A Clinical Prediction Tool for MRI in Emergency Department Patients with Spinal Infection

Area of Science:

• Orthopedics
• Infectious Disease
• Radiology

Background:

• Pyogenic spinal infection (PSI) is often missed at initial presentation, leading to increased patient morbidity and medical-legal risks.
• Diagnostic delays in PSI necessitate improved tools for early risk assessment and informed imaging decisions.

Purpose of the Study:

• To derive and validate a decision tool, the Spine Infection Risk Calculation Heuristic (SIRCH) score, for estimating the risk of PSI.
• To assess the performance of the SIRCH score in predicting PSI and its impact on magnetic resonance imaging (MRI) utilization.

Main Methods:

• A two-part prospective observational cohort study involving spine pain patients over a six-year derivation phase.
• Development of the SIRCH score using multivariable logistic regression, incorporating four key clinical variables: historical risk factors, fever, progressive neurological deficit, and C-reactive protein (CRP) levels.
• Validation of the SIRCH score's sensitivity in a separate disease-only cohort.

Main Results:

• The SIRCH score, with a threshold of ≥ 3, demonstrated high sensitivity (100%) and a positive predictive value of 40% in the derivation phase.
• The SIRCH score significantly reduced MRI utilization compared to using an elevated CRP level (P <0.001).
• In the disease-only cohort, the SIRCH score achieved 92% sensitivity for detecting PSI.

Conclusions:

• The SIRCH score offers a sensitive and effective method for estimating spinal infection risk.
• Implementing the SIRCH score can lead to more judicious MRI use compared to relying solely on elevated CRP or clinical suspicion.
• The SIRCH score aids in timely diagnosis and management of pyogenic spinal infections.