Limb measurements for height and bone mineral density estimation.

To estimate height and forearm bone mineral density (BMD), three simple sites of limb segments of 428 healthy subjects (321 females and 107 males) were measured. Mean length of upper arm length (UAL), lower arm length (LAL) and knee to floor height (KFH) were not different in various age groups (p > 0.05). Good correlations were obtained from all measurements compared with height (r = 0.789, 0.826 and 0.810 for UAL, LAL and KFH, respectively) and moderate correlations were found compared with BMD (r = 0.386, 0.461 and 0.458 for UAL, LAL and KFH, respectively). If physically disabled activity occurred in some segmental limbs, the others which have good ability can be used for height assessment. When applying the stepwise method for multiple regression analysis, combination of LAL, KFH and UAL showed the best correlation (r = 0.866) with height. It indicated that UAL, LAL and KFH measurements provide enough predictive ability for height estimation. Only LAL and KFH were the variables which predicted BMD equation with moderately predictive ability. In primary health care, simple long bone measurements with the patient's history should be a primary step for screening of high risk osteoporotic patients. In prevention of osteoporotic fracture, LAL and KFH may be a useful adjunct to other criteria in the discrimination of high risk patients.

Arm span, height and forearm bone mineral density in normal young and postmenopausal women.

The value of arm span and height measurements in the erect position were evaluated by (1) comparing them in individual young adults and postmenopausal women and (2) comparing them with forearm bone mineral density (BMD) using a Panasonic dual energy X-ray absorptiometry (DXA). The study group consisted of 301 healthy women composed of 165 young adults (20-40 years), 62 postmenopausal women and 74 women whose age was more than 40 and had regularity of menstruation. Arm span and height of the individual young adults and postmenopausal women were not significantly different (p>0.05). It indicated that arm span and height measurements were independent of normal aging. Height loss occurred in some elderly women who had some diseases or conditions which induced short stature. We suggested that arm span can be used to predict height during the younger adult life of elderly women and to predict the current height in patients who had a height loss problem. A simple linear regression equation of height and arm span has already been constructed. During the increase of years since menopause in postmenopausal women, height and arm span measurements were relatively constant, while BMD gradually declined. We also found that the greater the number of years since menopause, the higher the rate of BMD loss (%). Height and arm span had low correlation with BMD. Height or arm span alone can not be used to predict BMD, but may be combined with other risk factors of osteoporosis.

Diagnostic agreement of combined radiogrammetric analysis with texture analysis in the evaluation of bone density: a comparison with dual energy X-ray absorptiometry.

To assess the reliability and validity of combined radiogrammetric analysis with texture analysis (digital X-ray radiogrammetry: DXR) of metacarpals and distal forearm when compared to dual energy X-ray absorptiometry (DXA) of the distal radius. DXA readings from the supradistal and distal 1/10 of radius were compared with DXR obtained on the same day with a sample of 274 women and 87 men (average age 31.33 years, range 20-65 years). The results showed that both sites of forearm DXA scores were significantly correlated with DXR scores (r = 0.5211 and 0.6090, respectively). DXA scores were used as the standard, defining moderate fracture risk by a standardized t score <-2 and marked risk by t <-3. Following the current WHO definition of osteoporosis, t <-2.5 was also applied. Using cutting points of t <-2, <-2.5 and <-3 for DXA and t <-2 for DXR values, the sensitivity to osteopenia was 38.98 per cent, 40.63 per cent and 55.56 per cent respectively, with specificity of 94.36 per cent for all cutting points of DXA. At these cutting points, the corresponding false positive was 5.64 per cent and false negatives were 61.02 per cent, 59.37 per cent and 44.44 per cent, respectivety. It indicated that DXR measurement had low sensitivity but was appropriate with high specificity for discrimination of forearm osteoporosis.

The accuracy of bone mineral density at distal radius on non-forearm osteoporosis identification.

The accuracy of BMD at distal radius was evaluated in terms of sensitivity, specificity, false negative, false positive, predictive value of a positive (osteoporosis) and a negative (normal) test for non-forearm osteoporosis. 278 women (150 osteoporotic and 128 normal) were measured for both distal radius bone mineral density (BMD) using Panasonic (DXA-70) dual energy X-ray absorptiometry (DEXA) and non-forearm BMD using Hologic (QDR-4500) DEXA on the same day. The results showed that mean age, menopause age, height and weight in the osteoporotic group were not different from the healthy group (p=0.168, 0.091, 0.274 and 0.097, respectively). Mean BMD of both distal radius and lumbar spine in the normal women was significantly higher than that in the osteoporotic group (p<0.001, = 0.002, <0.001, respectively). While mean BMD of the hip, femoral neck and Ward's triangle in both groups was not different (p = 0.330, 0.874, 0.847, respectively). The sensitivity of BMD of the right radius was very high (90.00-95.45%) and specificity was moderately high (53.85-73.68%). While false negative (4.55-10.00%) was less than false positive (26.32-46.15%). The accuracy of right radius BMD when compared with spine, hip, femoral neck and Ward's triangle was 82.35, 66.66, 80.00 and 86.49 per cent, respectively. The sensitivity (85.00-96.67%), specificity (57.69-81.58%), false negative (3.33-15.00%) and false positive (18.42-42.31%) of left radius BMD had the same trend as right radius BMD. Accuracy of the left radius when compared with non-forearm BMD was 88.24, 66.67, 75.71 and 86,49 per cent, respectively. The predictive value of right radius osteoporosis was 73.68, 47.37, 77.78 and 89.66 per cent for detecting osteoporosis at spine, hip, femoral neck and Ward's triangle, respectively. The predictive value of normal right radius BMD was 93.33, 92.86, 87.50 and 75.00 per cent, respectively for normal non-forearm BMD. Moreover, the predictive value of left radius osteoporosis for identifying spinal, hip, femoral neck and Ward's triangle osteoporosis was (80.56, 47.22, 77.55 and 91.07%, respectively) and the predictive value of normal left radius BMD for identifying normal BMD at non-forearm sites (96.88, 90.00, 71.43 and 72.22%, respectively) was revealed. It indicated that forearm DEXA provides adequate accuracy for in vivo determination of spinal, femoral neck and Ward's triangle osteoporosis. However, there was inadequate accuracy and very low predictive ability for identifying hip osteoporosis.

Prediction of bone mineral density of lumbar spine, hip, femoral neck and Ward's triangle by forearm bone mineral density.

Bone mineral density (BMD) of dual energy X-ray absorptiometry (DEXA) at both radius and ulna were measured to evaluate the correlation of those and BMD at lumbar (L) spine, hip, femoral neck and Ward's triangle. The 64 simple linear regression analysis was calculated to postulate the predicted equation by using the BMD at supradistal, distal 1/10, distal 1/6 and distal 1/3 of both forearms as independent variables, while the dependent variables were BMD at L1-L4 spine, total hip, femoral neck and Ward's triangle. 115 patients aged between 41-79 years (mean age 55.97+/-8.34 years) from the menopausal clinic, Pramongkutklao Hospital, were scanned at both forearms by Panasonic (DXA-70) DEXA and at non forearm regions by Hologic (QDR 4500) DEXA on the same day. The results showed that the BMD of each of the 4 parts of both radius and ulna had positive correlation to those of L-spine, total hip, femoral neck and Ward's triangle with r = 0.4012 to 0.7032 (P<0.001 for all). The greater distal of the forearm, the better correlation of BMD to the non forearm BMD. The 64 simple linear regression equations were constructed with regression coefficient ranging from 0.6048 to 1.9011 (P<0.001 for all). When considering the non forearm BMD, the mean BMD at Ward's triangle significantly declined more rapidly than that of L-spine, total hip and femoral neck (P<0.05 for all). It indicated that there was an early change of BMD at Ward's triangle. However, this change followed the forearm BMD. Distal forearm BMD was the earliest sign of bone loss. We can predict non forearm BMD by supradistal and distal 1/10 of forearm BMD.

Evaluation of forearm bone mineral density: comparison of dominant and non-dominant forearms.

For evaluation of forearm bone mineral density (BMD), (1) BMD of corresponding sites of dominant and non-dominant forearms were compared and (2) characteristics of each of the 4 regions of interest (ROIs) including supradistal, distal 1/10, distal 1/6 and distal 1/3 along the long bone of both forearms were analyzed. One hundred and forty one women (79 normal and 62 osteoporotic) were recruited by randomized selection from the department of Nuclear Medicine of Phramongkutklao Hospital. Both dominant and non-dominant forearms of each subject were scanned by Panasonic (DXA-70) dual energy X-ray absorptiometry (DEXA) on the same day. Lumbar spine BMD was also measured by Hologic DEXA (QDR-4500) and WHO criteria for diagnosis of osteoporosis was applied for identifying osteoporosis and normal groups. The results showed that none of the corresponding sites of BMD of both forearms were significantly different (p>0.05 for all). The BMD from distal to proximal of each long bone (radius and ulna) of both forearms was gradually increased in osteoporosis and normal groups. Further distal sites of the forearms and lower BMD were found. Comparison between mean BMD at corresponding sites in normal and osteoporotic groups, had significantly different BMD at both radii for all ROIs (p<0.05). While BMD at corresponding sites of both ulna in the 2 groups was not significantly different (p>0.05). A great percentage change of mean BMD in the osteoporotic group was seen at supradistal and distal 1/10 of both forearms when using BMD in the normal group as control. We suggest that both distal radii especially at supradistal and distal 1/10 sites should be scanned in routine practice. The distal location of the forearms had a relatively smaller amount of surrounding soft tissue than the proximal.

The use of historical and anthropometric data as risk factors for screening of low mBMD & MCI.

To evaluate the risk factors which affect bone loss in screening for osteoporosis, interview of anamnestic data (age, marriage status, pregnancies, menopausal age, intake of calcium, vegetables, protein and coffee, excessive use of alcohol and smoking, sedentary habits, family history), medical data, surgical data, followed by measurement of anthropometric variables [weight, height, antero-posterior (AP) thickness at xiphoid level], blood examination (calcium, inorganic phosphorus, alkaline phosphatase), both postero-anterior (PA) hands and lateral thoraco-lumbar radiography were done in 1,182 normal volunteers aged 17-83. From PA hands radiographs, metacarpal bone mineral density (mBMD) and metacarpal index (MCI) were measured by computed X-ray densitometry (CXD) (Bonalyzer, Teijin Ltd., Tokyo). The results showed that the mean of menopausal age in Thai females was 48.86 +/- 3.09 years ranging from 39 to 55 years. The average number of children in their family was 2.10. Correlation among anthropometric variables, AP thickness was positive linear correlation to weight/height ratio (r = 0.7878, p-value < 0.005). Weight, AP thickness and body mass index (BMI) significantly increased with aging (r = 0.2456, 0.4489 and 0.3484, p-value < 0.005, 0.001 and < 0.005), but decreased with height (r = -0.1030, p-value = 0.001). Lower mBMD and MCI were associated with increased age, married female, increased pregnancies, increased AP thickness, decreased vegetable intake, increased protein intake and increased years after menopause. From a multiple regression analysis, the significant factors that can predict the MCI were years after menopause, sex, daily vegetable intake and hormonal replacement. The incidence rate of high risk of developing osteoporosis in females, no vegetable intake and no hormonal replacement subjects occurred 7.50, 2.22 and 2.63 times greater than in males, vegetable intake and hormonal replacement subjects, respectively. In postmenopausal women since 1-2, 3-5, 6-10, 11-15 and > 15 years, the incidence rate were 5.24, 14.51, 17.01, 20.86 and 29.76 times greater than the rate of premenopausal women. Concerning perimenopausal women, only 2 of all factors influenced the measured mBMD and MCI. The incidence rate of high risk of developing osteoporosis in women who intake protein > 30 g/d and intake medicine (corticosteroid) was 2.96 and 6.16 times greater than < 30 g/d protein intake and no medicine intake subjects.

Computed X-ray densitometry measurement of mBMD & MCI in normal Thai and osteoporotic patients.

To establish the reference values of age-related change of metacarpal bone mineral density (mBMD) and metacarpal index (MCI) in screening for osteoporosis, both postero-anterior (PA) hands and lateral thoraco-lumbar radiography were done on 1,182 normal volunteers aged 17-83. From PA hands radiographs, mBMD and MCI were measured by computed X-ray densitometry (CXD) (Bonalyzer, Teijin Ltd., Tokyo). Exclusion of the surgical menopause condition and the causes of affected bone loss, the results show that mean mBMD and MCI in various age groups were significantly different (p-value < 0.005 for both) in females. Both values increased gradually from age under 20 and peaked in the 30-39 years age group, then decreased gradually until age 50 and decreased markedly after age 50. The yearly rate of bone loss from the peak density detected by mBMD and MCI was 1.3 per cent and 1.6 per cent between aged 50-59, 1.6 per cent and 2.7 per cent in subjects aged 60-69, 1.3 per cent and 3.2 per cent in those aged 70-79. However, mBMD and MCI in males did not show a downward trend with age. It indicated that a screening program for early prevention of osteoporosis may be necessary only in females before, during and after menopause. Because 92.3 per cent of 39 osteoporotic subjects had abnormal CXD measurements lower than -2 standard deviations (SD) limit of mean mBMD in young healthy women (aged 20-40 years), this value appeared to constitute a satisfactory definition of "high risk of developing osteoporosis". The incidence rate of high risk of developing osteoporosis was 3.03 per cent in a normal young population, while the risk rate occurred 4.76, 13.14, 34.28, 47.30 and 47.00 per cent in subjects aged 40-49, 50-59, 60-69, 70-79 and > 80, respectively. Results confirmed the necessity of early prevention of osteoporosis in postmenopausal women. These measurements may be appropriate for mass screening to separate patients who have a greater risk for development of osteoporosis from those at lesser risk.