Improvement in Sleep Apnea Associated With Closure of a Patent Foramen Ovale

Abstract

Recent reports have documented an association between patent foramen ovale and obstructive sleep apnea. We report on a 51-year-old man with obstructive sleep apnea and recent stroke who was enrolled in a clinic trial evaluating the efficacy of closure of patent foramen ovale following ischemic stroke. He was randomly assigned to device closure. There was subjective dramatic improvement in sleep-apnea symptoms and objective improvement in polysomnographic testing after device implantation. Aside from a drop in apneas and hypopneas from 181 and 8 on the first polysomnogram to 19 and 0 on the second, there was no significant weight loss nor were there other significant changes in sleep parameters or medications. He stopped using continuous positive airway pressure 2 months after implantation and has had no recurrent sleep complaints during 18 months of follow-up. Further studies evaluating the relationship among patent foramen ovale, sleep apnea, and device implantation are warranted.

Citations:

Silver B; Greenbaum A; McCarthy S. Improvement in Sleep Apnea Associated With Closure of a Patent Foramen Ovale. J Clin Sleep Med 2007;3(3):295–296

Persistent patency of the foramen ovale is a common anatomic finding present in roughly 25% of the general population.¹ Data regarding an association between isolated patent foramen ovale (PFO) and stroke are conflicting, but systematic studies suggest no relationship.² Ongoing randomized trials are testing the hypothesis that PFO closure after stroke will reduce the risk of recurrent stroke. PFO has also been associated with other conditions such as platypnea-orthodeoxia syndrome,³ migraine with aura, decompression illness, and obstructive sleep apnea (OSA).^4,5 We report on a patient who, in the course of a clinical trial of PFO closure after stroke,⁶ had subjective and objective improvement in sleep apnea.

REPORT OF CASE

A 51-year-old right-handed man with a medical history of hyperlipidemia, depression, obesity, and OSA on continuous positive airway pressure (CPAP) was admitted with dysarthria and left sensorimotor deficit. Diffusion-weighted magnetic resonance imaging showed acute infarctions in the right parietal lobe and right insular cortex. Other than PFO with atrial septal aneurysm on transesophageal echocardiogram, no other potential source of stroke was identified. He was enrolled in an unmasked, randomized clinical trial comparing best medical therapy and percutaneous closure of the PFO; he was assigned to the closure arm of the trial. The device was implanted 2 months after the stroke.

Two months after randomization and 4 months after his stroke, the patient had stopped using CPAP. At 12 months, he reported that he no longer had daytime sleepiness. A full-night polysomnogram (PSG) 10 months before the stroke, and prior to treatment with CPAP, showed severe OSA with a total of 181 apneas and 8 hypopneas. The details of this PSG can be found in Table 1. Audio monitoring revealed loud snoring. A full-night PSG 2 months after randomization (and 4 days after discontinuation of CPAP) showed 19 apneas and 0 hypopneas. The details of this PSG can be found in Table 1. Audio monitoring revealed loud snoring. Both PSGs were performed in a sleep laboratory accredited by the American Academy of Sleep Medicine using the same acquisition equipment and monitoring montages. Both studies were interpreted by the same individual. The patient slept flat during both studies. At the time of randomization (2 months after the stroke), he was 175 cm tall and weighed 113 kg. Eighteen months after the stroke, his weight had increased to 116.6 kg, but he reported no return of daytime sleepiness. Medications used prior to the first PSG were pentoxifylline, fenofibrate, simvastatin, omeprazole, and rofecoxib. Medications used after his stroke included aspirin, omeprazole, simvastatin, and fluoxetine. The patient denied alcohol, narcotic, or benzodiazepine use at any time.

Table 1.

Polysomnographic Findings Before and After Device Implantation

Parameter	Study 1	Study 2
Time in bed, h	8.0	7.1
Total sleep time, h
Total	6.6	6.5
Supine	1.5	0.6
Nonsupine	5.1	5.9
Sleep stage, as % of total sleep time
1	55.7	61.9
2	28.9	27.1
3/4	1.5	0.0
REM	13.8	11.0
Mean SaO2, %, during
Wakefulness	96	93
NREM	93.2	92.6
REM	90.4	93.2
A+H, during entire sleep period, no.
Total	189	19
Supine	45	7
Nonsupine	144	12
Hypopneas, total no. during sleep period	8	0
AHI, no./h
Total	28.6	2.9
Supine	30.0	11.7
Nonsupine	28.2	2.0
Percentage of A+H, %
Obstructive	64.6	100
In supine position	23.8	36.8
Central	5.8	0
In supine position	0	NA
Mixed	29.6	0
In supine position	28.6	NA
Weight, kg	118	113

Polysomnographic findings 12 months prior to (Study 1) and 2 months after device implantation (Study 2). REM refers to rapid eye movement sleep; NREM, non-rapid eye movement sleep; A+H, apneas plus hypopneas; AHI, apnea-hypopnea index, the number of apneas and hypopneas per hour of sleep.

DISCUSSION

In this patient with PFO, device implantation was temporally associated with improvement in symptoms of sleep apnea and with objective marked reduction in apnea frequency during PSG monitoring. Although a placebo effect is possible, neither the patient nor the investigators expected changes in sleep-apnea symptoms going into the trial. The patient volunteered the changes in his sleep habits during follow-up office visits. Additionally, the changes on PSG during sleep would unlikely to be attributable to a conscious effect like placebo. Weight loss is a potential reason for improvement in symptoms.⁷ However, the patient's weight fluctuated by only 4.5 kg, and his symptoms of OSA did not recur when he returned to his prestroke weight at the 18-month follow-up visit. Other potential reasons for improvement, including significant decrease in percentage of rapid eye movement sleep between PSGs; change in supine sleep position between PSGs; use of CPAP within several days of the PSG; and discontinuation of narcotics, benzodiazepines, and alcohol, did not occur in this patient.

Pathophysiologically, PFO and sleep apnea may be related by hemodynamic or neurohumoral factors. Significant hypoxemia, particularly during Valsalva maneuvers, were observed in about one third of patients with OSA and PFO in 1 study.⁴ In another study, closure of a PFO resulted in immediate significant improvement in oxygenation and symptoms in patients with platypnea-orthodeoxia syndrome.³ Alternatively, unrecognized neurohumoral factors in deoxygenated blood may trigger sleep-apnea symptoms via passage to the arterial system through a PFO. Further studies evaluating the relationship among PFO, sleep apnea, and device implantation are warranted.