Can computed tomography scan findings predict “impending” aneurysm rupture?

Abstract

Listen

Translate article

Conclusion: Computed tomography (CT) signs of “impending rupture” are poor predictors of short-term aortic aneurysm rupture. Summary: A number of signs that are present on CT scans have been advocated in the radiologic literature as indicating increased risk of rupture of abdominal aortic aneurysms and have been suggested to indicate “imminent” or “impending” aneurysm rupture (AJR 1994;163:1123-9). These signs include the “crescent sign,” a curvilinear area of higher attenuation within the aneurysm that may represent hemorrhage into mural thrombus; ill defined or irregular aortic walls, discontinuity of circumaortic calcification, aortic bulges or blebs, and draping of the aorta over the spine. This study reviewed a medical records system from August 1994 to August 2004 looking for CT scan reports of impending rupture. Medical records were also reviewed for patient demographics, symptoms, comorbidities, and documented subsequent aneurysm rupture, operative findings, or both. If a rupture occurred ≤2 weeks of the index CT scan, it was defined as supporting the imminent rupture diagnosis. The record retrieval system identified 45 patients with aortic aneurysms and CT stigmata of impending rupture. Six patients were excluded because five had additional signs suspicious for leak and another had a previously repaired aneurysm and associated infection. Therefore, 39 intact aneurysms were evaluated, of which 26 (67%) were infrarenal, two (5%) were suprarenal, and 11 (28%) were thoracoabdominal. Mean aneurysm diameter was 6.8 ± 1.4 cm. Mean patient age was 74 years, and 49% were women. Ten patients underwent elective repair during the first 2 weeks that precluded adequate observation for early rupture. None of these patients had intraoperative signs of rupture. Two of the remaining 29 patients suitable for more long-term evaluation ruptured ≤72 hours of the CT scan. The positive predictive value of the diagnosis of impending rupture was therefore 6.9%. One other patient’s aneurysm ruptured 7 months after the initial CT scan. Twenty-six patients whose aneurysms did not rupture were observed over an average length of follow-up of 246 days (range, 14 days to 3 years). Of these patients, 14 were repaired electively 2 to 3 weeks after the CT scan, and 12 never underwent repair. These 12 were observed for a mean of 394 days without rupture. Comment: Every vascular surgeon will at one time be faced with the dilemma of what to do with a patient with a large abdominal aortic aneurysm and a radiologic diagnosis of impending rupture without clear evidence of leakage from the aneurysm. Since the risk of acute rupture in these patients seems quite low, the authors’ data suggest that an approach of expedient but not emergent operation is reasonable. Nevertheless, the weaknesses of these data are obvious in that only patients who had a diagnosis of impending rupture included in their official CT report were in the database. The data would have been more complete if the authors had also evaluated all patients with aortic aneurysm rupture and a previous CT scan. Perhaps some of these patients had signs of impending rupture that were not included in the official radiologic report.