Sleep Strips

[StillAnotherGuest]

Here's an interesting article:

Otolaryngol Head Neck Surg. 2006 Aug ;135 (2):265-8
A comparison of polysomnography and the SleepStrip in the diagnosis of OSA.

Kenny P Pang, Thomas A Dillard, Amy R Blanchard, Christine G Gourin, Robert Podolsky, David J Terris

OBJECTIVE: To investigate the role of a portable screening device (SleepStrip) in the diagnosis of obstructive sleep apnea (OSA).

METHODS AND MATERIALS: Prospective, nonrandomized double-blinded single cohort study at an academic health center. Patients with suspected OSA scheduled for an attended overnight Level I polysomnogram (PSG) and who consented to participate in the study wore the SleepStrip device at home the night after the PSG. The apnea-hypopnea index (AHI) determined by PSG was compared with the results of the SleepStrip recording.

RESULTS: Thirty-seven patients with a mean age of 52.1 +/- 12.2 years and mean body mass index of 35.7 +/- 5.2 participated in the study. The overall agreement between the AHI and the SleepStrip results using Cohen's Kappa value was 0.139 (P = 0.19). The sensitivity and specificity of the SleepStrip for diagnosing severe OSA when the AHI was >40 were 33.3% and 95% (P = 0.05). When the AHI was >25, the SleepStrip sensitivity and specificity were 43.8% and 81.3% (P = 0.26). The sensitivity and specificity of the SleepStrip for diagnosing OSA in patients with an AHI >15 were 54.6% and 70%, respectively (P = 0.26).

CONCLUSION: The SleepStrip has a low correlation with the AHI as measured by PSG. Further studies are needed before this device can be recommended as a screening tool for the diagnosis of OSA. EBM rating: B-2b.

[-SWS]

The sensitivity and specificity of the SleepStrip for diagnosing severe OSA when the AHI was >40 were 33.3% and 95% (P = 0.05). When the AHI was >25, the SleepStrip sensitivity and specificity were 43.8% and 81.3% (P = 0.26). The sensitivity and specificity of the SleepStrip for diagnosing OSA in patients with an AHI >15 were 54.6% and 70%, respectively (P = 0.26).

Of course, we'll want to see more independent studies as they come out. But I'm not surprised to see those sensitivity and specificity scores. I had predicted that type of disappointing sensitivity/specificity relationship when we first discussed sleep strips in the original thread:
viewtopic.php?t=14578&p=122896#122896

Very clearly, the sleep strip cannot serve as a standalone diagnostic method. However, I think the manufacturers portray these devices as "screening" tools rather than comprehensive "diagnostic" tools. And in my own opinion the sleep strips are not adequate as standalone screening devices either, based on those predictably lacking sensitivity/specificity scores.

If the sleep strips are to serve any useful screening or diagnostic role whatsoever, they probably need to serve as but a single element in an array of diagnostic/screening procedures. The rhetorical question for such a multi-component screening or diagnostic methodology would then become: what is the resulting system's final or overall correlation to the various severity gradients of SDB? To the various atypical presentations? To the SDB patient population as a whole? Undoubtedly there are plenty of outstanding challenges for researchers, developers, and practicing clinicians in sleep science.

There is still a drastic need to devise much better SDB screening in the field, in my opinion.

My other observation is that manufacturers so often seem to obscure key specifications, such as their own internally-derived sensitivity/specificity scores. SAG, is that last statement of mine off base? Or are the sales reps much more forthcoming with key technical specs than I suspect?

[StillAnotherGuest]

My other observation is that manufacturers so often seem to obscure key specifications, such as their own internally-derived sensitivity/specificity scores. SAG, is that last statement of mine off base? Or are the sales reps much more forthcoming with key technical specs than I suspect?

I suppose that one should pretty much assume that all the manufacturers put their best foot froward and hide their shortcomings. Generally not a big deal, really. I mean, as long as we remain vigilant to the limitations of devices like these, we can make informed decisions.

Speaking of which, in looking back at the posts regarding this device, I don't see where anyone pointed out that this is thermistor technology. With all the data touting pressure transducers over thermistors as being far superior in accuracy, this whole device is one huge step backwards. A single channel thermistor for a screener? Why bother? What does this thing offer that a good clinical eval doesn't? Other than the very strong possibility of a false negative?
SAG

[Snoredog]

Here's an interesting article:

Otolaryngol Head Neck Surg. 2006 Aug ;135 (2):265-8
A comparison of polysomnography and the SleepStrip in the diagnosis of OSA.

Kenny P Pang, Thomas A Dillard, Amy R Blanchard, Christine G Gourin, Robert Podolsky, David J Terris

OBJECTIVE: To investigate the role of a portable screening device (SleepStrip) in the diagnosis of obstructive sleep apnea (OSA).

METHODS AND MATERIALS: Prospective, nonrandomized double-blinded single cohort study at an academic health center. Patients with suspected OSA scheduled for an attended overnight Level I polysomnogram (PSG) and who consented to participate in the study wore the SleepStrip device at home the night after the PSG. The apnea-hypopnea index (AHI) determined by PSG was compared with the results of the SleepStrip recording.

RESULTS: Thirty-seven patients with a mean age of 52.1 +/- 12.2 years and mean body mass index of 35.7 +/- 5.2 participated in the study. The overall agreement between the AHI and the SleepStrip results using Cohen's Kappa value was 0.139 (P = 0.19). The sensitivity and specificity of the SleepStrip for diagnosing severe OSA when the AHI was >40 were 33.3% and 95% (P = 0.05). When the AHI was >25, the SleepStrip sensitivity and specificity were 43.8% and 81.3% (P = 0.26). The sensitivity and specificity of the SleepStrip for diagnosing OSA in patients with an AHI >15 were 54.6% and 70%, respectively (P = 0.26).

CONCLUSION: The SleepStrip has a low correlation with the AHI as measured by PSG. Further studies are needed before this device can be recommended as a screening tool for the diagnosis of OSA. EBM rating: B-2b.

[-SWS]

I suppose that one should pretty much assume that all the manufacturers put their best foot froward and hide their shortcomings. Generally not a big deal, really. I mean, as long as we remain vigilant to the limitations of devices like these, we can make informed decisions.

Thanks for the comment, SAG. Well, my hunch is that statement is especially true for you, and somewhat less true for other lab managers. And in a different vein of thought or comparison altogether, I kind of guess that statement will be more true for routine PSG test-suite architecting and somewhat less true for devising not-so-routine test suites.

Speaking of which, in looking back at the posts regarding this device, I don't see where anyone pointed out that this is thermistor technology. With all the data touting pressure transducers over thermistors as being far superior in accuracy...

For sure. I personally don't think the SleepStrip is the last or best screening technology you'll see. In fact, later on in that same thread you'll see an alternative implementation by Respironics that gets away from using thermistor technology altogether.

But even that single-channel Respironics offering is probably not the best and final screening device that we will see in the near future. Three very easy data channels to put together on one ultraportable and easy-to-wear screening device will probably include SpO2, snore, and pressure derived flow sensors. Overall screening sensitivity/specificity should improve compared to any one data channel. Again, in my opinion screening devices are not substitutes for comprehensive PSG diagnosis. The ideal is that they would get the right patients to the sleep labs more often than traditional clinical evaluations, which sadly aren't happening with adequate frequency or even of an adequate methodology.

A single channel thermistor for a screener? Why bother? What does this thing offer that a good clinical eval doesn't? Other than the very strong possibility of a false negative?
SAG

Well, take a peak at that pulse ox my cousin used. It too yielded a strong possibility of a false negative. And yet that pulse ox was the only thing that got him into the sleep lab for a much needed PSG. Where was that good old clinical eval? It was being performed as a lackluster clinical eval that fell even short of pulse oximtery. That easy-to-use pulse ox would have also sent me to the PSG more than a decade earlier than I finally went.

So sure, false negatives in screening tools mean you're missing quite a few "live ones". But taking a pass altogether means you're essentially missing everyone---at least all those who fail to acquire a good albeit elusive "clinical eval".

What an interesting challenge this simple feat of real-world screening can be, SAG. Right about now my last-place screening preference would be to have one of those staus quo clinical evals that kept missing me, my cousin, and everybody on the block for years. My next vote regarding preference might be a standalone screening device such as pulse ox or sleep strip (with yet better results than the clinical screening eval that falls far too short way too often). Lastly, I would prefer a multi-component screening suite that is easy to wear, ultra portable, and yields even higher correlative test scores than any single device. I truly think that's where the industry may be heading.

[StillAnotherGuest]

This sleep strip cannot possibly detect central or mixed apnea events which is why I feel everyone who suspects they may have SDB be screened at least once with a PSG if not to rule that out.

I know you meant to say "differentiate" instead of "detect".

What an interesting challenge this simple feat of real-world screening can be, SAG. Right about now my last-place screening preference would be to have one of those staus quo clinical evals that kept missing me, my cousin, and everybody on the block for years. My next vote regarding preference might be a standalone screening device such as pulse ox or sleep strip (with yet better results than the clinical screening eval that falls far too short way too often). Lastly, I would prefer a multi-component screening suite that is easy to wear, ultra portable, and yields even higher correlative test scores than any single device. I truly think that's where the industry may be heading.

I. too, have opinions on where the industry is heading. But it basically boils down to this:

A hammer is only as good as the guy swingin' it.

There's been sufficient data on at least the OSA part of sleep disorders out there for at least 13 years. There's so much out there now that it's like trying to keep your shoes clean while walking behind a cattle drive. For those of you who have a far better working knowledge of OSA than your medical guy, maybe what you should really consider is whether or not any of their tools are sharp. Cause what you're really saying is "My guy is cutting-edge on everything in the medical world, except whenever he gets to anything about sleep he tosses it."

Sleep-strip is appealing to those who are looking for a short cut because it reduces decision-making down to darn near nothing. At least that is what they want you to believe. Unfortunately. these screening devices end up in the hands of people who have no knowledge or understanding of what they're looking at or what the limitations are, even those devices that have potential.

Right, if you use oximetry in your screener, the chances of getting useful data become better. The University of British Columbia Hospital Sleep Disorders Program in their article in AoIM Diagnosis and Initial Management of Obstructive Sleep Apnea without Polysomnography was picking out over 95% of the moderate to severe OSA. But they weren't "diagnosing sleep disorders", they were "picking out moderate to severe OSA". If your RDI is <15 or you have co-morbidity, there will be no short cut. And again, in that study, in order to find 81 subjects they had to look at 2216 patients referred to their sleep center (not 2216 people on the street, 2216 people referred to their center, which means they all had a sleep issue) of which 2135 were ineligible for a myriad of reasons.

The study we really need to be doing is an all-factors analysis. And until that time, if it's the Accredited Sleep Centers with credentialed personnel that are putting up good outcomes numbers for screening and/or portable testing, then great, have Accredited Sleep Centers with credentialed personnel perform screening and/or portable testing.

Because that's a far cry from mail-order Sleep-strips.
SAG

[NightHawkeye]

First, my apologies for butting in here, however, I have one simple question which has been bugging me for quite some time now, and to which no one else has provided an adequate answer.

My next vote regarding preference might be a standalone screening device such as pulse ox or sleep strip (with yet better results than the clinical screening eval that falls far too short way too often). Lastly, I would prefer a multi-component screening suite that is easy to wear, ultra portable, and yields even higher correlative test scores than any single device.

-SWS, I agree totally that a multi-component screening device would be better than a single screening device such as an oximeter. However, I also read with great interest some time back your posting about how an oximeter saved the day for youself and for your cousin. Likewise, for me an oximeter allowed the diagnosis finally of apnea. Also, I judge from postings on this forum that the vast majority of apnea sufferers routinely experience many episodes of oxygen desaturation. Indeed, Medicare rules even require oxygen desaturation for declaration of hypopnea.

Since oxygen desaturation can be determined inexpensively using existing diagnostic equipment (i.e., an oximeter), and since the vast majority of apnea sufferers apparently present with significant oxygen desaturations, why couldn't an oximeter be routinely used as an inexpensive in-home screening device for sleep apnea? I daresay that, for a variety of reasons - including cost and convenience (as was evidenced in your case and also with your cousin), an oximeter screening would allow far more people to be diagnosed with sleep apnea than could ever hope to be diagnosed using PSG's. Diagnosis of nearly the entire population with oximeters could also occur quite quickly using oximeters - far more quickly and thoroughly than one could ever hope for with an overnight PSG in a sleep lab.

The advantages of using an oximeter as the first line diagnostic tool, rather than a PSG, seemingly far outweigh the disadvantages. The only real disadvantage I see is that a few folks would still need to go for a PSG to identify other potential sleep disturbances. I also tend to think that sleep lab business would actually pick up because of the far greater market penetration possible with oximeters and the resulting increased interest in sleep disturbances.

Other benefits seem likely to accrue from using an oximeter in this manner, but at this point I'll simply pose the question: Why not?

Regards,
Bill ( . . . again, my sincere apologies if I've failed to expunge biases and such from this post. My question is quite sincere.)

[-SWS]

-SWS, I agree totally that a multi-component screening device would be better than a single screening device such as an oximeter.

Well... of course, that ultra-portable easy-to-wear multi-component screening device just doesn't exist yet. Embedded snore detection and pressure-flow analysis would theoretically improve sensitivity compared to standalone pulse oximetry. But that theoretical sensitivity improvement would be at the expense of potentially skewing specificity (more false positives). For comprehensive SDB diagnosis, that potential specificity loss can be quite a technical caveat. Yet for screening toward the objective of getting patients into the PSG, achieving heightened sensitivity at the expense of marginal specificity reduction can be technically acceptable IMO.

However, I also read with great interest some time back your posting about how an oximeter saved the day for youself and for your cousin. Likewise, for me an oximeter allowed the diagnosis finally of apnea.

Bill, as far as my cousin is probably concerned, that idiomatic expression of yours just doesn't do my pulse oximeter justice! I say that because that pulse oximeter might have actually saved his life. BTW, I also referred to that incident about my cousin in my second post of this thread. If anyone is interested in following that story about my cousin's apnea-screening disaster, you can simply click the hyperlink I presented in my second post of this thread.

This particular thread has all the earmarks of a great discussion IMO. There are already several separate issues kicking around in this thread that underscore several key dilemmas in today's sleep industry. So in my next post I'm going to at least attempt to highlight what I personally believe are several separate yet crucial issues being discussed here. And I happen to believe that some of those worthy objectives may actually be diametrically opposed to yet other worthy objectives being discussed!

And the dilemma-ridden issues that are being discussed in this thread represent but a few "growth pains" in that relatively young branch of medicine known as sleep science. At least that is my take. None the less, I think this is already a very interesting thread!

[BrianRT]

Actually, the ultra-portable easy-to-wear multi-component screening device is already here. The Respironics Stardust II is a 7-channel recorder that the patient wears at home. It's self-contained so you can still be fully ambulatory too. It measures Sp02, Heart Rate, Airflow (with a nasal sensor that looks much like a standard nasal cannula) Chest Expansion with a velcro thermistor chest strap, patient position (tells if you're supine or non-supine during events), a snore indicator, and a Patient Event Marker that is pushed if you have to get out of bed (so we know you're up moving around if the results look pretty crazy). There is even a way to hook it into certain Auto-Cpaps for home titration too!

Already, there are a few insurance comp. (certain John Deere plans I know of) that will accept it IN LIEU of a first night study. At that point, they agree to pay for an Auto and pretty much consider that the '2nd night titration' also.

It is certainly much cheaper than a 2 night lab visit, esp. considering we do the Stardust screen for free at my DME. I personally feel that this is where the future is heading with OSA diagnostics and PAP titration.

[-SWS]

Actually, the ultra-portable easy-to-wear multi-component screening device is already here. The Respironics Stardust II is a 7-channel recorder that the patient wears at home.

Thanks, BrianRT!! Actually that Stardust II is a bit more cumbersome to wear and use than I had in mind. Not to say that the Stardust II has no merit. But that Stardust II aims for home diagnosis rather than screening. And that is but one of several key issues looming in this thread. And those key issues actually blend into each other, as in the case of the Stardust II. That's why I'm still going to at least attempt to discretely outline or list some of the greater issues that are coming into play in this discussion.

But while that Stardust II is admittedly a tad more cumbersome than I had in mind, its diagnostic objectives are loftier as well. Rather, I personally had in mind an even more convenient physical form factor comprised of: 1) something that sits on the face and is as sensory-unobtrusive as the SleepStrip, coupled with 2) a very small recording-pulse-oximeter sized box that travels well. Something relatively inexpensive that can be sent home unattended, albeit for SDB screening purposes rather than comprehensive diagnostic purposes. Inexpensive, nearly effortless, very small, and nearly doctor-office-ubiquitous is what I had in mind.

SDB screening is probably horrendously amiss down at the general practitioner's office right now! Those good doctors are essentially screening for every common disorder known to mankind. They're not screening only for apnea, and their not screening only for sleep disorders. SAG draws a rather keen metaphor of sleep-disorder screening and diagnosis being analogous to the skill-based practice of swinging a hammer. He thus points out that the hammer's aim, and final result, is only as good as the practitioner swinging that hammer. Engineers tend to envision seeing more pneumatic nailers in use.

Evolved technical implementations are always much better than initial technical implementations. Stay tuned in the field of sleep science and you'll see technical implementations getting better and better. In the interim, you're absolutely guaranteed to see some technical disappointments. I agree with you, Brian. I also think the future is heading toward more and more home-based sleep diagnostics---albeit with plenty of technical disappointments and medical-practice caveats along the way. Those are just some of the "growing pains" that I personally think are in store for sleep medicine.

[BrianRT]

I agree that it could be a bit cumbersome, but I guess it's a trade-off for all of the info it can generate. We use it as a screen to see if a full Polysom is indicated. I know that certain insurance plans consider it diagnostic, so it's all in what you do with it. One thing I like about that part is that with those insurance guidelines (again, only a few John Deere's that I know of) allow you to completely circumvent a sleep lab. Very cost effective compared to 2 nights at a lab, but I digress.

Technically, you could do your own 2-channel at home test with an airflow sensor (such as the Respironics RUSleeping (rusleeping.respironics.com)) and a pulse ox (I like the WristOX myself (http://www.nonin.com/products.asp?ID=19&sec=1&sub=6)) but you had to manually sync up the results yourself. As an aside, you can use the Stardust with as few as one channels, such as Pox only, or 2, like the Pox and flow cannula, and have it synched up automatically.

In and of itself, pulse oximetry can't tell you if you're having an apneic episode, just that you're having an O2 desat. And that could be due to a lot of different things (pneumonia, pulmonary embolism, left to right shunt etc. etc.) That's why I like to have at least one other bit of data to correalate it with.

So anyway, newbie to the forum here, so go easy on me I've been really impressed with a lot of answers and people on here. Hopefully I can make some positive contributions here and learn a good deal in the process, too. It's great when people take a proactive role in their care. A fair part of my day is just trying to convince people to wear the darn thing, lol.

[rested gal]

Brian, welcome to the board. You hopped in just in time for a very interesting discussion thread.

Fellow Tennessean, I see.

[-SWS]

So anyway, newbie to the forum here, so go easy on me.

Hey, Brian- Welcome to the forum and THANK YOU for coming to the forum!!! .

[BrianRT]

Thanks, guy!

So what part of TN, RG?

[-SWS]

I for one didn't forget about this excellent thread. I've been thinking about the various key (potentially "hot")underlying issues.

Bill, I most definitely haven't ignored your sincere question. I've been seriously thinking about it, because I think it's a very good question. You mentioned that you'd like to see pulse oximetry deployed wide scale for apnea diagnoses. I'm certainly not the correct person to answer your question because I don't have a medical background. But I personally think oximetry falls short, population-wide, as being sufficient for apnea "diagnoses" for several reasons already highlighted in this thread.

As a definitive means of diagnosing an entire patient population, pulse oximetry sensitivity and specificity both fall short. In fact, pulse oximetery is not exclusively used for SDB because there are quite a few other cardio-pulmonary disorders that entail desaturation. And that implies a need for "differential diagnosis" which transcends the scope of oximetry. Additionally, Bill, I think the many co-morbidities of apnea also influence diagnostic procedures, since some of those co-morbidities are cause while others are effect regarding pathophysiology. In my layman's way of thinking, it thus makes sense to at least attempt to diagnostically differentiate both the root cause and the extent of the problem(s) in physiology.

So Bill, that's what I think is going on in medical minds regarding why apnea is not necessarily viewed as an easy diagnosis. But also why they don't tend to think pulse oximetry constitutes sufficient diagnostic criteria across the patient population. However, here's where opinions may get a bit more interesting for you, Bill. What I kinda think you may be getting at is what Dr. Barbara Phillips refers to as the apnea candidate who even the "janitor at the sleep clinic can recognize" (see podcast link below). These would be the more overt presentations of apnea who are virtually guaranteed to flunk the pulse oximetry test exactly as you and I suspect they would, Bill. Dr. Phillips recommendation for those patients with overt apnea presentations? Skip the screening and diagnostic wait, and start these patients on CPAP trial straight away. Kind of reminds me of the medical concept of triaging when patient numbers become too vast for the resources on hand. Presumably these patients with overt apnea presentations will be evaluated for co-morbidities later down the road.

But in this series of podcast presentations linked to below, Phillips and Dement acknowledge the various problems of apnea under-diagnosis, too few PSG beds relative to apnea candidates, PSG-phobic patients, those challenging SDB patients not recognizable by the building janitor, the medical community not truly "knowing what apnea is" (regarding extensive etiology, pathophysiology, epidemiology), etc. They acknowledge reasons to change the current apnea health-care model, and they intelligently consider how ambulatory testing might either hinder or facilitate necessary changes. This is only one of very many tough-to-solve issues in a very young branch of medicine, in my own opinion

Here's a link to the thread with the Dement and Phillips podcasts (thanks to Padacheek):
viewtopic/t17923/Sleep-Review-PodCast-o ... udies.html

Okay, running out of time at my end. So I'm not going to go back and fix my typos or crappy sentence structure. I'm going to bed since we have two sleep professionals in this thread! .