Abstract Body: Background: Paradoxical embolism through a PFO is a common etiologic mechanism in the 18-60 yo patients with otherwise cryptogenic stroke (CS). However, its importance as a cause of stroke in the elderly is less clear. Older individuals have more competing causes of stroke, potentially reducing the relevance of PFO-associated stroke; but they have more venous thromboembolism and greater right-to-left shunting, potentially enhancing PFO-associated stroke frequency. Clinical management would be aided by estimates of how often a PFO in an older patient with CS is causal vs. incidental.
Methods: Through systematic search, we identified all case–control studies of the comparative frequency of PFO in patients with CS vs known stroke cause (KS). We performed a random-effects meta-analysis across these studies to determine the best estimate of the differential frequency of PFO in CS and KS among three age cohorts: 1) age 60 and older 2) age 50/55 and older; and age 70/75 and older. We then applied Bayes’ theorem to determine the probability that detected PFO is causal or incidental in these groups.
Results: The systematic search identified: for age 60+, 1 study enrolling 397 patients (190 CS, 207 KS); for age 50/55+, 5 studies enrolling 1203 patients (534 CS, 669 KS); and for age 70/75+, 3 studies enrolling 431 patients (171 CS, 260 KS). Methods of PFO detection were TCD in 47% of patients, TTE in 5%, and TEE in 48%. Crude cumulative rates of PFO detection were: age 60+, CS 68/190 (35.8%) vs. KS 44/207 (21.3%); age 50/55+, CS 198/554 (35.7%) vs. KS 81/669 (12.1%); and age 70/75+, CS 58/171 (33.9%) vs. KS 43/260 (16.5%). In the formal meta-analyses, summary odds ratios for PFO in CS vs. KS patients were: age 60+, 2.06 (95% CI, 1.32-3.23), p = 0.001; age 50/55+, 4.23 (95% CI, 2.44-7.32), p = 0.00001;.and age 70/75+ 3.77 (95% CI, 1.27-11.24), p = 0.02. With the application of Bayes’ theorem, the corresponding probabilities indicated that, when present, PFO in patients with CS is causal in: age 60+, 51.5% (95% CI, 24.2-69.0%); age 50/55, 78.2% (95% CI, 60.4-83.9%); and age 70/75+,61.5% (95%CI, 21.0-91.1%).
Conclusion: Although confidence intervals are wide reflecting moderate study sizes, current evidence indicates that in older patients with cryptogenic stroke and PFO, the PFO is causal in at least one half of patients. These findings support the performance of randomized trials of PFO closure in older patients with otherwise cryptogenic stroke.