New sepsis-associated morbidity and mortality in pediatric oncology patients

Sepsis is a leading cause of morbidity and mortality in children worldwide, yet the development of new morbidity after sepsis has not been clearly defined in high-risk subgroups such as children with cancer. Using the TOPICC (Trichotomous Outcome Prediction in Critical Care) multicenter cohort study dataset, we evaluated whether children with cancer have a higher risk of the composite outcome of death or new morbidity at hospital discharge compared to children without cancer. Among 854 children with sepsis, 88 patients (10.3%) had an underlying cancer diagnosis. Children with cancer were older (median 8.1 vs 3.7 years) and more frequently developed sepsis while in the hospital. The pattern of organ failure differed between groups, with less frequent invasive mechanical ventilation (26.1% vs 49.9%, p<0.001) but more frequent vasoactive infusions (47.7% vs 35.8%, p=0.03) in children with cancer compared to non-oncology patients. Children with cancer had an increased rate of death or new morbidity (22.7% vs 12.1%, p=0.006) compared to non-oncology patients. New morbidity (defined by ΔFSS score >2 points) occurred in 13.9% of cancer vs 6.9% of non-cancer survivors (p=0.03), and PICU mortality was similar between groups (10.2% vs 5.6%, p=0.09). Cancer diagnosis was independently associated with higher odds of death or new disability at discharge (adjusted odds ratio 3.71, p<0.001) in multivariable logistic regression, after adjusting for baseline FSS, baseline developmental delay, clinical concern for neurologic injury on PICU admission, and PICU supportive measures. These results suggest that children with cancer who develop sepsis are more likely to experience adverse outcomes at hospital discharge, even after accounting for baseline health and critical illness severity.