Off-label closure during CLOSURE study

The role of percutaneous closure of patent foramen ovale (PFO) in patients with cryptogenic stroke or transient ischemic attack remains controversial. Registry data have suggested considerable benefit of closure over medical therapy, but the prospective, randomized CLOSURE I trial found no benefit for device closure.

We compared patients enrolled into CLOSURE I to off-label closures performed during the study recruitment period at a single large institution and prospectively enrolled into an institutional registry of PFO closure. We also compared CLOSURE I patients at our institution to the reported characteristics of the entire study to ensure generalizability.

Between 11/3/2003 and 4/16/2007, there were 100 off-label closures and 33 patients randomized into CLOSURE I. Compared with off-label closure, patients in CLOSURE I were younger (41.6 ± 10.1 years vs 50.0 ± 14.0 years; P<.001) and had fewer cardiovascular risks including hypertension (12% vs 36%; P=.009), hyperlipidemia (24% vs 53%; P=.008), and coronary disease (3% vs 44%; P<.001). Degree of right-to-left shunting was considerably higher in off-label closures (28%, 14%, and 58% vs 45%, 30%, and 25% for mild, moderate, and severe, respectively; P=.026).

Off-label closures outnumbered patient recruitment into CLOSURE 3:1 at our institution during study recruitment. Certain demographic differences were expected (age over 60 was an exclusion for CLOSURE I), but vascular risks were considerably greater in the off-label group and may be important mechanistically. Large shunts were considerably more common in off-label patients, suggesting that higher-risk patients may have been preferentially closed off-label. These results suggest that the results of CLOSURE I may not apply to all patients with initial cryptogenic stroke.