Quantitative ultrasound technique at the phalanges in discriminating between uremic and osteoporotic patients

Quantitative ultrasound technique at the phalanges in discriminating between uremic and osteoporotic patients

European Journal of Radiology 60 (2006) 108–114 Quantitative ultrasound technique at the phalanges in discriminating between uremic and osteoporotic ...

169KB Sizes 0 Downloads 31 Views

European Journal of Radiology 60 (2006) 108–114

Quantitative ultrasound technique at the phalanges in discriminating between uremic and osteoporotic patients G. Guglielmi a,∗ , F. de Terlizzi b , F. Aucella c , A. Scillitani d a

Department of Radiology, Scientific Institute Hospital “Casa Sollievo della Sofferenza”, Viale Cappuccini 1, 71013 San Giovanni Rotondo, Italy b IGEA srl, Carpi, Italy c Division of Nephrology, Scientific Institute Hospital “Casa Sollievo della Sofferenza”, San Giovanni Rotondo, Italy d Division of Endocrinology, Scientific Institute Hospital “Casa Sollievo della Sofferenza”, San Giovanni Rotondo, Italy Received 21 October 2005; received in revised form 9 March 2006; accepted 2 May 2006

Abstract This study was conducted to test the ability of quantitative ultrasound technique (QUS) at the phalanges to discriminate between uremic and osteoporotic patients. Three groups of subjects (38 dialytic women, 16 osteoporotic women with vertebral fractures, 19 non-dialytic and non-fractured women) were recruited at the Department of Radiology at “Casa Sollievo della Sofferenza” Hospital, San Giovanni Rotondo, Italy. The groups were matched for age and body mass index (BMI). On all subjects the following measurements were performed: spinal BMD by QCT and by DXA, Femoral BMD by DXA, phalangeal QUS. For QUS measurements, the DBM Sonic (IGEA, Carpi, Italy) was applied to the metaphysis of the proximal phalanges of the last four fingers of the hand. Osteoporotic women with vertebral fractures showed significantly lower values of spinal BMD by QCT and DXA and Ward’s Triangle BMD with respect to hemodialytic patients (p < 0.005). All QUS values, except for BTT and SoS, showed lower values in osteoporotic women with respect to hemodialytic patients (p < 0.05). Control group showed higher values of AD-SoS, BTT and SoS than hemodialytic patients (p < 0.005) while the two groups did not differ for BMD values measured with both QCT and DXA. UBPI and FWA data showed a similar behaviour to DXA and QCT results, whereas BTT and SoS showed a completely different behaviour. AD-SoS was the only parameter that could effectively discriminate among the three groups (ANOVA, p < 0.0001). We conclude that phalangeal QUS can discriminate between hemodialysed patients and controls with similar bone mineral density, and can also discriminate between hemodialysed and osteoporotic subjects with vertebral fractures. Different characteristics of ultrasound signal can be ascribed to each bone tissue condition, enabling a clear differentiation of bone tissue changes occurring in menopause, osteoporosis and renal osteodystrophy. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Quantitative ultrasound; Dual energy X-ray absorptiometry; Quantitative computed tomography; Uremic and osteoporotic patients

1. Introduction Renal osteodystrophy is a common complication of end-stage renal failure patients on hemodialysis. It comprises a variety of abnormal bone remodelling processes, including high bone turnover disorders, such as osteitis fibrosa, or low bone turnover, such as aluminium osteopathy [1,2].



Corresponding author. Tel.: +39 0882 410686; fax: +39 0882 453861. E-mail address: guglielmi [email protected] (G. Guglielmi).

0720-048X/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2006.05.001

Renal osteodystrophy is a serious disease that strongly influences the morbidity and mortality of hemodialysed patients [4]; reliable diagnosis of this metabolic bone disease is therefore an important and difficult challenge for the nephrologist [3]. The possibility of a differential diagnosis of low bone mass clinical conditions, caused by osteomalacia, osteitis fibrosa, or the more common osteoporosis, is a very interesting task. It is known that all these pathologies are not differentiable by X-ray-based techniques such as dual energy X-ray absorptiometry (DXA) or quantitative computed tomography (QCT), which are able to measure bone mineral density (BMD) but not bone structure or alterations of the organic bone matrix [5]. Quantitative ultra-

G. Guglielmi et al. / European Journal of Radiology 60 (2006) 108–114

sound (QUS) technique has been proposed as a diagnostic tool to assess both skeletal mass and other qualitative/structural characteristics of bone. QUS has been demonstrated to have a good reproducibility [6,7], high discriminating power between normal and osteoporotic subjects with fractures [6,8–10], and also a fracture predictive power [11,12] similar to DXA. In particular, ultrasound technique applied to the phalanges has proved to be capable of well identifying bone pathologies different from primary osteoporosis: glucocorticoid induced osteoporosis [13], rheumatoid arthritis [14], hyperparathyroidism [15], Cushing’s syndrome [16], osteomalacia [17], osteogenesis imperfecta [18], and more recently genetic and environmental factors in monozygotic twins [19]. An interesting study demonstrated that QUS at the phalanges could differentiate between uremic patients with low bone turnover and high bone turnover [20]. Other studies applying phalangeal QUS technique to hemodialysed patients have shown promising results in terms of discrimination between patients and controls, and of association with parathormon (PTH), alkaline phosphatase, and dialytic age [21–24]. To our knowledge, no studies have been conducted on the capability of QUS at the phalanges to discriminate between uremic and osteoporotic patients. This pilot study was designed in order to address this issue. 2. Subjects and methods 2.1. Subjects Three groups of subjects (38 dialytic women, 16 osteoporotic women with vertebral fractures, 19 non-dialytic and nonfractured women) were recruited from the bone mineral unit of the Department of Radiology at “Casa Sollievo della Sofferenza” Hospital in San Giovanni Rotondo, Italy, where they were being studied for osteoporosis. The groups were matched for age and body mass index (BMI). In the dialytic group, mean duration of dialysis was 77 ± 69 months (range 1–275). Primary renal disease was identified according to the EDTA–ERA (European Dialysis and Transplant Association–European Renal Association) registry code 7. Among the 38 patients, seven suffered from glomerular diseases, eight from diabetes, four from polycystic kidney disease, eight from vascular disease, six from interstitial disease, and the remaining six from nephropathy of unknown origin. The diagnosis of osteoporosis was defined according to the WHO criteria [25]. The causes of secondary osteoporosis were excluded in all subjects; these had never been treated with drugs known to influence bone metabolism, nor did they receive any calcium or vitamin D. A lateral X-ray of the thoracolumbar spine was performed for all subjects to assess the presence of fractures. Vertebral fractures were diagnosed on visual inspection using the semiquantitative method previously described by Genant et al. [26]. According to this technique, fractures assessed on lateral thoracolumbar spine radiographs were defined as reductions of more than approximately 20% in anterior, middle, or posterior vertebral height and were graded by severity. The non-dialytic and non-fractured women were recruited in our outpatient clinic among blood donors. In this

109

group we excluded subjects who were taking drugs or had been affected by diseases influencing bone metabolism. On all subjects the following measurements were performed: spinal BMD by QCT and DXA, Femoral BMD by DXA, phalangeal QUS. The study was approved by our ethic review committee, and informed consent was obtained from all subjects. 2.2. Quantitative ultrasound measurements For QUS measurements, the DBM Sonic (IGEA, Italy) was applied to the metaphysis of the proximal phalanges of the last four fingers of the non-dominant hand. The device consists of a transmitting and a receiving probe attached to a caliper that measures the thickness of the phalanx. When the probe receives the ultrasound propagating through the phalanx, it generates an electric signal which is displayed on the device screen. The amplitude dependent speed of sound (AD-SoS) is the main parameter measured by the device and is calculated by dividing the thickness by the time between emission and reception of the signal by the probes. The reception time is calculated when the received signal reaches a fixed amplitude threshold. The device automatically calculates the average AD-SoS of the four measurements. Further, the analysis of the characteristics of the signal after it has crossed the finger was considered • Ultrasound bone profile index (UBPI), in units: a parameter derived on the basis of statistical models applied to a population of more than 10,000 women, using information on the patient’s risk of fracture at the moment of the examination [6]. • Fast wave amplitude (FWA), in mV: a measure of the amplitude of the first ultrasound signal reaching the transducer after transmission through the phalanx [6]. • Bone transmission time (BTT), in ␮s: the time window between the ultrasound signal transmitted through the bone tissue and the signal transmitted through the soft tissue; it is therefore independent of the soft tissue thickness. • Speed of sound (SoS), in m/s: the velocity of transmission of the ultrasound pulse through the phalanx independent of amplitude. 2.3. BMD measurements BMD was measured using anterior posterior dual X-ray absorptiometry (DXA, in g/cm2 ) at the lumbar spine (L2–L4) and at the hip with a Norland XR 26 scanner (Norland Instruments, USA). The standard procedures supplied by the manufacturer for scanning and analysis were used. Reference values for calculation of T-scores were used from the NHANES-III normative database. Single-energy quantitative CT measurement, in mg/cm3 , of the trabecular bone of the L1–L4 vertebrae of the spine was obtained using a Toshiba Xpeed CT (Toshiba Medical Systems Division, Japan) according to the method developed by Cann and Genant [27]. All fractured vertebral bodies identified on lat-

<0.0001 <0.005 <0.005 <0.05 <0.005 <0.05 n.s. n.s. <0.005 <0.0005 <0.05 <0.05 <0.05 n.s. n.s. 52 0.17 10 0.20 48 ± ± ± ± ± 1990 0.62 45 1.34 1988 54 0.15 7 0.30 66 ± ± ± ± ± 75 0.17 8 0.16 41 1925 0.53 42 1.15 1929 AD-SoS (m/s) UBPI (units) FWA (mV) BTT (␮s) SoS (m/s)

± ± ± ± ±

1879 0.42 35 1.17 1926

<0.05 <0.005 <0.05 <0.05 n.s. n.s. n.s. n.s. <0.0001 <0.005 n.s. n.s. 56 0.16 0.12 0.11 ± ± ± ± 138 0.90 0.74 0.63 25 0.13 0.08 0.09 ± ± ± ± 26 0.17 0.11 0.11 158 0.88 0.68 0.58 Spine BMD (QCT) (mg/cm3 ) L2–L4 BMD (DXA) (g/cm2 ) Neck BMD (DXA) (g/cm2 ) Troch BMD (DXA) (g/cm2 )

± ± ± ±

89 0.73 0.61 0.53

n.s. n.s. n.s. n.s. n.s. n.s. <0.05 n.s. n.s. n.s. n.s. n.s. 7 5 8 3.2 ± ± ± ± 59 156 66 27.2 7 7 6 4.0 ± ± ± ± 62 156 64 26.5 8 5 15 6.0

Control group (N = 19) mean ± S.D. Osteoporotic fractured women (N = 16) mean ± S.D. Dialytic patients (N = 38) mean ± S.D.

The short term in vivo precision errors, expressed as a coefficient of variation in percentage (CV%), were 0.5% for AD-SoS, 2.2% for UBPI, 16.5% for SDy and 1.8% for BTT in our centre [9]. The in vivo reproducibility of BMD by DXA and QCT was respectively 1.0% and 1.8% in our laboratory [10]. According to selection criteria there was no difference among the three groups for BMI, age and height; in the control group, subjects were significantly weighted with respect to dialytic patients (p < 0.05) (Table 1). All patients in the osteoporotic group had atleast one fracture (mean 1.8; range 1–3). Only three patients had fracture of L1 and this vertebra was excluded from the BMD evaluation. All the other subjects presented fractures at dorsal spine. Osteoporotic women with vertebral fractures showed significantly lower values of spinal BMD by QCT and DXA with respect to hemodialytic patients. All QUS values, except for BTT and SoS, showed lower values in osteoporotic women with respect to hemodialytic patients. Control group showed higher values of AD-SoS, BTT and SoS than hemodialytic patients while the two groups did not differ for BMD values measured with both QCT and DXA.

Variables

3.1. Precision

Table 1 Descriptive statistics of the three groups, compared by ANOVA and Bonferroni post hoc test

3. Results

t-test dialytic vs. osteoporotic (p value)

Differences of anthropometric or densitometric data among the three groups were assessed by one-way ANOVA and Bonferroni post hoc analysis. Bivariate associations between variables in hemodialytic patients were assessed by Pearson correlation or Spearman Rank test as appropriate. A p value of 0.05 was considered statistically significant. In hemodialytic patients multiple regression analysis was performed to evaluate the association of age, BMI, years since menopause, duration of dialytic treatment, alkaline phosphatase, osteocalcin, and PTH (independent variables) on bone mass measured at different sites and with different devices or ultrasonographic parameters (dependent variable). Stepwise backward method was used to optimise the linear model and a p ≤ 0.10 was considered as significant in the multiple regression analysis.

± ± ± ±

2.5. Statistical analysis

59 154 59 24.8

t-test dialytic vs. control (p value)

Total alkaline phosphatase activity was determined by a multichannel autoanalyzer. Serum intact PTH levels were measured by a two-site chemoluminometric immunoassay (Chiron Diagnostics, East Walpole, MA, USA) and serum bone osteocalcin was assayed by immunoradiometric assay for the intact molecule (ELSA-OST-NAT, Cis Biointernational, Gifsur-Yvette, France); intra- and inter-assay coefficients of variation were 3.8% and 4.7%, respectively.

Age (years) Height (cm) Weight (kg) BMI (kg/m2 )

t-test osteoporotic vs. control (p value)

2.4. Biochemical determinations

BMI: body mass index; AD-SoS: amplitude dependent speed of sound; UBPI: ultrasound bone profile index; FWA: fast wave amplitude; BTT: bone transmission time; SoS: speed of sound; QCT L1–L4: quantitative computed tomography; DXA L1–L4: dual X-ray absorptiometry measured at L1–L4; DXA FN: dual X-ray absorptiometry measured at femoral neck; DXA Troch: dual X-ray absorptiometry measured at great trochanter.

<0.0001 <0.05 <0.05 <0.05 <0.001

ANOVA (p value)

eral thoracolumbar spine radiographs were excluded from data analysis for both DXA and QCT measurements.

<0.001 <0.01 n.s. n.s.

G. Guglielmi et al. / European Journal of Radiology 60 (2006) 108–114

n.s. n.s. n.s. n.s.

110

G. Guglielmi et al. / European Journal of Radiology 60 (2006) 108–114

Fig. 1. Mean values and standard deviation of BTT in the analysed groups. * p < 0.05 with respect to hemodialysed group.

Fig. 2. Mean values and standard deviation of UBPI in the analysed groups. * p < 0.05 with respect to hemodialysed group.

Lastly, osteoporotic women with vertebral fractures showed significantly lower values of spinal BMD by QCT and DXA, and all QUS values with respect to control subjects. All values are reported in Table 1. It is interesting to note that BTT and SoS were not able to distinguish between hemodialysed patients and osteoporotic fractured subjects, whereas they did discriminate hemodialysed patients and controls with similar BMD (Fig. 1). On the contrary, UBPI and FWA, which were able to distinguish between hemodialysed patients and osteoporotic fractured subjects, were not able to discriminate between hemodialysed patients and controls (Fig. 2).

111

Fig. 3. Mean values and standard deviation of AD-SoS in the analysed groups. * p < 0.05 with respect to hemodialysed group.

UBPI and FWA data show a similar behaviour to the DXA and QCT results, whereas BTT and SoS show a completely different behaviour. AD-SoS is the only parameter able to discriminate effectively among the three groups (Fig. 3). Bivariate association between clinical variables and densitometric data in hemodialytic patients are reported in Table 2. Significant associations were found between densitometric variables and age or years since menopause; QCT and femoral neck BMD also correlate with months of dialysis; SoS and femoral neck BMD correlate with alkaline phosphatase. Multiple regression analyses showed that age, years since menopause and BMI are the most important predictors of all DXA BMDs and of UBPI; age, months of dialysis and PTH are the significant independent predictors of AD-SoS; age and PTH are the significant independent predictors of spinal QCT (Table 3). 4. Discussion This preliminary analysis has shown that all QCT, DXA and QUS techniques are able to distinguish between the osteoporotic subjects with vertebral fractures and the control group of women without vertebral fractures. It is clear that bone mineral density is sensibly lower in osteoporotic fractured subjects with respect to controls, and this is the main reason for the observed difference between the groups.

Table 2 Bivariate association between clinical variables and densitometric data of haemodialytic patients

AD-SoS UBPI FWA BTT SoS QCT L1–L4 DXA L1–L4 DXA FN DXA TROCH

Age

BMI

YSM

Months of dialysis

PTH

Alkaline phosphatase

Osteocalcin

−0.55** −0.67** −0.69** −0.32* −0.45** −0.75** −0.41** −0.71** −0.55**

−0.01 0.10 0.11 0.04 −0.05 −0.04 0.24 0.10 0.18

−0.16 −0.36* −0.40* 0.08 0.05 −0.67** −0.38* −0.56** −0.50**

0.05 0.23 0.20 −0.03 −0.04 0.46** 0.25 0.34* 0.29

−0.10 0.05 0.04 −0.26 −0.29 −0.01 −0.19 −0.26 −0.26

−0.28 −0.08 −0.10 −0.25 −0.33* −0.04 −0.17 −0.34* −0.28

−0.16 −0.09 0.04 −0.38 −0.23 −0.02 −0.09 −0.10 −0.06

YSM: years since menopause; BMI: body mass index; PTH: parathormon; AD-SoS: amplitude dependent speed of sound; UBPI: ultrasound bone profile index; FWA: fast wave amplitude; BTT: bone transmission time; SoS: speed of sound; QCT L1–L4: quantitative computed tomography; DXA L1–L4: dual X-ray absorptiometry measured at L1–L4; DXA FN: dual X-ray absorptiometry measured at femoral neck; DXA Troch: dual X-ray absorptiometry measured at great trochanter. * p < 0.05. ** p < 0.01.

112

G. Guglielmi et al. / European Journal of Radiology 60 (2006) 108–114

Table 3 Multivariate analysis for densitometric variables as dependent variables: age, BMI, years since menopause, months of dialysis, PTH, osteocalcin were independent variables introduced in the model, that was optimised by backward stepwise method Dependent variable

Independent variables

Partial r

Significance p

Overall r

Lumbar spine BMD (QCT)

Age PTH

−0.61 −0.25

0.0001 0.099

0.62

Lumbar spine BMD (DXA)

Age Years since menopause BMI

−0.28 0.56 0.36

0.069 0.0001 0.019

0.59

Neck BMD (DXA)

Age Years since menopause BMI Months of dialysis

−0.53 0.49 0.44 −0.34

0.0001 0.001 0.004 0.027

0.706

Trochanter BMD (DXA)

Age Years since menopause BMI

−0.35 0.64 0.36

0.020 0.0001 0.018

0.671

AD-SoS (QUS)

Age Months of dialysis PTH

−0.37 −0.31 −0.26

0.013 0.040 0.081

0.460

UBPI (QUS)

Age Years since menopause BMI

−0.41 0.32 0.26

0.006 0.032 0.089

0.495

Only significant variables (p < 0.10) are reported in the final model, together with partial correlation coefficient, significance and overall correlation coefficient of the final model.

In our study we investigated whether the abnormalities occurring in the bone tissue of hemodialysed patients – not only in terms of bone mineral density – could be assessed by QUS technique. In order to address this aim we compared the control group with the hemodialysed group. Taking into account X-ray-based densitometric techniques, these two groups, matched for age and BMI, were not different at all the investigated sites (spine and femur) in our sample. Studies on BMD in hemodialysed patients have shown contrasting results: BMD was found to be normal or partially decreased in appendicular skeleton, whereas it was increased at axial level [28–30]. When compared by means of QUS parameters, it was observed that the parameters BTT and SoS – the only two QUS parameters not dependent on amplitude, but only on the rapidity of the ultrasound wave to propagate through the bone tissue – were significantly lower in the hemodialysed group with respect to the control group of similar BMD. Similar results were obtained by Luisetto et al. [17] who compared a group of osteoporotic and osteomalacic women with similar BMD values as obtained by DXA at lumbar spine. They found lower values of velocity (not dependent on amplitude) in osteomalacic patients in comparison with osteoporotic patients. Since an abnormal bone metabolism in hemodialysed patients often leads to osteomalacia, the results obtained by Luisetto et al. [17] are in agreement with our findings. It is also interesting to note that the investigated QUS parameters behave in different ways: some of them, UBPI and FWA, which are influenced by the amplitude of the US signal and therefore are linked to the attenuation of the ultrasound pulse,

have the same power as the QCT and DXA parameters in discriminating dialysed from osteoporotic fractured subjects; while others, BTT and SoS, which are influenced only by the velocity of transmission of the ultrasound pulse through the bone and are thus linked to the elastic properties of the material of the bone, have a completely different behaviour, i.e. they do not discriminate between hemodialysed patients and fractured subjects, but do discriminate between dialytic patients and control group with similar BMD values. These observations seem to show that the attenuation of the ultrasound pulse is closely related to the density and structural characteristics of the bone, while the velocity and rapidity of propagation are more related to the elastic properties of the bone, and this can be observed in a clearer way in a clinical condition of two bones of similar density but different elastic properties, such as dialytic and normal bone. Lastly, AD-SoS, which is dependent on both velocity and attenuation, is the only parameter that can significantly discriminate among the three groups of subjects investigated. In several studies the high discriminating ability of AD-SoS between fractured and non-fractured women has been successfully tested [6,8–10]; other studies have also demonstrated that hemodialysed patients have lower AD-SoS values with respect to control subjects [20,24]. No studies have been conducted on various QUS parameters in hemodialytic populations. In this study we were able to investigate the behaviour of several QUS parameters in osteoporotic and hemodialysed patients in comparison with normal non-fractured subjects; we thus identified the characteristics of the ultrasound signal that alter in each of these three groups. From the clinical point of view, it is important to be able to identify by QUS at the phalanges

G. Guglielmi et al. / European Journal of Radiology 60 (2006) 108–114

some important changes in the elastic properties of the bone in dialytic patients which could not be revealed by DXA or QCT techniques, which usually measure only bone mineral density, without taking into account variation in bone tissue composition. And the changes in bone tissue composition are extremely frequent in dialytic patients, leading to higher fragility of the bones and predisposition to fractures. The correlation analysis of densitometric parameters with clinical variables in uremic patients has shown a close relationship of all QCT, DXA and QUS with age and years since menopause, which are therefore the main determinants of bone mass also in hemodialysed subjects; only spinal QCT correlates with duration of dialysis, and only SoS and femoral neck DXA correlate with alkaline phosphatase, but the correlation coefficients are low. In the multivariate regression model we observed that age, years since menopause and BMI are the significant determinants of the majority of densitometric variables (lumbar spine, femoral neck, trochanter, UBPI) while the significant determinants of AD-SoS are age, months of dialysis and PTH. Again, AD-SoS seems to be more specifically related to the hemodialytic condition, being correlated with duration of dialysis and PTH, indices of progression of the disease. Similar correlation coefficients were also found by Montagnani et al. [20], Pluskiewicz et al. [24] and Przedlacki et al. [22] with phalangeal QUS, confirming the reliability of our results. The main limitation of the study is the low number of subjects investigated, especially in the control and osteoporotic groups; and since our main interest in the study was the dialytic population we were unable to find a large cohort of postmenopausal women with those particular characteristics in our centre. We may conclude that phalangeal QUS can discriminate hemodialysed patients from controls with similar bone mineral density and can also discriminate between hemodialysed patients and osteoporotic subjects with vertebral fractures. Different characteristics of the ultrasound signal can be ascribed to each bone tissue condition, allowing a clear differentiation of bone tissue changes occurring in menopause, osteoporosis and renal osteodystrophy. Phalangeal QUS could be used together with DXA to better characterise the bone tissue in terms not only of density but also of elasticity and, ultimately, fragility, in various osteopenizing pathologies, such as osteoporosis and renal osteodystrophy.

References [1] Hory B, Drieke TB. The parathyroid-bone axis in uremia: new insight into old questions. Curr Opin Nephrol Hypertens 1977;6:40–8. [2] Sherrard DJ, Herez G, Pei Y, et al. The spectrum of bone disease in endstage renal failure—an evolving disorder. Kidney Int 1993;43:436–42. [3] Malluche HH, Faugere MC. Renal bone disease 1990: an unmet challenge for the nephrologist. Kidney Int 1990;38:193–211. [4] Malluche HH, Sawaya BP, Faugere MC. Dialysis: current status, contemporary limitations and future challenges. Kidney Int 1995;48(Suppl.):S37–9. [5] Guglielmi G, de Terlizzi F, Aucella F. Quantitative bone ultrasonography: state of the art and perspectives. G Ital Nefrol 2004;21:343–54. [6] W¨uster C, Albanese C, De Aloysio D, et al. Phalangeal osteosonogrammetry study (PhOS): age related changes, diagnostic sensitivity and discrimination power. J Bone Miner Res 2000;15(8):1603–14.

113

[7] Soballa T, Wuster C, Schlegel J, et al. Ultrasound transmission speed and ultrasound bone profile score (UBPS) of the phalanxes in normal women with osteoporosis. Horm Metab Res 1998;30:536– 41. [8] Drozdzowska B, Pluskiewicz W, de Terlizzi F. The usefulness of quantitative ultrasound at the hand phalanges in the detection of the different types of non-traumatic fractures. Ultrasound Med Biol 2003;29(11):1545–50. [9] Guglielmi G, Njeh CF, de Terlizzi F, et al. Phalangeal quantitative ultrasound, phalangeal morphometric variables, and vertebral fracture discrimination. Calcified Tissue Int 2003;72:469–77. [10] Guglielmi G, Cammisa M, De Serio A, et al. Phalangeal US velocity discriminates between normal and vertebrally fractured subjects. Eur Radiol 1999;9:1632–7. [11] Mele R, Masci G, Ventura V, de Aloysio D, Bicocchi M, Cadossi R. Threeyear longitudinal study with quantitative ultrasound at the hand phalanx in a female population. Osteoporosis Int 1997;7(6):550–7. [12] Gluer CC, Eastell R, Reid DM, et al. Association of five quantitative ultrasound devices and bone densitometry with osteoporotic vertebral fractures in a population-based sample: the OPUS study. J Bone Miner Res 2004;19(5):782–93. [13] Wuster C. The use of quantitative ultrasonometry in patients with glucocorticoid-induced osteoporosis. Front Horm Res 2002;30:20– 36. [14] Roben P, Barkmann R, Ullrich S, Gause A, Heller M, Gl¨uer CC. Assessment of phalangeal bone loss and erosions in patients with rheumatoid arthritis by quantitative ultrasound. Ann Rheum Dis 2001;60(7): 670–7. [15] Montagnani A, Gonnelli S, Cepollaro C, et al. Graphic trace analysis of ultrasound at the phalanges may differentiate between subjects with primary hyperparathyroidism and with osteoporosis: a pilot study. Osteoporosis Int 2002;13(3):222–7. [16] Tauchmanova L, Rossi R, Nuzzo V, et al. Bone loss determined by quantitative ultrasonometry correlates inversely with disease activity in patients with endogenous glucocorticoid excess due to adrenal mass. Eur J Endocrinol 2001;145(3):241–7. [17] Luisetto G, Camozzi V, de Terlizzi F. Use of quantitative ultrasonography in differentiating osteomalacia from osteoporosis: preliminary study. J Ultrasound Med 2000;19(4):251–6. [18] Cepollaro C, Gonnelli S, Pondrelli C, et al. Osteogenesis imperfecta: bone turnover, bone density, and ultrasound parameters. Calcified Tissue Int 1999;65:129–32. [19] Guglielmi G, de Terlizzi F, Torrente I, Mingarelli R, Dallapiccola B. Quantitative ultrasound of the hand phalanges in a cohort of monozygotic twins: influence of genetic and environmental factors. Skeletal Radiol 2005;34:727–35. [20] Montagnani A, Gonnelli S, Cepollaro C, et al. Quantitative ultrasound in the assessment of skeletal status in uremic patients. J Clin Densitom 1999;2(4):389–95. [21] Rico H, Aguado F, Revilla M, Villa LF, Martin J. Ultrasound bone velocity and metacarpal radiogrametry in hemodialyzed patients. Miner Electrolyte Metab 1994;20(3):103–6. [22] Przedlacki J, Pluskiewicz W, Wieliczko M, et al. Quantitative ultrasound of phalanges and dual-energy X-ray absorptiometry of forearm and hand in patients with end-stage renal failure treated with dialysis. Osteoporosis Int 1999;10(1):1–6. [23] Pluskiewicz W, Adamczyk P, Drozdzowska B, et al. Skeletal status in children, adolescents and young adults with end-stage renal failure treated with hemo- or peri-toneal dialysis. Osteoporosis Int 2002;13(5):353–7. [24] Pluskiewicz W, Przedlacki J, Drozdzowska B, Wlodarczyk D, Matuszkiewicz-Rowinska J, Adamczyk P. Quantiative ultrasound at hand phalanges in adults with end-stage renal failure. Ultrasound Med Biol 2004;30(4):455–9. [25] Anone. Consensus development conference: diagnosis, prophylaxis and treatment of osteoporosis. Am J Med 1993;94:646–50. [26] Genant HK, Jergas M, Palermo L, et al. Comparison of semiquantitative morphometric assessment of prevalent and incident vertebral fractures in osteoporosis: the study of osteoporosis fractures research group. J Bone Miner Res 1996;11:984–96.

114

G. Guglielmi et al. / European Journal of Radiology 60 (2006) 108–114

[27] Cann C, Genant HK. Precise measurement of vertebral mineral content using computed tomography. J Comput Assist Tomogr 1980;4:493–500. [28] Lechleitner P, Krimbacher E, Genser N, et al. Bone mineral densitometry in dialyzed patients: quantitative computed tomography versus dual photon absorptiometry. Bone 1994;15(4):387–91.

[29] Bianchi ML, Colantonio G, Montesano A, et al. Bone mass status in different degrees of chronic renal failure. Bone 1992;13: 225–8. [30] Chan TM, Pun KK, Cheng IK. Total and regional bone densities in dialysis patients. Nephrol Dial Transplant 1992;7:835–40.