Association between RoPE score and PFO grading on bubble echocardiography in cryptogenic stroke patients: a retrospective cohort study

Abstract

Introduction: Identification of high-risk anatomical and physiological features of a patent foramen ovale (PFO) is important for patient selection for transcatheter device closure of PFO in patients with cryptogenic stroke. Currently, there are no clinical screening tools in use that can be used in predicting high-risk PFO features before undertaking transoesophageal echocardiography.

Methods: This retrospective cohort study, conducted in a stroke unit in South East England, included 130 patients diagnosed with ischaemic stroke or transient ischaemic attack who were deemed as cryptogenic in nature following initial evaluation (≤55 years with no known risk factors or immediately identified underlying etiology). Patients underwent comprehensive diagnostic evaluations, including bubble echocardiography. The primary predictor, risk of paradoxical embolism (RoPE) score (≥6), was assessed for its association with a significant PFO, categorized as model 1 (≥small) and model 2 (≥moderate). Multivariable logistic regression models were used to estimate adjusted odds ratios for the relationship between RoPE score and PFO presence.

Results: Of the 130 patients, 47 had a known etiology, and 83 had cryptogenic stroke. The known etiology group had higher rates of hypertension, hyperlipidaemia, and non-stenotic atherosclerosis, while the cryptogenic group had more cortical strokes and higher RoPE scores. Multivariable analysis showed that a lower RoPE score (≤5) was associated with known etiology (aOR: 3.91, p < 0.01). RoPE scores ≥6 were significantly associated with both small and moderate PFOs (aORs: 5.39, p < 0.01 and 15.95, p < 0.01, respectively). Of 28 candidates for PFO closure, 20 underwent the procedure, all with high RoPE scores and large PFOs.

Discussion: This study reinforces the importance of a multidisciplinary approach in the evaluation and management of patients with PFO and suspected embolic stroke. While PFO is prevalent in both cryptogenic and non-cryptogenic stroke patients, its pathogenic role is highly context dependent. Our findings confirm that a high RoPE score (≥6) and a cortical stroke phenotype are independently associated with clinically relevant, higher-grade PFOs. Furthermore, patients selected for device closure consistently exhibited high RoPE scores and multiple high-risk anatomical features, aligning with current international guidelines. Importantly, low RoPE scores (≤5) were significantly associated with strokes of known etiology, underscoring the utility of the RoPE score not only in identifying likely PFO-related strokes but also in ruling out embolic mechanisms. These results support the integration of clinical scoring systems like RoPE for patient selection about the suitability for device closures as higher RoPE scores predict high-risk PFO and therefore minimize unnecessary interventions.

Conclusion: RoPE scores may be utilized in predicting high-risk anatomical and physiological features of PFO. However, larger prospective studies are needed to validate these findings and refine pre-transoesophageal echocardiography screening tools.