Because the long-term security of endovascular treatments remains uncertain, a follow-up study of the patients treated in the International Subarachnoid Aneurysm Trial was performed to compare the frequency, timing, and consequences of aneurysm recurrence. Patient data were reclassified by actual treatment performed. Aneurysm and patient characteristics, including occlusion grades, time and type of retreatment, and clinical outcomes, were compared. The relationship between these variables and late retreatment as a surrogate for recurrence was analyzed by means of the Cox proportional hazards model. Retreatment was performed in 191 of 1096 (17.4%) patients after primary endovascular coiling (EVT) and in 39 of 1012 patients (3.8%) after neurosurgical clipping. After EVT, 97 (8.8%) patients were retreated early and 94 (9.0%) late, 7 (0.6%) after rebleeding and 87 (8.3%) without. The mean time to late retreatment was 20.7 months. After neurosurgical clipping, 30 (2.9%) patients were retreated early and 9 (0.85%) late, 3 (0.3%) after rebleeding and 6 (0.6%) without. The mean time to late retreatment was 5.7 months. The hazard ratio (HR) for retreatment after EVT was 6.9 (95% CI=3.4 to 14.1) after adjustment for age (P=0.001, HR=0.97, 95% CI=0.95 to 0.98), lumen size (P=0.006, HR=1.1, 95% CI=1.03 to 1.18), and incomplete occlusion (P<0.001, HR=7.6, 95% CI=3.3 to 17.5). Late retreatment was 6.9 times more likely after EVT. Younger age, larger lumen size, and incomplete occlusion were risk factors for late retreatment after EVT. After neurosurgical clipping, retreatments were earlier; whereas EVT retreatments continued to be performed throughout the follow-up period. Short-term follow-up imaging is therefore insufficient to detect recurrences after EVT.
