Trends in Musculoskeletal Imaging in Trauma Patients: How Has Our Practice Changed Over Time?

Objectives: The purpose of the study was to review diagnostic imaging in trauma patients. We hypothesized that diagnostic musculoskeletal imaging has increased over time, but at a lesser rate than radiography performed for other purposes. Two trauma centers were compared. Design: Retrospective multicenter study. Setting: Two level 1 trauma centers. Patients/Participants: Five hundred patients per year from each trauma centers were reviewed for 2002, 2005, and 2008. Main Outcome Measurements: Effective doses [millisieverts (mSv)] and total charges for radiography were calculated. Results: Most imaging was performed within 24 hours of injury. In 2002, 15% of all radiographic studies were computed tomography (CT) scans compared with 33% in 2008 (P \u3c 0.0001). Center 1 used more CT, and center 2 used more projection (plain) radiography. The percentage of musculoskeletal CTs increased from 26% in 2002 to 49% in 2008 (P \u3c 0.0001), without change in patient acuity. The mean effective dose per patient was 17.3 mSv in 2002, 30.0 mSv in 2005, and 34.1 mSv in 2008 (P \u3c 0.001). The percentage of total dose attributable to musculoskeletal studies increased from 25% in 2002, to 29% in 2005, and 31% in 2008 (P \u3c 0.001). Mean total charges per patient were $4529 in 2002; $6922 in 2005; and $7750 in 2008 (P, 0.001), with higher 2008 mean charges at center 1 versus 2 ($8694 vs. $6806, P = 0.001), primarily because of more CT scans. Conclusions: The number of diagnostic imaging tests, radiation dose, and related charges in trauma patients increased over time at both trauma centers, with CT scans accounting for most of the radiation dose and costs. A shift toward more advanced imaging from conventional projection radiography was noted at both trauma centers. Effective dose per patient more than doubled over the course of study at center 1. By 2008, half of all radiographic studies were for musculoskeletal purposes. Previous studies have suggested an increased risk of cancer with exposures of 20-40 mSv, making the mean total radiation doses in excess of 30.0 mSv since 2005 of great concern. Variability in ordering patterns between the 2 centers with similar patient acuity suggests opportunity for discussion about indications for utilization, which could result in lower radiation doses and fewer expenses