Systematic review of preservation methods and clinical outcome of infrainguinal vascular allografts

Objective: We systematically reviewed clinical studies on the use of venous and arterial allografts for infrainguinal revascularization. We attempted to find evidence for the best infrainguinal vascular allograft by a systematic review of the available literature. Methods: An electronic search of the MEDLINE, EMBASE, and Cochrane databases was used to determine key articles from studies on the different types of vascular allografts used in infrainguinal reconstruction from 1966 to 2004. Articles were independently reviewed by using previously defined inclusion and exclusion criteria. Study results were gathered with cumulative primary patency as the primary end point. Secondary end points were major complications, graft disintegration, and major limb loss. Quantitative analysis was performed on the prospective randomized trials, and linear regression analysis was performed on cumulative primary patency. Fontaine's classification system was applied. Results: No systematic review of randomized controlled trials was found. Five randomized controlled trials, 3 prospective cohort or case series, and 15 retrospective case series with 3837 vascular allografts were found. Methods of allograft preservation were cryopreservation (5 studies), cold storage (3 studies), and glutaraldehyde preservation of human umbilical veins (15 studies). One-year cumulative primary patency rates were 13% to 79% for cryopreservation, 63% to 80% for cold storage, and 40% to 91% for glutaraldehyde. The weighted mean 1-year cumulative primary graft patency rate was 41% for cryopreservation, 71% for cold storage, and 70% for glutaraldehyde allografts. Four randomized trials on femoropopliteal bypasses demonstrated higher patency rates of glutaraldehyde-preserved human umbilical veins than polytetrafluoroethylene grafts. Statistical heterogeneity between studies (I-2 = 91.4%) was too high to perform a formal meta-analysis. The rate of major limb loss was 20% to 58% for cryopreservation, 10% to 69% for cold storage, and 0% to 65% for glutaraldehyde, and the percentage of graft disintegration was 2% to 6% for cryopreservation, 4% to 15% for cold storage, and 0% to 11% for glutaraldehyde. Conclusions: A firm conclusion could not be made because there were no studies available in which direct comparison was performed between different preservation methods of vascular allografts. In addition, heterogeneity of the individual studies hampered direct comparison of different types of vascular allografts. However, the overall graft performance of glutaraldehyde-preserved human umbilical vein allografts may be superior to that of other vascular allografts.