Frontiers in Neurology (Feb 2024)

PFO-spectrum disorder: two different cerebrovascular diseases in patients with PFO as detected by AI brain imaging software

  • Raluca Ștefania Badea,
  • Raluca Ștefania Badea,
  • Sorina Mihăilă-Bâldea,
  • Sorina Mihăilă-Bâldea,
  • Athena Ribigan,
  • Athena Ribigan,
  • Anca Negrilă,
  • Nicolae Grecu,
  • Nicolae Grecu,
  • Andreea Nicoleta Marinescu,
  • Florina Antochi,
  • Cristina Tiu,
  • Cristina Tiu,
  • Dragos Vinereanu,
  • Dragos Vinereanu,
  • Bogdan Ovidiu Popescu,
  • Bogdan Ovidiu Popescu

DOI
https://doi.org/10.3389/fneur.2024.1357348
Journal volume & issue
Vol. 15

Abstract

Read online

BackgroundPatent foramen ovale (PFO) is a prevalent cardiac remnant of fetal anatomy that may pose a risk factor for stroke in some patients, while others can present with asymptomatic white matter (WM) lesions. The current study aimed to test the hypothesis that patients with a PFO who have a history of stroke or transient ischemic attack, compared to those without such a history, have a different burden and distribution of cerebral WM hyperintensities. Additionally, we tested the association between PFO morphological characteristics and severity of shunt, and their impact on the occurrence of ischemic cerebral vascular events and on the burden of cerebral WM lesions.Patients and methodsRetrospective, case–control study that included patients with PFO confirmed by transesophageal echocardiography. Right-to-left shunt size was assessed using transcranial Doppler ultrasound. Cerebral MRIs were analyzed for all participants using the semi-automated Quantib NDTM software for the objective quantification of WM lesions. WM lesions volume was compared between patients with and without a history of stroke. Additionally, the anatomical characteristics of PFOs were assessed to explore their relation to stroke occurrence and WM lesions volume.ResultsOf the initial 264 patients diagnosed with PFO, 67 met the inclusion criteria and were included in the analysis. Of them, 62% had a history of PFO-related stroke/TIA. Overall burden of WM lesions, including stroke volume, was not significantly different (p = 0.103). However, after excluding stroke volume, WM lesions volume was significantly higher in patients without stroke (0.27 cm3, IQR 0.03–0.60) compared to those with stroke/TIA (0.08 cm3, IQR 0.02–0.18), p = 0.019. Patients with a history of PFO-related stroke/TIA had a tendency to larger PFO sizes by comparison to those without, in terms of length and height, and exhibited greater right-to-left shunt volumes.DiscussionWe suggest that PFO may be associated with the development of two distinct cerebrovascular conditions (stroke and “silent” WM lesions), each characterized by unique imaging patterns. Further studies are needed to identify better the “at-risk” PFOs and gain deeper insights into their clinical implications.

Keywords