The Apnea-Hypopnea Index (AHI) is a measure of severity in obstructive sleep apnea. It is calculated as the number of times per hour that the airway partially or completely collapses, associated with a drop in oxygen levels. Partial collapse is labeled a hypopnea, whereas complete collapse is labeled an apnea. CPAP treatment uses air pressure delivered through a mask to stent open the airway. If there continues to be partial or complete collapse despite CPAP, the basic idea, in theory, is that the pressure is inadequate.
The major caveat:
Where are you getting this AHI? Most likely, you’re reading it off a PAP machine. The upshot of this answer is that, as much as we all love numbers, such readings should not be trusted.
The AHI was initially validated in the setting of overnight polysomnography, i.e. a sleep lab. There a technician watches the tracings generated from a nasal pressure transducer, chest/abdomen effort belts, and continuous pulse oximetry. There are specific criteria for scoring apneas and hypopneas. The recommended criteria from the American Academy of Sleep Medicine for a hypopnea are now:
a. The peak signal excursions drop by ≥30% of pre-event baseline using nasal pressure...
b. The duration of the ≥30% drop in signal excursion is ≥10 seconds.
c. There is a ≥3% oxygen desaturation from pre-event baseline and/or the event is associated with an arousal.
There has been a lot of controversy in recent years as to whether that number in the c criterion should be 3% or 4% drop in pulse oximetry. Why should we care?
The number you’re getting off your PAP machine has no pulse oximetry to correlate with. They must therefore be using some other non-standard metric. How do they calculate this? This, unfortunately, is proprietary information known only to the PAP manufacturers. Occasionally there are validation studies of varying quality published, but there is no uniformity in the industry as to what algorithm is used, and there is very little transparency about where these numbers are coming from.
What, then?
Monitoring of sleep apnea treatment is symptom-based. If symptoms improve, good enough, keep with it. If there is inadequate improvement or recurrence of symptoms after improvement, things should be re-evaluated. Generally this involves one or more of the following:
assessing the interface (Is the mask leaking? That will cause inadequate pressure delivered to the airway.)
home overnight pulse oximetry
adjustment of the settings on an auto-titrating machine to allow for higher pressures
empirically increasing the pressure on a fixed-pressure CPAP
a repeat sleep study while using PAP, adjusting the pressure settings (i.e. a titration study)
Conclusion: Most likely, in the next few years these technologies will get better, and hopefully transparency and uniformity will improve. In the meantime, take the AHI number on your CPAP machine with a grain of salt.
Kryger, Roth, Dement. Principles and Practice of Sleep Medicine. Chapter 107: Positive Airway, Pressure Treatment for Obstructive Sleep Apnea-Hypopnea Syndrome. pp 1219-1232.
Epstein, LJE, et al. Clinical guideline for the evaluation, management and long-term care of obstructive sleep apnea in adults.. J Clin Sleep Med. 2009 Jun 15;5(3):263-76.
Caveat lector: In addition to the afore mentioned references, this answer contains my own impressions gleaned from training in the area.