Evaluation of Carpal Tunnel Release Using the Knifelight ® Instrument

INTRODUCTION

Carpal tunnel syndrome is one of the commonest upper limb conditions, with a prevalence estimated up to 3.7% (Papanicolaou et al., 2001). In Maine, USA, the carpal tunnel release rate is reported as varying from 0.82 to 2.87 per 1000 per year (Keller et al., 1998). In the United States 400,000 such operations are done each year, costing a total of 2 billion dollars (Palmer and Hanrahan, 1995). Most patients have a good result but there is a small incidence of unsatisfactory outcomes, usually relating to tenderness of the scar or pillar pain (Nancollas et al., 1995). In the last decade, less invasive techniques for carpal tunnel decompression have been developed in order to reduce the incidence of pillar pain and tender scars. Among such techniques is the Knifelight^® (Avci and Sayli, 2000). This is a new knife with its own battery-powered light source, which can be used to divide the flexor retinaculum through a small incision in the palm of the hand (Fig 1). A preliminary report was encouraging (Avci and Sayli, 2000) and we have therefore performed a prospective randomized controlled trial to evaluate the safety and efficacy of the Knifelight instrument in comparison with open carpal tunnel release.

PATIENTS AND METHODS

We obtained approval from our Hospital Ethical Committee for conduct of this study. All patients admitted for unilateral carpal tunnel decompression were considered for the study. After obtaining informed consent, patients were randomized just before surgery into two groups by the use of pre-filled blank envelopes. The patients in Group I underwent an open carpal tunnel decompression. This was performed under local anaesthetic infiltration with a pneumatic tourniquet. Through a longitudinal incision over the heel of the hand, which extended over the wrist crease, the flexor retinaculum was divided. No dissection was made in the carpal tunnel. The wound was closed with interrupted sutures and a compression bandage was applied. Patients in Group II had carpal tunnel decompression performed using the Knifelight^®. A small incision was made on a line between the middle of the wrist and the third web space, at the level of a line drawn along the palmar aspect of the fully abducted thumb (Fig 2). The distal edge of the flexor retinaculum was identified and blunt dissection was performed above and below this proximally as far as the wrist crease using a fine needle-holder. The two blunt skids of the Knifelight were placed above and below the ligament and the instrument was firmly but smoothly pushed proximally so that the blade between the two skids divided the flexor retinaculum. When the skids could be palpated at the level of the wrist crease, the operating theatre lights were turned out. As the instrument was withdrawn, its two lights (one on each skid) could be identified below the skin, thus confirming that the flexor retinaculum had been divided. The wrist was flexed after withdrawal of the instrument so that the cut edges of the ligament and the median nerve could be visualized, confirming complete division of the ligament. The wound was closed with two sutures and a pressure bandage was applied.

The degree of discomfort experienced by the patient during the operation was recorded as no pain, slight discomfort, moderate discomfort or pain. Patients were examined at 2 weeks when the sutures were removed and again at 6 weeks when they were asked about residual numbness or paraesthesia and the degree of discomfort at the operation site. The assessor could not be blinded as the incisions were different in the two groups. The wound and the heel of the hand were also assessed for local swelling, tenderness or scar hypertrophy and grip strength was measured with a Baseline^® hydraulic hand dynamometer (Fabrication Enterprises Inc., Irvington, NY, USA). The time of return to work (for those patients in work) was recorded as were any complications. Statistical analysis was performed using Chi-squared and t-tests.

RESULTS

All patients approached gave their consent to participation in the study. Details of the patients are given in Table 1. The operations took between 6 and 8 min, regardless of which procedure was used. All the operations were performed by the senior author. There was no significant difference in the levels of peroperative discomfort recorded by the patients. None, mild, moderate and severe discomfort was reported by 15, 21, five and two patients, respectively, in Group I (open carpal tunnel release) and 18, 16, 14 and one patient, respectively, in Group II (Knifelight procedure).

Complications in Group I included four patients with significant pillar pain, of whom three were referred for physiotherapy. One patient had a partial wound dehiscence, one unexplained thumb pain, one mild stiffness of the fingers and one transient numbness in the index finger. Complications in Group II included one patient with transient numbness of the index finger, one who developed significant reflex sympathetic dystrophy and one who had a superficial wound infection. In two cases in Group II there were technical difficulties, and the Knifelight could not be advanced through the flexor retinaculum due to excessive thickness of the retinaculum. These two patients were converted to an open procedure and thus withdrawn from the study.

Grip strength was better in Group II at 6 weeks, although not quite to a statistically significant degree. There were statistically significant differences in the scar tenderness and also in the time to return to work between the two groups (Table 2). There was no difference in the proportion of patients cured of their pre-operative carpal tunnel syndrome symptoms between the two groups.

DISCUSSION

Open carpal tunnel release is known to have a small but definite incidence of unsatisfactory results, usually because of scar tenderness and pillar pain. The mechanism of pillar pain is not clear but may result from small neuromas in the subcutaneous tissue. The heel of the hand is richly innervated by the palmar cutaneous branch of the median nerve, the palmar cutaneous branches of the ulnar nerve, the nerve of Henle and multiple transverse ulnar cutaneous branches (Szabo, 1999). Valls-Solé et al. (1995) have also postulated that the normal sliding movement of the median nerve as the wrist flexes and extends may be impaired if the nerve is involved with scar tissue.

Less invasive procedures have been recommended in an attempt to decrease the incidence of such complications, including carpal tunnel decompression through small incisions and, especially in the United States, endoscopic carpal tunnel decompression. However, it is still not clear whether these techniques have produced any improvement. Gerritsen et al. (2001) have performed a review of randomized controlled trials comparing these techniques with the gold standard of open carpal tunnel release. They concluded that the newer, less invasive, procedures are equally effective but that there is conflicting evidence about whether they produce an earlier return to work or any decrease in pillar pain. They found that endoscopic carpal tunnel release produced more transient nerve complications. Boeckstyns and Sørensen (1999) reviewed 54 publications and found that permanent nerve damage occurred in 28 out of 9516 endoscopic procedures, including nine injuries to the main trunk of the median nerve, compared with two out of 1203 open procedures. As a result of these data, and the fact that endoscopic carpal tunnel decompression is perceived as being technically difficult with a significant learning curve and with cost implications, it has failed to become widely accepted.

The advantage of the Knifelight technique is that the operation is technically easy and is done at least partially under direct vision. Our preliminary results have shown a decrease in the incidence of pillar pain and scar tenderness. As a result of this, there has been less need for supplementary treatment such as physiotherapy and anti-inflammatory medication. Also we found that the patients treated with Knifelight had a faster return to work. Although these measured improvements were not great in magnitude, it remains the case that small improvements in a very common operation can result in worthwhile benefits to the population as a whole. The main disadvantage of the Knifelight is the greater cost of the instrument, as it is a single use device. However, it is possible that the unit cost may decrease if the technique becomes more widespread.

In view of the encouraging preliminary results, we would suggest that larger studies with long-term follow-up are necessary to determine whether the clinical improvement is maintained over a period of time, and whether rare complications such as nerve injury have an acceptably low incidence.