Normal AHI and Oxygen Saturation Parameters for Sleep Apnea
An AHI (Apnea-Hypopnea Index) of less than 5 events per hour is considered normal, while an RDI (Respiratory Disturbance Index) of less than 5 events per hour also indicates absence of obstructive sleep apnea. 1, 2
Apnea-Hypopnea Index (AHI) Thresholds
The diagnostic classification based on AHI is consistently defined across multiple guideline-referenced studies:
- Normal/No OSA: AHI <5 events/hour 1
- Mild OSA: AHI 5-14 events/hour 1
- Moderate OSA: AHI 15-29 events/hour 2
- Severe OSA: AHI ≥30 events/hour 2
Some studies use alternative thresholds with AHI ≥10 events/hour as the cutoff for OSA diagnosis, though the AHI ≥5 threshold is more widely accepted in current practice 1. The American Academy of Sleep Medicine specifically indicates that AHI values ≥15/hour in REM sleep are considered moderate to severe 2.
Oxygen Saturation Parameters with Overnight Oximetry
Key Oximetry Metrics
Oxygen Desaturation Index (ODI): Normal is defined as ODI <5 events/hour (where events are ≥3% or ≥4% desaturations depending on the scoring criteria used) 3, 4, 5.
Average Oxygen Saturation (AO₂): Studies show mean values of approximately 92.7% ± 5.6% in mixed populations, with values >90% generally considered within normal range 6, 5.
Lowest Oxygen Saturation (LO₂): Mean values around 68.5% ± 19.3% are seen in OSA populations, though this metric shows high hour-to-hour variability 6, 5.
Time with SpO₂ <90% (T<90%): This represents the percentage of total sleep time with oxygen saturation below 90%. Normal individuals should have minimal time <90%, while OSA patients average 15.7% ± 24.2% of sleep time 6, 5.
Important Clinical Considerations
Hypopnea Scoring Variability
The definition of hypopnea significantly impacts AHI calculation. The American Academy of Sleep Medicine defines hypopnea as a ≥30% reduction in airflow for ≥10 seconds associated with either ≥3% oxygen desaturation or an arousal 7. However, Medicare guidelines require ≥4% desaturation for hypopnea scoring, which can substantially alter the AHI 8.
Critical pitfall: The choice of oximeter and its signal averaging time (typically 4-6 seconds) can affect whether patients meet diagnostic thresholds. One study found that oximeter choice caused 7 of 35 patients near the AHI threshold of 15 to be reclassified 8.
Best Oximetry Parameter for Clinical Use
T<90% (percent time with SpO₂ <90%) demonstrates the least hour-to-hour variability (Cronbach alpha coefficient 0.98) and may be the most reliable single metric for expressing oxygen desaturation severity 6. However, comprehensive assessment requires multiple parameters including lowest oxygen saturation, ODI, and T<90% 5.
Oximetry Limitations
Overnight pulse oximetry with normal ODI (<5 events/hour) does not exclude OSA. Visual correlation between episodic increases in heart rate variability and minimal oxygen desaturations on "normal" oximetry is an independent predictor of OSA (odds ratio 5.6) and should prompt polysomnography 3.
Oxygen saturation combined with airflow monitoring shows good agreement with polysomnography for AHI in non-OSA and mild OSA patients, but tends to underestimate AHI in moderate to severe cases 4.
REM-Specific Considerations
The global AHI may underestimate disease severity when OSA occurs predominantly during REM sleep. Patients may have a global AHI of 8/hour (below treatment threshold) but an AHI in REM of 29.1/hour (severe OSA requiring treatment) 2.