From March 1981 to March 1990, 61 patients with Stanford type B acute aortic dissection were initially treated by conservative therapy. Among these 61 patients, the dissected lumen became occluded due to thrombosis early after diagnosis in 25 patients (Group T) and remained patent in 36 patients (Group P). Twentythree patients in Group T (92%) and 22 patients in Group P (61%) were discharged without major complications related to acute aortic dissection. However, 2 patients in Group T (8%) and 14 patients in Group P (39%) required additional surgical therapy or died during hospitalization. The mean aortic diameter at the time of admission in Group T was smaller than that of Group P (38±3mm vs 43±7mm, p<0.05). During the observation period, there was a tendency for the diameter of the dissected aorta in Group T to decrease, but to increase in Group P. Long-term survival appeared to be better in Group T than in Group P, but there was no significant difference in the overall survival curve. Large aortic diameter at the time of admission and the presence of a true thoracic aortic aneurysm were major contributing factors influencing the prognosis. A long-term follow-up study showed that the dissected lumen reduced or disappeared in 14 of 23 patients in Group T (61%) but only 2 of 16 patients in Group P (12.5%). We concluded that the patients with small dissected aortas and thrombosed dissected lumens (Group T) can recuperate only with conservative therapy. However, patients with large dissected aortas and patent dissected lumen (Group P) may require surgical therapy even in Stanford type B aortic dissection.