BACKGROUND: Controversy exists regarding the best type of arteriovenous (AV) fistula to be formed in secondary and tertiary access procedures when primary fistulas have failed. This meta-analysis aimed to compare transposed brachial-basilic AV fistulas (BBAVFs) with upper limb AV prosthetic grafts. METHODS: A literature search of the MEDLINE and SCOPUS databases was performed to identify comparative studies reporting outcomes for both BBAVFs with upper limb AV prosthetic grafts. Meta-analysis techniques were applied to identify differences in outcomes between the two groups regarding primary and secondary 1-year failure rates. RESULTS: Eleven relevant studies, involving 1509 patients, met the inclusion criteria and were incorporated in the final analysis; however, only one was randomised controlled trial. The pooled odds' ratio (OR) estimate for the primary and secondary failure rates at 1 year was 0.67 (CI 0.41-1.09) and 0.88 (CI 0.69-1.12), respectively, showing no difference in the outcome between the two groups. The re-intervention rate was higher for prosthetic grafts (0.54 per BBAVF versus 1.32 per graft). In a small subgroup of two studies comparing BBAVFs with forearm grafts the pooled estimate for 1-year primary failure rate was in favour of the BBAVF group (OR 0.3, CI 0.15-0.58, p=0.0004) suggesting that forearm grafts were inferior having a 3-fold risk of failure at 1 year. CONCLUSION: This analysis supports the use of BBAVF early in difficult access cases prior to the use of prosthetic grafts. However, the latter conclusion is debatable due to heterogeneity, small size and non-randomised design of the included studies.
BACKGROUND: Controversy exists regarding the best type of arteriovenous (AV) fistula to be formed in secondary and tertiary access procedures when primary fistulas have failed. This meta-analysis aimed to compare transposed brachial-basilic AV fistulas (BBAVFs) with upper limb AV prosthetic grafts. METHODS: A literature search of the MEDLINE and SCOPUS databases was performed to identify comparative studies reporting outcomes for both BBAVFs with upper limb AV prosthetic grafts. Meta-analysis techniques were applied to identify differences in outcomes between the two groups regarding primary and secondary 1-year failure rates. RESULTS: Eleven relevant studies, involving 1509 patients, met the inclusion criteria and were incorporated in the final analysis; however, only one was randomised controlled trial. The pooled odds' ratio (OR) estimate for the primary and secondary failure rates at 1 year was 0.67 (CI 0.41-1.09) and 0.88 (CI 0.69-1.12), respectively, showing no difference in the outcome between the two groups. The re-intervention rate was higher for prosthetic grafts (0.54 per BBAVF versus 1.32 per graft). In a small subgroup of two studies comparing BBAVFs with forearm grafts the pooled estimate for 1-year primary failure rate was in favour of the BBAVF group (OR 0.3, CI 0.15-0.58, p=0.0004) suggesting that forearm grafts were inferior having a 3-fold risk of failure at 1 year. CONCLUSION: This analysis supports the use of BBAVF early in difficult access cases prior to the use of prosthetic grafts. However, the latter conclusion is debatable due to heterogeneity, small size and non-randomised design of the included studies.