Large Leak

Last night my ahi was 1.66 which I am very happy with. However sleepyhead shows I had 45% large leak. How is this possible? I would have expected the ahi to be through the roof with such a leak.

Need more information please before we can really comment.

First thing is...what machine are you using? Need to know that because different brands report leaks differently.

Second is ....just how bad (how deep into large leak territory did you go) exactly were the leaks? Barely into large leak territory or way deep into large leak territory???

We worry about leaks affecting therapy because we don't know if that nice low AHI is because nothing happened or maybe the leaks were so bad the machine simply couldn't sense the apnea events happening.

Since you posted a 45% of the time in large leak statistic ...I assume using OSCAR so how about posting the detailed report?
This format please
viewtopic/t158560/How-to-post-images-for-review.html

Thanks for the responses. The machine is a system one and I'me afraid the rest of your reply has gone way over my head.

speckledhen wrote: ↑

Sun Nov 14, 2021 11:30 am

Thanks for the responses. The machine is a system one and I'me afraid the rest of your reply has gone way over my head.

Respironics machines report large leak differently than the ResMed machines.
OSCAR defaults to the ResMed large leak threshold number for calculations of the percentage of time spent in large leak.
Large leak territory is totally different on a Respironics machine so that 45% statistic may not be an accurate reflection of time spent in large leak.

Look up on the actual EVENTS graph....see the LL line??? Any LL flagged areas up on the Events graph??
That's a machine flagged event and not OSCAR calculation....Machine data point trumps OSCAR generated statistic.

If you would post a screen shot of the detailed report we could tell you whether your large leak statistic of 45% means you really have a problem or not.
Without visually seeing exactly what is going on with the leak it's really impossible to give you more than an educated guess but most likely you aren't having nearly as much time deep in large leak territory for there to be much of a chance for the machine to be clueless and missing flagging events.

If you aren't seeing a lot of LL flagged time on the Events graph...I would either ignore the % of time spent in large leak statistic...or turn it off...or reset the red line threshold to something more accurate for a Respironics machine.

op states she is using sleepyhead. i know, same-same, but just so they don't get confused.