Effects of cardiac dysfunction on non-hypercapnic central sleep apnea

Introduction: Non-hypercapnic central sleep apnea (CSA) commonly occurs during nonrapid eye movement (non-REM) sleep in adults with congestive heart failure (CHF) and in some subjects without signs or symptoms of CHF. Hyperventilation, reduced lung volume, and circulatory delay are known to contribute to CSA, but to differing degrees depending on presence or absence of CHF. Aim: To determine whether the pattern of ventilation during sleep could be used to determine the presence of CHF. Methods: Full polysomnographs demonstrating CSA were examined in 10 consecutive subjects with CHF and in 10 without CHF. Ventilatory, apnea, and cycle lengths, and circulation time (from the onset of ventilatory effort to the nadir of oximeter trace) were measured from cyclic apneas during non-REM sleep. Results: The non-CHF group had a greater left ventricular ejection fraction (LVEF) (59.7+/-1.9% vs 19.2+/-2.2%). Circulation time (11.8+/-0.5 s vs 24.9+/-1.7 s; p<0.001) and cycle length (35.1+/-2.8 s vs 69.5+/-4.5 s; p<0.001) were significantly greater in the CHF group compared with the non-CHF group, but not apnea length (21.3+/-1.8 s vs 26.8+/-2.0 s; p=0.06). Ventilatory length to apnea length ratio (VL:AL) was uniformly >1.0 in the CHF group (mean, 1.65; range, 1.02 to 2.33), and in the non-CHF group <1.0 (mean, 0.66; range, 0.54 to 0.89). LVEF correlated negatively with both circulation time (r=-0.86; p<0.001) and cycle length (r=-0.79; p<0.001). Conclusion: The VL:AL ratio >1.0, as well as both circulation time >15 s and cycle length >45 s,can be used to recognize the presence of CHF in subjects with CSA.