Abstract
Background
Computed tomography (CT) imaging is considered as standard of care for transcatheter aortic valve replacement (TAVR) prothesis sizing. ECG-triggered high-pitch helical CT data acquisition on dual-source CT systems is associated with less contrast volumes, but CT image reconstruction of the aortic valve is limited to a single time point within the cardiac cycle. Although systolic CT imaging of the aortic valve is recommended due to the changes in aortic annulus area during the cardiac cycle, this recommendation is not supported by clinical outcome data.
Purpose
The study aimed to assess the impact of systolic vs. diastolic CT imaging of the aortic annulus for TAVR sizing on patient outcomes in a large series of patients.
Methods
In the study CT images of 1346 patients undergoing TAVR at our centre from 2013–2016 were re-evaluated. Patients were stratified into two groups with 0–35% and 36–99% of the RR-interval as systolic and diastolic CT imaging. Outcomes according to VARC2-criteria at 30 days and long-term survival were analysed.
Results
CTs of 1135 out of 1346 patients were analysed retrospectively, 278 (24.5%) of which were acquired during systole and 859 (75.5%) during diastole. Mean age was 80.6±7.6 years and 52.2% were female. Mean follow-up, available for 83.4% of patients, was 2.1±0.8 years. No significant difference in baseline characteristics was observed across both groups.
Aortic annular area measurements were significantly larger in systole (mean aortic anulus area: systole: 4.8±1.0 cm2; diastole: 4.5±1.0 cm2, p<0.01), resulting in larger implanted valves. Balloon-expandable valves were used in 69.4% of patients; the need for post-dilation did not differ between both groups (systole 4.7%, and diastole 7.5%; p=0.13).
There was no difference between groups concerning the combined endpoints for device success or early safety according to VARC2-criteria. Overall 30-day mortality was 3.4% (systole: 4.2% and diastole: 3.1%, p=0.35). Device failure occurred in 2.7% (systole: 2.9% and diastole: 2.7%, p=0.83). Permanent pacemaker implantation was required in 20.1% (systole: 18.7%, and diastole: 20.5%; p=0.55). Stroke rate at 30 days was 2.9%, n=33 and was similar in the two groups. The landmark analysis demonstrates the 30-day and 1-year mortality rates for both groups (Figure).
Figure 1. Landmark analysis mortality
Conclusion
The current analysis does not suggest that systolic CT imaging for TAVR sizing is associated with improved early or late outcomes. Accordingly, full-cycle CT imaging which usually is associated with higher contrast volumes, should be avoided in this elderly population, which often presents with reduced kidney function.
The Impact of the Cardiac Cycle on Aortic Annular Dimensions as Measured by Computed Tomography (CT) and the Implications for CT Sizing Criteria in Transcatheter Aortic Valve Replacement