To the Editor: A 50-year-old morbidly obese woman (weight, 183 kg; body mass index, 70 kg/m2) presented after 1 month of worsening dyspnoea. Risk factors for cardiovascular disease included hypertension, smoking and previous heavy alcohol consumption. She had left ventricular failure (thought to be due to diastolic dysfunction) and atrial fibrillation with rapid ventricular response. There was no evidence of infection or pulmonary embolism.

A transthoracic echocardiogram showed normal systolic function, although the images were poor. Measurement of arterial blood gases showed a compensated respiratory acidosis (Paco2, 58 mmHg; Pao2, 76 mmHg; pH, 7.39; bicarbonate, 34 mmol/L; base excess, 8 mmol/L). Administration of intravenous digoxin, frusemide and heparin led to some symptomatic improvement.

Cardiac monitoring showed atrial fibrillation, with ventricular rates of 120–170 beats/minute despite the digoxin treatment. Overnight, 70 pauses in heartbeat of up to 7 seconds were recorded (Box).

The constellation of the nocturnal rhythm disturbance, hypercapnia and obesity suggested sleep disordered breathing as a unifying diagnosis. Further questioning revealed symptoms indicative of obstructive sleep apnoea. Polysomnography, performed that night, showed a respiratory disturbance index (number of apnoeic or hypopnoeic episodes per hour) of 130 (normal value, < 15). The longest apnoeic episode was recorded at 34 seconds. The average minimum oxygen saturation was 84% during non-REM (rapid eye movement) sleep and 64% during REM sleep. The lowest saturations reached during non-REM and REM sleep were 64% and 61%, respectively. These results suggested severe obstructive sleep apnoea.1

The following night, her condition improved with application via mask of continuous positive airway pressure (CPAP) at a level of 12 cm H2O using room air. Cardiac monitoring continued to show atrial fibrillation with high ventricular rates, but no pauses in heartbeat occurred. This allowed treatment with atenolol and digoxin to control the ventricular rate.

At 6-week review, the patient was complying with home CPAP therapy and felt well. An electrocardiogram confirmed spontaneous reversion to sinus rhythm at 70 beats/minute.

There is increasing evidence to suggest a link between obstructive sleep apnoea and atrial fibrillation.2 Obstructive sleep apnoea is also a risk factor for hypertension, coronary artery disease and heart failure, all of which are known to precipitate atrial fibrillation.3,4 Transient heartblock has been found in 10% of patients with obstructive sleep apnoea. Pauses of up to 2 seconds are a predictable physiological response to inspiratory airflow obstruction and hypoxia. These pauses are exacerbated by negative chronotropes. Studies have shown that bradycardia can be effectively abolished with mask CPAP therapy.5

The dramatic response to CPAP therapy in this patient suggests that sleep apnoea caused the significant nocturnal ventricular pauses and had a role in the aetiology of the atrial fibrillation.