We studied five patients with chronic stable congestive heart failure (CHF), all of whom demonstrated recurrent apneas in association with Cheyne-Stokes respiration (CSR) during sleep. All five patients had symptoms consistent with a sleep apnea syndrome. Nasal continuous positive airway pressure (NCPAP) was administered at 8 to 12.5 cm H2O to all patients during sleep. The number of apneas fell from (mean +/- SE) 60 +/- 12/h of sleep on the control night to 9 +/- 7/h of sleep (p less than 0.01) on the NCPAP night, whereas mean nocturnal SaO2 rose from 88 +/- 2% on the control night to 92 +/- 2% (p less than 0.025) while on NCPAP. This was associated with resolution of symptoms of sleep apnea. In addition, resting left ventricular ejection fraction (LVEF) as measured by radionuclide angiography (RNA) rose from 31 +/- 8% while off NCPAP to 38 +/- 10% (p less than 0.05) while on NCPAP. Furthermore, all five patients experienced marked improvement in symptoms of heart failure from functional classes III and IV (New York Heart Association Classification) prior to NCPAP therapy to class II after NCPAP therapy was instituted. We conclude that, in certain patients, CSR during sleep associated with chronic CHF constitutes a sleep apnea syndrome, which can be alleviated by NCPAP. In addition, NCPAP therapy may lead to a reduction in cardiac dyspnea and improvement in left ventricular function.

Respiratory distress syndrome of the newborn, prematurity, and familial airway hyperreactivity may contribute to long-term pulmonary sequelae. We assessed the role of each by testing pulmonary function and airway reactivity in 11 prematurely born children who survived the respiratory distress syndrome and in 11 prematurely born children who had no neonatal respiratory disease, each of whom was paired with a sibling born at term. The subjects were between 7 and 12 years of age when studied. Airway reactivity was also assessed in their mothers. The group who had had the respiratory distress syndrome had higher ratios of residual volume to total lung capacity and lower values for forced expiratory volume in one second than did their siblings or normal controls (P less than 0.01). Expiratory flow was decreased in both groups born prematurely (P less than 0.02) and was related to neonatal exposure to oxygen (r = -0.71, P less than 0.02). The incidence of airway hyperreactivity was elevated in all groups, including the mothers. These data suggest that long-term pulmonary sequelae of the respiratory distress syndrome of the newborn are related to the disease, its treatment, or both, and to airway hyperreactivity. In prematurely born children without neonatal lung disease, the sequelae are related to airway hyperreactivity. The possibility of a relation between familial airway hyperreactivity and premature birth is suggested.

We studied the electromyographic activity of the diaphragm (EMGdi) and abdominal external oblique (EMGeo) muscles in response to progressive hypercapnia (HCVR) and hypoxia (HVR) in five normal males. The slopes of the regression lines relating log EMGdi activity to minute volume of ventilation (VE) were steeper during HVR runs than HCVR runs (mean +/- SE, 0.03201 +/- 0.00724 vs. 0.02729 +/- 0.00676, P less than 0.03). Phasic expiratory EMGeo activity was seen in 15 of 15 HCVR runs but in only 6 of 15 HVR runs. Furthermore, the maximum level of VE attained before the onset of EMGeo activity was significantly lower during HCVR runs than during HVR runs (23.1 +/- 2.5 vs. 34.8 +/- 4.01/min, P less than 0.003). We conclude that in awake humans 1) the diaphragm is activated to a greater extent by hypoxia than hypercapnia at a given VE and 2) hypercapnia causes a more consistent recruitment of abdominal expiratory activity at lower VE than does hypoxia.

The electromyographic activity of the diaphragm (EMGdi) and scalene muscle (EMGsc) was studied in the supine and upright positions, respectively, during hyperoxic progressive hypercapnic rebreathing (HCVR) in five healthy males. End-expiratory esophageal pressure (EEPes) was quantified on a breath-to-breath basis as a reflection of altered end-expiratory lung volume. There was no significant difference in the slopes of EMGdi, expressed as a percentage of maximum at total lung capacity vs. minute volume of ventilation (VI), between the supine and upright positions [0.79 +/- 0.05 (SE) vs. 0.92 +/- 0.17, respectively]. In contrast, the slope of the regression line relating EMGsc to VI was steeper in the upright than in the supine position (0.69 +/- 0.05 vs. 0.35 +/- 0.04, respectively; P less than 0.005). Positive EEPes at comparable VI at the ends of HCVRs were of greater magnitude upright than supine (3.27 +/- 0.68 vs. 4.35 +/- 0.60 cmH2O, respectively, P less than 0.001). We conclude that altering posture has a greater effect on scalene and expiratory muscle activity than on diaphragmatic activity during hypercapnic stimulation.