Triaging Acute Chest Syndrome Clinical Decision-Making Using Bedside SaO2/FiO2 Ratio

Introduction

Acute chest syndrome (ACS) is a life-threatening complication of sickle cell disease (SCD). ACS severity is inconsistently defined, typically using either the number of involved lung lobes on chest imaging with possible delays in chest imaging changes or the level of transfusion support, which can be subjectively applied. Additionally, the clinical course of ACS is difficult to predict. Thus, we evaluated the utility of the ratio of the peripheral oxygen saturation obtained from pulse oximetry to the fraction of inspired oxygen (SaO₂/FiO₂) in defining ACS severity and associations with clinical outcomes.

Methods

This retrospective observational study included all ACS admissions of patients 18 years or older with SCD at a single health center from January 1, 2017 to December 31, 2021. ACS was defined as having at least one symptom (fever, chest pain, hypoxemia, cough, wheezing, or shortness of breath) and a new opacity on chest radiography. Clinical and oxygenation data were collected via manual chart review. Arterial oxygen saturation (SaO₂) was approximated by the pulse oximetry (SpO₂) measurement recorded in the electronic medical record flowsheets. Every SpO₂ measurement from presentation in the ED until the first 96 hours after ACS diagnosis was recorded. FiO₂ was calculated based on the type of oxygen support. We stratified admissions by home oxygen support given the impact of supplemental oxygen on the SaO₂/FiO₂ ratio. We compared the SaO₂/FiO₂ ratio to two existing methods of defining ACS severity in the literature: 1) the number of opacified lobes on chest radiography (1-2 versus 3+ lobes) and 2) the type of transfusions (simple versus red cell exchange [RCE]). We also compared the SaO₂/FiO₂ ratio to the clinical outcome of intensive care unit (ICU) transfer, which was equivalent to the composite endpoint of ICU transfer or death. Three key timepoints were determined in the patient’s clinical course for determining the utility of triaging a patient based on their current SaO₂/FiO₂ ratio: 1) presentation in the emergency department (ED), 2) initiation of antibiotics for ACS, and 3) chest radiography confirmation of ACS diagnosis. Analyses were stratified by whether patients used home oxygen or not.

Results

Of the 227 ACS admissions during the study period, 54% were female, the mean age at admission was 28.9 years (SD 9.7), and 70% had Hgb SS. Admissions with an ICU transfer had a lower hemoglobin and platelet count at ACS diagnosis than those that remained on the floor (Hgb 7.3 vs 7.7 g/dL, p=0.03; platelet 303,000/µL vs 376,000/µL, p=0.01). Of the 227 admissions, 32 admissions (14%) were severe based on having 3+ lobes with opacities on chest radiography, and 45 admissions (20%) were severe due to receiving RCE. Home oxygen use was documented for 43 (19%) admissions. For admissions without home oxygen, chest imaging with 3+ opacified lobes had lower SaO₂/FiO₂ than those with 1-2 lobes (322 vs 383, p=0.006), and RCE was associated with lower SaO₂/FiO₂ than no RCE (293 vs 414, p<0.001). Similarly, admissions with home oxygen had lower SaO2/FiO2 when RCE was used (254 vs 303, p=0.007). For the 3 key timepoints regardless of home oxygen status, SaO₂/FiO₂ ratio did not correlate with subsequent ICU transfer at ED presentation (AUC 0.59), but it did correlate strongly at antibiotic initiation (AUC 0.78) and at ACS diagnosis (AUC 0.76). Thus, SaO₂/FiO₂ ratio cutoffs of 314 (sensitivity 54%, specificity 85%) for patients without home oxygen and 276 (sensitivity 65%, specificity 85%) for patients with home oxygen were calculated for triaging whether patients would transfer to the ICU or not. These cutoffs were highly associated with ICU transfer in multivariable multilevel models (314: aOR 12.2, 95% CI 2.9-51.6; 276: aOR 6.9, 95% CI 1.7-28.6).

Conclusions

The SaO₂/FiO₂ ratio is associated with ACS severity and clinical outcomes in hospitalized patients with SCD and warrants further evaluation. The recommended SaO₂/FiO₂ ratio cutoff of 314 for patients without home oxygen is equivalent to saturation of 91% on 2 L/min nasal cannula, and the recommended SaO₂/FiO₂ ratio of 276 for patients with home oxygen is equivalent to saturation of 91% on 3 L/min. This bedside measurement of oxygen status may be a clinically useful, inexpensive biomarker for assessing ACS severity and guiding clinical management such as transfer to an ICU. Further study of these cutoffs through a prospective multi-center study is indicated.