Quantitative assessment and clinical value of right-to-left shunt in patent foramen ovale using three-dimensional transesophageal echocardiography combined with right-heart contrast: a prospective study.

BACKGROUND

A patent foramen ovale (PFO) is a common congenital cardiac condition characterized by a persistent communication between the right and left atria, which can facilitate a right-to-left shunt (RLS). This study aimed to determine the diagnostic accuracy, reproducibility, and prognostic value of PFO shunt volume quantified by three-dimensional transesophageal echocardiography (3D-TEE) combined with right-heart contrast echocardiography.

METHODS

In this prospective study, 100 patients with suspected PFO were prospectively enrolled and allocated in a 7:3 ratio to a training cohort (n=70) and a validation cohort (n=30). Each participant underwent two-dimensional transthoracic echocardiography (2D-TTE), two-dimensional transesophageal echocardiography (2D-TEE), and 3D-TEE. The total shunt volume (V_total) was calculated using voxel-based grey-scale integration. The optimal cut-off value was determined in the training set with the Youden index and then applied to the validation set. A multi-modality "hard reference" standard served as the diagnostic gold standard for method comparison. Reproducibility was evaluated with intraclass correlation coefficients (ICCs). All patients were followed for 36 months; stroke or transient ischemic attack (TIA) risk was analyzed with Kaplan-Meier curves and Firth-penalized Cox models, and changes in migraine frequency and residual shunt volume were recorded.

RESULTS

Eight baseline 3D structural indices, including tunnel length, slit width, leaflet excursion and shunt volume, showed no significant difference between cohorts (P>0.05). V_total, V_peak and t_dur demonstrated excellent repeatability (ICC =0.91-0.94; coefficient of variation 6.96-8.27%). The optimal threshold of 22.93 µL yielded an area under the curve (AUC) of 0.92 in the training set and 0.89 in the validation set (DeLong, P=0.27), with a sensitivity of 92%, specificity of 89%, and a diagnostic odds ratio (DOR) of 89.78, outperforming 2D-TTE and 2D-TEE. Kaplan-Meier analysis showed a 3-year stroke/TIA-free survival of 79% for shunt volume ≥22.93 µL versus 96% below the threshold (12 events, log-rank P=0.01); V_total ≥23 µL remained an independent predictor [hazard ratio (HR) =5.27, P=0.04]. Migraine frequency fell significantly over follow-up (P=0.001), and residual shunt volume at six months was lower after closure than with medical therapy (P<0.001).

CONCLUSIONS

3D-TEE with right-heart contrast precisely quantifies PFO shunt volume; a threshold of 22.93 µL is reproducible and predicts both diagnostic status and long-term outcomes, supporting its use in stroke risk stratification, closure decision-making and therapeutic monitoring.