The long-term clinical outcome after anatomical reconstruction and tenodesis in the treatment of chronic anterolateral ankle instability was assessed in a retrospective multicentre study. The first group (AR) consisted of 25 patients (mean age at operation 22 yrs +/- 5.7) who underwent anatomical reconstruction and the second group (TE) of 29 patients (mean age 23 yrs +/- 6.6) who underwent tenodesis. For both groups, the mean follow-up period was 12.3 yrs (AR +/- 2.5 yrs, TE +/- 2.7 yrs). At physical examination, there were significantly more patients in the TE group (n=18) with a positive anterior drawer sign as compared with the AR group (n=7) (p=0.02). Medially located degenerative changes in the ankle joint as seen on standard radiographs were seen more often in the TE group (n=7) than in the AR group (n=1) (p=0.03). The mean talar tilt, 4.7 degrees in the AR group vs 6.9 degrees in the TE group, (p=0.02) and anterior talar translation, 2.9 mm in the AR group vs 4.3 mm in the TE group, (p=0.04) were significantly higher in the TE group at radiographic stress examination. According to the rating system developed by Good et al. (1975), significantly fewer patients in the TE group (n=8) had an excellent result as compared with the AR group (n=15) (p=0.03) and more patients in the TE group (n=9) had a fair or poor result (p=0.04) as compared with the AR group (n=2). We conclude that a tenodesis procedure does not restore the normal anatomy of the lateral ankle ligaments. When compared with anatomical reconstruction, a tenodesis leads to inferior results in terms of functional and mechanical stability, as well as overall satisfaction at long-term follow-up.