Introduction: Prior studies have reported higher retreatment rates (RR) for cerebral aneurysms (CAs) after endovascular coiling (EC) compared to surgical clipping (SC). However, RRs have largely been derived from studies in the early 2000s and may not represent current practice.
Methods: Using administrative data on all discharges from acute care hospitals in California (2005-2011) and Florida (2005-2014), we identified patients with ruptured and unruptured CAs who were treated with EC or SC. Retreatments within 3 months were excluded, to minimize the inclusion of planned retreatments. Logistic regression was used to assess factors associated with retreatment, and results are presented as OR [95% CI].
Results: Among 19,650 patients with CAs, 12,441 (63.3%) were treated with EC and 7,209 (36.7%) with SC. Mean age was 57 ± 13, 72% were female and 12% were black. Between 2005 and 2014, the use of EC increased relative to SC (50% vs. 81%, p<0.0001). Retreatment occurred in 1488 (7.6%) patients (10.1% vs. 3.2%, EC vs. SC), with 89% of retreatments within 2 years of index treatment. Retreatment was associated with age > 80 (OR 0.3 [0.2-0.5]), female sex (OR 1.6 [1.4-1.9]), black vs. white race (OR 0.8 [0.7-0.9]) and EC vs. SC (OR 3.5 [3.1-4.1]). Adjusted two-year RR decreased from 2005 to 2011 for patients with unruptured CAs treated with EC (8.7% vs. 6.8%, OR 0.95 [0.90-1.00] and Figure). Adjusted RR was unchanged for SC (mean RR 3.6% and 2.7%, unruptured and ruptured) and ruptured CAs treated with EC (mean RR 9.9%).
Conclusions: Analysis of two-year RR of CAs in a large real-world cohort demonstrates a continuous reduction in RR for unruptured CAs treated with EC in the last decade. These findings may reflect improving obliteration rates of EC for unruptured CAs.