Can a Septic Hip Decision Rule Aid in the Evaluation of Suspected Pediatric Musculoskeletal Infections?

Document Type


Publication Date



DOI: 10.1016/j.jemermed.2018.12.022


Background: Musculoskeletal (MSK) infections can be difficult to diagnose in acute care settings. The utility of clinical decision tools for pediatric MSK infections in an emergency department has not been well studied.

Objective: Our aim was to evaluate the performance of a septic hip clinical decision rule (CDR) in the evaluation of pediatric musculoskeletal infections.

Methods: We performed a retrospective study of children evaluated for an MSK infection in our emergency department from 2014 to 2016. Data collection included demographics, discharge diagnoses, and clinical/laboratory predictors from the CDR. A χ2 analysis and Wilcoxon rank-sum tests compared patients with and without MSK infections. Logistic regression analysis examined the predictors for MSK infections. A receiver operating characteristic (ROC) curve was calculated to evaluate the performance of the predictors.

Results: Of 996 evaluations included in the final analysis, 109 (10.9%) had MSK infections. In a multivariable model, an adjusted odds ratio (OR) was significant for fever (OR 3.9, 95% confidence interval [CI] 2.4-6.4), refusal to bear weight/pseudoparalysis (OR 4.4, 95% CI 2.7-7.1), and C-reactive protein (CRP) > 2.0 mg/dL (OR 5.4, 95% CI 3.2-9.1). The probability of infection was 75.1% with five predictors present, 1.9% for zero predictors, and 5.1% if one predictor was present. An ROC curve showed an area under the curve of 0.82, indicating moderate accuracy.

Conclusions: A septic hip CDR demonstrates a low predicted probability of an MSK infection with zero or one clinical predictor present and moderate predictability with all five predictors. Fever, refusal to bear weight/pseudoparalysis, and CRP > 2.0 mg/dL performed best and should alert providers to consider other MSK infections in addition to septic arthritis.

Journal Title

The Journal of emergency medicine





First Page


Last Page


MeSH Keywords

Adolescent; Arthritis, Infectious; C-Reactive Protein; Child; Child, Preschool; Decision Support Systems, Clinical; Emergency Service, Hospital; Female; Fever; Hip; Humans; Infant; Logistic Models; Male; Odds Ratio; Pediatrics; Predictive Value of Tests; Prospective Studies; ROC Curve; Retrospective Studies; Weight-Bearing


emergency department; musculoskeletal infection; pediatric

Library Record