Positive Airway Pressure Therapy With Adaptive Servoventilation Part 1: Operational Algorithms

Transcription

1 [ Contemporary Reviews in Sleep Medicine ] Positive Airway Pressure Therapy With Adaptive Servoventilation Part 1: Operational Algorithms Shahrokh Javaheri, MD, FCCP ; Lee K. Brown, MD, FCCP ; and Winfried J. Randerath, MD, FCCP The beginning of the 21st century witnessed the advent of new positive airway pressure (PAP) technologies for the treatment of central and complex (mixtures of obstructive and central) sleep apnea syndromes. Adaptive servoventilation (ASV) devices applied noninvasively via mask that act to maintain a stable level of ventilation regardless of mechanism are now widely available. These PAP devices function by continually measuring either minute ventilation or airflow to calculate a target ventilation to be applied as needed. The apparatus changes inspiratory PAP on an ongoing basis to maintain the chosen parameter near the target level, effectively controlling hypopneas of any mechanism. In addition, by applying pressure support levels anticyclic to the patient s own respiratory pattern and a backup rate, this technology is able to suppress central sleep apnea, including that of Hunter-Cheyne-Stokes breathing. Moreover, ASV units have become available that incorporate autotitration of expiratory PAP to fully automate the treatment of all varieties of sleep-disordered breathing. Although extremely effective in many patients when used properly, these are complex devices that demand from the clinician a high degree of expertise in understanding how they work and how to determine the proper settings for any given patient. In part one of this series we detail the underlying technology, whereas in part two we will describe the application of ASV in the clinical setting. CHEST 2014; 146 ( 2 ): ABBREVIATIONS : AHI 5 apnea hypopnea index ; AI 5 apnea index ; ASV 5 adaptive servoventilation ; CHF 5 congestive heart failure ; CSA 5 central sleep apnea ; EPAP 5 expiratory positive airway pressure ; HCSB 5 Hunter-Cheyne-Stokes breathing ; IPAP 5 inspiratory positive airway pressure ; IPS 5 inspiratory pressure support ; MV 5 minute ventilation ; PAP 5 p o s it ive ai r w ay pre ssu re ; PB 5 periodic breathing ; SRBD 5 sleep-related breathing disorder In the last two decades we have witnessed the introduction of sophisticated positive airway pressure (PAP) devices for treatment of sleep-related breathing disorders (SRBDs). 1 CPAP devices were introduced in and have proven effective in the treatment of OSA, yet adherence is less than ideal. CPAP devices are also frequently not effective in a number of conditions associated with central sleep apnea (CSA) such as congestive heart failure (CHF) and opioids therapy. 3-6 A variety of pressure generators with expanded capabilities have become available. The most advanced of these Manuscript received July 30, 2013 ; revision accepted February 19, AFFILIATIONS: From the College of Medicine (Dr Javaheri), University of Cincinnati, Cincinnati, OH; Department of Internal Medicine (Dr Brown), School of Medicine, The University of New Mexico, Albuquerque, NM; and Zentrum für Schlaf- und Beatmungsmedizin Aufderhöher (Dr Randerath), Institut für Pneumologie an der Universität Witten/ Herdecke, Klinik für Pneumologie und Allergologie, Krankenhaus Bethanien, Solingen, Germany. CORRESPONDENCE TO : Shahrokh Javaheri, MD, FCCP; shahrokhjavaheri@icloud.com 2014 AMERICAN COLLEGE OF CHEST PHYSICIANS. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: /chest Contemporary Reviews in Sleep Medicine [ 146#2 CHEST AUGUST 2014 ]

2 devices, using a mode of operation known as adaptive servoventilation (ASV), 1,7-9 are the subject of this article. Also called auto servoventilation or anticyclicmodulated ventilation, ASV represents the most recent development in PAP treatment of periodic breathing (PB) due to CHF and other conditions associated with CSA not suppressed by long-term use of CPAP. In this, the first of two how-to articles regarding ASV, we discuss the operational algorithms of three ASV devices; a subsequent article will discuss practical applications. A major motivation for the development of ASV ( Fig 1 ) was having a means to effectively treat CSA in CHF. Patients with CHF often suffer from a pattern of breathing first described by John Hunter, although the disorder is commonly referred to as Cheyne-Stokes breathing (for details, see Javaheri 10 ). This unique pattern, hereafter referred to as Hunter-Cheyne-Stokes breathing (HCSB), consists of repetitive cycles of crescendo-decrescendo changes in tidal breathing with interposed central apneas or hypopneas ( Fig 2 ). Decrescendo breathing, central hypopneas, and apneas decrease Pa o and increase Pa co 2 2, which in turn stimulate chemoreceptors thereby prompting the appearance of a crescendo in breathing. The resulting hyperventilation depresses Pa co 2 toward or below the apneic threshold, with consequent decrescendo breathing, central hypopnea, or apnea. 11 ASV devices counterbalance ventilatory instability by modulating the degree of inspiratory pressure support (IPS), providing positive IPS when tidal volume wanes and withdrawing that support when ventilation is excessive according to the prevailing airflow or minute ventilation (MV) ( Fig 3 ). This component of ventilation that is anticyclic to the periodicity of the patient s own breathing acts to Figure 1 The various components of ASV. ASV 5 adaptive servoventilation; auto 5 automatic; CSA 5 central sleep apnea; EPAP 5 expiratory positive airway pressure; HCSB 5 Hunter Cheyne-Stokes breathing; IPS 5 inspiratory pressure support. dampen the oscillations in ventilatory drive that underlie PB. ASV devices also apply a fixed or variable end expiratory PAP (EPAP) to suppress obstructive events. These devices apply mandatory breaths in a timed backup mode to abort any frank apneas ( Fig 1 ). These multiple strategies for suppressing a variety of SRBD events make ASV technology potentially effective in the treatment of complex SRBDs that are a combination of both central and obstructive events. Although (to date) data on autotitrating devices remain conflicting regarding improving adherence, the variable IPS could perhaps be better tolerated than conventional PAP devices and that may eventually translate into improved quality of life and survival of patients with mixed p atter ns of SR BD. In theory, ASV devices should be more effective than conventional PAP devices for suppressing PB. Furthermore, SRBD may exhibit a complex pattern consisting of both CSA/HCSB and obstructive events, first recognized in early studies of patients with CHF and confirmed subsequently Our practice has been to classify sleep apnea as either predominantly obstructive or predominantly central to guide therapy. However, the predominant mechanism of sleep apnea may vary throughout the night along with changes in sleep stage and position, as well as fluid shifts 23 due to the recumbent sleeping position. The severity and the phenotype of SRBD may also change long-term from adjustments in medications, changes in body weight, and acute decompensation or progression of CHF. Given this reality, ASV devices with autotitrating IPS and EPAP may offer a significant therapeutic advantage. CPAP fails to control CSA/HCSB in almost 50% of patients with CHF 3,4 ( Fig 4 ); furthermore, patients have difficulty adhering to CPAP even when it has effectively suppressed PB. In the Canadian Continuous Positive Airway Pressure for Patients with Central Sleep Apnea and Heart Failure (CanPAP) trial, adherence to CPAP at 1 year averaged 3.5 h/night, 4 which is inadequate even though patients with heart failure tend to exhibit short sleep time. 13,14 Meanwhile, an important post hoc analysis of the Canadian data demonstrated that those patients with CHF whose CSA was suppressed by CPAP had improved survival, whereas those in whom CPAP was ineffective did poorly. 4 We have maintained that, in the latter category of patients, CPAP should not be recommended 5 and that this group would most likely benefit from treatment with ASV. Currently, two manufacturers market ASV devices in the United States (ResMed and Philips Respironics, Inc), journal.publications.chestnet.org 515

3 Figure 2 A 5-min epoch of a baseline polysomnogram showing HCSB in a patient with systolic heart failure. Montage from top to bottom as follows: pressure transducer flow, chest effort, abdominal effort, SaO 2. SaO2 5 arterial oxyhemoglobin saturation measured by pulse oximetry. See Figure 1 legend for expansion of other abbreviation. while a third manufacturer markets a different machine outside of North America (Weinmann Medical Technology [Weinmann Geräte für Medizin GmbH 1 Co. KG]). These devices differ in operation, and it is critical that technologists and sleep physicians become familiar with their algorithms. Common Features of ASV Devices These devices have three common features: auto-ips, auto-epap, and auto-backup rate ( Fig 1 ). IPS is variable ( Fig 3 ), being minimal (and even zero) during hyperpneic phases of PB and when breathing is stable. In contrast, IPS increases during periods of diminished ventilation. To achieve this, instantaneous airflow is continuously measured in a moving time window. An average metric of breathing (either peak inspiratory airflow or MV) is computed over the duration of this window such that contribution of breaths gathered prior to the window progressively decreases or is deleted as new breaths are constantly added. Using this metric, ASV devices augment ventilation when mean peak inspiratory airflow or MV falls below a target value, but IPS is minimal or absent when ventilation is stable or excessive compared with the target. All ASV devices also initiate timed mandatory breaths, potentially aborting the course of impending apneas. Finally, EPAP is present to control obstructive events and can be set to an autotitrating mode similar to the same manufacturer s autotitrating PAP. Operation of VPAP SV, VPAP Adapt Enhanced, and VPAP Adapt These devices are derived from the AutoSet CS originally introduced by ResMed. Michael Berthon-Jones is credited with originating the technology and David Bassin for modification of the algorithm. To our Figure 3 A 5-min epoch showing the operation of an ASV device. When the patient s airflow increases (hyperpnea), the pressure delivered by the device decreases (less IPS) and when the patient s airflow decreases (hypopnea) the pressure from the device increases (more IPS). The device algorithm is, thus, anticyclic to the patient s own ventilation. The pressure support varies between 0 cm H 2 O and 4.5 cm H 2 O. See Figure 1 and 2 legends for expansion of abbreviations. 516 Contemporary Reviews in Sleep Medicine [ 146#2 CHEST AUGUST 2014 ]

4 Figure 4 A 5-min epoch of a patient with systolic heart failure. Note: HCSB is present despite the use of CPAP. Montage from top to bottom as follows: flow channel generated from CPAP device, chest effort, abdominal effort, SaO 2, CPAP, 6 cm H 2 O. See Figure 1 and 2 legends for expansion of abbreviations. knowledge, Teschler and colleagues 24 were the first to successfully employ this device. VPAP Adapt (the prefix S9 has been dropped) is the currently available version of ResMed ASV and will be the focus of this article. In all ResMed ASV devices, instantaneous inspiratory airflow is measured and integrated to calculate ventilation. A low-pass filter with a time constant of 3 min provides average weighted MV, giving higher weight to more recent breaths and progressively less weight to breaths recorded earlier ( Fig 5 ). Typically, three time constants encompass most of the necessary information reflecting recent ventilation. Using this continuously updated value, a target of 90% to 95% of the recent average ventilation is calculated. When the actual ventilation decreases below the target, an integral controller increases IPS proportional to the amount ventilation is below the target, and when actual ventilation increases above the target, IPS decreases proportionally to the amount it is above the target. Maximum IPS is constrained such that it can be no higher than 25 cm H 2 O minus the prevailing EPAP, that is, EPAP + maximum IPS 25 cm H 2 O. The Figure 5 The graph shows the ResMed ASV method for weighted averaging of ventilation to calculate recent average minute ventilation. The graph shows the most recent 6 min, although averaging actually extends back indefinitely, with exponentially smaller contributions to the average the longer the interval. The most recent 6 min contribute about 86% of the total. In each box on the graph (representing 1 min), the numbers under the curve represent the proportion the minute contributed to the overall total. See Figure 1 legend for expansion of abbreviations. general form of the servo control algorithm for changing IPS is: IPS 5 gain (target MV 2 actual MV) calculation interval. Gain is 0.3 cm H 2 O/(L/min)/s (converting flow to pressure). Calculation is performed 50 times per second, that is, the calculation interval is 0.02 s. Making the calculation interval short results in the IPS increasing or decreasing more smoothly, preventing abrupt changes. The VPAP Adapt device can be set to one of three modes: ASV auto (engaging the automatic EPAP algorithm), ASV (fixed EPAP), or CPAP. In the two ASV modes, VPAP Adapt operates under an algorithm almost identical to its predecessor, although with four major differences: (1) the ability to automatically titrate EPAP (only in ASV auto mode); (2) allowance of IPS values of zero; (3) improved patient-machine synchronization, tolerating lower breathing rates < 10 breaths/min and expiratory pauses; and (4) modified algorithms that help prevent inappropriate increases in IPS by limiting the maximum rate at which the target ventilation can rise ( L/min/s). This is known as the slew rate limit, an engineering term defined as the maximum rate of change of output per unit of time. Also, the algorithm now facilitates more rapid declines in IPS when values of IPS remain stable (. 90 s) but are significantly above the minimum IPS. This rapid fall in IPS is accomplished by lowering the percentage of average MV that is targeted and the time constant of the ventilation filter (decreasing the moving time window progressively from 3 min to 1 min, thus, giving more weight to the most recent ventilation values). In previous ResMed devices, minimum IPS was 3 cm H 2 O. 7 Currently, the minimum and maximum IPS can be set by the operator at will. We, typically, set minimum IPS to 2 cm H 2 O to 3 cm H 2 O, but for patients with reduced chest wall compliance such as in obesity, we choose minimum IPS of 5 cm H 2 O to facilitate better ventilation. We set maximum IPS about 8 cm H 2 O to 10 cm H 2 O above the minimum IPS and adjusted later as necessary. journal.publications.chestnet.org 517

5 When VPAP adapt is used in ASV mode, EPAP must be set at a fixed level by the operator. When used in ASV auto mode, EPAP varies automatically within clinicianspecified limits. If limits are not set, the default algorithm varies EPAP between 5 cm H 2 O and 15 cm H 2 O. The EPAP algorithm responds to apneas, airflow limitation, and the presence of snoring ( Table 1 ). An apnea is recognized when MV decreases below 75% of baseline ventilation. In the absence of upper airway narrowing, EPAP decreases exponentially toward minimum with a time constant of 20 min. For hypopneas, defined as MV below 50% of the baseline, IPS increases within the minimum and maximum range. There are three choices when the clinician specifies EPAP range. The default setting activates the autotitrating EPAP algorithm. We remain concerned that this algorithm might react inappropriately, though unlikely, and raise EPAP to levels that adversely affect hemodynamics, particularly in patients with CHF. To eliminate this possibility, we commonly set a range for EPAP, best determined using data from a previous CPAP or bilevel PAP titration. We suggest that the minimum EPAP be set to 5 cm H 2 O and that one of two strategies be used for setting the maximum EPAP. One strategy uses the value of EPAP that eliminated obstructive apneas in the supine position during rapid-eye-movement sleep on the previous titration. However, upper airway resistance values may persist at levels high enough to prevent inspiratory PAP (IPAP) during the next inspiration from effectively augmenting airflow. Therefore, setting EPAP about 2 cm H 2 O higher than the value that eliminated obstructive apneas in the previous titration may be a safer option. An alternative strategy sets the maximum EPAP to the level that suppressed all obstructive events (apneas, hypopneas, inspiratory airflow limitation, and snoring) in the previous titration. This usually means a higher maximum EPAP that may constrain the range of minimum and maximum IPS available to the operator resulting in an unsuccessful ASV result. In addition, a lower maximum EPAP may enhance patient comfort and is less likely to adversely affect the hemodynamics of patients with CHF. 5 Appropriate titration of EPAP settings should be individualized and is critical to a successful outcome. 6,25 EPAP takes priority over pressure support, since a patient with significant upper airway obstruction may not be ventilated successfully even when the algorithm ramps up to the maximum of IPS. In all ResMed ASV devices, the backup rate default is 15 breaths/min. The device also detects the patient s breathing rate by computing a moving average of breathing period in a manner similar to that for calculating MV, though over several breaths rather than 3 min, and can adapt to prevailing breathing rates. When ventilation is inadequate in spite of maximum support, or if excessive leak or other factors lead to failure in maintaining an adequate detectable MV, the backup rate will move toward 15 breaths/min by default. Inspiratory time, expiratory time, and rise time are determined by the device algorithm using fuzzy logic and cannot be set externally. VPAP Adapt does not provide an option for expiratory pressure relief. However, a ramp function is available to reduce EPAP to a starting level which gradually rises to the fixed or minimum EPAP setting. While ramping, minimum IPS is applied and along with the subtherapeutic EPAP, sleep-onset respiratory events may occur. In this device, the rise in EPAP is a linear function until the minimum EPAP is reached. The clinician should be alert to this possibility, since an excessively long ramp time may result in sleep disruption due to arousals from these events. A short or no ramp time may be preferable with this device. To guide therapy, a smart card provides much information including apnea hypopnea index (AHI) ( Table 1 ) and apnea index (AI) without differentiating central from obstructive events, compliance, and leak ( e-appendix 1 ). It may be best to assume that all residual apneas are obstructive in mechanism when considering a change in device settings. The AHI computed by all ASV devices is based on airflow alone, and there are no published studies validating this approach. Consequently, we interpret the reported AHI from the perspective of clinical status. If residual AHI is low and the patient is doing well, further studies are not warranted. High residual AHI values need further investigation, which may include another laboratory titration. Excess leak (see e-appendix 1 ) should prompt reevaluation of the size and type of interface as well as pressure changes as appropriate. It should be emphasized that rapid changes in leak during the course of a night render device estimates of ventilation unreliable, irrespective of the brand of ASV. Operation of BiPAP autosv Advanced System One Credit is given to Pete Hill, Mike Kane, Greg Matthews, Ben Shelly, and Heather Ressler at Philips Respironics for creation of the first ASV device equipped with automatic EPAP titration. The BiPAP autosv Advanced System One servoventilation device, the most recent version as well as the original BiPAP auto SV and its 518 Contemporary Reviews in Sleep Medicine [ 146#2 CHEST AUGUST 2014 ]

6 TABLE 1 ] An Overview of the Algorithms of the Three Adaptive Servoventilation Devices Parameter VPAP Adapt BiPAP autosv Advanced System 1 SomnoVent CR Manufacturer Resmed Respironics Weinmann Target parameter of MV Peak inspiratory flow Relative minute volume the algorithm Calculation of target parameters Threshold for inspiratory pressure support Average weighted ventilation Time constant 3 min 90%-95% of recent average MV Average peak flow calculated in 4 min moving window 90-95% of peak flow in absence of SRBD 60% percentile during SRBD Regulation of IPS based on current IPS, slope of IPS and slope of relative minute volume between last 2 breaths No threshold, regulation intends to stabilize relative MV, but not on a predefined level Maximum inspiratory pressure, cm H 2 O Maximum pressure support 252 prevailing EPAP 252 prevailing EPAP 20 2 prevailing EPAP Minimum pressure support EPAP range, EPAP mode 4-15 (default) 4-25 (default) automatic 4-17 (default) automatic EPAP responds to ASV mode: EPAP set manually at a fixed level ASV auto: automatic Apnea, flow limitation and snoring Apnea, hypopnea, flow limitation and snoring Can also be set manually up to 17 Apnea, hypopnea, flow limitation and snoring Early expiratory pressure No Yes Yes relief Definition of apnea Reduction of MV 75% Flow reduction 80% Reduction of MV 80% Definition of hypopnea Reduction of MV 50% Flow reduction 40% Reduction of MV 40% for 2 breaths Backup frequency Auto mode: default 15/min Adaptation to patients breathing in a moving window similar to minute moving window, though over several breaths Backup frequency cannot be set manually Auto mode: default 15/min Adaptation in moving window (12 breaths) Typical rate 3 to 5 breaths below patient s respiration Minimum 8/min (up to 10/min) Fixed backup frequency possible PIF < 50% compared with average PIF Automatic backup frequency: 80% of average breathing rate with highest weight on last breaths Fixed backup frequency possible, minimum 8/min (Continued) journal.publications.chestnet.org 519

7 TABLE 1 ] (continued) Parameter VPAP Adapt BiPAP autosv Advanced System 1 SomnoVent CR Inspiration time, expiration time, rise time Ramp Automatic application of inspiration time and rise time + (while ramping, the rise in EPAP is a linear function) 3 levels of rise time from end EPAP to maximum IPAP + (while ramping, EPAP rises more quickly than otherwise if obstructed events occur) Depending on current IPS, volume and breathing rate application of inspiration time and rise time Smart card + + ASV 5 adaptive servoventilation; EPAP 5 expiratory positive airway pressure; IPS 5 inspiratory pressure support; MV 5 minute ventilation; PIF 5 peak inspiratory flow; SRBD 5 sleep-related breathing disorder. + Advanced variant, monitors peak inspiratory flow using a pneumotachograph in a 4 min moving window. 8 In theory, this peak flow should correlate with tidal volume and, when multiplied by breathing rate, MV. Two primary statistics are computed that set the target ventilation limits of the controller ( Fig 6 ). The low ventilation limit is derived by computing 95% of the moving window mean peak inspiratory flow value and is used in the absence of SRBD. In the presence of SRBD, the high limit tracks a value 60% above the moving window mean peak inspiratory flow. The algorithm determines these values repeatedly over time; IPS increases when measured flow is below the target and IPS is decreased when flow is above the target. The general form of the servo control algorithm for changing IPS for any given breath is: IPS 5 IPS of the previous breath 1 gain 3 (target peak inspiratory flow 2 current breath s peak inspiratory flow). Gain is a variable that converts flow into pressure and is computed based on Figure 6 BiPAP autosv devices monitor a 4-min moving window of peak flow values. This histogram illustrates the relationship between the mean value (point A), 95% of the mean value (point B), and the 60% value (point C). Points B and C define the range of peak flows delivered by the device. In the presence of sleep-related breathing disorder (SRBD), the target value will act to regain ventilatory stability by temporarily ramping to the high target (point C) and after 30 s will ramp to the lower value of point B L/min. recent patient data. The BiPAP autosv Advanced algorithm is capable of withdrawing IPS entirely, 8 a feature shared with the current ResMed VPAP Adapt, and may benefit some patients during parts of the night when breathing is normal and no support is needed (eg, during rapid-eye-movement sleep in patients with heart failure and CSA/HCSB). Maximum IPS levels are 25 cm H 2 O minus the prevailing EPAP. To determine EPAP, BiPAP autosv Advanced analyzes the airflow signal to assess airway patency in a manner similar to Philips current autotitrating CPAP devices. Obstructive apneas prompt progressive EPAP increases by increments of 1 cm H 2 O over a 10-s period. To avoid overshooting, two apneas, hypopneas, or a combination thereof must occur for EPAP to increase. Apnea is recognized when inspiratory peak airflow decreases 80%; hypopnea is recognized when airflow decreases 40% but < 80%. Obstructed vs open airway apneas are distinguished based on the airflow response to a machine-triggered breath. If peak flow decreases 80% and there is no airflow measured in response to a mandatory breath, the event is characterized as an obstructive apnea and EPAP increases. If airflow is detected in response to a mandatory breath, the event is classified as an open airway apnea and only IPS is increased. In the case of hypopnea, when there is a decrease in peak flow by 40%, IPS increases. However, when two hypopneas within a 3-min window are not corrected by IPS, EPAP increases 1 cm H 2 O. With stable breathing, optimal EPAP is determined by proactive searching which occurs interspersed within periods of fixed EPAP using similar algorithms employed by Philips Auto CPAP. Proactive searching includes pressure increases to identify potential improvement in flow morphology and pressure decreases to ensure that minimal efficacious pressure is delivered. 520 Contemporary Reviews in Sleep Medicine [ 146#2 CHEST AUGUST 2014 ]

8 Proactive searches complete both positive and negative searches in approximately 25 min. If evidence of airflow limitation or other obstructive event (apnea, hypopnea, snoring) occurs, EPAP will increase automatically at least 1 cm H 2 O every 1 min to 3 min depending on the type of obstruction. Automatic adjustment of EPAP is confined from 4 cm H 2 O to 25 cm H 2 O, unless set otherwise. The backup rate may be set to two different modes: auto or fixed rate. If auto mode is enabled, the device synchronizes with the patient s intrinsic rate. Monitoring a moving window of the last 12 spontaneous breaths, it computes two sets of parameters: the breath period and exhalation time. These are trended, and timing thresholds are computed as fractions of these periods. Timing of subsequent breaths is compared with these thresholds, and mandatory breaths are delivered if spontaneous breaths do not occur within the calculated parameters, typically, three to five breaths below the prevailing native breathing rate. A minimum breathing rate is enforced which begins at eight breaths/min and ramps upward to 10 breaths/min in the continued absence of spontaneous breathing. Breathing rate can also be set at a fixed level by disengaging the auto mode and specifying the desired rate. The BiPAP autosv Advanced incorporates Bi-Flex, which if enabled inserts expiratory pressure relief. Levels of one, two, or three progressively engage pressure drops proportional to the patient s airflow. This decrease in early EPAP still maintains maximum IPS ( Fig 7 ) but decreases mean airway pressure and may enhance hemodynamics and comfort. End-expiratory pressure is not affected by this decrease in early EPAP. Another feature of BiPAP autosv Advanced is rise time, available when Bi-Flex is not enabled. Rise time is the period of time in milliseconds for the pressure to increase 67% of difference from end EPAP to maximum inspiratory pressure. Following an exponential response, complete pressure increase occurs in roughly three times this duration. Zero is the fastest rise time which progressively prolongs as the level increases from one to three. Rise time should be adjusted to provide the most comfortable setting for the patient. Meanwhile, a prolonged rise time (and inspiratory time) improves flow distribution. The default is 300 ms. A ramp function reduces EPAP to a desired level (default is 4 cm H 2 O), which will gradually increase to the prescribed pressure setting (minimum EPAP or the fixed value of EPAP). During the period of ramping, Figure 7 Pressure levels vary throughout the breathing cycle. IPAP is applied during the whole inhalation period. It is reduced to the lowest level during early expiration (early EPAP). The difference between IPAP and early EPAP (IPAP-Early-EPAP) represents the degree of IPS. The device varies the pressure level at the end of expiration EEP in a manner similar to that of autotitrating CPAP devices to overcome upper airway obstruction. EX 5 expiration; IN 5 inspiration; IPAP 5 inspiratory positive airway pressure. See Figure 1 legend for expansion of other abbreviations. only minimum IPS is applied. However, importantly in this device, if obstructed events occur while ramping, the EPAP will increase more quickly than otherwise. To guide therapy, a smart card provides much information including AHI derived from airflow ( Table 1 ), central and obstructive AI, adherence data, and leak (e-appendix 1 ). Operation of the SomnoVent CR The anticyclic modulated ventilation of the SOMNOvent CR (Weinmann Medical Technology) is regulated based on flow monitored by a pneumotachograph and integrated to compute current MV. As usual, IPS equals IPAP minus EPAP. IPS automatically varies to counterbalance changes in MV while EPAP automatically varies to eliminate obstructive events ( Table 1 ). Unless set to a fixed value, the SOMNO vent CR adjusts EPAP in a manner similar to that of other autotitrating PAP devices. 9,26 In the case of upper airway obstruction, EPAP automatically increases based on the cumulative sum of obstructions within a 2-min epoch. If needed, EPAP can be increased manually up to about 17 cm H 2 O. journal.publications.chestnet.org 521

9 IPS increases with hypopneas and decreases with hyperventilation ( Fig 3 ). To regulate IPS, average MV is calculated by a low-pass filter every 2 min throughout the night, in a moving window giving 50% weight to the preceding 2 min and a weight of 50% to an earlier interval. For each current breath, actual MV is related to the average MV as a percentage to calculate relative MV of the current breath. IPS reacts anticyclically ( Fig 3 ), with a decrease in IPS when relative MV increases and increasing support with falls in relative MV. Additionally, the device contains a stability-targeted, proactive anticyclic regulation algorithm which monitors any waxing or waning trend in MV. SOMNO vent CR attempts to predict future values of MV, forecasting hypopneic or apneic phases even when MV starts decreasing but is still above threshold values. Thus, IPS can proactively increase, resulting in IPS regulation that is intended to occur earlier and at a more gentle rate, perhaps enhancing patient comfort and adherence. The maximum IPAP this device can produce is 20 cm H 2 O, and IPS can go no higher than the difference between current EPAP and maximum IPAP. Minimum IPS is always 0 cm H 2 O, unless set otherwise; this allows anticyclic regulation even in the presence of overall hyperventilation. SOMNOvent CR also provides a trilevel pressure profile ( Fig 7 ) wherein expiratory pressure is varied between a lower level at the first part of expiration (early EPAP) and a higher level at the end of expiration. As with the Philips device, this capability may allow for a lower overall mean airway pressure level that may enhance patient comfort and preserve hemodynamic function in patients with CHF. Current average spontaneous breathing rate is calculated by a low-pass filter giving the highest weight to the last breath and decreasing weights to of the previous two breaths. The automatic backup rate generates a first mandatory breath at 80% of the average spontaneous breathing rate but will go no lower than eight breaths/min. During apneas, mandatory breaths are applied automatically. The frequency of these breaths depends on the patient s baseline breathing rate, or the operator can manually choose a fixed breath rate. IPS is constrained to at least 5 cm H 2 O for mandatory breaths to maintain sufficient tidal ventilation. If ventilation is successful (at least 80% relative minute volume), the automatic backup rate is decreased by 5% to allow for the resumption of spontaneous breathing. Alternatively, a fixed backup rate can be used. As mask characteristics are detected automatically, interfaces of all manufacturers can be used without having to change any device settings. To guide therapy, SOMNO vent CR has download availability, providing much information including AHI, AI, hypopneas, intentional leak ( e-appendix 1 ), and adherence. In summary, there are three ASV devices, two of which are available in the United States and all of which use algorithms and other technology that vary from one to the others. It is critical that sleep physicians and technologists be aware of these differences so that these unique modalities may be applied to the best benefit of patients. In the second of these reports, we will expand on these differences, advise on the best available techniques for determining proper settings, and compare and contrast the results of using this technology in various patient groups as reported in the literature. Acknowledgments Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Javaheri has received research grants from Philips Respironics, Inc (Koninklijke Philips N.V.) and honoraria for lectures from ResMed and Philips Respironics, Inc (Koninklijke Philips N.V.). Dr Brown has chaired, co-chaired, and continues as a member of the Polysomnography Practice Advisory Committee of the New Mexico Medical Board and serves on the New Mexico Respiratory Care Advisory Board. He currently receives no grant or commercial funding pertinent to the subject of this article. Dr Randerath has received fees for speaking and research funds from companies producing positive airway pressure devices including Weinmann Medical Technology (Weinmann Geräte für Medizin GmbH + Co. KG), Philips Respironics, Inc (Koninklijke Philips N.V.), and ResMed companies. Additional information: The e-appendix can be found in the Supplemental Materials section of the online article. References 1. Harris N, Javaheri S. Advanced PAP therapies. In: Mattice C, Brooks R, Lee-Chiong T, eds. Fundamentals of Sleep Technology. Philadelphia, PA : Lippincott, Williams and Wilkins ; 2012 : Su l liv an CE, Iss a FG, B er t hon-jones M, Eve s L. R e ve rs a l of obstructive sleep apnoea by continuous positive airway pressure applied through the nares. Lancet ;1(8225): Jav a he r i S. Effects of continuous positive airway pressure on sleep apnea and ventricular irritability in patients with heart failure. Circulation ;101(4): Arzt M, Floras JS, Logan AG, et al ; CANPAP Investigators. Suppression of central sleep apnea by continuous positive airway pressure and transplant-free survival in heart failure: a post hoc analysis of the Canadian Continuous Positive Airway Pressure for Patients with Central Sleep Apnea and Heart Failure Trial (CANPAP). Circulation ;115(25): Javaheri S. CPAP should not be used for central sleep apnea in congestive heart failure patients. J Clin Sleep Med ;2(4): Jav a he r i S, Ma li k A, Smit h J, C hung E. Ad apt ive pre ssu re supp or t servoventilation: a novel treatment for sleep apnea associated with use of opioids. J Clin Sleep Med ;4(4): Brown LK. Adaptive servo-ventilation for sleep apnea: technology, titration protocols, and treatment efficacy. Sleep Med Clin ;5(3): Contemporary Reviews in Sleep Medicine [ 146#2 CHEST AUGUST 2014 ]

10 8. Javaheri S, Goetting MG, Khayat R, Wylie PE, Goodwin JL, Par t has ar at hy S. The performance of two automatic servoventilation devices in the treatment of central sleep apnea. Sleep ;34 (12 ): Randerath WJ, Galetke W, Stieglitz S, Laumanns C, Schäfer T. Adaptive servo-ventilation in patients with coexisting obstructive sleep apnoea/hypopnoea and Cheyne-Stokes respiration. Sleep Med ;9(8): Javaheri S. Heart failure. In: Kryger MH, Roth T, Dement WC, Saunders WB, eds. Principles and Practices of Sleep Medicine. 5th ed. Philadelphia, PA : 2011 : Javaheri S, Dempsey JA. Central sleep apnea. Compr Physiol ;3(1): Javaheri S, Parker TJ, Wexler L, et al. Occult sleep-disordered breathing in stable congestive heart failure. Ann Intern Med ;122(7): Javaheri S, Parker TJ, Liming JD, et al. Sleep apnea in 81 ambulatory male patients with stable heart failure. Types and their prevalences, consequences, and presentations. Circulation ;97(21): Javaheri S. Sleep disorders in systolic heart failure: a prospective study of 100 male patients. The final report. Int J Cardiol ;106(1): Sin DD, Fitzgerald F, Parker JD, et al. Risk factors for central and obstructive sleep apnea in 450 men and women with congestive heart failure. Am J Respir Crit Care Med ;160(4): Vazir A, Hastings PC, Dayer M, et al. A high prevalence of sleep disordered breathing in men with mild symptomatic chronic heart failure due to left ventricular systolic dysfunction. Eur J Heart Fail ;9(3): Zhao ZH, Sullivan C, Liu ZH, et al. Prevalence and clinical characteristics of sleep apnea in Chinese patients with heart failure. Int J Cardiol ;118(1): Christ M, Sharkova Y, Fenske H, et al. Brain natriuretic peptide for prediction of Cheyne-Stokes respiration in heart failure patients. Int J Cardiol ;116 (1 ): Oldenburg O, Lamp B, Faber L, Teschler H, Horstkotte D, Töpfer V. Sleep-disordered breathing in patients with symptomatic heart failure: a contemporary study of prevalence in and characteristics of 700 patients. Eur J Heart Fail ;9(3): MacDonald M, Fang J, Pittman SD, White DP, Malhotra A. The current prevalence of sleep disordered breathing in congestive heart failure patients treated with beta-blockers. J Clin Sleep Med ;4(1): Schulz R, Blau A, Börgel J, et al ; Working Group Kreislauf und Schlaf of the German Sleep Society (DGSM). Sleep apnoea in heart failure. Eur Respir J ;29(6): Tremel F, Pépin J-L, Veale D, et al. High prevalence and persistence of sleep apnoea in patients referred for acute left ventricular failure and medically treated over 2 months. Eur Heart J ;20(16): Yumino D, Redolfi S, Ruttanaumpawan P, et al. Nocturnal rostral fluid shift: a unifying concept for the pathogenesis of obstructive and central sleep apnea in men with heart failure. Circulation ;121(14): Teschler H, Döhring J, Wang YM, Berthon-Jones M. Adaptive pressure support servo-ventilation: a novel treatment for Cheyne- Stokes respiration in heart failure. Am J Respir Crit Care Med ;164(4): Morganthaler TI. The quest for stability in an unstable world: Adaptive Servoventilation in opioid induced complex sleep apnea syndrome. J Clin Sleep Med ;4(4): Randerath WJ, Galetke W, Kenter M, Richter K, Schäfer T. Combined adaptive servo-ventilation and automatic positive airway pressure (anticyclic modulated ventilation) in co-existing obstructive and central sleep apnea syndrome and periodic breathing. Sleep Med ;10(8): journal.publications.chestnet.org 523