Rheumatology Advance Access originally published online on April 20, 2007
Rheumatology 2007 46(6):1036-1037; doi:10.1093/rheumatology/kem087
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
LETTERS TO THE EDITOR |
Do Tc-99m-diphosphonate bone scans have any place in the investigation of polyarthralgia?
1Kennedy Institute of Rheumatology, Imperial College London, 1 Aspenlea Road, London W6 8LH, 2Department of Rheumatology, Charing Cross Hospital, Fulham Palace Road, London W6 8RF and 3Department of Nuclear Medicine, Charing Cross Hospital, Fulham Palace Road, London W6 8RF, UK.
Correspondence to: Dr B. Fisher, Department of Rheumatology Charing Cross Hospital Fulham Palace Road London W6 8RF, UK. E-mail: benjamin.fisher{at}imperial.ac.uk
SIR, polyarthralgia with few clinical signs is a frequent complaint encountered by rheumatologists. In this situation Tc-99m-diphosphonate bone scans (BS) may be requested to rule out inflammatory arthritis. Whether the evidence justifies this approach is open to question. We conducted a retrospective audit to determine if BS influenced our practice. In the 6 months between April and September 2005, 44 BS were requested by our department. Of these, 25 were for investigation of polyarthralgia. Despite having equal numbers of lists, 20 of these were requested by specialist registrars and only 5 by consultants. Nuclear Medicine radiologists reported 3 as normal, 20 as having ‘degenerative’ or unspecified changes, and only 2 as having changes consistent with inflammatory disease. Changes reported as ‘inflammatory’ were deemed non-significant by the referring clinicians upon review of both patients. Of these 25 patients, 24 sets of notes were available for review. Only 4/24 had synovitis (mild or ‘possible’) prior to BS; none of whom had increased uptake in the affected joints. Where a pre-test diagnosis (inflammatory or non-inflammatory) was recorded in the notes (17 patients), this generally accorded with the BS report. In only three patients did it differ, one with palindromic symptoms, one who received i.m. steroid 10 days prior to BS, and one who had no recorded evidence to support a diagnosis of inflammatory arthritis. Only 4/24 scans highlighted joint regions not recorded in the notes as being affected; these comprised three shoulders, one acromioclavicular joint and two elbows. However, despite having BS thought to be negative for inflammatory disease, only 3/24 were discharged following this and further investigations (ultrasound or MRI) were requested in nine. Overall, we found little evidence that BS influenced diagnosis or management. We believe that the disproportionate number of requests from trainees reflects a misunderstanding of the sensitivity and specificity of BS for inflammatory arthritis. Non-specific uptake may lead to further investigation rather than patient reassurance.
Tc-99m-diphosphonates localize to bone and uptake is increased on blood pool and delayed images in both synovitis and sub-chondral bone damage (the hyperaemia of synovitis is shared by juxta-articular bone and vice versa, due to anastomotic vessels) [1]. Consequently, an increase in peri-articular uptake is not specific for synovitis and must be interpreted in the clinical context. Furthermore, Tc-99m-diphosphonates are taken up by all joints to a variable degree, making interpretation difficult [2]. The number of false positives is unknown but it is clear that BS are unable to reliably distinguish between ‘active’ and ‘inactive’ rheumatoid arthritis [2, 3]. This lack of specificity has been further highlighted by comparing different imaging modalities (ultrasound and MRI) [4]. Computerized image analysis yields only marginal improvement [5]. Despite these limitations, does a positive BS still have value in predicting joint damage in inflammatory arthritis? A careful study in early rheumatoid arthritis found that erosions were predated by persistent scintigraphic activity in affected joints [5]. Of note, however, almost all finger joints that eroded were active on clinical examination. The sensitivity of clinical examination was inferior to scintigraphy in the foot joints, but as analysis was by joint rather than by patient and as the scintigraphic score was derived from multiple BS, the implications for routine clinical practice are not clear.
Such difficulties led to the suggestion that a negative joint scan is of greater importance in the investigation of polyarthralgia [1]. This appears to be supported by one retrospective study; after 3.6 yrs mean follow-up, none of 22 patients with a negative BS had evidence of inflammatory joint disease [6]. Of interest, however, is that 19 of these patients had a pre-scan diagnosis of non-inflammatory joint disease. The remaining three had joint tenderness or pain on passive motion, but no synovial thickening or effusion. The eventual diagnosis in these patients was SLE, polymyalgia rheumatica and undifferentiated connective tissue disease. It is known that SLE patients with synovitis or arthralgia may have normal BS [7]. The excellent correlation between pre-study diagnosis and negative BS is supported by our study.
In conclusion, Tc-99m-diphosphonate BS suffer from lack of specificity and add little to clinical evaluation. Furthermore, they lack the detailed anatomical information provided by MRI and ultrasound. Musculoskeletal ultrasound in particular, has the advantage of being cheap and safe and is now becoming more widely available. In addition, there is no ionizing radiation with either MRI or ultrasound. In comparison, the typical radiation dose of a BS is 3.6 mSv (equivalent to 90 chest X-rays) and the typical cost has been set at around £148. We argue that their role in the evaluation of polyarthralgia has been largely superseded.
The authors have declared no conflicts of interest.
 |
References |
|---|
 Top References |
|---|
- Rosenthall L. Nuclear medicine techniques in arthritis. Rheum Dis Clin North Am (1991) 17:585–97.[Web of Science][Medline]
- Jamar F, Houssiau FA, Devogelaer JP, et al. Scintigraphy using a technetium 99m-labelled anti-E-selectin Fab fragment in rheumatoid arthritis. Rheumatology (2002) 41:53–61.
[Abstract/Free Full Text] - de Bois MH, Arndt JW, van der Velde EA, Pauwels EK, Breedveld FC. Joint scintigraphy for quantification of synovitis with 99mTc-labelled human immunoglobulin G compared to late phase scintigraphy with 99mTc-labelled diphosphonate. Br J Rheumatol (1994) 33:67–73.
[Abstract/Free Full Text] - Backhaus M, Kamradt T, Sandrock D, et al. Arthritis of the finger joints: a comprehensive approach comparing conventional radiography, scintigraphy, ultrasound, and contrast-enhanced magnetic resonance imaging. Arthritis Rheum (1999) 42:1232–45.[CrossRef][Web of Science][Medline]
- Mottonen TT, Hannonen P, Toivanen J, Rekonen A, Oka M. Value of joint scintigraphy in the prediction of erosiveness in early rheumatoid arthritis. Ann Rheum Dis (1988) 47:183–9.
[Abstract/Free Full Text] - Shearman J, Esdaile J, Hawkins D, Rosenthall L. Predictive value of radionuclide joint scintigrams. Arthritis Rheum (1982) 25:83–6.[Web of Science][Medline]
- Van de Wiele C, Van den Bosch F, Mielants H, Simons M, Veys EM, Dierckx RA. Bone scintigraphy of the hands in early stage lupus erythematosus and rheumatoid arthritis. J Rheumatol (1997) 24:1916–21.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||