Peripheral artery disease (PAD) remains a major cause of morbidity and future cardiovascular events despite advancement in the surgical interventions and optimal medical therapy. The aim of our study is to evaluate the efficacy and safety of anticoagulation (AC) therapy for reducing cardiovascular and limb events in patients with PAD. PUBMED, Medline, and Cochrane Library were searched through 2020 for randomized clinical trials comparing major adverse cardiovascular events (MACE) and risk of major bleeding (MB), between AC and standard of care (SOC) therapy, among patients with PAD. Meta-analysis was performed using weighted pooled absolute risk difference (RD) with 95% confidence interval (CI) and fixed effects model for overall and sub-groups of full dose (FD) and low dose (LD) AC therapies. Amongst 17,684 patients from 7 different studies, the addition of AC to SOC therapy was associated with MACE reduction (RD -0.022, 95% CI -0.033 to -0.012, p <0.001) and increased MB (RD 0.02, 95% CI 0.014 to 0.025, p <0.001). For FD, MACE reduction was (RD -0.021, 95% CI -0.042 to 0.001, p = 0.061) and MB (RD 0.036, 95% CI 0.025 to 0.047, p <0.001). For LD, MACE reduction was (RD -0.023, 95% CI -0.035 to -0.011, p <0.001) and MB (RD 0.011, 95% CI 0.005 to 0.017, p <0.001). In conclusion, addition of AC to the current SOC therapy can mitigate future MACE events in patients with PAD albeit at risk of increased bleeding. LD AC is associated with an efficacy/safety net benefit compared to FD AC therapy.
Peripheral arterial disease (PAD) remains a major cause of morbidity, mortality, and disability across the world despite the recent advancement in medical, endovascular, and surgical therapies. PAD patients are at a higher risk of cardiovascular and limb events despite optimal medical therapy. Although current medical and surgical societal guidelines have advocated single or dual antiplatelet therapy (SAPT and DAPT) for these patients. , Recent randomized controlled trials (RCTs) have found superior outcomes with the addition of oral anticoagulation (AC) to antiplatelet. , Addition of AC to anti-platelet therapy in patients with PAD has been shown to provide an additional anti-ischemic benefit albeit at the cost of higher bleeding risk. More recent randomized clinical trials have shown a similar reduction of ischemic events with the addition of low-dose rivaroxaban to antiplatelet therapy for patients with PAD. , Therefore, we constructed a study level meta-analysis, comparing AC combination therapy to the current standard of care (SOC) antiplatelet therapy. In addition, we aimed to evaluate full-dose (FD) versus low-dose (LD) , , AC on safety and efficacy outcomes.
Methods
A search was conducted in the MEDLINE, Cochrane Library, and the Embase database through 2020 and clinical trials were identified by searching the keywords, “peripheral artery disease,” “peripheral artery intervention,” and Boolean operator and “anticoagulation” by 2 independent authors (HK, RM). Studies were included if they were RCT by design, compared AC therapy to antiplatelet therapy, and reported individual cardiovascular and limb events. All studies in the SOC arm used either single antiplatelet or dual antiplatelet (ePAD only), whereas the AC arm included AC with antiplatelet or without antiplatelet (Dutch BOA trial only). Studies were excluded if they were observational, nonrandomized, retrospective, subgroup of the original study or abstract. Figure 1 depicts a summary of our systematic review for inclusion, exclusion, and final selection of studies for the meta-analysis whereas Table 1 summarizes inclusion and exclusion criteria for each individual study. The study did not require institutional review board approval being a systemic review of the previously published RCT. The study is registered with international, prospective register for systematic review protocols (Prospero; CRD42020190815).
| VOYAGER PAD | COMPASS | ePAD | Wave | Jivegard | Johnson | Dutch BOA | |
|---|---|---|---|---|---|---|---|
| Year | 2020 | 2018 | 2018 | 2007 | 2005 | 2002 | 2000 |
| Sample size | 6,564 | 4,996 | 203 | 2,161 | 284 | 831 | 2,690 |
| Median follow-up, (years) | 2.3 | 1.7 | 0.5 | 2.9 | 1 | 3 | 1.8 |
| Age, (years) | 67 ± 6 | 67.6 ± 8.5 | 67.3 ± 9.5 | 64 | 73.5 ± 9 | 64 ± 8 | 69 ± 10 |
| Male | 74% | 71% | 71% | 73.6% | 55.2% | 99% | 63% |
| Aspirin dose | 100 mg | 100 mg | 100 mg | 81-325 mg | 75 mg | 325 mg | 80 mg |
| AC type and dose | Rivaroxaban (2.5mg) | Rivaroxaban (2.5mg) | Edoxaban (60mg) | VKA (INR 2-3) | Dalteparin (5000 IU) | VKA (INR 1.4-2.8) | VKA (INR: 3-4.5) |
| Composite primary efficacy outcome (MACE) | CVD, CVA, MI, ALI, AMP | CVD, CVA, MI, ALI | CVD, CVA, MI, TLR, AMP | CVD, CVA, MI, ALI, AMP | (-) | (-) | CVD, CVA, MI, AMP |
| Primary safety outcome (major bleeding) | + | + | + | + | + | + | + |
Qualified studies were reviewed by 2 independent authors (HK, RM) to assess suitability. A total of 7 studies were selected after discrepancies were resolved and mutual consensus among all the co-authors for final inclusion. Study characteristics and results were extracted by the authors and incorporated in a standardized form. Data extracted from the studies included the first author’s name, publication year, country of origin, sample size, gender, and age of study participants. Also, years of follow-up, method to assess events, the number of cases, and participants in each treatment subgroup were recorded and summarized in Table 1 . The data underlying in this article will be shared on reasonable request to the corresponding author.
Primary efficacy endpoint was defined as major adverse cardiovascular events (MACE) as a composite of cardiovascular death (CVD), myocardial infarction (MI), stroke (or cerebrovascular accident [CVA]), limb events (revascularization, amputation) and net benefit. Primary safety outcome was defined as major bleeding (MB) according to trial definition and summarized in supplementary data Table S1 whereas Supplementary Table S2 and S3 summarize individual outcomes of interest reported by each trial. Secondary endpoints include individual rates of all-cause death (ACD), cardiovascular death, CVA, target vessel/limb revascularization (TRL/TVR), and major amputation (AMP). The net clinical benefit (NCB) of AC compared with SOC was calculated using the following formula: NCB = (MACE rate on AC – MACE rate on SOC ) – (major bleeding rate on AC – major bleeding rate on SOC).
The data were analyzed using the Comprehensive Meta-Analysis package V3 (Biostat, Englewood, NJ, USA). The Mantel-Haenszel method was used for calculating the weighted pooled absolute risk difference under the fixed effects model. The data are reported as RD with 95% confidence interval (CI). To access the heterogeneity of the treatment effect among the different RCT included, Cochran’s Q and I 2 statistic were calculated under the random effects model and reported for each individual outcome in the results section respectively. Publication bias was assessed using the Egger’s test and a p value < 0.05 considered significant for potential bias (Supplementary Table S4). Sensitivity analysis was performed using the one study removed technique to explore the effect of individual study on overall outcomes (Supplementary Table S5 and Supplementary Figure 4 to 11). To further evaluate the effect of dose of AC used, we also conducted a sub-group analysis for FD and low dose LD AC therapy. Forest plots were used to represent within group, and overall effect size along with their respective 95% CI. A p value <0.05 was considered significant for our analysis.
Results
A total of 7 studies were included in the final analysis. Figure 1 summarizes the selection criteria for the final studies included in the meta-analysis while Figure 2 represents the summary of primary efficacy (MACE) and safety (MB) outcomes with their respective forest plots. Figure 3 represents the NCB or harm associated with AC as compared to SOC therapy. The basic demographic parameters of the AC and SOC groups and a summary of the trials included in the study is summarized in Table 1 . The individual mortality, ischemic, and limb events are presented in the supplementary material. The study included a total of 17, 684 subjects, of whom 8,843 were assigned to AC group and 8,841 to SOC group, mean age 68 +/- 3 years. There are 4 studies in the FD and 3 studies in the LD AC group.
