Peripheral artery disease (PAD) is rising in worldwide along with population aging. Patients with PAD experience chronic leg ischemia that impairs walking ability. Obstruction to distal blood flow to leg muscle due to atherosclerotic lesions characterizes peripheral artery disease. However, a simple model of reduced blood supply due to arterial stenosis does not fully explain functional limitations in patients with PAD. Impaired endothelial function, reduced microcirculatory flow, and altered skeletal muscle structure and function all have been associated with reduced walking ability in patients with PAD.¹ Heightened oxidative stress leading to impaired mitochondrial function has been demonstrated in muscle biopsies in PAD.² Exercise interventions improve a broad array of pathogenic aspects in PAD resulting in marked improvement in clinical status; thus, structured exercise therapy is the first-line treatment for patients with symptomatic PAD.³ However, patients with PAD frequently suffer from residual functional impairment and decline over time.

Article, see p 589

In this issue of Circulation Research, McDermott et al⁴ report the results of a pilot study of a cocoa beverage intervention in patients with PAD to gain evidence regarding functional benefits and potential mechanisms. Cocoa is a nutrient-rich food that contains bioactive flavonoids including epicatechin that augments vasodilation and reduces oxidative stress.⁵ Prior clinical evidence supported a small acute benefit of dark chocolate on treadmill walking in patients with PAD along with changes in systemic oxidative stress.⁶ The current study was a randomized pilot study designed 6-month comparison of an epicatechin-rich cocoa beverage compared with placebo beverage on 6-minute walk distance in patients with PAD. As anticipated, cocoa beverage increased epicatechin and theobromine levels. Overall, the patients in the cocoa group as compared with the placebo group had an improvement in walk distance. Surprisingly, there was considerable variability in the placebo group walk distance between the 2 follow-up measurements with a greater decline at 2.5 hours after beverage consumption (24 meters) than at 24 hours (3 meters). The difference between the 2 time points does not seems to be attributable to the frequently observed phenomenon of greater effect of polyphenol-rich foods at the timepoint of maximal circulating levels as the cocoa group had a similar improvement at 2.5 hours (18 meters) and 24 hours (15 meters) after beverage consumption. Assuming that the true decline in the placebo group was the average of the 2 measurements (13 meters), there would still be a difference in the cocoa group of >30 meters, a distance generally recognized as clinically important.⁷ There was no change in the secondary outcome of maximal treadmill walking time. Though there was an absolute difference in the accelerometer measured physical activity in daily life, this difference did not achieve statistical significance due to considerable variability in responses.

In addition to the functional assessments, the study reports multiple measurements both systemic and at the level of the calf muscle to address potential mechanisms that may drive the impact of cocoa in patients with PAD. In contrast with prior studies of cocoa in other patient groups, there was no change in endothelial function measured by flow-mediated dilation of the brachial artery that may reflect differences in the study beverage or the advanced vascular disease in patients with PAD.⁸ There was a tendency for an improvement in calf perfusion as measured by magnetic resonance imaging and a corresponding increase in skeletal muscle capillary density indicating that there may be an effect of cocoa on microvascular parameters. Prior animal model studies have indicated that epicatechin enhances angiogenesis along with higher VEGF-A (vascular endothelial growth factor-A) levels.

In the calf muscle biopsies of the patients with PAD, there was a relative increase of mitochondrial cytochrome c activity in the cocoa group that was largely produced by a decline in the placebo group. This findings suggests that cocoa may protect the muscle from deterioration in mitochondrial respiratory capacity. There was no evidence of differences in either PGC1α or citrate synthase activity suggesting that the effect on mitochondrial function may not have been mediated by a change in mitochondrial biogenesis. The number of central nuclei, an indicator of muscle injury, also had a lower degree of increase in the cocoa treated group without any indication of augmented muscle regeneration measured by embryonic+ fibers or Pax7+ fibers. In contrast to a study of epicatechin in a murine model, there were no changes in the pro- and anti-muscle growth factors follistatin and myostatin in the calf muscle biopsies further indicating that muscle growth did not produce the changes in walk distance with the cocoa intervention.⁹ Overall, the muscle biopsy findings bolster the concept that cocoa may have persistent protective effects in patients with PAD.

Oxidative stress induced by repeated episodes of ischemia and reperfusion is a potential modulator of muscle dysfunction in PAD. In both animal models and human studies of PAD, oxidative stress markers are higher in skeletal muscle and predict clinical events.¹⁰ Interestingly, the markers of oxidative stress such as nitrotyrosine and 4-HNE (hydroxynonenal) levels were not altered in the muscle after 6 months cocoa consumption. Antioxidant functions of flavonoids and polyphenols have been postulated as mediating the potential vascular protective effects. However, it should be noted that a certain level of oxidants is essential to promote angiogenesis by redox-induced signaling. Mouse studies show that antioxidants may inhibit ischemic limb vascularization. Also, the effects of oxidants or antioxidants in the ischemia are not necessarily same in the myocardium and skeletal muscles.¹¹ It seems likely that the protective effects of cocoa will not be fully explained through antioxidant properties.

It is critical to emphasize that the present study reports that results of a pilot intervention with multiple secondary and exploratory end points that used a one-sided P value of 0.1 to evaluate potential effects. The sample size is also small. Therefore, all the findings are not conclusive evidence of the benefits of cocoa in patients with PAD but represent support for a definitive study. Even though the study was randomized, the small number of patients led to baseline differences in clinical characteristics including obesity, smoking, and race and residual confounding may have persisted. The numbers of patients with paired muscle biopsies was in a smaller, selected group generating a greater possibility that chance, confounding or bias may have influenced the results. The lack of change in treadmill walking time may reflect the timing of the testing after the completion of the study intervention. There was also an unexpectedly higher rate of cardiovascular events in the cocoa group that warrants careful monitoring in future studies in patients with PAD. The ongoing COSMOS (Cocoa Supplement and Multivitamin Outcome Study), randomized trial of cocoa extract pills in over 20 000 adults, will provide important evaluation of the impact on cardiovascular event.¹²

Novel treatments to offer our patients with PAD to help them walk farther are greatly needed. Overall, at the baseline visit, the patients with PAD groups had impaired walk distance at baseline (334 meters), which is below the tenth percentile of normal values for men and women aged 70 to 80 years, emphasizing the importance of developing clinical interventions in this patient group.¹³ Taken together, the pilot study by McDermott and colleagues provides compelling preliminary evidence to support a potential benefit of epicatechin-rich cocoa on walking ability along with protection from worsening of calf muscle perfusion, skeletal muscle injury, and mitochondrial dysfunction. Additional clinical studies are warranted to evaluate the clinical efficacy of cocoa and other polyphenol-based therapies in patients with PAD. Some day, we may be able to prescribe eating more chocolate to our patients with PAD.

Dr Hamburg is supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under Award Numbers 5P50HL120163 and U54HL120163 and R01HL115391 and an American Heart Association Strategically Focused Research Network in Cardiometabolic Disease 20SFRN35120118. Dr Matsui is supported by R01HL133013. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Dr Hamburg has consulting relations with Merck, Bayer, and Sanifit and has equity interest in Acceleron Pharma.

For Sources of Funding and Disclosures, see page 601.

随着人口老龄化，外周动脉疾病（PAD）在全球范围内呈上升趋势。患有 PAD 的患者会出现慢性腿部缺血，从而损害行走能力。由于动脉粥样硬化病变导致腿部肌肉远端血流受阻是外周动脉疾病的特征。然而，动脉狭窄导致血液供应减少的简单模型并不能完全解释 PAD 患者的功能限制。内皮功能受损、微循环流量减少以及骨骼肌结构和功能改变都与 PAD 患者的行走能力降低有关。¹在 PAD 的肌肉活检中证实了导致线粒体功能受损的氧化应激升高。²运动干预可改善 PAD 的一系列致病因素，从而显着改善临床状况；因此，结构化运动疗法是症状性 PAD 患者的一线治疗。³然而，患有 PAD 的患者经常遭受残留的功能障碍，并随着时间的推移而下降。

文章，见第 589 页

在本期《循环研究》中，McDermott 等人⁴报告了可可饮料干预 PAD 患者的初步研究结果，以获得有关功能益处和潜在机制的证据。可可是一种营养丰富的食物，含有生物活性黄酮类化合物，包括可增强血管舒张和减少氧化应激的表儿茶素。⁵先前的临床证据支持黑巧克力对 PAD 患者在跑步机上行走的小幅急性益处以及全身氧化应激的变化。⁶目前的研究是一项随机试验研究，旨在对 PAD 患者 6 分钟步行距离内富含表儿茶素的可可饮料与安慰剂饮料进行为期 6 个月的比较。正如预期的那样，可可饮料增加了表儿茶素和可可碱的含量。总体而言，与安慰剂组相比，可可组患者的步行距离有所改善。令人惊讶的是，安慰剂组的步行距离在 2 次随访测量之间存在相当大的差异，饮用饮料后 2.5 小时（24 米）比 24 小时（3 米）下降幅度更大。两个时间点之间的差异似乎不是由于经常观察到的现象，即在最大循环水平的时间点富含多酚的食物效果更大，因为可可组在 2.5 小时（18 米）和 24 小时有类似的改善。饮用饮料后数小时（15 米）。假设安慰剂组的真正下降是 2 次测量的平均值（13 米），可可组中仍然存在 > 30 米的差异，这一距离通常被认为具有临床意义。⁷最大跑步机步行时间的次要结果没有变化。尽管加速度计测量的日常生活中的身体活动存在绝对差异，但由于反应的相当大的可变性，这种差异没有达到统计学意义。

除了功能评估外，该研究还报告了系统性和小腿肌肉水平的多项测量，以解决可能推动可可对 PAD 患者影响的潜在机制。与之前对其他患者组的可可研究相比，通过血流介导的肱动脉扩张测量的内皮功能没有变化，这可能反映了研究饮料的差异或 PAD 患者的晚期血管疾病。⁸通过磁共振成像测量，小腿灌注有改善的趋势，骨骼肌毛细血管密度相应增加，表明可可对微血管参数可能有影响。先前的动物模型研究表明，表儿茶素可增强血管生成以及更高的 VEGF-A（血管内皮生长因子-A）水平。

在 PAD 患者的小腿肌肉活检中，可可组线粒体细胞色素 c 活性相对增加，这主要是由于安慰剂组的下降所致。这一发现表明，可可可以保护肌肉免受线粒体呼吸能力的恶化。没有证据表明 PGC1α 或柠檬酸合酶活性存在差异，这表明对线粒体功能的影响可能不是由线粒体生物发生的变化介导的。中央核的数量（肌肉损伤的指标）在可可处理组中也有较低程度的增加，而没有任何通过胚胎+纤维或 Pax7+ 纤维测量的肌肉再生增强的迹象。与小鼠模型中表儿茶素的研究相比，⁹总体而言，肌肉活检结果支持了可可对 PAD 患者可能具有持久保护作用的概念。

反复发作的缺血和再灌注引起的氧化应激是 PAD 肌肉功能障碍的潜在调节剂。在 PAD 的动物模型和人体研究中，骨骼肌中的氧化应激标志物较高，可预测临床事件。¹⁰有趣的是，食用可可 6 个月后，肌肉中的氧化应激标志物，如硝基酪氨酸和 4-HNE（羟基壬烯醛）水平没有改变。黄酮类化合物和多酚的抗氧化功能已被假定为介导潜在的血管保护作用。然而，应该注意的是，一定水平的氧化剂对于通过氧化还原诱导的信号传导促进血管生成是必不可少的。小鼠研究表明，抗氧化剂可以抑制缺血性肢体血管形成。此外，氧化剂或抗氧化剂在缺血中的作用在心肌和骨骼肌中不一定相同。¹¹可可的保护作用似乎无法通过抗氧化特性得到充分解释。

需要强调的是，本研究报告了使用单侧P值 0.1 以评估潜在影响。样本量也很小。因此，所有研究结果都不是可可对 PAD 患者有益的确凿证据，但代表了对确定性研究的支持。尽管该研究是随机的，但少数患者导致包括肥胖、吸烟和种族在内的临床特征存在基线差异，残留的混杂因素可能仍然存在。进行配对肌肉活检的患者人数较少，选择的组产生更大的可能性，即机会、混杂或偏见可能影响结果。跑步机步行时间没有变化可能反映了研究干预完成后测试的时间安排。可可组的心血管事件发生率也出乎意料地高，这需要在未来对 PAD 患者的研究中进行仔细监测。正在进行的 COSMOS（可可补充剂和多种维生素结果研究）是对 20 000 多名成年人进行可可提取物丸的随机试验，将对心血管事件的影响进行重要评估。¹²

非常需要为我们的 PAD 患者提供新的治疗方法，以帮助他们走得更远。总体而言，在基线访视时，PAD 组患者在基线时的步行距离受损（334 米），低于 70 至 80 岁男性和女性正常值的百分之十，强调了制定临床干预措施的重要性这个病人组。¹³总之，McDermott 及其同事的初步研究提供了令人信服的初步证据，支持富含表儿茶素的可可对步行能力的潜在益处，以及防止小腿肌肉灌注恶化、骨骼肌损伤和线粒体功能障碍。有必要进行额外的临床研究来评估可可和其他基于多酚的疗法对 PAD 患者的临床疗效。有一天，我们也许可以给患有 PAD 的患者开出多吃巧克力的处方。

Hamburg 博士得到美国国立卫生研究院国家心脏、肺和血液研究所的支持，奖励编号为 5P50HL120163 和 U54HL120163 和 R01HL115391，以及美国心脏协会心脏代谢疾病战略重点研究网络 20SFRN35120118。松井博士得到 R01HL133013 的支持。内容完全由作者负责，并不一定代表美国国立卫生研究院的官方观点。

Hamburg 博士与默克、拜耳和 Sanifit 有咨询关系，并拥有 Acceleron Pharma 的股权。

有关资金来源和披露信息，请参见第 601 页。