increment of 14% in the normal group.

The relative decreases in S-BGP at day 15 of 14% in the estrogen-substituted group versus 3% in the late menopausal group were significantly different (p < 0.05) (Fig. 2).

DISCUSSION

The estrogen-substituted women and the normal postmenopausal women were intended to be age matched. The difference in mean age between the two groups was 3.2 years and significant. However, we do not consider this small difference in mean age to be of any importance, compared with the difference in sex hormone status between the two groups.

Biochemical markers of bone formation and resorption were normal in the patients with osteoporosis. But BMC and BMD were lower at nearly all sites measured. This lower bone mass could have been achieved by at least two mechanisms. It could be the result of a low peak bone mass and normal age- and menopause-related bone loss during the years since peak bone mass was achieved. Peak bone mass could have been as in the normals, combined with an increased age- or menopause-related bone loss. This study was not designed to elucidate the relative importance of the two mechanisms in the development of osteoporosis, but only to examine one possible explanation for a higher than normal bone loss.

The osteoporotic women displayed no differences in either calcium homeostasis or thyroid status in comparison with normal postmenopausal women. The hypothesis of increased levels of thyroid hormones in osteoporosis could therefore not be confirmed in this study.

The estrogen-substituted postmenopausal women had significantly lower levels of biochemical markers of bone turnover than the normal unsubstituted postmenopausal women. The level of these markers were as in the premenopausal women (data not shown)67 as also reported by others.39–44 This low turnover would, despite the fact that the balance at individual bone remodeling unit level still may be negative, reduce the loss of bone.45–49 In accordance with this, BMC and BMD were higher at nearly all sites measured. The estrogen-substituted postmenopausal women had lower than normal S-phosphate, but otherwise their calcium homeostasis was normal. They had elevated S-TT4, but S-FT4 was normal, in accordance with the stimulatory effect of estrogen on the thyroid binding proteins.50,51

The three groups responded nearly identically to stimulation with thyroid hormone. After increasing the level of T3 3-fold, bone resorption was increased, evaluated by U-OHP and S-ICTP with 50%, and was still increased in the estrogen-substituted and the normal women 1 week after stimulation was stopped. The relative increase in the osteoporotic women was not different from the increase in the other groups. From these data, it is not possible to dissect whether the increased activity is caused by increased activity of the already active resorbing osteoclasts, increased resorptive activity due to recruitment of new osteoclasts, or a combination of both.

S-BGP revealed an increase of the same magnitude as the resorptive markers. This increase is primarily caused by increased activity of already functioning osteoblasts, since initiation of new bone formation does not take place unless preceded by resorption.52 The other two markers of bone formation, S-TAP and S-BAP, did not exhibit any changes. Theoretically, this could be explained by the known increase in the clearance of alkaline phosphatase in hyperthyroid disease.53 Another argument for this theory is that the increase in S-BGP from day 15 to day 57 is also seen in S-TAP and S-BAP. This increase in formative biochemical markers from day 15 to day 57 suggests that during the 8 days of stimulation with thyroid hormone many new remodeling sites have been initiated, and they have reached the formative phase as expected 8 weeks later.54,55

The overall effect on bone remodeling evaluated by biochemical markers after 7 days of thyroid hormone excess is stimulation of the already on-going resorption and formation and initiation of new remodeling sites. No differences in the response could be demonstrated among the three groups.

Calcium homeostasis is known to be affected by thyroid hormone excess. An increase in S-calcium and S-phosphate and a decrease in S-PTH is reported in some studies of hyperthyroidism.55–65 After 1 week of thyroid hormone stimulation, the osteoporotic women displayed a small but significant increase in S-calcium, S-PTH was significantly decreased, but S-phosphate was unchanged. The relative increase in S-calcium was not significantly different from the change in S-calcium in the normal women, but the decrease in S-PTH was significantly different. One could speculate that the osteoporotic women, despite no relative differences in markers of bone turnover, had a more pronounced stimulation of bone resorption, causing a burst of calcium out of the bone and inducing the reduction in S-PTH. Other possible explanations could be that the osteoporotic women had a higher sensitivity of S-PTH toward S-calcium or that the bone formation is impaired and unable to use calcium in the same magnitude in the formative process, as is released during the resorptive activities. The estrogen-substituted postmenopausal women on the contrary displayed a reduction in both S-calcium and S-phosphate without changes in S-PTH.

It is unlikely that the small changes in S-calcium, S-phosphate, and S-PTH are of any importance at all. All groups had an increase in urinary excretion of phosphate in contrast to what would be expected, if the decrease in S-PTH was of physiological importance. The increase in U-phosphate is most likely a direct effect of thyroid hormones on the kidneys; however, efflux of phosphate from soft tissues and the bone after stimulation of resorptive activity is also a possible mechanism behind the increased renal excretion of phosphate.66

This study does not provide any evidence for altered concentrations of circulating thyroid hormones or sensitivity of the bone remodeling system to thyroid hormone during estrogen substitution or in primary osteoporosis. In osteoporosis, the biochemical markers were as in normals, and the only difference from normal women was the increased S-calcium and the decreased S-PTH after stimulation with thyroid hormone. If this is a sign of increased sensitivity of the resorptive bone cells to thyroid hormones, causing a more pronounced increase in resorption than in normal postmenopausal women, it cannot be of great importance since the relative increases in the otherwise sensitive biochemical markers were not different from the increases seen in the normals.

From this study, it can be concluded, that the bone remodeling system displays unaltered sensitivity toward thyroid hormones during estrogen substitution and in women with spinal osteoporosis. In a similar study we performed in premenopausal and early postmenopausal women, we found unaltered sensitivity toward thyroid hormones across menopause.67 It is not possible from these studies to conclude anything about thyroid hormone sensitivity in patients during the development of osteoporosis. However, we do not find it likely that sensitivity could be altered just for a short period of time, since sensitivity is based upon a number of receptors and the efficiency of these in bringing up changes in the cell. This study did not support the theory that osteoporosis could be caused by increased levels of thyroid hormones in the circulation or by increased sensitivity of bone toward thyroid hormones.

Acknowledgements

This study was supported by the P.A. Messerschmidt og hustrus Fond, the Danish Medical Research Council, the Erik Hørslev og hustru Birgits Fond, the Hafnia Fond, the Købmand Sven Hansen og hustru Inas Fond, the Dagmar Marshalls Fond, the Den almindelige Danske Lægeforenings Fond, the Lægekredsforeningen for Aarhus Amt, the Beckett-Fonden, the Direktør E. Danielsens og hustrus Fond, the Løvens Kemiske Fabriks Fond, the Torben Frimodt og Alice Frimodts Fond, the Oda og Hans Svenningens Fond, and the Mimi and Victor Larsens Fond. The technical assistance of Ilse Rasmussen, Donna Lund, Birthe Weinell, Jane Lundkvist, Hanne Damgaard, Mette Carstens, Birthe Gosvig, and Hanne Mengel is gratefully acknowledged.

Ancillary