Improved Visualization of Heart Valves With 4D-CT Using a Novel Motion Correction Algorithm

Abstract

OBJECTIVE: Motion artifacts remain a limitation of cardiac CT when assessing both native and prosthetic heart valves, particularly during phases of rapid motion. This may lead to reduced image quality, potentially impairing the evaluation of valvular movement and pathology. The purpose of this study was to evaluate whether a novel reconstruction-based motion correction algorithm improves image quality and reduces motion artifacts of native and prosthetic heart valves compared with standard reconstructions in photon-counting detector CT (PCD-CT). MATERIALS AND METHODS: In this retrospective single-center study, 57 patients [21 women; median age, 71 (interquartile range, 57 to 78 y)] who underwent clinically indicated 4D cardiac CT for suspected native and prosthetic valvular disease were included. Images were acquired using retrospectively electrocardiography (ECG)-gated spiral dual-source PCD-CT without ECG-pulsing. Images were reconstructed in 10% steps throughout the RR-interval using both standard (TrueStack) and an otherwise identical reconstruction using the novel motion correction algorithm (ZeeFreePlus). Subjective image quality and motion artifacts were independently assessed by 2 blinded, experienced readers using a 4-point visual analog scale for 3 categories: valve frame/annulus, valve leaflets, and valvular pathology. Objective analysis included quantification of the volume of artifacts throughout the cardiac cycle. RESULTS: A total of 68 valves were evaluated, including 57 prosthetic valves (22 transcatheter aortic valve replacements and 19 surgical valves, 11 mitral valves, and 5 tricuspid valve interventions) and 11 native valves. Valvular pathology was present in 29 cases, with the most prevalent being hypoattenuated leaflet thickening in 14/29 (48%). Motion-corrected reconstructions significantly improved image quality and reduced motion artifacts across all cardiac phases and all categories (all, P<0.001). The largest improvements were observed during systole, particularly at 10% of the RR-interval, with significantly greater gains than during diastole (P<0.05). Improvements were consistent across aortic and atrioventricular prosthetic valves and translated into improved visualization of valvular pathology (P<0.001). Objective quantitative analysis corroborated the subjective image quality findings by demonstrating reductions in the volumes of motion-induced artifacts when using the algorithm (P<0.001). CONCLUSIONS: A novel motion correction algorithm substantially reduced motion artifacts and improved image quality in both native and prosthetic heart valve imaging with PCD-CT, with the largest improvements in systole. These benefits resulted in enhanced visualization of valvular pathology, potentially reinforcing the growing clinical role of cardiac CT for the assessment of valve disease.