Full text of DOI:10.1007/BF03345061

J. Endocrinol. Invest. 23: 727-731, 2000
Bone turnover in hyperthyroidism before and after
thyrostatic management
G.C. Isaia*, C. Roggia*, D. Gola*, M. Di Stefano*, G. Gallone**, G. Aimo***,
P. Ardissone*, and M. Mussetta*
*Department of Internal Medicine, University of Turin; **Endocrinology Division, Ospedale Mauriziano;
***Baldi e Riberi Laboratory, Azienda Ospedaliera San Giovanni Battista, Turin, Italy
ABSTRACT. Hyperthyroidism is associated with en-
hanced osteoblastic and osteoclastic activity, and
patients frequently have low bone mineral density
and high bone turnover. The aim of this study was
to examine the bone formation and resorption
markers trend in 12 female patients, before and
after normalization of thyroid activity. The follow-
ing measurements were made at baseline and 1
and 6 months after hormone normalization in-
duced by methimazole treatment: total alkaline
phosphatase (ALP), bone alkaline phosphatase
(BALP), collagen type C-terminal propeptide (PICP),
osteocalcin (BGP), telopeptide (ICTP), urinary-hy-
droxyproline/urinary creatinine (uOHP/uCreat), uri-
nary calcium/urinary creatinine (uCa/uCreat) and
deoxypyridinoline crosslinks (D-Pyr). Compared
with controls, all of these parameters were signif-
icantly increased (ALP p=0.014; BALP p=0.0001;
PICP p=0.013; BGP p=0.009; ICTP p=0.0001;
uOHP/uCreat p=0.002; uCa/uCreat p=0.044;
crosslinks p=0.0001). After treatment the values
of ALP, BALP and PICP in hyperthyroid patients
showed an initial slight increase and then a signifi-
cant downwards trend (ALP p=0.008, BAP p=
0.001, PICP p=0.026). Furthermore, resorption
markers showed a significant decrease (uOHP/
uCreat p<0.005 and D-Pyr p<0.008). As regards
lumbar BMD patients, measurements were signif-
icantly reduced in comparison with the control
group (p=0.005). Six months after serum thyroid
hormones level normalization, we observed a sig-
nificant increase (p=0.014 vs baseline). Both neo-
formation and resorption markers are useful to as-
sess pathological bone turnover and bone in-
volvement in hyperthyroidism. They could also be
employed to monitor the effect of antithyroid
treatment on bone and to indicate if bone antire-
sorption therapy should be considered.
(J. Endocrinol. Invest. 23: 727-731, 2000)
©2000, Editrice Kurtis
INTRODUCTION
Morphometric and dynamic studies have shown
that increased secretion results in a significant ex-
cess of resorption that markedly disturbs the bal-
ance between resorption and formation (1, 11, 12).
The bone remodeling changes observed in hy-
perthyroidism have been attributed to increased
cAMP production, and the probably prostaglan-
din-related release of resorption-stimulating fac-
tors (13-15). In fact, an increase of IL (interleukin)-
6 and IL-8 was found in hyperthyroid patients (16),
even though the precise pathogenetic meaning
of cytokines on bone metabolism is unclear; yet,
data obtained on osteoblasts stimulated in vitro
by T3 (17) seem to show that osteoporosis in hy-
perthyroidism is closely associated with serum IL-
6 levels.
Thyrotoxicosis enhances bone turnover by increas-
ing the number of osteoclasts and resorption sites.
Both cortical and trabecular bone thus display a
substantial increase in both superficial osteoclastic
resorption and osteoblastic bone formation (1, 2).
This increased resorptive-formation ratio signifi-
cantly reduces bone mineral density (BMD) (3, 4),
and makes thyrotoxicosis a risk factor for osteo-
porosis and its associated fractures (5, 6).
Normalization of thyroid hormone levels with an-
tithyroid drugs is followed by significant BMD in-
crements after 1-2 years (7-10).
Key-words: Bone densitometry, bone formation, bone markers, bone re-
sorption, hyperthyroidism.
Functional and structural bone changes in hyper-
thyroidism are normally associated with elevated
levels of serum alkaline phosphatase (ALP) and os-
teocalcin (BGP), two biological bone formation
Correspondence: Prof. Giancarlo Isaia, Department of Internal Medicine,
University of Turin, Corso A.M. Dogliotti 14, 10126 Torino, Italy.
Accepted March 31, 2000.
727

G.C. Isaia, C. Roggia, D. Gola, et al.
markers (9, 18, 19). Direct stimulation by thyroid
hormone of osteoblasts possessing nuclear T3 re-
ceptors has been demonstrated (20). The resorp-
tion markers urinary hydroxyproline (uOHP) and
pyridinoline are also elevated and their levels sig-
nificantly correlate with the circulating thyroid hor-
mone concentration (19, 21, 22). Garnero et al.
have shown that urinary pyridinoline cross-link ex-
cretion is a highly sensitive marker for altered bone
metabolism, whereas BGP osteocalcin and bone al-
kaline phosphatase (BALP) increments are less im-
pressive (22).
Since biochemical bone turnover markers are of
particular interest in the investigation of bone dis-
orders associated with thyroid diseases, in this study
we measured their plasma and urinary levels to de-
termine their changes in hyperthyroid patients be-
fore and during treatment and to define the effects
of hyperthyroidism on bone turnover.
assay the free hormone fractions (CV within assay
was 3.5% for fT3 and 3.7% for fT4; CV between as-
say was 6.5% for fT3 and 5.0% for fT4). TSH levels
were quantified by the Biocode TSH IRMA kit (CV
within assay 6.2% and between assay 7.1%; sensi-
tivity 0.05 (μU/ml). Blood phosphorus and calcium
were evaluated enzymatically and ALP by the IFCC
method (ALP CV 2.6%).
uCa was assayed by photonic absorption, while
uOHP was tested by HPLC with a fluorometric de-
tector. Results of both markers were expressed as a
ratio to creatinine.
PTH was determined by RIA methods using Incstar
Corporation kits (CV within assay 5%; CV between
assay 7.3%). Also BGP was measured on serum af-
ter centrifugation and storage at -20 C by RIA meth-
ods (Incstar Corporation kits) (CV within assay 3.8%
and between assay 4.8%).
The bone isoenzyme of ALP was tested by an IRMA
method from Hybritech (CV within assay 4.2%; CV
between assay 7.2%). PICP and ICTP were deter-
mined using the RIA method provided by Orion
Diagnostica (CV within assay 2.1% and between as-
say 4.1%; CV within assay 4.4% and between assay
4.9%, respectively).
MATERIALS AND METHODS
We examined 12 female patients aged 20-68 (5
post-menopausal) with a clinically evident and bio-
logically confirmed (high thyroid hormone level and
TSH inhibition) hyperthyroidism due to Graves’dis-
ease (3 patients) and multinodular goiter (9 pa-
tients). None of them had been treated with anti-
thyroid drugs or other drugs known to influence
bone calcium metabolism (including hormonal re-
placement therapy), and they were all free from
acute or chronic diseases.
The urinary D-Pyr crosslinks were measured by an
enzyme immunoassay provided by Metra Biosy-
stems (CV within assay 4.3% and between assay
4.6%).
Lumbar BMD was measured by means of the DXA
technique (Hologic QDR 1000 system); our CV is
1.2% in vivo and 0.5% in vitro.
The controls were 12 age-matched, clinically healthy
with normal plasma TSH and thyroid hormones wom-
en; the 5 post-menopausal controls did not differ
from the 5 post-menopausal hyperthyroid patients
for what regards years since menopause.
The following parameters were evaluated at the
baseline in both groups:
Statistics were obtained by using Student’s t test
for paired and unpaired data.
RESULTS
The blood calcium, phosphorus and PTH levels of
hyperthyroid patients were not significantly differ-
ent from those of the controls and remained with-
in the normal limits throughout the study.
Patients’ TSH, thyroid hormone, bone marker and
BMD are compared with those of the controls in
Table 1.
Figure 1 shows the percent variations of all of the
parameters in patients vs controls in basal condi-
tions.
A) blood FT3, FT4, TSH, calcium and phosphorus,
ALP, intact PTH, medium molecule PTH, BGP,
BALP, C1 procollagen (PICP), telopeptide (ICTP);
B) urinary deoxypyridinoline (D-Pyr) crosslinks, urinary
hydroxyproline/urinary creatinine (uOHP/uCreat),
urinary calcium/urinary creatinine (uCa/UCreat) and
urinary phosphorus (2-hour morning urine);
C) L2-L4 bone density.
All of the patients received for 2 months a mean
dose of 17.5 mg/day methimazole (range 10-20
mg) and their thyroid hormones levels were mea-
sured monthly; then the mean dose was of 10 mg
(range 5-15 mg).
All of the parameters were reassessed 1 month and
6 months after thyroid compensation.
The Amerlex-mAb FT3 and FT4 kits were used to
No significant correlation between FT3 and FT4 and
the turnover markers was found in the patient
group. The temporary trend of bone formation mar-
kers (ALP, BALP, BGP, PICP) are shown in Figure 2,
those of the resorption markers (ICTP, OHP/uCreat
and D-Pyr crosslinks) in Figure 3. Total ALP base-
line was higher (p=0.014) in patients than in con-
trols (Table 2) and strongly increased, although not
728

Bone turnover in hyperthyroidism
BALP (mcU/l)
ALP (U/l)
Table 1 - TSH, FT3, and FT4 basal values in hyperthyroid pa-
200
150
100
50
50
*
tients.
*
40
ꢀ
30
20
10
0
ꢀꢀ
Basal 1 month 6 months
PICP (mcU/l)
Patients
N. 1
TSH (μU/ml) FT3 (pg/ml)
FT4 (pg/ml)
48.9
0
0.01
0.00
19.7
21
Basal 1 month 6 months
BGP (ng/ml)
N. 2
52.1
N. 3
0.00
19
49.3
20
15
10
5
300
200
100
0
*
ꢀꢀꢀ
N. 4
0.00
21.4
22.3
20
52.7
*
*
N. 5
0.01
56.4
0
Basal 1 month 6 months
Basal 1 month 6 months
N. 6
0.02
47.2
ꢀꢀ
ꢀꢀꢀ
*NS; •p=0.008 vs 1 month; p=0.001 vs 1 month; p=0.026 vs 1 month
N. 7
0.00
17
43.5
Fig. 2 - Bone formation marker patterns in hyperthyroid pa-
tients: baseline and one and six months after thyroid hormone
normalization.
N. 8
0.00
20.3
21
49
N. 9
0.01
51
N. 10
N. 11
N. 12
Mean values
Normal range
0.00
25.2
18.8
20.9
20.55 2.02
2.2-4.7
59.7
0.00
47.3
uOHP/Creat (mg/g Creat)
ICTP (mcU/l)
10
8
80
60
40
20
0
0.00
50.1
6
0.00 0.01
0.3-5
45.97 13.54
8-20
ꢀ
ꢀ
4
ꢁ
2
0
Basal 1 month 6 months
Basal 1 month 6 months
D-Pyr crosslinks (nM/nM Creat)
20
ꢀ
p=0.0001 vs 1 baseline
ꢁp=0.003 vs 1 baseline
ꢂp=0.003 vs baseline
ꢂꢂp=0.008 vs baseline
significantly, one month after compensation before
decreasing (Fig. 2).
15
10
5
ꢂ
ꢂꢂ
Similar patterns were displayed by the other mark-
ers: the patients’ baseline BALP was significantly
higher than in the controls (p<0.0001) (Table 2)
and rose one month from normalization before
falling within the normal range after six months
(Fig. 2).
The patients’ higher baseline PICP (p=0.013 vs the
controls) displayed a slight significant change dur-
ing the study (p=0.026 vs one month) (Fig. 2).
The patients’ BGP baseline was slightly higher
than in the controls (p=0.009) (Table 2). The de-
0
Basal 1 month 6 months
Fig. 3 - Bone resorptions marker patterns in hyperthyroid pa-
tients: baseline and one and six months after thyroid hormone
normalization.
crease after normalization was not significant
(Fig. 2).
The ICTP baseline was significantly higher than in
the controls (p=0.0001) (Table 2). Substantial de-
creases after compensation resulted in levels sig-
nificantly below the baseline (p=0.0001) (Fig. 3).
The uOHP/uCreat ratio was higher than in the con-
trols (p=0.002) (Table 2) and fell significantly after
normalization (Fig. 3).
Similar patterns were displayed by uCa/uCreat and
the D-Pyr crosslinks (Table 2). The latter was signifi-
cantly higher (p=0.0001) than in controls (Table 2)
and both fell significantly (p=0.003) one month after
compensation and mantained this decrease after six
months (p=0.008) (Fig. 3).
300
250
200
150
100
50
0
Lastly, the baseline lumbar BMD was significantly
lower in the patients (p=0.005) and rose slightly,
but significantly, six months after compensation
(Fig. 4).
ALP BALP PICP BGP ICTP uOHP uCa D-Pyr
crosslinks
Fig. 1 - Percent differences between patient and control bone
turnover marker baselines.
729

G.C. Isaia, C. Roggia, D. Gola, et al.
Table 2 - Mean SD in controls and hypertension patients.
Controls
Basal
hyperthyroid patients
TSH (μU/ml)
3.53 0.46
2.82 0.49
10.45 0.9
77 26.9
0.00 0.00
p<0.0001
p<0.0001
p<0.0001
p=0.014
p=0.0001
p=0.013
p=0.009
p=0.0001
p=0.002
p=0.44
FT3 (pg/ml)
20.66 2.17
50.69 4.73
139.4 64.19
33.23 14.07
234.5 104.09
9.68 5.50
FT4 (pg/ml)
ALP (U/l)
BALP (μU/l)
12.21 7.08
122.8 36.67
3.7 1.74
PICP (μU/l)
BGP (ng/ml)
ICTP (μU/l)
2.71 1.02
19.99 6.65
0.07 0.05
6.5 1.42
10.20 3.22
80.30 48.85
0.19 0.17
uOHP/Creat (mg/g Creat)
uCa/Creat (mg/mg Creat)
D-Pyr Crosslinks/Creat (nMol/mMol Creat)
BMD (g/cm2)
19.65 8.91
0.85 0.2
p=0.0001
p=0.05
1.010 0.18
1.2
measuring samples from the same subject in different
analytical batches (25). Our data suggest that BGP, at
least measured by this method, is less able than BALP
to indicate changes in bone metabolism.
It is clear that hyperthyroidism shifts the balance of
bone remodeling from formation to resorption (13,
19, 21, 22).
The bone resorption markers, on the other hand,
fell in step with the reduction in thyroid hormones
and quickly became normal when compensation
was obtained (21, 22).
D-Pyr displayed the most rapid and significant re-
duction. This behavior as observed by other authors
(16) is different from that of bone formation mark-
ers and suggests the persistence of the increase of
bone formation ever after attainment of the euthy-
roid state.
Also the L2-L4 BMD, significantly lower in hyperthy-
roid patients than in controls, raised significantly in a
very short time (26) and so we can conclude that spe-
cific bone management therefore is not required.
As pointed out by Siddiqi et al. however, the exact
role of bone markers in hyperthyroidism remains to
be defined (9).
1
•
Hyperthyroids
Controls
**
*
0.8
0.6
0.4
0.2
0
*p=0.005 vs controls
**p=0.039 vs controls
ꢀp=0.014 vs baseline
Baseline
6 months
Fig. 4 - BMD patterns in controls and in hyperthyroid patients
(baseline and six months after thyroid hormone normalization).
DISCUSSION
Hyperthyroidism is accompanied by marked chan-
ges in bone remodeling that may lead to substan-
tial BMD reduction.
In our patients the transient increase in specific bone
formation markers (ALP, BALP and PICP) observed
one month after the achievement of euthyroidism was
evidence of enhanced osteoblastic activity, probably
reflecting an attempt of bone turnover to become
positive, in response to increased resorption; alterna-
tively it can be explained by the persistent slight in-
crease in cortical bone turnover (23).
REFERENCES
In contrast, BGP concentrations did not change sig-
nificantly and, at least with the method used in the
present study, are of little assistance in the evaluation
of bone turnover. Other workers, however, have re-
ported significant BGP increases in hyperthyroid pa-
tients and a significant correlation between FT3 and
BGP baseline (9, 11, 18, 24). This discrepancy may be
due to the variability of different BGP assays and of
1. Melsen F., Heriksen E.F., Charles P.
Morphometric and dynamic studies of bone change
in hyperthyroidism.
Acta Pathol. Microbiol. Scand. 1977, 85: 141-150.
2. Meunier P.J., Bianchi G.G.S., Edouard C.M., Bernard
J., Courpron P., Vignon G.
Bony manifestations of thyreotoxicosis.
Orthop. Clin. North Am. 1972, 3: 745-774.
730

Bone turnover in hyperthyroidism
3. Riggs B.L., Melton III L.J.
15. Moselkilde L., Eriksen, E.F., Charles P.
Involutional osteoporosis.
Effects of thyroid hormones on bone and mineral
metabolism.
N. Engl. J. Med. 1990, 314: 1676-1686.
Endocrinol. Metab. Clin. North Am. 1990, 19: 35-63.
4. Baran D.T., Braverman L.E.
Editorial: Thyroid hormones and bone mass.
J. Clin. Endocrinol. Metab. 1991, 72: 1182-1183.
16. Siddiqi A., Monson J.P., Wood F., Besser G.M.,
Burrin J.M.
Serum cytokines in thyrotoxicosis.
J. Clin. Endocrinol. Metab. 1999, 84: 435-439.
5. Grant D.J., Marion E.T., McMurdo M.E., Mole P.A.,
Paterson C.R.
Is previous hyperthyroidism still a risk factor for os-
teoporosis in post-menopausal women?
Clin. Endocrinol. 1995, 43: 339-345.
17. Siddiqi A., Burrin J.M., Wood D.F., Monson J.P.
T3 regulates the production of IL-6 and IL-8 in hu-
man bone marrow stromal and osteoblast-like cells.
J. Endocrinol. 1998, 157: 453-461.
6. Bauer D.C., Cummings S.R., Tao J.L., Browner W.S.,
SOFR Group.
18. Lee M.S., Kim K.Y., Lee M.C., Cho B.Y., Lee H.K.,
Koch C.S., Min H.K.
Hyperthyroidism increases the risk of hip fracture. A
prospective study.
J. Bone Miner. Res. 1992, 7: S121 (Abstract no. 166).
Negative correlation between the change in bone
mineral density and serum osteocalcin in patients
with hyperthyroidism.
7. Rosen C.J., Adler R.A.
J. Clin. Endocrinol. Metab. 1990, 70: 766-770.
Longitudinal changes in lumbar bone density in thy-
rotoxic patients after attainement of euthyroidism.
J. Clin. Endocrinol. Metab. 1992, 75: 1531-1534.
19. Mc.Leod J.M., Mc.Hardy K.C., Harvey R.D., Duncan
A., Reid I.W., Bewsher P.D., Robin S.P.
The early effects of radioiodine therapy for hyper-
thyroidism on biochemical indices of bone turnover.
Clin. Endocrinol. (Oxf.) 1993, 38: 49-53.
8. Lalau J.D., Sebert J.L., Marie A., Fournier A.,
Quichaud J.
Effect of thyreotoxicosis and its treatment on miner-
al and bone metabolism.
J. Endocrinol. Invest. 1996, 9: 491-496.
20. Rizzoli R., Poser J., Bargi U.
Nuclear thyroid hormone receptors in cultured bone
cells.
9. Siddiqi A., Burrin J.M., Nooan K., James I., Wood
D.F., Price C.P.
Metabolism 1986, 35: 71-74.
A longitudinal study of markers of bone turnover in
Graves’ disease and their value in predicting bone
mineral density.
21. Harvey R.D., Mc.Hardy K. C., Reid I.W.,Paterson F.,
Beusher P.D., Duncan A., Robins S.P.
Measurement of bone collagen degradation in hyper-
thyroidism and during thyroxine replacement therapy
using pyridinium cross-links as specific urinary markers.
J. Clin. Endocrinol. Metab. 1991, 72: 1189-1194.
J. Clin. Endocrinol. Metab. 1997, 82: 753-759.
10. Oikawa M., Kushida K., Takahashi M., Ohishi T.,
Hoshino H., Suzuki M., Ogihara H., Ishigaki J., Inoue T.
Bone turnover and cortical bone mineral density in the
distal radius in patients with hyperthyroidism being trea-
ted with antithyroid drugs for various periods of time.
Clin. Endocrinol. 1999, 50: 171-176.
22. Garnero P., Vassy B., Bertholin A., Riou J.P., Delmas P.D.
Markers of bone turnover in hyperthyroidism and the
effects of treatment.
J. Clin. Endocrinol. Metab. 1994, 78: 955-959.
11. Mundy G.R., Shapiro J.L., Baudelin J.C., Canalis E.
M., Raisz L.G.
23. Mosekilde L., Melsen F.
Effect of antithyroid treatment on calcium-phospho-
rus metabolism in hyperthyroidism. II: Bone histo-
morphometry.
Direct stimulation of bone resorption by thyroid
hormones.
J. Clin. Invest. 1976, 58: 529-534.
Acta Endocrinol. 1978, 87: 751-758.
12. Marcus R.
24. Minisola S., Pacitti M.T., Romagnoli E., Rosso R.,
Carnevale V., Caravella P., Scillitani A., Dicembrino F.
Clinical validation of a new immunoradiometric as-
say for intact human osteocalcin.
Cyclic nucleotide phosphodiesterase from bone:
characterization of the enzyme and studies of inhibi-
tion by thyroid hormones.
Endocrinology 1975, 96: 400-408.
Calcif. Tissue Int. 1999, 64: 365-469.
13. Mundy G.R., Luben R.A., Raisz L.G., Oppenheim J.J.,
Buell D.M.
25. Hannon R., Blumsohn A., Naylor K., Eastell R.
Response of biochemical markers of bone turnover
to hormone replacement therapy: impact of bio-
logical variability.
Bone resorbing activity in supernatants of lymphoid
cell lines.
N. Engl. J. Med. 1974, 290: 867-871.
J. Bone Miner. Res. 1998, 7: 1124-1132.
14. Gowen M., Wood D.D., Ihrie E.F., McGuire M.K.B.,
Russel G.G.
26. Jodar E., Munoz-Torres M., Escobar-Jimenez F.,
Quesada M., Luna J., Olea N.
An interleukin-like factor stimulates bone resorption
in vitro.
Nature 1983, 306: 378-380.
Antiresorptive therapy in hyperthyroid patients: lon-
gitudinal changes in bone and mineral metabolism.
J. Clin. Endocrinol. Metab. 1997, 82: 1989-1994.
731