Complication rates associated with antegrade use of vascular closure devices: a systematic review and pooled analysis

Vascular closure devices (VCDs) have become a mainstay in endovascular interventions in recent years. On-label use only allows for retrograde interventions originating at the common femoral artery (CFA). However, off-label use in antegrade and/or superficial femoral artery (SFA) access has become more common in the treatment of peripheral arterial disease. Despite this, there is a paucity of comparative literature assessing individual VCD safety profiles and relative complication risks of CFA vs SFA applications.

MEDLINE and EMBASE systematic searches were performed from inception to October 2019 to identify studies assessing VCD use in antegrade CFA and SFA interventions. Abstract selection and data extraction were performed by two independent reviewers. Primary outcomes included bleeding-related complications, vessel occlusion or stenosis, embolization, pseudoaneurysm formation, or arteriovenous fistula formation.

Twenty-four unique studies with 4124 vascular closure events via six unique VCDs were included (Angioseal, Exoseal, Femoseal, Glubran 2, Mynx, Starclose; 3698 CFA, 426 SFA). Pooled occurrence of all complications across closure devices used in the CFA and SFA ranged from 0.9% (Mynx) to 7.4% (Starclose) and 0% (Perclose) to 10.1% (Starclose), respectively, depending on VCD type. Clinically meaningful differences were identified between devices with a trend toward significance (P = .08 CFA, P = .07 SFA). Individual devices demonstrated a range of bleeding complication rates from 0.4% (Mynx) to 7.2% (Femoseal) for the CFA site and from 0% (Perclose) to 6.4% (Starclose) for the SFA site (P = .01 and P = .03, respectively). Significant heterogeneity between studies precludes definitive characterization of the device as cause for variations in bleeding complication rates. Pooled complication rates did not differ between the CFA and SFA arms (4.6% vs 5.8%, P = .56). Bleeding complication rates also did not differ between the CFA and SFA arms (3.6% vs 3.6%, P = .98).

Clinically meaningful differences in overall pooled complications were identified between VCDs with a trend toward significance. Significant differences between VCDs exist with respect to bleeding risk. However, this must be interpreted with caution as these differences could be secondary to interstudy heterogeneity. Finally, no difference was identified between antegrade SFA and CFA VCD use with respect to overall complication and bleeding risks.