This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ryan, P J Search for Related Content PubMed PubMed Citation Articles by Ryan, P J

Bone densitometry in the management of Colles' fractures: which site to measure?

Osteoporosis Unit, Department of Nuclear Medicine, Medway Maritime Hospital, Windmill Road, Gillingham, Kent ME7 5NY, UK

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
This study assessed whether patients with Colles' fractures can be evaluated by measurement of peripheral bone mineral density (BMD) instead of the more established axial BMD technique. 205 consecutive patients who had sustained a recent Colles' fracture were invited to attend for BMD measurement at the lumbar spine, hip and forearm by dual energy X-ray absorptiometry (DXA) using a Hologic QDR4500C instrument. Hologic normal reference ranges were used for spine and ultradistal radius BMD, and the NHANES 111 reference range was used for the femoral neck. 158 patients attended, of whom only 61 had a T-score below -2.5 at one of the three measured sites (84 T<-2.0, 112 T<-1.5). 36 patients had bilateral forearm fractures and did not undergo a forearm measurement. After excluding these patients and using a T-score of -2.5 as a diagnostic threshold, a spine and hip measurement strategy missed 26% of osteoporotic patients and a forearm only strategy missed 33%. If a forearm strategy is adopted, with axial BMD performed only on those patients with an ultradistal radial T-score of <-1.5 and >-2.5, 27 further scans would be required and only four patients with T<-2.5 at the spine or hip would be undetected. Where axial BMD measurement is performed, this should be accompanied by forearm BMD measurement in some patients with Colles' fractures. A peripheral bone mineral densitometry strategy using forearm BMD alone will miss many individuals with osteoporosis. Therefore where forearm, and probably any peripheral scanning method, only is used for patient assessment there should be close liaison with an axial BMD centre with agreed joint protocols, as some patients should be advised to have an additional axial BMD measurement following their peripheral BMD measurement.

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
The distal forearm, along with the spinal vertebrae and hip, is a common site for osteoporotic fracture. Epidemiological studies have shown that patients with Colles' fractures have an increased risk of other fractures at typical osteoporotic sites [1, 2]. However, studies using bone densitometry have indicated that not all patients with Colles' fractures have osteoporosis [3]. This is supported by epidemiological models [4]. With respect to bone density it is recognized that the widely used diagnostic threshold of T<-2.5 does not encompass all individuals whose fractures are at least partly related to low bone mass. Furthermore, the threshold of T<-2.5 is not necessarily a therapeutic threshold. Some reviews have advised intervention at higher bone mineral density (BMD) values [5]. Conversely, in all patients the value of therapy must be considered in the wider clinical context, which may mean some drug treatments are not advocated despite low bone mass.

Patients with distal forearm fractures deserve further assessment to determine whether they would benefit from osteoporosis therapy. In many centres this is mostly by bone mineral densitometry of the spine and hip using dual energy X-ray absorptiometry (DXA). There has, however, been a recent proliferation of peripheral measurement devices, some utilizing DXA and others using different techniques such as ultrasound. Peripheral devices clearly have utility in assessing future fracture risk but their results relate poorly to axial BMD measurements [6, 7]. This is partly due to biological factors as, although osteoporosis is a generalized condition, there is marked heterogeneity of the skeleton, but is also owing to measurement issues such as accuracy. As a consequence, it is not uncommon to find patients with osteoporosis at one measured site but not at others [6]. Most centres performing axial bone mineral densitometry measure at least two sites, which helps reduce this problem, although additional sites of measurement would inevitably detect more osteoporotic individuals. Peripheral measurements generally measure a single site and are more likely to produce misleading results, such as reassuring the patient who has a normal result but has osteoporosis elsewhere in the skeleton. Some studies have considered that peripheral devices could be used as a pre-screen for axial BMD measurement, but this has undergone only limited assessment in mixed groups of patients [8]. Since different conditions preferentially affect different skeletal sites, it is premature to consider that peripheral BMD could be used in all patients who are suspected to have osteoporosis.

This study was performed to assess the relative merits of axial and forearm BMD in patients presenting with Colles' fractures. The chosen patients had a common clinical presentation and formed a group in which peripheral measurement should have high clinical utility.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Patients with typical osteoporotic, or any low trauma, fractures were invited by letter to attend for bone densitometry as part of a new approach to the management of osteoporosis, called the Medway Fracture Initiative. Letters from orthopaedic consultants regarding all patients attending the fracture clinic were copied to an osteoporosis specialist nurse who analysed them to select patients for follow-up. All patients with Colles' fractures were invited to attend for bone densitometry. This work examines the results on a series of consecutive patients with Colles' fractures who were invited to attend the osteoporosis unit. Those who accepted the offer had bone mineral densitometry within 3 months of injury. Measurements were made by DXA using a Hologic QDR 4500C instrument (Hologic, Waltham, MA) at the lumbar spine (L1–L4), femoral neck and contralateral ultradistal radius. Results were compared with normal, young adult reference ranges and expressed as T-scores. Hologic reference ranges were used at the spine and forearm, with the widely accepted NHANES 111 reference range used at the femoral neck. Analysis was made of the number of osteoporotic individuals (T<-2.5) detected by axial BMD (spine and hip) and forearm BMD measurement. This analysis was also performed for different T-score cut-off values. The merits of BMD measurements at axial sites following a single forearm measurement vs an axial only measurement strategy was examined.

	Results

Top Abstract Introduction Methods Results Discussion References

 
205 patients were invited to attend the osteoporosis unit for bone densitometry and advice on osteoporosis where relevant. Of those invited, 158 patients, 143 female and 15 male, attended for measurement, with an average age of 65.5 years (SD 12.2). Patients not attending were generally older and either unable to attend or did not wish to come to the osteoporosis unit. The number of patients in different age groups is shown in Figure 1. As expected, the vast majority of patients were over 50 years of age. In 36 patients the forearm site was not examined owing to a history of bilateral forearm fracture. Of these, 18 patients were osteoporotic at the spine or hip (T-2.5), 21 had a T-score -2.0 and 26 had a T-score -1.5. When considering all three sites, 61 of 158 patients had a T-score -2.5, 84 T-2.0 and 112 T-1.5. Tables 1 and 2 contain a summary of the numbers of patients below different T cut-off scores at different sites or combinations of sites. Tables 3 and 4 show the same data after exclusion of the 36 patients without forearm values. Table 5 contains summary data from patients with measurements at all three sites, examining how many would fail to be classified as being below a particular T cut-off score using different measurement strategies.

View larger version (23K): [in this window] [in a new window]   Figure 1. Patient demographics, divided into those with T-2.5 and those with T>-2.5 at any of the measured sites (spine, hip or forearm). pts, patients.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Other possible strategies are suggested by the results of this work. Using the widely advocated T<-2.5 as a treatment threshold [9] in a centre performing spine and hip DXA, additional forearm measurements are only required in those in whom the T-score at the spine or hip is -1.5 and -2.0. With this approach only six patients in whom T<-2.5 at the forearm would have remained undetected, with 30 extra forearm scans required. This approach presupposes analysis of the results at the spine and hip before a decision on the forearm measurement is made. It may be simpler to perform forearm scans on all patients with Colles' fractures.

For centres performing forearm densitometry only, results from this study suggest that only patients in whom -2.5T-1.5 would require axial scanning. This is a modest fraction of the overall patients. This study implies only an extra 27 scans and only four patients with T<-2.5 at the spine or hip that would be undetected. Similar conclusions can be drawn for other T-score thresholds that may be used to decide which patients require therapy. This presupposes that a T<-2.5 threshold, or indeed any other forearm BMD threshold, is used in a similar way for diagnosis and therapy decisions as measurements at axial sites. While this may be reasonable for forearm values, it may be less so at other peripheral sites. In most patients, a further weakness of the approach of using peripheral measurements alone is the poor utility of peripheral sites for monitoring therapies such as hormone replacement therapy or bisphosphonates, where changes in bone mass with treatment are generally below the threshold required for confidence that a significant change has occurred.

It must be remembered that T-score thresholds do not identify an equivalent percentage of the population at each skeletal site for a variety of reasons, such as differing rates of alteration of BMD with age, different spreads of values and approaches to smoothing data. It is therefore a major assumption that the same T-score threshold can be used for diagnostic and treatment decision at different sites [10, 11], although this is generally not considered in practice.

With regard to peripheral scanning, the above approach does not necessarily apply in exactly the same way to other forearm scanners or other peripheral devices. However, a similar method is likely to be appropriate, with cut-off values varying for different devices. Each provider of peripheral scanning would therefore need to form a joint approach with their nearest axial provider. Whether the same cut-off values would apply in different clinical scenarios would require further examination. Patients with vertebral fractures appear to have lower spine measurements than forearm measurements [12]. Some secondary causes of osteoporosis appear to affect the skeleton differently from post-menopausal osteoporosis. For example, in some studies oral corticosteroids preferentially affect the spine and ribs compared with the peripheral radius [13], primary hyperparathyroidism predominately affects cortical bone and forearm BMD compared with spine BMD [14], and anorexia nervosa can produce osteoporosis with little disturbance of ultrasound parameters [15].

It is of interest in this study that few osteoporotic patients were detected at the femoral neck, which is increasingly advocated as the most important site of measurement, with some authors suggesting a total hip T-score of -2.5 be used to define osteoporosis [16]. This view has been supported by data from bisphosphonate trials of alendronate and risedronate indicating that intervention with medication produces a reduction in fracture risk only in those with a T-score of -2.5 at the hip [17]. The NHANES 111 reference range is widely regarded as the best acquired reference data but produces lower normal values than the Hologic reference range, which may be closer to UK reference values [18, 19]. Certainly using the hip measurement alone, as advocated by some authors, would have yielded a surprisingly small number of osteoporotic individuals. This is in contrast to the study by Earnshaw et al [3] in which more patients were osteoporotic at the femoral neck than at the spine or radius. Based on the present data, a case could be made for omitting the hip measurement and performing spine and forearm bone densitometry only.

47 patients did not attend for densitometry and it is possible that these would have attended for a peripheral measurement if available more locally. Axial bone densitometry providers may need to recognize this. Our work suggests, however, that if peripheral scanning devices are used there are some patients who must be referred for axial BMD measurement, and peripheral bone densitometry providers must form a relationship with axial providers to ensure that there are appropriate protocols. Ignoring this approach is not acceptable, as some patients will be falsely reassured when they may otherwise be suitable for one of the proven therapies that can prevent fractures. Conversely, axial providers also need to consider forearm BMD as part of the investigation of Colles' fracture patients. Further consideration of the widespread use of the NHANES 111 hip reference range is advised, as this may not be appropriate in a UK population.

Received for publication January 24, 2001. Revision received July 23, 2001. Accepted for publication August 13, 2001.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Lauritzen JB, Schwarz P, McNair P, Lund B, Transbol I. Radial and humeral fractures as predictors of subsequent hip, radial or humeral fractures in women, and their seasonal variation. Osteoporosis Int 1993;3:133–7.[Medline]
2. Wasnich RD, Davis JW, Ross PD. Spine fracture is predicted by non-spine fractures. Osteoporosis Int 1994;4:1–5.
3. Earnshaw SA, Caute SA, Warley A, Hosking DJ. Colles' fractures of the wrist as an indicator of underlying osteopenia in post menopausal women. A prospective study of bone mineral density and bone turmover rate. Osteoporosis Int 1998;8:53–6.[Medline]
4. Melton LJ 3rd, Thamer M, Ray NF, et al. Fractures attributed to osteoporosis: report from the National Osteoporosis Foundation. J Bone Miner Res 1997;12:16–23.[Medline]
5. Eddy DM, Johnstone Jr C-C, Cummings SR, et al. Osteoporosis. Review of the evidence for prevention, diagnosis, treatment and cost effectiveness analysis. Osteoporosis Int 1998;8(Suppl.4):S7–80
6. Mole PA, McMurdo MET, Paterson CR. Evaluation of peripheral dual energy x-ray absorptiometry: comparison with single photon absorptiometry of the forearm and dual energy absorptiometry of the spine or femur. Br J Radiol 1998;71:427–32.[Abstract]
7. Patel P, Blake GM, Jefferies A, Sauterean Chaudley PM, Fogelman I. A comparison of a peripheral DXA system with conventional densitometry of the spine and femur. J Clin Densitom 1998;1:235–44.
8. Jones T, Davie MWJ. Bone mineral density of the distal forearm can identify patients with osteoporosis at spine or femoral neck. Br J Rheumatol 1998;37:539–43.[Abstract/Free Full Text]
9. Compston JE. Bone densitometry and decision making. J Clin Densitom 1998;2:5–9.
10. Arlot ME, Sornay Rendu E, Farnero P, Vey Martz B, Delmas PD. Apparent pre and post menopausal bone loss evaluated by DXA at different sites. Women in the Offly cohert. J Bone Miner Res 1997;12:683–90.[Medline]
11. Faulkner KG, Von Stetten E, Miller P. Discordance in patient classification using T scores. J Clin Densitom 1999;2:343–50.[Medline]
12. Ryan PJ, Blake GM, Herd R, Parker J, Fogelman I. Post menopausal vertebral osteoporosis: can dual energy X-ray absorptiometry forearm bone density substitute for axial measurements? Br J Rheumatol 1994;33:546–9.[Abstract/Free Full Text]
13. Sambrook PN, Jones G. Corticosteroid osteoporosis. Br J Rheumatol 1995;34:8–12.[Abstract/Free Full Text]
14. Wishart J, Horowitz M, Need A, Nordin BE. Relationship between forearm and vertebral mineral density in postmenopausal women with primary hyperparathyroidism. Arch Intern Med 1990;150:1329–31.[Abstract/Free Full Text]
15. Zipfels M, Russel JD, Gross G, Touys S, Herzog W, Beumont PJU. Quantitative digital ultrasound (QUS) and bone mineral density (BMD) in patients with anorexia nervosa. J Bone Miner Res 1999;14(Suppl. 1):S257.
16. Kanis JA, Gluer CC. An update on the diagnosis and assessment of osteoporosis with densitometry. Osteoporosis Int 2000;11:192–202.[Medline]
17. Miller P, Roux C, McClung M, Adani S, Eastell R, Pack S, et al. Risedronate reduces hip fractures in patients with low femoral bone mineral density. Arthritis Rheum 1999;42(Suppl. 9):287.
18. Chen Z, Maricic M, Lund P, Tener J, Gluck O. How the new Hologic hip normal reference values affect the densitometric diagnosis of osteoporosis. Osteoporosis Int 1998;8:423–7.[Medline]
19. Ahmed AIH, Blake GM, Rymer JM, Fogelman I. Screening for osteopenia and osteoporosis. Do the accepted normal ranges lead to over diagnosis? Osteoporosis Int 1997;7:432–8.[Medline]

This article has been cited by other articles:

				M. L. Gourlay and S. A. Brown Clinical Considerations in Premenopausal Osteoporosis Arch Intern Med, March 22, 2004; 164(6): 603 - 614. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ryan, P J Search for Related Content PubMed PubMed Citation Articles by Ryan, P J