Test Description

SaO2 (Arterial Oxygen Saturation) measures the percentage of hemoglobin molecules in arterial blood that are bound to oxygen. It is directly measured via co-oximetry as part of arterial blood gas analysis, making it more accurate than pulse oximetry (SpO2), which estimates saturation using light absorption.

SaO2 is a critical indicator of oxygen-carrying capacity and reflects how efficiently the lungs are oxygenating blood. It provides immediate information about tissue oxygenation and guides oxygen therapy, mechanical ventilation adjustments, and assessment of respiratory failure.

SaO2 vs SpO2: Key Differences

  • SaO2 (Arterial Oxygen Saturation): Directly measured from arterial blood sample using co-oximetry. Gold standard for oxygenation assessment. Measures actual oxygen bound to hemoglobin.
  • SpO2 (Pulse Oximetry): Non-invasive estimate using light absorption through tissue. Correlates well with SaO2 in most situations but can be unreliable in shock, hypothermia, CO poisoning, methemoglobinemia, severe anemia, or with nail polish/dark skin pigmentation.

Oxygen-Hemoglobin Dissociation Curve

The relationship between PaO2 and SaO2 follows a sigmoid-shaped (S-shaped) curve, not a linear relationship. This has critical clinical implications:

  • Flat portion (upper plateau): SaO2 remains >90% even as PaO2 drops from 100 to 60 mmHg. This provides a "safety margin" - small drops in PaO2 don't immediately compromise oxygen delivery.
  • Steep portion (lower curve): Below PaO2 of 60 mmHg (SaO2 90%), the curve becomes very steep. Small decreases in PaO2 cause large drops in SaO2, rapidly compromising oxygen delivery.
  • Clinical significance: Once SaO2 drops below 90%, the patient is on the steep part of the curve and oxygenation can deteriorate rapidly.
Quick Reference
  • Normal Range: 95-100%
  • Hypoxemia: <90%
  • Critical Hypoxemia: <75%
  • Primary Use: Assessment of oxygen-carrying capacity and tissue oxygenation status
  • Sample Type: Arterial blood (measured directly from ABG)
  • Key Distinction: SaO2 (measured) vs SpO2 (pulse oximetry estimate)
  • Clinical Pearl: "SaO2 90% = PaO2 60 mmHg" - critical point on dissociation curve
Clinical Significance

Factors Affecting Oxygen Saturation

Multiple physiologic factors shift the oxygen-hemoglobin dissociation curve, altering the relationship between PaO2 and SaO2:

Right Shift (Decreased Affinity - O2 Released More Easily):

  • Decreased pH (Acidosis): Bohr effect - acidosis promotes oxygen release to tissues
  • Increased temperature (Fever): Heat increases oxygen unloading
  • Increased 2,3-DPG: Chronic hypoxia, anemia, high altitude cause 2,3-DPG increase
  • Increased PaCO2: Hypercapnia shifts curve right
  • Clinical effect: Lower SaO2 for given PaO2 (appears worse), but oxygen delivery to tissues improved

Left Shift (Increased Affinity - O2 Held More Tightly):

  • Increased pH (Alkalosis): Alkalosis causes hemoglobin to hold oxygen tighter
  • Decreased temperature (Hypothermia): Cold reduces oxygen release
  • Decreased 2,3-DPG: Stored blood, septic shock
  • Decreased PaCO2: Hypocapnia shifts curve left
  • Fetal hemoglobin: Higher oxygen affinity than adult hemoglobin
  • Clinical effect: Higher SaO2 for given PaO2 (appears better), but impaired oxygen delivery to tissues

Causes of Low SaO2 (Hypoxemia)

Respiratory Causes:

  • Ventilation-Perfusion (V/Q) Mismatch: Pneumonia, pulmonary embolism, COPD, asthma
  • Shunt: Atelectasis, pulmonary edema, ARDS, right-to-left cardiac shunt
  • Diffusion Impairment: Interstitial lung disease, pulmonary fibrosis
  • Hypoventilation: Drug overdose, neuromuscular disease, CNS depression
  • High Altitude: Decreased atmospheric oxygen (low FiO2)

Cardiac Causes:

  • Low Cardiac Output: Heart failure, cardiogenic shock
  • Right-to-Left Shunt: Congenital heart disease (Tetralogy of Fallot, etc.)

Hemoglobin Abnormalities:

  • Carbon Monoxide Poisoning: COHb binds hemoglobin with 200x affinity of oxygen
  • Methemoglobinemia: Ferric iron (Fe3+) cannot bind oxygen
  • Severe Anemia: Reduced oxygen-carrying capacity

Causes of SaO2/SpO2 Discrepancy

Conditions where SpO2 (pulse oximetry) is unreliable or falsely reassuring:

CO Poisoning (Carbon Monoxide):

  • SpO2: Falsely NORMAL or HIGH (90-100%) - pulse oximeter cannot distinguish carboxyhemoglobin (COHb) from oxyhemoglobin
  • SaO2 (measured): LOW - co-oximetry directly measures COHb separately
  • Clinical pearl: Patient may appear "cherry red" with normal SpO2 but severely hypoxic. Always obtain ABG with co-oximetry if CO poisoning suspected.

Methemoglobinemia:

  • SpO2: Fixed at ~85% regardless of actual oxygenation - methemoglobin absorbs both red and infrared light equally
  • SaO2: Variable depending on actual oxygen saturation
  • Clinical pearl: "Chocolate brown" blood. Causes include nitrites, benzocaine, dapsone, sulfonamides.

Other Causes of SpO2 Unreliability:

  • Shock/Poor Perfusion: Inadequate pulse signal for pulse oximeter
  • Severe Anemia: Insufficient hemoglobin for accurate measurement
  • Hypothermia: Vasoconstriction impairs peripheral perfusion
  • Nail Polish/Artificial Nails: Interferes with light transmission (especially blue, green, black)
  • Dark Skin Pigmentation: May cause falsely high SpO2 readings
  • Motion Artifact: Movement during measurement
Critical Warning: Never rely on pulse oximetry alone in suspected CO poisoning, methemoglobinemia, or shock states. Obtain arterial blood gas with co-oximetry for direct measurement of SaO2, COHb, and MetHb.
Interfering Factors

Factors Affecting SaO2 Measurement (Co-oximetry)

Pre-Analytical Factors

  • Air Bubbles in Sample: Can falsely elevate PaO2 and SaO2
  • Delayed Analysis: Cellular metabolism continues, decreasing PaO2/SaO2 if not analyzed quickly or kept on ice
  • Excessive Anticoagulant: Dilutes sample, affecting results
  • Venous Contamination: Falsely low PaO2 and SaO2 if venous blood mixed with arterial sample

Patient Factors Affecting Pulse Oximetry (SpO2) Accuracy

  • Poor Perfusion: Shock, hypothermia, peripheral vascular disease, vasoconstriction
  • Motion: Shivering, seizures, patient movement
  • Ambient Light: Bright surgical lights, fluorescent lighting
  • Nail Polish: Especially blue, green, black colors (remove or use alternate site)
  • Anemia: Hgb <5 g/dL may affect accuracy
  • Skin Pigmentation: Dark skin may cause falsely high readings
  • Edema: Peripheral edema can impair signal

Hemoglobin Variants and Dyshemoglobinemias

  • Carboxyhemoglobin (CO poisoning): SpO2 falsely normal; SaO2 (co-oximetry) accurate
  • Methemoglobin: SpO2 reads ~85%; SaO2 may be different
  • Fetal Hemoglobin: Generally does not significantly affect pulse oximetry
  • Intravascular Dyes: Methylene blue, indocyanine green can cause false desaturation readings

Medications Causing Methemoglobinemia

  • Topical Anesthetics: Benzocaine, lidocaine (high dose)
  • Antibiotics: Dapsone, sulfonamides, trimethoprim
  • Antimalarials: Primaquine, chloroquine
  • Other Drugs: Nitrates, nitrites, metoclopramide, phenazopyridine
  • Chemical Exposures: Aniline dyes, nitrobenzene
Clinical Pearls
"90-60 Rule": SaO2 of 90% corresponds to PaO2 of approximately 60 mmHg. This is the critical inflection point on the oxygen-hemoglobin dissociation curve. Below this point, the curve is steep and oxygenation deteriorates rapidly.
Steep Part of Curve: Once SaO2 drops below 90%, small decreases in PaO2 cause large drops in saturation. This is where patients "fall off the cliff" and decompensate quickly. Aggressive oxygen therapy is critical.
SpO2 Unreliability in Shock: Never trust pulse oximetry in shock states, hypothermia, or poor perfusion. Obtain ABG with co-oximetry for accurate SaO2 measurement. SpO2 can read normal while patient is profoundly hypoxic.
CO Poisoning Pearl: Carbon monoxide poisoning shows normal or high SpO2 (90-100%) but LOW SaO2 on ABG co-oximetry. Pulse oximeter cannot distinguish COHb from O2Hb. Always check COHb level with co-oximetry if CO exposure suspected. Patient may look "cherry red" despite severe tissue hypoxia.
Methemoglobinemia Pearl: SpO2 reads fixed at ~85% regardless of supplemental oxygen. Blood appears "chocolate brown" and doesn't turn red with oxygen exposure. Causes include benzocaine spray (common in procedural sedation), dapsone, and nitrites. Treat with methylene blue 1-2 mg/kg IV.
Pulse Oximetry Limitations: SpO2 accuracy decreases when SaO2 <80%. Pulse oximeters are calibrated using healthy volunteers (ethical reasons prevent testing at very low saturations). Below 80%, readings become increasingly unreliable.
Right Shift is Adaptive: In chronic hypoxia (COPD, high altitude), the oxygen-hemoglobin curve shifts right via increased 2,3-DPG. While SaO2 may appear lower for a given PaO2, oxygen is released more easily to tissues - this is a beneficial adaptation.
A-a Gradient Correlation: Use SaO2 with PaO2 to assess for V/Q mismatch or shunt. Normal A-a gradient with low PaO2/SaO2 suggests hypoventilation or high altitude. Widened A-a gradient suggests intrinsic lung pathology (pneumonia, PE, ARDS).
Target Saturations: In most patients, target SaO2 94-98%. In COPD patients at risk for CO2 retention, use lower target of 88-92% to avoid suppression of hypoxic ventilatory drive. In neonates, avoid hyperoxia (SaO2 >95%) to prevent retinopathy of prematurity.
Don't Trust SpO2 in Dark Skin: Recent studies show pulse oximeters overestimate SaO2 in patients with dark skin pigmentation, leading to missed hypoxemia. When in doubt, obtain ABG for direct SaO2 measurement, especially in critically ill patients.
Venous vs Arterial: Central venous oxygen saturation (ScvO2) from central line is typically 70-75% (lower than arterial). Mixing venous blood in arterial sample will falsely lower SaO2. Ensure bright red, pulsatile arterial flow when obtaining ABG.
Oxygen Toxicity Risk: While treating hypoxemia is critical, prolonged exposure to high FiO2 (especially 100% oxygen) can cause oxygen toxicity and absorptive atelectasis. Once SaO2 >92%, titrate FiO2 down to lowest level maintaining adequate oxygenation.
References
  1. Kratz, A., Ferraro, M., Sluss, P. M., & Lewandrowski, K. B. (2004). Laboratory reference values. New England Journal of Medicine, 351, 1548-1564.
  2. Lee, M. (Ed.). (2009). Basic skills in interpreting laboratory data. Ashp.
  3. Farinde, A. (2021). Lab values, normal adult: Laboratory reference ranges in healthy adults. Medscape. https://emedicine.medscape.com/article/2172316-overview?form=fpf
  4. Nickson, C. (n.d.). Critical Care Compendium. Life in the Fast Lane • LITFL. https://litfl.com/ccc-critical-care-compendium/
  5. Farkas, Josh MD. (2015). Table of Contents - EMCrit Project. EMCrit Project. https://emcrit.org/ibcc/toc/
Back to ABG Panel All Lab Values
Medical Disclaimer
  • For Educational Purposes Only: This content is intended for educational reference and should not be used for clinical decision-making.
  • Not a Substitute for Professional Judgment: Always consult your local protocols, institutional guidelines, and supervising physicians.
  • Verify Before Acting: Users are responsible for verifying information through authoritative sources before any clinical application.
AI Assistance Notice
The clinical content and references are curated and reviewed by myself; however, AI was used to assist in organizing, paraphrasing, and formatting the information presented.