Defining Severe Shigellosis in the Enterics for Global Health Study: A Comparison of Leading Diarrhea Severity Definitions Among Children With Shigella Diarrhea

Background Shigella vaccine development is a critical priority due to high burden and long-term outcomes among children. Choosing a clinical efficacy endpoint for Shigella vaccine trials requires comparing existing diarrhea severity definitions among children with Shigella. Methods Six-to 35-month-old children presenting with diarrhea were enrolled at 7 EFGH study sites. Among children with Shigella, 6 diarrhea severity definitions (Global Enteric Multicenter Study [GEMS] moderate-to-severe diarrhea [MSD], modified Vesikari score [MVS], MVS +/- dysentery, Clark score, MAL-ED score, and Shigella mortality score) were examined against 2 outcomes: change in length/height-for-age z-score (ΔLAZ/HAZ) over 3 months and death or hospitalization within 14 days. We compared the performance of each severity score to predict outcomes using linear regression and measures of diagnostic accuracy. Results Moderate or severe diarrhea was negatively associated with ΔLAZ/HAZ by MVS (-0.07 z-score, 95% CI -0.10 to -0.03), MVS +/- dysentery (-0.05, 95% CI -0.09 to -0.02), and MAL-ED score (-0.04, 95% CI -0.07 to 0.00). GEMS MSD (93.6%), MAL-ED (87.2%), MVS +/- dysentery (85.1%), and MVS (80.9%) had high sensitivity in predicting death or hospitalization. Conclusions MVS, MVS +/- dysentery, and MAL-ED score identified adverse outcomes following Shigella diarrhea and are viable options for a vaccine trial case definition.