Oxygen Desaturation & Sleep Apnea

Oxygen saturation, often written as SpO₂, is the percentage of hemoglobin in the blood that is carrying oxygen. A value of 95% means 95% of the measured hemoglobin is oxygenated. Normal awake saturation is typically 95–100%.

During sleep, brief dips in oxygen saturation are common and not always concerning. However, frequent, deep, or prolonged drops — especially those that coincide with breathing pauses — may be associated with obstructive sleep apnea. The pattern and frequency of these drops, not a single reading, are what matter clinically.

What this means: Oxygen saturation is a key data point on a sleep study report, but a single number does not tell the whole story.

Oxygen nadir is the lowest oxygen saturation level recorded during the entire sleep study. It describes the single deepest dip. A nadir of 82%, for example, means oxygen dropped to at least that low at one point during the night.

Nadir provides a sense of the worst oxygen event, but it does not capture how often drops occurred or how long they lasted. A person could have a single deep dip and otherwise normal oxygen, or repeated deep dips throughout the night — the nadir alone cannot distinguish these patterns.

The Oxygen Desaturation Index (ODI) is the average number of times per hour that oxygen saturation drops by a defined amount from its baseline. Common definitions count a drop of 3% or 4% as a desaturation event.

ODI is a rate, similar in structure to AHI, but it counts oxygen dips rather than breathing events. ODI and AHI are related — most hypopneas cause an oxygen drop — but they are not identical. Some breathing events (particularly those that cause arousals without significant oxygen change) may not appear in the ODI, and some oxygen dips may not meet the full criteria for a scored hypopnea.

Because different reports may use different drop thresholds (3% vs. 4%), the ODI value can vary even for similar raw data. This is a consequence of the scoring definition, not an error.

Some sleep study reports include the total time (or percentage of sleep time) that oxygen saturation remained below a defined threshold, such as 90% or 88%. This measure captures the cumulative burden of low oxygen, not just individual dips.

For example, a report might state that oxygen was below 90% for 12 minutes, or for 4% of total sleep time. This measure adds context that a rate like ODI does not — it shows how sustained the low oxygen periods were, not just how often they occurred.

Different studies and laboratories may use different thresholds, so the specific number depends on the definition applied.

Beyond the summary numbers, the depth and duration of individual oxygen dips matter. A brief, shallow dip to 92% is different from a prolonged drop to 78%, even if both count as a single desaturation event in some scoring systems.

The pattern of dips also matters — whether they are isolated or clustered, whether they occur in specific sleep stages or positions, and how quickly oxygen recovers after each event. These details are part of why a sleep study report requires professional interpretation rather than self-reading.

Several factors can cause oxygen numbers to differ between reports:

• Desaturation threshold: A 3% or 4% drop may be used to define an event, changing the ODI.
• Threshold for time-below: Reports may use 90%, 88%, or other cutoffs.
• Device sensitivity: In-lab oximetry and home test sensors may differ in accuracy and sampling rate.
• Night-to-night variation: Oxygen patterns can vary from one night to the next.

These differences are expected and are part of why results are interpreted in context rather than as standalone numbers.

Oxygen data does not stand alone. A healthcare provider interprets oxygen measures alongside:

• The AHI or related index
• Symptoms and daytime functioning
• Medical history, including cardiovascular and lung conditions
• Event type (obstructive vs. central)
• Position and sleep stage effects

There is no universal oxygen number that determines severity or treatment for every person. What is significant depends on the individual's health, the pattern of oxygen changes, and the overall clinical picture. For more on how oxygen relates to health consequences, see sleep apnea and low oxygen.

Consumer finger pulse oximeters and wearable devices can show general oxygen trends. They may be useful for noticing patterns worth discussing with a provider. However, they have important limitations:

• They are not diagnostic tools and cannot diagnose sleep apnea.
• They may have accuracy limitations, especially at lower saturation ranges or with movement.
• They do not measure airflow, breathing effort, or sleep stages.
• They cannot determine a 'safe' oxygen level for an individual.

If you notice concerning oxygen trends on a consumer device, or if you have symptoms of sleep apnea, a professional sleep evaluation is the appropriate next step. Learn more in our sleep testing guide.