Elsevier

Journal of Clinical Densitometry

Volume 12, Issue 1, January–March 2009, Pages 17-21
Journal of Clinical Densitometry

Original Article
Short-Time In Vitro and In Vivo Precision of Direct Digital X-ray Radiogrammetry

https://doi.org/10.1016/j.jocd.2008.10.005Get rights and content

Abstract

Digital X-ray radiogrammetry (DXR) calculates peripheral bone mineral density (BMD) from hand radiographs. The aim of this study was to examine in vitro and in vivo precision for the new direct digital version of DXR, a development of the conventional DXR. The in vitro precision for direct DXR was tested on 4 different X-ray equipments, based on 31 radiographs of the same phantom. The in vivo precision was based on duplicate hand radiographs from both hands in 39 individuals. For the 4 X-ray equipments, in vitro precision ranged from 0.14% to 0.30%, expressed as coefficient of variations (CV%) and from 0.0012 to 0.0028 g/cm2, expressed as smallest detectable difference (SDD). The precision was correlated to the resolution of the radiographic equipment (r = 0.95, p = 0.05). The corresponding values for the in vivo precision for mean values of both hands were: CV% = 0.46%; SDD = 0.0046 g/cm2, and least significant change (LSC%) = 1.28%. The DXR-BMD for 1 of the X-ray equipments differed 1.1% from the overall mean. The precision for direct DXR was highly satisfactory both in vitro and in vivo. DXR-BMD values may differ between the radiographic equipments, and follow-up measurements should be performed with the same X-ray equipment.

Introduction

Dual-energy X-ray absorptiometry (DXA) measurements at hip and spine are considered the gold standard for detection and management of osteoporosis 1, 2, 3. Because DXA is considered as a resource-demanding method, there is a need for less expensive and more feasible methods to measure bone mineral density (BMD). In the past decade, several peripheral BMD measurements have been developed (4), including digital X-ray radiogrammetry (DXR) (5). DXR calculating BMD at metacarpal hand bones on radiographs is a further development of the classic radiogrammetry method first described in 1960 (6). DXR-BMD has been shown to be a good predictor of fractures (7) and radiographic joint damage in rheumatoid arthritis (RA) 8, 9, and is also a response variable for anti-inflammatory treatment 10, 11.

For the DXR equipment that assesses bone density from scanned conventional hand radiographs (Pronosco X-posure, Sectra, LinKöping, Sweden), the short-time precision is shown to be superior compared with DXA assessed at hand, hip, and spine 12, 13. The DXR method has been further developed, and DXR-BMD can now also be analyzed directly from digital hand radiographs (direct DXR).

The precision for direct DXR has not been fully investigated, and the objective of this study was to examine both in vitro and in vivo reproducibility for the direct version of DXR.

Section snippets

Methods

Direct DXR (Sectra, Linköping, Sweden), a further development of the DXR method developed for BMD calculations on conventional hand radiographs, was used to measure hand BMD on radiographs made in digitized format. DXR is a computer version of the traditional radiogrammetry technique (6), and has been described in detail 5, 14. The computer automatically recognizes regions of interest around the narrowest part of the second, third, and fourth metacarpal bone on hand radiographs and measure

In Vitro

The in vitro precision data (variance, CV%, and SDD) for the 4 X-ray equipments are shown in Table 1. The DXR-BMD mean for all measurements was 0.470 mg/cm2. The DXR-BMD mean for the most diverging X-ray equipment, the Sectra MicroDose, differed 1.1% from this overall mean (Table 1 and Fig. 1).

The precision for the 4 X-ray equipments tested ranged from 0.14% to 0.30% for CV% and from 0.0012 to 0.0028 g/cm2 for SDD (Table 1). The resolution and the precision of the equipment correlated

Discussion

The short-time precision, both in vitro and in vivo, for direct DXR-BMD was highly satisfactory. The utility of DXR-BMD in research and clinical practice has been demonstrated in several studies 7, 8, 19, 20, and our observation supports the fact that DXR-BMD is a robust measure.

The in vitro precision for the 4 tested X-ray equipments expressed as CV% ranged from 0.14% to 0.30%. The reproducibility was dependent on the resolution capacity of the tested X-ray equipments as shown in Table 1.

The

References (30)

  • M.L. Bouxsein et al.

    Digital X-ray radiogrammetry predicts hip, wrist and vertebral fracture risk in elderly women: a prospective analysis from the study of osteoporotic fractures

    Osteoporos Int

    (2002)
  • M. Hoff et al.

    Cortical hand bone loss after one year in early rheumatoid arthritis predicts radiographic hand joint damage at 5 and 10 year follow-up

    Ann Rheum Dis

    (2008)
  • A. Stewart et al.

    Predicting erosive disease in rheumatoid arthritis. A longitudinal study of changes in bone density using digital X-ray radiogrammetry: a pilot study

    Rheumatology

    (2004)
  • G. Haugeberg et al.

    Reduced loss of hand bone density with prednisolone in early rheumatoid arthritis: results from a randomized placebo-controlled trial

    Arch Intern Med

    (2005)
  • M. Hoff et al.

    Adalimumab therapy reduces hand bone loss in early rheumatoid arthritis: explorative analyses from the PREMIER study

    Ann Rheum Dis

    (2008)
  • Cited by (27)

    • Imaging in Rheumatic Diseases

      2016, Kelley and Firestein's Textbook of Rheumatology: Volumes 1-2, Tenth Edition
    • Potential sources of quantification error when retrospectively assessing metacarpal bone loss from historical radiographs by using digital X-ray radiogrammetry: An experimental study

      2014, Journal of Clinical Densitometry
      Citation Excerpt :

      DXR-BMD measurements are highly reproducible when conducted in accordance with device labeling. Their coefficient of variation of about 0.4% (8,9) compares favorably with the approx 0.9% of dual-energy X-ray absorptiometry (DXA) of the hand (10), the 0.7% of computed tomography (11), and the 1–3% of DXA hip (12). The high reproducibility of DXR makes it suitable for evaluating treatment effect because very small changes in BMD can be detected with a high statistical power.

    • Tibial subchondral bone mineral density: Sources of variability and reproducibility

      2013, Osteoarthritis and Cartilage
      Citation Excerpt :

      The amplitude of sBMD values was calculated by the highest value obtained for all subjects minus the lowest value. The short-term reproducibility (i.e., immediate reposition) values were analysed using the coefficient of variation (CV)16,32, the standardized CV (sCV)33–35, the RMS CV36–38, the intra-class coefficient of correlation (ICC) calculated with MedCalc software (version 12.3.0.0) with the specific set on “consistency” for one observer as recommended by Fermanian in 200539, the smallest detectable difference (SDD)32 and the least significant changes (LSC)32, according to the formulae in Appendix. The intra-observer reproducibility was analysed using the same formulae.

    • Comparison Between Amputation-Induced Demineralization and Age-Related Bone Loss Using Digital X-Ray Radiogrammetry

      2012, Journal of Clinical Densitometry
      Citation Excerpt :

      The ROIs were displayed on the computer monitor for illustration of the positioning; manual ROI correction was not possible as the system precluded interactive manipulation. Several studies have shown that the used algorithm reliably identifies the diaphysis shafts and also the outer and inner cortical edges of the three middle metacarpals (1,16–21). The outer bone diameter was measured as the entire width of the second, third, and fourth metacarpal bones including the two cortical regions and the medullary cavity.

    • Imaging Modalities in Rheumatic Diseases

      2012, Kelley's Textbook of Rheumatology: Volume 1-2, Ninth Edition
    • The ability of hand digital X-ray radiogrammetry to identify middle-aged and elderly women with reduced bone density, as assessed by femoral neck dual-energy X-ray absorptiometry

      2010, Journal of Clinical Densitometry
      Citation Excerpt :

      A further advantage of the DXR method is the excellent reliability, as both short- and long-term precisions have been shown to be excellent and superior to other quantitative bone measure devices, including DXA (42,43). For DXR cadaver phantoms, ex vivo precision has been reported to be as low as 0.14%, whereas in vivo precision has been found to be 0.46% (43). The significance of a method to be used, for example, in monitoring bone density, depends on the precision of the device, on the size of bone changes over time, and the responsiveness of the variable to treatment.

    View all citing articles on Scopus

    Competing interests: Johan Kälvesten is an employee of the Sectra. None of the other authors have competing interests.

    View full text