Live Recordings (Sleep Medicine Trends 2026)-19 APAP

Video Transcription

Video Summary

The speaker reviews the “good, bad, and ugly” of Auto-CPAP (APAP). APAP can effectively treat many patients, sometimes with faster access and lower mean pressures, but some patients remain undertreated—often without clinicians realizing it—because device-reported residual AHI is only an estimate and can be inaccurate. Devices rely mainly on airflow signals (no EEG sleep/wake, no oximetry), so they may over-score events during wake or leak, and under-score events, especially in REM where breathing is irregular. Classification of apneas as “clear airway” vs “obstructive/closed airway” is imperfect because devices don’t measure respiratory effort.<br /><br />Algorithms differ substantially across manufacturers (e.g., aggressiveness of pressure increases; behavior during leak). High leak can reduce accuracy and may even stop titration on some devices. The speaker argues against default wide ranges (4–20 cm H₂O) and recommends tailoring min/max pressures using P90/P95 and clinical issues (REM clusters, intolerance, treatment-emergent centrals).<br /><br />Because cloud summaries are limited, he highlights free tools (OSCAR) and alternatives (SleepHQ) that read SD-card waveform data and can integrate inexpensive oximetry, helping confirm whether residual events are real, identify REM-related clusters, leak patterns, and types of periodic breathing (not all is Cheyne–Stokes). If problems persist, consider switching to fixed CPAP or performing a targeted lab titration.

Keywords

Auto-CPAP (APAP)

residual AHI accuracy

airflow-based event detection limitations

REM-related under-treatment

leak effects on titration

manufacturer algorithm differences

pressure range optimization (P90/P95)

OSCAR/SleepHQ SD-card waveform analysis