Externally validating the Mortality In Spinal Infection-20 score: A retrospective cohort study in New Zealand.

Area of Science:

• Neurosurgery
• Infectious Diseases
• Public Health

Background:

• Spinal epidural abscess (SEA) is a serious condition requiring accurate prognostic tools.
• The "mortality in spinal infection" (MSI) score is one such tool proposed for predicting patient outcomes.
• External validation of prognostic scores is crucial for their clinical utility.

Purpose of the Study:

• To externally validate the "mortality in spinal infection" scoring system in a New Zealand population with spinal epidural abscess.
• To assess the predictive accuracy of the MSI score for one-year mortality in SEA patients.

Main Methods:

• Retrospective cohort study of 140 patients diagnosed with spinal epidural abscess between February 2009 and January 2022.
• Calculation of the MSI score for all included patients.
• External validation using receiver operating characteristic (ROC) curves and calibration plots.

Main Results:

• The MSI score showed limited predictive ability, with an area under the ROC curve of 0.67 (95% CI 0.57-0.79), failing to meet the >0.80 criterion.
• A score ≥11 was associated with higher mortality (33% vs 10%), but with low sensitivity (0.28) and positive predictive value (0.33).
• The score demonstrated high specificity (0.92) and negative predictive value (0.90).

Conclusions:

• The MSI score demonstrated limited ability to discriminate mortality risk in this New Zealand cohort of spinal epidural abscess patients.
• The current MSI score requires refinement or further development to improve its predictive accuracy.
• Prospective, multicentre data collection is recommended for developing more robust predictive tools for SEA outcomes.