The discipline of catheter-based closure of patent foramen ovale (PFO) is still in its “childhood”. Ushered in by the seminal paper from Bridges et al in 1992,1 it is still hampered by amateurism, ghosts of old misconceptions, apprehension about new methods requiring permanent implants and, most of all, lack of adequate exposure to the discipline in the professional world. The paper prompting this commentary is a step in the right direction. The fact is, catheter-based closure of a PFO is the most simple and innocuous therapeutic intervention in invasive cardiology, shy of, perhaps, the insertion of a temporary cardiac pacemaker. Hence, the old saying “see one, do one, teach one” fits as closely as ever. What makes it so simple? The PFO lies in the direct extension path of the inferior vena cava, as it is supposed to. In about half the cases, a conventional guidewire being pushed up the vena cava will pass the PFO spontaneously. The other half of PFOs are easily cannulated with very few exceptions, provided a few anatomical and technical details are known to the operator. They can be learned in less than one hour. All current devices feature a left-sided disk or retainer which invariably pulls the valve-like PFO shut when opened in a left atrium and pulled back against the left side of the interatrial septum. The analogy of sticking a folded umbrella through a door ajar, opening the umbrella on the other side and pulling it back explains this principle well. This example also explains that balloon measurement of the PFO is not required. The width of the maximum opening is hardly relevant. The length of the tunnel may be important for devices with a fixed interdisk distance, as the right-sided disk might get partially stuck in the tunnel, preventing it from unfolding properly (a cosmetic, but not necessarily a functional problem). It is of little importance with devices that are adjustable (automatically in the case of the Amplatzer device and operator-guided with the Premere device). The final seat of the device resembles a cufflink in a buttonhole, highly unlikely to dislodge. Once implanted, the patient requires very little aftercare. The venous puncture can be cared for by the patient holding a cotton swab against it. There is no objection to the patient walking out of the catheterization laboratory, leaving the hospital an hour later and engaging in a tennis game the same day. The need for follow-up drug treatment has never been examined, but some platelet inhibition for a few months and a recommendation for prophylaxis against endocarditis until full overgrowth of the device (weeks to months) is standard. Stripping the intervention down to its essentials, it can be performed in ten minutes by a single operator without the need for assistance with echocardiography or anesthesiology. The risk of cardiac perforation is practically zero if soft wires with a J-tip and compliant (not overly large) introducers are used. Air embolism remains as rare as with coronary interventions when using introducers that can be easily flushed (without the dead space of sidearms). The puncture site problems remain the only bane of this procedure. Not using an arterial line or a second venous line for intracardiac echocardiography (reported as the only reason for puncture site problems in the paper), and looking for the vein while carefully avoiding arteries in the groin, reduce these noncardiac problems to a few percent. The paper by Aslam et al conspicuously demonstrates that a small-volume institution and consequently small-volume operators can be launched with a few proctored cases and then go on doing an excellent job with further cases on their own, be they few or many. However, it also shows that such small-volume centers tend to stagnate rather than follow the evolution of the procedure, unless they maintain close contact with large-volume centers. This particular center made the transition from the initially-available device to the next device, which proved significantly more user-friendly. An analogy about these two devices has been described: it’s like driving a tractor to church on Sundays, and then getting hold of a limousine. You immediately abandon the tractor, though it got you to church every Sunday before. Large-volume centers have realized a couple of things and reacted to them. (1) Rather than replacing transesophageal echocardiography by intracardiac echocardiography during the procedure, echocardiography can be forgone altogether, reducing procedural duration and complications. (2) Transthoracic echocardiography is insufficient to assess the late result of the procedure. It lacks precision to define the closure rate of devices and to distinguish between the performance of individual devices. The highest complete closure rate (when assessed by 6-month transesophageal echocardiography with a protracted Valsalva maneuver and a bubble test sending the bubbles at the end of the Valsalva maneuver) can reasonably be expected to be about 90% with the Amplatzer occluder, and between 60% to 90% with the remainder of devices. While a device is likely to provide some protection even in the situation of continued bubble transit, the individual situation should be assessed and discussed with a patient with residual shunt. It is inadequate to just perform a transthoracic echocardiogram with its poor yield to detect clots attached to the device or residual shunt, and then report this to the patient as the valid result of the intervention. Finally, and most importantly, the clinical indications for these procedures are a mess. The interventional cardiologist usually learns about potential candidates via the neurologist and echocardiographer. The neurologist still works with the old tool box of explained strokes and cryptogenic (unexplained) strokes. Illogically, he puts the PFO-mediated stroke into the basket of cryptogenic stroke, although he acknowledges that the stroke is explained by the PFO. While it will take a decade to correct this in the textbooks, modern doctors should not wait for this to happen before changing their terminology. This is far more than a semantic problem. It excludes the patients who most need the procedure, i.e., the elderly. When looking for the PFO as the possible cause for a stroke only after having excluded all other possible causes, hardly a patient above the age of 60 will make it to that step. Notwithstanding, the PFO only starts to develop its true evil potential at about that age. Venous thrombosis, virtually absent in children (PFO has no pathogenic potential in them), and still rare in young and middle-aged adults, increases exponentially after the age of 55 years.2 Moreover, in analogy, while small clots having crossed the PFO usually go unnoticed unless they hit a brain artery, they may also cause havoc in a coronary artery.3 Operators capable of closing a PFO should consider this act when confronted with the suspicion of an embolic myocardial infarction not only in young but also in older people. In summary, it is unlikely, that all PFOs will be closed (about 25% of the population) any time soon, unless a harmless vaccination-type procedure becomes available. It is less unlikely that we will be screening for dangerous PFOs (associated with a very flimsy septum primum, a situation called atrial septal aneurysm) and close those (about 4% of the population) before a stroke or myocardial infarction has irreversibly damaged the patient, i.e., in early adulthood. It is possible that prognostically less compelling indications for PFO closures will be adopted following easy-to-gather randomized evidence (migraine, deep sea diving). It is an acknowledged fact that a PFO can cause prognostically-important events and even death. The question remains whether the current and future techniques of closing it prevent these effects and do this without creating, themselves, an equal or even larger number of them. While we are waiting for the respective randomized results (be ready for a long wait in light of the relatively young patients included in the pertinent trials since their PFO-related event risk will not start to significantly increase until one or several decades), PFO closure may remain restricted to the current indications. Even reserving the procedure for people with a PFO who have already had an event compatible with a paradoxical embolism, and even honoring the irrational prerequisite that all other stroke causes be excluded, the number of patients to be treated per time corresponds to about 10% of the current number of patients undergoing PCI. This is the current volume at some leading centers, while a few have already gone beyond, bowing to the principle that “lack of evidence of usefulness is not necessarily evidence of lack of usefulness”.
References 1. Bridges ND, Hellenbrand W, Latson L, et al. Transcatheter closure of patent foramen ovale after presumed paradoxical embolism. Circulation 1992;86:1902-1908. 2. Anderson FA, Jr., Wheeler HB, Goldberg RJ, et al. A population-based perspective of the hospital incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism. The Worcester DVT Study. Arch Intern Med 1991;151:933-938. 3. Mehan VK, Wahl A, Walpoth N, Meier B. Instant percutaneous closure of patent foramen ovale in patients with acute myocardial infarction and normal coronary arteries. Catheter Cardiovasc Interv 2006;67:279-282.