Hip radiography for the investigation of nontraumatic, short duration hip pain presenting to a children's emergency department.

OBJECTIVE

To assess the diagnostic accuracy of routine hip x-ray in 2- to 11-year-old children presenting to the emergency department with recent onset, nontraumatic hip pain and to assess the predictive value of other clinical presenting variables.

METHODS

Retrospective chart review of children with hip pain of less than 2 weeks duration, presenting to a metropolitan children's emergency department in New Zealand. Data were extracted on radiography results, diagnosis on final discharge from hospital, and other clinical details. Diagnostic categories were divided into urgent pathology (those requiring immediate treatment) and nonurgent pathology (those requiring follow up and those requiring no treatment or follow up). Contingency analysis and logistic regression were performed on variables to assess ability to predict urgent pathology.

RESULTS

Hip x-rays were performed in 99% of 310 eligible patients. Of these, 3% had some degree of significant abnormality reported on hip x-ray. Final diagnoses were transient synovitis in 86% of patients, osteomyelitis or other significant musculoskeletal sepsis in 8%, slipped upper femoral epiphysis in 1%, and other musculoskeletal disorder in 4%, with no cases of septic arthritis. In children 9 years or older, the likelihood ratio with a positive x-ray was 17, compared with 6 in children younger than 9 years. Fever, weight-bearing status, and age were the most important predictors of urgent pathology.

CONCLUSIONS

In children younger than 9 years presenting with recent onset, nontraumatic hip pain, hip x-ray is of limited value. In children 9 years or older, hip x-ray may be useful to exclude slipped upper femoral epiphysis. Fever, weight bearing status, and age may better predict urgent pathology.