IBMS/ECTS 2001 - PROGRAM and ABSTRACTS
POSTER PRESENTATIONS
All poster presentations will be displayed throughout the conference.
The suffix after the abstract number indicates the poster session in which the poster will be attended
(W - Wednesday, T - Thursday, S - Saturday).
Click on the abstract number to view the poster session.
Osteoporosis: Diagnosis
YOUNG ADULT BONE MINERAL DENSITY IN CROATIA
S. Cvijetic*, S. Bolanca, D. Dekanic
Institute for Medical Research and Occupational Health, Ksavesrka cesta 2, Zagreb, Croatia
Aim: in order to establish a national reference database, bone mineral density (BMD, g/cm2) of the lumbar spine (L2-L4) and of the left femoral neck of randomly chosen subjects, aged 20-49 years, was measured by dual x-ray absorptiometry (Lunar DPX).
Subjects and Methods: subjects of both sex, across the age range 20-49 years from the city of Zagreb participated in the study. The part of the participants was randomly selected from the demographic lists; the other group comprised of subjects who had been referred to the densitometry and who fulfilled the criteria for establishment of reference data for bone mineral density. Height and weight were measured in all subjects. There were 28% women who were in postmenopause. Bone mineral density (BMD, g/cm2) of the lumbar spine (L2-L4) and the left femoral neck was measured by dual x-ray absorptiometry (Lunar DPX).
Results: The mean BMD values did not differ significantly between five-year age groups in both sexes. In women there was an apparent decline in BMD after 40 years of life.
Conclusion: There is a tendency of BMD fall after the age of menopause in women. The peak bone mass is obtained between 25 and 30 years in women and between 20 and 25 years in men.
| BMD data in 5 year age groups | Men (N=248) | Women (N=246) | ||
| Age | L2-L4 | Neck | L2-L4 | Neck |
| 20-24 | 1.202 | 1.076 | 1.198 | 0.995 |
| 25-29 | 1.232 | 1.049 | 1.202 | 1.002 |
| 30-34 | 1.224 | 0.998 | 1.206 | 0.987 |
| 35-39 | 1.206 | 1.046 | 1.191 | 0.969 |
| 40-44 | 1.195 | 0.967 | 1.162 | 0.955 |
| 45-49 | 1.775 | 0.927 | 1.113 | 0.932 |
WHEN SHOULD DENSITOMETRY BE REPEATED IN HEALTHY PERI- AND POSTMENOPAUSAL WOMEN? THE DANISH OSTEOPOROSIS PREVENTION STUDY.
B. Abrahamsen1*, N. Nissen1, A. P. Hermann2, B. Hansen3, O. Bärenholdt4, P. Vestergaard2, C. L. Tofteng3, S. Pors Nielsen4
1Dept of Endocrinology, Odense Univ. Hosp
2Univ. dept. of Endocrinology, Aarhus Amtssygehus
3Osteoporosis Centre, H:S Hvidovre Hospital
4Dept of Clin. Physiology, Hilleroed Hospital, Denmark
Introduction: Oestrogen / SERM treatment should be considered in postmenopausal women with BMD 1 SD or more below the age-matched reference (T- or Z-score < -1). It is unclear, however, how soon densitometry should be repeated. The aim of this study was to determine the need for repeat DXA within 5 years in untreated postmenopausal women in order to predict declines of T- or Z-score to below –1.
Population and methods: 925 healthy women (age 51.2y [SD:2.9], spine BMD 1.1019 [0.136], Z-score +0.25 [1.23] who were followed for 5 years without HRT. DXA of spine, hip, forearm and whole-body was done at inclusion and after 1, 2, 3 and 5 years (Hologic QDR-1000/W og -2000).
Results: The annual rate of loss in SD-units was 0.12[0.10] at the spine (1.3%), 0.10[0.09] at the femoral neck (1.2%) and 0.07[0.09] at the UD forearm (1.0%). The table below shows the risk of developing Z < -1 as a function of follow-up time and baseline Z-score.
For T-scores the risk (spine) for developing T below –1 in patients with baseline T-scores of [0 to 1] was 4.1% at 3 and 8.6% at 5y. For baseline T-score in the [–1 to 0] range the risk was 48.8% at 3 and 60.7% at 5 years.
Conclusion: The loss rate in SD-units was higher in the spine than in the femoral neck or forearm. In accordance with this, Z-scores below –1 were reached earlier at the spine. The need for repeated DXA depends strongly on baseline BMD. Thus, DXA does not need to be repeated for 5 years if the baseline Z-score is positive, beacuse the risk of developing Z <-1 is less than 1%. In contrast, if the Z-score is below 0 at baseline a new DXA scan should be done after one or at most two years.
| Risk of developing low Z-score | |||||
| BMD | Risk of Z-score < - 1 | ||||
| Site | Baseline Z-score | 1 y | 2 y | 3 y | 5 y |
| Spine | < -1 (n=137) | 91.5% | 90.0% | 88.1% | 86.1% |
| -1 to 0 (n=275) | 14.9% | 21.4% | 23.8% | 29.8% | |
| 0 to 1 (n=279) | 0 | 0.4% | 0.8% | 0.4% | |
| > 1 (n=233) | 0 | 0 | 0 | 0 | |
| Neck | < -1 (n=162) | 87.7% | 87.7% | 89.9% | 82.7% |
| -1 to 0 (n=281) | 11.4% | 15.7% | 18.1% | 19.9% | |
| 0 to 1 (n=263) | 0 | 0 | 0.4% | 0 | |
| > 1 (n=163) | 0 | 0 | 0 | 0 | |
DOES SOCIOECONOMIC DEPRIVATION INFLUENCE A PATIENT'S LIKELIHOOD OF BEING REFERRED FOR BONE DENSITOMETRY?
J. Saunders1, A. Johansen1,2*, M. Stone2, W. Evans3, S. Jones1, R. Lyons1
1Collaboration for Accident Prevention and Injury Control, Welsh Combined Centres for Public Health, University Of Wales College of Medicine, Cardiff, UK
2Bone Research Unit, Academic Department of Geriatric Medicine, University of Wales College of Medicine, Cardiff, UK
3Department of Medical Physics, University Hospital of Wales, Cardiff, UK
Introduction
The likelihood of a person being referred for bone densitometry depends on local availability of the technology, as well as on factors such as their age, sex, medical history, and perceived risk of fracture. In practice patient demand is an important additional factor. We set out to determine whether a patient's socioeconomic status affects their likelihood of being referred for dual X-ray absorptiometry (DXA).
Methods
442,200 people living in Cardiff and the Vale of Glamorgan are served by DXA systems in the two local hospitals. Primary and secondary care physicians have open access to DXA assessment. During a twenty-three month period (July 1998 - May 2000) we identified all of the 85,756 females aged 45 and over who underwent DXA assessment. After exclusion of duplicate entries we identified 1,173 who were scanned, for whom demographic details including a post-code were recorded. We used individuals’ post-codes to determine the electoral ward in which they were resident. For individual electoral wards we calculated the overall incidence of fracture, and the rate of referral of DXA, and compared these figures with the published Townsend socioeconomic deprivation score for that ward.
Results
The 50 electoral wards showed wide variation in DXA referral rates (0.68 to 11.83 per 1,000 people per year, mean 6.49, SD 2.56), and Townsend deprivation scores (-7.57 to 9.76, mean -0.13, SD 4.54). There was significant correlation between fracture incidence and Townsend deprivation score (r=0.52, p<0.01), but DXA referral rates were unaffected by socioeconomic status (r=0.038, p=ns).
Conclusion
There is marked variation in referral rates between adjacent electoral wards, but this is not a reflection of the age, fracture risk or socioeconomic status of local residents. It appears to be more a manifestation of individual primary care physicians' attitude and clinical practice.
BONE MINERAL DENSITY IN LUMBAR SPINE AND FEMORAL NECK IN THE FEMALE POPULATION OF MOSCOW
N. V. Toroptsova*, N. V. Demin, L. I. Benevolenskaya
Institute of Rheumatology, Moscow, Russia
The purpose of the study was to determine normal bone mineral density (BMD) values in women using dual-energy X-ray absorptiometry (DEXA).
Material and methods: 400 women aged 15-85 years, stratified in 10-year age groups, had bone mass measurements at both lumbar spine (L1-L4) and femoral neck sites (Hologic QDR-4500A). Subjects with suspected conditions affecting bone metabolism or receiving any treatment affecting bone mineralisation were excluded.
Results: The highest value for lumbar spine BMD was found within the 25-34 years age group (1.026 g/cm2), being significantly lower before the age of 25 years and after the age of 55 years. Lumbar spine BMD shows no significant change through the age range of 35-54 years. The highest value for femoral neck BMD was found in the 20-24 years age group (0.842 g/cm2), values become statistically lower after the age of 55 years.
Conclusion: our study presents the first data on normal bone mineral density values in women in Russia. We determined peak bone mass for lumbar spine and femoral neck. Our data will be used for generating standard curves for bone mineral density in Russian population.
P305 T
Withdrawn
CORRELATION BETWEEN CENTRAL AND PERIPHERAL SKELETON BONE MINERAL DENSITY (BMD) DURING PEAK BONE MASS FORMATION
R. Carvajal, G. Riera-Espinoza*, M. Naressi, G. Velásquez, J. Rámos
UNILIME, Universidad de Carabobo, Hospital Universitario Dr. Angel Larralde, Valencia, Venezuela
Peak Bone Mass (PBM) formation is crucial to obtain an adequate bone mass through life. It is one of the main determinant of possible fractures in the late postmenopausal period. Estimation of formation or bone growth rate during childhood, adolescence and young adulthood represent singular importance in determine standards values for each population group and defining strategies that enhances PBM acquisition. Peripheral DEXA measurements are portable easy applicable and cheaper than central ones. The aim of this study was to show the correlation between BMD at lumbar and femoral neck with BMD at distal and ultradistal forearm.
Bone Mineral Density was measured to 89 subjects, 10-25 year-old, 42 male and 47 females, by DEXA (LUNAR-DPX) at lumbar spine L2-l4 and femoral neck. At forearm distal and ultradistal regions were measured by Osteometer DTX-200.
Distal and ultradistal values were strongly correlated (r= 0.902, p<0.000). Distal was significantly associated with L2-L4 and femoral neck (r= 0.757 and r= 0.689 respectively, p<0.000 for both). Ultradistal also was correlated with L2-l4 (r= 0.696, p<0.000) and femoral neck (r= 0.755, p<0.000)
CONCLUSION: During peak bone mass formation (10-25 year-old) peripheral measurements of BMD by DEXA (Osteometer DTX-200) at distal and ultradistal regions of the forearm are well correlated (r values around 0.7) with central BMD DEXA values at lumbar spine and femoral neck
MENOPAUSAL AGE AND BMI INFLUENCE CORRELATION BETWEEN LUMBAR AND FEMORAL BMD
M. Di Stefano*, I. Grosso, F. Navone, C. Roggia, S. Casalis, G. C. Isaia
Department of Internal Medicine - University of Torino, Italy
Aim of this study was to evaluate the correlation between BMD at lumbar spine and at various femoral subregions in women of different BMD, BMI and menopausal ages.
We measured both lumbar (L2-L4) and left femoral BMD by DXA technique (Hologic QDR 4500A) in 330 women (69 pre and 261 postmenopausal, age range 38-65 years). None of patients had diseases nor took drugs affecting phosphocalcium metabolism and none showed the presence of vertebral fracture, scoliosis, osteoarthrosis, calcifications or other conditions able to reduce the accuracy of measurement. We divided the patients into different groups on the basis of their lumbar BMD (<-2.5 SD, -1.5 SD to -2.5 SD, > -2.5 SD), BMI (<20, 20 to 25, 25 to 30, >30) and menopausal age (premenopause, < 5 years, 5 to 10 years, > 10 years since menopause). The Table shows the correlation (Pearson’s coefficient) between BMD of lumbar spine and of femoral subregions according to menopausal age (p<0.05 for all correlation).
Moreover, after dividing the women on the basis of their BMI (<20, 20 to 25, 25 to 30, >30), we found the higher correlation in the group with BMI > 30 (r=0.81, p<0.001 between lumbar and total femoral BMD).
Osteoporotic patient showed lower correlation coefficient than normal subjects (respectively r=0.55 and r=0.73, p<0.001 between lumbar and total femoral BMD).
On the basis of our data we can conclude that correlation between BMD of lumbar spine and of various femoral subregions significantly decreases according to the increase of postmenopausal age, and that in subjects with BMD less than -2.5 SD these correlation coefficients are lower. Patient with BMI > 30 showed the better correlation.
Moreover, osteoporosis is characterized not only by reduced BMD, but also by different amount of bone loss among spine and hip. It is possible to hypothyze that oestrogen reduction can increase the differences between BMD at various skeletal sites.
| Lumbar vs. | All patients | Premen. | Men. 0 to 5 | Men. 6 to 10 | Men. >10 |
| Neck | 0.64 | 0.72 | 0.65 | 0.54 | 0.51 |
| Troch | 0.63 | 0.71 | 0.64 | 0.48 | 0.52 |
| Inter | 0.64 | 0.69 | 0.63 | 0.51 | 0.47 |
| Total | 0.63 | 0.74 | 0.61 | 0.58 | 0.50 |
| Ward's | 0.61 | 0.72 | 0.59 | 0.63 | 0.52 |
DXA-ASSESSED REFERENCE CURVES BETWEEN BONE AND MUSCLE INDICATORS FOR A DIFERENTIAL DIAGNOSIS BETWEEN "PHYSIOLOGICAL" AND TRUE OSTEOPENIAS
J. L. Ferretti1,2*, R. F. Capozza1, G. R. Cointry1, H. Plotkin2, J. R. Zanchetta2,3
1Rosario University School of Medicine, Rosario, Argentina
2IDIM/FIM, Buenos Aires, Argentina
3USAL University School of Medicine, Buenos Aires, Argentina
The strength (and indirectly the mass) of bones is determined by the intrinsic stiffness and the spatial distribution of the mineralized matrix. Both these determinants are inversely related through a feedback mechanism (bone mechanostat). Bone mass / strength losses can only come from a disturb in that biomechanical regulation. The stimulus for the mechanostat comes from the strains determined by the mechanical usage of the skeleton (regional muscle contractions). As an evidence of that, the bone / muscle masses are linealy related, showing the same slope for any gender, age or body habitus. The setpoint of the system, genetically determined, is sensitive to systemic factors (hormones, drugs, etc). As an evidence of that, the intercepts of that relationship varies with gender and reproductive status. Therefore, bone mass can only be lost (ostepenia) because of a prolonged inactivity or weightlessness (reduced input, "physiologic" osteopenia), or a genetic or systemic disorder (shifted setpoint, "true" osteopenia). Only if a biomechanical compromise (bone fragility) can be demonstrated the condition should be regarded as a "disuse" or a "true" osteoporosis, respectively. DXA cand measure bone mass and diagnose an osteopenia (not an osteoporosis because it does not measaure tissue strength or distribution. However, DXA is able to assess all bone, fat, and lean (proportional to muscle) masses. Thus, DXA allows correlating the BMC and "muscle" mass and approaching a differential diagnosis between "physiologic" and "true" osteopenias (appropriate or inappropriate bone/ muscle mass proportion) as pre-conditions for diagnosing a "disuse" or a "true" osteoporosis, respectively.
We developed adequate reference BMC (raw or fat-adjusted values) / lean-mass charts for whole-body DXA determinations (XR-26, Norland, Wisconsin) in normal Argentine boys and girls (n=545), men (n=228), and pre- and post-menopausal women (n=330, 347) showing the 1-99 percentiles. A simple graphic procedure estimates any percentile and would approach a differential diagnosis between physiologic and true osteopenias according to a suitable reference limit. Appropriate factors allow the data transformation for use with different densitometers.
PEAK BONE MASS IN A VENEZUELAN POPULATION. PRELIMINARY REPORT
G. Riera-Espinoza*, R. Carvajal, M. Naressi, G. Velásquez, J. Ramos
UNILIME, Universidad de Carabobo, Hospital Universitario Dr. Angel Larralde, I.V.S.S, Valencia, Venezuela.
Peak Bone Mass (PBM) and progressive bone loss in men and women specially after menopause are the main risk for osteoporotic fractures that occurs later in life.
89 healthy subjects, 10-23 year-old were recruited for Bone Mineral Density assessment. BMD was measured by DEXA (LUNAR DPX) at femoral neck and lumbar spine (L2-L4) and by DEXA (Osteometer-200) at distal and ultradistal forearm.
Subjects were divided into two groups: 42 males and 47 females. In the age range 10-23 year-old males reach the highest bone mass at age 19 in both areas L2-L4 and femoral neck, 80% of that value was obtained at age 15-17 year-old. Women did so at older age, 22 year-old also at both sites: L2-L4 and femoral neck, and 80% of PBM was reached at younger age than men, 12-13 year-old. Mean values by quinquennium were for males: L2-L4 10-14: 0.883±0.15, 15-19: 1.181±0.15, 20-23: 1.121±0.08, Femoral neck 10-14: 0.947±0.18, 15-19: 1.115±0.1, 20-23: 1.032±0.11. In women L2-L4 10-14: 0.925±0.13, 15-19: 1.126±0.09, 20-23: 1.198±0.07, Femoral Neck 10-14: 0.823±0.16, 15-19: 1.006±0.09, 20-23: 1.016±0.07
Conclusion. As a preliminary report Peak Bone Mass in Venezuelan population 10-23 year-old is achieved at 19 year-old in men (L2-L4: 1.296±0.01 gr/cm2, femoral neck 1.195±0.004 gr/cm2) and at 22 in women (L2-L4: 1.250±0.09 gr/cm2, femoral neck: 1.049±0.06 gr/cm2). 80% of that PBM is achieved later in men, 16-17 year-old, than women, 12-13 year-old.
INFLUENCE OF DIFFERENT DXA DEVICES ON DETECTION OF OSTEOPOROSIS IN PATIENTS SAMPLE
P. Kasalicky1,2*, J. Rosa2,3, J. Kocian2
1Postgraduate Medical School, Ist Med.Clin.
2MEDISCAN densitometric center
3Reumatologic institute
We compared results of DXA densitometry obtained in MEDISCAN densitometric center during years 1999 and 2000. This busy center is one of regional DXA centers in Prague, used by various referring specialists and GPs.
During the year 1999 we measured 3791 patients by Norland-XR 26 device. BMD in the region of femoral neck and L spine (L2-L4) was measured and calculated by software version 2.5.1., with US normative population supplied by Norland. After replacing DXA device in our densitometric center by Lunar Prodigy (May 2000) patients were measured only with this device. In period V/2000-XI/ 2000 we measured 3358 patients, for analysis was used German reference population supplied by Lunar.
We had few data for comparing the same patients measured subsequently on both devices, so only groups of patients, measured during different period, were compared. Indications and specialisation of reffering physicians didn’t change in these periods and so we supposed similar proportion of osteoporotic patients measured on both devices. We compared number of patients classified as osteoporotic after measuring by Norland device and by Lunar device, according to T-score and WHO osteoporosis definition. In Norland group 34% of all patients had T-score under –2.5 in region of left femoral neck, 16% in region L2-L4, in both region it was 12%. In Lunar Prodigy group these numbers were: 14%, 22% and 7% resp. In Norland group of patients with age over 70 years had osteoporosis 70% in region of femoral neck, 28% in L spine and 25% in both regions, in Prodigy 32% in femoral neck,35% in L spine and 16% in both regions. In group over age 75 this difference was even marked: 74% patients in Norland group were classified as osteoporotic, compared with 35% in Lunar group. In L spine 25% of patients measured by Norland and 37%measured by Lunar were classified as osteoporotic.
So in our sample the type of DXA device and reference population were the most important factors for classification as osteoporotic, with impact on therapy. Problems with explanation of results of patients measured on Norland with described osteoporosis in the hip region and subsequently only osteopenic or normal results obtained by Lunar Prodigy could be expected.
SEVERE OSTEOPENIA IN A YOUNG BOY WITH KOSTMANN'S CONGENITAL NEUTROPENIA TREATED WITH G-CSF: SUGGESTED ETIOLOGY AND TREATMENT
R. Vassilopoulou-Sellin*, R. Sekhar, S. Culbert, W. K. Hoots, M. J. Klein, H. Zietz
University of Texas M D Anderson Cancer Center, Houston, Texas, USA
Kostmann's syndrome, a congenital disorder causing impairment of myeloid differentiation characterized by severe neutropenia, which can be treated with recombinant human granulocyte colony-stimulating factor (G-CSF). We present a 13-year old-boy with Kostmann's syndrome, treated with G-CSF from age 3.5 years. His growth and development, was normal, although complicated by intermittent infections. He was referred to Endocrinology with severe osteopenia at the spine and hips (lumbar spine BMD 0.486 g/cm2; Z-score -3.6). Ionized calcium level at the upper limit of normal (1.28 mmol/L; range, 1.13-1.32 mmol/L), suppressed intact PTH level (12 pg/ml; range, 10-65 pg/ml), and low 1,25-dihydroxy vitamin D (1,25-VitD) level (21 pg/ml; range, 24-65 pg/ml). He had elevated urinary deoxypyridinoline (DPD) crosslinks (46.9 nmol/mmol creatinine; range, 2-34 nmol/mmol creatinine), elevated osteocalcin level (200 ng/ml; normal, 20-80 ng/ml) and alkaline phosphatase level (236 IU/ml; normal, 38-126 IU/ml). Because of clinical concern for his skeletal health, he received intravenous pamidronate. One month later, the iPTH and DPD crosslinks were in the normal range (54 pg/ml and 17.7 nmol/mmol creatinine respectively) and the 1,25 VitD level was elevated (111 pg/ml). Three months after treatment, BMD increased at the lumbar spine (+30.86%), femoral necks (left, +20.02%; right, +17.98%), and total hips (left, +18.40%; right, +15.94%). Seven months after bisphosphonate therapy, there was a return toward pretreatment levels with increasing urinary DPD crosslinks (28.7 nmol/mmol creatinine) and decreasing iPTH (26 pg/ml). While BMD continued to increase, the magnitude of improvement was attenuated as well (lumbar-spine, +4.8%; left total hip, +1.2% and right total hip +2.4% relative to BMD at 3 months). Eight months after treatment, iPTH was suppressed at 14 pg/ ml. We hypothesize that prolonged administration of G-CSF as treatment for Kostmann's syndrome is associated with increased bone resorption mediated by osteoclast activation leading to bone loss. The osteopenia may be successfully treated with antiresorptive bisphosphonates, resulting in marked improvement in BMD. Regular measurement of iPTH can serve to monitor the magnitude and duration of the therapeutic response and the need for BMD reassessment and, perhaps, retreatment.
RATING OF THE MICRO-CT MEASUREMENT OF HARD TEETH SUBSTANCES FOR THE OSTEOPOROSIS-DIAGNOSIS
L. Kettler1,2*, E. Paulisch1, K. P. Lange1, M. Giehl2, W. Gowin2, D. Felsenberg2
1Humboldt Universität zu Berlin, Germany
2Freie Universität Berlin, Germany
In the following discourse an attempt has been made to verify two different diseases using the Micro-CT diagnostic method, and a special examination object - the human tooth. The diseases examined are osteoporosis, from the orthopedic and radiologic point of view and from the dentist's point of view the tooth fracture for persons suffering from osteoporosis. In former scientific works, for example by Türb and Gobetti (1996) the problem of "cracked tooth syndrome" was examined, however a possible relationship with osteoporosis was not taken into consideration. Until now there has been no radiologic examination taking in perticular the human tooth as parameter for the manifestation of osteoporosis. Up to now a relation between spontaneous cracked tooth of patients older than 50 years and a disease of osteoporosis could not be proven, thus making any indication with explanation models, valuable. If it could be proved that osteoporosis also attacks teeth, the following practical application would result of it. A tooth which had to be extracted for dental reasons could be radiologically examined by Micro-CT. In total 14 teeth of 9 different proportionary patients have been measured by means of Micro-CT. The low number of examined samples results from the time consuming examination with the Micro-CT of approximately 5 hours per examined tooth plus necessary interpretations of received data in three dimensions. Teeth were chosen for following criterions: 1. The existing of an already manifested OPTG (osteoporosis manifestation on the jaw). 2. The diagnosis received through patients questioning if they suffer from osteoporosis. 3. Healthy teeth not damaged by caries which had be extracted by parodontal reasons. The advantage of Micro-CT however is that the structure is not damaged, in contrast to a section and the examined piece theoretically can be shown infintesimally often in all three dimensions. Summarized it can be declared, that further scientifical clinical studies with a resembling design are making an evidence based evaluation necessarry and in addition to this being from a practical clinical signification.
ELECTROALGOMETRY. DECREASE OF SKIN IMPEDANCE AS A MEASURE OF BONE AND JOINT PAIN
T. Fujita1,2*, Y. Fujii1, A. Miyauchi3, Y. Takagi3
1Calcium Research Institute, Osaka, Japan
2Katsuragi Hospital, Osaka, Japan
3National-Hyogo Chuo Hospital, Hyogo, Japan
Evaluation of musculoskeletal pain, one of the most annoying manifestations of metabolic bone and joint diseases in old age, osteoporosis and osteoarth ritis, has so far depended entirely on subjective complaints under the influence of psychoemotional factors. Although analgesic effect is one of the most important aims of treatment to improve the quality of life, it has not been used as a reliable endpoint because of a lack of an objective method of evaluation. Galvanic skin response was successfully used to differentiate analgesic calcitonin and placebo effect by us. As another electrodermal phenomenon associated with pain, fall of skin impedance was measured under various strains on the spine and knee to assess its relationship with subjective pain on visual rating scale (VRS). Skin impedance was measured by using Skin Impedance Meter D46-03 manufactured by Fukuda Electric Company, Tokyo in 22 normal subjects and 178 patients with osteoporosis, osteoarthritis of the knee and/or spondylosis deformans in a quiet air conditioned test room. Skin impedance at rest was stable in a single subject within 1 hour with a coefficient of variation of 2.3%. With advance in age, an increase of skin impedance at rest was noted, with slightly higher values in females than in males. When the test subjects stood up, bent forward, squatted, walked or climbed up and down the stairs, the skin impedance fell by 20 to 60% parallel to the subjective pain. Intermittent administration of 33 to 200 mg/day etidronate gave a marked analgesic effect especially in degenerative joint disease, as indicated by VRS and fall of skin inpedance which appears to be useful as electroalgometry. In spondylosis deformans, this was accompanied by alleviation of spinal deformity suggested by a decrease of intraindividual variation of L1 to L4 BMD in DXA, indicating a parellel analgesic and reparative effects of etidronate.
LOW COST AND SENSITIVE SCREENING METHOD FOR EVALUATING POST-MENOPAUSAL OSTEOPOROSIS
M. Anburajan1*, C. Rethinasabapathy1, M. Paul Korath1, G. N. S. Prasad2, B. G. Ponnappa1, T. N. R. Panicker1, A. Govindan1, K. Jagadeesan1
1K.J. Research Foundation, K.J. Hospital, 927, P.H. Road, Chennai -600 084, Tamil Nadu, India
2Dept. Of Physis, Anna University, Chennai
OBJECTIVE: Establishing empirical link between total hip BMD by DXA and clavicle radiogrammetry from chest radiograph that gives results of high sensitivity and specificity for predicting total hip BMD and facilitating evaluation of osteoporosis in developing countries.
MATERIALS & METHODS: 50 south Indian women, aged 16-84 years studied. Patients with secondary bone diseases excluded. Pre-menopausal: (n=21, M=31.0±8.8 years); Post-menopausal: (n=29, M=64.3±11.2 years). Hip BMD (g/ cm2) measured by DXA. Standard PA chest radiograph taken and clavicle radiogrammetry made. WHO's diagnostic criteria used. Data analysis done with SPSS/PC software package.
RESULTS: In 45 non-fracture pre- and post-menopausal women, total hip BMD by DXA was correlated significantly with combined cortical thickness (CCT) and %CCT of the clavicle (r=0.73, p<0.001; r=0.74, p<0.001). There was statistically significant negative correlation between total hip BMD and age (r=-0.64, p<0.001). Clavicle CCT in cm (X2), clavicle %CCT (X1) and age in years (X3) used as independent variables in a multiple linear regression analysis to predict total hip BMD g/cm2 (Y, dependent variable) using: Y= A + B1 X1 + B2 X2 + B3 X3------I; Obtained coefficients were: A=0.505563, B1=0.003938, B2=0.370914 and B3=-0.0005. This multiple linear regression equation had low sensitivity of 76.5% and specificity of 97.0% for evaluating osteoporosis. In order to improve sensitivity further, relative weights of coefficients changed by trial and error. Final values of A=0.47 and B2=0.420914 increased sensitivity of equation-I to 82.4% with specificity of 94.0%. Hence, finalised KJH; Anburajan's empirical formula is: Y= 0.47 + 0.003938 (X1) + 0.420914 (X2) - 0.0005 (X3)----II. Positive and negative predictive value of the above formula was 87.5% and 91.2% respectively.
CONCLUSION: The KJH; Anburajan's formula is useful to predict hip BMD from cheap and widely available standard chest radiograph with good sensitivity and specificity - a boon to developing nations.
Visit: www.horizon-solution.com/osteoporosis
REDUCED LUNG FUNCTION IN WOMEN WITH VERTEBRAL FRACTURES
J. A. Falch*, C. Bugge
Osteoporosis Clinic, Aker Hospital, Oslo, Norway
Vertebral fractures will result in an increased thoracic kyphosis with deformation of the thorax. We studied whether this would influence the lung function measured by spirometry in osteoporotic women. We also wanted to see if the fracture induced loss of body height would influence the calculation of expected lung function.
Seventy women with vertebral fractures (OP group) were evaluated by spirometry, body height and measurement of head-wall distance as an estimate of kyphosis. As a control group, 66 women 67 years of age without fractures had the same measurements. Per cent expected vital capacity (VC) and one second forced expiratory volume (FEV1) were calculated using a Norwegian reference material for the spirometry data. The age of the women with fracture was 73(8) years (mean(SD)). Their height loss was 6.9(4.3) cm (range 0-18 cm). Expected VC calculated with actual height (VC-AH) was 75(18)%, but using the historical height of the patients, the expected VC (VC-HH) was 66(18)% (p<0.0001). Expected FEV1 calculated the same ways were 74(17)% and 67(18)%, respectively. In the 66 normal women expected VC was 91(21)% and FEV1 89(17)%. These values were significantly higher compared to the OP group (p<0.0001). In the OP group, the head-wall distance was 5.0(4.7) cm (range 0-15 cm). This distance correlated to VC-AH, FEV1-AH, VC-HH, and FEV1-HH with the following r-values: -0.21, -0.42, -0.40, and -0.55.
In women with vertebral fractures, a reduced lung function estimated by VC and FEV1 is found, and the lung function is negatively correlated to the degree of kyphosis. Whether the expected lung function should be calculated from the actual or historical body height is unknown. However, an even worse lung function is found if historical height is used. It is probable that the expected lung function calculated from a Norwegian reference material will be underestimated in elderly women, as our control group had both VC and FEV1 approximately 10% lower than expected.
APPLYING A SIMPLE CLINICAL TOOL TO IDENTIFY OSTEOPOROSIS AMONG JAPANESE WOMEN
S. Fujiwara1*, N. Masunari1, G. Suzuki1, B. Sedrine2, J. Y. Reginster2
1Radiation Effects Research Foundation, Hiroshima, Japan
2WHO Collaborating Center for Public Heath Aspects of Rheumatic Disorders, Liege, Belgium
The objective of this study was to determine if a simple scoring index for predicting osteoporosis (femoral neck BMD T < -2.5) can apply to Japanese women. A simple scoring index had been developed earlier, based on multiple variable regression modeling using data from postmenopausal women in Asian countries other than Japan. The final index contained only 2 variables (age and body weight) and achieved 91% sensitivity and 45% specificity. The area under the curve (AUC) was 0.79. We applied this model to Japanese women in a population-based study (Adult Heath Study (AHS)). The AHS recruited a cohort of about 20,000 people in Hiroshima and Nagasaki based on the 1950 Japanese national census; this cohort has received biennial health examinations since 1958. The subjects of the present analysis are 1,127 women aged 50 years and older (average age 65±9.2 years) who received both spine and femoral neck BMD measurements between 1994-1995. Three risk categories were designated, based on the risk index values; 25% of the AHS population was categorized as high risk, 50% as moderate risk, and 25% as low risk. Almost half (44%) of the high risk patients had osteoporosis (hip T-score < -2.5) – physicians might be advised to measure BMD to confirm diagnosis. Among women with low risk, only 1% had osteoporosis, and BMD measurements are probably not necessary. Approximately 10% of the moderate risk women had osteoporosis; the decision to measure BMD for this category may vary by other potential risk factors. In Japan, spine BMD is commonly used when diagnosing osteoporosis. When using spine BMD for diagnosis, the percentages of osteoporosis in the three risk categories were almost same as those using femoral neck BMD. In summary, the scoring index had acceptable predictive ability and should be easy to use for Japanese women. This free and simple risk assessment tool could help clinicians and patients actively assess osteoporosis and determine the need for BMD measurements and intervention before fractures occur.
AN ALGORITHM FOR THE VISUAL AND QUANTITATIVE IDENTIFICATION OF PREVALENT VERTEBRAL FRACTURES
G. Jiang1*, N. A. Barrington2, R. Eastell1
1Bone Metabolism Group, University of Sheffield, UK
2Diagnostic Imaging, Northern General Hospital, Sheffield, UK
The identification of vertebral fractures from radiographs is difficult due to the presence of deformities that have not resulted from fractures. These non-fracture deformities include degenerative changes, subtle wedging (anterior and lateral), Scheuermann’s disease, Schmorl’s nodes, and tumours, including angioma. Fractures always include the vertebral endplate and can be considered to be wedge (endplate and anterior ring epiphysis), concave (endplate alone) and compression (endplate, anterior and posterior ring epiphyses). The aims of this study were to identify non-fracture vertebral deformities and then compare visual and morphometric definitions of vertebral fracture. The first morphometric definition (QM1) is based on a reduction in vertebral height ratios by more than 3 standard deviations below the mean height ratio,(these are calculated from each population using a statistical trimming approach) and the second definition (QM3) is based on a 15% reduction in standardised vertebral heights. In a population-based sample of 372 women ages 50 to 85 years, we identified 26 women with non-fracture deformity alone, 23 women with fracture deformity, and 325 with no deformities. Bone mineral density was only low in the group with fracture deformity. We found better agreement between visual diagnosis with QM3 (kappa: 0.819, 95% CI: 0.711~0.926) than with QM1 (kappa: 0.690, 95% CI: 0.566~0.814). In a clinic-based sample of 80 women ages 48 to 87 years referred with osteoporosis we carried out the same approach of identifying non-fracture deformities and then excluding those vertebrae from further analysis. We found better agreement between visual diagnosis with QM3 (kappa: 0.935, 95% CI: 0.912~0.959) than with QM1 (kappa: 0.539, 95% CI: 0.476~0.601). However, the prevalence of fracture was so high in this population that the estimates of mean and standard deviation using the trimming approach gave incorrect results. We then applied the mean and standard deviation from the population-based group and found a better agreement between visual diagnosis and QM1 (kappa: 0.950, 95% CI: 0.929~0.970). We conclude that similar results can be obtained by visual diagnosis and quantitative morphometry (QM3) if non-fracture deformities are identified first.
COMPARISON OF ATTITUDES TO OSTEOPOROSIS IN TWO REGIONS OF THE EUROPEAN COMMUNITY
J. C. Taylor1*, G. P. R. Clunie2, A. M. Scarponi3, D. O'Reilly4, F. Vecchini3, E. Mannarino3, J. C. Chamberlain1, H. Fitz-Clarence1
1Osteoporosis Unit, Centre for Rheumatology, UCLH, London, UK
2Rheumatology Dept., Ipswich Hospital NHS Trust, Ipswich, Suffolk, UK
3Osteoporosis Centre, University of Perugia, Umbria, Italy
4Rheumatology Dept., West Suffolk NHS Trust, Bury St. Edmunds, Suffolk, UK
Although diagnostic facilities and prioritisation of health care can vary, little is known of the variation in understanding of this disease among medical practitioners. We surveyed two regions of the European Community assessing differences in attitudes, opinions and knowledge of general practitioners (GPs) regarding osteoporosis.
Equivalent questionnaires were sent to GPs in Suffolk (SU), UK and Umbria (UM), Italy. These areas have a low population density with access to specialist advice for osteoporosis.
245/345 GPs (71%) responded from SU and 179/496 (36%) from UM after 2 mailings. There were more single handed GPs in UM. GPs in both regions rated osteoporosis as an important problem (90% SU, 91% UM). However, GPs remain unconvinced about the effectiveness of treatment (76% SU, 55% UM). Lifestyle advice and prescription of calcium and vitamin D was frequent, but GPs in UM prescribed considerably less HRT (66% SU, 36% UM). Alendronate was used more frequently than etidronate in UM (alendronate 40%, etidronate 23%), but the reverse was true for SU (alendronate 16%, etidronate 42%). 67% of GPs in UM had access to DEXA and 98% of GPs used the service to diagnose and monitor patients. In SU 56% of GPs had access to DEXA (68% use it for diagnosis and 47% for monitoring). Less GPs (36%) in UM were aware of guidelines than GPs in SU (55%). GPs in both regions consider osteoporosis a primary care issue (87% SU, 76% UM), but would prefer a one stop assessment in hospital with subsequent follow up in general practice. In general GPs consider management of secondary osteoporosis should include hospital care (63% SU, 83% UM), but would prefer shared care.
Prescribing habits differ between both regions particularly in the use of HRT and preferred bisphosphonate. The difference in availability of DEXA machines for both countries may account for differences in access to bone densitometry, but GPs in the UK underuse such facilities. GPs are unconvinced about the efficacy of therapy for osteoporosis, although further education may influence change. Our findings show a similarity in the attitudes of GPs in the UK and Italy. GPs are aware of the public impact of osteoporosis and are willing to participate in the management of patients with osteoporosis.
VALIDATION OF A SIMPLE CLINICAL TOOL TO IDENTIFY ASIAN WOMEN WITH OSTEOPOROSIS
R. Rajatanavin*, K-S. Tsai, S-P. Chan, A. Kung, Q-R. Huang, M-I. Kang, S. B. Chionh, W. Ben Sedrine, J-Y. Reginster
The Osteoporosis Risk Assessment Tool for Asia Research Group, World Health Organization Collaborating Center for Public Health Aspects of Rheumatism
We previously developed a risk assessment tool for identifying osteoporosis (femoral neck BMD T < -2.5) among women in Asia, using risk factors obtained by questionnaire. The tool was based on multiple variable regression modeling using data from postmenopausal women in Asian countries other than Japan. The full index with 11 variables achieved 95% sensitivity (sens) and 47% specificity (spec). The area under the curve (AUC) was 0.85. Removing all variables except patient age and weight did not materially reduce predictive ability (sens = 91%, spec = 45%; AUC = 0.79), but substantially improved simplicity. The objective of the current study was to compare the performance of several other published risk assessment indices that had been developed in primarily Caucasian populations to the performance of our tool in postmenopausal Asian women. Other published risk indices included the Osteoporosis Risk Assessment Index (ORAI), the SOFSURF index, and the Simple Calculated Osteoporosis Risk Estimation (SCORE). These published indices involve calculations using 3 to 6 risk factors. The results among our Asian sample were: sens = 90%, spec = 46%, AUC = 0.77 for the SOFSURF index, sens = 90%, spec = 33%, AUC = 0.77 for the SCORE index, and sens = 84%, spec = 52%, AUC = 0.76 for the ORAI. Thus, the performance of other indices in this Asian population was similar to that of our index based on age and weight alone. In summary, our risk assessment index performed as well as other indices that involve calculations using larger numbers of risk factors. Our index is very easy to use - risk can be tabulated by age and weight, so that calculations are not necessary. This free and simple risk assessment tool could help clinicians actively assess osteoporosis and determine the need for intervention or BMD measurements before fractures occur.
VERTEBRAL MORPHOMETRY AND THORACIC KYPHOSIS
D. D. Daniele Diacinti*, E. D. Emilio D'Erasmo, S. M. Salvatore Minisola, E. T. Ernesto Tomei, G. F. M. GianFranco Mazzuoli
University "La Sapienza", Rome, Italy
Hyperkyphosis and loss of height are common in the elderly population, particulary in osteoporotic women. The aim of this longitudinal study was to evaluate with morphometric radiography (MRX) the contribution of decrease vertebral body heights to the thoracic kyphosis.
Materials and methods: 156 asymptomatic healthy women (26 premenopausal) with an age range of 46-74 years were enrolled after were excluded vertebral fractures on lateral thoracic films centring on level T7 (tube-to-film distance of 115 cm). The height and weight of each patient was documented. In all women studied was measured lumbar bone mineral density (LS-BMD) by dual-energy x-ray absorptiometry (DXA) using the Hologic QDR-4500 densitometer (Hologic, USA). DXA examination and lateral spine radiographs were repeated after two years. On each lateral thoracic radiograph the kyphotic angle was calculated with the method of Fon et al. (1).The films were digitized by means of a scanner and then was performed the vertebral morphometry from T4-T12 using specified software (QR-Verona).The computer automatically calculated the anterior, middle and posterior vertebral bodies heights (Ha, Hm, Hp), the ratios of heights of single vertebrae (Ha/Hp, Hm/Hp, Hp/Hpp), as well as the sum of vertebral body heights (AHs, MHs and PHs).
Results: after 24 month we observed a decrement of the vertebral heights respect to basal values, small (-1.2 mm) in premenopausal sample, higher in postmenopausal women (-3.5mm), especially within seven years post-menopausal: -4.1mm versus -2.9mm (p<0.02). The highest AHs decrement respect to PHs (-2.9mm versus –1.4mm) resulted in a significantly lower AHs/PHs ratio both in premenopausal (0.995 versus 0.991; p<0.05) and postmenopausal women (0.991 versus 0.984; p<0.001). AHs/PHs ratio correlated significantly with angle of kyphosis (r=-0.49; p<0.001).
Conclusion: this study demonstrates that vertebral heights decrease with aging and menopause. The AHs/PHs ratio may be used as index of anterior spine wedging and so of thoracic kyphosis.
1) Fon GT, et al. AJR 1980 ; 134:979-983
ACCURACY IN WEIGHT MEASUREMENT WITH HOLOGIC QDR 4500 ACCLAIM
B. Sutter*, O. Legrand, E. Meys, F. Bougon, P. Hardouin
Institut Calot, Berck/Mer, France
Aim of the study: dual-energy X-ray absorptiometry (DXA) is selected as a reference method for both assessing body composition and measuring the changes in body composition. Nevertheless, numerous studies have raised errors in measurement depending on hydratation, software, and hardware. We tested a validation method comparing scale weight (Sw) and DXA estimated weight (DXAw).
Material & method: we measured Sw and DXAw in 184 subjects (86 males and 98 females) aged 36.7±12.6 yrs (13.7 - 79.7) weighing 72.3±18.1 kg (30.4 - 146.2) with an HOLOGIC 4500 Acclaim(r), software V8.26a.
Results: difference between DXAw & Sw was minus 1.24±2.0 kg (p less than 0.001), ranging from +1.34 to minus 12.97 kg (when expressed as a percentage, minus 1.32±1.85%, ranging from +2.32% to minus 8.94%). Difference correlated closely (polynomial, order2: see figure) with scale weight [r2 0.869], Body Mass Index [r2 0.835), DXA derived fat [r2 0.772], and poorly with total lean [r2 0.209], and not with BMC, height, sex, or age.
Sw can be estimated from sDXA as:
Sw = 0.0015*[DXAw]2 + 0.8694*DXAw + 2.5738 [r2 0.998]
Conclusion: In clinical practice, fan-beam DXA underestimate weight, especially for patient over 75 kg. This error depends on total weight, BMI, and fat estimated weight, suggesting a fat effect. From our experience, a validation procedure is recommended for DXA when evaluating body composition especially for monitoring. Comparisons with scale weight establish a 1st step for this procedure.
DETERMINATION OF CA ISOTOPE RATIOS IN METABOLIC STUDIES USING SECTOR FIELD INDUCTIVELY COUPLED PLASMA MASS SPECTROMETER (HR-ICP-MS)
M. P. Field, M. Cifuentes, R. M. Sherrell, S. A. Shapses*
Rutgers University, New Brunswick, USA
The precise and accurate determination of Ca isotope ratios in biological samples is imperative in limiting the expense of enriched isotopes used in human metabolic tracer studies. The Ca isotope spectra obtained from an ICP-MS (using conventional spray chamber) operated in low resolution mode (R=300) is subject to numerous polyatomic, isobaric and doubly charged interferences. The most severe polyatomic interferences can be resolved at resolutions greater than 3500. Here we evaluate the ELEMENT (Finnigan MAT, Bremen, Germany), a HR-ICP-MS, for the rapid (50 samples/day) determination of Ca isotope ratios in 20 fold dilutions of raw urine and oxalate pre-concentrates. Settling the magnet at 42Ca, and electrostatically scanning the 42 to 48 mass range, optimizes scanning speed for the determination of precise isotope ratios. Good precision (<±0.5%, 1-sigma) and accuracy (±0.5%) for all Ca isotope ratios are obtained at medium resolution (4300), sufficient to resolve all polyatomic interferences. Reducing water based interferences on 42Ca (ArHH and MgO) and 44Ca (CO2 and SiO) by sample desolvation (MCN-6000, CETAC, Omaha, NE) provides excellent precision (±0.06% 1-sigma) and accuracy (±0.1%) for 42/44 and 43/44 in low resolution (R=300). Furthermore removal of matrix by Ca oxalate pre-concentration yields similar precision for 42/44 and 43/44 and reduces interferences previously prohibiting the determination of 46/44 and 48/44 ratios. This new analytical method is applied to our on-going clinical trials where 0.017 mg/kg of 42Ca (IV) and 0.012 mg/kg 43Ca (oral) is administered to postmenopausal women. Our preliminary results in 11 women (61±7 years, body mass index of 28.1±1.5 kg/m2) show a baseline fractional absorption rate of 0.264±0.070. The percent urinary enrichments at 24h from the IV and oral doses are 3.01±0.71 and 3.25±0.73, respectively. In summary, to our knowledge these are the most precise Ca isotope ratios in diluted urine or oxalate pre-concentrate that have been reported using a single collector ICP-MS; the rapid analysis makes this especially suitable for larger studies of calcium absorption and kinetics.
TRABECULAR BONE VOLUME AND BONE MINERAL DENSITY CORRELATION IN LUMBAR SPINE
O. Cvijanovic1*, Z. Crncevic-Orlic2, I. Kristofic1, J. Spanjol1, S. Zoricic1, I. Maric1, D. Bobinac1
1Department of Anatomy, University of Rijeka, Rijeka, Croatia
2Department of Endocrinology, Clinical Hospital Rijeka, Rijeka, Croatia
The objective was to determine correlation between trabecular bone volume (BV/TV) and bone mineral density/content(BMD, BMC) in third lumbar vertebrae (L3) obtained from 45 normal autopsy cases aged from 40 to 80 years (23 females and 22 males) divided in 4 separate age and sex related groups (ranged 40-60 and 60-80 yrs). Vertebrae were analyzed by dual-energy X-ray absorptiometry (DEXA) and histomorphometry. For the histomorphometric analysis vertical trabecular bone cylinders 7 mm thick obtained by drilling the central part of the vertebra were fixed in PFA, embedded undecalcified in MMA, cut into 7 microm thick sections, and stained with Goldner trichrome. Estimates of trabecular bone volume (BV/TV) were assesed by using an image analyzer (VAMS, Issa, Zagreb). BMD and BMC measurements in different age and sex-related groups were compared with BV/TV values measured in the same groups. Our results revealed that BMD and BV/TV correlation was strong in a group of women aged from 60 to 80 years (r= 0.9, p<0.05) and in the same age group of men (r=0.99, p<0.05). BMC and BV/TV correlation in above mentioned groups was (r= 0.92, p<0.05) and (r= 0.91, p< 0.05), respectively. In a both groups of younger women and men a high correlation between BMD and BV/TV was determined (r=0.99, p<0.05) and (r=0.86, p<0.05), respectively. In the same groups BMC and BV/TV correlated poorly (r=0.42, p<0.05) and (r=0.54, p<0.05), respectively. Moreover, BMD and BMC were found significantly higher in younger men and women when compared to elderly (p<0.001). In conclusion, there is a strong correlation between bone mineral density /content and trabecular bone volume, predominantly in elderly of men and women.
INTRAINDIVIDUAL COEFFICIENT OF VARIATION OF LUMBAR SPINE DENSITY AS AN INDEX OF SPONDYLOTIC DEFORMITY
T. Fujita1,2*, Y. Fujii1, A. Miyauchi3, Y. Takagi3
1Calcium Research Institute
2Katsuragi Hospital
3National Sanatorium Hyogo Chuo Hospital
The severity of spondylosis deformans, as widely seen as osteoporosis after middle age, giving rise to severe backache seriously interfering with the quality of life, has been difficult to evaluate quantitatively. Deformity of the spine characterizing spondylosis deformans including osteophyte formation and hyperostosis may give artifactually high values of bone mineral density (BMD) measured by dual energy absorptiometry (DXA) in anterior-posterior direction. Thus the intraindividual coefficient of variation (CD) among L1-L4 rises parallel to the degree of spondylotic deformity. Young subjects without spondylotic changes gave a mean CD of 3-7%, whereas it was 10-30% or more in patients with spondylosis deformans. The degree of spondylotic deformity in the plain lumbar spine X-ray picture showed a highly significant correlation with the CD of L1-L4 BMD. In a 27 months randomized, double blind, prospective study at Katsuragi Hospital to compare effects of 900 mg Ca supplement as Active Absorbable Algal Calcium (AAACa or Advacal) and calcium carbonate with placebo in elderly women with a mean age of 80 years, a significant increase of lumbar spine BMD over that on placebo was noted on AAACa but not on calcium carbonate (Calcif Tissue Int 58:226-230, 1996). The results of DXA measurement was reevaluated to calculate CD of L1-L4 BMD to test the effect of these agents on spondylotic deformity. In the 18th month CD was significantly decreased on AAACa from that on placebo (p=0.0346) and in the 27th month, CD on AAACa was lower than that on placebo (p=0.0052) and that on calcium carbonate (p=0.0193) by Fisher's PLSD. AAACa apparently reversed the spondylotic deformity along with increasing BMD. Suppression of PTH secretion and bone resorption probably inhibits calcium entrance into cartilage and susequent degeneration and disappearance reversing the advance of degenerative joint disease. This is the first attempt to quantify the spondylotic changes for clinical monitoring.
BONE MINERAL DENSITY AND BODY COMPOSITION ASSESSMENT IN CLINICAL PRACTICE
T. O. Chernova
Russian Endocrinology Research Centre, Moscow
During the past decades the implementation of bone densitometry in clinical practice made it possible to assess BMD in health and disease and to monitor the treatment.
PURPOSES: 1) To assess the clinical value of dual-femur acquisition in comparison with single-femur acquisition and the clinical value of spine densitometry in cases of bone deformities; 2) To assess the clinical significance of T- and Z-Scores in comparing the clinical data of obese and normal patients; 3) To assess the perspectives of lateral morphometry in clinical practice; 4) To find the new modalities for clinical usage of body composition data.
METHODS: About 6000 patients were examined over the period of 4 years. We assessed all parameters of BMD and body composition using DEXA (Lunar Expert XL, USA).
CONCLUSIONS: 1) There were clinically insignificant differences between dominant and non-dominant proximal femur DEXA results. Dual-femur assessment is very important in special cases of problems with one femur or spine deformities and in cases of early diagnostics of osteopenia among risk population. The obtained data show that in cases of kyphosis, scoliosis, etc the spine acquisition has no clinical significance and is not acceptable for the therapy monitoring; 2) T-Score is important parameter, but Z-Score is more important in assessment of obese patients; 3) Lateral Morphometry (LM) is rather difficult for performance and assessment, it is necessary to assess it simultaneously with x-ray examinations or CT and LM may be useful for scientific research, but its usage for clinical practice is limited; 4) Body composition (BC) is a perspective research in the assessment and therapy monitoring of obesity along with the possible future implementation in the research of aging, insulin resistance and meno- and andropause treatment.
LIMITATIONS OF PROXIMAL FEMUR AGE-ADJUSTED BONE MINERAL DENSITY (BMD) IN WOMEN WITH VERTEBRAL FRACTURES
A. Mylonakis*, D. Hadjidakis, P. Katsavochristos, K. Papaxoinis, M. Sfakianakis, S. A. Raptis
2nd Department of Internal Medicine-Propaedeutic, Research Institute and Diabetes Center, Athens University, 'Evangelismos' and 'Evgenidion' Hospitals, Athens, Greece
Recently a lot of concern has been raised about the appropriate interpretation of bone mineral density (BMD) measurements in persons with vertebral osteoporotic fractures. In order to investigate whether the proximal femur absorptiometry may offer misleading evidence, 123 postmenopausal women, were classified according to their age-adjusted (z-score) femoral neck (FN) BMD status [low (<-1.5) or normal (>-0.5)], and divided into 4 groups: group LN (low BMD without vertebral fractures), n= 29, age= 60.7±7.6 years (mean±1SD), BMI= 22.8±2.3 kg/m², group LF (low BMD with at least 1 vertebral fracture), n= 13, 63.2±11.2 years, 24.2±2.5 kg/m², group HN (normal BMD without any fracture), n= 72, 66.8±6.5 years, 28.1±3.9 kg/m² and group HF (normal BMD with at least 1 vertebral fracture), n= 25, 70.9±4.9 years, 28.6±3.9 kg/m². None of the women suffered from any disease or had received any medication affecting bone metabolism. BMD measurements were performed at the FN, Ward’s triangle (WD) and trochanter (TR). Thirty one percent of women with low BMD and 26% with normal BMD suffered from at least 1 vertebral fracture. HF had significantly higher age and years since menopause (YSM) than HN (p<0.05). LF had significantly higher YSM and BMI than LN (p<0.05). Absolute BMD values of FN and WD were significantly lower in HF than in HN (p<0.05). However z-scores at any anatomic area didn’t differ significantly between the same groups. Neither BMD values nor z-scores presented any significant difference between LN and LF in any area. BMD of FN and WD was negatively correlated to age at menopause (AAM) in LN and LF (r= -0.48 to -0.65, p<0.05 to 0.01) but not in HN and HF. CONCLUSIONS: Age and interval since menopause seem to participate to fracture susceptibility. Later menopausal age seems to be associated with lower BMD values of FN and WD only in women with osteopenia. Since FN z-score values did not differ between groups with and without fractures, caution should be exercised in the interpretation of femur age-adjusted densitometry and spine x-rays should always be requested irrespective of age.
TOTAL HIP OR FEMORAL NECK? DIFFERENCES IN WHO CLASSIFICATION IN POSTMENOPAUSAL WOMEN
J. L. Mansur
Centro de Endocrinologia y Osteoporosis La Plata, Argentina
Hologic and the last versions of Lunar equipments (IQ, Prodigy) measure the bone mineral density (BMD) of "Total Hip". This region includes femoral neck, trochanter and shaft (or intertrochanter). The ICSBM recommends to use standardized BMD of total hip (TH) instead of femoral neck (FN), since TH correlates with hip fracture in a similar way as FN, but with better precision (as the area of TH is bigger). Actually half of the abstracts presented (ASBMR 2000) are based in TH and the other half in FN, and NOF recommendations use the word "hip" or "proximal femur". Objective: To study characteristics of TH and to know if the WHO suggested classification differs if we choose TH or FN. Patients and methods: 300 postmenopausal women without HRT or treatment were measured with Lunar DPX (software IQ). CV:TH=0.9%, FN=1.5%. Results: TH is the addition of 3 regions that measure areas of different sizes: 1) Shaft: 14.30 cm(SD:1.1)=44.5%; 2) Troch: 13.03 cm(SD: 2.1)=40.6%; 3) FN: 4.78 cm(SD: 0.4)=14.9%; (In Hologic systems the size of shaft is bigger and troch is smaller) Total hip: 32.10 cm(SD: 2.6)=100%. Ward area is included in FN and Troch. We find that 84.8% of the patients were in the same category (WHO classification), discordance:15.2%. BMD was lower in FN in 10.3% and in TH in 4.8% of the patients. Correlation R of TH: vs FN: 0.93; vs Troc: 0.96; vs Shaft: 0.98; vs Ward: 0.91. Discussion: It is amazing that people use nearly as synonymous the regions FN and TH (or as NOF recommendations "hip" or "femur proximal") when there are regions with a different proportion of trabecular and cortical bone. As when we study a patient the objective is to know the risk of fracture, we propose to observe the T-Score of both regions and to inform the worst. Since 10.3% are in a better category if we use total hip, we suggest not to use this region as a routine.
DETERMINATION OF BONE MINERAL DENSITY OF LIMBS AND PELVIC IN NORMAL SUBJECTS OF HENAN PROVINCE
Z. C. Liu*, X. H. Wang
Henan Electric Hospital, Zheng Zhou 450052, China
Abstract To evaluate bone mineral density (BMD) of normal subjects in Henan province of China. BMD of upper and lower limbs and pelvic in 417 healthy subjects was measured by Hologic 4500W dual energy X-ray absorptiometry.
Results showed that the peak bone values of BMD were in group, aged 30-39 years of both sexes. Then the BMD diminished with aging, especially for females; BMD was higher in male than in female of any age group; BMD was lower in left upper limb than in right upper limb; BMD was similar in both legs. We conclusion that this study provide useful reference data for prevention and cure of osteoporosis in this region.
P329 T
Withdrawn
NURSE-LED PRIMARY CARE BASED OSTEOPOROSIS SCREENING OF POST-MENOPAUSAL WOMEN USING PERIPHERAL BONE DENSITOMETRY AND RISK ASSESSMENT QUESTIONNAIRE
A. H. Franklin-Stevens1*, L. Cheng2, M. Froome2, J. Taylor1, M. E. Shipley1, J. C. Chamberlain1, H. Fitz-Clarence1
1Osteoporosis Unit, Centre for Rheumatology, UCLH, London, UK
2Riomed Ltd., Southampton, UK
The effectiveness of population based osteoporosis screening in primary care in relation to cost and benefit remains a controversial area. However, utilising peripheral densitometry and risk assessment to identify those at high risk of future osteoporotic fracture may be more beneficial. This study therefore aims to determine the effectiveness of osteoporosis screening of post-menopausal women in primary care.
2,558 women aged between 60-80 years from five general practice databases in North London were sent postal invitations to attend for osteoporosis screening at their surgery. Subjects attending either the walk-in or appointment only clinic completed a nurse administered risk assessment questionnaire and had os calcis BMD measurement using peripheral densitometer DXT200 LUNAR (T- score more than -0.6 SD = normal, T score between -0.6 and -1.6 SD = osteopaenia, and a T score less than -1.6 SD = osteoporosis). Patients within the osteopaenic or osteoporotic range subsequently received lifestyle advice. All results were forwarded to their GPs.
816 women (31%) attended for screening. 173 women (21.2%) were identified as osteoporotic, 248 (30.4%) osteopaenic, and 395 (48.4%) normal. Among the osteopaenic and osteoporotic patients, 33% have had previous low trauma fracture, 20% smoked, 6% had low BMI and 4.2% had prolonged use of oral steroids (more than 7.5mg daily, for over three months). Other potential risk factors did not reach significant numbers.
Targeted primary care based osteoporosis screening (in a multi-ethnic London population), using both peripheral densitometry and risk assessment questionnaire is of benefit. It revealed a significant proportion of patients (52%) as having osteoporosis or osteopaenia. This study therefore offered an effective strategy for identifying 'high fracture- risk' patients actively requiring preventive and therapeutic treatment, who in our opinion, could otherwise have been missed. Moreover, it played an educational role in increasing osteoporosis awareness amongst patients and primary care physicians and health professionals.
A much larger primary care group investigation to enable multivariate risk factor and BMD analysis is in progress.
FRACTURE CASE HISTORY AT OSTEOPOROSIS, RELATION TO CURRENT RESULTS IN THE QCT AND PQCT
K. Abendroth
Regional Experts Group for Osteoporosis in Saxony and Thuringia, Jena, Germany
The Regional Experts Group for Osteoporosis (REGO)in Saxony & Thuringia has introduced our osteoporosis risk questionnaires in Tampere 2000 (Calcif Tissue Int 2000, 66, suppl 1: S110, P-230/231). The evaluation of 1000 questionnaires of patients from the osteoporosis diagnostics (no population screening) yielded a special weighting of the information of fractures after the 40th year of life into reference to the densitometric results. An assignment of the 14 different risk answers of the questionnaire to normal densitometric results or to osteoporosis succeeded most clearly with this bone fracture case history. Therefore to this reconciles some interesting results. Details on 5 fractures typical of the osteoporosis were analysed. Forearm, upper arm, rib, femoral neck and vertabral body fractures were included. In the analysed group of 1000 patients two thirds of the densitometric results were examined at this with the QCT technology and 1/3 with the pQCT technology. At pQCT technology the measurements of total BMD and trabecular BMD were judged to be separated. How frequent a corresponding fracture case history is to the results of a densitometric definite osteoporosis connectedly. The table 1 gives the answer on this question.
The fracture case history has a various relevance for the assignment of the results of the three densitometric measurement procedures. The QCT measuring is proceeded the most accurate at this. At the pQCT measuring the determination of the complete density of the ultra distal radius has the better assignment to the fracture case history as the trabecular bone density. These statements concern the fracture risk at osteoporosis, not the densitometric osteoporosis diagnostic to himself.
| Table 1: The percentage of the
corresponding fracture (fx) case history which is connected to a current densitometric result according to an osteoporosis is indicated. |
|||||
| forearm fx |
upper arm fx |
rib fx |
fem.neck fx |
vertebral fx |
|
| QCT-trabecular BMD | 84 | 81 | 80 | 87 | 87 |
| pQCT-total BMD | 80 | 84 | 93 | 85 | 67 |
| pQCT-trabecul. BMD | 66 | 61 | 67 | 73 | 42 |
COMPARISON OF PA AND LATERAL LUMBAR SPINE BONE DENSITY MEASUREMENTS FOR DISCRIMINATION OF WOMEN WITH VERTEBRAL FRACTURES
L. Ferrar*, R. Eastell
University of Sheffield, Sheffield, UK
The Hologic Acclaim fan-beam densitometer has the advantage that PA and lateral dual-energy x-ray absorptiometry (DXA) scans of the lumbar spine may be acquired without repositioning of the patient. Lateral bone mineral density (BMD) measurements may have better diagnostic sensitivity compared to postero-anterior (PA) measurements, but this approach has not been evaluated using the Hologic Acclaim. Our aims were to evaluate PA, lateral and lateral volumetric bone density at the lumbar spine for discrimination of women with vertebral fractures using 3 approaches. These were comparison of Z scores, comparison of receiver operating characteristic area under the curve (AUC), and logistic regression analysis.
We acquired paired PA and supine lateral lumbar spine DXA scans (Hologic Acclaim) in 188 postmenopausal women ages 50 to 88 (mean 65±8 years). Of these, 59 (mean age 70±8 years) had radiologically-diagnosed vertebral fractures. Normal BMD values for vertebrae L2-3 were calculated for the remaining 129 women with no fractures (mean age 63±8 years), after adjusting for age and weight.
In women with vertebral fractures, there were no significant differences in mean BMD Z scores for PA (Z = -1.20, SE = 0.10, 95% CI = -1.39, -0.98 g/cm2), lateral (Z = -0.98, SE = 0.09, 95% CI = -1.17, -0.79 g/cm2), or lateral volumetric (Z = -0.94, SE = 0.12, 95% CI = -1.17, -0.71 g/cm3). There were also no significant differences in AUC for PA (AUC = 0.83, SE = 0.03, 95% CI = 0.73, 0.85), lateral (AUC = 0.82, SE = 0.03, 95% CI = 0.76, 0.87), or lateral volumetric (AUC = 0.80, SE = 0.04, 95% CI = 0.73, 0.85). From logistic regression analysis, the increased risk of vertebral fracture per 1 SD decrease in BMD was 4.2 (95% CI = 2.6, 6.8) for PA, 4.6 (95% CI = 2.8, 7.6) for lateral and 3.6 (95% CI = 2.3, 5.6) for lateral volumetric.
Based on our results using these 3 statistical approaches, we found no evidence that lateral or lateral volumetric lumbar spine BMD measurements provide improved diagnostic discrimination.
P333 S
Withdrawn
BONE MASS ASSESSMENT BY DIGITAL X-RAY RADIOGRAMMETRY IN SPANISH POPULATION. COMPARISON WITH DUAL X-RAY ABORPTIOMETRY AND QUANTITATIVE ULTRASOUND
C. Gómez1*, L. Casado2, J. C. Barredo3, A. Iglesias3, V. Valencia4, M. Naves1, J. L. Fernández1, J. B. Díaz1, J. B. Cannata1
1Bone and Mineral Research Unit, Instituto Reina Sofía de Investigación, Hospital Central de Asturias, Oviedo, Spain
2Health Centre of Vallobín
3Health Centre of Ventanielles
4Health Centre of Castrillón, Asturias, Spain
Digital X-ray Radiogrammetry (DXR) estimates bone mass using an average of cortical and bone width measurements at the radius, ulna and metacarpal region by digital analysis of a X ray film of the hand and forearm.
The aim of this study was to establish 1) the Spanish population database for this technique, 2) to compare it with other populations, c) to investigate its performance compared with Quantitative Ultrasound (QUS) and Dual X-ray absorptiometry (DXA) in peripheral and axial bone.
A population-based sample of 230 people, aged 20 to 80 years old, of both sexes were enrolled. We performed DXR (X-Posure System, Pronosco) of the non dominant hand, QUS (Sahara, Hologic) of the non dominant calcaneus and DXA at the lumbar spine and right hip. The "in vivo" precision, expressed as CV was 0.9% for DXR, 3.4% for QUS and 1.05 for DXA. Peak bone mass occurred in the fourth decade in both sexes (0.626±0.019 g/cm2 in men and 0.569±0.037 g/cm2 in women), the female population values were significant lower (-0.5 to -1.2 SD Z-Score) respect to previous data published in USA and German women. No reference data of men were found. DXR-BMD was correlated better with total hip DXA-BMD (r=0.54 and r=0.64; p<0.05, in women and men respectively) than with calcaneus BUA or SOS (r=0.3 to 0.6; p<0.05, in both sexes). However, we found differences in the performance of the three techniques. Whereas hip DXA was technically correct in the 100% of the subjects, the DXR was possible in 90% of women and in less than 70% of older men. The QUS was optimal in around 90% of women and 75% of men, having more problems in older men. In conclusion, the Spanish DXR-BMD values for women was lower that other Caucasian population, its use was technically more difficult in men.
HIGH SHORT-TERM IN-VIVO PRECISION OF THE PRONOSCO X-POSURE SYSTEMTM ON HAND X-RAYS
A. B. Helboe1*, K. Juul1, A. Rosholm1, P. B. Andersen2, N. H. Bjarnason3
1Pronosco A/S, Vedbaek, Denmark
2CCBR, Aalborg, Denmark
3CCBR, Ballerup, Denmark
The precision of a method for measurement of bone mineral density (BMD) is important when interpreting repeated measurements over time, e.g. to monitor postmenopausal bone loss or follow the effect of therapy. The Pronosco X-posure System is a bone densitometer using the Digital X-ray Radiogrammetry (DXR) technology, and the BMD estimate generated by the system is referred to as DXR-BMD. Version 2 of the system uses a plain radiograph covering the hand, and is identical to Version 1 with exception of the regions of interest (ROI) used for the DXR-BMD calculation. Version 2 uses a subset of the original ROIs – including the second through fourth metacarpal – and it provides the DXR-BMD as well as metacarpal index (MCI), cortical thickness and bone width as output.
For version 1, the CV% was 0.65% (0.58%-0.75%) for the whole population.
In the present study the short-term precision of the Pronosco X-posure System Version 2 is investigated. The study used a balanced design with each participant having three radiographs taken with repositioning of the forearm in between each capture. The study was carried out in two populations: 20 pre-menopausal women between the ages of 30 and 40 and 20 postmenopausal women above the age of 64. The short-term in-vivo precision errors were calculated as root-mean-square averages of standard deviations of repeated measurements and expressed as the coefficient of variation (CV) with the corresponding 90% confidence intervals. The results for Version 2 are presented in the table below.
The short-term in-vivo precision error of the Pronosco X-posure System Version 2 has been shown to give very acceptable results in both pre- and postmenopausal women and compared to Version 1, the short-term in-vivo precision error of the Pronosco X-posure System Version 2 seems optimized.
| DXR-BMD and MCI Results | ||
| Pre-menopausal | Post- menopausal | |
| N | 20 | 20 |
| Mean age (years) | 35.2 | 68.2 |
| Mean DXR-BMD (g/cm2) | 0.598 | 0.499 |
| DXR-BMD CV (%) | 0.28 (0.22- 0.34) | 0.44 (0.36-0.51) |
| MCI CV (%) | 0.26 (0.21- 0.32) | 0.46 (0.37-0.56) |
DIGITAL X-RAY RADIOGRAMMETRY ON HAND X-RAYS OBTAINED ON MAMMOGRAPHIC X-RAY EQUIPMENT
M. Friis1*, N. Skovgaard1, K. Juul2, D. S. Nissen2
1Department for Diagnostic Radiology and Ultrasound, Glostrup Hospital, Denmark
2Clinical Research and Development, Pronosco, Denmark
Digital X-Ray Radiogrammetry (DXR) is a new technique using automated image analysis of digitized X-rays of the hand and thereby calculating DXR-BMD using information from the three middle metacarpal bones.
The Pronosco X-posure System™ V.2 MAMMO is a new system based on the DXR technology which uses hand X-rays acquired on mammographic X-ray equipment (X-rays).
In the present study the Pronosco X-posure System™ V.2 MAMMO was compared to the established Pronosco X-posure V.2 RAD system which uses hand X-rays acquired on conventional X-ray equipment (conventional X-rays). Moreover the short-term in vivo precision error of the Pronosco X-posure System™ V.2 MAMMO was estimated.
The study used an open cross sectional design consisting of an equivalence part, a short-term in vivo precision part and a clinical performance part. One hundred and eight women participated in the equivalence part having one conventional X-ray and one X-ray taken. Thirty postmenopausal women participated in the in vivo precision part having 3 X-rays taken with repositioning of the hand in between each capture. The clinical performance part consisted of the women included in the equivalence part and the in vivo precision part plus 10 women having one X-ray taken. In total 148 women between the age of 20-79 years participated in the study.
The results of the equivalence study demonstrated a high degree of correlation (r = 0.995) between DXR-BMDV.2 MAMMO and DXR-BMDV.2 RAD measurements.
The short-term in vivo precision error of the Pronosco X-posure System™ V.2 MAMMO was found to give very acceptable results with a CV% for DXR-BMDV.2 MAMMO of 0.36%.
All X-rays were acquired at constant 14 mAs irrespective of the thickness of the womens hands. Since the system was able to calculate DXR-BMD values for all 213 X-rays acquired during the study the clinical performance of the Pronosco X-posure System™ V.2 MAMMO was found very satisfactorily.
DIGITAL X-RAY RADIOGRAMMETRY ON HAND X-RAYS
N. Baadegaard*, R. Linde, O. Wendt, A. Rosholm
Pronosco A/S, Vaedbaek, Denmark
Based on a plain radiograph of the hand the Pronosco X-posure System™ estimates BMD of the distal forearm by a combined computerized radiogrammetric analysis. The system is able to provide values for BMD, metacarpal index (MCI) and porosity. This presentation describes the methodology of DXR-BMD and results obtained on DXR-BMD in technical sensitivity and reproducibility studies using the Pronosco X-posure System™ version 2 RAD, which was approved by the FDA in October 2000.
A series of image analysis modules find the rough location of the three middle metacarpals by template matching algorithms, and subsequently the regions of interest (ROIs) are selected. The ROIs are selected in a coupled fashion by sliding three boxes up and down on the three metacarpal shafts until the overall narrowest location is found. In each ROI the outer and inner edge of the cortex are determined. The average cortical thickness ti and bone width Wi are determined for each metacarpal i and the bone volume per area is computed using: VPAi = pi * ti * (1 – ti / Wi). A combined VPA is obtained as a weighted average of the individual VPA's and the DXR-BMD estimate is computed as DXR-BMD= c * (1 – P)* VPA. (P is the estimated three-dimensional porosity and c is a scaling factor).
The robustness of the Pronosco X-posure System™ version 2 RAD has been tested in two comprehensive technical studies. A study using anthropomorphic phantoms investigated the sensitivity of the BMD measuring algorithm to changes in the settings and conditions of the conventional X-ray equipment. The study concluded that realistic deviations from a standard protocol for the capturing conditions did not influence the estimated BMD value. In a reproducibility study the BMD measurement between sites and operators were investigated. It was found that there was a difference of 2.8 mg/cm2 (0.5% of the mean BMD value of young adult women) when performing two repeated measurements on two arbitrary different X-ray installations. The difference using a single installation and one operator was 1.9 mg/cm2 (0.3% of the mean BMD value of young adult women). Thus, successive measurements on the same patient may as well be performed at different X-ray installations since the main variation on the BMD assessment is due to variation of operators.
DIFFERENT PERFORMANCES OF BONE DENSITOMETERS IN THE REGION OF NORTH HOLLAND
K. P. Staal1*, J. C. Roos2, R. A. Manoliu3, P. J. Kostense4, P. Lips1,4
1Department of Endocrinology, Academic Hospital, Vrije Universiteit, Amsterdam, The Netherlands
2Department of Nuclear Medicine, Academic Hospital, Vrije Universiteit, Amsterdam, The Netherlands
3Department of Radiology, Academic Hospital, Vrije Universiteit, Amsterdam, The Netherlands
4Institute for Research in Extramural Medicine (EMGO-Institute), Vrije Universiteit, Amsterdam, The Netherlands
Since the introduction of dual X-ray absorptiometry (DXA) to measure the bone mineral density the differences between the machines have been recognised as a difficult problem. Differences in equipment, calibration and reference populations cause differences in diagnosis of osteoporosis. We tested in the region North Holland, including Amsterdam, all the available DXA-machines (1 Norland, 2 Lunar DPX, 2 Lunar Experts, 1 Hologic 1000, 3 Hologic 2000 and 8 Hologic 4500). Two anthropomorphic phantoms (Hologic lumbar spine and hip phantom) and 1 semi-anthropomorphic geometrically defined phantom (European Spine Phantom (ESP)) were used. Every phantom was measured 5 times at baseline and after 2 months. Between every measurement the phantoms have been replaced. The results show that there was no significant drift after a 2 months period. Also as known by different studies we found that the Lunar machines measured higher values in comparison with other brands. The coefficients of variation over repeated measurements were up to 1.0% except in the femoral neck (between 1.6 and 4.6%) and in the high density region of ESP (between 0.9 and 3.7%). Interpretation of the measured values by the software of the tested machines expressed in T-scores show more variation, in the lumbar spine up to 0.8 SD for men and 0.9 SD for women and in the total hip region up to 1.5 SD for men and 0.8 SD for women. The variations in T-scores are mainly caused by the different reference populations installed on the tested DXA-machines. There were in our group 5 different reference populations for the lumbar spine as well for the hip used. The observed differences in T-scores may cause a shift in diagnosis from normal to osteopenic and from osteopenic to osteoporosis from 10-20% between the different hospitals. Standardisation of reference populations is of upmost importance to achieve more uniformity in bonedensitometry measurements and diagnosis of osteoporosis in different hospitals.
ROLE OF QUANTITATIVE EXAMINATION OF BONES (SUNLIGHT OMNISENSE) IN DISCRIMINATION OF WOMEN WITH LOW FRACTURE-RISK DESPITE LOW BMD (DEXA)
L. Papierska*, W. Misiorowski, S. Zgliczyñski
Medical Center for Postgraduate Medicine, Department of Endocrinology, Warsaw, Poland
We have performed the ultrasound examination (Sunlight Omnisense) of radius and proximal phalanx III in 40 women with a low mineral bone density found in DEXA (L2/L4 T-score<-2.5 and distal radius T-score<-3) but without any fractures despite the serious traumas. All women were postmenopausal, but any other risk factors for osteoporosis were not found. We have found significantly higher SOS values in examined group than in fractured controls.
Our finding indicates the role of ultrasound examination of bones as additional factor in predicting the risk of fractures.
AGE- AND GENDER-RELATED CHANGES IN THE HUMAN SPINE AND ILIAC CREST ASSESSED BY STATIC HISTOMORPHOMETRY
J. S. Thomsen*, E. N. Ebbesen, L. Mosekilde
Department of Cell Biology, University of Aarhus, Aarhus, Denmark
The aims of the study were to assess age- and gender-related changes in the human spine and the iliac crest by use of established static histomorphometry, and to correlate the histomorphometric measures between these two skeletal sites.
The material comprised normal human lumbar vertebral bodies (L2) from 24 females (19- 96 years) and 24 males (23-95 years). Standard 7 mm transiliac crest bone biopsies were also obtained from these individuals. 9-mm-thick medio-lateral slices from L2 and the whole iliac crest bone biopsies were embedded in methylmetacrylate, stained with aniline blue, and scanned at a high resolution into a computer. With a computer program the following static histomorphometric measures were determined: trabecular bone volume; marrow and bone space star volume; anisotropy; node-strut analysis; trabecular thickness, trabecular number; trabecular separation; and trabecular bone pattern factor. Furthermore, the connectivity density of the trabecular network was determined.
In the vertebrae, the BV/TV decreased from 15% to 6% from 20 to 90 years of age, which is identical to the results we have previously found on a smaller non-gender-specific material. The other static histomorphometric parameters also confirmed the previously found values. In this larger material it was possible to compare the results from the two genders. In the vertebral bodies we found no gender related differences in any of the static histomorphometric measures. In the iliac crest, the BV/TV decreased from 30% to 10% from 20 to 90 years of age. At this skeletal site there was a gender related difference in anisotropy and in connectivity density. Concerning these two parameters, the data from males were uncorrelated with age. In contrast, the anisotropy increased significantly with age for females. Furthermore, it was found that the connectivity density of younger females was higher than for males and declined significantly with age. In general there was a correlation between data from the vertebral body and the iliac crest, although the correlations were not strong.
In conclusion, there were no gender related differences in the static histomorphometric parameters in the vertebral body. In the iliac crest only anisotropy and connectivity density showed gender related differences.
AGE-RELATED CHANGES IN SPINAL CANCELLOUS MICROARCHITECTURE ARE ZONE-SPECIFIC
L. Mosekilde*, E. N. Ebbesen, J. S. Thomsen
Department of Cell Biology, University of Aarhus, Aarhus, Denmark
We have recently shown that there is an age-related decrease in BV/TV in the human spine. Our observations were based on analyses on whole vertical sections (half a vertebral body). However, inspection of the sections clearly shows that there are three discrete zones in the vertebral body: one central zone and one zone under each of the endplates. The aim of the present study was to quantify the age-related changes in the static histomorphometric parameters in these zones.
The material comprised normal human lumbar vertebral bodies (L2) from 24 females (19- 96 years) and 24 males (23-95 years). 9-mm-thick vertical slices from L2 were embedded in methylmetacrylate, stained with aniline blue, and scanned at high resolution into a computer. With a computer program, the vertebral sections were automatically subdivided into 3 zones of equal height with parallel demarcation lines. Due to symmetry, the two sub-endplate zones were considered together in the analyses. With a computer program, the following histomorphometric measures were determined: trabecular bone volume; marrow and bone space star volume; anisotropy; node-strut analysis; parallel plate analysis; and trabecular bone pattern factor. Data from males and females were pooled in this analysis, as we have previously shown that there are no gender specific differences in the histomorphometric parameters at this site.
BV/TV in the sub-endplate zones was higher than in the central zone in the younger individuals (17% versus 12%), but the age-related decline was larger than in the central zone. Consequently BV/TV became similar for the two zones with age (6%). Similar age- related changes were found for trabecular number. Marrow space star volume showed lower values for the endplate zones than for the central zone for young individuals, but the values increased more rapidly, so no difference between zones was seen in elderly individuals.
In conclusion, the histomorphometric analyses confirm the visual impression that there are three distinct zones in the vertebral bodies in younger individuals and that this difference gradually diminishes with increasing age. Furthermore, the loss of bone density is significantly greater in the sub-endplate zones than in the central zone.
THE NATURE OF CHANGES IN VERTEBRAL BODY TRABECULAR BONE IN OSTEOPOROSIS
A. Boyde
University College, London, UK
Background: Microscopic images of porous bone are used both to study and to sell osteoporosis (OP). Many are figments of imagination. Materials and Methods: unembedded thick, plane-parallel slices of lumbar vertebral bodies, all cells removed, or osteoid removed to expose mineralisation front. Unembedded cancellous bone samples have great 3D depth and SEM is therefore the preferred imaging box. The aim of the present work was to enhance information content using separate recording from multiple backscattered electron (BSE) detectors in combination with motion parallax. Each image field is documented with at least four scans with as many separate detectors. Data is interpolated to emulate acceptance angles into 8, 16 or 32 etc. virtual detectors. Playing resultant images in sequence continuously sweeps the apparent direction of illumination and powerfully increases depth interpretation. Further enhancement is achieved through spectral colour coding of the direction of illumination allowing wider sectors of total data sets to be exploited simultaneously. In addition, the sample may be moved systematically such that image sequence replay emulates linear motion parallax from ‘flying past’ the surface. Findings: In 2mm thick slices of normal and 4mm thick slices of porotic vertebral trabecular bone studied from both sides, SEM inspection can cover 80% of surfaces in the clearest context: contrast this with minuscule fractions of total tissue surface in routine LM histomorphometry, which are also discontinuous and lack 3D context, yet engendered the ‘data base’ upon which most judgements about bone remodelling phenomena are founded. Conclusions: The activation-resorption-formation cum basic multicellular unit (ARF BMU) theory is widely defended against obvious exceptions which may disprove the rule in OP. Resorption with no subsequent repair is rife. That repair which does occur is not site coupled, and repair by microcallus formation can be generalised within bones with no external signs of collapse.
BONE RESORPTION IN POSTMENOPAUSAL WOMEN WITH NORMAL AND LOW BMD ASSESSED WITH BIOCHEMICAL MARKERS SPECIFIC FOR TELOPEPTIDE DERIVED DEGRADATION PRODUCTS OF COLLAGEN TYPE I
J. Y. Reginster1,2*, A. N. Taquet2, C. Christiansen3, E. Gamwell-Henriksen4, O. Bruyere2, J. Collette2, S. Christgau4
1WHO Collaborating Center for Public Health Aspects of Osteoarticular Disorders
2Bone and Cartilage Metabolism Unit, University of Liège, Liège, Belgium
3Center for Clinical and Basic Research, Ballerup, Denmark
4Osteometer BioTech, Herlev, Denmark
Biochemical markers of bone resorption can provide an estimate of the current state of bone turnover. Such a measurement can be used clinically to predict the risk of an individual to experience osteoporosis-related fractures (prognostic tool) and to assess the response of an osteoporotic patient to an antiresorptive therapy (monitoring tool). Our aim was to assess clinical relevance of four currently marketed biochemical markers of bone resorption, based on the measurement of degradation products derived from the telopeptides of type I collagen. In a cohort of 984 women, stratified for menopause, age and bone mineral density, the following markers were measured: urinary cross-linked N-telopeptide of type I collagen (NTx), and the level of breakdown products of type I collagen C-telopeptide in serum (S-CTx), and in urine by ELISA (U-CTx-E), and RIA (U-CTx-R). Furthermore the ratio (alpha/beta) between the alphaL form of CTx measured in the CTx RIA and the betaL form measured in the ELISA was calculated. The mean difference was calculated for each marker in women with osteopenia (Op) or osteoporosis (PMO) (WHO definition) compared to healthy premenopausal women (Pre) and to postmenopausal women with normal bone mass (N Post). Serum CTx but not urinary CTx or NTx was significantly elevated in N Post compared to Pre women. All marker values were significantly elevated in Op and PMO subjects compared both to Pre and N Post women. Compared to premenopausal values, the largest elevation in both Op and PMO women was observed for serum CTx. The alpha/beta CTx ratio was elevated in post- compared to pre-menopausal women, but there was no difference in the ratio between N Post, Op or PMO women. The results suggests that S-CTx, which was the marker with the highest elevation at menopause, may be best suited for monitoring antiresorptive therapy for prevention and/or treatment of postmenopausal osteoporosis which aims at returning bone turnover to premenopausal levels.
BONE TURNOVER MARKERS DO NOT IDENTIFY BONE STATUS IN POSTMENOPAUSAL WOMEN
M. Muñoz-Torres*, P. Mezquita-Raya, F. López-Rodríguez, F. Escobar-Jiménez
Bone Metabolic Unit, Endocrinology Division, University Hospital "San Cecilio", Granada, Spain
Postmenopausal osteoporosis is characterized by an uncoupling of the remodelling proccess, in which resorption is increased, that leads to bone loss. The evaluation of remodelling can be assessed by determining bone formation and resorption markers as a complement of the bone mineral density (BMD) measurements. However, in clinical settings utility of these markers of bone turnover in postmenopausal osteoporosis has not been well established.
AIM: To determine the performance of markers of bone formation and bone resorption in patients with postmenopausal osteoporosis in order to discriminate BMD status.
PATIENTS AND METHODS: we studied 160 women (61±7 yr). All were Caucasian, community-dwelling, ambulatory, in a good health except for osteoporosis and did not have secondary causes for low BMD or medications that might affect bone density. We determined serum levels bone specific alkaline phosphatase (b-ALP: Tandem-R OstaseTM, Hybritech), total alkaline phosphatase (t-ALP), osteocalcin (OC), tartrate resistant acid phosphatase (TRAP), urinary cross-linked C-telopeptide of collagen type I or -Crosslaps corrected by urinary creatinine (CTX: CrossLapsTM, Osteometer Biotech), and BMD at lumbar spine (LS) and femoral neck (FN) measured by DEXA (Hologic QDR1000).
RESULTS: using the WHO densitometric criteria, 51.2% of our population were osteoporotic. We did not find statistically significant differencies either in the formation or resorption indexes in comparison with the osteoporotic and normal BMD groups. There were no differencies when considering osteoporosis, osteopenia or normal BMD. No marker showed a significant correlation with BMD (Z-Score or gr/cm2), except CTX, that was weakly correlated with BMD at lumbar spine (r=-0.156; p=0.048).
CONCLUSIONS: Our data show an overlap in the results of clasical and recent biochemical markers of bone remodelling when comparing between osteoporotic and normal BMD postmenopausal women. In clinical practice, bone turnover markers do not allow the identification of patients with low bone mass.
SHOULD BE FEASIBLE TO EXCHANGE 2-H FASTING URINE AND 24-H URINE SAMPLES FOR DETERMINATION OF CLASSICAL RESORPTION BIOCHEMICAL MARKERS?
J. Ferrer-Cañabate*, I. Tovar, P. Martinez
Servicio AA.CC. H.U. Virgen Arrixaca, Murcia, Spain
Objective: We performed a comparative study on the equality of the determination of two classical biochemical markers of bone resorption in two different urine samples: 24-h (expressed as mg/24h) and 2-h fasting morning urine samples (expressed as urinary creatinine ratio (mg/mg creat)).
Material and methods: Fifty untreated physiologic or surgical postmenopausal women (age 48.9±5.3, range 30-62) were included. Samples were kept at -80ºC until use. Urinary calcium was assessed by photometrical colour test and creatinine by Jaffe's photometrical cinetic colour test, both of them in a Hitachi 747 Automatic Analyzer. OH-proline was assessed by an Hypronosticon(c) manual test. The statistical linear regression (y=ax+b) of the results were performed with the SPSS Statistical Software Program.
Results: See Table.
Discussion: There are not statistical linear regression between not only Ca in 2-h fasting urine and 24-h urine samples but neither OH-proline. So, we had better not to supply or exchange these specimens, because we would be running the risk that not to obtain similar results.
| a | (a)95% | b | (b)95% | R2 (p<0.05) | |
| Ca | 268.69 | 209.1-328.2 | 57.9 | -135.2-251.1 | 0.0075 |
| OHpro | 24.69 | 1.59-47.8 | 19.88 | 11.99-27.76 | 0.1566 |
BIOCHEMICAL MARKERS IN MENOPAUSAL WOMEN
J. Ferrer-Cañabate*, I. Tovar, P. Martinez
Servicio AA.CC. H.U. Virgen Arrixaca
OBJECTIVE
Measurement of bone marker levels has a role in the assessment of the osteoporotic risk in postmenopausal women. The aim of this work was to evaluate which marker of bone formation or resorption is better to evaluate this risk.
MATERIAL AND METHODS
55 untreated surgical or physiologic postmenopausal women (age 48.9±5.3, range 30-62), 13 perimenopausal women (46.2±3, 43-53) and 27 young healthy women (31±6.95, 19-44) were included in this study, all of them with normal hepatic and renal functions. The diagnosis of osteoporosis or osteopenia was a BMD (DEXA, lumbar spine). 2-h fasting urine samples and sera were collected and kept at -80ºC until use (markers in urine were expressed as creatinine ratios). Markers of bone formation (serum osteocalcine (BGP), ECLIA, Roche; and serum bone alkaline phosphatase isoenzyme, ELISA, Metra) and markers of bone resorption (U-Ca/Cre, automatized photometric colour test, Boehringer Mannheim; U-NTx, ELISA, Osteomark, Ostex; serum CTx (Crosslaps), ECLIA, Roche; and F-DPD, ELISA, Metra) were measured, besides FSH, LH and other control biochemical markers. All the statistical analysis were performed with the SPSS Software Program.
RESULTS
The best correlation was found between CTx and BGP, followed by CTx and NTx (table 1). Moreover, the Student test showed that the best markers to discern the bone turnover between peri-, postmenopausal and young control women were CTx and NTx (table 2).
|
|
DETERMINATION OF BONE ALKALINE PHOSPHATASE ENZYME (BAP): ELECTROPHORESIS VS. EIA
J. Ferrer-Cañabate*, I. Tovar, P. Martinez
Servicio AA.CC. H.U. Virgen Arrixaca
OBJECTIVE
The aim of this work was to perform a comparative study between two methods for determination of a bone formation marker, the bone alkaline phosphatase enzyme (BAP): the classical electrophoretic separation of AP enzymes in a buffered agarose gel and an immunoassay utilizing monoclonal anti-BAP antibodies., in order to know their equality and correspondence as measurements of this indicator of osteoblastic activity.
MATERIAL AND METHODS
Fifty-five untreated postmenopausal women were included in this study, mean age 48±5.1 years, all of them with normal hepatic functions according to their serum transaminases. Mean total alkaline phosphatase enzyme was 180±57.58 U/L (range 105-401). Sera was separated from blood by centrifugation and was frozen in glass tubes at –80ºC until use. We used two methods of determination: the first one, an IsopalR Alkaline Phosphatase (AP) Isoenzyme Electrophoresis kit (Beckman) with incubation with IsopalR Neuraminidase looking for bone-liver bands differentiation, and the second one an immunoassay in a microtiter strip utilizing monoclonal antibodies (Alkphase-BR, Metra Byosistem). All statistical analysis were realized with the SPSS Statistical Software Program.
RESULTS
Linear regression results were showed in table (equation type y=ax+b).
Correlation of Pearson was 0.85 (p<0.01), and Bland-Altman graphic (means vs. differences) showed that differences were bigger in accordance with the biggest mean results.
DISCUSSION
The EIA determination can supply perfectly to the electrophoretic one, because althoug is more expensive, it is also more sensitive, with less within- and between-run CV and, if hepatic function is normal, with cross-reaction with the hepatic isoenzime less than 15%.
| a | (a)95% | b | (b)95% | R2 (p<0.05) | |
| Electroph. vs EIA | 3.001 | 2.428-3.574 | 8.806 | -5.969-23.580 | 0.722 |
THE LEVELS OF VITAMIN K ANALOGUES IN OSTEOPOROTIC AND OSTEOARTHRITIS PATIENTS SAMPLES DETERMINED BY THE SENSITIVE AND SIMPLE HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY WITH ECD METHOD.
H. Wakabayashi1*, H. Hirata1, K. Onodera2, T. Yamamoto3, T. Tanizawa3, S. Sato4, T. Kusano4
1Niigata College of Pharmacy, Niigata, Japan
2Okayama University, Okayama, Japan
3Shinrakuen Hospital, Niigata, Japan
4Tominaga-Kusano Hospital, Sanjo, Japan
To investigate the physiological and pathophysiological roles of vitamin K analogues in the bone metabolism, especially in the osteoporosis, we have developed the sensitive and simple analysis system of vitamin K1 and K2 analogues (MK-4,5,6,7,8,9,10) in the bone and serum. After the separation of vitamin K analogues on a reversed-phase column (Shodex C18, 250 x 4.6 mm I.D.) using a mixture of ethanol-methanol (1:1, v/v) as the mobile phase, the analogues were reduced once in a platinum catalyst reduction column (10 x 4.6 mm I.D.) on-line, then monitored quantitatively by electrochemical detector (EICOM ECD-300) with a glassy carbon working electrode operated in the oxidation mode (+0.6V vs. Ag/AgCl). The detection limits (signal-to-noise ratio: 3) of vitamin K analogues were 2-10 pg. We also investigated the extraction procedures for the vitamin K analogues from bone and serum. Quantitative recoveries from bone and serum were obtained in the range of 80-101% for vitamin K analogues. We could measure the circulating levels of vitamin K analogues both osteoporotic patients (n=28) and normal subjects (n=21). It is notable that the circulating levels of vitamin K1 and vitamin K2 (MK-7) in the osteoporotic patients were significantly lower than in normal subjects. We also determined the contents of vitamin K analogues in the bone obtained from the patients with osteoporosis or osteoarthritis. High levels of vitamin K1 and MK-4 were found in trabecular bone taken from femoral head of osteoporotic patients. While, the levels of all vitamin K analogues in both cortical and trabecular bone of the osteoarthritis patients differed from humeral head, femoral head, and patella, respectively. We believe that the developed analysis method for the determination of vitamin K analogues in biological sample is a powerful tool for investigating physiological and pathophysiological roles of vitamin K analogues in the bone metabolism.
URINARY EXCRETION OF B, CA, MG AND P IN POSTMENOPAUSAL WOMEN WITH AND WITHOUT OSTEOPOROSIS
E. Arévalo González*, O. M. Alarcón, M. Mora, J. R. Vielma, L. Linares de M., G. Hernández de Avila
Universidad de Los Andes, Mérida, Venezuela
Boron (B) is necessary to transform the vitamin D into their active form, it intervenes in the consolidation of the fractures, in the appropiate bone calcification and in the metabolism of Ca, P and Mg. Furthermore, the supply of B can help in the prevention of the osteoporosis. In the present work we analyzed the relationship between the urinary concentration of B and the concentrations of Ca, Mg and of P in serum and urine of 11 healthy postmenopausal women (Group I), without metabolic bone diseases and of 34 postmenopausal women with osteoporosis (Group II). Patients did not suffer from diabetes and renal or hepatic diseases. It was determined the urinary excretion of B, Ca and Mg by atomic emission spectroscopy with plasm coupled by induction: Ca and Mg in serum by flame atomic absortion spectroscopy: serum Ca2+ for potentiometry with selective ion electrode and P and Creatinine in serum and urine by molecular absortion spectrphotometry. A significant difference was demonstrated (p<0.05) in the concentration of B and P in the urine of the studied groups. No statistically significant relationships were demonstrated among the urinary concentrations of B and serum and urinary concentrations of Ca, Mg and P in postmenopausal women with osteoporosis.
SERUM TARTRATE-RESISTANT ACID PHOSPHATASE 5B AS A MARKER OF BONE RESORPTION IN ORCHIDECTOMIZED RATS
S. L. Alatalo*, Z. Peng, J. M. Halleen, H. K. Väänänen
Institute of Biomedicine, Department of Anatomy, University of Turku, Turku, Finland
Bone-resorbing osteoclasts express high amounts of tartrate-resistant acid phosphatase (TRAP) 5b and secrete it into the circulation. Orchidectomy (ORC) decreases bone mineral density and mechanical strength and may be a useful model for male osteoporosis. We used orchidectomized rats to study serum TRAP 5b as a marker of bone resorption. Trabecular bone mineral density (BMD) was measured from the left tibia shaft with peripheral Quantitative Computed Tomography (pQCT) before the operation and 5, 11, 17, 24, 40, 70, 110, 150 and 180 days after the surgery. Tail blood samples were taken at the same time points and serum TRAP 5b activity was measured with a specific immunoassay. BMD was significantly lower in ORC group compared with Sham group already at 11 days after the operation. Serum TRAP 5b activity was significantly elevated in ORC group at 5 days after the operation, returned to the Sham level at 17 days, and decreased below the Sham level at all later time points. At 180 days after the operation, trabecular bone volume was approximately 80% decreased in ORC group compared with Sham group. Relative osteoclast number (compared with trabecular bone area) was slightly higher in ORC group, suggesting that further bone loss still occurs at this time point. However, the absolute number of osteoclasts was significantly lower in ORC group, and correlated significantly with serum TRAP 5b activity at the same time point. This data suggests that serum TRAP 5b is a useful marker to determine the absolute bone resorption rate in rat ORC model.
SERUM TARTRATE-RESISTANT ACID PHOSPHATASE 5B IS A SPECIFIC AND SENSITIVE MARKER OF BONE RESORPTION
J. M. Halleen1*, S. L. Alatalo1, A. J. Janckila2, S. Cheng3, H. Suominen3, H. Woitge4, M. J. Seibel4, H. K. Väänänen1
1University of Turku, Turku, Finland
2Veterans Affairs Medical Center, Louisville, KY, USA
3University of Jyväskylä, Jyväskylä, Finland
4University of Heidelberg, Heidelberg, Germany
Osteoclasts express high amounts of tartrate-resistant acid phosphatase (TRAP) and secrete it into the circulation during bone resorption, suggesting that serum TRAP may be a useful marker of bone resorption. However, there are two forms of TRAP in the circulation, named as TRAP 5a and TRAP 5b, of which TRAP 5b is derived from osteoclasts and TRAP 5a from some other, yet unidentified sources. We have developed a monoclonal antibody O1A using TRAP 5b purified from human osteoclasts as antigen. O1A was used as a capture antibody in a solid-phase immunofixed enzyme activity assay. O1A was specific for TRAP in both human bone and serum in Western analysis. The immunoassay did not detect TRAP 5a or any other acid phosphatases from human serum samples. In a 6-month double-blinded placebo-controlled study, serum TRAP 5b activity decreased 48% after hormone-replacement therapy, whereas no change was observed in the placebo-group. Compared with healthy premenopausal women, serum TRAP 5b activity was significantly increased in healthy postmenopausal women and in patients with osteopenia, osteoporosis, active Paget's disease of bone, and in breast cancer patients with overt bone metastases. Unlike other serum markers of bone resorption, TRAP 5b activity did not accumulate into the circulation in patients with renal or hepatic failure. These results suggest that serum TRAP 5b is a specific and sensitive marker of bone resorption.
IS FRACTURE RISK RELATED TO THE RELATIVE RATES OF BONE FORMATION VERSUS RESORPTION? A PROSPECTIVE STUDY
J. W. Davis1, P. D. Ross2*, W. Knowlton1, H. Katagiri3, R. D. Wasnich1
1Hawaii Osteoporosis Center, Honolulu, HI, USA
2Merck & Co., Inc., Rahway, NJ, USA
3Tottori University, Yonago, Japan
We previously reported that biochemical markers of bone turnover at baseline predicted the incidence of new vertebral and nonvertebral fractures among 504 Japanese-American women living in Hawaii. Those results suggested that both formation and resorption markers can provide an indication of bone turnover, and that increased bone turnover is associated with increased fracture risk. This analysis examines whether the difference between markers of formation and resorption (a potential indicator of the relative bone balance) is associated with the incidence of new fractures. New vertebral fractures were defined as a decrease in vertebral height of at least 15% compared to baseline. Deoxy-pyridinoline (DPD), and osteocalcin (OC) were measured using commercial assays (Quidel). Marker levels were converted to Z-scores, and the difference in Z-scores was calculated for OC minus DPD. During 3 years follow-up, 63 women developed new fractures. In separate analyses, increasing levels of both DPD and OC were associated with progressively greater fracture risk. However, the difference between OC and DPD was not significantly associated with fracture risk. Additional adjustment for calcaneus BMD did not substantially alter these findings. The difference in Z-scores may not accurately quantify true differences in bone balance because of the added measurement error of combining two test results or other factors. Moreover, this study may have lacked the power to detect a significant association between relative bone balance and fracture risk. The findings confirm previous reports that increased bone turnover, as measured by some individual markers of both formation and resorption, is associated with increased fracture risk.
EFFECTS OF THE TNF-ALPHA ANTIBODY INFLIXIMAB ON SERUM MARKERS OF BONE TURNOVER AND MINERAL METABOLISM IN PATIENTS WITH RHEUMATOID ARTHRITIS
H. P. Dimai1*, T. Müller2, S. Eder2, J. Hermann2
1Department of Internal Medicine/ Division of Endocrinology, Graz, Austria
2Department of Internal Medicine / Rheumatology, Graz, Austria
Background: Many, if not most, pathological features in rheumatoid arthritis (RA) can be ascribed to the action of several proinflammatory cytokines that are expressed in the synovia and pannus, such as tumor necrosis factor-alpha (TNF-alpha), interleukin-1 (IL-1), chemokines etc. In vitro observations in the early 1980s led to the hypothesis that TNF-alpha drives the production of other proinflammatory cytokines, thus playing a key-role in pathogenesis of chronicity of the disease. On the other hand, it has been shown that TNF-alpha also plays a key role in osteoclast-differentiation and activity. This mechanism led to the concept, that TNF-alpha may also be involved in pathogenesis of osteoporosis in patients with RA. Treatment of these patients with the anti-TNF-alpha chimeric monoclonal antibody infliximab ameliorates joint destruction, as assessed by histology in an experimental system. However, so far it has not yet been investigated, whether treatment with infliximab affects markers of bone turnover and bone and mineral metabolism.
Subjects and methods: We designed a longitudinal study including 10 patients with RA who were not responding to treatment with methotrexate. Markers of bone turnover, such as osteocalcin, beta cross-laps or osteoprotegerin were determined twice before starting therapy with infliximab, and three times within two weeks thereafter. Furthermore, at the same intervals, parameters of mineral metabolism, such as intact parathyroid hormone (iPTH) and serum electrolytes, have been investigated. The data were compared to those obtained from 10 age- and weight-matched controls. The obtained data will be presented at the meeting.
CHANGES IN BONE MASS AND BONE TURNOVER MARKERS FOLLOWING PREGNANCY AND LACTATION IN MOTHERS OF TERM AND PRETERM INFANTS
C. M. Smith1*, R. C. Coombs2, R. Eastell1
1Bone Metabolism Group, University of Sheffield, UK
2Neonatal Unit, Northern General Hospital, Sheffield, UK
Bone turnover increases during the last trimester of pregnancy and so mothers of preterm infants may have lower bone turnover and hence less bone loss as a result of subsequent lactation.
The aim of this study was to compare changes in bone mass and turnover in lactating and non-lactating women who delivered term or preterm infants.
Total body (TB) and sub-region (cortical and trabecular sites) bone mass was assessed in 45 women (36 preterm, 9 term) using DXA (Hologic QDR 4500A) following parturition and again 6 months later (in 26 subjects). Infant feeding method was determined by personal choice. Serum and urine markers (osteocalcin, type I procollagen amino-terminal propeptide (PINP) and total and free pyridinium crosslinks were measured at each time point.
Significance is denoted as; * p<0.05, ** p<0.01. At 6 months following delivery 13 subjects had predominantly breast-fed (BF) and 13 predominantly formula fed (FF). Mean age at weaning was approximately 4 months post delivery in each group. 4 subjects had not resumed menses. There were no differences in bone turnover markers or bone mass postpartum or in response to lactation between term and preterm mothers. The BF group had a greater % decrease in BMD at the hip* and femoral neck** and % increases at predominantly cortical sites, which were significant at the arm*. The FF group showed no change at any sub-regional site. At the total body both groups showed a non-significant increase in BMD (0.96 and 0.56% respectively). Following backwards stepwise regression 45.5%* of the variation in % change BMD at the total hip was explained by predominant feeding method, maternal height and multiple pregnancy. There were significant associations between bone turnover markers postpartum and changes in bone mass eg. baseline PINP was associated with % change BMD at the pelvis (r = 0.61*).
Loss of trabecular bone mass during lactation has been widely reported. However, an increase in cortical bone mass following lactation has not and may represent endosteal formation in lactating women. There was no effect of delivering or breastfeeding preterm infants on either bone turnover markers or bone mass.
THE EFFECT OF STORAGE AND MAILING ON STABILITY OF URINARY NTX
R. A. Hannon1*, I. B. Catch2, A. Price2, R. Eastell1
1Bone Metabolism Group, Division of Clinical Sciences (NGHT), University of Sheffield, Sheffield, UK
2Clinical Chemistry, Northern General Hospital, Sheffield, UK
Urinary NTX may be used in clinical practice to monitor response to treatment and to provide additional information for diagnosis. Since urinary NTX exhibits large circadian variation all urine samples must be collected at the same time of day, usually as first or second voids. However, once collected, the samples may not arrive in the laboratory for several hours or days or may be mailed to the laboratory. The aim of this study was to investigate the stability of urinary NTX under such conditions. Spot urine samples were collected from 22 healthy subjects (2M,20F), mean age 34 years (range 20 to 62 years). Each sample was divided into five 1ml aliquots and one 5ml aliquot. One 1ml aliquot (baseline sample) was stored immediately at -20 deg C. The remaining four 1ml aliquots were kept at room temperature (20-25 deg C) for different lengths of time (24h,48h,72h and 96h) and then stored at -20 deg C until analysis. The 5ml aliquot was sent to another laboratory and returned to our laboratory by mail (total 72 h) and then stored at -20 deg C until analysis. Urinary NTX was measured by the automated Vitros ECi assay and urinary creatinine (Cr) by Vitros dry slide chemistry (Ortho-Clinical Diagnostics). All samples from an individual were measured in the same analytical batch. NTX,Cr and NTX/Cr levels for the different conditions, expressed as percentage of baseline, were analysed by repeated measures ANOVA followed by Dunnettt's multiple comparison test.
Storage for up to 48 hours or mailing of urine samples does not have a significant effect on levels of NTX/Cr. Longer storage results in a small (13%) but significant increase in NTX/Cr.
| Time at room temp | 24h | 48h | 72h | 96h | Mailed (72h) |
| NTX, %baseline | 104.5 | 103.3 | 108.7* | 109.3* | 104.1 |
| Cr, %baseline | 99.5 | 98.7 | 98.6* | 96.6** | 99.5 |
| tdNTX/Cr, %baseline105.1 | 104.5 | 110.3* | 113.6** | 105.3 | |
| *p<0.05 compared to baseline **p<0.001 compared to baseline |
|||||
IMPACT OF SEASON, MEASUREMENT INTERVAL AND RANDOM VARIABILITY ON REPRODUCIBILITY OF MARKERS OF BONE TURNOVER
A. C. Eagleton*, R. A. Hannon, R. Eastell, A. Blumsohn
Bone Metabolism Group, University of Sheffield, UK
The aim of this study was to address the relative importance of season, interval between measurements (serial correlation) and random factors on the repeat-test reproducibility of several markers of bone turnover. Blood samples and 24hr urine samples were collected at a sampling interval of 4 weeks over 56 weeks from healthy men and premenopausal women (6M 6F mean 32.9yrs, range 24 –44). Samples from each subject were assayed in a single analytical batch.
Seasonality was assessed by ANOVA and cosinor analysis with Hotteling’s t-test. Components of variability were determined by nested ANOVA (CVa, CVi, CVg). Effect of interval between sampling on CVi+a was determined by semivariogram analysis.
There was no significant effect of time of year on any marker except TRAcP (ANOVA P<0.05). There was also no significant evidence of seasonality for any analyte by cosinor analysis (Hotteling’s T2 at P<0.05; Table). The upper 95% confidence limit of the amplitude of any possible seasonal component was < 12% of the mean concentration for all markers. Components of random variation for each marker over the whole study period were determined by nested ANOVA (Table). Semivariograms were constructed for each analyte to study the effect of sampling interval on CVi+a. All analytes showed only slight evidence of serial correlation at an interval of 28 days. For example, based on this analysis, estimated CVi+a for betaCTX was 18.7% at a sampling interval of 28 days and 23.8% at an interval of 308 days.
In conclusion: 1) within-subject variability of markers is consistent with that reported in previous studies, 2) there is no evidence to suggest that season is an important contributor in this population, 3) serum betaCTX is more reproducible than urinary markers of bone resorption 4) at a test interval of 28 days serial correlation between measurements is not substantial.
| Marker | CVg | CVi+a | Seasonality | ||
| Amplitude% | Hottelings T2 | P | |||
| IBAP | 33.0 | 9.0 | 2.0 | 0.033 | 0.83 |
| PICP | 22.2 | 13.4 | 2.9 | 0.225 | 0.19 |
| OC | 28.0 | 14.2 | 1.6 | 0.051 | 0.75 |
| TRAcP | 15.4 | 9.7 | 3.5 | 0.099 | 0.58 |
| TDpd/Cr | 39.8 | 26.6 | 3.2 | 0.149 | 0.35 |
| Serum bCTX | 39.0 | 19.7 | 4.6 | 0.422 | 0.15 |
A COMPARISON OF THE IMPACT OF FEEDING ON THE BIOLOGICAL VARIABILITY OF BONE TURNOVER MARKERS
J. A. Clowes1*, R. A. Hannon1, T. S. Yap1, N. Hoyle2, A. Blumsohn1, R. Eastell1
1Division of Clinical Sciences (NGHT), University of Sheffield, UK
2Hoffman La Roche, Penzberg, Germany
We have previously demonstrated that feeding results in a small decrease in biochemical markers of bone turnover. These markers may be valuable for the monitoring of treatment of metabolic bone disease. The aim of our study is to examine the affects of feeding on parameters of variability especially least significant change. Twenty healthy premenopausal women were randomised in a cross over study to fast (F) or breakfast (NF) for a total of 10 days. All subjects fasted from midnight and had an identical breakfast at either 0800 (NF) or 1000 (F) on alternate days. Subjects collected a daily two hour second void urine sample between 0800 and 1000 and had a daily 0900 blood sample.
Resorption markers were urine and serum N-terminal telopeptide of type I collagen - uNTX and sNTX (Osteomark by ELISA), urine immunoreactive free deoxypyridinoline - ifDPD (Pyrilinks-D by ELISA), urine C-terminal telopeptide of type I collagen - uCTX (Osteometer by ELISA) and serum CTX (beta-Crosslaps/serum by Roche Elecsys). Formation markers were procollagen type I N-terminal propeptide - PINP (Roche Elecsys) and Osteocalcin - sOC - (ELSA-Osteo by IRMA).
Absolute values in the F and NF states were compared using a paired t test and the log of the subject mean. F and NF analytical (CVa) and individual (CVi) coefficient of variation were calculated using analysis of variance. The LSC was calculated as 2.77*20.5*CVi+a for a significance level of P=0.05. F and NF within subject variances were analysed using an F test, log of the subject mean and 80 degrees of freedom.
Each marker demonstrated differences in variability (CVa, CVi, LSC) in subjects, however overall there was no major increase in variability in the NF state. The systematic effects of feeding that result in a decrease in the absolute values of markers, mean however, that for consistency of results it is preferable to study patients in the fasting state.
| Bone Marker | % Difference | F CVa | NF CVa | F CVi | NF CVi | F LSC | NF LSC |
| sOC | 4.1** | 5.0 | 4.2 | 6.3 | 10.2~ | 22 | 30 |
| sPINP | 3.8**** | 2.4 | 1.0 | 10.6 | 10.6 | 30 | 29 |
| uifDPD | 7.5**** | 6.4 | 10.1 | 10.2~ | 8.9 | 33 | 37 |
| uNTX | 7.9* | 9.8 | 7.0 | 43.7~ | 29.9 | 124 | 85 |
| uCTX | 7.0** | 10.6 | 11.8 | 24.6 | 18.4 | 74 | 60 |
| sNTX | 8.5**** | 13.7 | 8.5 | 12.2~ | 7.1 | 51 | 31 |
| sCTX | 17.8**** | 1.5 | 3.4 | 19.1~ | 26.3 | 53 | 73 |
| *P<0.05, **P<0.01, ****P<0.0001 (Paired t test). ~P<0.05 (F test). | |||||||
CONSIDERATIONS REGARDING SOME BIOCHEMICAL MARKERS OF BONE METABOLISM AND OF CALCIOTROPIC HORMONES IN A GROUP OF ELDERLY SUBJECTS
I. Zosin1*, P. Bottermann2, D. Ivan2, C. Crista1, D. Muresan1
1Clinic of Endocrinology, University of Medicine and Pharmacy, Timisoara, Romania
22nd Medical Clinic, Medical Faculty of Technical University, Munich, Germany
The study group included 52 subjects aged 56 - 95 years (mean 75.42; F/M ratio = 2.7 / 1), issued from a senior home.
We excluded the cases bed – ridden or those treated with drugs having bone impact.
There were performed the following determinations: a profile of phospho – calcium metabolism, creatinine, urea – N, bone specific alkaline phosphatase (BAP), 25-OH-D3, 1,25-(OH)2-D3, PTH and deoxypyridinoline (DPD) excretion. The samples were drawn in the middle of June. All cases had a normal renal function. The values of serum calcium are partially at the lower limit of normal ranges. The mean of BAP was 51.39±30.57 U/ml (generally normal values). Only four aged subjects (over 70 years old) presented significantly increased values.
The mean value of 25 OH D3 was 10.39±9.53 ng/ml (decreased in most of subjects). 1,25(OH)2D3 levels lay between normal limits excepting four persons (values < 20 pg/ml).
PTH showed a mean of 64.19±57.26 pg/ml. PTH values were dissimilar (more of 48% cases with increased levels, sometimes significantly high).
The DPD excretion presented a mean value of 7.51±10.31 (moderate increase in both sexes).
The obtained data point out complex and unconcordant perturbations of some bone markers and of calciotropic hormones in healthy elderly.
An important percentage of subjects demonstrate features of secondary hyperparathyroidism with peculiarities in relation with age and subjects condition (senior home residents).
CALACANEAN ULTRSOUND FOR CLASSIFICATION OF OSTEOPOROSIS
H. D. Bolosiu*, S. P. Simon, C. R. Bolosiu, H. Popoviciu
Osteoporosis Unit, Rheumatology Department, University of Medicine and Pharmacy "I.Hatieganu", Cluj-Napoca, Romania
Background. Among many methods developed to investigate osteoporosis (OPR) the most used are dual x-ray absorptiometry (DXA) and quantitative ultrasound (QUS). They are aimed to measure different parameters at different sites of the skeleton. While any measure of the bone mineral density (BMD) is a strong indicator of OPR and osteoporotic fractures, the differences in T-score results obtained by various techniques should be considered when applying the WHO criteria for OPR.
Aim of the study. Our objective was to compare QUS of calcaneus to traditional DXA of the axial skeleton in order to find out which QUS parameters and at what values better discriminate patients supposed to suffer from OPR.
Material and methods. One hundred forty eight consecutive patients (126 female and 22 male) aged 19-81 years were referred for evaluation of bone loss and underwent examination. Ninety-eight of the female patients were post-menopausal. All patients were measured using DXA (DPX-alpha, Lunar) and QUS (Sahara, Hologic). Subjects were classified as normal (17.2%), or having either osteopenia (OPN) (32.7%) or OPR (50.1%) on their lowest T-score according to the WHO criteria.
Results. The QUS estimated calcanean BMD was significantly different from that measured by DXA at the axial skeleton. From the directly measured US parameters, attenuation (BUA) showed the best correlation to the DXA ones. BUA significantly differentiates OPR from OPN at threshold of 56dB/MHz (p<0.0001) but did not do so between the latter and normal. (p=0.25). We have found a correlation between QUS and DXA T-scores (R=0.69) in our group. Classification results of various US thresholds compared to DXA showed the optimal agreement for QUS threshold of T= -1.7 and T=0. At the T-score equal to or less than -1.7 QUS identifies OPR with 76% specificity (66% sensitivity) while at T-score equal to or greater than 0 the method discriminate OPN from normal (78% sensitivity, 42% specificity).
Conclusion. We believe QUS T-scores of l.7 and 0 fairly identify OPR and normal subjects respectively. Patients falling between these values should be subsequently measured by DXA for better classification.
QUANTITATIVE ULTRASONOMETRY (QUS) OF THE OS CALCIS IN POSTMENOPAUSAL WOMEN WITH SPINE AND HIP FRACTURE
P. Hadji*, M. Kalder, K. Bock, U. S. Albert, G. Emons, K. D. Schulz
Philipps-University of Marburg, Marburg, Germany
Quantitative ultrasonometry (QUS) of the Os Calcis has been shown to predict hip fracture in late postmenopausal women, and vertebral and forearm fracture in early postmenopausal women. Speed of sound (SOS), broadband ultrasound attenuation (BUA) and Stiffness index (SI) of the Os Calcis were measured using the Achilles ultrasonometer (GE/Lunar). Osteoporosis risk factors were assessed by a detailed questionnaire. We examined 1314 normal women from 48 to 79 years, with a mean age 60±7.5 years. In addition, we examined women of similar age, of whom 80 had suffered a hip fracture and 40 a spine fracture.
The short-term precision in vivo expressed as the coefficient of variation was 1.2% for BUA, 0.2% for SOS, and 1.3% for SI. A total of 813 women were measured at both the right and left heel. There was high correlation between the two sides (r=0.80 to 0.93) (p<0.001), with no systematic offset. The ultrasound variables decreased significantly (p<0.001) with age in healthy women; the annual decrease was -0.4% for BUA, -0.07% for SOS, and -0.7% for SI. BUA, SOS and SI discriminated (p<0.001) between fracture and non-fracture subjects, but the fracture groups were 2 to 4 years older. The T-score in the controls averaged -2.1 while that in the fracture patients averaged about -3.0. After control for age, years since menopause, and body size, BUA, SOS as well as the SI remained significantly lower (11 to 12% for SI) in women with fracture. The Z-score was -0.8 (p<0.01) in spine fracture cases, and -0.9 (p<0.001) in hip fracture patients. QUS provides a gradient of fracture risk comparable to x-ray densitometry of the axial skeleton, and gives comparable Z-scores and T-scores in younger postmenopausal women; it provides a precise, radiation-free, low-cost, and rapid method for fracture risk assessment in clinical practice.
PREDICTION OF FRACTURE LOAD: A CHICKEN TIBIA MODEL OF IMPENDING PATHOLOGICAL FRACTURE USING ULTRASOUND DENSITOMETRY
R. S. Yang1*, C. H. Lin1, T. H. Huang2, S. J. Lin1
1National Taiwan University, Taipei, Taiwan
2National Taiwan Normal University, Taipei, Taiwan
In this study we have investigated the impact of speed of sound (SOS) by ultrasound densitometry and regional bone mineral density (rBMD) by dual-energy x-ray absorptiometry (DXA) on the prediction of the fracture load of chicken tibia with a drill hole in the midshaft. The periosteum and soft tissues of these chicken tibias were removed. They were grouped into 5 groups with a matched length, diameter and weight. Group 1 (3.5 mm) and Group 3 (5 mm) tibias were drilled at the convex side of midshaft. Group 2 (3.5 mm) and Group 4 (5 mm) were drilled at the side of midshaft. The Group 5 tibias were not drilled and used as controls. The cortical thickness was measured and drill hole ratio (DDR) was calculated as drill hole size/diameter on the plain radiographs. The rBMD was measured at midshaft (1.95 cm x 1.95 cm). The SOS on the cave side of tibia was measured using the Sunlight Omnisense 7000S ultrasound bone densitometer. The fracture loads of these tibias were assessed by a three point bending test using an MTS-BIONIX 858 testing system. The fracture load for Group 1 tibias was 359.7±41.3N, for Group 2, 252.7±55.9 N, for Group 3, 325.3±75.8 N, for Group 4, 218.5±149.7 N, and for Group 5, 377.1±83.0 N (p<0.05). Analysis of covariance (ANOCOVA) and multiple regression analysis using DDR and cortical thickness as regressors showed a significantly negative effect of DDR on the fracture load (r2=0.26, p<0.05). The addition of regional BMD as a regressor showed no impact on the regression results (r2=0.34, p<0.05). On the other hand, the addition of SOS as a regressor showed a larger correlation coefficient (r2=0.41, p<0.005). In addition, the DDR showed a negative effect (p<0.01) whereas the SOS showed a significantly positive effect (p<0.05) on the fracture load. The results suggested that SOS instead of rBMD showed its better effect to predict the fracture load. Therefore, the use of DDR and SOS may improve the prediction of the fracture load of this impending pathological models.
QUANTITATIVE ULTRASOUND OF OS CALCIS OF WOMEN IN RURAL MOUNTAINOUS HONDURAS
W. Bronson1,2*, J. McMillen1, E. Brooks3, W. Gondring1
1Heartland Health Systems, St. Joseph, MO, USA
2U. of Missouri- Columbia, Columbia, MO, USA
3Missouri Western State College, St. Joseph, MO, USA
The purpose of this study was to determine the bone mineral density (BMD) of women in an isolated, rural, mountainous region of the Department of Yoro, Honduras. In a previous study we reported what appears to be a low prevalence of fractures of the hip, wrist and spine among postmenopausal women in this region. This study further explores our initial observation.
Heel BMD's of 225 women were determined using a Lunar Achilles Express heel ultrasonometer powered by a Honda EU 1000i generator. Seven clinics were held at 5 sites in the region. Questionnaires were administered.
One hundred and nine postmenopausal women were identified. Only 4 women over age 59 had a heel BMD in the normal range based on WHO criteria. No women over age 69 had a heel BMD in the normal range. The average age at menopause was 46.75 years, s.d. 6.56. None of these postmenopausal women received hormone replacement therapy.
In conclusion, postmenopausal BMD loss at the os calcis resulting in osteopenia and osteoporosis by WHO criteria is the norm in this population. Without body size adjustment, heel BMD differences do not explain our previous observation of fewer fractures of the hip, wrist and spine in this rural mountainous region. Hip geometry, stature, greater neuromuscular fitness of the elderly, or other non-bone mass factors may be more important.
| T-Scores by Decade | ||||||
| Age | n | <-2.5 | -2.4 - -1.0 | -0.9 - 0.0 | 0.1 - 1.0 | >1.0 |
| 20-29 | 33 | 9 | 15 | 5 | 4 | |
| 30-39 | 40 | 9 | 14 | 13 | 4 | |
| 40-49 | 44 | 1 | 17 | 14 | 9 | 3 |
| 50-59 | 54 | 4 | 21 | 12 | 14 | 3 |
| 60-69 | 26 | 5 | 17 | 2 | 2 | |
| 70-79 | 18 | 5 | 13 | |||
| 80-89 | 7 | 5 | 2 | |||
| 90-99 | 1 | 1 | ||||
EVALUATION OF CALANEAL BONE DENSITY WITH HANDY QUANTITATIVE ULTRASOUND DEVICE IN A JAPANESE POPULATION
K. Yoh1*, H. Kisimoto2, H. Oota2, I. Gorai2, J. Hasimoto2, S. Yosimoto2, Y. Nakatsuka2
1Orthop Dept. Hyogo College of Medicine
2CM100 Multicenter Study Group
Normative data are essential for the clinical utility of quantitative ultrasound (QUS). The aim of this study was to provide normative data of QUS values in Japanese men and women. As part of a multicenter study in Japan we collected normative data for a handy QUS device (CM100, Furuno, Japan). Short-term precision was also evaluated. CM100 is a mobile handy size QUS device with a gel-coupled (dry) system which can measure speed of sound (SOS) in calcaneus. 2760 Japanese healthy women and 1108 Japanese healthy men aged between 20 to 90 years were enrolled in this study.
Precision error using Phantom was 0.15% and precision error (%CV) in vivo was 0.27%(0.17%~0.43%). The young adult mean (YAM) of SOS were 1525±30m/s and 1530±29.9m/s for men (n=426) and women (n=739), respectively. The peak value of SOS was in the age of 20 to 25 years both in men and in women for 1534±26m/s and 1537±32m/s, respectively. The lowest value of SOS was in the age of 85 to 89 years both in men and women for 1506±16m/s and 1461±13m/s, respectively. The decrease of SOS started from 20 to 25 years same as in men and in women. Specially in women, the decreasing rate of SOS in age from 50 to 59 years was higher than another generation. The age-related decline in the QUS values in men and in women showed some difference to age-related decline observed at the lumbar spine using DXA. In conclusion, the handy QUS device CM100 is highly reproducible and this investigation presents normative data of QUS value of men and women in a Japanese population.
THE IDENTIFICATION OF FRACTURE PATIENTS AT RISK OF OSTEOPOROSIS USING ULTRASOUND AS A RISK ASSESSMENT TOOL
H. J. Veevers1*, I. Stewart1,2, D. P. Montgomery1,2, A. Doyle3
1Blackpool Wyre & Fylde Community Health Service NHS Trust, Blackpool, UK
2Blackpool Victoria Acute Trust, Blackpool, UK
3Blackpool Primary Care Group, Blackpool, UK
Low trauma fracture is an accepted risk factor for future hip fracture. An Audit of Osteoporosis in General Practice in Blackpool had identified that only approximately 50% of patients with a low trauma fracture had been considered at risk of Osteoporosis.
A study was undertaken to identify and evaluate patients presenting to an acute hospital with a low trauma fracture, for example, wrist, humerous and ankle, for being at risk of Osteoporosis. Identification of other risk factors such as previous fractures, early menopause/hysterectomy, corticosteroid use and other drug or medical history was by patient questionnaire and patients both male and female between the ages of 40-80 were offered a heel ultrasound scan (HOLOLOGIC SAHARA).
Those patients with an ultrasound 'T' score of -1.5 and below or between -1.0 and -1.5 with other major risk factors were referred for Dexa of hip and lumbar spine. Patients were also offered education and advice about Osteoporosis.
THE LOW TRAUMA FRACTURES WHICH ARE LISTED ARE THOSE WHO HAVE BEEN SCANNED USING DEXA.
87 wrist fractures
14 humeral
8 ankle
6 spinal
24 others
RESULTS FROM THE FIRST 12 MONTHS
363 patients have been assessed with ultrasound.
201 have been referred for a Dexa scan
139 have been scanned to date
76 of those patients scanned using Dexa have been considered for drug therapy
ULTRASOUND AND DUAL X-RAY ABSORPTIOMETRY IN A POPULATION OF WOMEN
N. Sierra1*, M. J. Moro2, M. Díaz-Curiel2, A. Rapado2
1Centro de Salud Alicante, Fuenlabrada, Spain
2Fundación Jiménez Díaz, Madrid, Spain
The purpose of this study was to determine the correlation of different methods for the measurement of bone quality in a population sample of 79 spanish women.
Subjects and methods: Velocity of ultrasound (SOS) and Broadband ultrasound attenuation (BUA) were measured at the calcaneus (Lunar Achilles, Germany), SOS was measured at the phalangeal (DBM Sonic 1200 IGEA, Italy), and bone mineral density (BMD) of spine, femoral neck and ultradistal radius were measured by DEXA (Hologic QDR4500). The women were classified in four groups. Group 1: thirteen females with Colles fracture (aged 50-58 years); Group 2: seven females receiving corticoids chronic treatment (aged 50-65 years); Group 3: seventeen females premenopausal (aged 40-54 years); Group 4: forty-two postmenopausal women (aged 45-72 years). Pearson´s correlation were used for analysis.
Results: In group 1, the SOS of calcaneus showed a correlation with SOS of phalanges (r=0.58, p=0.04) and additionally with the DEXA of femoral neck (r=0.49, p=0.05) and radius (r=0.78, p=0.001). BUA of calcaneus only correlated with DEXA of radius (r=0.52, p=0.05). In group 2, SOS of calcaneus showed correlation with all the DEXA measurements: femoral neck (r=0.68, p=0.05), lumbar spine (r=0.93, p=0.002) and radius (r=0.59, p=0.001). In group 3, the only correlation was SOS of calcaneus with SOS of phalanges (r= 0.6, p=0.01) and DEXA of radius (r=0.514, p=0.03). In group 4 the SOS of phalanges showed a correlation with SOS of calcaneus (r=0.31, p=0.04), DEXA of femoral neck (r=0.35, p=0.02), lumbar (r=0.55, p=0.001) and DEXA of radius (r=0.58, p=0.0001). Similarly SOS of calcaneus correlated with DEXA at all location: femoral neck (r=0.64, p<0.0001), lumbar (r=0.56, p=0.0001) and radius (r=0.44, p=0.003).
Conclusions: In general, ultrasound measurements, especially SOS, had a good correlation with all the BMD measurements by DEXA. We found significant correlation between these methods for group of women with corticoids therapy and postmenopausal women, which had a higher prevalence of osteoporosis. We concluded from this transversal study that ultrasounds, specially at the calcaneus, could be a valid methodology for osteoporosis assesment.
PATIENTS WITH ELECTIVE ARTHROPLASTIES: A HIGH-RISK POPULATION FOR OSTEOPOROSIS
B. J. Edwards1*, D. Stulberg1, L. Madison1, C. Wilson2
1Northwestern University, Chicago, Illinois, USA
2Northwestern Memorial Hospital, Chicago, Illinois, USA
Osteoarthritis, the most common arthritis in humans affects approximately 40 million Americans and is the second most common cause of disability. Joint pain, swelling, and falls are common. Rheumatoid arthritis affect joints, exacerbated by the use of glucocorticoids. As these diseases progress, pain and joint deformity will become disabling, in these cases arthroplasty is indicated.
Method: Patients admitted for arthroplasty were evaluated by the Osteoporosis Program. 103 patients were seen Oct 1999-Sept 2000. Female 98 (95%), male 5 (5%). Age 67±10 yrs, 80% of patients were 60-80 years of age. All were community dwelling, cognitively intact and ambulatory.
Diagnosis of osteoporosis: 22% of cases on admission, 34 (33%) reported a previous non-traumatic fracture. Radiographic osteopenia was present in 25% of cases. Heel ultrasound (Metra Biosystems) was used with 30% of the remaining patients presenting a T score<-1.5, and 20% a T score <-1.0.
Results: Our study shows a high prevalence of osteoporosis (50%) and osteopenia (20%) in a group of patients with severe arthritis. Obesity is not protective of osteoporosis in these patients. Lifestyle factors and chronic medical conditions are common, as well as is the use of glucocorticoids. Established osteoporosis was present high percentages in spite of low diagnosis rates. Patients with severe arthritis have unique risk factors for osteoporosis which are poorly understood. Presence of an osteoporosis program helps to address the issues raised by this high-risk patients.
Conclusion: Patients with severe arthritis are a high-risk group of patients for osteoporosis, falls and fractures. Their risk factors are unique. An inpatient osteoporosis program increases diagnosis and treatment rates in this group of patients with severe arthritis.
EUROPEAN PEDIATRIC REFERENCE DATA FOR QUANTITATIVE ULTRASOUND OF FINGER PHALANGES
R. Barkmann1*, W. Rohrschneider2, G. Baroncelli3, R. Lorenc4, R. Tormo5, C-C. Glüer1
1Medizinische Physik, UKK Kiel, Germany
2Pädiatrische Klinik der Universität Heidelberg, Germany
3Department of Reproductive Medicine and Pediatrics, Pisa, Italy
4Children's Mem. Health Institute, Warsaw, Poland
5Hospital Materno-Infantil "Vall d'Hebron", Barcelona, Spain
Quantitative Ultrasound (QUS) of finger phalanges is an interesting tool for the assessment of juvenile bone status due to its lack of radiation and the measurement on an easily accessible site, which is known to be affected by growth disorders. However, in order to study skeletal disorders, normal reference data for the pediatric age range are required. We started collecting data from five European centers in order to establish a comprehensive European pediatric reference database.
About 2500 healthy boys and girls have been measured in five centers in Germany, Italy, Poland and Spain using the DBMSonic 1200 or DBMSonic Bone Profiler (IGEA, Italy). Besides the standard parameter Amplitude Dependent Speed of Sound (AD-SoS) a new parameter was evaluated, the Bone Transmission Time (BTT). BTT is the difference between the transmission times in the phalanx and in pure soft tissue. It only depends on bone properties and is not affected by the width of the soft tissue, which in general affects speed of sound measurements. All centers were included in a quality control procedure using a recently developed cross-calibration phantom.
Highly significant positive, gender-specific correlations could be observed between AD-SoS resp. BTT and age in all centers. Differences in the calibration of single devices, which could be as large as one population standard deviation of the young population, could be substantially reduced using the cross-calibration procedure to less than 25% of a population standard deviation.
This is to our knowledge the most comprehensive database of pediatric reference data of phalangeal QUS including the new parameter BTT. The introduction of a cross-calibration procedure assures proper use of these data on all devices. This is an open study and more measurements and other centers will be included.
QUANTITATIVE ULTRASOUND (QUS): COMPARISON OF AN IMAGING DEVICE WITH ANOTHER ULTRASOUND DEVICE AND WITH DENSITOMETRIC MEASUREMENTS
A. Peretz, P. Bergmann*
CHU Brugmann, Free University of Brussems, Belgium
INTRODUCTION. DXA measurements are recognized as the gold standard for the diagnosis of osteoporosis. QUS is a useful tool in the detection of osteoporosis. There is a large variety of ultrasound devices with technical differences between them. AIM. To compare two devices measuring US transmission through the calcaneum and their ability to identify patients with low DXA. MATERIAL and METHODS. 109 women, 64±11 years old, underwent QUS measurements using Achilles + (Lunar, Maddison) and Ubis 5000 (DMS, France) devices. DXA of the hip (BMDH) was performed using a QDR 4500 machine (Hologic, Waltham, MA). RESULTS. The 3 QUS parameters of both devices showed similar absolute values for speed of sound (SOS) while bone ultrasound attenuation (BUA) was lower with UBIS. They were significantly correlated between them and with BMDH (p<0.0001). ROC curves were constructed to evaluate the performance of US parameters to detect with the best sensitivity and specificity low BMDH (T score= -2SD). Using Stiffness, the 2 US devices performed equally. The threshold giving the best sentivity and specificity was 44 U (Achilles +) and 57 U (Ubis). Whatever the device used, patients belonging to the highest tertile (T3) of QUS had a negative predictive value of 88% for osteopenia. In the lowest tertile, the positive predictive value for osteopenia was only ~ 50%. Concordance in classifying the patients between tertiles of QUS parameters was evaluated using kappa scores (table).
CONCLUSION. 1. Both devices give parameters which are well correlated between them and with DXA; 2. Both devices have the same predictive value for low and normal BMD; 3. An efficient use of QUS would be to exclude from DXA measurements patients in the highest tertile; 4. Using BUA, concordance to classify patients in this tertile using both QUS devices is good; it is medium if stiffness is used.
| BUA | SOS | Stiffness | |
| T1 vs T2 vs T3 | 0.58 | 0.61 | 0.59 |
| T1 vs T2 + T3 | 0.61 | 0.75 | 0.80 |
| T3 vs T1 + T2 | 0.75 | 0.62 | 0.60 |
USEFULNESS OF QUANTATIVE ULTRASOUND FOR EARLY DETECTION OF BONE MASS CHANGE
H. B. Park*, D. M. Kim, C. W. Ahn, S. S. Jeong, J. H. Nam, B. S. Cha, H. C. Lee, K. B. Heo, S. K. Lim
A University, Yonsei, Korea
Hyperthyroidism is a well known cause of altered bone metabolism, characterized by increase in both osteoclastic and osteoblastic activity with predominance of bone resorption and resulting in decreased bone mass. However, the effect of antithyroid treatment on BMD with hyperthyroidism is still controversial. In addition, it has not been clearly defined whether the bone loss in hyperthyroidism is reversible or irreversible and when the bone mass was increased during antithyroid treatment. The aim of this study is to evaluate the bone mass change (SOS) at diagnosis, after 3, 6, and 9 months of medical treatment by QUS (Omnisense sunlight) prospectively. Seventy newly diagnosed patients with Graves' disease were studied longitudinally, every 3months interval, for 9months after commencing antithyroid treatment with methimazole or PTU. We measured age, fT4, TSH respectively and that included the patients who had normal range of thyroid function (fT4) within two or three months after antithyroid treatment. We excluded the patients who recurred, or became hypothyroidism. Of these patients, we measured SOS (speed of sound) and T-score of Tibia, distal Radius by QUS in 40 patients (male 15, female 25) with hyperthyroidism at diagnosis, 3months, 6months and 9months after treatment. Bone mass changes (%DSOS) were as follows: [(final bone mass-initial bone mass)/initial bone mass]x100.
Hyperthyroidism in young patients, even severe hyperthyroidism, bone mass was rapidly increased after six months of antithyroid treatment. This increament was continuous for nine months but degree of increament in 6months of antithyroid treatment was greater than that of 9months of antithyroid treatment. However, bone mass was decreased in old men and postmenopausal women after 6months of antithyroid treatment, even after 9months of antithyroid treatment. This results indicate that age and menopausal state, in addition to thyroid function, are important to determine bone mass in hyperthyroidism.
| Table: Bone mass change in hyperthyroidism | |||||||
| 3month(%) | 6month(%) | 9month(%) | |||||
| Radius | Tibia | Radius | Tibia | Radius | Tibia | ||
| Male | -0.63 | -0.2 | -0.98 | -1.3 | 3.43 | 3.5 | |
| Female | Menopause | -1.1 | -0.8 | -0.4 | -0.1 | -1.0 | -1.5 |
| Premenopause | -0.2 | -0.6 | 0.3 | 0.2 | 2.3 | 2.6 | |
DIAGNOSTIC OF OSTEOPOROSIS: CUT-OFF LEVELS BASED ON CALCANEOUS ULTRASOUND WITH TWO DIFFERENT DEVICES
A. Diez-Pérez*, J. Puig, M. J. Peña, L. Mellibovsky, J. Carbonell, X. Sanz, X. Nogués
Hospital del Mar-UAB and URFOA-IMIM, Barcelona, Spain
Non-invasive diagnosis of osteoporosis (OP) is currently based on WHO criteria (Kanis, 1994) using dual-energy absorptiometry (DXA) as reference standard. Ultrasound devices (US) are a potential alternative because are inexpensive, portable and convenient, both in hospital and primary care setting. However, no clear diagnostic criteria or cut-off values for clinical decision making have been established for US. We analyze diagnostic criteria of osteoporosis for two US instruments.
Bone mass in 138 postmenopausal women, age mean (SD) 58.7 (10.4), attending a menopause unit in a teaching hospital were measured with a DXA (Hologic QDR 4500 SLTM) and two US instruments (Hologic SaharaTM, [128 cases], and Lunar AchillesTM [104 cases]). Diagnostic of osteoporosis was accepted for T-score at femoral neck measured by DXA equal or below -2.5. Calcaneal US measurements were BUA (broadband ultrasound attenuation), SOS (speed of sound), Stiffness and T-score. For each US value level, sensitivity, specificity and positive likelihood ratio (+LR) were calculated with respect to DXA. Values of LR were categorized as a diagnostic of OP for +LR >10 (very high probability), and possible diagnosis for +LR between 5 and 10 (high probability). Percent of classified patients with this method is 45.8% for Sahara T -score and 50% for Lunar T-score. Reference values for diagnostic of OP are for Sahara T-score values below -1.9 and for Lunar T-score values below -2.6.
We conclude that, for these US values, diagnosis of OP is of very high probability and can support a reliable clinical decision without a DXA measurement in a substantial proportion of patients.
CORRELATION OF ULTRASONOMETRIC BONE QUALITY WITH OSTEOPOROTIC VERTEBRAL FRACTURES IN WOMEN WITH RHEUMATOID ARTHRITIS (RA)
R. Dreher
Hospital for Rheumatic Diseases, Bad Kreuznach, Germany
Aims
Comparison of ultrasonometric parameters (QUS, SOS, BUA) of the calcaneal bones with the bone density values (L2-L5, femoral neck) in women with RA and women with non inflammatory syndroms and association with osteoporosis fractures/deformities of the lumbar vertebrae.
Method: QUS (Lunar Achilles) was measured on both calcaneal bones in a total of 95 women without diagnosed osteoporosis (Mean age 60±11 years, BMI 0.27±5 kg/m2 RA:,57 women (mean age 59±12 years, BMI = 26±4 kg/m2, NID: 38 women (mean age 62±11 years, BMI 28±6 kg/m2). The results were compared with DXA measurements on lumbar vertebrae and the neck of the femur (LUNAR) and correlated with fractures of lumbar vertebrae.
Osteoporosis: DXA T-score femoral neck < - 2.5 SD
Lumbar vertebrae fractures: more than 20% wedge or central deformity
Statistics: linear Pearson correlations (Two sided, univariate and multivariate logistic regression analysis.
Results: Women with (RA) show significantly lower ultrasonometry values (QUS, SOS, BUA) of the calcaneal bones as compared to women with NID. Women with RA and with demonstrable fractures (n=16) of the lumbar vertebrae show a significant lowering of QUS and SOS on the calcaneus compared to nonfractured women with RA (n=41). RA women with and without demonstrable fractures do not differ significantly with regard to the DXA values.
Significant associations of all ultrasonic parameters (QUS, BUA, SOS: p<0.001) with osteoporosis and fracture can be demonstrated in RA women, in contast to L2-L4 and femoral neck DXA which is not associated with osteoporosis and fractures.
Conclusion: Risk for osteoporosis and osteoporotic fractures is increased in RA women with low ultrasonic parameters of calcaneal bones. The ultrasonographic quality of the bone (QUS, SOS, BUA) but not the content of calcium salts (DXA) is the major predictor for vertebral fractures in women with rheumatoid arthritis.
PREVALENT APPENDICULAR FRACTURES IN ELDERLY WOMEN WITH NORMAL DEXA BONE MASS ARE ASSOCIATED WITH LOW ULTRASOUND MEASUREMENTS
A. Marangou1,2*, A. Devine1, S. S. Dhaliwal2, R. L. Prince1,2
1Department of Medicine, University of Western Australia
2Dept Endocrinology and Diabetes, Sir Charles Gairdner Hospital, Perth, Western Australia
The pathophysiology and aetiology of multiple fractures in people with normal DEXA BMD has not been well investigated. The aims of this study were to examine the factors that may be associated with multiple appendicular fractures in women with normal Dexa bone density (BMD).
The study group consisted of 204 women age 75±2y (mean±SD) who had been assigned a Total Hip BMD T score greater than or equal -1.0 and classified according to prevalent fracture status: 0/1 fracture (n=179) and 2 or more fractures (n=25).
Calcaneum QUS (Lunar Achilles) & DEXA(Hologic 4500A) BMD at the hip and bone turnover markers (osteocalcin, alkaline phosphatase and urinary dexoyprydinoline/creatinine ratio) were measured. Age at menopause, average daily calcium and alcohol intake, smoking history and body mass index (BMI) were ascertained. Significant differences between the fracture groups were determined by Student's t test.
BMD at Total Hip & Femoral neck and bone turnover markers were not different between the two groups. There was no difference between the two groups for age at menopause, calcium & alcohol intakes, years smoked and BMI. QUS measures were significantly lower in the multiple fracture group by approx. 1 SD compared to those in the none or one fracture group (SOS 1510±24 versus 1529±24 metres per second, P<0.01; BUA 102±8 versus 106±7 decibels/Mega Hertz; stiffness 71±11 versus 79±10%, P<0.05).
We conclude that patients with prevalent appendicular fractures with normal DEXA BMD may have qualitative differences in bone that can be detected by QUS.
