Objective: To test the validity of a set of defined and tested rules for decisions on use of radiography for acute ankle injuries.
Design: Prospective survey.
Setting: Community hospital emergency department managing 42000 visits annually.
Participants: A non-consecutive sample of 318 adults and children presenting during 1 year was evaluated by 25 family physicians in part-time emergency practice.
Interventions: Participating physicians interpreted the Ottawa ankle rules for all enrolled patients and ordered radiographs in 96% of cases.
Main outcome measures: Sensitivity of the Ottawa ankle rules to predict whether radiography is required for acute ankle injuries.
Results: Of the 318 cases, 22 incomplete records were excluded, leaving 259 records of adults and 37 records of patients younger than 16 accepted for analysis. Of 34 adult patients with identified fractures, only one was predicted by the rules not to require radiographs. Sensitivity of the rules was 0.971 (confidence interval [CI] 0.914 to 1.00), specificity was 0.302 (CI 0.242 to 0.362), positive predictive value was 0.174 (CI 0.120 to 0.228), and negative predictive value was 0.986 (CI 0.971 to 1.00). Radiography could have been reduced by 26.3% had the rules been applied. Of the 37 children, seven had fractures. All were properly identified by the rules. Radiography in this group could have been reduced by 22%.
Conclusions: This study validates the Ottawa ankle rules and supports their use. Further research on how the rules apply to children is required.