Antiplatelet agents for preventing thrombosis after peripheral arterial bypass surgery

Background Peripheral arterial disease (PAD) may cause occlusions (blockages) in the main arteries of lower limbs. One treatment option is bypass surgery using autologous (the patient's own tissue) vein graft or prosthetic (artificial) graft. A number of factors influence occlusion rates in these patients, including the material used. To prevent graft occlusion patients are usually treated with antiplatelet, antithrombotic drugs, or a combination of both. Objectives To determine the effects of antiplatelet agents for the prevention of thrombosis in people with lower limb atherosclerosis who were undergoing femoropopliteal or femorodistal bypass grafting. Outcomes included the overall success of therapy (graft patency and limb salvage rates) and complications of treatment. Search methods For this update the Cochrane Peripheral Vascular Diseases Group Trials Search Co-ordinator searched the Specialised Register (last searched June 2014) and the Cochrane Central Register of Controlled Trials (CENTRAL) (2014, Issue 5). We sought additional trials through screening the reference lists of relevant papers. Selection criteria Two review authors, RB and AL, independently reviewed studies found in the search and evaluated them based on the inclusion and exclusion criteria, resolving disagreements through discussion. Data collection and analysis RB and AL independently extracted details of the selected studies for the update. We compared the treatment and control groups for important prognostic factors and differences described. If any data were unavailable, we sought further information from study authors. We synthesised data by comparing group results. We addressed unit of analysis issues by subgroup analysis. Main results We include 16 studies with 5683 randomised participants. Nine different treatment groups were evaluated: aspirin (ASA) or aspirin and dipyridamole (ASA/DIP) versus placebo or nothing (six studies); ASA or ASA/DIP versus pentoxifylline (two studies); ASA/DIP versus indobufen (one study); ASA or ASA/DIP versus vitamin K antagonists (two studies); ASA/DIP versus low molecular weight heparin (one study); ticlopidine versus placebo (one study); ASA versus prostaglandin E1 (one study); ASA versus naftidrofuryl (one study); and clopidogrel and ASA versus ASA alone (one study). The treatment comparisons were evaluated separately, and, where possible, we performed subgroup analysis for venous grafts and prosthetic grafts and at different follow-up time points. The quality of evidence was low to moderate as many of the treatment comparisons had very few studies to contribute data, several of the included studies had unit of analysis issues, the treatment dosages varied between studies, and data for many outcomes important to this review were not given in any of the studies, or differed greatly between studies. Overall study quality was moderate, with the largest problem being that the majority of studies did not describe their methods of randomisation, allocation concealment or blinding of outcome assessors, leading to risk ratings of 'unclear'. The other main issue with study quality was studies not blinding participants or personnel. The treatment comparison with the most number of included studies, which allowed for robust conclusions, was that of aspirin (ASA) or ASA and dipyridamole (ASA/DIP) versus placebo or nothing, covered by six studies. For this treatment group, there was improved graft patency in the ASA or ASA/DIP treatment group, odds ratio (OR) 0.42 (95% confidence interval (CI) 0.22 to 0.83; P = 0.01; 952 participants). This effect was not seen for venous grafts alone at any of the time points, but was observed for all time points in prosthetic grafts, including the final time point of 12 months (OR 0.19, 95% CI 0.10 to 0.36; P <0.00001; 222 participants). Only a single study evaluated secondary patency, for which there was no difference between treatment groups. For the comparison ASA or ASA/DIP versus placebo or nothing there was no difference for any of the side effects, including general, gastrointestinal, bleeding and wound/graft infection. Amputations, cardiovascular events and mortality were also similar between the treatment groups. The comparison of ASA or ASA/DIP versus vitamin K antagonists included two studies, one of which was very large, with over 2000 participants. There were no differences between treatment for primary graft patency at three, six, 12 or 24 months, and there was also no evidence of a difference for limb amputation, cardiovascular events or mortality. One large study (851 participants) evaluated clopidogrel and ASA versus ASA alone, and for all grafts there was no evidence of a difference of primary patency at 24 months. There was evidence of increased total bleeding in the clopidogrel and ASA group (OR 2.65, 95% CI 1.69 to 4.15) from an increase in mild (OR 2.34, 95% CI 1.37 to 4.00), and moderate bleeding (OR 4.13, 95% CI 1.37 to 12.45), but no difference in severe or fatal bleeding. There was no difference between the treatment groups for limb amputation or mortality. For the remaining treatment comparisons there is not currently enough evidence to draw any robust conclusions about the efficacy or safety of the treatment on graft patency after peripheral bypass. Authors' conclusions Antiplatelet therapy with aspirin or with aspirin plus dipyridamole had a beneficial effect on primary patency of peripheral bypass grafts compared to placebo or no treatment. This effect was not evident when evaluating venous grafts alone, but antiplatelet therapy did have a beneficial effect on patency in those who had prosthetic grafts. There was no evidence of differences in side effects (including general, gastrointestinal, bleeding or infection), amputation, cardiovascular events or mortality between the treatment groups. However, the number of participants included in this analysis might be too small to detect a statistically significant effect for side effects, amputation, cardiovascular morbidity or mortality. We found no difference in primary graft patency when aspirin or aspirin with dipyridamole was compared to a vitamin K antagonist or when clopidogrel with aspirin was compared to aspirin alone. However, there was evidence of increase bleeding in the clopidogrel with aspirin group for the latter comparison. The remaining six treatment comparisons did not include enough data to draw any robust conclusions about their efficacy or safety at this time.