Daridorexant did not worsen breathing in severe untreated sleep apnoea

Design

randomised, double-blind, placebo-controlled, two-period crossover trial at a single sleep centre

Population

16 adults with severe obstructive sleep apnoea without insomnia, mean baseline AHI 51.2 events per hour

Primary outcome

treatment difference in apnoea-hypopnoea index, with nocturnal oxygen saturation secondary

Effect

AHI difference -3.7 events per hour (one-sided 95% CI up to +4.2); saturation -0.12% (one-sided 95% CI to -0.6); total sleep time +32.5 minutes (90% CI 6.9-58.2)

A randomised, double-blind, placebo-controlled two-period crossover trial at a single sleep centre gave 16 adults with severe obstructive sleep apnoea and no insomnia either daridorexant 50 mg - the maximum therapeutic dose - or placebo each evening for five nights. Mean baseline apnoea-hypopnoea index was 51.2 events per hour and mean oxygen saturation during sleep 92.1%.

The prespecified thresholds for a clinically meaningful harm were a rise in AHI of 10 events per hour or more, or a fall in nocturnal saturation of 2% or more. Neither was reached: the treatment difference in AHI was -3.7 events per hour (one-sided 95% CI up to +4.2) and in saturation -0.12% (one-sided 95% CI down to -0.6). Daridorexant increased total sleep time by 32.5 minutes (90% CI 6.9 to 58.2), shortened latency to persistent sleep by 10.3 minutes, and showed a non-significant reduction in wake after sleep onset. Four adverse events occurred, all mild, none respiratory.

The clinical problem this addresses is a real and common one: 30 to 50% of patients with sleep apnoea also have insomnia symptoms, and prescribers have avoided hypnotics in them on the reasonable assumption that sedation worsens obstruction. That assumption has been the default rather than a finding. This trial does not overturn it for sedative hypnotics as a class - benzodiazepines and Z-drugs are pharmacologically different - but for a dual orexin receptor antagonist at full dose over five nights in severe disease, breathing did not deteriorate.

• Treat this as reassurance about orexin antagonists specifically, not about hypnotics generally
• Do not extrapolate to benzodiazepines or Z-drugs, which were not studied
• Continue to diagnose and treat the apnoea itself - this trial says nothing about CPAP adherence
• Note the duration: five nights, not chronic use, and 16 patients at one centre
• Screen for untreated sleep apnoea before prescribing any hypnotic, which this does not change

Why it matters

Insomnia coexists with sleep apnoea in a third to a half of patients, and has been left untreated on an assumption nobody had tested at full dose in severe disease.

Don't overread it

Sixteen patients, five nights, one centre, one drug class - not a general clearance for hypnotics in sleep apnoea.

The statistics, in plain English

The design here is a non-inferiority-style safety question, and it is reported as one-sided confidence intervals against prespecified harm thresholds - the AHI interval reaching only +4.2 events per hour excludes the 10-event rise that was defined as meaningful. That is the correct way to test for absence of harm, but the trial's size is the limit: 16 patients in a crossover design can exclude a large respiratory effect and nothing smaller, and cannot address what happens beyond five nights. The sleep benefits are reported with 90% rather than 95% intervals, a weaker standard, and were secondary.

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.