COMMENTARY: The Trouble with Bubbles

Why do residual shunts exist after patent foramen ovale (PFO) closure devices are placed? Possible explanations include the use of a device too small for the defect, inability of the placed device to adequately conform to the defect, leakage through the device, septal mobility limiting adequate fusion of the device to the tissue, presence of septal perforations and remodeling or trauma to the septal tissue. In this issue of the Journal of Invasive Cardiology, Zajarias et al¹ identify septal mobility as being predictive of a residual shunt after placing an Amplatzer PFO occluder device. Although this relationship may concur with many interventionalists’ empiric observations, several questions seem to be hiding in the shadows of this study. Probably the most important is a fundamental one: Is residual shunting after PFO closure clinically relevant? Clinical events after PFO closure are infrequent, occurring in 1–3% of cases, and are usually mild, presenting as transient ischemic attacks (TIAs) rather than permanent neurologic events.² Perhaps the dominance of TIAs as opposed to permanent neurologic events is due to the elimination of large defects, leading to clinically smaller implications. Perhaps it is sufficient to get rid of large shunts, thereby preventing large clots from crossing into the left atrium. Large shunts, which are more likely to be associated with clinical events, have stronger associations with clinical events.³ PFO closure devices are particularly effective in reducing large shunts.⁴ Of course, the residual events could be for other reasons unrelated to a residual shunt as well. After all, approximately 50% of young patients with a cryptogenic stroke do not have a PFO, strongly suggesting there are other factors causing their clinical events. Some individuals who undergo PFO closure will undoubtedly have experienced their original neurologic event due to mechanisms unrelated to their PFO, so a recurrent event is likely to be for reasons also unrelated to the PFO. In fact, the data supporting the clinical relevance of a residual shunt are tenuous. Windecker et al, evaluating 80 patients with PFO closure, did find that residual shunt at the time of the procedure was associated with a four-fold increase in recurrent events compared with those without a residual shunt.⁵ However, subsequently, the same investigators looking at a larger series of patients failed to discern an association with residual shunt at 6 months and a recurrent neurologic event.⁶ A review of studies evaluating effectiveness of treatment options for paradoxical embolization in patients with a PFO failed to find an association of residual shunt with recurrent neurologic events.² Even assuming that residual shunt is predictive of recurrent events, methodological challenges exist in defining the presence or magnitude of a residual shunt. Residual shunting changes with time, usually, but not always, diminishing with the healing response. So the timing of the assessment is a consideration. Then, the challenges of assessing right-to-left shunting prior to device closure remain relevant. These include the impact of Valsalva, the ability to obtain a good quality Valsalva, upper versus lower extremity contrast injection, the echocardiographic modality being used, the optimal planes for visualizing the bubbles (which may vary from individual to individual), and so forth.⁷ In addition, the devices can cause shadowing in the left atrium, which may decrease the sensitivity of the procedure (more relevant for intracardiac echo assessments). For example, refer to Figure 2 in the article by Zajarias et al to see shadowing caused by the device. (Of course, since this is a TEE image, the shadowing is in the right atrium, of lesser importance in evaluating for residual shunting, but this is a nice illustration of the shadowing caused by this particular device). With this background, Zajarias et al in this issue of the Journal of Invasive Cardiology address what anatomic features might contribute to the finding of a residual shunt on transthoracic echocardiography 24 hours after PFO closure, when the Amplatzer PFO occluder device was employed.¹ The strongest association was with the floppiness of the atrial septum. Patients with atrial septal aneurysms were more likely to have residual shunts, but even patients with milder forms of atrial elasticity had an increase in residual shunts. This might be due to the floppiness of the tissue preventing good device/tissue coaptation, or simply because PFOs are larger when associated with atrial septal aneurysm.³ However, the association does not necessarily imply causation. It is not necessarily true that the residual leaking is of etiological significance, even if there is a relationship between atrial septal aneurysm and residual leaking and recurrent events. Perhaps whatever caused the floppiness of the atrial septum to begin with is of etiological importance. To elaborate, perhaps floppiness of vascular tissue is the relevant feature for reasons that are currently not recognized, not the associated residual shunting. Suppose, for example, patients with atrial septal aneurysms are more likely to develop paroxysmal atrial fibrillation, and that is the true etiology of the recurrent events. If this hypothesis were true, the association with residual shunting might be an epiphenomena, or a coincidental, not a causative, relationship. Interestingly, Zajarias et al virtually ignore another anatomic condition that has been identified as being associated with residual shunts: a long tunnel. Long tunnels have the potential to distort a closure device, leading to residual shunting.^8,9 Perhaps that is due to the challenge of defining a long tunnel, or perhaps the Amplatzer device is less likely than other currently-available devices to have a residual shunt due to device distortion through a long tunnel. Is this meaningful? As more devices are introduced to close PFOs, one may imagine that residual shunting may become a surrogate for effectiveness of a device. We need to consider whether that is an appropriate surrogate, and not simply assume that residual shunting is a marker for a more effective device. We must be cautious regarding too much reliance on such a surrogate until there are data confirming a clinical relevance of that surrogate. Such clinical relevance is currently lacking for residual shunting after PFO closure. There may be other advantages of one device or closure strategy that may offset a slightly lower risk of complete closure, provided large residual shunts are avoided. These advantages may include cost, ease of use in implantation, lower risks of complications or device fracture, or even the ability to avoid leaving a device behind while closing the PFO. Although the information being provided by Zajarias et al is useful in helping us understand the mechanisms of residual shunting after PFO closure, we must remember to look upon this understanding only in the context of its clinical relevance.

References 1. Zajarias A, Thanigaraj S, Lasala J, Perez J. Predictors and clinical outcomes of residual shunt in patients undergoing percutaneous transcatheter closure of patent foramen ovale. J Invasive Cardiol 2006;18:533–537. 2. Khalry P, O’Donnell CP, Landzberg MJ. Transcatheter closure versus medical therapy of patent foramen ovale and presumed paradoxical thromboemboli. Ann Intern Med 2003;139:753–760. 3. Kerut EK, Norfleet WT, Plotnick GD, Giles TD. Patent foramen ovale: A review of associated conditions and the impact of physiological size. J Am Coll Cardiol 2001;38:613–623. 4. Anzola GP, Mirandi E, Casilli F, Onorato E. Does transcatheter closure of patent foramen ovale really “shut the door?”: A prospective study with transcranial Doppler. Stroke 2004;35:2140–2144. 5. Windecker S, Wahl A, Chatterjee T, et al. Percutaneous closure of patent foramen ovale in patients with paradoxical embolism — Long-term risk of recurrent thromboembolic events. Circulation 2000;101:893–898. 6. Scwherzmann M, Wahl A, Nedeltchev K, et al. Patients at risk for recurrent embolism after percutaneous closure of patient foramen ovale for presumed paradoxical embolism (Abstract). J Amer Coll Cardiol 2004;7A. 7. Devuyst G, Piechowski-Jozwiak B, Karapanayiotides T, et al. Controlled contrast transcranial Doppler and arterial blood gas analysis to quantify shunt through patent foramen ovale. Stroke 2004;35:859–863. 8. Ruiz CE, Alboliras ET, Pophal SG. The puncture technique: A new method for transcatheter closure of patent foramen ovale. Catheter Cardiovasc Interv 2001;53:369–372. 9. Tande AJ, Knickelbine T, Chavez I, et al. Transseptal technique of percutaneous PFO closure results in persistent interatrial shunting. Catheter Cardiovasc Interv 2005;65:295–300.