Detection of Endoleaks after Endovascular Repair of Abdominal Aortic Aneurysm

Detection of Endoleaks after Endovascular Repair of Abdominal Aortic Aneurysm

Alla M. Rozenblit, MD ; Michael Patlas, MD ; Ayala T. Rosenbaum, MD ; Takao Okhi, MD ; Frank J. Veith, MD ; Mitchell P. Laks, MD ; Zina J. Ricci, MD

1 From the Departments of Radiology (A.M.R., M.P., A.T.R., M.P.L., Z.J.R.) and Surgery (T.O., F.J.V.), Albert Einstein College of Medicine and Montefiore Medical Center, 111 E 210th St, Bronx, NY 10467. Received May 14, 2002; revision requested July 3; revision received August 5; accepted September 25.

PURPOSE: To assess unenhanced and delayed phase computed tomographic (CT) images combined with arterial phase images for detecting endoleaks after endovascular treatment for abdominal aortic aneurysm (AAA).

MATERIALS AND METHODS: CT scans were retrospectively evaluated for the presence of endoleaks after endovascular treatment of AAAs in 33 patients with endoleak (positive group) and 40 patients without evidence of endoleak or aneurysm enlargement (negative group). All patients underwent unenhanced and biphasic contrast material–enhanced CT. The CT scans were reviewed in the following combinations: (a) arterial phase and unenhanced scans (uniphasic/unenhanced set), (b) arterial and delayed phase scans only (biphasic set), and (c) arterial and delayed phase scans with unenhanced scans (complete set). Each set was reviewed by two radiologists blinded to the diagnosis of endoleak. Findings were recorded as positive, negative, or indeterminate for endoleak.

RESULTS: Within the positive group, endoleaks were diagnosed with the uniphasic/unenhanced, biphasic, and complete image sets in 30 (91%), 32 (97%), and 33 (100%) patients, respectively. With the uniphasic/unenhanced set, three (9%) endoleaks (seen only on delayed phase images) were missed. With the biphasic set, one (3%) endoleak was interpreted as indeterminate. Within the negative group, uniphasic/unenhanced, biphasic, and complete image sets were negative for endoleaks in 100%, 80%, and 100% of patients, respectively. With the biphasic set, results were indeterminate in 20% of cases.

CONCLUSION: A delayed CT acquisition enables detection of additional endoleaks, while an unenhanced acquisition helps eliminate indeterminate results. Thus, both acquisitions contribute to accurate diagnosis of endoleaks when combined with an arterial phase acquisition.