ORIGINAL ARTICLE 
Annals of Nuclear Medicine Vol. 12, No. 1, 15-20 1998 
Relationship between skeletal uptake of 99mTc-HMDP and bone mineral density in elderly women 
Yusuke KIGAMI, Itsuo YAMAMOTO, Hideo OHNISHI, Masahiko TAKADA, Ryouji MATSUSHITA, Yasuyo HAMANAKA, Toyotsugu OTA and Rikushi MORITA 
Department of Radiology and Division of Nuclear Medicine, Shiga University of Medical Science 
The relationship between bone mineral density in elderly women and the pattern of skeletal uptake of 99mTc-HMDP, especially in regard to skull uptake, was investigated. The whole-body skeletal uptake (WBSU) and whole-body skeletal tracer distribution patterns were studied in 86 disease-free women on bone scintigraphy with 99mTc-hydroxy-methylene-diphosphonate (HMDP). Bone scans were quantified by setting regions of interest (ROI) and bone mineral density (BMD) was assessed by dual-energy X-ray absorptiometry in all patients. WBSU and the skeletal distribution pattern were compared with bone mineral densities of the entire skeleton as well as selected regions. WBSU was high in the elderly and negatively correlated with regional bone mineral densities (r = -0.403 to -0.534). Among the regions, uptake by the skull increased with age more than in other regions in women and had the highest negative correlation with the bone mineral density. The skull uptake correlated negatively with total body BMD (r = -0.583) and with lumbar BMD (r = -0.561 , p < 0.0001). Our results show that increased radionuclide uptake in bone scintigraphy, especially skull uptake was associated with decreased bone mineral density in elderly women, so that, increased skull uptake in elderly women would be a scintigraphic sign of post-menopausal or senile osteopenia. 
Key words: bone scintigraphy, skull, bone mineral density, osteoporosis, menopause 

INTRODUCTION 
DIFFUSELY ENHANCED radionuclide accumulation in the skull on bone scintigraphy, which is sometimes termed "Hot Skull," is often observed in elderly women without any apparent skeletal diseases or symptoms related to bone (Fig. 1). This was originally found in patients with breast cancer, who underwent cytotoxic chemotherapy, and was attributed to the chemotherapy,1,2 but in routine clinical examination, diffuse increase in radionuclide uptake in the skull is seen in women without chemotherapy. Our previous study which quantified skeletal radionuclide uptake demonstrated a substantial increase 
Received August 7, 1997, revision accepted October 2, 1997. For reprint contact: Itsuo Yamamoto, M.D., Department of Radiology and Division of Nuclear Medicine, Shiga University of Medical Science, Seta-Tsukinowa-cho, Otsu, Shiga 520-2 192, JAPAN. E-mail : iyama@belle.shiga-med.ac.jp 
in skull uptake after 50 y.o. in women, suggesting a relationship with the menopause.3 Increased skull uptake in bone scintigraphy is common in some metabolic bone diseases such as hyperparathyroidism, osteomalacia and hyperthyroidism4-7 Although details of the mechanism of this increased skull uptake remain unclear, the skull bone, as non-weight bearing bone, might be sensitively affected by systemic metabolic change. In this study we studied apparently normal elderly women who were examined with routine bone scintigraphy, by semi-quantifying the image, applied bone densitometry to correlate it with skeletal uptake in bone scan and discussed the significance of skeletal uptake in bone scintigraphy. 

MATERIALS AND METHODS 
Subjects: Patients who underwent bone scintigraphy, but had no focal abnormal uptakes in two consecutive years without any medication, and patients with no skeletal symptoms at their first pretreatment bone survey were selected and asked to undergo measurement of bone densitometry. One hundred and thirty-seven patients agreed to enter this study, but after measurement of bone density, twenty-one patients were considered to be inappropriate because of the presence of osteoarthritis, scoliosis, prosthesis or other artifactual causes. Another thirty patients were also excluded due to factors which could possibly influence the results, such as the administration of drugs and the presence of other chronic diseases or the presence of abnormal focal uptakes in their bone scintigram from non-malignant diseases. Finally, eighty-six women (age range 28-75, mean: 51.5 +- 11.6 (mean +- SD), weight 52.4 +- 8.6 kg (mean +- SD), height 154.0 +- 5.4 cm (mean +- SD)) were included in this study and their data were further analyzed. Of these, seventy patients had breast cancer, 14 had other malignant diseases, and 2 had nonmalignant diseases. None received estrogen or other drugs which could possibly influence bone metabolism, such as corticosteroids, progesterone, calcitonin or methotrexate. Approximately two-thirds of the patients were in a disease-free state after therapy and one-third were undergoing initial examination before therapy. Patients who had any metabolic bone diseases such as primary hyperparathyroidism, hyperthyroidism, renal osteodystrophy, osteomalacia, multiple myeloma or other skeletal disorders possibly influencing the results were excluded. All were outpatients, and severely ill patients were also excluded. Bone scintigraphy: On all subjects bone scintigraphy was performed semi-quantitatively to find regional skeletal uptakes. 99mTc-hydroxy-methylene-diphosphonate (HMDP; CLEAR BONE, Mediphysics Co. Ltd., Osaka, Japan) was used for bone scintigraphy in all patients. The labeling procedure for the radiopharmaceutical was consistent and the labeling efficiency was checked to assure constant high quality. 99mTc-HMDP was counted in a Curimeter (Capintec Japan, Tokyo, Japan) and 555 MBq of the radiopharmaceutical was injected intravenously with repeated drawing and injecting to avoid leaving the remnant of the tracer. Four hours later, after complete urination, 10 minute whole body bone scintigrams, with both anterior and posterior projections were obtained by means of a dual whole-body scinticamera with a large field of view (51 x 38 cm; GCA 901A/W2, Toshiba Medical Co., Ltd., Tokyo, Japan) equipped with a high resolution collimator. The protocol for scanning was kept constant, throughout the study: patient positioning, scan speeds (20 cm/min) and other parameters were kept constant. The distance from the nearest body surface to the scinticameras was kept at 5 cm. Data were recorded on an optical memory disk (OD-01A, Toshiba Medical Co., Ltd., Tokyo, Japan). Quantification of bone scintigrams: Skeletal uptakes of 99mTc-HMDP were analyzed with a data processor (GMS-550U. Toshiba Medical Co., Ltd., Tokyo, Japan) as described previously.3 In both anterior and posterior views of a whole body scintigram, regions of interest (ROI) were set over selected bone regions as shown in Fig. 2, excluding soft tissue and the urinary tract. Patients with abnormally high uptakes from various skeletal disorders such as arthritis, fracture, sinusitis and/or dental diseases were pre-excluded from the analyses. A simple mean of radioactivities (counts) from both views was calculated after adjusting for the radionuclide half-life decay, and defined as the whole-body skeletal uptake (WBSU). The soft tissue uptake was also monitored by subtracting the activities of the urinary tract (kidney and bladder) and WBSU from the total body uptake. Soft tissue activity was less than one-tenth of WBSU and had a small effect on the WBSU. Bone uptake in the head, thoracic region and lower legs was expressed as the ratio of the uptake in each region to WBSU. Reproducibility of the procedure for quantitation of bone scan was checked by comparing analyzed results in two distinct examinations which were performed within a year in 10 patients and showed 3.5-10.2% of C.V. in regional skeletal uptakes and 5.5% of that in WBSU. Bone mineral densitometry: Bone mineral density was assessed by dual-energy X-ray absorptiometry (DXA) with DPX-L (Lunar Co. Ltd., Madison, WI, USA) and was performed within four weeks of bone scintigraphy by a trained technician using constant scanning protocols. Calibration was performed daily prior to measurement for patients. Total body bone mineral density (BMD), anterior-posterior lumbar BMD (integral L2 to L4 BMD; L2-4 BMD) and femoral neck BMD were assessed according to the standard protocols: lumbar BMD was assessed in a position which reduces lumbar lordosis and femoral neck BMD was determined at 30 degree inner rotation of the right leg, placing water-equivalent material at the outside of the lateral thigh. Head BMD was calculated from the total body measurement by setting the region of interest on the head, which was defined as the region over the line between both olecranon processes (Fig. 2), and lumbar BMD and femoral neck BMD were each measured with protocols for regional measurement. Average in vivo precision values for these measurements are 1.2% for lumbar BMD, 2.0% for femoral neck BMD, 1.1 % for total body BMD and 1.8% for the head BMD (determined by five separate measurements in three healthy volunteers). Correlation among skeletal radionuclide uptakes and regional BMDS was studied. DataAnalysis.' The association between BMDS and radionuclide uptake was examined by simple regression with STAT VIEW SE (Abacus Concepts Inc., Berkeley, CA). Differences between two groups were tested with unpaired t-test (Welch's t-test). A p value less than 0.05 is considered to be significant. 

RESULTS 
Whole body skeletal uptake (WBSU) and absolute counts of head uptake as a function with age are shown in Fig. 3. Consistent with the previous study,3 WBSU and head uptake increased from the fifth decade to the seventh decade. Furthermore, relative uptake by the head (a ratio of the head uptake to WBSU) increased with age until the seventh decade (significantly different in the seventh decade from those in the fourth and fifth decades). WBSU was negatively correlated with regional bone mineral densities (r = -0.471 to -0.561, p < 0.0005) (Table 1). Lumbar BMD showed the highest negative correlation (r = -0.561 , p < 0.0001 ), followed by femoral neck BMD (r = -0.534, p < 0.0001 ). Detailed insight of the ROI data revealed that uptake by the head had the highest negative correlation with bone mineral density. The head uptake correlated negatively with total body BMD (r = -0.583, p < 0.0001) and with lumbar BMD (r = -0.561 , p < 0.0001 ). The radionuclide uptake by the thorax and legs showed a lower correlation with bone mineral densities (r = -0.327 to -0.510). When regional uptakes were expressed as ratios relative to WBSU, only the head uptake showed a significant correlation with BMDS (Table 1, Fig. 4). To validate the effects of age, the correlation between radionuclide uptake and bone mineral density was examined in each decade group. Although the numbers in each group were relatively small, a tendency to a negative correlation in each age group was observed (Fig. 4). 

DISCUSSION 
In this study, we have confirmed the previous results that showed an age-related increase of skeletal uptake in 99mTc-HMDP scintigraphy in women3,8,9 and for the first time showed that the regional skeletal uptake correlates negatively with bone mineral density assessed by DXA. Higher radionuclide accumulation showed lower bone mineral density. Among regional bone densities, the correlation with the radionuclide uptake was high in total body BMD, lumbar BMD and femoral neck BMD. Among the regional uptakes in bone scans, uptake by the head showed the highest correlation with BMDs. Our method estimating WBSU was a modification of a method originally shown by D'Addabbo et al. which makes it possible to calculate WBSU without waiting for 24 hours or harvesting the urine.10 At 4 hours, radioactivity still remains in the soft-tissue, but by excluding the major source of extra-skeletal activity, namely the activity from the urinary tract, WBSU can be obtained and the influences of remaining activity in the soft tissue is small.3,10 There can be individual variation in tissue gamma-ray attenuation due to differences in body size, but in our present study differences in body size were relatively small (coefficient variations in body weight and height were 16.4% and 3.5%, respectively) and thus errors should be small. Setting of the ROI was done in as uniform a way as possible and variation in the calculation of regional uptake was also small. An increase in skeletal uptake with aging in bone scintigraphy was seen in some previous studies; whole-body radionuclide retention measured at 24 hours after the injection was higher in elderly people.8.9 This was attributed either to a decrease in renal function with aging9 or to enhanced bone metabolism in the aged.8 The latter study also demonstrated enhanced bone loss in those who had high whole-body skeletal retention of the radionuclide. Both explanations for the increased skeletal uptake in elderly subjects may be true: radionuclide uptake by bone in bone scintigraphy is influenced by renal function as well as bone metabolic turnover rate.7 A decrease in renal function might partly contribute to increased WBSU in elderly women in our study, but some part could be explained by increased skeletal turnover in the elderly. Our previous study, which showed a correlation of WBSU with levels of bone metabolic markers,3 supports this. There is accumulating evidence which shows increased bone turnover in elderly women.11-14 Increased skeletal turnover causes more bone loss in the elderly,15,16 in whom bone formation cannot completely compensate for bone resorption, so that lower BMD in those with higher WBSU could be attributed to enhanced skeletal turnover in the elderly. Increased skull uptake in normal elderly women is a 1,2,17,18 In conwell-known finding in bone scintigraphy. trast to WBSU, increased skull uptake is not attributable to decreased renal function, since it is unusual for decreased radionuclide clearance to affect only skull uptake. Increased skull uptake is age-dependent and cannot be attributable to any artifacts in obtaining a bone scan, such as disproportionate tissue attenuation of radioactivity. In addition, increased skull uptake is rarely found in elderly men.3,17 This was originally discussed in relation to chemotherapy in patients with breast cancer 1,2 and once in connection with hyperostosis cranii in younger women,18 but in one study,17 increased skull uptake was related to the post-menopausal bone metabolic change. Recent studies show that postmenopausal bone loss is exaggerated in those with increased bone turnover11 and bone mass is lower in those with high bone metabolic markers.19 The present findings combined with the previous results indicate the increased skull uptake would be a sign of bone loss in post-menopausal or elderly women. Increased skull uptake is a common finding in patients with high-turnover bone diseases such as hyperpara-thyroidism and osteomalacia.4,5,6 The reason why skull uptake increases in such diseases is unclear, but the skull, being a non weight-bearing bone, might be early affected by a systemic mineral metabolism change and could show the first and prominent sign of general bone metabolic change as seen in "salt and pepper appearance" in the skull X-ray photograph in hyperparathyroidism.20 It is therefore probable that the increased skull uptake in elderly women is attributable to enhanced systemic bone metabolism. To ultimately determine the meaning of increased skull uptake in women, a prospective study to assess the rate of bone loss would be necessary.21 
In summary, we have shown that increased radionuclide uptake in bone scintigraphy is associated with decreased bone mineral density in elderly women, especially increased skull uptake associated with decreased lumbar BMD, so that increased skull uptake in elderly women would be a sign of post-menopausal or senile osteopenia. 

REFERENCES 
1. Creutzig H, Dach W. The "sickle-sign" in bone scintigraphy. Eur J Nucl Med 6: 99-101 , 1981. 
2. Roos JC, van IJ, van BM. Oei HY, van RP. The hot skull: malignant or feminine? Eur J Nucl Med 13: 207-209, 1987. 
3. Kigami Y, Yamamoto I. Ohnishi H, Miura H. Ohnaka Y, Ota T, et al. Age-related change of technetium-99m-HMDP distribution in the skeleton. J Nucl Med 37: 815-818, 1996. 
4. Sy WM. Bone scan in primary hyperparathyroidism. J Nucl Med 15: 1089-1091 , 1974. 
5. Wiegmann T. Rosenthall L. Kaye M. Technetium-99m-pyrophosphate bone scans in hyperparathyroidism. J Nucl Med 18: 231-235, 1977. 
6. Fogelman I, Carr D. A comparison of bone scanning and radiology in the evaluation of patients with metabolic bone disease. Clin Radiol 31 : 321-326, 1980. 
7. Fogelman I. Collier BD. Brown ML. Bone scintigraphy: Part 3. Bone scanning in metabolic bone disease. J Nucl Med 34: 2247-2252, 1993. 
8. Fogelman I. Bessent R. Age-related alteration in skeletal metabolism; 24 hour whole-body retention of diphosphonate in 250 normal subjects. J Nucl Med 23: 296-300, 1982. 
9. Martin P, Schoutens A, Manicourt D, Bergmann P. Fuss M, Verbanck M. Whole body and regional retention of 99mTc-labeled pyrophosphate at 24 hours: physiological basis of the method for assessing the metabolism of bone in disease. Calcif Tissue Int 35: 37-42, 1983. 
10. D'Addabbo A, Rubini G. Mele M. Lauriero F. A new method for assessing 99mTc-MDP bone uptake from a bone scan image: quantitative measurement of radioactivity in global skeletal regions of interest. Nucl Med Commun 13: 55-60, 1992. 
11. Christiansen C. Riis BJ. Rodbro P. Prediction of rapid bone loss in postmenopausal women. Lancet 1: 1105-1108, 1987. 
12. McLaren AM. Hordon LD, Bird HA. Robins SP. Urinary excretion of pyridinium crosslinks of collagen in patients with osteoporosis and the effects of bone fracture. Ann Rheum Dis 51 : 648-651 , 1992. 
13. Delmas PD. Biochemical markers of bone turnover for the clinical investigation of osteoporosis. Osteoporos Int 1 : 81-86, 1993. 
14. Kasai R. Okumura H, Yamamura T. Statistical analysis of age- and sex-related changes of serum osteocalcin. J Bone Miner Met 11 : 7-12, 1993. 
15. Uebelhart D, Schlemmer A, Johansen JS, Gineyts E, Christiansen C, Delmas PD. Effect of menopause and hormone replacement therapy on the urinary excretion of pyridinium cross-links. J Clin Endocrinol Metab 72: 367-373, 1991. 
16. Christiansen C. Riis BJ, Rodbro P. Screening procedure for women at risk of postmenopausal osteoporosis. Osteoporosis Int 1: 35-40, 1990. 
17. Senda K, Itoh S. Evaluation of diffusely high uptake by the calvaria in bone scintigraphy. Ann Nucl Med 1: 23-26, 1987. 
18. Jacobsson H. Haverling M. Hyperostosis cranii. Radiography and scintigraphy compared. Acta Radiol 29: 223-226, 1988. 
19. Miura H. Yamamoto I. Yuu I. Kigami Y. Ohta T. Yamamura Y, et al. Estimation of bone mineral density and bone loss by means of bone metabolic markers in postmenopausal women. Endocrine J 46: 797-802, 1995. 
20. Edeiken J, Hodes PJ. ln Roentgen Diagnosis of Diseases of Bone, Williams & Wilkins, Baltimore Md., p. 445, 1973. 
21 . Fogelman I, Bessent R, Cohen HN, Hart DM, Lindsay R. Skeletal uptake of diphosphonate. Method for prediction of post-menopausal osteoporosis. Lancet 2: 667-670, 1980.