Structured Follow-up Imaging is Useful after Thoracic Aortic Surgery

Purpose: Although postoperative follow-up imaging after aortic surgery is recommended by guidelines, its clinical utility is not well documented. We implemented a structured follow-up imaging protocol at our Aortic Center in 2017, and sought to describe clinical outcomes and radiologic findings through this program.

Methods: All patients who survived to discharge after open thoracic aortic surgery for aneurysm (n=466) or dissection (n=93) between 01/2017-07/2021 were included. Prior to discharge, all received instruction on clinical and imaging follow-up per our Aortic Center protocol. The extent of aortic replacement was root in 301 (53.8%), ascending in 28 (5.0%), arch in 103 (18.4%), and root-to-arch in 128 (22.9%). Of 559 patients, 179 patients (32.0%) followed at our center and received scheduled imaging and were compared against patients who did not. Survival was analyzed by the method of Kaplan and Meyer with follow-up status as a time dependent covariate to account for the inherent survival bias associated with postoperative follow-up (“immortal time bias”). Reintervention was assessed with death as a competing risk using the Fine-Gray subhazard function. Scheduled imaging was reviewed for aortic growth, pseudoaneurysm, and perigraft density, and the incidence and outcomes of these findings was documented.

Results: After aortic surgery, the cumulative incidence of follow-up imaging was 31.5%, 36.9%, and 38.6% at 1, 2, and 3 years postoperatively. Patients with follow-up were more likely to have a dissection and fewer comorbidities, but were similar in regards to socioeconomic factors and distance to hospital. After matching and accounting for survival bias, patients with follow-up had a higher reintervention rate (26.0% vs. 9.0%, p=0.036) with similar survival (98.7% vs. 95.2%, p=0.110) at four years (Figure A, B). For aneurysm patients, the cumulative incidence of aortic growth ≥3 mm/year was 9.0% at 1 year, 21.1% at 2 years, and 29.1% at 3 years for patients with aneurysms. For dissection, the cumulative incidence of growth ≥3 mm/year was 59.8% at one year and 72.9% at two years. A pseudoaneurysm was noted in the scheduled scans of 6 patients (2.8%), resulting in one reoperation. Perigraft density was observed on all postoperative CT images and generally decreased in size on subsequent imaging; of 11 patients with a perigraft density ≥20 mm, one developed graft infection requiring reoperation. The cumulative incidence of the composite outcome of pseudoaneurysm, perigraft density ≥20 mm, and growth ≥3 mm/year on imaging was 47.9% at three years (Figure C).

Conclusion: Implementation of a structured follow-up program resulted in low compliance. Follow-up imaging was associated with increased rates of aortic reintervention. Clinically relevant radiologic findings were common on scheduled imaging and were continually detected throughout the 5-year follow-up period. Our study provides novel data to support structured follow-up imaging after thoracic aortic surgery.

Identify the source of the funding for this research project: None