Endovascular Retreatment of Previously Clipped Intracranial Aneurysms: An Individual Patient Data Meta-analysis

BACKGROUND AND OBJECTIVES: To evaluate the effectiveness and safety of endovascular retreatment of previously clipped intracranial aneurysms (IAs). METHODS: Systematic searches of Medline, Embase and Cochrane Central were performed. The primary outcome was good functional outcome (modified Rankin Scale 0-2) at the last follow-up. Secondary outcomes included complete occlusion, and perioperative complications. One-stage individual patient data meta-analysis was performed, adjusted using generalized linear mixed models with prespecified covariables and study as a random effect. RESULTS: Twenty-six studies reporting 413 patients with 414 previously clipped IAs retreated endovascularly were included. Totally, 249 (60.1%), 39 (9.4%), 39 (9.4%), 81 (19.6%) and 6 (1.4%), underwent simple coiling, balloon-assisted coiling, stent-assisted coiling, flow diversion, and disruption, respectively. Of these, 88.6% and 11.4% IAs were located anteriorly and posteriorly. Mean size of the IAs at retreatment was 6.8 mm. Most (70.4%) of previously clipped IAs were unruptured at retreatment, mainly due to regrowth (n = 48), remnants (n = 49) or recurrences (n = 41), when reported. With clinical follow-up of 26.8 months, proportions of good functional outcomes, improved/unchanged neurological outcomes, and deaths were 77.3% (95%CI: 72.15; 81.74), 88.2% (95%CI: 64.57; 96.85), and 6.9% (95%CI: 4.19; 11.18), respectively. With angiographical follow-up of 19.4 months, the proportion of complete occlusion was 74.4% (95%CI: 64.38; 82.29). Multivariate analyses showed that a prolonged interval (>1 month) between clipping and endovascular retreatment was associated with good functional outcome (odds ratio 7.37, 95%CI: 2.16; 29.94) whereas posteriorly located IAs were associated with perioperative complications (odds ratio 8.05, 95%CI: 1.45; 50.48). CONCLUSION: Endovascular retreatment of previously clipped IAs can be accomplished safely and effectively in well-selected patients. The indications for retreatment need to be carefully weighed against the natural history of a previously clipped IA.