Pulse Oximetry-ABG Discrepancy: Carbon Monoxide Poisoning
The most likely cause of a 100% SpO2 with a PaO2 of 59 mmHg is carbon monoxide (CO) poisoning, where carboxyhemoglobin falsely elevates the pulse oximetry reading while true oxygen delivery remains critically impaired.
Understanding the Discrepancy
Pulse oximeters cannot detect carboxyhemoglobin or methemoglobin, leading to falsely normal or elevated readings despite severe hypoxemia 1. The device measures light absorption at only two wavelengths and calculates the oxygenated fraction of available hemoglobin, but carboxyhemoglobin absorbs light similarly to oxyhemoglobin, causing the oximeter to register it as oxygen-saturated hemoglobin 1.
Key Physiologic Principles
- PaO2 of 59 mmHg represents severe hypoxemia requiring immediate intervention, as values <60 mmHg indicate significant oxygen deficiency 2
- Despite this critically low PaO2, the oxyhemoglobin dissociation curve shows that even a PaO2 of 70 mmHg would maintain saturation above 93% under normal conditions 1
- A PaO2 of 59 mmHg should correspond to an SaO2 of approximately 88-90%, not 100% 1
Differential Diagnosis Algorithm
Primary Consideration: Carbon Monoxide Poisoning
- Standard pulse oximetry cannot differentiate carboxyhemoglobin, as noted by the American College of Physicians 3
- CO binds hemoglobin with 200-250 times greater affinity than oxygen, displacing oxygen while appearing "saturated" to the pulse oximeter
- This creates the classic discrepancy: normal or high SpO2 with low PaO2 and actual tissue hypoxia
Secondary Consideration: Methemoglobinemia
- Methemoglobin also causes falsely elevated pulse oximetry readings 1
- Typically causes SpO2 to read around 85% regardless of actual oxygenation, though can vary
- Less likely to produce a 100% reading compared to CO poisoning
Technical/Equipment Issues (Less Likely)
- Pulse oximeters tend to overestimate true arterial O2 saturation by approximately 2.75% in critically ill patients 4
- In hypoxemic patients (SaO2 <90%), the mean overestimation increases to 4.9% 4
- However, this degree of overestimation cannot fully explain a 100% reading with PaO2 of 59 mmHg
Clinical Management Approach
Immediate Actions
- Obtain co-oximetry on the ABG sample to directly measure carboxyhemoglobin and methemoglobin levels 3
- Either arterial or venous blood can measure COHb levels, as they are similar 2
- Do not rely on pulse oximetry for clinical decision-making in this scenario 3
Risk Factors to Assess
- Exposure history: enclosed space fires, faulty heating systems, vehicle exhaust, industrial exposure
- Active smoking status: smokers have higher rates of occult hypoxemia (13% false negative rate vs 10% overall) 5
- Skin pigmentation: Black patients have higher rates of SpO2-SaO2 discrepancies (aOR 1.35) and occult hypoxemia (aOR 1.22) 6
Critical Clinical Pitfalls
Common Errors to Avoid
- Never assume a normal SpO2 negates the need for ABG analysis in critically ill or symptomatic patients 3
- A SpO2 >94% is necessary to ensure SaO2 ≥90% in most clinical scenarios 7, 5
- In patients with suspected CO poisoning, 50% of cases with SpO2 90-93% actually have SaO2 <90% 4
When to Obtain ABG Instead of Relying on Pulse Oximetry
- Any patient on supplemental oxygen therapy 3
- SpO2 ≤94% warrants reflex ABG measurement 5
- Suspected CO poisoning, methemoglobinemia, or cyanide toxicity 3
- Critically ill patients requiring precise oxygenation assessment 3, 8
- Active smokers or patients with significant comorbidities (higher Charlson Comorbidity Index scores predict discrepancies) 6
Outcome Implications
Occult hypoxemia (PaO2 <60 mmHg with SpO2 >88%) is associated with significantly worse outcomes, including:
- Longer ICU length of stay (IRR 1.12) 6
- Higher SOFA scores at 24 hours (β = 0.31) 6
- Dramatically increased in-hospital mortality (aOR 1.73) 6
This underscores the critical importance of recognizing and correctly diagnosing this discrepancy pattern, as delayed recognition of true hypoxemia directly impacts patient survival.