• Journal: Anesthesiology

• PMID: 35231085

• Type of study: Retrospective cohort study

• Hypothesis

  • There will be a greater prevalence of occult hypoxemia in anesthetized patients who self-reported race/ethnicity other than White 

• Methods

  • Single-center

    • 46,253 patients spanning 12 years

    • 151,070 paired SaO2-SpO2 for 46,253 patients

  • Adult patients who received an anesthetic with at least 1 ABG

  • Racial/ethnic groups

    • White, Black, Hispanic, Asian, Other

  • Primary Outcome

    • Occult hypoxemia: SpO2 >92% and ABG O2 saturation <88%

    • SpO2 value

      • Calculated as the mean SpO2 value during the 5 minute interval starting 10 mins before ABG drawn

• Results

  • Racial/ethnic group populations

    • White (47.8%), Black(11.2%), Hispanic (13.6%), Asian (5.6%), Other (21.8%)

  • Primary outcome - Prevalence of occult hypoxemia

    • Overall: 1.3%

    • White: 1.1%, Black: 2.1%, Hispanic: 1.8%, Asian: 1.1% , Other: 1.2%

    • Black (OR 1.44) and Hispanic (OR 1.31) race/ethnicity associated with an increased risk of occult hypoxemia as compared to White

• Strengths

  • Large # of patients

  • Large # of paired SpO2/SaO2 values

  • Generalizability

  • Large # of demographic factors studied

• Weaknesses

  • Single-center

  • Only included cases with arterial lines

  • Don't know exact SpO2 when ABG drawn

  • Didn't directly measure skin pigmentation

  • "Other" race/ethnicity category

  • Used 2 different brands of pulse-oximeters

  • No info on other potentially important variables (ex: vasopressor use, placement of pulse-ox probe)

• Key Point: The prevalence of occult hypoxemia is greater in patients who self-report as Black or Hispanic as compared to White.

Anesthesiology. 2022 May 1;136(5):688-696