ABSTRACT
Importance: Multisystem Inflammatory Syndrome in Children (MIS-C) associated with SARS-CoV-2 infection is thought to be driven by a post-viral dysregulated immune response, where interleukin 6 (IL-6) might have a central role. In this setting, IL-6 inhibitors are prescribed as immunomodulation in cases refractory to standard therapy. Objective: To compare plasma IL-6 concentration between critically ill children with MIS-C and sepsis. Design: A retrospective cohort study from previously collected data. Setting: Individual patient data were gathered from three different international datasets. Participants: Critically ill children between 1 month-old and 18 years old, with an IL-6 level measured within 48 h of admission to intensive care. Septic patients were diagnosed according to Surviving Sepsis Campaign definition and MIS-C cases by CDC criteria. We excluded children with immunodeficiency or immunosuppressive therapy. Exposure: None. Main Outcome(s) and Measure(s): The primary outcome was IL-6 plasma concentration in MIS-C and sepsis group at admission to the intensive care unit. We described demographics, inflammatory biomarkers, and clinical outcomes for both groups. A subgroup analysis for shock in each group was done. Results: We analyzed 66 patients with MIS-C and 44 patients with sepsis. MIS-C cases were older [96 (48, 144) vs. 20 (5, 132) months old, p < 0.01], but no differences in sex (41 vs. 43% female, p = 0.8) compared to septic group. Mechanical ventilation use was 48.5 vs. 93% (p < 0.001), vasoactive drug use 79 vs. 66% (p = 0.13), and mortality 4.6 vs. 34.1% (p < 0.01) in MIS-C group compared to sepsis. IL-6 was 156 (36, 579) ng/dl in MIS-C and 1,432 (122, 6,886) ng/dl in sepsis (p < 0.01), while no significant differences were observed in procalcitonin (PCT) and c-reactive protein (CRP). 52/66 (78.8%) patients had shock in MIS-C group, and 29/44 (65.9%) had septic shock in sepsis group. Septic shock had a significantly higher plasma IL-6 concentration than the three other sub-groups. Differences in IL-6, CRP, and PCT were not statistically different between MIS-C with and without shock. Conclusions and Relevance: IL-6 plasma concentration was elevated in critically ill MIS-C patients but at levels much lower than those of sepsis. Furthermore, IL-6 levels don't discriminate between MIS-C cases with and without shock. These results lead us to question the role of IL-6 in the pathobiology of MIS-C, its diagnosis, clinical outcomes, and, more importantly, the off-label use of IL-6 inhibitors for these cases.