Evidence from Cochrane systematic reviews for effects of antithrombotic drugs for lower-limb revascularization. A narrative review

ABSTRACT BACKGROUND: Peripheral arterial disease (PAD) is characterized by progressive narrowing of the arterial lumen, resulting from atherosclerotic plaques. Treatment for PAD aims to control atherosclerosis and improve blood flow. Use of antiplatelet agents and anticoagulants has played important roles in helping to prevent occlusions and stenosis. OBJECTIVE: To evaluate the evidence from Cochrane systematic reviews regarding the accuracy, effectiveness and safety of use of anticoagulants and antiplatelets in lower-limb revascularization, in patients with peripheral arterial disease. METHODS: Systematic reviews found through searches in the Cochrane Library were included. Two authors evaluated whether the reviews found were in line with the inclusion criteria for this investigation. A qualitative synthesis of their findings was presented. RESULTS: Three systematic Cochrane reviews were included. Patients who underwent prosthetic bypass surgery probably presented greater benefit from use of antiplatelets, and patients who underwent vein revascularization probably presented greater benefit from use of anticoagulants. Patients who received endovascular treatment benefited from both antiplatelet and anticoagulant treatment. However, the reliability of the results found was impaired because at the time when these reviews were published, there was no mandatory assessment using the GRADE criteria. CONCLUSION: Despite the evidence found, it is necessary for these reviews to be updated in order to evaluate the degree of certainty of the results found.


INTRODUCTION
Peripheral arterial disease (PAD) is characterized by progressive narrowing of the arterial lumen resulting from atherosclerotic plaques on the artery walls. The disease has a prevalence of 18% in 50-year-old patients, and this reaches 29% in patients over 70 years of age. 1 Its incidence has been increasing over recent decades due to population growth, aging of the population and increased incidence of diabetes mellitus and smoking. 2 These patients are at higher risk of mortality due to acute myocardial infarction (AMI) and stroke. Thus, treatment of PAD offers additional prevention for cardiovascular events. 3 PAD commonly leads to intermittent claudication of the lower limbs, characterized by muscle pain during muscle activity, caused by restricted blood flow to the muscles recruited. This condition improves after a brief rest. As blood flow restriction increases, pain at rest and gangrene arise.
In addition, these patients have significantly reduced quality of life due to restricted mobility. 4 The aim in treating PAD is to control the risk factors for atherosclerosis and bring symptomatic relief through improving blood flow. In more advanced cases of the disease, revascularization of the lower limbs is required through surgery or percutaneous transluminal angioplasty.
Anticoagulant and antiplatelet agents play an important role in helping to prevent occlusions and stenosis, both for patients who have started treatment and for those undergoing revascularization, thus sustaining the clinical improvement. 5,6 Platelets participate in the process of hemostasis and pathogenesis of atherosclerotic disease. The presence of endothelial injuries exposes the subendothelial extracellular matrix to contact with platelets. 7 This mechanism promotes platelet recruitment, adhesion, activation and aggregation, to form a prothrombotic surface that promotes formation of clots and fibrous plaques and the ensuing thromboembolic complications. The latter can lead to acute myocardial infarction, stroke and peripheral vascular occlusions. 7

OBJECTIVES
The objective of this study was to summarize the evidence from Cochrane systematic reviews on the safety and effectiveness of use of antiplatelet agents and anticoagulants in lower limb revascularization in PAD.

Study design and location
This was a narrative review of Cochrane systematic reviews developed within the Evidence-Based Health postgraduate program at Escola Paulista de Medicina (EPM), Universidade Federal de São Paulo (UNIFESP).

Types of study
Only Cochrane systematic reviews were included, and only the latest version of the review was considered. Reviews that had been excluded from the Cochrane Library and systematic review protocols (reviews in progress) were not included.

Types of participants
We included reviews in which the participants presented the diagnosis of chronic PAD of the lower limbs and who had undergone revascularization of the lower limbs by means of venous or prosthetic bypasses or through angioplasty.

Types of interventions
We included studies in which patients underwent any treatment with antiplatelet agents or anticoagulants, and relevant outcomes were assessed in relation to the evolution of the disease.

Types of outcomes
Outcomes relating to the evolution of the disease and to morbidity, mortality and safety of treatments were considered.

Search for studies
We performed a systematic search in the Cochrane Database of Systematic Reviews (via Wiley) on July 11, 2020. The search strategy is detailed in Table 1.

Selection of studies
Two authors (SVMD and RLGF) evaluated and selected the titles and abstracts of systematic reviews regarding their agreement with the eligibility criteria of this study. Any occurrences of disagreements were resolved by a third researcher (WI).

Presentation of results
The findings of the systematic reviews were summarized and presented narratively.

RESULTS
The initial search resulted in retrieval of 370 systematic Cochrane reviews, of which three met the inclusion criteria for this review. [8][9][10] We present, in summary, the main methodological characteristics and the most relevant results of the reviews included. In Table 2, we present a summary of the main findings. Reviews involving medications that are no longer available worldwide for clinical use were excluded from this study.
For each of the studies included, the objectives and the outcomes assessed that were relevant to this review are demonstrated. At the end, we presented the authors' opinion in relation to the result.

VKA versus placebo
Benefits with VKA Occlusion in venous bypasses at 6 months; occlusion in prosthetic bypasses at 5 years; limb loss in venous bypasses at 5 years; limb salvage for venous bypasses at 6 months; limb loss for venous or prosthetic bypasses at 3, 6 and 24 months and 5 years; mortality for venous or prosthetic bypasses at 12 months

No difference among intervention groups
Occlusion in venous bypasses at 3, 12 and 24 months and 5 years; occlusion in prosthetic bypasses at 3, 6, 12 and 24 months; limb loss in venous or prosthetic bypasses at 3, 6, 12, 24 months and 5 years for prosthetic; mortality in venous or prosthetic bypasses at 3, 6, 12 and 24 months and 5 years

ASA plus DIP versus indobufen, for venous bypass or prosthetic bypass
No differences were found among the groups for the following outcomes: primary graft patency at 1, 3, 6, 9 or 12 months (evaluated in 1 RCT with 112 patients).

ASA or ASA plus DIP versus vitamin K antagonists (VKA), for venous bypass or prosthetic bypass
No differences were found among the groups for the following outcomes: primary graft patency at 3, 6, 12 or 24 months and limb amputation (evaluated in 2 RCTs with 2,781 patients); or cardiovascular events and mortality (evaluated in 1 RCT with 2,690 patients).

ASA versus prostaglandin (E1), for venous bypass
Early occlusion: no difference among the intervention groups (evaluated in 1 RCT with 100 patients).

ASA versus naſtidrofuryl
No differences were found among the groups for the following outcomes: primary graft patency, dizziness, gastric bleeding and mortality (evaluated in 1 RCT with 99 patients). No differences were found among the groups for the following outcomes: amputation for venous bypass and primary venous bypass patency at 24 months (evaluated in 1 RCT with 598 patients); mortality for venous or prosthetic bypasses (evaluated in 1 RCT with 851 patients); minor bleeding for prosthetic bypass, mild bleeding for prosthetic bypass and major bleeding for prosthetic bypass (evaluated in 1 RCT with 253 patients); or major bleeding for venous bypass (evaluated in 1 RCT with 598 patients).

Conclusions from this review
Treatment with ASA or ASA plus DIP had a positive effect on the patency of patients undergoing venous and prosthetic bypass surgery, especially on prosthetic bypasses, with benefit for outcomes at 1, 3, 6, 9 and 12 months. For venous bypass, the benefit was observed only for the 12-month endpoint. However, this superiority needs further investigation.
A single study evaluated the effect of ticlopidine versus placebo in venous bypass surgery. The results showed improvements at 6, 12 and 24 months after treatment. Thus, the only antiplatelet agent with better effect on venous bypass patency was ticlopidine.
The association of clopidogrel plus ASA versus ASA alone, which was evaluated in a single study, did not show any favorable effects for treatment with ASA plus Clopidogrel for outcomes at 24 months in venous bypass surgery, but favorable results were reported for prosthetic bypass surgery.
The results from this review need to be carefully interpreted since many of the analyses were obtained from a single study and with different dosages.

ANTITHROMBOTIC AGENTS FOR PREVENTING THROMBO-SIS AFTER INFRAINGUINAL ARTERIAL BYPASS SURGERY
This review 9 aimed to determine the efficacy of antithrombotic agents in patients with peripherical arterial disease (intermittent claudication and critical ischemia) who underwent femoropopliteal or femorodistal bypass surgery. Fourteen RCTs (total of 4,970 participants) were included. The main findings were as follows. Limb loss in venous bypass surgery: benefit through treatment with VKA for outcomes at 5 years (OR: 0.29; 95% CI: 0.14 to 0.6; 2 RCTs; 179 patients); no differences among the intervention groups for outcomes at 3, 6, 12 or 24 months.   RCT; 300 patients).

LMWH versus placebo
Occlusions in venous or prosthetic bypass surgery: no differences among the intervention groups in outcomes at 1, 3 or 12 months (evaluated in 1 RCT with 207 patients).

UFH versus antithrombin
Occlusions in venous or prosthetic bypass surgery: lower number of occlusions (benefit) through UFH in the intraoperative period (OR: 55.0; 95% CI: 1.86 to 1622.6; 1 RCT; 13 patients); no differences among the intervention groups at one month.

LMWH versus dextran plus heparin
Early graft occlusion: no differences among the intervention groups (evaluated in 1 RCT with 277 patients) for venous or prosthetic bypasses.

Conclusion from this review
This review suggested that the effectiveness of antithrombotic medications depends on the type of graft used (venous or prosthetic).
Patients undergoing venous bypass surgery are likely to benefit more from VKA than from platelet inhibitors, while patients undergoing prosthetic bypass surgery are likely to benefit more from antiplatelets (ASA) than from VKA. To support this information and rule out divergences, studies that are more homogenous, with larger samples and detailed descriptions of the participants' characteristics, are needed.
Regarding treatment with LMWH versus UFH, the results were marginal and further studies with more patients are needed, for better comparisons.
For LMWH versus ASA plus DIP, better patency was found in the LMWH group, but this advantage was observed for patients with critical ischemia and not for patients with claudication. Further studies are needed in order to confirm these results.

ANTIPLATELET AND ANTICOAGULANT DRUGS FOR PRE-VENTION OF RESTENOSIS/REOCCLUSION AFTER PERIPH-ERAL ENDOVASCULAR TREATMENT
This review 10 evaluated whether any antithrombotic drug was more effective in preventing restenosis or reocclusion after endovascular treatment, in comparison with any antithrombotic drug, no treatment, placebo or vasoactive drugs. Twenty-two RCTs (3,529 participants in total) were included. The main findings were as follows. No differences were found among the groups for the following outcomes: amputations, mortality and bleeding at a puncture site at 1 month (evaluated in 1 RCT with 223 patients).

Clopidogrel plus ASA versus LMWH followed by warfarin
Occlusions/restenosis: no differences among the intervention groups at 24 hours or 1, 6, 12 or 18 months (evaluated in 1 RCT with 103 patients).

Ticlopidine versus VKA
Occlusions/restenosis: no differences among the intervention groups at 12 months (evaluated in 1 RCT with 197 patients).
Side effects: higher number of side effects through use of ticlopidine

Conclusions from this review
The authors concluded that there was limited evidence to suggest that occlusions/restenosis after six months of endovascular treatment were reduced through use of antiplatelet drugs, in comparison with placebo/controls.
The best evidence points to use of dipyridamole plus ASA, cilostazol plus ASA and low molecular weight heparin (with or without ASA).
Cilostazol plus ASA was superior to ticlopidine, and LMWH plus ASA was superior to ASA, in assessing restenosis/occlusion at 12 months after the start of intervention.
The clinical trials included in this systematic review were small, and the side effects could not be consistently evaluated.
A single study 11 evaluated the use of ASA plus clopidogrel compared with LMWH plus warfarin, but without evidence of superiority for either of the treatments. Use of ASA plus clopidogrel seems to have off-label indications mainly in the United States. RCTs evaluating use of ASA plus clopidogrel for acute coronary syndrome showed significant reductions in cardiovascular deaths (AMI or stroke), but they showed that this treatment led to increased risk of major bleeding.

DISCUSSION
This study included three Cochrane systematic reviews. [8][9][10] No evaluation using the GRADE (Grading of Recommendations, Assessment, Development and Evaluation) criteria was done in any of the reviews included in this study. This evaluation is now mandatory for publication of systematic reviews in the Cochrane Library.
The results in Table 2 can be used for guidance for healthcare professionals and managers, but it needs to be borne in mind that future studies may substantially change the results that have been published so far.
For patients revascularized with prostheses, there was a benefit from use of ASA plus clopidogrel, and especially from use of ASA alone or ASA plus DIP. This benefit was not observed in relation to venous grafts. 8 For venous bypass surgery, the best results were observed through use of ticlopidine. For both venous and prosthetic bypass surgery, the results need to be carefully interpreted because they were based on few studies. 8 Patients undergoing venous bypass surgery are likely to benefit more from VKA than from platelet inhibitors. Comparing use of LMWH versus use of ASA plus DIP, better results were found in the LMWH group, but these results were observed in patients with critical ischemia and not in patients with claudication. Further studies are needed to confirm these results.
Patients undergoing prosthetic bypass surgery are likely to benefit more from antiplatelet (ASA) than from VKA. 9 For patients undergoing endovascular procedures, the best results came from use of ASA plus DIP, from cilostazol plus ASA and from LMWH. Cilostazol plus ASA was superior to ticlopidine; and LMWH plus ASA was superior to ASA.

CONCLUSION
This study included three Cochrane systematic reviews that provided evidence regarding use of antiplatelets and anticoagulants in lower-limb revascularization in patients with lower-limb PAD. Patients who underwent venous or prosthetic bypass surgery were included, as well as patients who received endovascular treatments.
We noticed that patients undergoing prosthetic graft revascularization presented better outcomes through use of antiplatelet agents. Patients undergoing venous graft revascularization showed better outcomes through use of anticoagulants. In patients undergoing endovascular treatment, use of both antiplatelet and anticoagulant medication proved to be beneficial.
However, the lack of use of the GRADE approach in the reviews included compromised the certainty of our evaluation of the evidence. Updates to these reviews are needed in order to assess the implications of these treatments for clinical practice, among patients with PAD.