Patent foramen ovale is not associated with an increased risk of stroke recurrence

Abstract

Our data suggest that the risk of recurrent stroke in subjects with PFO is not significantly increased in comparison with subject without it.

Recurrent Stroke and Patent Foramen Ovale

“…After retrieving the full-text version of the aforementioned 21 studies, 5 studies were excluded because the end points of medically treated patients were mixed with the end points of surgically treated (with percutaneous or open closure) patients, [26][27][28][29][30] 2 studies because of the absence of comparison group (patients without PFO or no shunt stratification) 31,32 and 1 study because the reported end point was the combined outcome of stroke, TIA, or death. 33 The remaining 13 studies 24,[34][35][36][37][38][39][40][41][42][43][44][45] and the available data from our study cohort were included both in the qualitative and quantitative synthesis, according to the presence of PFO (PFO present, PFO absent), the shunt size (small right-to-left shunt [RLS] versus moderate or large RLS, small or moderate RLS versus large RLS), and the available outcome data (stroke or TIA, stroke; Figure 1). The characteristics of the included studies are summarized in Table 1.…”

“…TCD or TEE protocol was sufficiently presented in 12 studies while the description was unavailable in 2 studies. 24,42 Loss to follow-up was reported in 8 studies 34,36,37,40,41,[43][44][45] and could not be evaluated in 2 studies. 24,25 Only 4 studies reported no losses to follow-up after initial recruitment.…”

“…24,25 Only 4 studies reported no losses to follow-up after initial recruitment. 35,38,39,41 Finally, a funding source (commercial or public) was acknowledged in 3 studies. 36,41,45 …”

“…Third, the mean followup period was not the same for all studies and in many cases for individual patients within studies. Patients were prospectively followed for ≈2 years in 6 studies, 34,36,40,42,43,45 <2 years in 2 studies, 24 and >2 years in the remaining 6 studies 35,[37][38][39]41,44 (Table 1). Fourth, the presence of atrial septal aneurysm had a variable incidence among the included echocardiographic studies (range, 8.5%-41%) and could also vary widely according to the shunt size (Table in the online-only Data Supplement).…”

“…In 6 of these studies, the percentage of patients lost to followup was small (0.3%-4.9%) 34,36,37,40,43,44 while in the other 2 was considerably large (16.7% and 17.2%). 41,45 Sixth, selection bias cannot also be excluded in studies that enrolled patients nonconsecutively, or in those studies that excluded patients after initial recruitment. Additional causes for possible selection bias and false PFO prevalence rates in the individual studies are related to the fact that not all stroke or even cryptogenic stroke patients undergo TEE or TCD and that some patients have an insufficient transtemporal acoustic window on TCD examination.…”