A long-term follow-up (3.5 years mean) of 106 operated ankle fractures out of 167 patients revealed 73.7% excellent and good, 16% acceptable and 10.3% poor results. Anatomically exact reduction was shown to be a prerequisite for good results and was seen in 77.2% of all cases. By neglecting biomechanical principles an insufficient reconstruction led to a poor final outcome. But even in 11% of the excellent operatively treated fractures late arthrosis developed. Statistical analysis showed that the onset of posttraumatic arthrosis was correlated with and predicted best by the number of single lesions. Dorsiflexion testing was the clinical parameter correlating best with the extent of the injury, the degree of arthrosis and the subjective complaints. This underlines the importance of assessment of dorsiflexion as a clinical screening test for remaining sequelae after ankle fractures.