A systematic review and meta-analysis of endovascular angiosomal revascularization in the setting of collateral vessels

Endovascular procedures for targeted treatment of lower extremity wounds can be subdivided as direct revascularization (DR), indirect revascularization (IR), and IR via collateral flow (IRc). Although previous systematic reviews assert superiority of DR when compared with IR, the role of collateral vessels in clinical outcomes remains to be defined. This systematic review and meta-analysis aims to define the usefulness of DR, IR, and IRc in treatment of lower extremity wounds with respect to (1) wound healing, (2) major amputation, (3) reintervention, and (4) all-cause mortality. A meta-analysis was performed in accordance with PRISMA guidelines. Ovid MEDLINE was queried for records pertaining to the study question using appropriate Medical Subject Heading terms. Studies were limited to those using DR, IR, or IRc as a primary intervention and reporting information on at least one of the primary outcomes of interest. No limitation was placed on year of publication, country of origin, or study size. Studies were assessed for validity using the Newcastle-Ottawa Scale. Study characteristics and patient demographics were collected. Data representing the primary outcomes—wound healing, major amputation, reintervention, and all-cause mortality—were collected for time points ranging from one month to four years following intervention. A meta-analysis on sample size-weighted data assuming a random effects model was performed to calculate odds ratios (ORs) for the four primary outcomes at various time points. We identified 21 studies for a total of 4252 limbs (DR, 2231; IR, 1647; IRC, 270). Overall wound healing rates were significantly superior for DR (OR, 2.45; P = .001) and IRc (OR, 8.46; P < .00001) compared with, IR with no significant difference between DR and IRc (OR, 1.25; P = .23). The overall major amputation rates were significantly superior for DR (OR, 0.48; P < .00001) and IRc (OR, 0.44; P = .006) compared with IR, with DR exhibiting significantly improved rates compared with IRc (OR, 0.51; P = .01). The overall mortality rates showed no significant differences between DR (OR, 0.89; P = .37) and IRc (OR, 1.12; P = .78) compared with IR, with no significant difference between DR and IRc (OR, 0.54; P = .18). The overall reintervention rates showed no significant difference between DR and IR (OR, 1.05; P = .81), with no studies reporting reintervention outcomes for IRc. Both DR and IRc offer significantly improved wound healing rates and major amputation r