The Ottawa Ankle Rules (OAR) are criteria for predicting ankle fractures in adults allowing for insignificant fractures, defined as small avulsion fractures. Because the clinical significance of avulsion fractures and Salter-Harris type I fractures in children is unclear, we sought to prospectively evaluate the use of the OAR in children and to determine whether different criteria should be used for predicting ankle fractures in children.
In this prospective study, patients younger than 18 years presenting to a pediatric emergency department (ED) with an acute nonpenetrating ankle injury were eligible for study participation. Information on 22 clinical variables was recorded on a standardized data sheet. The OAR were included but not specifically identified on the data sheets. A standard ankle radiographic series was obtained on all subjects. All fractures were considered to be significant. Follow-up phone calls were performed to assess final diagnosis and outcome. Sensitivity and specificity of OAR and other potential criteria for predicting ankle fractures in children were calculated.
A total of 195 patients with ankle injuries were evaluated. The mean age of patients was 12.6 years. Forty fractures (21%) were identified. The sensitivity of OAR was 83% (95% CI, 65–94%), specificity was 50% (95% CI, 41–59%), positive predictive value was 28%, and negative predictive value was 93%. Three independent factors were significantly associated with ankle fractures: inability to walk immediately after the event, inability to bear weight for four steps in the ED, and tender deltoid ligament. If one or more of these factors were present, sensitivity for predicting ankle fractures was 93% (95% CI, 78–99%), specificity was 27% (95% CI, 20–36%), positive predictive value was 23%, and negative predictive value was 95%.
The OAR cannot be applied to children with the same sensitivity as adults.
From the University of Colorado Health Sciences Center, The Children’s Hospital, Denver, Colorado (K.D. Clark); and Fowler-Kennedy Sports Medicine Center, London, Ontario, Canada (S. Tanner).
Presented as a poster at the Rocky Mountain Conference in Emergency Medicine, January 13, 1999, Snowmass, Colorado.
Address for reprints: Kathryn D. Clark, Box B251, The Children’s Hospital, Denver, CO 80218; e-mail: Clark.Kathryn@tchden.org
The authors thank Mary Samson, MS (University of Colorado Health Sciences Center, Denver, Colorado), and Todd MacKenzie, PhD (Research Institute, The Children’s Hospital, Denver, Colorado), for assistance with statistical analysis.