Outcomes of immunocompromised children hospitalized for Influenza, 2010-2021, the Canadian Immunization Monitoring Program Active (IMPACT)

Abstract

ObjectivesTo evaluate immunocompromising conditions and subgroups of immunocompromise as risk factors for severe outcomes among children admitted for influenza. MethodsWe performed active surveillance for laboratory-confirmed influenza hospitalizations among children ≤16 years old at the 12 Canadian Immunization Monitoring Program Active hospitals, during 2010–2021. Logistic regression analyses were used to compare outcomes between immunocompromised and non-immunocompromised children, and for different subgroups of immunocompromise. The primary outcome was intensive care unit (ICU) admission; the secondary outcomes were mechanical ventilation and death. ResultsAmong 8982 children, 892 (9.9%) were immunocompromised; these patients were older (median, 5.6 (IQR, 3.1–10.0) vs. 2.4 (1–6) years; p < 0.001) than non-immunocompromised children, had a similar frequency of comorbidities, excluding immunocompromise and/or malignancy (38% (340/892) vs. 40% (3272/8090); p 0.2), but fewer respiratory symptoms, such as respiratory distress (20% (177/892) vs. 42% (3424/8090), p < 0.001).In multivariable analyses, immunocompromise (adjusted odds ratio (aOR), 0.19; 95% CI, 0.14–0.25) and its subcategories immunodeficiency (aOR, 0.16; 95% CI, 0.10–0.23), immunosuppression (aOR, 0.17; 95% CI, 0.12–0.23), chemotherapy (aOR, 0.07; 95% CI, 0.03–0.13), and solid organ transplantation (aOR, 0.17; 95% CI, 0.06–0.37) were associated with decreased probability of ICU admission in children admitted for influenza. Immunocompromise was also associated with a decreased probability of mechanical ventilation (aOR, 0.26; 95% CI, 0.16–0.38) or death (aOR, 0.22; 95% CI, 0.03–0.72). ConclusionImmunocompromised children are overrepresented among hospitalizations for influenza, but have a decreased probability of ICU admission, mechanical ventilation, and mortality following admission. Admission bias precludes generalizability beyond the hospital setting.