ResMed ASV machine

Adaptive Servo Ventilation: ResMed vs. Respironics

A Table of Comparisons

by Lawrence Martin, MD

(Board certified in Pulmonary and Sleep Medicine)

Return to Home Page: CPAP, BiPAP, ASV

INTRODUCTION. Currently there are two manufacturers of ASV machines: ResMed and Philips Respironics, usually referred to simply as Respironics. The table below compares the two companies' machines along several aspects, from history to clinical indications to cost.

Click here for list of all abbreviations
Note: Mozilla Firefox and other browsers may not support tables format. This web site is best viewed with Internet Explorer.

Adaptive Servo Ventilation


ResMed ASV

Respironics AutoSV



For years there was just CPAP and BiPAP to treat sleep apnea patients. In 2001 a paper appeared about a novel treatment for one form of central sleep apnea: Teschler H, et al. Adaptive pressure support servo-ventilation: a novel treatment for Cheyne-Stokes respiration in heart failure. Am J Respir Crit Care Med 2001;164:614-9. The study used ResMed Corporation's new ASV machine to deliver variable pressures to patients with Cheyne-Stokes respiration -- the waxing and waning breathing often seen during sleep in heart failure patients. Soon the machine was used for treating all types of central sleep apnea as well as complex sleep apnea, a condition not recognized until after the machine was already in use.

"ASV" as a label and a particular methodology to adjust pressures was patented by ResMed. A few years later, another large equipment manufacturer, Philips Respironics, came out with their own ASV-type machine; they call it "autoSV." Auto SV is also an adaptive servo ventilator device that uses a different algorithm to adjust bilevel pressures. A newer model is called Auto-SV Advanced (discussed below).

ASV is but a way station on the road to newer and smarter machines. Expense (about $6000 per machine) limits their wider use, but no doubt these machines will become cheaper and at some point replace our basic CPAP/BiPAP units.


The basic difference between the two devices is the way the patient's breathing is tracked so that IPAP and/or EPAP can be adjusted. Per ResMed's fact sheet, the algorithm for ResMed's VPAP Adapt SV (as their machine is called):

"uses three factors to achieve synchronization between pressure support and the patient's breathing.
1. The patient's own recent average respiratory rate, including the ratio of inspiration to expiration and the length of any expiratory pause.
2. The instantaneous direction, magnitude, and rate of change of the patient's airflow, which are measured at a series of set points during each breath.
3. A backup respiratory rate of 15 breaths per minute."

In other words, the algorithm is proprietary and unique to ResMed. To highlight this point, below is a quote about the ResMed algorithm from Brown LK. Whither adaptive servo-ventilation? Current Opinion in Pulmonary Medicine 2010;16:527-29:

"The [ResMed] VPAP Adapt SV relies on a complicated estimate of recent average ventilation using mask pressure and total airflow to obtain target minute ventilation and determines respiratory phase using an estimate of intantateous respiratory airflow (and may also utilize 'fuzzy' catergories of resipratory phase). Furthermore, a host of internal constants and parameters govern the microprocessor's decisions with respect to pressure changes and cycling times, none of which are user-adjustable or explicitly stated."

The Philips Respironics FAQ states that the device tracks flow. "The flow signal is analyzed and a target flow is calculated. If the patient reaches the flow target, the device does not offer any additional pressure support. If the patient does not reach the flow target the device will dynamically change pressure support breath to breath." Omitted from this FAQ is that the device tracks peak flow over a 4-minute cycle, as shown below.
Respironics peak flow algorithm A peak flow target is established around the 4-minute average and the machine changes the air delivery as needed, to deliver 95% of the target, as shown below.
Respironics flow adjusted

If these graphs make the Respironics algorithm seem simpler than the ResMed model, Dr. Brown assures us it is not (Brown LK. Whither adaptive servo-ventilation? Current Opinion in Pulmonary Medicine 2010;16:527-29). He writes: "...this device's algorithm employs an even geater number of internal parameters [than ResMed's machine] that are not explicitly disclosed and not adjustable. ...The [Respironics] Advanced model adds automatic titration of expiratory positive airway pressure to the generator's capabilities, utilizing an algorithm borrowed from their auto-CPAP machines."

Despite the fact that the proprietary algorithms are rather opaque, both machines seem to work well in their stated objectives of treating central sleep apnea and complex sleep apnea. As such, they compete in the market for treatment of these conditions. And while each company claims its methodology is superior to the other, there is no head-to-head study. From the standpoint of the sleep medicine specialist, neither machine is clearly superior or inferior. Since ResMed was first out with an ASV machine, most of the articles in the medical literature are based on ResMed's device, but there are now several publications for autoSV as well. More importantly, this is a rapidly evolving field, and new machines will likely appear soon, from these two companies as well as other competitors.

Pressure Curves

ASV is a form of bilevel ventilation or "BiPAP". Click here to review BiPAP The difference is that ASV varies IPAP and/or EPAP to adjust ventilation as needed by the patient. With traditional BiPAP the pressures are set and don't vary. With ResMed's ASV the EPAP is fixed and the IPAP varies depending on the respiratory rate and air flow, measured continuously.

Below is example of ASV delivered from ResMed's ASV machine. The figure is from a ResMed brochure, which explains the graph as follows: "As breathing resumes and total ventilation exceeds the target,
- pressure support is rapidly reduced back towards the minimum 3 cm H2O, and
- this reduces the likelihood of over-ventilation and hypocapnia, which can lead to vocal cord closure and further apneas."
ResMed ASV pressure curve
Click here for larger image, then back space to return to table. In this figure the top graph is pressure and bottom is flow. Note that EPAP -- end-expiratory pressure -- is set at 5 cm H2O. The IPAP -- inspiratory positive airway pressure -- is variable. Initially it is 3 cm H2O, or 'pressure support 3', sometimes labeled PS 3. When central apnea ensues the machine senses less air flow and ratchets up the IPAP, eventually reaching the pre-set limit of 15 cm H2O. The prescription for this patient would read: "IPAP 5, PS 3-10."

Below is portion of a polysomnogram showing treatment of Cheyne-Stokes breathing with ASV. Figure is from: Banno K, Okamura K, Kryger MH: Adaptive Servo-Ventilation in patients with idiopathic Cheyne-Stokes breathing. J Clinical Sleep Medicine 2006; Vol 2, pages 181-186.

ASV on Polysomnogram
Click here for full size image, then back space to return to table.

ASV is a form of bilevel ventilation or "BiPAP". Click here to review BiPAP The difference is that ASV varies IPAP and/or EPAP to adjust ventilation as needed by the patient. With traditional BiPAP the pressures are set and don't vary. With Respironics AutoSV, the IPAP and/or EPAP vary depending on a rolling average of the patient's peak air flow (see Methodology, above).

Respironics currently has two machines: AutoSV and Advanced AutoSV. With regular AutoSV the EPAP is fixed and the IPAP varies. With Advanced AutoSV both EPAP and IPAP can vary. Below is the pressure curve from Repironics regular AutoSV machine, taken from a Respironics bronchure.

CPAP pressure curve
Click here for larger view, then back space to return to table.

This is a rapidly evolving technology and new machines will likely come on the market soon. At this writing there is no head-to-head comparison of machines, and neither methodology has been proved superior.

Principal Indications

There are two principal indications:
1) Central sleep apnea
2) Complex sleep apnea

1) Central sleep apnea can basically be divided into Cheyne-Stokes (C-S) and non-Cheyne Stokes breathing. C-S breathing pattern - waxing and waning respirations punctuated by periods of no respiratory effort and no air flow - is usually seen in patients with congestive heart failure or cerebrovascular disease. Non C-S central sleep apnea does not show the waxing and waning respirations, just abrupt cessation of all respiratory effort and air flow - is seen in patients taking chronic opiates, in patients with no obvious cause, and as the manifestation of Complex Sleep Apnea.

2) Complex Sleep Apnea is diagnosed when a patient with obstructive sleep apnea is treated with CPAP or BiPAP, and the treatment itself elicits central apneas.

1) Central sleep apnea
2) Complex sleep apnea

See comment under ResMed

If complex sleep apnea ("CompSA") is a result of CPAP, BiPAP may be tried, though it is not usually successful. When CompSA doesn't respond to BiPAP, or is a result of BiPAP in the first place, there are two choices:
a) Continue CPAP or BiPAP to treat the OSA, hoping that the central apneas will remit over time (they sometimes do);
b) Start the patient on ASV.

There is no clear standard about when to start ASV on patients who manifest CompSA.

Clinical Studies

The first clinical article on ASV appeared in 2001:
Teschler H, et al. Adaptive pressure support servo-ventilation: a novel treatment for Cheyne-Stokes respiration in heart failure. Am J Respir Crit Care Med 2001;164:614-9.

Over the next decade more articles appeared showing that ResMed's ASV could treat not only Cheyne-Stokes respiration, but a syndrome that wasn't even recognized in 2001: Complex Sleep Apnea, which is the intrusion of central sleep apneas when some patients are treated for OSA with CPAP or BiPAP. For later reviews and studies with ResMed's ASV device, see:

Allam JS, et al. Efficacy of servoventilation in treatment of complex and central sleep apnea syndromes. Chest 2007 Dec;132:1839-46.

Banno K, Okamura K, Kryger MH: Adaptive Servo-Ventilation in patients with idiopathic Cheyne-Stokes breathing. J Clinical Sleep Medicine 2006; Vol 2, pages 181-186.

Morgenthaler TI, et. al. Adaptive Servoventilation Versus Noninvasive Positive Pressure Ventilation for Central, Mixed, and Complex Sleep Apnea Syndromes. SLEEP, Vol. 30, No. 4, pages 468-475, 2007.

Beneficial effects of adaptive servo ventilation in patients with chronic heart failure. Circ J Jul 29, 2010.

For studies with Respironics' AutoSV, see:

Artz M, et al. Effects of dynamic bilevel positive airway pressure support on central sleep apnea in men with heart failure. Chest 2008 July;134:61-6.

Randerath WJ, et al. Adaptive servo-ventilation in patients with coexisting sleep apnoea/ hypopnoea and Cheyne-Stokes respiration. Sleep Med 2008 Dec;9:823-30

Again, there are no head-to-head studies. Related articles of interest include:

Eckert DJ, at al. Central sleep apnea: pathophysiology and treatment. Chest. 2007;131:595-607

Gay PC. Complex sleep apnea: it really is a disease. J Clin Sleep Medicine 2008 Oct 15;4:403-05.

Kuzniar TJ, Morgenthaler TI. Treatment of complex sleep apnea syndrome. Curr Treat Options Neurol 2008;Sept 10;336-41.

Javaheri S, et al. The prevalence and natural history of complex sleep apnea. J Clin Sleep Med 2009 June 15;5:212-4.

Brown LK. Whither adaptive servo-ventilation? Current Opinion in Pulmonary Medicine 2010;16:527-29.

Brown SE, et al. A Retrospective Case Series of Adaptive Servoventilation for Complex Sleep Apnea. Journal of Clinical Sleep Medicine, 2011;7:187-195.

What kind of mask is used to deliver ASV?

ResMed recommends their full face mask. Shown below is the ResMed Quattro Pro full face mask.
positive pressure mask

Respironics states any mask used for CPAP should work with their autoSV machine. Below is the Respironics Comfort Gel full face mask.
positive pressure mask

Mask type and size is best determined in the sleep lab, when a patient is being titrated on an ASV machine.

Prescription Examples

ResMed VPAP Adapt SV:
  • Backup rate: 12 breaths/minute
  • Rise time: (the time it takes the device to change from EPAP to IPAP. Settings range from 1 [fastest] to 6 [slowest]. The range of values corresponds to 1/10 seconds, so that a rise time of 4 = 0.4 seconds)
  • Inspiratory time: 2 seconds (range is 0.5 to 3.0 seconds)
  • EPAP [expiratory positive airway pressure] = 7.0 cm H2O
  • Minimum PS [pressure support] = 4.0 cm H2O
  • Maximum PS = 13.0 cm H2O
  • Maximum pressure for device (generally <=30 cm H2O)
  • Mask: ResMed Quattro Full Face Mask (recommended by company for their ASV machine), with heated humidity
Below is the ResMed ASV machine.
ResMed ASV machine

Respironics currently has TWO devices on the market, the regular AutoSV and the AutoSV Advanced. Respironics recommends that physcians now prescribe ONLY the AutoSV Advanced, so it seems likely that the regular machine will be phased out soon. Below are prescription examples for both.

Regular AutoSV
For the regular AutoSv you can use the same prescription example as listed under ResMed. A detailed approach to prescribing is provided in the Respironics pdf file.
Below is the Respironics auto SV machine. Respironics-auto-sv

Prescribing AutoSV Advanced
The same settings as for the regular Respironics AutoSV and the ResMed ASV machine, with one important addition. Respironics autoSV Advanced allows the EPAP to vary, so you have to specify a minimum and maximum EPAP. For example, to the settings for Respironics regular autoSV you would add:

  • ...
  • EPAP-min = 7 cm H2O
  • EPAP-max = 12 cm H2O
  • ...
For more information see autoSV Advanced.

The technology is evolving and this section should change as new machines (and methods of delivering ASV) come on the market.

Medicare Coverage Guidelines

(To some extent these guidelines also influence most commercial insurance coverage)

The following information is from the CMS web site on Respiratory Assist Devices

for Central sleep apnea, defined as:
1. Study showing AHI > 5 events/hr. and
2. Central apneas/hypopneas greater than > 50% of the total apneas/hypopneas, and
3. Central apneas or hypopneas >=5/hr., and
4. Symptoms of either excessive sleepiness or disrupted sleep.

for Complex sleep apnea
CompSA is a form of central sleep apnea specifically identified by the persistence or emergence of central apneas or hypopneas upon exposure to CPAP or an EO470 [Bi-level] device when obstructive events have disappeared. These patients have predominantly obstructive or mixed apenas during the diagnostic sleep study occurring at >= 5 times per hour. With use of a CPAP or EO470 [Bi-level], they show a pattern of apneas and hypopneas that meets the definition of CSA described above.

Same criteria as for ResMed

NOTE: CMS defines an hypopnea as "an abnormal respiratory event lasting at least 10 seconds associated with at least a 30% reduction in thoracoabdominal movement or airflow as compared to baseline, and with at least a 4% decrease in oxygen saturaiton." This definition makes the CMS criteria for central sleep apnea and CompSA confusing, because hypopneas are not traditionally scored as obstructive or central, but simply 'hypopneas'. Thus by CMS criteria a patient could have persistently high AHI from obstructive hypopneas after a CPAP/BiPAP titration, yet qualify for an ASV machine. One doubts that was the intent of the stated criteria.

Perhaps even more amazing is that all the DME companies quote the CMS guidelines, and never stop to explain the role of hypopneas in defining central sleep apnea. For example, see the Respironics brochure that details how to prescribe their auto SV unit.

Web sites for more information


ResMed technical fact sheet

ResMed VPAP Adapt SV

Complex sleep apnea (ResMed)

ResMed Tutorial on complex sleep apnea (with audio)


Frequently asked questions

How AutoSV works

How to prescribe autoSV

Please refer to 'Clinical Studies' for discussion of ASV in general.


ResMed's VPAP Adapt SV is not sold new on the internet, but is probably the same price as Respironics AutoSV machine, which is $5800 at In addition there is the cost of the mask, and also the sleep studies needed before any machine can be prescribed.

At the Remstar Auto SV is $5800 for the machine alone. In addition there is the cost of the mask, and also the sleep studies needed before any machine can be prescribed.

There are many web sites for purchasing CPAP machines, including:, CPAP-Supply and The CPAP Shop.

In clinical practice patients usually get their machine from a DME (durable medical equipment) company, following a physician's prescription. The DME cost of positive pressure devices is generally higher than the cost when purchased on the internet, because the DME companies provide service to the machine and training to the patient. Even so, DME companies will generally accept whatever the insurer pays, as long as it covers their cost. The insurer will sometimes rent the machine from the DME company, sometimes buy it outright, and there may or may not be a deductible to the patient. Thus there is no simple answer.

ABBREVIATIONS (alphabetical order)
ASV - adaptive servo ventilation
autoSV - automatic servo ventilation
BiPAP - bilevel positive airway pressure
cm H2O - centimeters of water
COPD - chronic obstructive pulmonary disease
CPAP - continuous positive airway pressure
CSA - central sleep apnea
E - expiration or expiratory
EEP - end-expiratory pressure
EPAP - expiratory positive airway pressure
I - inspiration or inspiratory
IPAP -inspiratory positive airway pressure
max - maximum
min - minimum
NIV - non-invasive ventilation
OSA - obstructive sleep apnea
PEEP - positive end-expiratory pressure
PS - pressure support
PSV - pressure support ventilaton
S - spontaneous
S/T - spontaneous/timed
SV - servo ventilation
VPAP - variable positive airway pressure

Return to Table: ResMed vs. Respironics
Return to Home Page: CPAP, BiPAP, ASV
Return to Lakeside Press home page

Forward any comments to:

Copyright © 2010 Lawrence Martin
First posted August 30, 2010
Last revised May 2, 2011