Back to Karmody Posters

OBJECTIVE: After endovascular aortic aneurysm repair (EVAR), the Society for Vascular Surgery recommends computed tomography (CT) scan within 30 days, followed by annual imaging. We sought to describe long-term adherence to surveillance guidelines among US Medicare beneficiaries, and determine patient and hospital factors associated with incomplete surveillance. METHODS: We analyzed fee-for-service Medicare claims for patients receiving EVAR from 2002 to 2005, and then collected all relevant postoperative imaging through 2009. Additional data included patient comorbidities and demographics, yearly hospital AAA repair volume, and Medicaid eligibility. Allowing a grace period of three months, complete surveillance was defined as at least one CT or ultrasound every 15 months after EVAR. Incomplete surveillance was categorized as gaps for intervals >15 months between consecutive images, or lost to follow-up if >15 months elapsed after last imaging. RESULTS: Our cohort comprised 9,503 patients. Median follow-up duration was 4.5 years. CT scan within 30 days of EVAR was performed in 3,090 (32.5%) patients. Median time to post-operative CT was 37 days (IQ range 25 - 94 days). Complete surveillance was observed in 51.0% of patients. For this group mean follow-up time was shorter than for those with incomplete surveillance (3.3 ± 2.1 vs. 5.1 ± 1.5 years, p<0.001). Among those with incomplete surveillance, follow-up became incomplete at 2.8 ± 1.3 years, with 53.0% lost to follow-up, 60.9% with gaps in follow-up (mean gap length 760 ± 325 days), and 13.9% with both. Patients with incomplete surveillance had fewer images in each postoperative year (Table 1), and fewer total images during follow-up (5.4 ± 3.6 vs. 6.8 ± 4.8, p<0.001). In a multivariable logistic regression, complete surveillance was less likely for patients with Medicaid eligibility (OR 0.79, 95% CI 0.68 - 0.92, p=0.002) and those treated in high volume hospitals (OR 0.89, 95% CI 0.80 - 0.99, p=0.039). CONCLUSIONS: Postoperative imaging after EVAR is highly variable and only a small proportion of patients meet current surveillance guidelines. Additional studies are necessary to determine if variability in postoperative surveillance impacts long-term outcomes.

Mean number of images per follow-up year by surveillance type

Years after EVAR	Complete Surveillance	Incomplete Surveillance	P value
0 - 1	2.54	2.07	<0.001
1 - 2	1.66	0.94	<0.001
2 - 3	1.62	0.81	<0.001
3 - 4	1.67	0.69	<0.001
4 - 5	1.28	0.59	<0.001
5 - 6	1.28	0.58	<0.001
6 - 7	1.15	0.47	<0.001
7 - 8	0.84	0.31	<0.001