Although the majority of ankle injuries are radiographed, fewer than 15 percent of these patients are found to have fractures of the ankle or foot. In 1992, the Ottawa rules were developed to rule out fractures of the malleolus and mid-foot. The rules are based on the patient's ability to walk four steps immediately after the injury and assessment of localized tenderness at specific sites, such as the posterior edge or tip of either malleolus, the navicular, and the base of the fifth metatarsal (see Table 1). The ability of the Ottawa rules to rule out fractures of the ankle and mid-foot has been validated in several clinical settings, but the results of these studies have not been systematically evaluated. Bachmann and colleagues reviewed the available data to establish the accuracy of the Ottawa rules.

A search of electronic and other databases and input from experts identified 27 studies of the use of the Ottawa rules in ankle injuries. Two independent researchers assessed the studies for quality, and data was pooled for the meta-analysis. The analysis included data on 15,581 adults and children. The pooled sensitivity from all studies was 97.6 percent (95 percent confidence interval [CI], 96.4 to 98.9). Combined fractures and cases assessed more than 48 hours after injury had slightly lower sensitivities. The overall specificity was 31.5 percent (95 percent CI, 23.8 to 44.4), with the lowest specificities in children and combined fractures. The authors calculated negative likelihood ratios for negative results in the most common clinical situations (see Table 2), assuming a 15 percent prevalence of fracture. Based on these data, they estimate that in most subgroups, less than 2 percent of patients with a negative Ottawa rules assessment actually have a fracture of the ankle or mid-foot. The probability of radiograph-confirmed fracture in children following a negative Ottawa rules assessment is estimated at 1.22 percent.

The authors stress that studies varied in the types of patients assessed and that tests based on subjective assessment (palpation) by different physicians have significant variability. Nevertheless, the Ottawa rules have an extremely high sensitivity for foot and ankle fractures and could reduce radiograph use by 30 to 40 percent. The authors call for widespread use of the rules in primary care and further studies to assess their performance in the primary care setting.