Wrist & Hand

Pins or plates for the distal radius: a meta-analysis

Cast or plate for wrist fractures in the over 60s

Carpal tunnel release in patients receiving workers’ compensation: beware poor outcomes

Patients who may have been injured at work present special challenges to surgeons, and we should all be wary of predicting outcome when there is litigation in the air. There is always a suspicion that a background of workers’ compensation (WC) may confound the usual expectations of recovery, either consciously or unconsciously, on behalf of both the patient and the surgeon, and there is certainly evidence that our suspicions might be well-founded. A group from El Paso, Texas (USA) reviewed 25 studies reporting the outcomes of 1586 wrists in the WC cohort and 2781 in the non-WC cohort, and found that patients undergoing a carpal tunnel release in the context of workers’ compensation took almost five weeks longer to return to work, were 16% less likely to return to pre-surgery occupation, and experienced nearly three times the reported average number of operative complications.⁴ Disturbingly, the workers’ compensation group were also less likely to have appropriate physical examination findings and electrodiagnostic tests confirming the diagnosis preoperatively. The authors make the valid observation that patients may have a potential financial interest to prove their WC claim, and therefore may end up requesting surgery for a condition that they do not strictly have according to physical examination and objective studies. On the surgical side, surgeons may feel pressured by the patient to perform what may be unnecessary surgery and, in addition, there is a potential for physicians to perform procedures outside the standard of care if financially incentivized to do so. This may seem an odd concept to some, but around the globe it can be true that worker’s compensation surgery is more lucrative than standard cases. In the United Kingdom, where litigation is often required to obtain compensation, we should bear in mind the finding that ongoing litigation was an independent predictor of delayed return to work. While not always possible, patients should be encouraged to settle their claim prior to intervention. The reason for the poorer outcomes in WC patients can be difficult to unpick; it may be that failings in the preoperative assessment means that patients in the WC group without carpal tunnel syndrome are getting surgery and biasing the results, or it may be that WC patients have a social or financial motive to report poorer outcomes, either consciously or subconsciously. Here at 360, we think that these factors should all be borne in mind while treating each case on its individual merits, but it is certainly the case that we are less likely to do harm to our patients if we have a confident diagnosis before choosing to operate.

Coincidence or complication? Trigger digit after carpal tunnel release

From time to time, someone has a successful carpal tunnel release but then develops a trigger finger. A group from Taichung City (Taiwan) have performed a systematic review that included nine studies and reported the outcomes of a total of 5654 carpal tunnel release operations, in order to address the question: Is this a coincidence?⁵ Remarkably, trigger finger occurred within six months of surgery in 8.5% of cases on average, and the incidence was as high as 32% in one study. Most patients who developed a trigger digit did so within the first six months following surgery; the thumb and ring finger were the most commonly involved digits. Various reasons for this finding are suggested by the authors, including the increased risk of both conditions in those with osteoarthritis, inflammatory arthropathies, and diabetes. The effect of these conditions is impossible to evaluate from this study, however, as the analysis comprised both papers that included and papers that excluded some of these conditions. The effect of type of surgical approach is also unclear; a number of the included studies involved high proportions of patients who underwent an endoscopic rather than open carpal tunnel release but, as no subgroup analysis was performed, it is not possible to identify the relative incidence, or to identify whether this is a risk factor. Aside from local postoperative inflammation contributing to stenosis, it has also been suggested that the release of the ligament itself may be causative, and that the slight bow-stringing of tendons post-release may increase pressure on the flexor pullies. Regardless of aetiology, it would seem that the patient should be counselled for the risk of developing a trigger digit as part of the consent process for carpal tunnel release. Here at 360, we might even go so far as to suggest that the patient should be counselled for release of a mild trigger finger if one exists at the same time as carpal tunnel release.

Cost description of clinical examination and MRI in wrist ligament injuries

We must always remember in these times of austerity that no healthcare budget is unlimited. As such, every time we spend money on any investigation, there is less money available to spend on treatment. In many areas, MRI scanning is now so readily available that there is a temptation for money to be spent on this investigation, whereas, on reflection, perhaps the diagnosis could be made without this expense. We were particularly interested to see a study from Gothenburg (Sweden) where the cost of MRI scanning was calculated and measured against the result of examination by an experienced hand surgeon.⁶ The authors describe the results of investigations undertaken at six radiology departments that, combined, undertook over 400 MRI investigations for suspected wrist ligament injuries. They found that the additional cost of the difference between MRI and a clinical examination by hand surgeon, plus indirect cost for patients, was over €950 000 per annum for a catchment population of 1.7 million. They also advocate early expert examination to facilitate timely acute repair of injuries, such as triangular fibrocartilage complex (TFCC) tears, the outcome of which is thought to be best when performed early. Indirect costs are notoriously difficult to estimate accurately, and extrapolation to other healthcare systems is also challenging, as the intrinsic costs of MRI scans and of surgeon time may be quite variable. If accurate, however, this saving would equate to approximately £36 million per annum in a country the size of the United Kingdom.

Safety of corticosteroid injection for carpal tunnel syndrome

Good medical practice mandates that we should always consider nonoperative treatment before surgery, which fits with the original Hypocratic oath of ‘first do no harm’. There is generally pretty good evidence, although mostly from case series, that steroid injections have some benefit, if not always durable, for patients with carpal tunnel syndrome. This treatment, of course, also carries both major and minor hazards of its own, such as intraneural injection or tendon rupture. In this study, a team from Canterbury (United Kingdom) asked patients about side effects and complications at six weeks following 689 carpal tunnel injections.⁷ The injections were given in the context of a carpal tunnel treatment pathway, and all those performing the injection were specially trained to do so. The technique utilized was based on an adaption of Racasan and Dubert’s technique, with the injection performed 1 cm or more proximal to the distal wrist crease, and radially rather than through the flexor carpi radialis tendon, with care taken to avoid the radial artery. Triamcinolone (40 mg) was injected without adding local anaesthetic. As usual, patients were asked to report paraesthesias, which could indicate the needle contacting the median nerve. In their series, they identified no cases of tendon rupture or intraneural injection. The group also retrospectively analyzed over 9500 injections performed in their centre over a 12-year period, which included the cases above. Overall, they identified one case of likely intraneural injection, one tendon rupture, and two cases of digital gangrene leading to partial loss of digit. They therefore advise the risk of severe side effects such as tendon rupture or intraneural injection being < 0.1%, minor persistent local effects such as subcutaneous atrophy or depigmentation being 2%, and transient effects such as pain, bruising, or facial flushing being between 15% and 20%. There has been much debate about the utility of ultrasound-guided carpal tunnel injections in avoiding intraneural injection but from this study it seems that we can be reassured that, at least if given by an experienced injector, the intervention is safe.

Radioscapholunate fusion long-term follow-up

Severe degenerative arthritis around the wrist can be difficult to treat, with simple and effective solutions in short supply for a condition that carries with it some significant comorbidity. Since wrist replacement remains experimental (with a material risk of complications in the short term and little outcome data on longer-term durability, not to mention issues of bearing wear), we should consider whether a patient’s own cartilage on subchondral bone is a better option than a metal or plastic bearing. Radioscapholunate fusion is an option for patients when there is established radiocarpal arthritis with preservation of the midcarpal joints. Sparing the midcarpal joints in the treatment can also preserve the ‘dart-thrower’s movement’. However, although in common usage, there is little hard clinical data to support its use. It is encouraging to see a series from patients in Adelaide (Australia) that shows good long-term outcomes in a cohort of these patients.⁸ The authors followed 17 patients with a minimum of ten years’ follow-up, 15 of whom were satisfied with their outcomes, with mean outcome scores as follows: visual analogue scale for pain score, 2.1/10; Quick Disabilities of the Arm, Shoulder and Hand (DASH) score, 29; and Modified Mayo Wrist score, 60. There were two conversions to wrist fusion in the course of the follow-up of the study, both within the first two years. A cohort of these patients underwent distal pole of scaphoid and excision of the triquetrum in addition. The authors state that patients with excision of the triquetrum had a mean radial-ulnar arc of motion increase of 10°, but this in itself was not statistically significant. While we cannot conclude that a benefit exists from this evidence alone, the probability of a type II error is high with such a small group size. The mean radiographic joint space for the scaphocapitate joint was 1.7 mm and lunocapitate joint was 1.3 mm at latest follow-up, indicating long-term radiographic preservation of these joints. In this study, patients who obtained a good result at two years were likely to achieve a good long-term outcome and, therefore, this seems a good technique in suitable patients compared with other, limited options.

Iliac bone grafting for scaphoid nonunion with avascular necrosis

In the previous issue of 360, we reported a paper that challenged the commonly accepted dogma that the proximal pole of the bone is ischaemic in scaphoid nonunion, that an MRI should be performed to demonstrate this, and that subsequent surgical reconstruction in the presence of avascularity requires insertion of a vascularized graft.⁹ There are certainly firm believers in the hand surgery community that this approach is necessary, but it is also true that the adoption of these assumptions leads to expensive MRI scans and complex surgery, for which only a few aficionados might feel comfortable. In this issue, we report another paper challenging these presumptions, this time from Seoul (South Korea).¹⁰ This group reports the outcomes of a series of 24 patients, all with proximal pole avascular necrosis treated operatively with a non-vascularized iliac crest autograft. The avascularity in all cases was confirmed preoperatively using MRI and intraoperatively by the clinical absence of punctuate bleeding. The operative technique of the Fisk–Fernandez wedge-shaped corticocancellous graft was used in 11 patients; cancellous bone grafting was used in 13 patients. The vast majority (n = 22/24) of the surgeries went on to heal. The two who did not have a successful outcome had both undergone previous surgeries; they subsequently underwent a further procedure using the Fisk–Fernandez technique and both then went on to union. Although relatively small series, these two papers in consecutive 360 issues include a total of 55 patients. It seems that we can therefore conclude from these papers that the issue of avascularity may be overblown, and that routine MRI scanning and vascularized bone grafting should be avoided.