This multicenter retrospective cohort study evaluated whether obesity is associated with intensive care unit (ICU) length of stay (LOS) in children admitted to pediatric ICUs with critical asthma. The investigators also assessed whether socioeconomic markers modify this relationship. The analysis used data from the Virtual PICU Systems (VPS) database covering admissions from 2010 through 2020.
The cohort comprised 16,412 children aged 2–19 years from 53 centers with sufficient data to assign body mass index (BMI) percentile, race, and insurance type. Children with chronic medical conditions and those missing relevant data were excluded. The dataset and time window were derived from entries in the VPS database; no interventions were applied as part of this observational analysis.
The primary outcome was the hazard of ICU discharge among survivors, modeled as a time-to-event variable with ICU LOS as the event time. Secondary outcomes included use of invasive or noninvasive mechanical ventilation (MV) within the first hour of ICU admission and MV use after the first hour. The study used multivariable Cox proportional hazards models for the primary outcome and logistic regression models for the binary ventilation outcomes.
Models were adjusted for potential confounders: age, sex, race, insurance type (as a proxy for socioeconomic status), pneumonia diagnosis, transfer status, and overall severity of illness as measured by the Pediatric Risk of Mortality (PRISM) III score.
Obesity was associated with a lower hazard of ICU discharge among survivors, with a hazard ratio (HR) of 0.87 (95% CI, 0.83–0.91). A HR less than 1.0 indicates a relatively reduced rate of discharge and is consistent with a longer ICU LOS for children classified as obese compared with non-obese peers in the adjusted models.
This association remained after adjustment for the listed demographic, clinical, and severity variables, supporting an independent relationship between obesity and prolonged ICU stay in this population.
Obesity was associated with increased odds of receiving mechanical ventilation. Specifically, obese children had higher odds of requiring MV within the first hour of ICU admission (odds ratio [OR] 1.24; 95% CI, 1.12–1.39) and higher odds of MV after the first hour (OR 1.63; 95% CI, 1.39–1.92).
These associations were observed in models that adjusted for age, sex, race, insurance type, pneumonia diagnosis, transfer status, and PRISM III score, indicating that the increased need for ventilatory support among obese children with critical asthma was not explained by those measured confounders.
The authors evaluated whether race or insurance type modified the association between obesity and ICU outcomes. Overall, race and insurance type did not significantly modify the relationship between obesity and longer ICU LOS or increased ventilatory support. An exception was noted among Black children: obesity was not associated with longer ICU LOS in this subgroup (HR 0.95; 95% CI, 0.89–1.03; p = 0.22). The report states that otherwise the findings were predominantly consistent across racial and socioeconomic groups.
In this large multicenter sample of children with critical asthma admitted to PICUs, obesity was independently associated with longer ICU stays and higher likelihood of needing invasive or noninvasive mechanical ventilation, both within the first hour and later during the ICU stay. These results persisted after adjustment for demographic factors, a pneumonia diagnosis, transfer status, and severity of illness.
The authors conclude that obesity contributes to increased ICU resource use and respiratory support requirements among children with critical asthma, with these associations largely consistent across racial and socioeconomic strata but with a noted lack of association for ICU LOS within the subgroup of Black children.
One author (Dr. Bhalla) reported institutional funding from the National Heart, Lung, and Blood Institute and support from the National Institutes of Health for related research (R01HL173488 and K23HL153756). The remaining authors disclosed no potential conflicts of interest.
Note: This summary and analysis are based exclusively on the abstract and metadata as reported in the source. Additional methodological details, subgroup analyses, and limitations beyond what is presented in the abstract were not reported in the source material provided.