BackgroundAnkle arthrodesis is a treatment of choice for advanced tibio-talar disease unresponsive to conservative treatment. Using arthroscopy to perform this procedure minimises soft-tissue trauma while providing similar outcomes to those of open surgery. Union rates have ranged across studies from 85% to 100%. The objective of this study was to assess the potential influence on union of the number of screws used for arthrodesis fixation. HypothesisThe working hypothesis was that using three screws for arthrodesis produced a higher union rate than did using only two screws. Material and methodsThis single-centre retrospective comparative study included 111 cases of arthroscopic ankle arthrodesis (in 108 patients) carried out between February 1994 and October 2012. The number of screws was two in 75 cases and three in 36 cases. Union was assessed on radiographs taken 2, 6, and 12months postoperatively. ResultsMean age at surgery was 55.8years. After 12months, union was achieved in 87.4% cases overall. The non-union rate was 16% with two screws and 5.6% with three screws. Three-screw fixation was associated with a significantly higher rate of union of the medial gutter after 6months and of the lateral gutter after 12months. DiscussionOur findings support the use of three screws for fixation of arthroscopic tibio-talar arthrodesis. Adding a third screw seems associated with a lower risk of non-union and a shorter time to union. These effects can be ascribed to greater stability of the construct. Level of evidenceLevel IV, retrospective study.