Pulse oximetry has become inseparable from prehospital assessment. The probe goes on, a number appears and, within seconds, we have an apparently objective measure of how well the patient is oxygenating.
It is quick, non-invasive and extraordinarily useful. It can also be falsely reassuring.
Can we trust a normal SpO₂—or are we asking the number to tell us more than it actually measures?
WHAT IS THE PULSE OXIMETER ACTUALLY MEASURING?
A pulse oximeter estimates the percentage of haemoglobin binding sites occupied by oxygen from the different light-absorption characteristics of oxygenated and deoxygenated haemoglobin.4
The displayed value is an estimate of peripheral oxygen saturation. It is not a direct measurement of arterial oxygen partial pressure, carbon dioxide, adequacy of ventilation, haemoglobin concentration, cardiac output, tissue perfusion or cellular oxygen use.
OXYGENATION IS NOT VENTILATION
Oxygenation describes movement of oxygen into the blood. Ventilation describes movement of air into and out of the lungs and elimination of carbon dioxide.
A patient can maintain an apparently normal SpO₂ while ventilation deteriorates, particularly when supplemental oxygen is administered. Supplemental oxygen can delay recognition of hypoventilation by pulse oximetry; oxygen may preserve saturation while carbon dioxide rises and respiratory effort falls.3
THEORY IN ONE LINEA normal SpO₂ does not tell us that ventilation or oxygen delivery is normal.
SATURATION IS NOT OXYGEN CONTENT
A profoundly anaemic patient may have an SpO₂ of 99%. The available haemoglobin may be almost completely saturated while total oxygen-carrying capacity remains substantially reduced. An acceptable saturation also does not confirm adequate cardiac output or tissue perfusion.
WHEN A NORMAL NUMBER CAN MISLEAD
1. Supplemental oxygen can conceal hypoventilation
Clinical deterioration should outweigh a reassuring saturation when the two conflict. Respiratory rate, depth, consciousness and capnography may reveal deterioration that saturation alone cannot detect.
A patient has received opioid analgesia and is now drowsier than expected. They are receiving supplemental oxygen and have an SpO₂ of 98%. Their breathing is slow and shallow and consciousness is deteriorating.
The immediate problem is ventilation, not saturation.A patient is rescued from an enclosed-space fire. They report headache, nausea and dizziness. Their SpO₂ is 99%.
That reading does not exclude carbon monoxide poisoning.2. Carbon monoxide can produce false reassurance
Conventional two-wavelength pulse oximetry cannot reliably distinguish oxyhaemoglobin from carboxyhaemoglobin. A normal SpO₂ therefore cannot exclude carbon monoxide poisoning.1
3. Poor perfusion can weaken the signal
Cold extremities, vasoconstriction, shock, movement and probe position can degrade signal quality. Waveform quality, pulse correlation, perfusion, movement and consistency with the clinical picture should be assessed before accepting a reading.7
4. Accuracy is not equal across all skin tones
Pulse oximetry may overestimate arterial oxygen saturation in people with darker skin pigmentation, particularly during hypoxaemia. The evidence supports greater caution near decision thresholds rather than rejection of pulse oximetry.258
5. A single reading hides the trajectory
The isolated number is identical. The clinical meaning is not. A sudden fall of more than 3% should prompt fuller assessment even within a target range.6
WHEN ‘NORMAL’ BECOMES A DIAGNOSIS
Once a clinician sees an SpO₂ of 98%, it can become an anchor around which the rest of the assessment is interpreted. Yet a pulse oximeter does not assess work of breathing, recognise exhaustion, listen to air entry, measure carbon dioxide, identify anaemia or understand the mechanism of exposure.
CLINICAL MEANINGA normal saturation is one piece of reassuring evidence. It is not a respiratory diagnosis.
NORMAL FOR WHOM?
BTS guidance recommends 94–98% for many acutely unwell adults not at risk of hypercapnic respiratory failure, and generally 88–92% for those with recognised risk while awaiting blood-gas information.6 Population targets support decision-making but cannot replace baseline, trajectory, treatment context or clinical judgement.
TRUST THE READING—BUT KNOW WHAT YOU ARE TRUSTING IT TO TELL YOU
- 01Is the signal technically credible?
- 02Is the patient receiving supplemental oxygen?
- 03Could abnormal haemoglobin affect the reading?
- 04Does the reading fit the patient?
- 05What is the trend?
- 06What has SpO₂ not measured?
SpO₂ can tell you something important. It cannot tell you everything important.
BEYOND THE BRIEF
Confirm that the signal is credible. Look at the trend. Consider whether oxygen is masking hypoventilation. Ask whether the clinical problem is one that pulse oximetry can detect at all.
REFERENCES & FURTHER READING
- Barker SJ, Tremper KK. The effect of carbon monoxide inhalation on pulse oximetry. Anesthesiology. 1987;66(5):677-679.
- Fawzy A, et al. Racial and ethnic discrepancy in pulse oximetry. JAMA Internal Medicine. 2022;182(7):730-738.
- Fu ES, et al. Supplemental oxygen impairs detection of hypoventilation by pulse oximetry. Chest. 2004;126(5):1552-1558.
- Jubran A. Pulse oximetry. Critical Care. 2015;19:272.
- Martin D, et al. Effect of skin tone on pulse oximetry accuracy: a systematic review. British Journal of Anaesthesia. 2024;132(5):947-959.
- O'Driscoll BR, et al. BTS guideline for oxygen use in adults in healthcare and emergency settings. Thorax. 2017;72(Suppl 1):ii1-ii90.
- Poorzargar K, et al. Accuracy of pulse oximeters in poor peripheral perfusion: a systematic review. Journal of Clinical Monitoring and Computing. 2022;36(4):961-973.
- Sjoding MW, et al. Racial bias in pulse oximetry measurement. New England Journal of Medicine. 2020;383(25):2477-2478.
Educational disclaimer. The Prehospital Brief provides educational discussion of evidence relevant to prehospital practice. It does not replace local clinical guidelines, organisational policy, individual clinical judgement or appropriate senior clinical advice.