BACKGROUND: Pediatric osteoarticular infection can cause severe morbidity. Some infectious loci may be difficult to identify clinically, and there may be more than one. There is little agreement regarding the appropriate use of preoperative magnetic resonance imaging (MRI) in this setting. After noting an unacceptably high rate of unplanned returns to the operating room for recurrent infection, clinicians at a tertiary care children's hospital noticed many patients had adjacent foci of infection on postoperative MRI. As a result, patients experienced prolonged treatment courses and multiple surgeries. An interdisciplinary team instituted practice guidelines whereby all patients with suspected osteoarticular infection underwent MRI for planned debridement during a reserved morning slot with a surgical suite on hold to proceed directly to surgery if indicated. Images were reviewed in real time to form the surgical plan. Young patients that required sedation for MRI were taken to surgery under the same anesthetic used for MRI without being awakened. The purpose of our retrospective study is to determine if implementing the practice guidelines for acute management of osteoarticular infection reduced unplanned returns to the operating room. METHODS: A total of 93 patients with osteoarticular infection were included in this study. A total of 40 cases, group A, were treated before implementing practice guidelines; 53 cases, group B, were treated after implementing practice guidelines. Our primary outcomes of interest were the identification of adjacent infections prior to surgery and need for repeat surgery, either planned or unplanned. RESULTS: Implementation of these guidelines reduced repeat surgery from 50% of patients to <27% (P=0.0099). Of patients requiring repeat surgery, 85% (n=17) were unplanned in group A versus 60% (n=9) in group B (P=0.0099). Adjacent infections were identified in 47.5% (n=19) of patients in group A, versus 60% (n=32) in group B. Adjacent infections were known before surgery in 32% (n=6) of patients in group A versus 72% (n=23) in group B. There were no statistically significant differences in initial patient characteristics or sites of infection. CONCLUSIONS: Implementing these guidelines reduced the need for repeat surgery in this population. It is difficult to predict with sufficient accuracy which patients need preoperative MRI. While resource intensive, preoperative MRI appears to offer substantial benefit in preoperative planning.