Clinical utility of the MN severity criteria in predicting recurrent Clostridioides difficile infection

Clostridioides difficile infection (CDI) frequently recurs and remains a major clinical challenge. The MN severity criteria, a Japanese scoring system for assessing acute CDI severity, have not previously been evaluated for predicting recurrence.

We conducted a single-center retrospective cohort study between January 2011 and July 2025. The MN severity score was calculated at diagnosis. CDI recurrence was defined as symptomatic infection with a positive stool toxin test within 8 weeks after initial clinical response. Multivariable logistic regression and receiver operating characteristic (ROC) analyses were performed to evaluate predictors of recurrence.

Among 558 eligible patients, 139 (24.9%) experienced CDI recurrence. Patients with recurrence had significantly higher MN severity scores at diagnosis (median 9 vs. 8, p < 0.001). An MN score ≥9 predicted recurrence with an area under the ROC curve of 0.70 (sensitivity 69.1%, specificity 63.3%). In multivariable analysis, an MN score ≥9 was independently associated with recurrence (adjusted odds ratio [aOR] 4.76, 95% confidence interval [CI] 2.97-7.61). Hematologic malignancy (2.28, 1.21-4.28) was also independently associated with recurrence, whereas use of potassium-competitive acid blockers was protective (0.33, 0.12-0.90). Higher MN severity scores were additionally associated with a shorter time to recurrence and a lower clinical cure rate.

The MN severity score, originally developed to assess acute CDI severity, also independently predicts 8-week recurrence. Incorporating the MN criteria into routine clinical practice may facilitate early identification of patients at high risk for recurrence and support individualized post-treatment strategies, including enhanced monitoring and recurrence-prevention interventions.