A study finds ASV superior in patients with comorbid insomnia and sleep apnea.

Insomnia, one of the most common sleep disorders, is often thought of as a psychiatric problem treated with therapy or medications. But emerging research shows many cases of insomnia have a physical cause—specifically, obstructive sleep apnea (OSA) or upper airway resistance syndrome (UARS).

This phenomenon is referred to as “complex insomnia,” COMISA, or comorbid insomnia. A new study, published in The Lancet’s EClinicalMedicine and funded by ResMed Science Center, looked at both the prevalence and treatment of complex insomnia. The results revealed OSA frequently disrupts sleep brainwaves among people with insomnia—who are completely unaware of this underlying process. Moreover, their physicians and therapists rarely suspect a sleep breathing etiology.1

The prospective randomized control trial also revealed the superiority of adaptive servoventilation (ASV) in these patients. ASV attained insomnia remission rates measured on the well-known Insomnia Severity Index in nearly 70% of cases compared to only one-quarter of fixed CPAP users.

According to lead author, sleep medicine specialist Barry Krakow, MD, ASV may offer a far more efficient treatment for complex insomnia than sleeping pills; indeed, drugs may actually worsen the sleep breathing problems and thus aggravate the insomnia in the process. In past research, the group has shown that nearly 90% of chronic insomnia patients failing sleep aids were suffering from undiagnosed OSA or UARS.2

In the current project, the researchers selected archetypal insomnia patients like those presenting to primary care or mental health clinics. They experienced chronic insomnia on average for a decade and failed numerous therapies, including drugs and behavioral interventions. To remove any OSA bias at the outset, patients with obvious abnormal sleep breathing risks or symptoms (such as obesity, daytime sleepiness, or witnessed apneas) were excluded. Remarkably, even though 660 patients were initially considered due to a primary complaint of a chronic insomnia disorder, 455 were immediately excluded due to these sleep breathing markers, yet their providers had not taken the steps to diagnose a potential sleep-breathing disorder. Because these patients are so similar to what’s seen in the general population, the prevalence of complex insomnia appears to be astonishingly higher than typically described in the scientific literature.

In a phone interview with Sleep Review, Krakow explained the treatment phase of the study where two sets of insomnia patients were compared for a 14-week trial. One group used ASV, the other CPAP. (The transcript has been lightly edited for clarity.)

Sleep Review (SR): What are the most important findings of this study?

Barry Krakow, MD: Both CPAP and ASV improve sleep quality, but it’s clear ASV improves sleep quality significantly more efficaciously than CPAP. Once you target sleep quality, a lot of other things fall into place: sleep fragmentation decreases so insomnia decreases, then more benefits accrue such as decreased trips to the bathroom and fewer awakenings along with daytime benefits of less fatigue and enhanced quality of life.

The remarkable thing is that in just the first two weeks of the study, ASV patients were already reporting more substantial gains in sleep quality than CPAP patients.

SR: Why is ASV more effective than CPAP?

BK: No surprise; it’s all about the breathing. ASV is more effective in eliminating all breathing events including the flow limitation events, aka RERAs. So, by the end of the study ASV patients average normalized airflow nearly 90% of the time whereas CPAP could only attain a level of normalized airflow 70% of the time.

SR: So if the key is effectively treating RERAs, and other sleep centers keep using CPAP without these optimal results, why do they keep using CPAP?

BK: Great question, but you’re assuming most sleep centers pay attention to RERAs; I suspect that assumption is flawed, because for the past decade we’ve treated thousands of patients on CPAP who sought second opinions at our center, and none of these patients had ever heard of the terms RERA or flow limitation. They knew about hypopneas and apneas but not these breathing events linked to upper airway resistance. I should add most sleep centers still won’t diagnosis UARS.

SR: What would make sleep doctors change their practice models, not just to consider other types of PAP devices, but also to actually schedule sleep studies on these people with insomnia to test for OSA or UARS?

BK: I think a major barrier is that sleep doctors were blindsided years ago when insurance companies rejected the idea that a sleep study was needed for insomnia patients. The great irony here for those of us in sleep medicine in the 1980s and 1990s, insomnia was actually a major criteria to raise suspicions for an OSA diagnosis. Since then, probably more than 50% of sleep doctors have never been trained to suspect a link between the two conditions, even though the late, great Dr Christian Guilleminault discovered the connection almost 50 years ago. So, putting it bluntly, the field of sleep medicine needs a wakeup call to the enormous prevalence of complex insomnia and that the vast majority of insomniacs showing up in various clinics and at sleep centers need a sleep study, and the sleep study needs to measure more than just apneas and hypopneas. Then PAP therapy needs to be offered.

SR: Do you have any specifics to move sleep professionals in this direction?

BK: Surprisingly, I don’t think a large proportion of sleep doctors attempt regular use of a PAP device. I would encourage them to try CPAP for one night and then ASV for one night and see how they stack up. Of course, a patient-centric model is a must. If you work with chronic insomnia patients who are failing therapies, prescription sedatives, OTC sleep aids or behavioral interventions, both patients and providers should zoom in on sleep quality (unrefreshing sleep reports are a dead giveaway). If sleep quality is compromised, then it should be clear something must underlie this condition, and there is no greater threat to the quality of one’s slumber than sleep-disordered breathing.

Once you cross this diagnostic barrier, the patient is still not home free, because these patients are going to receive initial exposure to CPAP, which often causes traumatizing experiences due to the difficulties people report in breathing out against the fixed pressurized air.

Because my initial work in sleep medicine involved nightmare patients with PTSD, I became acutely aware of the suffering these individuals experienced when they first attempted CPAP. In fact, these trauma survivors were invaluable resources as they taught us that we had to find another way. Sadly, many sleep professionals, not to mention arcane insurance guidelines, are wedded to the belief that CPAP works for everybody. If that were true, then why is CPAP compliance the single most problematic aspect in the treatment of OSA? The answer is that CPAP does not work or work well except on prototypical sleep apnea patients, but that hasn’t stopped many sleep centers from attempting to force CPAP on patients who will never be able to adapt to it.

SR: Could you provide more details again on why ASV is superior to CPAP? Also, I understand you’ve raised the question on whether the use of CPAP could be medical malpractice in certain patient cohorts. Could you expand on that?

BK: First let’s dispel one myth. Unfortunately, many individuals working in sleep medicine have persuaded themselves that advanced PAP therapy such as ASV or autobilevel [ABPAP] are all about marketing and therefore about making greater profits with more expensive machines. As we have repeatedly demonstrated reversal of CPAP failure cases by using advanced PAP, we would like to think consideration would be given to how much money is saved by turning a failure into a regular user. That information alone should whet the appetite of sleep professionals searching for better ways to solve CPAP failure, as well as insurance companies eager to see better outcomes in their patients, not to mention the bonus of health cost-savings.

The biggest question about PAP modes revolves around how patients breathe out against incoming air. And, the answer may surprise many sleep professionals, because they probably haven’t given it that much attention, otherwise they would have ceased using CPAP as we did back in 2005. What we learned is that patients either self-report the discomfort of trying to breathe out against CPAP…what we call subjective expiratory pressure intolerance or EPI. Or, in the sleep lab we observe irregularities on the expiratory limb of the airflow curve (objective EPI). Once you take note of these poor responders, you switch the patient to bilevel modes, and you will often see improvement straightaway. We have found during the past 10 years that either ABPAP or ASV are the most effective in resolving this issue.

Last, keep in mind we’ve only mentioned expiration. The second part of this equation, once you prevent or eradicate EPI, is to then raise inspiratory pressures to more effective levels to eliminate the flow limitation. Because ABPAP and ASV devices have special proprietary algorithms to solve these problems, we found we could fine-tune the settings by overriding these algorithms in the sleep lab environment.

If you appreciate this approach to care, then you need to develop a great deal of sensitivity when you attempt to start PAP therapy on a PTSD or other anxiety patient. If you force CPAP on such a patient, they will not only reject it, but they will engage in classic avoidance behavior and drop out of care for months or years or longer. In my view, we have unintentionally traumatized the patient with CPAP, causing them both short-term and long-term harm. Regrettably, this model of care approaches a medical malpractice scenario, because the provider offered the patient no other PAP modes at the get-go to rescue the patient from experiencing the phenomenon of “drowning in air.”

SR: Sounds like there’s a conflict in the way you and others are practicing sleep medicine versus some of the standards in the field?

BK: Which brings up something of a sore spot, because we have been very surprised that the American Academy of Sleep Medicine [AASM] has shown so little interest in promoting more research on complex insomnia. The AASM will occasionally cover this topic at the annual SLEEP conference. But it’s never promoted as one of the most important areas of discussion, which is shocking in that comorbid insomnia and sleep apnea is arguably the single most common sleep disorder combination diagnosed and treated at the vast majority of sleep centers.

I believe once the AASM assumes a more assertive position on the nature and prevalence of complex insomnia, then more sleep specialists will reassess their current approach to the condition. At that point, I would expect to see a lot more push back against the poorly written insurance policy statements we all confront in everyday practice.

SR: What are some additional research questions that still need to be investigated?

BK: There are many. For example, one of the most important aspects of insomnia is to recognize that it’s a complex disorder. It has psychological and physiological aspects. Even though we had stellar results for the ASV patients, we still saw the need for additional therapies for many of these individuals, such as cognitive behavioral therapy for insomnia [CBT-I].

An immediate question is: How do you integrate this type of treatment plan into the clinical realm? Do you start with CBT-I and then get a sleep study? Do you start with a sleep test, then treat the OSA/UARS with PAP, and then do CBT-I? You can easily imagine that one’s specific field of healthcare strongly influences how to proceed.

Another big gap in knowledge is that no one’s ever determined what constitutes a perfect or optimal titration on a PAP device. As I alluded to earlier, it seems there is scant interest in the concept of normalized breathing. That strikes us as odd. We’ve seen normal sleepers in our sleep lab, and we were quite excited to see that their breathing looked identical to the sleep breathing of someone obtaining a great and consistent response to PAP. I can recall only one research study examining this concept,3 which we believe is critical because our results show how ASV produced substantially more time with normalized airflow signals compared to CPAP.

SR: Are there any other big challenges?

BK: The findings do pose a challenge to virtually all types of clinics that insist on viewing insomnia as a psychological disorder. Think about the psychiatric clinics and primary care clinics that often jump to prescription sleeping pills faster than a bullet train to nowhere. Because of the constrained environments in these clinical settings, how will these types of providers come to realize the incredible importance of physiological sleep testing?

This challenge affects all the behavioral sleep medicine professionals and the psychologists. They provide CBT-I services that are crucial components of therapy, but I believe it’s a mistake to delay sleep testing. Sleep apnea affects so many different organ systems in the body that adversely influence physical and mental health. Just because these types of patients are initially viewed through a psychological lens should not preclude raising questions early on about sleep apnea. In fact, I would predict mental health cohorts will eventually generate the largest number of referrals to sleep centers because the insomnia, so common in mental disorders, is going to turn out to also be fueled in part by sleep-disordered breathing.

Unfortunately, the wakeup call to providers and insurers to change referral practices may require repeated alarms going off, because fields like psychiatry and psychology have their own unique administrative hassles to manage in dealing with insurers. Indeed, I’ve broached the topic of inserting sleep disorder centers into 10 different psychiatric hospitals. Over the last decade, I’ve reached out to some of the most prestigious facilities in the country, and you may or may not be surprised to know they all respond the same way: “We can’t bill for it.”


1. Krakow B, McIver ND, Ulibarri VA, et al. Prospective randomized controlled trial on the efficacy of continuous positive airway pressure and adaptive servo-ventilation in the treatment of chronic complex insomnia. EClinicalMedicine. 2019 Aug 8;13:57-7.
2. Krakow B, Ulibarri VA, McIver ND. Pharmacotherapeutic failure in a large cohort of patients with insomnia presenting to a sleep medicine center and laboratory: subjective pretest predictions and objective diagnoses. Mayo Clin Proc. 2014 Dec;89(12):1608-20.
3. Condos R, Norman RG, Krishnasamy I, et al. Flow limitation as a noninvasive assessment of residual upper-airway resistance during continuous positive airway pressure therapy of obstructive sleep apnea. Am J Respir Crit Care Med. 1994 Aug;150(2):475-80.