Author + information
- Stefano F. Rimoldi, MD,
- Sebastian R. Ott, MD,
- Emrush Rexhaj, MD,
- Robert von Arx, MD,
- Stefano F. de Marchi, MD,
- Roman Brenner, MD,
- Urs Scherrer, MD,
- Bernhard Meier, MD,
- Matthias Gugger, MD,
- Yves Allemann, MD and
- Christian Seiler, MD∗ ()
- ↵∗University Hospital, CH-3010 Bern, Switzerland
Patent foramen ovale (PFO) is a prevalent embryologic remnant with insufficient post-natal adhesion of the cardiac atrial septum primum and secundum. Among many other conditions, it has been associated with obstructive sleep apnea (OSA) syndrome. The mechanism operative in these conditions is temporary right-to-left atrial shunting across the “open door” of the septum primum, which leads to short bouts of arterial de-oxygenation. The effect of PFO closure on OSA has not been investigated to date. We tested whether PFO closure in OSA patients led to a lowering of the apnea–hypopnea index (AHI) and to improved cardiovascular function.
This prospective open-label study included 40 consecutive patients with newly diagnosed OSA. Fourteen of them had PFOs diagnosed by transesophageal contrast echocardiography; these patients underwent initial device closure. Twenty-six patients did not have PFOs. Conventional treatment for OSA was postponed for 3 months in both groups, and polysomnographic and cardiovascular examinations were performed at baseline and at the end of the follow-up period. The following endpoints were compared intra- and interindividually: AHI (primary endpoint); oxygen de-saturation index (ODI) (drops in oxygen saturation >3% points per hour); systemic arterial blood pressure; and Doppler echocardiographic parameters. During follow-up, AHI decreased from 38.6 ± 16.0 to 30.4 ± 16.1 events per hour in the PFO closure group (p = 0.0034), and it changed from 33.9 ± 29.8 to 38.6 ± 26.2 events per hour in the no PFO group (p = 0.29) (Figure 1). AHI change (follow-up minus baseline value) differed significantly between the groups: –7.9 ± 10.4 in the PFO closure group and +4.7 ± 13.1 in the no PFO group (p = 0.0009). The following parameters improved significantly in the PFO closure group, whereas they remained unchanged in the no PFO group: ODI; a decrease in nocturnal systolic blood pressure (Figure 1) and daytime blood pressure; right ventricular-to-right atrial systolic pressure gradient (Figure 1); and left ventricular diastolic function.
This first clinical investigation on the effect of PFO closure on OSA found a reduction of approximately 8 AHI events per hour in response to PFO closure. This reduction was accompanied by a mitigated ODI, by a nocturnal systemic blood pressure reduction of 5 mm Hg, and by lowered pulmonary pressure values with enhanced left ventricular diastolic function.
A causal relation between PFO and OSA would be evident if a positive effect on OSA of PFO closure could be systematically documented. So far, the efficacy of PFO closure on OSA parameters has only been implied in case reports (1–3).
Pathophysiologically, PFO appears to have an exacerbating effect on phasic breathing in OSA. The crucial element in this context is short events of atrial right-to-left shunts with ensuing episodes of hypoxemia, which aggravate the already disturbed central breathing regulation in OSA. The initiating incident in the cascade of phasic breathing in OSA is the first apnea in the context of muscular relaxation during rapid eye movement sleep. Rising arterial pressure of carbon dioxide (PCO2) in the context of this apnea induces breathing efforts against the closed glottis, which briefly elevates right atrial pressure above left atrial pressure and leads to shunting of de-oxygenated blood to the systemic side in cases of a PFO. Hypoxemia impairs endothelial function in the pulmonary and in the systemic circulation, and it acts as a central respiratory stimulant, which influences the PCO2-steered respiratory regulation differently from other stimulants (4). Hypoxemia lowers the eupneic PCO2 level in the context of hyperventilation without concomitant reduction in the apneic PCO2 threshold, thus destabilizing the system and rendering it more prone to ensuing apnea phases (4). In this scenario, the elevated sympathetic tone with increased systemic blood pressure is related to the events of arousal at the end of the apnea phases. Our study results support this concept by showing a normalization of the estimated pulmonary pressure in response to PFO closure, together with augmented diastolic function of the left ventricle and reduced nocturnal systolic blood pressure.
The main study limitations are the nonrandomized design, which was appropriate in the context of this first-in-man study, the low number of patients included, and the lack of female patients.
Please note: Dr. Meier has received research grants to the Department of Cardiology and speaker fees from St. Jude Medical. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Rimoldi and Ott contributed equally to this study.
- American College of Cardiology Foundation