Select Page

The impetus to develop the REPAP protocol derives from the recognition CPAP failure often emerges because continuous pressurized air provokes expiratory pressure intolerance (EPI) when settings are raised to titrate out RERAs. In accordance with AASM standards, a sleep laboratory must attempt to titrate out RERAs, the most difficult to treat of the three main breathing events (apneas, hypopneas, RERAs). To solve this problem, we learned several years ago the sophisticated algorithms embedded within the auto-adjusting technology of ABPAP and ASV coupled with an assist through the manual override of the sleep technologist were sufficient meet AASM standards to eliminate RERAs without triggering EPI.

Although we alluded to this phenomenon in our initial PAP-NAP paper1 published in 2008, it was not until four years later we provided more definitive data2 and commentary3 on the topic to explicitly point out the failure of the CPAP mode of air delivery. Since then and more recently we published several papers describing the use of manual titrations for advanced auto-adjusting technology.

With that backdrop, we turn to the Results section of the REPAP article where we will delve into two components necessary to achieve these objectives: change the mode of PAP delivery and therefore change the pressure settings on the new mode. In addition, as implied by the REPAP concept, these mode and pressure setting changes can occur at any repeat titration, but the more common scenario is a patient arrives at their new and final mode early in the process, which means future retitrations focus on changes in pressure settings. Then again, due to specific insurance constraints, it is not unusual to encounter a patient who shows central apneas on successive retitrations spread out by intervals of months if not longer who only qualified for an ASV device at a much later point in the process. Without insurance constraints, many patients would have benefited from an earlier transition to ASV, and we know this point is valid anecdotally in light of a few patients who opted out of insurance coverage to buy an ASV device out-of-pocket, and all of whom reported immediate benefits unequivocally superior to any prior PAP technology used in the past.

Naturally, changing modes precedes changing pressures, so that topic comes next in the paper. In the first round of retitrations 227 of 273 patients exhibited intractable pressure intolerance, subjectively or objectively or both forms. Forty-five patients switched because of intractable residual breathing events. Only one patient in the entire chart review remained on their initial device of auto-CPAP (APAP). The pattern of new prescriptions included the following breakdown: ABPAP [48.9%], ASV [30.8%], and fixed BPAP [19.9%]. The latter proportion reflected our chart review dating back to 2006 when we had yet to start our more aggressive efforts to transition patients to auto-adjusting technology. The user rates in each of the three groups included: BPAP (35 of 53 [66.0%], ASV (65 of 82 [79.3%], and ABPAP (109 of 130 [83.8%].

Looking at the 158 patients in the sub-group who underwent more than one procedure, their final (most recent) retitrations demonstrated much greater proportions for the auto-adjusting technology: ASV [65.8%], ABPAP (42 [26.6%], BPAP [4.4%], and iVAPS [3.2%]. To better understand these percentages, it is important to recognize greater than 90% of these patients had been on ABPAP at some point before their last retitrations. Thus, the predominance of ASV in this sub-group is an overestimation. In other words, in our clinical practice, ABPAP almost always encompasses more than 50% and often as much as 80% of initial prescriptions when a mode is changed. ASV numbers pick up over time when ABPAP fails and a patient qualifies for complex sleep apnea (10 central apneas; CAI > 5; CAI/AHI ratio > 50%). In this sub-group, the proportion of users again showed relatively consistent numbers in all groups: ABPAP [92.9%], BPAP [85.7%], and ASV [77.6%].

Pressure setting changes are somewhat more difficult to describe because switching from CPAP to auto-adjusting modes means we are describing a system with one pressure setting changing to systems with 3 to 4 pressure settings. For ResMed devices, which were used almost exclusively in this research, ASV settings include Maximum and Minimum Pressure Support and Max and Min Expiratory Pressure while ABPAP uses Max Inspiratory Pressure, Minimum Expiratory Pressure, and Pressure Support. By focusing on changes in expiratory pressures there is some parallel between CPAP settings and the EPAP (expiratory) settings on the advanced devices. Both increases and decreases of EPAP were observed, the former when the original CPAP pressures were not effectively eliminating hypopneas and apneas, and the latter when EPI caused disruption to sleep quality. This point explains how tricky expiratory adjustments during titrations may become: too high might solve breathing problems but provoke EPI; too low might prevent EPI but breathing problems persist.

Initially, EPAP settings were decreased in 65% of the sample while 30% needed an increase in EPAP. Only a handful of patients needed no initial change in EPAP settings, yet nearly all patients ultimately received adjustments in their inspiratory settings in the fine-tuning effort to normalize (“round”) the airflow curve. Inspiratory pressure settings tended to be more straightforward in terms of eliminating the last vestiges of residual breathing events, usually RERAs (or flow limitation) and hypopneas of lesser severity.

The remainder of the results delved into various statistical analyses in an attempt to tease apart the reasons 72% of patients were able to transition from CPAP failure to PAP use. The bottom line appeared to be two simple pronouncements from the patients: they slept better and reported easier adaptability to the PAP device. From a research standpoint, the questions would be what exactly caused the patient to report sleeping better and anticipating easier adaptability. As you will see, the answer seems straightforward but proving the answer with the scientific method could not really be achieved in this type of research design.

Our take of course starts with the simple effects of ABPAP or ASV to eliminate the discomfort of expiratory pressure intolerance coupled with a normalization of the airflow signal never previously experienced by the patient while using CPAP. For these things to prove so important, however, we would like to see objective data to corroborate these changes. Not surprisingly, then, were several data points showing exactly this phenomenon. Among patients who eventually became users after their first retitrations the following improvements were observed: greater time spent in REM and less time spent in Stage 1 NREM (the most superficial stage of sleep); decreased sleep onset latency, and increased sleep efficiency (time asleep/time in bed). These changes were noted in comparison to those who remained non-users. Among the group undergoing multiple retitrations, similar analyses revealed these improvements in users: greater total sleep time, increased sleep efficiency, decreased sleep onset latency, less time awake after sleep onset, less Stage 1 NREM and more REM time. AHI was also significantly lower and there was a statistical trend for a lower RDI. Again, all these improvements were noted in comparison to those who remained non-users.

Two other factors associated with less use are worth mentioning because others have reported the same findings: married patients have higher rates of use than single patients; and individuals suffering from depression have lower rates than those without depression.

We now arrive at the final part of the paper in which we discuss (Discussion) several facets of the research with a special interest in how other sleep professionals could make use of our findings, the potential impact of the REPAP approach on current practices, and how to further research the concepts in the REPAP protocol.

One of the most obvious ways to take advantage of the REPAP protocol is to recognize its simplicity. To quote our paper, “Parsimoniously, discomfort (defined as physically distressing sensations and worse sleep) caused PAP failure. However, new technology enhanced comfort; and, greater comfort resulted in enhanced adaptability and use, which aligns with a rationale to make PAP more user-friendly to improve self-efficacy and ultimately increase adherence.” Said another way, it is obvious huge numbers of patients fail CPAP, and therefore it should not be a difficult approach to inquire of these individuals to what extent they are not achieving optimal results and whether they experience discomfort when breathing out against pressurized air. The former question will often imply RERAs may not have been effectively titrated away, and the latter question almost always signals expiratory pressure intolerance. Either problem or both is sufficient grounds to consider the value of the REPAP protocol.

However, this simplicity does not necessarily help to overcome three pernicious obstacles in the field of sleep medicine that will block efforts to encourage patients to enter a REPAP protocol. First, most sleep specialists are trained to believe in the “one-and-done” model of titrations as well as the “one-size-fits-all” mentality alleging CPAP works for all patients. Both conceptualizations are not only incomplete, but there is no evidence whatsoever to imagine they are genuinely valid ways of understanding OSA patients’ capacity to use PAP therapy. Second, sleep specialists are further constrained because of anxiety and fear in dealing with insurance carriers who tend to believe the same two points: a single titration and CPAP for all! (Let’s not forget though the foresight of Medicare when several years ago it mandated the necessity of retitrations in non-compliant patients, thus promulgating an exceptionally patient-centric model of care very much aligned with the REPAP protocol.) Last, insurance carrier obsessions with the HST model of care is already preventing sleep professionals from even attempting retitrations, because in the world of HST logic, data are interpreted to support the view APAP is all that will ever be needed. If a patient is not achieving optimal results, the only likely explanations are mask fit problems, mask leak, or mouth breathing, all of which require only adjustments to the mask or headgear and chinstraps. Even when residual breathing events and poor outcomes persist, insurance gatekeepers frequently fight with sleep professionals to thwart any opportunity to visit the sleep laboratory for a new evaluation, either for diagnostic data or a retitration or both.

Forging ahead I am very hopeful other sleep professionals will not only cite our work, but perhaps more practically, our efforts will encourage patients, other sleep professionals and insurance gatekeepers to actually read the ideas and details found in the article. If the information were to disseminate into our field through various means, it seems reasonable more patients would eventually be able to avail themselves of a REPAP protocol when they are not achieving optimal results or when they continue to report discomfort with single-level CPAP technology. And, let’s not forget that the AASM provides plenty of support for this approach in their current and past guidelines on the subject.4-5 Yet, as we mentioned in the paper, this area seems to be a blind spot among many sleep professionals who for the reasons stated above continue to discount or simply ignore the potential value of retitrations as well as advanced PAP technology.

Like all innovative studies, more research and replication of findings are crucial to expanding knowledge in the field. Our protocol reflects a lower level of evidence as it was a retrospective chart review comparing a non-random sample of users versus non-users. The essential research to confirm or reject the value of a REPAP protocol would be a prospective randomized controlled study to compare two groups: one receiving treatment as usual in which no retitrations would likely occur and the other a REPAP group where retitrations are the proposed practice pattern. In one sense, this research has already been done all over the universe of sleep centers and laboratories, because undoubtedly a fair number of patients having sought care for CPAP failure eventually made it back to the sleep lab. Nonetheless, prospectively investigating a well-defined group of CPAP failure cases and then randomly assigning them to REPAP or treatment as usual would go a very long way in explaining how retitrations might be especially beneficial to such cases, albeit one of the principal components of our model is the implementation of a manual titration of auto-adjusting technology in the sleep lab. Thus, another research experiment that would prove necessary will compare two groups of patients returning to the sleep lab due to CPAP failure, one randomized to repeat a CPAP titration and the other receiving advanced PAP devices, manually titrated, during the procedure.

Finally, one of the last points we wrote about in the paper is the critical distinction between patient education and patient experience. As we have long advised, no amount of education can overcome certain physiological problems such as residual breathing events and expiratory pressure intolerance. Ultimately, we consistently and repeatedly observe a sizeable proportion of patients need to experience the experience of successful use of PAP. By going through the laboratory experience to be titrated with a more advanced PAP device, we see close to an 80% chance or greater that individuals will feel and sense the auto-adjusting dual pressure device is easier to adapt to and more comfortable. In this process, it is very common for the individual to report the next morning having slept better, again another favorable and motivating experience the patient has now experienced. These events must be distinguished from the process of talking to a patient and educating him or her about PAP. Instead, the REPAP protocol offers patients the actual experience of a good response to PAP, and in our clinical care model we have not been surprised to witness that “actions” speak louder than “words.”

In the final post on the REPAP protocol, which will likely appear a few weeks hence, we will discuss some of the interesting commentary that emerged with the publication of the paper. To date there have been two main reviews or op-eds on the paper, one published in Respiratory Care alongside our paper and the other published in the Sleep Review trade magazine. We’ll discuss both pieces and the interesting points unique to each one.


  1. Krakow B, Ulibarri V, Melendrez D, Kikta S, Togami L, Haynes P. A daytime, abbreviated cardio-respiratory sleep study (CPT 95807-52) to acclimate insomnia patients with sleep disordered breathing to positive airway pressure (PAP-N AP). J Clin Sleep Med. 2008 Jun 15;4(3):212-22.
  2. Krakow B, Ulibarri VA, Romero E. 2012 Adaptive servo-ventilation therapy in a case series of patients with co-morbid insomnia and sleep apnea. Journal of Sleep Disorders: Treatment and Care
  3. Krakow B, Ulibarri VA, Sanchez JN, Kikta S, McIver N, Melendrez D. Driving on “auto”: hands-on is more effective than hands-free. J Clin Sleep Med. 2012 Jun 15;8(3):343-4. doi: 10.5664/jcsm.1934.
  4. Kushida CA, Chediak A, Berry RB, Brown LK, Gozal D, Iber C, et al. Clinical guidelines for the manual titration of positive airway pressure in patients with obstructive sleep apnea. J Clin Sleep Med 2008;4(2):157-171.
  5. Gay P, Weaver T, Loube D, Iber C. Evaluation of positive airway pressure treatment for sleep related breathing disorders in adults. Sleep 2006;29(3):381-401.