AHI 3% vs. AHI 4%

[SleepingStormtrooper]

I recently had a new sleep study done to try out an ASV. Does someone understand the significance of AHI 3% versus AHI 4%? These two numbers from the study were wildly different: AHI3% = 7.83/hr, while AHI4% = 0.57/hr.

The clinical meaning of these numbers was helpfully spelt out in the study report:
*AHI 3% "is defined as the number of hypopneas, with either 3% oxygen desaturation or an arousal, plus the number of apneas per hour of sleep."
*AHI 4% "is defined as the number of hypopneas associated with greater than or equal to 4% SpO2 desaturations plus the number of apneas per hour of sleep."

Is one measure more standard? Which one is a better comparison to the value estimated by my CPAP? (Given that they just use your breathing data and don't account for oxygen saturation.)

[palerider]

I recently had a new sleep study done to try out an ASV. Does someone understand the significance of AHI 3% versus AHI 4%? These two numbers from the study were wildly different: AHI3% = 7.83/hr, while AHI4% = 0.57/hr.

The clinical meaning of these numbers was helpfully spelt out in the study report:
*AHI 3% "is defined as the number of hypopneas, with either 3% oxygen desaturation or an arousal, plus the number of apneas per hour of sleep."
*AHI 4% "is defined as the number of hypopneas associated with greater than or equal to 4% SpO2 desaturations plus the number of apneas per hour of sleep."

Is one measure more standard? Which one is a better comparison to the value estimated by my CPAP? (Given that they just use your breathing data and don't account for oxygen saturation.)

Neither of those numbers/definitions is something I've ever heard of... they're not standard.

The way they're being used is pretty well spelled out in what you pasted up there... but I'll try to rephrase.

When you have apnea, you may end up not breathing long enough for your oxygen levels to drop. (not always, but it frequently happens).

AHI is the number of *events* per hour, on average. But it doesn't tell you how long those events were... so you could be having a fair number of them, (7.83) per hour, but they're short ones, and so your oxygen didn't drop much... but at some point, you had some longer ones, but not as many... at that point your ahi was only 0.57, but because they were longer, your oxygen dropped more.

However, other than someone that just loves numbers, none of that is really very significant to anything, and I'd just ignore it

[SleepingStormtrooper]

Thanks! Good point about how long the events are. My sense is that both measures are commonly used by sleep labs, as I also saw them reported on an earlier sleep study by a different lab. I would love to know which is the better comparison with the value from my CPAP. One number would represent a drastic improvement, while the other would mean a possible minor improvement.

[palerider]

Thanks! Good point about how long the events are. My sense is that both measures are commonly used by sleep labs, as I also saw them reported on an earlier sleep study by a different lab. I would love to know which is the better comparison with the value from my CPAP. One number would represent a drastic improvement, while the other would mean a possible minor improvement.

Neither mean anything at all, and have no correlation at all to your cpap reports, since your cpap doesn't measure oxygen desaturation.

the 4% number is worse, even though the AHI is a lower number, because your organs and brain are more deprived of oxygen.

[SleepingStormtrooper]

No, sorry, I understand what these definitions mean, and that a CPAP of course doesn't measure O2 levels. But given a CPAP is reporting an "AHI", I would think the machines are at least trying to correlate with one of these (more accurate?) lab definitions. I mean otherwise, how would a doctor decide whether a patient appears better on an bipap, asv, etc. over what they're using currently?

[Pugsy]

The 3 % vs 4 % pertains to the criteria used to meet sleep study standards used for diagnosis
http://adventures-in-hosehead-land.blog ... -test.html
Once the diagnosis is established it's time to move on to treatment and the goal is no more desats due to sleep disordered breathing.

[jnk...]

They are choosing to specify what AHI "is" according to two differing scoring criteria for hypopneas--a situation that is a pain in the backside of sleep labs/centers.

AASM continues to push for 3%/arousal-hypopnea-based AHI, but unfortunately, many payers will still only accept the 4%-hypopnea-based AHI for paying for someone to be allowed to try treatment.

The AASM continues to recommend scoring hypopneas in adults when there is a ≥ 3% oxygen desaturation from pre-event baseline and/or the event is associated with an arousal. However, it is acceptable for accredited sleep centers to score hypopneas in adults when there is a ≥ 4% oxygen desaturation from pre-event baseline.

AASM sleep centers must specify in the PSG report whether hypopneas were scored using the recommended rule 1A or the acceptable rule 1B:

Recommended
1A. Score a respiratory event as a hypopnea if ALL of the following criteria are met:

a. The peak signal excursions drop by ≥30% of pre-event baseline using nasal pressure (diagnostic study), PAP device flow (titration study), or an alternative hypopnea sensor (diagnostic study).

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.

OR

Acceptable
1B. Score a respiratory event as a hypopnea if ALL of the following criteria are met:

a. The peak signal excursions drop by ≥30% of pre-event baseline using nasal pressure (diagnostic study), PAP device flow (titration study), or an alternative hypopnea sensor (diagnostic study).
b. The duration of the ≥30% drop in signal excursion is ≥10 seconds.
c. There is a ≥ 4% oxygen desaturation from pre-event baseline.

Please note that the criterion involving arousals is included in 1A and excluded from 1B.

Color added by me.

https://aasm.org/aasm-clarifies-hypopne ... -criteria/

But home-machine-reported "hypopneas" are guesstimates based on nothing more than flow, so they have little-to-nothing to do with the real hypopneas of either definition above.

Home machine data is for trending--not for diagnosing or for exact assessment of treatment in direct relation to PSG data.

[SleepingStormtrooper]

Interesting... I guess I was thinking of the CPAP AHI as just a less accurate estimate of what they measure in the lab. I didn't realize how disconnected the two scores may be. Thanks for the clarifications everyone!

Incidentally, I find it interesting that in the recommended/3% definition, arousals can be used to define hypopneas, but not in the other definition. Makes me wonder whether arousals were the thing driving the difference between my two scores, or if it was just that I was often hitting 3% desaturation but not quite reaching 4% desaturation.

[palerider]

No, sorry, I understand what these definitions mean, and that a CPAP of course doesn't measure O2 levels. But given a CPAP is reporting an "AHI", I would think the machines are at least trying to correlate with one of these (more accurate?) lab definitions. I mean otherwise, how would a doctor decide whether a patient appears better on an bipap, asv, etc. over what they're using currently?

No, I'm sorry, they don't.. The numbers are nothing more than statistics that about what happened during your sleep study.

(why is it people ask a question, then argue vehemently when the answer isn't what they want to hear?)

Cpaps measure your AHI purely based on breathing disturbances... did you stop breathing? *apnea*... did your respiratory flow volume drop to 50% or so? hypopnea... It has nothing to do with your O2, so sleep study numbers that are tied to O2.... *are irrelevant*.

That's like trying to glean some meaning out of your AHI while you're on your back, or side, or stomach... beyond "hey, I should sleep on my side". (ok, that's not a very good example.. but, meh)

[SleepingStormtrooper]

@palerider, I appreciate the responses. I wasn't arguing, just trying to make sure the question I was asking was getting addressed. The situation you and jnk... describe is, well, confusing (and I don't doubt now that it's the case). There are 2 lab numbers containing the label "AHI", which apparently can give wildly different values. And a CPAP also gives you an "AHI", yet it sounds like that number shouldn't be compared for any purpose with the other two. That's not great labeling. It would be natural to want to compare the AHI one is getting each night to the (other) AHI reported from the lab test, as one way to see if the treatment is helping.

[jnk...]

. . . It would be natural to want to compare the AHI one is getting each night to the (other) AHI reported from the lab test, as one way to see if the treatment is helping.

Natural, though not strictly practical for correlation purposes, in view of the lack of O2 info and arousal info.

While the nomenclature of apnea hypopnea index is same for both PAP device AHI and the polysomnography (PSG) AHI, the definitions vary significantly. The AHI determined during a PSG is based on standardized definitions recommended by the American Academy of Sleep Medicine, while the AHI reported by the PAP device, usually with the “smart card” from the machine, is based on a non-standard proprietary definition of the device manufacturer.--https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5381855/

But the bottom line, as is often noted on this board, is that the key to it all is optimal titration, in which case . . .

Our results suggest that in a majority of patients who were optimally titrated in the sleep laboratory, the PAP device continued to show optimal control at home.--Same study, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5381855/

Experience on this board has shown that the trending data from home machines are very valuable for fine-tuning pressure choice/range, regardless of the accuracy of the one-night lab/center titration that mostly just finds a single, ballpark starting pressure. That is, in my opinion, the much more important issue, not so much the fact that proprietary "hypopnea" definitions used by home-treatment-machine manufactures differ from PSG and among the manufacturers themselves.

[palerider]

There are 2 lab numbers containing the label "AHI", which apparently can give wildly different values. And a CPAP also gives you an "AHI", yet it sounds like that number shouldn't be compared for any purpose with the other two.

That *IS* what i said in the first response.

However, other than someone that just loves numbers, none of that is really very significant to anything, and I'd just ignore it

That's not great labeling. It would be natural to want to compare the AHI one is getting each night to the (other) AHI reported from the lab test, as one way to see if the treatment is helping.

If you say so... Sounds to me like you just can't let go of the idea.

Better idea, forget the lab results, focus on what you have *now*, and improve that.

[jnk...]

focus on what you have *now*, and improve that.

As Einstein never said about his AHI: "Get it as low as possible, but not lower."