- Email: [email protected]
Axial bone mineral density in patients with acromegaly
Axial Bone Mineral Density in Patients with Acromegaly Marek Bolanowski, MD, PhD, Wojciech Wielgus, MD, Andrzej Milewicz, MD, PhD, Roman Marciniak, MD, PhD
Rationale and Objectives. The purpose of this study was to assess the influence of long-term elevated levels of growth hormone and insulin-like growth factor I on bone mineral density (BMD) of the spine in patients with acromegaly. Materials and Methods. The spinal BMD in 40 patients with acromegaly was measured with quantitative computed tomography. The result was expressed as the mean of six measurements of consecutive vertebral bodies and as a z score. Twenty age- and sex-matched healthy persons also underwent spinal BMD measurement and served as controls. Results. In 36 of the 40 patients, the BMD z score ranged within 2 standard deviations (SDs) above or below the mean. Only two patients (one man and one woman) had enhanced BMD loss (below 2 SD). Two other female patients had BMD values of more than 2 SD above the mean. Differences between patient subgroups were not statistically significant, but a statistically significant (P < .05) positive correlation was found between basal serum concentration of growth hormone and spinal BMD for the entire patient group. No correlation between BMD and basal serum concentration of insulin-like growth factor I was found.
Conclusion. Hypersecretion of growth hormone in patients with acromegaly, regardless of other factors, has no evident effect on BMD of the lumbar spine. Key Words. Acromegaly; osteoporosis; densitometry.
Acromegaly is characterized by enhanced skeletal growth and soft-tissue enlargement, usually caused by a pituitary tumor that produces growth hormone in excess. Despite a strong action by growth hormone as an antiresorptive factor, secondary osteoporosis is a feature of acromegaly. The cause of this osteoporosis, however, is not clear. The stimulatory effect of growth hormone on bone formation may be offset by the even larger increase in bone resorption. Insulin-like growth factor I (somatomedin C) is thought to be a main mediator of hormonal influences on bone cells, and it has a crucial role in the maintenance of normal bone mass (1-4). The purpose of this study was to assess the influence
Acad Radio12000; 7:592-594
1Fromthe Departmentsof Endocrinology(M.B., A.M.) and Radiology (W.W., RM.), MedicalUniversityof Wroclaw, Sklodowskiej-Curie68, 50369 Wroclaw, Poland.ReceivedApril 21, 1997; revisionrequestedJune 6; final revisionreceivedFebruary21,2000; acceptedMarch 9. Address correspondence to R.M.
©AUR, 2000
592
of long-term elevated levels of growth hormone and insulin-like growth factor I on bone mineral density (BMD) of the spine in patients with acromegaly.
Patients
This study was performed with 40 patients (20 men and 20 women) with acromegaly at a different stages of the disease. Patients ranged between 25 and 67 years of age (mean, 48 years), and all patients underwent quantitative computed tomography (CT) to measure spinal BMD. Patient data are presented in Table 1. The diagnosis of acromegaly was established on the basis of typical clinical features, an elevated basal serum growth hormone concentration that was not suppressed by oral glucose load, an elevated basal serum insulin-like growth factor I concentration, and the presence of pituitary adenoma on CT scans or magnetic resonance images. Fourteen patients were examined before receiving any therapy, 17 after successful surgical treatment, three after
Table 1 Findings-in Study Patients with Acromegaly
Patients (n) Mean age (y) Mean duration of disease (y) Basal growth hormone concentration (mlU/mL)* Basal insulin-like growth factor I concentration (ng/mL)*
Total
Men
Women
40 48 10 65 _+38
20 55 12 75 _+40
20 41 6 68 + 39
878 + 205
925 +_241
836 + 189
*Data presented as mean + standard deviation.
surgery followed by radiation therapy, and six after surgery followed by oral or intramuscular administration of bromocriptine. Nine patients had hyperprolactinemia, and seven had hypogonadism. Six patients were postmenopausal women. Twenty age- and sex-matched, healthy subjects (10 men and 10 women) also underwent spinal BMD measurement with the same method and served as controls.
Quantitative CT of the Lumbar Spine S p i n a l B M D was expressed as m i l l i g r a m s o f c a l c i u m h y -
droxyapatite (Ca-HA) per milliliter within the vertebral bodies T12 to L5. Measurements were obtained with quantitative CT by using a Somatom HiQ S (Siemens, Erlangen, Germany) with the Osteo Programme. The final value of the trabecular BMD in an individual patient (mean value calculated from the analysis of six consecutive vertebral bodies) was expressed both as tile concentration of Ca-HA (in milligrams per milliliter) and as the number of standard deviations (SDs) below or above the mean normal population value (age- and sex-related z score) as established by the manufacturer. The Siemens reference phantom (Ca-HA, 0 and 200 mg/mL) was used, and scans were obtained with a protocol of 80 kV, a section thickness of 10 ram, and an x-ray dose of approximately 30 gSv (below the natural dose rate per month).
Statistical Analysis The mean values and SD were calculated and compared between the patient subgroups studied with the Student t test. Correlation coefficients were estimated according to the Pearson method.
The mean values of the T12 through L5 trabecular BMD measurements in the individual patients are shown in Figure 1 for men and in Figure 2 for women. The mean z score
values _+SD of the spinal BMD studies, expressed as the number of SD from the mean of the normal population (for all patients and for men and women separately), are shown in Table 2. The differences among patient subgroups were not statistically significant (Table 2). There was, however, a statistically significant and positive correlation between basal serum growth hormone concentration and spinal BMD in the whole group (r = 0.63, P < .05). No correlation between spinal BMD and basal serum insulin-like growth factor I concentration was found. In 36 of the 40 patients, the BMD z score ranged from 2 SD above to 2 SD below the mean. Only two patients (one man and one woman) had enhanced BMD loss (below 2 SD). Two other female patients had BMD more than 2 SD above the mean. )ISCUSSIOI~
We have analyzed the measurements of trabecular BMD in the lumbar spines of 40 patients with acromegaly and compared those measurements with others made in control subjects. The results of this study are interesting, because one might expect to find a clear effect of acromegaly on bone density. There have been suggestions that osteoporosis in patients with acromegaly results from concomitant hyperprolactinemia and hypogonadism rather than from excess production of growth hormone alone (5), but we found no correlation between low BMD and hyperprolactinemia or hypogonadism in the study patients. The small number of patients in our study with either hyperprolactinemia or hypogonadism probably caused these unexpected results. We could not confmn the observation of Diamond et al (5) that vertebral BMD in patients with active acromegaly, as assessed with dual photon absorptiometry, was decreased compared with that in healthy subjects, regardless of their gonadal status. Other studies have revealed normal spinal BMD as measured with dual x-ray absorptiometry
593
250
250
200
.J 200
o
"'"'""'"--..
o o
-:7, 1so .
•
o .q~.•
.................
~E
CJE
"''--...@...
0
•
~ ............
o ÷t so
""~" ~O . . . . . . . . . . ~ •
50
o
.~ & 15o .¢ o
-" ......
o
~
e'""e-#.
"'"'"-..9 -.~• "'--~-~
•o
10o
mean
•
•
•
.
"~ .....
SD
-I
0
o
......... +1 SD
e~"'~k~
mean
......................
-I SD
0
20
3~
20
4~
5~
6~
7~
Age (y)
Age (y)
Figure 2.
M e a n values of spinal trabecular B M D in 20 female patients with a c r o m e g a l y and 10 healthy controls. • = patients, o = controls.
Figure 1. M e a n values of spinal trabecular B M D in 20 male patients with a c r o m e g a l y and 10 healthy controls. • = patients, o = controls.
Table 2 Spinal BMD in Patients with Acromegaly and in Healthy Control Subjects
A c r o m e g a l y (n = 40) A c r o m e g a l y and hyperprolactinemia (n = 9) A c r o m e g a l y and h y p o g o n a d i s m (n = 7) A c r o m e g a l y and m e n o p a u s e (n = 6) Controls (n = 20)
Total
Men
Women
+0.14 + 1.2
+0.18 + 1.4
+0.06 _+ 1.1
+0.1 + 1.4
+0.6 + 1.6
- 0 . 5 + 1.1
- 0 . 7 3 + 1.2
- 2 . 0 +_ 0.4
- 0 . 2 + 1.0
+0.26 + 1.6
...
+0.26 + 1.6
+0.12 + 1.1
+0.15 + 1.2
+0.09 + 1.3
N o t e . - - D i f f e r e n c e s between patients and controls are not statistically significant. All values are mean (z score) + SD. Values in parentheses are the n u m b e r of patients in each group.
within the lumbar spine and either elevated measurements in the proximal femur or normal measurements in both sites (6,7). The latter study (7) did not reveal any difference between eugonadal and hypogonadal patients or any short-term changes in BMD after successful treatment. On the contrary, Ezzat et al (8) have shown a reduction in vertebral trabecular bone mass as measured with quantitative CT in hypogonadal patients with acromegaly. In conclusion, hypersecretion of growth hormone in patients with acromegaly, regardless of other factors, has no evident effect on BMD of the lumbar spine. 3EFERENCE,c Albright F, Reifenstein EC. The parathyroid glands and metabolic bone disease: selected studies. Baltimore, Md: Williams & Wilkins, 1948; 188196.
594
2. Nabarro JDN. Acromegaly. Clin Endocrino11987; 26:481-512. 3. Johansson AG, Lindh E, Blum WF, Koilerup G, Sorenson OH, Ljunghall S. Effects of growth hormone and insulin-like growth factor 1 in men with idiopathic osteoporosis. J Clin Endocrinol Metab 1996; 81 : 44-48. 4. Nishiyama K, Sugimoto T, Kaji H, Kanatani M, Kobayashi T, Chihara K. Stimulatory effect of growth hormone on bone resorption and osteoclast differentiation. Endocrinology 1996; 137:35-41. 5. Diamond T, Nery L, Posen S. Spinal and peripheral bone mineral densities in acromegaly: the effects of excess growth hormone and hypogonadism. Ann Intern Med 1989; 111:567-573. 6. Kotzmann H, Bernecker P, HLibsch P, et al. Bone mineral density and parameters of bone metabolism in patients with acromegaly. J Bone Miner Res 1993; 8:459-465. 7. Ho PJ, Barkan AL, Shapiro B. Bone mineral density of the axial skeleton in acromegaly. J Nucl Med 1992; 33:1608-1612. 8. Ezzat S, Melmed S, Endres D, Eyre DR, Singer FR. Biochemical assessment of bone formation and resorption in acromegaly. J Clin Endocrinol Metab 1993; 76:1452-1457.
Rationale and Objectives. The purpose of this study was to assess the influence of long-term elevated levels of growth hormone and insulin-like growth factor I on bone mineral density (BMD) of the spine in patients with acromegaly. Materials and Methods. The spinal BMD in 40 patients with acromegaly was measured with quantitative computed tomography. The result was expressed as the mean of six measurements of consecutive vertebral bodies and as a z score. Twenty age- and sex-matched healthy persons also underwent spinal BMD measurement and served as controls. Results. In 36 of the 40 patients, the BMD z score ranged within 2 standard deviations (SDs) above or below the mean. Only two patients (one man and one woman) had enhanced BMD loss (below 2 SD). Two other female patients had BMD values of more than 2 SD above the mean. Differences between patient subgroups were not statistically significant, but a statistically significant (P < .05) positive correlation was found between basal serum concentration of growth hormone and spinal BMD for the entire patient group. No correlation between BMD and basal serum concentration of insulin-like growth factor I was found.
Conclusion. Hypersecretion of growth hormone in patients with acromegaly, regardless of other factors, has no evident effect on BMD of the lumbar spine. Key Words. Acromegaly; osteoporosis; densitometry.
Acromegaly is characterized by enhanced skeletal growth and soft-tissue enlargement, usually caused by a pituitary tumor that produces growth hormone in excess. Despite a strong action by growth hormone as an antiresorptive factor, secondary osteoporosis is a feature of acromegaly. The cause of this osteoporosis, however, is not clear. The stimulatory effect of growth hormone on bone formation may be offset by the even larger increase in bone resorption. Insulin-like growth factor I (somatomedin C) is thought to be a main mediator of hormonal influences on bone cells, and it has a crucial role in the maintenance of normal bone mass (1-4). The purpose of this study was to assess the influence
Acad Radio12000; 7:592-594
1Fromthe Departmentsof Endocrinology(M.B., A.M.) and Radiology (W.W., RM.), MedicalUniversityof Wroclaw, Sklodowskiej-Curie68, 50369 Wroclaw, Poland.ReceivedApril 21, 1997; revisionrequestedJune 6; final revisionreceivedFebruary21,2000; acceptedMarch 9. Address correspondence to R.M.
©AUR, 2000
592
of long-term elevated levels of growth hormone and insulin-like growth factor I on bone mineral density (BMD) of the spine in patients with acromegaly.
Patients
This study was performed with 40 patients (20 men and 20 women) with acromegaly at a different stages of the disease. Patients ranged between 25 and 67 years of age (mean, 48 years), and all patients underwent quantitative computed tomography (CT) to measure spinal BMD. Patient data are presented in Table 1. The diagnosis of acromegaly was established on the basis of typical clinical features, an elevated basal serum growth hormone concentration that was not suppressed by oral glucose load, an elevated basal serum insulin-like growth factor I concentration, and the presence of pituitary adenoma on CT scans or magnetic resonance images. Fourteen patients were examined before receiving any therapy, 17 after successful surgical treatment, three after
Table 1 Findings-in Study Patients with Acromegaly
Patients (n) Mean age (y) Mean duration of disease (y) Basal growth hormone concentration (mlU/mL)* Basal insulin-like growth factor I concentration (ng/mL)*
Total
Men
Women
40 48 10 65 _+38
20 55 12 75 _+40
20 41 6 68 + 39
878 + 205
925 +_241
836 + 189
*Data presented as mean + standard deviation.
surgery followed by radiation therapy, and six after surgery followed by oral or intramuscular administration of bromocriptine. Nine patients had hyperprolactinemia, and seven had hypogonadism. Six patients were postmenopausal women. Twenty age- and sex-matched, healthy subjects (10 men and 10 women) also underwent spinal BMD measurement with the same method and served as controls.
Quantitative CT of the Lumbar Spine S p i n a l B M D was expressed as m i l l i g r a m s o f c a l c i u m h y -
droxyapatite (Ca-HA) per milliliter within the vertebral bodies T12 to L5. Measurements were obtained with quantitative CT by using a Somatom HiQ S (Siemens, Erlangen, Germany) with the Osteo Programme. The final value of the trabecular BMD in an individual patient (mean value calculated from the analysis of six consecutive vertebral bodies) was expressed both as tile concentration of Ca-HA (in milligrams per milliliter) and as the number of standard deviations (SDs) below or above the mean normal population value (age- and sex-related z score) as established by the manufacturer. The Siemens reference phantom (Ca-HA, 0 and 200 mg/mL) was used, and scans were obtained with a protocol of 80 kV, a section thickness of 10 ram, and an x-ray dose of approximately 30 gSv (below the natural dose rate per month).
Statistical Analysis The mean values and SD were calculated and compared between the patient subgroups studied with the Student t test. Correlation coefficients were estimated according to the Pearson method.
The mean values of the T12 through L5 trabecular BMD measurements in the individual patients are shown in Figure 1 for men and in Figure 2 for women. The mean z score
values _+SD of the spinal BMD studies, expressed as the number of SD from the mean of the normal population (for all patients and for men and women separately), are shown in Table 2. The differences among patient subgroups were not statistically significant (Table 2). There was, however, a statistically significant and positive correlation between basal serum growth hormone concentration and spinal BMD in the whole group (r = 0.63, P < .05). No correlation between spinal BMD and basal serum insulin-like growth factor I concentration was found. In 36 of the 40 patients, the BMD z score ranged from 2 SD above to 2 SD below the mean. Only two patients (one man and one woman) had enhanced BMD loss (below 2 SD). Two other female patients had BMD more than 2 SD above the mean. )ISCUSSIOI~
We have analyzed the measurements of trabecular BMD in the lumbar spines of 40 patients with acromegaly and compared those measurements with others made in control subjects. The results of this study are interesting, because one might expect to find a clear effect of acromegaly on bone density. There have been suggestions that osteoporosis in patients with acromegaly results from concomitant hyperprolactinemia and hypogonadism rather than from excess production of growth hormone alone (5), but we found no correlation between low BMD and hyperprolactinemia or hypogonadism in the study patients. The small number of patients in our study with either hyperprolactinemia or hypogonadism probably caused these unexpected results. We could not confmn the observation of Diamond et al (5) that vertebral BMD in patients with active acromegaly, as assessed with dual photon absorptiometry, was decreased compared with that in healthy subjects, regardless of their gonadal status. Other studies have revealed normal spinal BMD as measured with dual x-ray absorptiometry
593
250
250
200
.J 200
o
"'"'""'"--..
o o
-:7, 1so .
•
o .q~.•
.................
~E
CJE
"''--...@...
0
•
~ ............
o ÷t so
""~" ~O . . . . . . . . . . ~ •
50
o
.~ & 15o .¢ o
-" ......
o
~
e'""e-#.
"'"'"-..9 -.~• "'--~-~
•o
10o
mean
•
•
•
.
"~ .....
SD
-I
0
o
......... +1 SD
e~"'~k~
mean
......................
-I SD
0
20
3~
20
4~
5~
6~
7~
Age (y)
Age (y)
Figure 2.
M e a n values of spinal trabecular B M D in 20 female patients with a c r o m e g a l y and 10 healthy controls. • = patients, o = controls.
Figure 1. M e a n values of spinal trabecular B M D in 20 male patients with a c r o m e g a l y and 10 healthy controls. • = patients, o = controls.
Table 2 Spinal BMD in Patients with Acromegaly and in Healthy Control Subjects
A c r o m e g a l y (n = 40) A c r o m e g a l y and hyperprolactinemia (n = 9) A c r o m e g a l y and h y p o g o n a d i s m (n = 7) A c r o m e g a l y and m e n o p a u s e (n = 6) Controls (n = 20)
Total
Men
Women
+0.14 + 1.2
+0.18 + 1.4
+0.06 _+ 1.1
+0.1 + 1.4
+0.6 + 1.6
- 0 . 5 + 1.1
- 0 . 7 3 + 1.2
- 2 . 0 +_ 0.4
- 0 . 2 + 1.0
+0.26 + 1.6
...
+0.26 + 1.6
+0.12 + 1.1
+0.15 + 1.2
+0.09 + 1.3
N o t e . - - D i f f e r e n c e s between patients and controls are not statistically significant. All values are mean (z score) + SD. Values in parentheses are the n u m b e r of patients in each group.
within the lumbar spine and either elevated measurements in the proximal femur or normal measurements in both sites (6,7). The latter study (7) did not reveal any difference between eugonadal and hypogonadal patients or any short-term changes in BMD after successful treatment. On the contrary, Ezzat et al (8) have shown a reduction in vertebral trabecular bone mass as measured with quantitative CT in hypogonadal patients with acromegaly. In conclusion, hypersecretion of growth hormone in patients with acromegaly, regardless of other factors, has no evident effect on BMD of the lumbar spine. 3EFERENCE,c Albright F, Reifenstein EC. The parathyroid glands and metabolic bone disease: selected studies. Baltimore, Md: Williams & Wilkins, 1948; 188196.
594
2. Nabarro JDN. Acromegaly. Clin Endocrino11987; 26:481-512. 3. Johansson AG, Lindh E, Blum WF, Koilerup G, Sorenson OH, Ljunghall S. Effects of growth hormone and insulin-like growth factor 1 in men with idiopathic osteoporosis. J Clin Endocrinol Metab 1996; 81 : 44-48. 4. Nishiyama K, Sugimoto T, Kaji H, Kanatani M, Kobayashi T, Chihara K. Stimulatory effect of growth hormone on bone resorption and osteoclast differentiation. Endocrinology 1996; 137:35-41. 5. Diamond T, Nery L, Posen S. Spinal and peripheral bone mineral densities in acromegaly: the effects of excess growth hormone and hypogonadism. Ann Intern Med 1989; 111:567-573. 6. Kotzmann H, Bernecker P, HLibsch P, et al. Bone mineral density and parameters of bone metabolism in patients with acromegaly. J Bone Miner Res 1993; 8:459-465. 7. Ho PJ, Barkan AL, Shapiro B. Bone mineral density of the axial skeleton in acromegaly. J Nucl Med 1992; 33:1608-1612. 8. Ezzat S, Melmed S, Endres D, Eyre DR, Singer FR. Biochemical assessment of bone formation and resorption in acromegaly. J Clin Endocrinol Metab 1993; 76:1452-1457.