A Rapid MRI Protocol for the Evaluation of Acute Pediatric Musculoskeletal Infections

Abstract

Background: Acute musculoskeletal infection affects >1 in 6,000 children in the United States annually. Magnetic resonance imaging (MRI) is the gold standard for the diagnosis of musculoskeletal infection, but it traditionally requires contrast and anesthesia for children, delaying management. A rapid MRI protocol involves MRI without anesthesia and with limited non-contrast sequences optimized for fluid detection and diffusion-weighted images to identify abscesses. We hypothesized that a rapid MRI protocol would improve imaging and treatment efficiency for pediatric patients undergoing musculoskeletal infection evaluation without substantially affecting accuracy. Methods: This was a single-center, retrospective study of patients undergoing evaluation for musculoskeletal infection before (60 patients in the traditional cohort [TC]) and after (68 patients in the rapid cohort [RC]) implementation of the rapid MRI protocol. Sociodemographic and clinical variables were extracted from electronic health records, and statistical comparisons were performed. Results: The anesthesia rates were 53% for the TC and 4% for the RC, and the contrast administration rates were 88% for the TC and 0% for the RC. The median time to MRI after ordering was 6.5 hours (95% confidence interval [CI], 5.0 to 8.6 hours) for the TC and 2.2 hours (95% CI, 1.4 to 3.6 hours) for the RC (p < 0.01). The median duration of MRI was 63.2 minutes (95% CI, 56.8 to 69.6 minutes) for the TC and 24.0 minutes (95% CI, 21.1 to 29.5 minutes) for the RC (p < 0.01). The median hospital length of stay was 5.3 days (95% CI, 3.7 to 6.9 days) for the TC and 3.7 days (95% CI, 1.9 to 4.1 days) for the RC (p < 0.01). The median hospital charges were $47,309 (95% CI, $39,137 to $58,769) for the TC and $32,824 (95% CI, $22,865 to $45,339) for the RC (p < 0.01). Only 2 positive cases of musculoskeletal infection in the RC were missed on the initial imaging, but these instances were not attributable to the rapid protocol itself. Although 10 of 68 rapid MRI scans resulted in nondiagnostic outcomes due to patient motion, only 6 of 68 required repeat MRI with anesthesia. Conclusions: In patients evaluated for musculoskeletal infection, the rapid MRI protocol eliminated contrast and minimized anesthesia while improving MRI access and decreased scan and interpretation times, hospital length of stay, and hospital charges. The rapid MRI protocol had high sensitivity for diagnosing musculoskeletal infection and a low rate of imaging failure. Level of Evidence: Diagnostic Level III. See Instructions for Authors for a complete description of levels of evidence.