Every year around mid-December we work on polishing a series of brief abstracts to be submitted for publication and presentation at the annual SLEEP conference; the next event will be held June, 2016 in Denver, CO. This year, we have six staff members, including myself, who are writing six abstracts. By way of background, it may be of interest to know that my other five colleagues on these submissions are all registered polysomnography technologists with varying degrees of experience, ranging from 3 to 9 years of collaboration with our sleep center and sleep research institute.
Of special interest, every one of these five technologists is currently applying or anticipating applying to graduate school in such areas as physician assistant programs, neuropsychology and other biomedical sciences. I wanted to mention this background, because our center almost always hires individuals to train at our center to become sleep technologists, but they also possess the ambition to move on to higher levels of education in health sciences. As such, each person is highly motivated in working on abstract submissions, not only because of genuine interest in research but also because it is valuable training and experience for graduate level education. In the past, many of our technologists have conducted poster presentations and on rare occasion, their works have been accepted for an oral presentation.
In this post, I am going to summarize briefly (briefer than the abstract itself) some of the key points from each of the projects as well as a more thorough explanation of the rationale for choosing each area of research. Regrettably, I cannot include specific data points from each abstract, because it would then fall into the category of “pre-publication,” which would void our ability to publish the abstracts at all. Nonetheless, the qualitative information provided here should give you the gist of the findings and how they fit with our interests in these areas.
Our abstracts divide up into two main categories: relationships between insomnia and sleep-disordered breathing; and, PTSD and sleep. Three abstracts were submitted for each of these two areas.
The combination of insomnia and sleep-disordered breathing was termed, “complex insomnia” by Dominic Melendrez and myself in an abstract at the 2000 SLEEP conference and again in our paper by the same name in Biological Psychiatry in 2001. Since then, we have published many papers on the topic, most of them focusing on the high prevalence of complex insomnia and a lesser amount addressing the treatment of the breathing problem to determine effects on the insomnia problem. In past work, we have shown that treatments ranging in scope from nasal dilator strips to CPAP appear useful in treating OSA/UARS in insomnia patients; and, most importantly, insomnia symptoms decrease in moderate to marked fashion.
Most of our treatment research in the complex insomnia area has dealt more with sleep maintenance insomnia (staying asleep difficulties) than sleep onset insomnia (difficulty falling asleep). Moreover, the sleep onset problem is considerably less common among patients seeking treatment at sleep centers; where sleep maintenance problems present at least 10 to 20 times more frequently. Nonetheless, we have been using advanced PAP technologies such as ASV or ABPAP for more than 5 years, so we decided to look back to find a sample of sleep onset insomnia patients who were using treatment to see whether or not they also reported any improvements in their bedtime insomnia symptoms.
In our chart review, we found more than 50 but less than 100 patients (again, sorry for not providing more precise details). These patients were very severe sleep onset insomniacs who on average took more than 2 hours to fall asleep most nights. And, all the patients included in the study reported very clear impressions affirming psychological factors as the major factors in their insomnia. For example, nearly all the patients suffered from racing thoughts at bedtime or otherwise did not know how to turn off their minds. Others engaged in clock watching behavior that caused intense frustrations about not sleep, which in turn led to losing sleep over losing sleep.
Among this sample, we found that at follow-up, three times as many patients were compliant with PAP therapy than those non-compliant. And, most clinically relevant, patients using PAP more consistently and for longer hours showed the most robust treatment effects; the insomnia reductions were essentially double in magnitude for global insomnia ratings and for the specific problem of sleep onset insomnia compared to the insomnia reductions in the partial users of PAP therapy. Clinically, the regular PAP user group dropped below moderate status of an insomnia problem, which had originally presented in the severe to very severe range; whereas the partial users still suffered from insomnia in the moderate to severe range.
The good news from this abstract is that even some use of PAP therapy was associated with reductions in insomnia, globally and at sleep onset; whereas, consistent use of PAP therapy was associated with dramatic improvements. Notice I used the word “associated with” because these were what are known as retrospective chart reviews, which is a lower level of evidence.
To increase the level of evidence, this year we started a prospective, randomized controlled trial, comparing CPAP with ASVAuto in chronic insomnia patients with OSA/UARS. The study is posted at www.ClinicalTrials.gov Identifier: NCT02365064 and is receiving funding by ResMed Science Center. In the protocol, we hope to prove at least three things. First and foremost, this study will be the first to only use PAP therapy in a randomized controlled study to evaluate effects on insomnia. Thus, whatever the outcomes between groups, we will likely provide definitive evidence that PAP therapy, regardless of type, can reduce insomnia. Second, from our clinical experience, we have found that insomnia patients gravitate rapidly toward ASV and away from CPAP, because the former is so much more comfortable to use than the latter. As such, patients find adaptation much easier with ASV than with CPAP. Thus, we anticipate long-term results are likely to show more sustained insomnia improvements with ASV. Finally, another aspect of insomnia is the pervasive complaints of non-restorative sleep, and again our clinical practice shows that ASV is much more effective in decreasing the flow limitation events in OSA/UARS patients, which typically yields greater objective sleep consolidation and subjective reports of greater sleep quality the morning after first use and subsequent use.
In our first abstract on the RCT, we looked at a small number of patients who had completed the 6-week mark of the study. The study runs approximately 15 weeks. In comparing ASV and CPAP patients, we found no differences in the changes in insomnia. Still, global insomnia dropped in both groups by 50% from their baseline severity. In fact, on average, the patients no longer met criteria for an insomnia disorder when using a cut-off score widely applied to the instrument known as the Insomnia Severity Index (ISI), the main measure in our study.
Despite the lack of differences on the ISI scale in this very early phase of the study, we did see differences between the two groups regarding the effectiveness of eliminating the flow limitation events (RERAs in the technical vocabulary of sleep medicine). The ASV group had dramatically lowered their breathing event index to levels no longer consistent with a diagnosable case of sleep-disordered breathing, either OSA or UARS; whereas the CPAP patients still showed many flow limitations and therefore would have still met diagnostic criteria for UARS.
It will be very interesting to see what occurs through the end of the study, given that there are clear differences on the objective findings (breathing events) for the CPAP and ASV groups; whereas the subjective findings (insomnia) are virtually the same for both groups. Will these changes hold up at the end of the study and with more patients involved in the final sample? We are hoping to complete the study by the end of 2016 or sometime soon thereafter.
The second abstract on the RCT is in some ways more revealing than the first, because we were not surprised by the finding of PAP decreasing insomnia severity since we and several other groups have shown similar findings, albeit mostly in uncontrolled studies and often with retrospective chart review designs. In this second abstract, which relates more to the concept of nonrestorative sleep, we showed that ASV patients reported less insomnia-related impairment and greater gains in quality of life than the CPAP group. Some of these findings were statistically significant and others were called trends, meaning nearly significant statistically but not quite. Regardless of significance, however, the differences between the two groups showed what are called medium to large effects. These terms reflect clinically meaningful levels of change. And, in finding this distinction, we immediately wondered whether or not this finding might have been related to the greater reduction in breathing events in the ASV group than in the CPAP group.
As we gather more information from more patients in the study, we will closely monitor these changes. And, it cannot go without saying that these findings while of magnitudes consistent with real clinical change, the number of patients in the sample and the short period to date demands a very cautious interpretation of the data. It is a very exciting project with which to be engaged, and we very much look forward to the final results so we can publish definitive observations about the project. For now, the current findings should be viewed as speculative interpretations.
In the next post, we will delve into the three areas of PTSD sleep, including how to help these patients use PAP therapy more regularly, how PAP therapy may provide a treatment for their nightmares, and how sleep disorders in general may factor into one of the three core symptoms of PTSD, known as avoidance behavior.