We've updated our privacy policy.

Sleep Heart Health Study

6.6.2.8 Dealing with ambiguous respiratory events

If changes in amplitude are reduced by < 30%, but a ≥ 10 s. period of a clearly discernible reduction in the amplitude of respiratory signals from baseline is observed, then score events when:

  • they occur as part of a series of events (that do meet the 30% reduction criteria) (ResFigures 7a, 7n, and 7o).
  • they are associated with desaturations of at least 2% (ResFigures 7b, 7m, 7p).

Apneas/hypopneas immediately following large breaths or movements are not scored, unless they are part of the cycle (there are other respiratory events before and following). Such isolated events may be “sighs” or artifact (ResFigures 7c). When such an event is noted after a movement/large breath and appears to “trigger” a series of events, the first respiratory event is not scored (unless apneas/hypopneas were scored before the movement/large breath) (ResFigure 7d and 7e).

Determining whether to score one long event or two short events (i.e., after an initial decrease in amplitude of a breathing signal, there is some increase, but not to baseline, and then a fall again): In these cases, the oxygen saturation channel will be checked to determine whether one or more than one event should be marked. If the event is punctuated by a clear decrease in oxygen saturation followed by a rise or a stabilization, and then a decrease again, two events are marked; otherwise if the event is characterized by one steady progressive fall in desaturation, than one event is scored (ResFigure 7f, 7g, 7h, 7i, and 7j).

Periods of hyperventilation followed by long periods of hypoventilation: If periods of hypoventilation have clearly visible beginnings and endings (i.e., are “discrete”) and are associated with at least 2% desaturation, score as hypopneas regardless of their length (they can last up to a few min.). (Note: This is commonly observed in REM.) (ResFigure 7k and 7l).


National Sleep Research Resource
Sleep Heart Health Study