CARPAL-TUNNEL RELEASE - A PROSPECTIVE, RANDOMIZED ASSESSMENT OF OPEN AND ENDOSCOPIC METHODS

To define the role of two-portal endoscopic carpal-tunnel release as a method for the treatment of compression of the median nerve at the wr ist, a prospective, randomized, multicenter study was performed on 169 hands in 145 patients. Either open or endoscopic carpal-tunnel releas e was performed in all of the patients who had clinical signs and symp toms consistent with carpal tunnel syndrome, had not responded to or h ad refused non-operative management, and had had electrodiagnostic stu dies consistent with carpal tunnel syndrome. Follow-up evaluations wer e performed at twenty-one, forty-two, and eighty-four days. At the end of the follow-up period, both the open and endoscopic methods had res ulted in high levels of achievement of the primary outcomes (relief of pain and paresthesias). The numbness and paresthesias were relieved i n eighty (98 per cent) of eighty-two hands in the open-release group c ompared with seventy-seven (99 per cent) of seventy-eight hands in the endoscopic-release group. This parameter was not recorded for three h ands in the open-release group or six hands in the endoscopic-release group. The satisfaction of the patients with the procedure, graded on a scale of 0 to 100 per cent, averaged 84 per cent in the open-release group compared with 89 per cent in the group that had had endoscopic release. We found no significant differences between the two groups wi th regard to the secondary quantitative-outcome measurements, includin g two-point discrimination, postoperative interstitial-pressure data f or the carpal canal, Semmes-Weinstein monofilament testing, and motor strength. The open technique resulted in more tenderness of the scar t han did the endoscopic method. Thirty-two (39 per cent) of eighty-two hands in the open-release group and fifty (64 per cent) of seventy-eig ht hands in the endoscopic-release group were not tender at eighty-fou r days. This parameter was not recorded for three hands in the open-re lease group and six hands in the endoscopic-release group. The open me thod also resulted in a longer interval until the patient could return to work (median, twenty-eight days, compared with fourteen days for t he open-release and endoscopic-release groups). Four complications occ urred- in the endoscopic carpal-tunnel release group: one partial tran section of the superficial palmar arch, one digital-nerve contusion, o ne ulnar-nerve neurapraxia, and one wound hematoma. An additional comp lication - transection of the common digital nerve of the ring and sma ll fingers and the medial 20 per cent of the median nerve - occurred d uring the pre-study trial period, in a hand that had had endoscopic ca rpal-tunnel release. Preliminary analysis suggests that functional out comes are achieved more quickly when the endoscopic method is used. Ho wever, the greater rate of complications indicates that intraoperative safety must be improved before endoscopic carpal-tunnel release is pe rformed on a widespread basis.