Carotid artery revascularization using second generation stents versus surgery: a meta-analysis of clinical outcomes

Introduction: Meta-analyses and emerging randomized data indicate that second-generation ('mesh') carotid stents (SGS) may improve outcomes versus conventional (single-layer) stents but clinically-relevant differences in individual SGS-type performance have been identified. No comparisons exist for SGS versus carotid endarterectomy (CEA). Evidence acquisition: Thirty-day death (D), stroke (S), myocardial infarction (M), and 12-month ipsilateral stroke and restenosis in SGS studies were meta-analyzed (random effect model) against CEA outcomes. Eligible studies were identified through PubMed/EMBASE/COCHRANE. Forest plots were formed for absolute adverse evet risk in individual studies and for relative outcomes with each SGS deign versus contemporary CEA outcomes as reference. Meta-regression was performed to identify potential modifiers of treatment modality effect. Evidence synthesis: Data were extracted from 103,642 patients in 25 studies (14 SGS-treated, 41% symptomatic; nine randomized controlled trial (RCT)-CEA-treated, 37% symptomatic; and two Vascular Quality Initiative (VQI)-CEA-treated, 23% symptomatic). Casper/Roadsaver and CGuard significantly reduced DSM versus RCT-CEA (-2.70% and -2.95%, P<0.001 for both) and versus VQI-CEA (-1.11% and -1.36%, P<0.001 for both). Gore stent 30-day DSM was similar to RCT-CEA (P=0.581) but increased against VQI-CEA (+2.38%, P=0.033). At 12 months, Casper/Roadsaver ipsilateral stroke rate was lower than RCT-CEA (-0.75%, P=0.026) and similar to VQI-CEA (P=0.584). Restenosis with Casper/Roadsaver was +4.18% vs. RCT-CEA and +4.83% vs. VQI-CEA (P=0.005, P<0.001). CGuard 12-month ipsilateral stroke rate was similar to VQI-CEA (P=0.850) and reduced versus RCT-CEA (-0.63%, P=0.030); restenosis was reduced respectively by -0.26% and -0.63% (P=0.033, P<0.001). Twelve-month Gore stent outcomes were overall inferior to surgery. Conclusions: Meta-analytic integration of available clinical data indicates: 1) reduction in stroke but increased restenosis rate with Casper/Roadsaver, and 2) reduction in both stroke and restenosis with CGuard MicroNET-covered stent against contemporary CEA outcomes at 30 days and 12 months used as a reference. This may inform clinical practice in anticipation of large-scale randomized trials powered for low clinical event rates (PROSPERO-CRD42022339789).