EARLY RESULTS OF CONVENTIONAL VERSUS TWO-PORTAL ENDOSCOPIC CARPAL TUNNEL RELEASE A prospective study C. DUMONTIER, C. SOKOLOW, C. LECLERCQ and P. CHAUVIN From the lnstitut de la Main, Institut Franfais de Chirurgie de la Main and Unitd de Biostatistique d'Informatique Mddicale, Hrpital Saint-Antoine, Paris, France The authors compare in a prospective, randomized study the early outcome of carpal tunnel release using either a conventional palmar open release (n=40) or a two-portal endoscopic release (n=56). Both groups were similar. No statistically significant differences were found regarding pain, disappearing of paraesthesiae or time to return to work. However, better recovery of grip strength was observed in the endoscopic group at 1 and 3 months. No surgical complications were observed in either group. Journal of Hand Surgery (British and European Volume, 1995) 20B." 5:658 662 Carpal tunnel release, whether endoscopic or conven- tional, has proved clinically and electromyographically effective in carpal tunnel syndrome (Chow, 1993; Genba et al, 1993; Friol et al, 1994). Many authors have reported that endoscopic carpal tunnel release reduces post-operative pain, increases pinch and power grip recovery, and allows for earlier return to work, but by 6 months no differences have been demonstrated from conventional surgery (Brown et al, 1993; Chow, 1993; McDonough and Gruenloh, 1993; Bande et al, 1994; Erdmann, 1994). However, in most published studies, no control groups were available or groups were not similar (Chow, 1993; McDonough and Gruenloh, 1993; Friol et al, 1994; Slattery, 1994). From May 1992 to May 1993 we performed a single centre randomized prospective study of the early outcome after carpal tunnel release. MATERIALS AND METHODS During 1 year, every patient suspected of having idio- pathic carpal tunnel syndrome was randomly assigned to conventional or endoscopic carpal tunnel release. Randomization was made during the first consultation using a disposable examination sheet on which the type of surgery to be done was noted. Only patients with more than 1 month of follow-up were included in the analysis. This represents 96 patients (40 open versus 56 endoscopic) out of 103 who were initially treated (43 open versus 60 endoscopic). Conventional surgery was performed using a 3 cm to 4 cm palmar approach along the axis of the fourth ray. In no case did the incision extend proximally to the distal wrist crease. Neither external nor internal neurolysis was performed and no tenosynovectomy was done. The endoscopic group was treated using the extra-bursal modification of the two- portal Chow technique (Chow, 1993). In both groups, bulky dressings were removed within 1 week. No splint- age or physiotherapy was used in either group. Distributions between the two groups were compared using the chi-square test, means using an analysis of variance. RESULTS The distribution of the two groups is shown in Table 1: there is no statistical difference when comparing their sex distribution, age, profession, presence of morning finger stiffness and grip strength data before surgery. 79 patients were examined at 2 weeks (35 open/44 endos- copic), 62 at 1 month (24/38), 58 at 3 months (30/28) and 20 were examined at 6 months (12/8). All patients were asked to come back at 6 months, but in private practice, it is not always possible to recall the patients Table 1--Description of the two groups of patients Conventional g r o u p Endoscopic group P value Sex: (male/female) Age: (mean ± sd) Profession Manual workers Clerical workers, unoccupied or retired Morning finger stiffness (yes/no) Jamar dynamometer Right (in Kg, mean_+sd) Left 4/36 7/49 0.79 50.7 _ 13.98 53.4 ± 14.49 0.08 21 24~ 16 31 0.22 15/18 28/22 0.35 28.48 _+11.77 28.49 + 12.90 0.99 30.65 + 12.98 29.14 ± 13.90 0.29 sd = standard deviation. 658