Wrist & Hand

Is bone scanning helpful in the diagnosis of CRPS?

Complex Regional Pain Syndrome (CRPS) remains an enigma. While all agree that the condition exists and causes ongoing refractory pain combined with disabling loss of function, the cause and pathophysiology are unclear. There is a broad agreement that, although imperfect, the Budapest Criteria, reached as a consensus statement, represent the most reliable way of reaching the diagnosis of CRPS type 1. However, clinicians from around the globe also use a range of other criteria, and innately we all feel more comfortable with an investigation to ‘confirm the diagnosis’, especially in this somewhat nebulous condition. There is controversy associated with the use of bone scintigraphy, and the possibility of the ‘imperfect reference test’ effect having an impact on the overall accuracy of the diagnostic test. With this in mind, this review team from Zurich (Switzerland) did their utmost to establish the diagnostic accuracy of bone scintigraphy in the diagnosis of CRPS.¹ They designed a systematic review of the literature and Bayesian meta-analysis to allow for estimation of the test accuracy of bone scintigraphy, and were able to account for the potential contamination of the results due to the imperfect nature of the Budapest Criteria, which in this case is the reference test. The results require a little untangling, however, they essentially establish that the sensitivity and specificity appear to be about 0.80 and 0.85, respectively, when the inaccuracy of the reference test is taken into account. Although this might make bone scanning seem appealing, when the authors turned the question on its head and repeated the analysis just using studies that included the reference test, they found a significantly lower sensitivity (0.55) but much higher specificity (0.94). This suggests that bone scintigraphy cannot be used to reach a diagnosis, however, it may be very useful in order to rule out a diagnosis of CRPS.

Do injections other than steroid work for tendinopathies?

Does therapy help contractures?

It is almost impossible to go through a hand clinic as a patient with a genuine problem without being treated by the hand therapist and, as often as not, this will be to tackle the difficult problem of stiffness in the hand. The therapist will usually effect a programme of stretches, occasionally supplemented by a splint. Despite the almost ubiquitous use of hand therapy and stretching, the evidence may not quite match up with current practice. A review team from Sydney (Australia) undertook a Cochrane review to establish if there was benefit in stretching as a treatment for contracture.⁶ In one of the largest reviews we have seen, the authors were able to identify 49 studies with relevant results reporting the outcomes of 2135 participants. Impressively, overall the majority of studies were at low risk of bias. The take home message is very clear. In those studies available for inclusion in the review, there was no evidence of benefit in patients with neurological conditions (mean difference 2°; 26 studies with 699 participants) or non-neurological conditions (0.2°; 19 studies with 925 participants). It turns out that in the literature as it stands (which is of reasonable quality), there is no clinically important effect of stretching exercises on joint mobility. So perhaps much time and effort and unrealistic hope could be saved by reconsidering this omnipresent but perhaps fruitless intervention.

Who does well following post-operative decompression?

Generally, it is said that the earlier decompression surgery for entrapment neuropathy is done, the more favourable a clinical outcome will be obtained, however, there may be more to it than just timing. Surgeons in Aichi (Japan) report their outcomes of 83 patients, all of whom underwent surgical decompression for cubital tunnel decompression.⁷ Although the authors rather enthusiastically state “this retrospective study was designed to investigate prognostic factors for postoperative outcomes for cubital tunnel syndrome using multiple logistic regression analysis with a large number of patients”, for this kind of analysis 83 patients really isn’t very many. The authors undertook a thorough review of the literature and established that potential prognostic factors for outcomes include demographic factors (sex, age, BMI, workers’ compensation status), surgical factors (type of surgery, disease duration) and some comorbidities (cervical lesion, presence of diabetes mellitus), along with pre-operative severity. Outcomes were assessed two years following surgery, and Messina’s criteria were used as the outcome measure. The bottom line is that the authors were able to demonstrate pre-operative severity as the only independent variable associated with poor prognosis. None of the other factors demonstrated any significant association with poor prognosis. While this paper has been correctly written and the statistics are appropriate, the significant limitation of such a small number of patients with so many covariates included in their analysis means that the results have very little meaning.

Mobility of the scaphoid in casts

Collagenase under the randomised trial spotlight

There has been much published and discussed in the medical literature, and gracing the pages of 360, concerning the value or otherwise of collagenase clostridium histolyticum (CCH) injection. Although much has been written and many healthcare dollars have been spent on this topic, one of the difficulties that we face in establishing the benefit or otherwise of CCH is the lack of a well conducted randomised controlled trial (RCT). Clinical trialists in Silkeborg (Denmark) have stepped up to the plate, and published their own direct head-to-head comparison with needle fasciotomy.⁹ Fifty patients were recruited into the study and results were available at final follow-up of two years. The needle fasciotomy group (n = 21) had percutaneous needle fasciotomy of their isolated proximal interphalangeal joint Dupuytren contractures, while the CCH group were treated with collagenase injections. The primary outcome measure was clinical improvement in objective contracture measurement. The authors also recorded secondary outcome measures of change in contracture, recurrence, adverse events, complications, and Disabilities of the Arm, Shoulder, and Hand (DASH) questionnaire score. We were surprised, given the size of the trial, that the authors identified some definite differences between the two groups. A greater than 50% improvement in contracture was maintained in 29% of needle fasciotomy patients, and in just 7% of collagenase patients, at two years of final follow-up. There was also a significant benefit in complication profile favouring the needle fasciotomy (24% complications versus 93%). The data presented here, although from a small study with some obvious issues with randomisation procedure, do suggest that in the longer term the needle fasciotomy is a better option.

Carpal coalitions

One of the most common normal variants in the hand are carpal coalitions. These are usually an incidental finding, although they can on occasion cause symptoms as they can interfere with the normal biomechanics of the wrist and hand. Given the frequency of these coalitions, there is relatively little known about them, and, in particular, how many are incidental findings and how many identified due to their symptoms. A study team in Boston, Massachusetts (USA) undertook the mammoth task of searching through 1119 posteroanterior wrist radiographs to identify carpal coalitions and what the indications for the radiographs were.¹⁰ They divided their patients according to indication for radiograph: wrist pain (623 wrists); non-traumatic wrist pain (175 wrists); and other reasons (321 wrists). Perhaps surprisingly, 8.8% of patients had a carpal coalition, and they were equally frequent in patients with wrist pain, be it traumatic or atraumatic. However, wrist pain was less common in patients with no trauma or pain. Given the findings here, although carpal coalitions were seen on wrist radiographs, it seems unlikely that many were symptomatic – especially given the equal incidence in traumatic and atraumatic painful wrists. The lower incidence in the third (non-painful) group requires a little more research to unpick. However, for the time being we would caution clinicians and patients alike who attribute painful wrists to a coalition, as there may be another occult underlying pathology which will be ignored.