Accuracy of CPAP event classifications

[Sleeprider]

The central events you zoomed in on may or may not be central apnea. I have strong doubts, mostly because they are random events that do not show any of the more alarming characteristics of central apnea pattern.  These may be sleep transition, position changes or obstructive that leave the airway open enough to appear as clear airway apnea.  On the other hand, particularly at the end of the night, the events are clearly obstructive and accompanied by flow limitations with flat-topped inspiratory peaks.  This suggests the pressure is too low to keep the airway patent, and I would expect some notable drop in SpO2 will accompany these flow limitations.  I have seen this kind of breathing pattern respond to simple increases in pressure, and also to pressure support, where inhale pressure is maintained higher than exhale pressure.  

In the second graph I have copied from your post, you can see several apnea where there is a vertical mark near the apnea flag.  Philips machines do not use forced oscillation technique (FOT) to probe the airway like Resmed, but uses individual pressure pulses to determine if the airway is open or obstructed.  Those hash marks show where the pulses occur. During a truly central event, we usually see the respiration flow rate chart deflect when the pulse is applied. This shows the airway is open and the air pulse can pass into the deeper airway and rebound, while with an obstructed, the appearance is usually less visible. Most of the pressure pulses we see here have very little effect, and the shallow, irregular breathing is clearly obstructed and flow limited.  I don't think your son is experiencing a lot of central apnea, but it would be the task of a bilevel titration study to evaluate if the breathing volume can be restored through bilevel pressure support, or if it needs a backup rate to maintain the respiration rate.  I suspect the former.  The two ways this may fall out is that your son responds to higher CPAP pressure and breaths normally, or he ends up benefiting from inhale pressures higher than exhale pressure to maintain a normal breathing wave-form.  Only failing those two approaches would a device like ASV with a backup rate be considered.  First things first, I expect the next step will be simple increased pressure.

[tedvpap]

Looks to me like the problem is still obstructive. Have you thought about trying auto mode? By setting the minimum pressure to 6 but setting the maximum to 10 or so, then the machine can raise the pressure when it determines more pressure is needed.