Introduction The Penumbra SMART COIL® System includes a novel generation of embolic coils comprising complex and WAVE shape properties with varying levels of softness to promote dense packing and durable long-term occlusion. We present the final analysis of 997 patients at 68 U.S./Canada sites with up to 1-yr follow-up in the SMART registry to assess the utility of the SMART COIL System in the treatment of intracranial aneurysms and other malformations.

Materials and Methods The SMART registry is a prospective, multi-center registry assessing the embolization of neurovascular lesions using the Penumbra SMART COIL System. Procedures must employ ≥75% of SMART, PC400, or POD coils to meet inclusion criteria. SMART registry endpoints included retreatment rates through 1-yr (±6-mo) follow-up, procedural device-related serious adverse events (SAE), and the ability to achieve adequate occlusion at immediate post-procedure.

Results This analysis of 997 patients (72.1% female; mean age 59.6±13.0 yr) included cerebral aneurysms (91.0%, 907/997); arteriovenous malformations (1.1%); fistulae (4.1%); and other lesions (3.8%). Aneurysms, of which 31.8% were ruptured, were small (≤10 mm) in 85.3% (774/907) of patients, large (>10 mm to 25 mm) in 14.4% (131/907), and giant (>25 mm) in 0.2% (2/907); 63.5% (555/874) of patients had wide-neck aneurysms (dome-to-neck ratio <2 or neck width ≥4 mm).

For all lesions, median coil deployment time, defined as the time from the first coil deployed until the last coil detached, was 16.0 min (IQR 8.0–32.0), and median time of fluoroscopic exposure was 37.0 min (IQR 24.0–56.0). Stent-assisted coiling was performed in 5.7% of patients, and balloon-assisted coiling was performed in 18.7% of patients. Median packing density (calculated for aneurysms only) was 29.1% [IQR 21.0–38.8].

For aneurysms, Raymond Occlusion Class (RROC) I and II was achieved in 79.7% (719/902) of cases at immediate post-procedure and in 90.1% (643/714) at 1-yr follow-up, with a recanalization rate of 12.8% (91/710) and a retreatment rate of 7.1% (52/733). In multivariate models, a large or giant aneurysm size (odds ratio [OR]=2.45, p=0.0030), balloon-assisted coiling (OR=1.84, p=0.0195), ruptured aneurysm status (OR=3.91, p<0.0001), and post-procedural RROC III (OR=1.91, p=0.0180), were predictors of incomplete occlusion (RROC III) at 1-yr follow-up. Large or giant aneurysms (OR=2.05, p=0.0426), smoking (OR=0.5, p=0.0186), and balloon-assisted coiling (OR=2.21, p=0.0111) were also predictive of retreatment through 1-yr follow-up.

For non-aneurysm lesions, complete angiographic occlusion was achieved in 85.2% (23/27) of lesions post-procedure, with a recanalization rate of 6.1% (3/49) and a retreatment rate of 1.9% (1/53) at 1-yr follow-up.

Overall, procedural device-related serious adverse events (SAE) were observed in 2.6% (26/997) of subjects.

Conclusion The final results of the SMART Registry, a prospective registry with 997 subjects, demonstrate that the SMART COIL System achieves safe and adequate embolization in a wide variety of neurovascular lesions with low retreatment rates over 1 yr. Large aneurysm size and balloon-assisted coiling were predictive of incomplete occlusion at 1-yr and retreatment through 1-yr follow-up.

Disclosures A. Spiotta: None. M. Park: None. R. Bellon: None. B. Bohnstedt: None. C. Schirmer: None. R. DeLeacy: None. D. Fiorella: None. B. Woodward: None. B. Baxter: None. S. Kale: None. O. Zaidat: None. P. Sunenshine: None. A. Yoo: None. M. Kabbani: None. K. Liu: None. R. Starke: None. A. Reeves: None. K. Snyder: None. T. Sivapatham: None. T. Dumont: None.