Pulse Oximetry and Arterial Saturation Difference in Pediatric COVID-19 Patients: Retrospective Analysis by Race

Authors

Fabio Savorgnan
Adel Hassan
Nirica Borges
Sebastian Acosta

Publication Date

6-1-2023

Journal

Pediatric Critical Care Medicine

DOI

10.1097/PCC.0000000000003208

PMID

36825900

PMCID

PMC10226388

PubMedCentral® Posted Date

2-24-2023

PubMedCentral® Full Text Version

Post-print

Published Open-Access

yes

Keywords

Humans, Child, Retrospective Studies, Oxygen, COVID-19, Oximetry, Hypoxia, disparities, occult hypoxemia, oxygen saturation, race, severe acute respiratory syndrome coronavirus 2

Abstract

OBJECTIVES: Pulse oximetry (Sp o2 ) may overestimate arterial oxygen saturation (Sa o2 ) in blood laboratory testing. This study aimed to assess Sp o2 -Sa o2 difference in relation to race (i.e., patient self-reporting as Black or White), occult hypoxemia, and length of stay (LOS) in pediatric patients with COVID-19.

DESIGN: Single-center retrospective study in pediatric COVID-19 patients. We used multivariable linear regressions to examine the association between race and oximetry measurements and between occult hypoxemia and LOS. Oximetry bias was defined using Sp o2 and Sa o2 data according to approved comparisons. Occult hypoxemia was defined as Sp o2 greater than 92% and Sa o2 less than 88%.

SETTING: Quaternary pediatric hospital.

PATIENTS: Pediatric COVID-19 patients admitted to Texas Children's Hospital between May 2020 and December 2021.

INTERVENTIONS: None.

MEASUREMENTS AND MAIN RESULTS: There were 2713 patients with complete physiological data in the analysis. Of the total, 61% were Black, and 39% were White. Oximetry bias was greater in Black compared with White patients ( p < 0.001), and this bias increased as the oxygen saturations decreased ( p < 0.001). Black and White patients had a 12% and 4% prevalence of occult hypoxemia, respectively ( p < 0.001). LOS was not associated with oximetry bias or occult hypoxemia once controlled for the level of support (intensive care, respiratory, circulatory).

CONCLUSIONS: We found an oximetry bias in the measurement of Sp o2 with respect to Sa o2 in symptomatic hospitalized pediatric patients with the diagnosis of COVID-19. Furthermore, race is related to an increased oximetry bias. However, we did not find a relationship between oximetry bias and the LOS in the hospital in this cohort of patients.