ABSTRACT
BACKGROUND: Many emergency physicians order chest x-rays (CXRs) for pediatric patients who present with a chief complaint of chest pain despite a paucity of research to support this testing, which exposes patients to radiation, cost, and delays. OBJECTIVES: This study aimed to begin development of a decision making tool that will allow emergency physicians to selectively obtain CXR films in pediatric patients presenting with chest pain. METHODS: We performed a retrospective cohort study of 400 consecutive pediatric patients with a chief complaint of chest pain and reviewed charts to determine how many received a CXR and which clinical characteristics were present in all patients. Chest radiograph findings were graded for significance as follows: (1) no or minor clinical significance: normal result in the CXR film without effect on the immediate evaluation of a patient; (2) moderate clinical significance: only impact on plan for follow-up; and (3) major clinical significance: result in the CXR film directly affects immediate management. We then evaluated each chart for historical or examination findings that might identify criteria associated with positive radiographic findings to propose a set of criteria that could lead to the development of a decision rule that allows a reduced utilization while having a high sensitivity for clinically significant positive findings on CXR film. RESULTS: Of the 400 pediatric patients reviewed, 63.5% (n = 254) received a CXR in the emergency department (ED). Of those receiving a CXR, only 8.26% (n = 21) had a finding that affected either ED management or follow-up planning. The criteria that would have identified all patients with positive results in the CXR films were abnormal vital signs, shortness of breath, palpitations, presence of comorbidities, abnormal or unilateral breath sounds, history of trauma, murmur, or cough. CONCLUSIONS: This pilot study demonstrates the potential for a decision rule to eliminate both cost and radiation exposure by using defined criteria to determine the need for a CXR in pediatric ED patients. We identified 8 simple criteria that would have identified all children who benefited from a CXR in this study. The next phase of this study will prospectively evaluate the utility of each of the criteria as part of a draft decision rule.