Residual Hypopneas at 10 cm H2O: Pressure Problem or Leak Problem?
A 58-year-old woman (BMI 33) with treated hypertension is undergoing in-laboratory CPAP titration after a diagnostic study showing an AHI of 44/hr and a nadir SpO2 of 82%. She is fitted with a nasal pillow mask. After 90 minutes she is asleep in supine REM at 10 cm H2O. The device reports a 95th-percentile leak of 52 L/min and a residual AHI of 16/hr, all hypopneas. Flow limitation is visible on the flow-time trace, and the SpO2 nadir is 85%. She wakes with a dry mouth and nasal burning; the humidifier is set at 3 of 5. Over the next 40 minutes the technologist increases pressure to 12 and then 14 cm H2O. Leak climbs to 58 L/min, residual AHI is 19/hr, and she awakens twice, once removing the mask. No prior titration exists and the order reads only "titrate to eliminate events."
1. Is the residual AHI of 16/hr at 10 cm H2O evidence that the pressure is inadequate? Explain what the data indicate and what must be corrected first.
Read the worked answer
No. The most defensible reading is that this is a leak problem masquerading as a pressure problem.
Why. A 95th-percentile leak of 52 L/min is well above the generally accepted threshold of about 24 L/min for a large, unacceptable leak. With a nasal pillow interface and complaints of dry mouth, the leak is most likely oral, meaning the delivered pharyngeal pressure is below the set pressure for a substantial fraction of the night. Under those conditions the device cannot distinguish hypopneas caused by insufficient pressure from hypopneas caused by pressure loss through the mouth. The rising leak and rising residual AHI as pressure increased (16 to 19/hr) is the key signal: if pressure were simply too low, events should fall, not rise. Higher pressure with a mouth leak often worsens leak, arousal, and event counts.
Counter-checking. Before concluding, confirm the leak value is not an artifact (disconnected tubing, poor sensor, mask seated incorrectly), confirm the epoch is truly supine REM and not wake, and look at whether the hypopneas cluster after arousals. Also consider positional worsening and nasal obstruction from the low humidifier setting.
Order of operations. The leak must be brought under control before any pressure interpretation is valid. Chasing events with pressure in the presence of a large leak produces an inappropriately high prescription and a patient who cannot tolerate therapy.
2. Give an ordered plan for the remaining 2 hours, with decision points at each step.
Read the worked answer
Step 1 - Return to 10 cm H2O and stop escalating. The higher pressures added arousal and leak without benefit; there is no justification for continued increases until the leak is addressed.
Step 2 - Fix the interface and humidification. Check pillow size and strap tension, inspect for nares irritation, and raise humidification to 4–5 and confirm heated tubing is on to limit condensation and nasal drying. Recheck leak for 15–20 minutes.
Step 3 - Address oral leak. If leak remains above 24 L/min with dry mouth, add a chin strap or switch to an oronasal mask. With nasal obstruction or a strong mouth-breathing pattern, the oronasal mask is the more reliable choice. Allow 20–30 minutes for leak to settle at 10 cm H2O with the patient supine and asleep.
Decision point A. If leak falls below 24 L/min and the residual AHI falls to 5/hr or less, the endpoint is met at 10 cm H2O and no further increase is needed.
Decision point B. If leak is controlled but obstructive hypopneas, flow limitation, or desaturations persist in supine REM, now increase pressure by 1 cm H2O at 5-minute intervals to a maximum of about 15 cm H2O, reassessing leak at each step.
Decision point C. If leak cannot be controlled with reasonable mask interventions, or the patient cannot sleep, document the technical limitation, prescribe the best-validated pressure with a plan for mask refitting, and consider home auto-titrating PAP with download review rather than an arbitrary pressure. Notify the ordering physician of the leak-limited study.
3. After switching to an oronasal mask, leak is 10 L/min and residual AHI is 4/hr at 11 cm H2O with a nadir of 90% and 8 minutes below 88%. Is this an acceptable titration endpoint? Discuss what you would record and how the prescription should be written.
Read the worked answer
Acceptability. Yes. A residual AHI of 4/hr is below the customary 5/hr target, the leak of 10 L/min is within acceptable limits, and the nadir of 90% is at the threshold generally considered adequate. Eight minutes below 88% is a brief, non-sustained desaturation and does not by itself justify nocturnal oxygen, particularly when it may have occurred during an arousal or before the final pressure was reached.
Confirm before finalizing. Verify that these values were collected in supine REM with the patient asleep, that the desaturation minutes did not all occur at a lower pressure, and that no significant flow limitation remains on the flow trace. If supine REM was not captured at 11 cm H2O, consider a modest increase (1 cm H2O) only if events or desaturation persisted there.
What to record. Prescribed pressure 11 cm H2O; interface type and size (oronasal); humidification setting; leak 10 L/min; residual AHI 4/hr; supine REM status; oxygen saturation profile including the 8 minutes below 88%; reason for change from nasal pillow to oronasal mask; the failed higher-pressure trial and why it was abandoned.
Prescription and follow-up. Write the order as fixed CPAP at 11 cm H2O with heated humidification and the specified mask. Recommend a data download at 2–4 weeks to confirm leak and residual AHI on the new interface, with clinical follow-up for mask comfort. If download shows leak above 24 L/min or residual AHI above 5/hr, a repeat titration or a switch to auto-titrating PAP at a defined range is preferable to an empirical pressure increase.