In a recently released study in Pediatrics (10.1542/peds.2019-2829), Dr. Itay Ayalon and colleagues on behalf of the AWARE (Assessment of Worldwide Acute 3 Kidney Injury, Renal Angina and Epidemiology) study group examined the interesting question of whether weight status and 28-day mortality in critically ill pediatric patients are related. The authors hypothesized that obesity, as compared to underweight, normal weight and overweight, would be associated with higher mortality and worse outcomes. To study this question, they completed a secondary analysis of a multicenter, international study that had enrolled critically ill children and young adults, the AWARE trial, whose primary objective was to define the epidemiology of acute kidney injury. AWARE enrolled a total of 6,821 children and young adults ages 3 months to 25 years whose predicted stay in one of 32 pediatric intensive care units (PICUs) across the globe was at least 48 hours. AWARE data were collected prospectively and were comprehensive with respect to patient characteristics and hospital course.
The authors define the weight groups by BMI adjusted for age and sex, and explain their statistical approach using multivariable logistic regression in clear terms. In this international group, 28.5% were underweight, almost half were normal weight (44.3%) and overweight and obese children and young adults represented 11.4% and 15.8% of the cohort respectively. The median age of those studied was 5.3 years. As planned for the AWARE study, a separate sub-analysis was conducted for those with sepsis.
So, I have to admit, I was completely unprepared for the results – I assumed that this was yet another study which would demonstrate the risks associated with obesity and overweight. And I was completely wrong! It was actually most dangerous to be underweight, and this result was found both in the main cohort (1.8-fold increased risk of 28-day mortality [95% 11 confidence interval (CI): 1.2–2.8]) and the sepsis sub-cohort (2.9-fold increase 13 [95% CI: 1.2–6.9]). Being overweight or obese was not protective as compared to normal weight, but on the other hand, did not increase risk of death. The study also examined the effect of co-morbidities and admission diagnoses on 28-day mortality, and on additional secondary outcomes, both for the entire group and by weight cohort. The results are fascinating because they are so counterintuitive. The Discussion leads us through some controversial and interesting areas, such as the association of fluid overload and mortality: underweight participants in this study were more likely to develop fluid overload than those in the other weight groups. This is a thought provoking study, and will be of interest to all practitioners, especially as we struggle to address the obesity epidemic in our own country.
