OSA and PTSD: Has the Tipping Point Arrived?

Recent clinical and research developments lead may indicate we are getting closer to the tipping point on a more universal recognition of the unusual and unexpected relationship between obstructive sleep apnea (OSA) and posttraumatic stress disorder (PTSD).
Earlier this year, I was collaborator on a symposium submission on sleep disorders in PTSD to be presented at the annual conference of CHEST (American College of Chest Physicians), but which was rejected soon thereafter. Yet, months later or roughly 6 weeks before the October conference in Montreal, Canada, we were invited by a co-chairmen of the entire conference proceedings to present the symposium as previously submitted. The notification was so short that two members of the original symposium speakers could no longer attend. But, four sleep specialists (Drs. Bernard Roth, Vincent Mysliwiec, Christopher Lettieri, and I) were able to deliver the lectures. While I was the only sleep doc in private practice, the other three physicians worked at the most prominent military sleep medical centers in the country: Madigan Army Medical Center Sleep Clinic and Lab, Fort Lewis, WA (Dr. Roth); Sleep Disorders Center at Wilford Hall, Lackland Air Force Base, San Antonio, TX (Dr. Mysliwiec); and Walter Reed National Military Medical Center Sleep Clinic and Laboratory, Bethesda, MD (Dr. Lettieri).
The main idea covered in our symposium was the pervasiveness of sleep disorders in PTSD patients, including how frequently OSA and UARS present in a manner that often steers patients away from imagining they might need a full sleep medicine workup. To be clear, in the example of insomniacs, this observation does not reflect physicians’ attempts to limit their care. Rather, because sleep disorders are most commonly understood in the context of psychological distress, particularly as it relates to insomnia in PTSD patients, there is less emphasis or urgency in referring such patients to a sleep for testing. Germane to this post, all the presentations discussed or mentioned the higher than expected rates of OSA/UARS in PTSD patients and how these patients manifest greater difficulties in attempting to use PAP therapy.
My talk reviewed our experiences with patients using ABPAP or ASV therapy instead of conventional CPAP devices. In a case series we have been compiling on 100 consecutive OSA/UARS patients with moderately severe or greater PTSD, we showed a 61% compliance rate among those using either of these advanced PAP technologies. We also described another 20% of the sample who were using PAP at a sub-threshold level, approximately 14 hours/week or just 6 hours below the numerical definition of compliance of 20 hours/week. These percentages compared favorably to past reports of CPAP compliance in PTSD patients where in some cases the rate was as low as only 30%.
In another development this year, two review articles were published on the co-occurrence of OSA and PTSD. In Jaoude et al’s study, published in the journal Annals of American Thoracic Society, 12 studies are cited that provide evidence for an association between OSA and PTSD, and 4 studies looking at similar PTSD populations did not show an association. No commentary is listed on what type of respiratory sensors were used in the study, but the authors pointed out the difficulties in attempting to score RERAs in trauma survivors who might already being suffering significant spontaneous arousals (i.e. sleep fragmentation) directly related to the PTSD condition itself. The study provided a schematic highlighting the potential role of sleep fragmentation in creating susceptibility to worsened upper airway collapsibility, and also point to possible neuroendocrine changes triggered by arousal activity that could further aggravate the PTSD process. Two of their main recommendations were calls for additional research in PTSD cohorts with co-morbid sleep-disordered breathing to better understand the relationships between the two disorders as well as to explore the best treatment options in these complex patients.
In our article, published in Sleep Medicine Reviews, we looked at 15 studies that specifically examined rates of sleep apnea in PTSD patients. Some studies either due to their being conducted in the 1990s or to other factors tested patients only with the older thermistor technology instead of the more recently accepted apparatus, the nasal cannula pressure transducer (NCPT), the latter considered a much more accurate way to measure hypopneas and flow limitation breathing events. We reported on this key distinction by noting that 11 of the studies using NCPT measured prevalence rates averaging 62% of PTSD patients with sleep-disordered breathing; in contrast, the 4 studies relying on thermistor technology showed only 15% with OSA. In the review, we discussed how the older technology and the related studies may have inadvertently created an impression some years ago that OSA/UARS was not common in PTSD patients.
Just this month, I received a phone call from a prominent sleep otolaryngologist who was treating a military veteran patient for OSA. The patient was attempting to connect the dots between his service in Vietnam and resultant PTSD and then subsequent development of OSA. He was questioning how to respond to the individual and having come across our research contacted me to discuss the appropriate responses. In actuality, he had already sorted out the most reasonable answers to the patient’s questions, but we had an interesting conversation that confirmed his own insights.
The gist of the argument arising in these scenarios is a patient contending he or she had no awareness of sleep apnea until after finishing military service and therefore wonders whether the condition would be deemed “service-connected.” Over the past few years, this issue has received considerable attention as a huge influx of sleep apnea claims have been observed among veterans. The controversy relates to the scenario described by his patient, that is, if he could connect the dots between PTSD and OSA he might be more likely to receive government disability payments as part of his benefit package.
As noted above in both review articles, schematic representations were published suggesting pathways through which a PTSD patient might develop sleep-disordered breathing. In these pathways, a common finding would be that PTSD itself causes severe sleep fragmentation, which in turn has been shown to destabilize the upper airway, all of which would lead to susceptibility for frank breathing events such as apneas, hypopneas or flow limitations. There is even one well-known, scientific study conducted in a non-PTSD sample of patients that showed artificially-induced sleep fragmentation leads to greater collapsibility tendencies of the upper airway. The amazing fact from this study is the sample was comprised of completely normal sleepers with normal breathing patterns. Yet, after just one night of sleep fragmentation, triggered by a bell tone delivered 300 times during the night, these individuals subsequently demonstrated signs of worse breathing on the very next night of sleep.
The problem, however, in declaring that PTSD causes OSA is that no study has ever been conducted to show this exact effect. Rather, the findings are all associations. For example, it is clear that many PTSD patients suffer from OSA or UARS, which raises serious questions on how the two conditions are related. But, the type of study needed to prove the connection beyond mere association, would need to include a large sample of patients in a prospective design followed over an extended period of time. The best example would arguably include baseline sleep testing on several thousand soldiers before a deployment; and then a year or two later, sleep test all the soldiers again. If the theory we are discussing holds, then soldiers who developed PTSD from combat or other stressors from the deployment should manifest a significantly higher rate of OSA/UARS than soldiers who for whatever reasons did not develop PTSD.
This type of study is expensive, labor-intensive, and time-consuming, but nevertheless it would most likely answer the question does PTSD cause sleep-disordered breathing? In our paper, we also showed a schematic describing the potential for not only PTSD itself but also posttraumatic nightmares and insomnia to cause intense sleep fragmentation, which based on the model above, might lead to increasing upper airway collapsibility. Testing then in future studies might look at all three factors, PTSD, nightmares and insomnia, to gain greater clarification how these mental health and sleep disorder process interact to worsen sleep breathing.
Summing up, it is very clear that many patients, physicians, and psychologists are increasingly interested in the possible connections between OSA/UARS and PTSD. The stakes here are quite high, because on the one hand, the disability cost factor would clearly be enormous if it were discovered
PTSD does in fact cause sleep apnea. However, I would also argue on the other hand that finding out such information might lead to a dramatic transformation in how we treat PTSD patients. Imagine for a moment the marked shift that would occur if all PTSD patients were required to undergo sleep studies as soon they were diagnosed with this incapacitating mental health disorder. Conceivably, by solving their sleep problems early in the course of treatment, there is every reason to believe it would reduce the severity of their condition and therefore might reduce their healthcare costs considerably over the long-term.

As described in a prior post, PTSD nightmares were reduced in a dose-response relationship to hours of CPAP use. CPAP of course was used on these patients because they suffered from OSA. Thus, if PAP therapy is so powerful it can reduce chronic nightmare problems in PTSD patients, we can begin to contemplate how this sleep-oriented approach would greatly impact those suffering from this vexing mental health condition. We only need to consider that nightmares on their own have repeatedly been linked to worse quality of life, worse psychiatric distress, greater risks of suicidal ideation and of course worse quality of sleep. Thus, from this example, we see how important it will be to conduct more research on PAP therapy in these patients.
I very much look forward to the time when patients, psychologists and physicians are all on the same page in the way they look at this physiological sleep dimension of PTSD. Whether or not we are at the tipping point, we certainly seem to be much closer to a place of much greater awareness in just these past few years. The acceleration of awareness clearly is coinciding with a faster rate of published works on the topic, and this latter finding is a particularly useful marker to follow going forward. Just using a simple search guide such as "Sleep Apnea" on NCBI will help you follow the most recent research publications in PubMed.

Barry Krakow MD


Dr Krakow’s 27 years of sleep research have focused on the complex relationship between physiological and psychological sleep disorders. Dr Krakow currently operates private sleep medical center, Maimonides Sleep Arts & Sciences, Ltd., and serves as Classic SleepCare’s paid Medical Director.

Leave a comment

Comments have to be approved before showing up.