Safely Reducing Unnecessary Radiographs in Suspected Pediatric Musculoskeletal Injuries

Abstract

Background While radiographs are a critical component of diagnosing musculoskeletal (MSK) injuries, they are associated with radiation exposure, patient discomfort, and financial costs. Our study initiative was to develop a system to diagnose pediatric MSK injuries efficiently while minimizing unnecessary radiographs. Methods This was a quality improvement trial performed prospectively at a single level one trauma center. A multidisciplinary team with leaders from pediatric orthopedics, trauma surgery, emergency medicine, and radiology created an algorithm delineating which x-rays should be obtained for pediatric patients presenting with MSK injuries. The intervention was performed in the following 3 stages: stage 1: retrospective validation of the algorithm, stage 2: implementation of the algorithm, and stage 3: sustainability evaluation. Outcomes measured included number of extra radiographs per pediatric patient and any missed injuries. Results In stage 1, 295 patients presented to the pediatric emergency department with MSK injuries. A total of 2148 radiographs were obtained, with 801 not indicated per the protocol, for an average of 2.75 unnecessary radiographs per patient. No injuries would have been missed using the protocol. In stage 2, 472 patients had 2393 radiographs with 339 not indicated per protocol, averaging 0.72 unnecessary radiographs per patient, a significant reduction from stage 1 (P < 0.001). There were no missed injuries identified on follow-up. In stage 3, improvement was sustained for the subsequent 8 months with an average of 0.34 unnecessary radiographs per patient (P < 0.05). Conclusions Sustained reduction of unnecessary radiation to pediatric patients with suspected MSK injuries was accomplished through the development and implementation of a safe and effective imaging algorithm. The multidisciplinary approach, widespread education of pediatric providers, and standardized order sets improved buy-in and is generalizable to other institutions. Level of Evidence: III