Pediatric Acute Respiratory Distress Syndrome in South African PICUs: A Multisite Point-Prevalence Study

• Published in: Pediatric critical care medicine Vol. 24; no. 12; pp. 1063 - 1071
• Main Authors: Morrow, Brenda M, Lozano Ray, Eleonora, McCulloch, Mignon, Salie, Shamiel, Salloo, Asma, Appel, Ilse N, Du Plooy, Elri, Cawood, Shannon, Moshesh, Porai, Keeling, Kathryn H, Solomon, Lincoln J, Hlophe, Sbekezelo, Demopoulos, Despina, Parker, Noor, Khan, Ayesha Bibi, Naidoo, Kuban D, Argent, Andrew C
• Format: Journal Article
• Language: English
• Published: United States 01-12-2023
• ISSN: 1529-7535
• DOI: 10.1097/PCC.0000000000003330

To describe the prevalence of pediatric acute respiratory distress syndrome (pARDS) and the characteristics of children with pARDS in South African PICUs.
Observational multicenter, cross-sectional point-prevalence study.
Eight PICUs in four South African provinces.
All children beyond the neonatal period and under 18 years of age admitted to participating PICUs.
None.
Clinical and demographic data were prospectively collected on a single day of each month, from February to July 2022, using a centralized database. Cases with or at risk of pARDS were identified using the 2015 Pediatric Acute Lung Injury Consensus Conference criteria. Prevalence was calculated as the number of children meeting pARDS criteria/the total number of children admitted to PICU at the same time points. Three hundred ten patients were present in the PICU on study days: 166 (53.5%) male, median (interquartile range [IQR]) age 9.8 (3.1-32.9) months, and 195 (62.9%) invasively mechanically ventilated. Seventy-one (22.9%) patients were classified as being "at risk" of pARDS and 95 patients (prevalence 30.6%; 95% CI, 24.7-37.5%) fulfilled pARDS case criteria, with severity classified as mild (58.2%), moderate (25.3%), and severe (17.6%). Median (IQR) admission Pediatric Index of Mortality 3 risk of mortality in patients with and without pARDS was 5.6 (3.4-12.1) % versus 3.9 (1.0-8.2) % (p = 0.002). Diagnostic categories differed between pARDS and non-pARDS groups (p = 0.002), with no difference in age, sex, or presence of comorbidities. On multivariable logistic regression, increasing admission risk of mortality (adjusted odds ratio [aOR] 1.02; 95% CI, 1.00-1.04; p = 0.04) and being admitted with a respiratory condition (aOR 2.64; 95% CI, 1.27-5.48; p = 0.01) were independently associated with an increased likelihood of having pARDS.
The 30.6% prevalence of pARDS in South Africa is substantially higher than reports from other sociogeographical regions, highlighting the need for further research in this setting.