Beyond the Obvious: Evaluating Incidence and Causes of False Positive Patent Foramen Ovale Diagnoses in Cryptogenic Ischemic Stroke—A Retrospective Analysis

Background: Transesophageal echocardiography (TEE) is regarded as the gold standard for diagnosing patent foramen ovale (PFO). However, the occurrence of false positive PFO diagnoses is common in clinical practice, which may lead to unnecessary percutaneous interventions, as PFO occluders can only be placed when PFO presence is unequivocally confirmed. Various medical societies recommend a systematic protocol for the thorough assessment of the interatrial septum. Purpose: This study aimed to evaluate the incidence of false positive PFO diagnoses and analyze the underlying causes of misdiagnosis by comparing imaging data. Methods: In this retrospective observational study, we included 346 patients who were diagnosed with PFO following a cryptogenic ischemic stroke from January 2012 to December 2021. We first assessed the incidence of false positive PFO diagnoses. Subsequently, we compared stored imaging sequences from the screening TEE examinations with recommended diagnostic protocols outlined in European and American guidelines, identifying sources of diagnostic error. Results: The patient cohorts were divided into two groups: Group A consisted of 20 patients (5.8%) with PFO misdiagnoses, while Group B included 326 patients (94.2%) with accurate PFO occlusions. Our review of guideline compliance revealed that imaging planes in the mid-esophagus were used significantly more frequently than modified transgastric probe positions in the upper esophagus (71-96% vs. 1-2%). Three-dimensional imaging of the interatrial septum (IAS) was rarely utilized (0-3%), and the stepwise enlargement of the transducer angle in 15° increments was only partially documented. Comparisons between the two subgroups indicated reduced utilization of the bicaval viewing plane in the middle esophagus for the false positive group (Group A: 50% vs. Group B: 87%, p<0.001). The bubble test was performed in 80% of the misdiagnosis group using the short axis view, compared to 63% in the control group, while the bicaval view was utilized 30% of the time in the misdiagnosis group versus 47% in the control group (p=0.13 for both). IAS presentation in x-plane mode was consistently low across both groups (Group A: 20% vs. Group B: 17%, p=0.76). Conclusion: The incidence of false-positive PFO diagnoses before procedures was found to be 5.8% over the past decade. To mitigate such misdiagnoses, we identified three critical quality criteria that should be incorporated into all TEE examinations: 1) 2D imaging in the short axis view at the aortic valve level (30-75°), 2) 2D-imaging in the bicaval axis view (90-120°), and 3) a bubble test performed in conjunction with these views utilizing x-plane mode.