Applying the bedside SaO2/FiO2 ratio to adolescents with acute chest syndrome to triage ICU transfer Free

Introduction We previously described the relationship between the diagnostic SaO₂/FiO₂ ratio – the ratio of the peripheral oxygen saturation by pulse oximetry (SpO₂) to the fraction of inspired oxygen (FiO₂) – and transfer to the intensive care unit (ICU) for adult patients with sickle cell disease (SCD) hospitalized with acute chest syndrome (ACS). This was the first study to characterize the SaO₂/FiO₂ ratio in SCD as a potential bedside triage tool, but it was restricted to adult patients (Wesevich et al., Blood Adv, 2025). In this study, we explore the utility of the SaO₂/FiO₂ ratio in adolescents with SCD.

Methods In this cross-sectional study, we included all ACS hospitalizations of adolescent patients (ages 12-17) with SCD at the University of Chicago from January 1, 2017, to September 30, 2024. Hospitalizations were included if ACS was diagnosed via standard criteria of new opacities on chest radiography plus fever, cough, shortness of breath, chest pain, or hypoxemia. We also included hospitalizations treated for ACS even if chest radiography lacked new opacities given the time delay in opacification and the pediatric clinical practice of minimizing radiation exposure leading to fewer repeat chest imaging than adult patients. Demographics, oxygenation data, and clinical data collected. Body mass index (BMI) was categorized to underweight, normal, overweight, and obese using standard pediatric percentile cutoffs. The SaO₂/FiO₂ ratio was calculated at ACS diagnosis based on the SpO₂ measurement and any methods of supplemental oxygenation and corresponding flow rates or FiO₂ settings. If ACS was treated without new opacities, then SaO₂/FiO₂ was calculated starting at the first dose of antibiotics. Clinical data included the number of opacified lobes on diagnostic chest radiography (0-2 vs 3+), red cell exchange [RCE] use, ICU transfer, length of stay (LOS), and 28-day readmission. The SaO₂/FiO₂ ratio cutoff for ICU transfer was calculated using the Youden Index. Multilevel models of logarithmically transformed SaO₂/FiO₂ ratios clustered hospitalizations by patient.

Results A total of 68 patients had 119 hospitalizations for ACS. Most patients were male (62%) and had Hb SS (82%) and normal BMI (72%); the mean age at admission was 15.1 years (SD 1.8). The median SaO₂/FiO₂ ratio at the time of ACS diagnosis was 388 (IQR 331-452). There were 25 ACS hospitalizations (21%) where the patient was treated based on symptoms without an initial new opacity; those hospitalizations had a similar median diagnostic SaO₂/FiO₂ ratio as the cases with new opacities (372 vs 388, p=0.9). The 4 hospitalizations with 3+ opacified lobes at diagnosis (4%) had a lower median diagnostic SaO₂/FiO₂ ratio than those with 0-2 opacified lobes (261 vs 388, p=0.001). The 7 hospitalizations that went on to receive RCE (6%) had a lower median diagnostic SaO₂/FiO₂ ratio than those without RCE (334 vs 388, p<0.001). The 15 hospitalizations that required an ICU transfer (13%) had a lower median diagnostic SaO₂/FiO₂ ratio than those that did not (268 vs 394, p<0.001). The Youden Index method determined the optimal diagnostic SaO₂/FiO₂ratio cutoff for ICU transfer was 338, which had 80% sensitivity and 79% specificity. Hospitalizations with a diagnostic SaO₂/FiO₂ ratio below 338 had 24.0-times higher hazard of ICU transfer than those above that cutoff after adjusting for demographic and clinical factors (p=0.005). When using the previously described adult ICU transfer cutoff of 310, hospitalizations had 41.1-times higher hazard of ICU transfer than those above the cutoff. Diagnostic SaO₂/FiO₂ ratio was not significantly associated with LOS after ACS diagnosis or 28-day readmission.

Conclusions The median diagnostic SaO₂/FiO₂ ratio in adolescent patients with SCD was higher than the adult population previously studied. Lower SaO₂/FiO₂ ratio was associated with more opacified lobes on chest radiography, RCE use, and ICU transfer. The adolescent SaO₂/FiO₂ ratio cutoff of 338 corresponds to SpO₂ below 98% on 2 L/min nasal cannula, which is more conservative than the proposed adult cutoff of 310 (< 90% on 2 L/min). Despite this, the adolescent cutoff had a higher sensitivity than the adult cutoff (80% vs 63%). This simple bedside triage tool should be prospectively evaluated in adolescents admitted with ACS to guide ICU transfer clinical decision-making.