By Jeffrey Zimmet, MD, PhD
Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory, San Francisco VA Medical Center
SYNOPSIS: In this retrospective cohort study using administrative data from a U.S. payor source, only 58.6% of patients undergoing patent foramen ovale closure had the procedure done for the approved indication of stroke or systemic embolism. A significant proportion of patients were outside the recommended age range.
SOURCE: Goldsweig AM, Deng Y, Yao X, et al. Approval, evidence, and “off-label” device utilization: The patent foramen ovale closure story. Circ Cardiovasc Qual Outcomes 2024;17:e010200.
Patent foramen ovale (PFO) closure currently is approved in the United States for adults up to 60 years of age who have experienced cryptogenic stroke that is thought to be associated with the PFO. Off-label use of these devices is thought to be common but has not previously been characterized. Accordingly, Goldsweig and colleagues reported on a retrospective cohort study performed to define trends in PFO closure use and procedure indications. Using data from the OptumLabs Data Warehouse for the period from 2006 through 2019, researchers identified 5,315 patients as undergoing PFO closure. Of the patients studied, 51.8% were female, and 29.2% were older than age 60 years.
Over the course of the study period, rates of PFO closure correlated somewhat with published study results, with a decline in the 2010 to 2015 period following some negative trials, and then an increase again in 2015 to 2019 after positive study results and U.S. Food and Drug Administration (FDA) approval. Stroke or systemic embolism — the primary approved indication for the procedure — was the listed indication in 58.6% of cases. Transient ischemic attack (TIA) was the listed indication in 10.2% of cases, and migraine was listed in 8.8% of cases. Older patients and male patients were less likely to have an indication of migraine.
A significant proportion of patients (17.6%) were reported to have atrial fibrillation (AF) at baseline. An additional 11.9% of patients developed AF subsequent to the procedure, with just more than half of new AF cases showing up in the first year after the procedure. A separate analysis in the subset of patients with American Hospital Association data showed no association of PFO closure rates with variables such as hospital size, number of patients using Medicaid, or whether hospitals were teaching centers. The authors concluded that PFO closure rates during the study period correlated weakly with clinical data and FDA approval status, and that almost half of patients underwent the procedure for indications not covered by FDA approval.
COMMENTARY
The history of PFO closure for treatment of cryptogenic stroke has been long and winding, with multiple negative trials preceding the positive evidence that led, eventually, to the current FDA approval. Ultimately, three positive trials of PFO closure all were published in the same issue of the New England Journal of Medicine in 2017 (RESPECT, Gore REDUCE, and CLOSE), with a fourth trial (DEFENSE-PFO) presented the following year. Assuming approximately 25% of all individuals have a PFO, the central question in each case of cerebral ischemia is whether an identified PFO is an incidental bystander or represents the etiology for stroke. These trials stacked the deck in favor of closure by excluding older patients as well as those with large-artery atherosclerotic disease or dissection, AF or another established cardioembolic source, small-vessel occlusive disease, or a hypercoagulable disorder requiring anticoagulation. Three of the four trials also included only patients with higher-risk PFOs, selecting only those with hypermobile or aneurysmal interatrial septum or larger shunts. For the most part, patients older than 60 years of age were not enrolled, and only patients with imaging-defined large-vessel stroke were included. Trials of PFO closure for the treatment of migraine have had mixed results, and this remains an unapproved indication for a closure procedure.
The perspective on these study results depends heavily on the lens through which one views them. There is no doubt that the use of administrative data is imperfect at best and can lead to false conclusions because of inherent limitations or to errors in coding. For example, this study looked for evidence of a stroke diagnosis only in the two-year period prior to PFO closure. Some patients may have experienced a stroke more than two years prior to closure but were not referred for the procedure until after positive data and regulatory approval. Taking a longer view or including data from more recent years might have changed this particular result.
Another issue is that diagnostic codes currently cannot distinguish between PFO and atrial septal defect (ASD). Adults with ASD commonly have AF, and this may partially explain why a surprising number of patients undergoing closure had AF at baseline. However, PFO closure remains a significantly more common procedure in adults compared to ASD closure, so the likelihood is that these data pertain mainly to PFO. Migraine remains an unproven indication for PFO closure, and this indication should be uncommon in practice.
Despite the limitations of administrative data, the suggestion that nearly one-quarter of patients undergoing this procedure had no stroke, systemic embolism, TIA, or even migraine raises the concern that some patients may be undergoing unnecessary procedures, such as prophylactic PFO closure for primary prevention of stroke. Ongoing trials ultimately may expand the indications for these devices, but for now this information presents an opportunity to promote the use of this technology for appropriate indications.
