Current practice guidelines recommend surgical repair of large thoracic aortic aneurysms to prevent fatal aortic dissection or rupture, but limited natural history data exist to support clinical criteria for timely intervention.

Of 3247 patients with thoracic aortic aneurysm registered in our institutional Thoracic Aortic Center Database, we identified and reviewed 257 nonsyndromic patients (age, 72.4±10.5 years; 143 female) with descending thoracic or thoracoabdominal aortic aneurysm without a history of aortic dissection in whom surgical intervention was not undertaken. The primary end point was a composite of aortic dissection/rupture and sudden death. Baseline mean maximal aortic diameter was 52.4±10.8 mm, with 103 patients having diameters ≥55 mm. During a median follow-up of 25.1 months (quartiles 1–3, 8.3–56.4 months), definite and possible aortic events occurred in 19 (7.4%) and 31 (12.1%) patients, respectively. On multivariable analyses, maximal aortic diameter at baseline emerged as the only significant predictor of aortic events (hazard ratio=1.12; 95% confidence interval, 1.08–1.15). Estimated rates of definite aortic events within 1 year were 5.5%, 7.2%, and 9.3% for aortic diameters of 50, 55, and 60 mm, respectively. Receiver-operating characteristic curves for discriminating aortic events were higher for indexed aortic sizes referenced by body size (area under the curve=0.832–0.889) but not significantly different from absolute maximal aortic diameter (area under the curve=0.805).

Aortic size was the principal factor related to aortic events in unrepaired descending thoracic or thoracoabdominal aortic aneurysm. Although the risk of aortic events started to increase with a diameter >5.0 to 5.5 cm, it is uncertain whether repair of thoracic aortic aneurysms in this range leads to overall benefit, and the threshold for repair requires further evaluation.