Clinical Usefulness of the "MN Criteria" - the Clostridioides difficile Infection Severity Scoring System - in the Japanese Setting

Abstract

Objective The severity of Clostridioides difficile infection (CDI) is an important prognostic factor. The "MN criteria," proposed in Japan in 2017, attempted to remedy the shortfalls in the reported guidelines proposed globally to determine CDI severity. We therefore assessed the accuracy of the MN criteria and validated the important factors associated with predicting CDI severity. Methods Sixty-six CDI cases were investigated retrospectively at a Japanese University Hospital from January 2015 to December 2018. The fulminant cases were screened out, and the non-fulminant cases were classified according to their severity stages using the nine variables included in the MN criteria. Clinical events, such as death within 28 days, colectomy, and admission to the intensive care unit, were evaluated. First, the sensitivity and specificity of the MN criteria for predicting clinical events were determined. The relationships between clinical events and the explanatory variables were then evaluated through univariate and multivariate analyses. Results The screening of the fulminant cases and classification of the non-fulminant cases into mild/moderate and severe/super severe cases resulted in a sensitivity of 1.00 and a specificity of 0.89. Univariate and multivariate analyses revealed a significant association of the serum albumin (Alb) level as well as white blood cell (WBC) count with clinical events. Conclusion The findings provide evidence supporting the accuracy of the MN criteria in predicting CDI severity and show that the Alb and WBC are important variables in predicting CDI severity.

Keywords: Clostridioides difficile; Clostridioides difficile infection; MN criteria; severity scoring system.

Figure 1.

Flow chart of the disposition of the cases included in this study according to the MN criteria. Alb: serum albumin; CDI: Clostridioides difficile infection; eGFR: estimated glomerular filtration rate; WBC: white blood cell

Figure 2.

Kaplan-Meier curve of the incidence of clinical events in the non-fulminant (n=56) and fulminant cases (n=10) and in the mild/moderate (n=17) and severe/super severe cases (n=5). (A) Clinical events were more likely to occur earlier and with greater incidence in the fulminant cases [median, 2 days (IQR, 2-3 days)] than in non-fulminant cases [median, 10 days (IQR, 7-21 days)]. (B) Clinical events occurred early to late [median, 10 days (IQR, 6-16.5 days)], and the incidence tended to be high in the severe/super severe cases.

Figure 3.

Correlation between the four explanatory variables and clinical events analyzed by quantification II. The contribution ratios of these variables were as follows: Alb, 44%; WBC, 38%; eGFR, 13%; and BT, 6%; the partial correlation coefficients for these variables were 0.86, 0.83, 0.60, and 0.33, respectively. Alb: serum albumin, BT: body temperature, eGFR: estimated glomerular filtration rate, WBC: white blood cell