The pulse oximetry gap in carbon monoxide intoxication

Ann Emerg Med. 1994 Aug;24(2):252-5. doi: 10.1016/s0196-0644(94)70137-7.


Study objective: Pulse oximetry has been reported to be falsely elevated in the presence of carbon monoxide (CO). However, the degree to which pulse oximetry overestimates measured oxyhemoglobin saturation (O2Hb) has not been investigated in patients with CO exposure. This study quantifies the effect of CO on pulse oximetry and O2Hb in a series of patients with elevated carboxyhemoglobin (COHb) levels.

Methods: A prospective case series of 25 pulse oximetry measurements, with concurrent arterial blood gas sampling, were obtained on 16 adults with CO exposure.

Results: COHb levels (mean, 16.1%; SD, 11.6%; range, 2.2% to 44%) did not significantly correlate with pulse oximetry saturation (mean, 97.7%; SD, 1.5%; range 96% to 100%) (r = .45; P = .1 [NS]). Compared with COHb, a pulse oximetry gap (mean, 17.5%; SD, 1.5%; range, 2.3% to 42%), defined as pulse oximetry saturation minus O2Hb, yielded a linear regression model: pulse oximetry gap = 1.82 + 0.94 x COHb (SEM = 0.07; F = 204; R2 = .90; P < .0001).

Conclusion: Oxygen saturation as measured by pulse oximetry failed to decrease to less than 96% despite COHb levels as high as 44%. Regression between the pulse oximetry gap and COHb suggests that pulse oximetry overestimates O2Hb by the amount of COHb present. Pulse oximetry is unreliable in estimating O2Hb saturation in CO-exposed patients and should be interpreted with caution when used to estimate oxygen saturation in smokers.

MeSH terms

  • Adult
  • Carbon Monoxide Poisoning / blood*
  • Carboxyhemoglobin / analysis*
  • Humans
  • Linear Models
  • Oximetry*
  • Oxyhemoglobins / analysis*
  • Prospective Studies


  • Oxyhemoglobins
  • Carboxyhemoglobin