Expiratory pressure relief undoes some of what CPAP is for

Design

four-step study: randomised-order in-laboratory titration with and without the algorithm, followed by home periods on fixed and automatic CPAP

Population

29 adults with severe obstructive sleep apnoea on nasal CPAP; mean age 63, BMI 37, median AHI 58 events/h

Primary outcome

upper airway patency (peak inspiratory flow), titrated CPAP level, residual apnoea-hypopnoea index and adherence

Effect

peak inspiratory flow 0.19 vs 0.38 L/s (p<0.001); titrated pressure 11.40 vs 9.17 cmH2O (p<0.001); residual AHI on fixed CPAP 3.7 vs 1.3 events/h; adherence unchanged at about 7 hours

Expiratory pressure alleviation algorithms drop the delivered pressure during expiration to make CPAP more comfortable, and they are switched on by default on many devices. The physiological objection has always been obvious: the pharynx narrows and can collapse during expiration, so lowering pressure then is lowering it at the moment it may be needed. This study tested that in 29 adults with severe obstructive sleep apnoea — median apnoea-hypopnoea index 58 events per hour — across four steps, two in the sleep laboratory and two at home.

With the algorithm on, peak inspiratory flow during flow limitation halved, from 0.38 to 0.19 L/s. Manually titrated therapeutic pressure rose from 9.17 to 11.40 cmH2O — so the device needs over 2 cmH2O more to achieve the same control, which is a direct cost to the comfort it was meant to buy. Residual apnoea-hypopnoea index was higher with the algorithm on both fixed CPAP (3.7 versus 1.3 events per hour) and automatic CPAP (1.4 versus 0.8).

And adherence, the entire justification, did not change: about seven hours a night either way, in both modes. That is the finding that matters. A comfort feature that costs pressure and control and buys no extra use is not a trade-off — it is a loss.

• Check whether expiratory pressure relief is enabled in any patient with a higher-than-expected residual apnoea-hypopnoea index on download.
• If it is on during titration, the pressure you prescribe will be higher than the patient actually needs with it off.
• Adherence was already excellent in this group at seven hours; the result may differ in patients who are struggling to tolerate therapy at all.
• Residual events of 3.7 per hour is still technically controlled disease — the effect is on the margin, not a treatment failure.
• Auto-CPAP partly compensated by raising pressure, which is why the residual difference was smaller in that mode.

Why it matters

The feature is enabled by default on many machines and nobody looks at it when residual events are high.

Don't overread it

These were adherent patients with severe disease — this does not establish that the feature never helps someone who cannot tolerate CPAP otherwise.

The statistics, in plain English

Twenty-nine patients is small, but this is a within-patient crossover design where each person is their own control, which gives far more precision per participant than a parallel trial. The physiological result — halved peak inspiratory flow, p<0.001 — is large and mechanistically coherent, which is the strongest part. The adherence comparison is the weakest: with 29 patients and means around seven hours, a difference of half an hour would not have been detected, so 'no effect on adherence' means 'no large effect'.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.