See related article, p 3543

In 2012, investigators from the Tufts Friedman School of Nutrition estimated that 45.4% of deaths from cardiometabolic disease (ie, heart disease, stroke, and type 2 diabetes) and 44.5% of deaths from ischemic stroke in the United States were associated with suboptimal intake of 10 dietary factors included too little intake of fruit, vegetables, nuts and seeds, whole grains and seafood and too much intake of red meat, processed meat, sugar-sweetened beverages, and sodium.¹ These estimates are approximated by other evidence^2,3 which, taken together, suggest that any campaign to lower the burden of stroke and heart disease will need to include a major push to improve diet quality. The idea that “food is medicine,”⁴ has been proposed to encourage the health care community to see diet support as part of mainstream clinical therapeutics and public policy. It is reasonable to ask, however, if the evidence for diet therapy is developed well enough to give it an equal role next to conventional drug and device therapies for primary and secondary stroke prevention.

Keys⁵ first brought attention to the effect of diet pattern on risk for vascular disease in the Seven Countries Study. Started in 1958, the study showed that southern Europeans had a lower incidence of coronary heart disease than residents of the United States and Northern Europe and that this difference could not be fully explained by serum cholesterol, blood pressure, smoking, age, or physical activity. The Seven Countries Study pointed to the Cretan Mediterranean dietary pattern as possibly important. That diet was high in vegetables, fruit, and pasta and low in saturated fat.⁵

Subsequent research confirmed findings from the Seven Countries Study and discovered underpinnings of the Mediterranean-type diet.^6–12 A higher ratio of monounsaturated/saturated fat, and higher consumption of fruits and vegetables,¹³ legumes, nuts,^14,15 fiber from cereal,¹⁶ olive oil, and fish¹⁷ all contribute to lower risk for cardiovascular disease.¹⁸ So does moderate consumption of alcohol^19,20 and lower intake of dairy and meat products. Since the work of Keys,⁵ the beneficial effects of the Mediterranean-type have been examined in several European and US cohorts.^11,12,21 In the HALE project (Healthy Ageing: A Longitudinal Study in Europe Population) of European populations, adherence to a Mediterranean-type diet was associated with a specific reduction in cardiovascular mortality (hazard ratio, 0.71 [95% CI, 0.58–0.88]).⁸ Among US nurses, those in the top quintile for the Alternate Mediterranean Diet Score, compared with those in the bottom quintile, were at lower risk for coronary heart disease (relative risk, 0.71 [95% CI, 0.62–0.82]), sudden cardiac death (adjusted relative risk, 0.60 [95% CI, 0.43–0.84]), and stroke (relative risk, 0.87 [95% CI, 0.73–1.02]).^11,22

Evidence on alcohol and dairy illustrate the complexity and controversy in the field of nutrition science and vascular disease. Contrary to earlier research on alcohol that suggested a benefit for moderate use, a recent multicenter case-control study demonstrated that alcohol was inversely associated with coronary heart disease risk, but positively associated with risk for stroke. The safety of alcohol for stroke remains uncertain.²³ Dairy products are a good source of protein, vitamins, and minerals, especially calcium and magnesium, and for this reason are recognized as a nutrient-rich food that should be part of a healthy diet.²⁴ Cheese, whole milk, and certain other dairy products, however, are also high in saturated fat. The United States Department of Agriculture and other sources of nutrition advice seek to balance the potential harms of saturated fat with the benefits of other dairy product components by recommending consumers substitute no-fat and low-fat dairy products for full fat products.²⁵

Evidence on diet and risk for stroke is consistent with the evidence for diet and risk for other vascular disease events. Reduced risk for stroke, particularly ischemic stroke, has been reported with greater adherence to a high-quality diet,^26–34 a vegetarian diet,³⁵ and greater consumption of specific food, such as fruits and vegetables,^36,37 fish,^38–40 fiber,^41,42 and potassium.^43–45 Higher consumption of salt⁴⁶ and diets emphasizing fried food, concentrated sweets, eggs, and processed meats⁴⁷ or red meat⁴⁸ have been associated with increased risk for stroke. Estimates of how much stroke risk can be reduced with a healthy diet, such as a Mediterranean style diet, range from 18% to 42% in US cohorts.^28,29

Very few observational studies have examined the influence of diet among people with established vascular disease. However, in survivors of myocardial infarction in the Nurses’ Health Study and the Health Professionals Follow-up Study, the adjusted hazard ratio was 0.76 (95% CI, −0.60 to 0.96) for all-cause mortality and 0.73 (95% CI, 0.51–1.04) for cardiovascular mortality, comparing extreme quintiles of the Alternative Healthy Eating Index.⁴⁹ Similar findings were reported in this cohort for adherence to a Mediterranean style dietary pattern as measured by the Alternative Mediterranean Diet score.⁵⁰

In this issue of Stroke, Ikehara et al⁵¹ report one of the largest studies on the association between a specific food and risk for stroke. Peanuts are a legume, not to be confused with a tree nuts like almonds or walnuts, and are rich in calories, protein, vitamins, and fiber. A 28-g (1 ounce) serving of about 30 to 60 shelled nut pieces delivers 160 kcal of energy, 7 g protein, 2 grams of fiber, and over 5% of the United States Department of Agriculture daily value for several vitamins and minerals, including thiamin, niacin, folate, magnesium, and zinc. They are also a good source of healthy fat: a 28-g serving of peanuts will contain 14 total grams of fat but only 2 g are saturated and the rest are monounsaturated or polyunsaturated.⁵²

The investigators analyzed data from participants in the JPHC (Japan Public Health Center–Based Prospective Study on Cancer and Cardiovascular Diseases). The JPHC assembled 2 cohorts from 1989 to 1991 (cohort I) and 1993 to 1994 (cohort II). Participants were all residents age 40 to 59 years (cohort 1) and 40 to 69 years (cohort 2) from several public health regions. The present study includes a subset of the JPHC participants who completed a food frequency questionnaire in 1995 (cohort I) or 1998 to 1999 (cohort II) and who were free of cancer, stroke, or myocardial infarction. Follow-up began on the date of the food questionnaire. After excluding some outliers and participants with missing nutritional data, the final cohort was nearly 75 000 people. Outcome were ascertained by monitoring hospital admissions.

The average participant consumed about 1.5 peanuts a day (1.5 g); the range was none in quartile 1 to 4.3 in quartile 4. Men and women with higher peanut consumption tended to be younger, healthier, and more physically active. They also ate more fish, fruit, and vegetables. During a median follow-up of 15 years, the investigators observed 3599 strokes and 849 ischemic heart disease events (myocardial infarction or sudden death). Higher peanut consumption was associated with reduced risk for ischemic stroke in quartile 2 (hazard ratio, 0.82 [95% CI, 0.72–0.93]), quartile 3 (hazard ratio, 0.86 [95% CI, 0.77–0.96]), and quartile 4 (hazard ratio, 0.80 [95% CI, 0.71–0.90]) compared with quartile 1 after adjustment for many covariates including consumption of vegetables, fruit, and fish. No significant association was observed between peanut consumption and risk for hemorrhagic stroke or ischemic heart disease.

The finding of a beneficial effect of peanuts is almost too good to believe, especially when you consider that the dose of peanuts was only about 0.7 to 4.3 peanuts daily in those who demonstrated benefit and the effect did not increase incrementally across quartiles. Before becoming too doubtful, however, there is other research to support the findings. Ikehara et al⁵¹ remind us of a combined analysis of the US Nurses’ Health Study and the Health Professions Follow-up study; participants who consumed peanuts ≥2× a week had a 10% reduction in risk for stroke.⁵³ At least one other large cohort study also suggests a specific benefit of peanuts on stroke risk.⁵⁴ In contrast, results from the Nurses’ Health Study and the Health Professionals Follow-Up Study do not show an effect of peanuts on stroke mortality,¹⁵ and research on all types of nuts combined is mixed.^55–57

It is possible that any beneficial effect of a small number of peanuts is amplified by isocaloric replacements.⁵⁸ That is to say that people who eat peanuts may eat them as part of an overall healthy diet pattern in which peanuts replace less healthy foods. Replacement of red meat with other dietary sources of protein has been associated with reduced risk for stroke.⁴⁸

The findings of observational research, such as the peanut study here, do not support class 1 recommendations in professional guidelines for prevention of stroke; class 1 requires phase III randomized clinical trials with cardiovascular outcomes. The 2021 American Heart Association guideline for prevention of stroke in patients with stroke and transient ischemic attack gives a class 2a recommendation for a Mediterranean diet as “reasonable”⁵⁹; it gives stronger (class 1) recommendations for statin therapy, antiplatelet therapy, certain diabetes drugs, carotid surgery, and exercise. The discrepancy reflects the lack of high-quality randomized clinical trials evidence for the effectiveness of a healthy diet. Since the Seven Countries Study, only 2 adequately powered studies of diet for vascular prevention have been reported: the older Lyon Diet-Heart Study in patients with coronary heart disease⁶⁰ and the more recent PREDIMED Study (Prevención con Dieta Mediterránea) in patients at high risk.⁶¹ Both studies showed that diet prevented major cardiovascular events. In PREDIMED, a Mediterranean-type diet, supplemented with mixed nuts or olive oil, reduced risk for stroke, a secondary end point, by 39% (hazard ratio, 0.61 [95% CI, 0.44–0.86]).⁶¹

Neither the National Institute of Neurological Disorders and Stroke nor the National Heart Lung and Blood Institute is currently funding phase III diet trials with cardiovascular end points.⁶² One or more trials in patients after stroke seems like a compelling proposition: these patients are at high risk for vascular disease, dietary therapy is inexpensive, perfectly safe, has almost no contraindications, and is likely to be appealing to patients. The hazard ratios observed in the study by Ikehara et al⁵¹ are similar to those in trials of some proven therapies for secondary prevention of stroke and large enough to be tested in an randomized clinical trial of reasonable size. However, it probably makes more sense for funding agencies to prioritize trials using a Mediterranean-type diet because patients with stroke eat foods in combination, and the treatment effect for a Mediterranean-type diet in observational work and in the PREDIMED trial is large and would require a smaller sample size.

To conduct a definitive phase III diet trial in individual patients with stroke, the National Institutes of Health and the research community would need to overcome methodologic challenges, particularly with participant selection, selection of the experimental and comparison maneuvers, site preparation, implementation of the maneuver, and accounting for treatment crossover.⁶³ This would not be easy. If trials confirm a treatment effect on hard end points, however, lifestyle treatment facilities might finally enter the fabric of health care, funded as fully as surgical suites, pharmacies, and imaging centers. I predict that many would serve peanuts at check-in.

Disclosures None.

The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.

For Disclosures, see page 3553.

参见相关文章，第 3543 页

2012 年，塔夫茨弗里德曼营养学院的研究人员估计，美国 45.4% 的心脏代谢疾病（即心脏病、中风和 2 型糖尿病）死亡和 44.5% 的缺血性中风死亡与次优摄入有关在 10 个饮食因素中，水果、蔬菜、坚果和种子、全谷物和海鲜摄入量过少，以及红肉、加工肉类、含糖饮料和钠摄入量过多。¹这些估计值与其他证据^{2,3 相近}，综合起来表明，任何降低中风和心脏病负担的运动都需要大力推动改善饮食质量。“食物就是药物”的理念，⁴已提议鼓励医疗保健社区将饮食支持视为主流临床治疗和公共政策的一部分。然而，我们有理由质疑饮食疗法的证据是否足够完善，使其在卒中一级和二级预防中与常规药物和器械疗法相比具有同等作用。

Keys ⁵首先在七国研究中关注饮食模式对血管疾病风险的影响。始于 1958 年的研究表明，南欧人的冠心病发病率低于美国和北欧的居民，而且这种差异不能完全用血清胆固醇、血压、吸烟、年龄或体力活动来解释. 七国研究指出克里特地中海饮食模式可能很重要。这种饮食富含蔬菜、水果和意大利面，饱和脂肪含量低。⁵

随后的研究证实了七国研究的结果，并发现了地中海式饮食的基础。^6-12较高的单不饱和脂肪/饱和脂肪比例，以及较多食用水果和蔬菜、^{13 种}豆类、坚果、^14,15来自谷物的纤维、¹⁶橄榄油和鱼¹⁷都有助于降低患心血管疾病的风险。¹⁸适度饮酒^19,20和减少乳制品和肉制品的摄入量也是如此。自从 Keys 的工作⁵以来，已经在几个欧洲和美国的队列中研究了地中海型的有益影响。^11,12,21在欧洲人口的 HALE 项目（健康老龄化：欧洲人口纵向研究）中，坚持地中海式饮食与心血管死亡率的特定降低相关（风险比，0.71 [95% CI，0.58–0.88]） . ⁸在美国护士中，与最低五分之一相比，替代地中海饮食评分最高的五分之一的护士患冠心病的风险较低（相对风险，0.71 [95% CI，0.62–0.82]），突发心源性死亡（调整后的相对风险，0.60 [95% CI，0.43-0.84]）和中风（相对风险，0.87 [95% CI，0.73-1.02]）。^11,22

关于酒精和乳制品的证据说明了营养科学和血管疾病领域的复杂性和争议。与早期关于酒精的研究表明适度使用有益，相反，最近的一项多中心病例对照研究表明，酒精与冠心病风险呈负相关，但与中风风险呈正相关。酒精对中风的安全性仍然不确定。²³乳制品是蛋白质、维生素和矿物质（尤其是钙和镁）的良好来源，因此被认为是一种营养丰富的食物，应该成为健康饮食的一部分。²⁴然而，奶酪、全脂牛奶和某些其他乳制品的饱和脂肪含量也很高。美国农业部和其他营养来源的建议旨在通过建议消费者用无脂和低脂乳制品替代全脂产品来平衡饱和脂肪的潜在危害和其他乳制品成分的好处。²⁵

关于饮食和中风风险的证据与饮食和其他血管疾病事件风险的证据一致。据报道，更坚持高质量的饮食，^26-34素食，³⁵和更多地食用特定食物，如水果和蔬菜，^36,37鱼，^{38- 40}纤维、^41,42和钾。^43–45食盐摄入量较高⁴⁶以及强调油炸食品、浓缩甜食、鸡蛋和加工肉类⁴⁷或红肉的饮食⁴⁸与中风风险增加有关。在美国队列中，通过健康饮食（例如地中海式饮食）可以降低多少中风风险的估计范围为 18% 至 42%。^28,29

很少有观察性研究检查饮食对已确诊血管疾病患者的影响。然而，在护士健康研究和卫生专业人员随访研究中的心肌梗塞幸存者中，全因死亡率的调整风险比为 0.76（95% CI，-0.60 至 0.96）和 0.73（95% CI， 0.51–1.04）心血管死亡率，比较替代健康饮食指数的极端五分之一。⁴⁹在这个队列中报告了类似的发现，用于坚持地中海风格的饮食模式，这是通过替代地中海饮食评分衡量的。⁵⁰

在本期Stroke 中，Ikehara 等人⁵¹报告了一项关于特定食物与中风风险之间关联的最大研究之一。花生是一种豆类，不要与杏仁或核桃等树坚果混淆，并且富含卡路里、蛋白质、维生素和纤维。一份 28 克（1 盎司）的约 30 至 60 颗带壳坚果可提供 160 大卡能量、7 克蛋白质、2 克纤维，以及超过美国农业部每日价值的 5% 的多种维生素和矿物质，包括硫胺素、烟酸、叶酸、镁和锌。它们也是健康脂肪的良好来源：一份 28 克花生将含有 14 克总脂肪，但只有 2 克是饱和脂肪，其余为单不饱和或多不饱和脂肪。⁵²

研究人员分析了 JPHC（基于日本公共卫生中心的癌症和心血管疾病前瞻性研究）参与者的数据。JPHC 从 1989 年到 1991 年（队列 I）和 1993 年到 1994 年（队列 II）收集了 2 个队列。参与者是来自多个公共卫生地区的所有年龄在 40 至 59 岁（队列 1）和 40 至 69 岁（队列 2）的居民。本研究包括在 1995 年（队列 I）或 1998 年至 1999 年（队列 II）完成食物频率问卷并且没有癌症、中风或心肌梗塞的一部分 JPHC 参与者。随访从食物问卷日期开始。在排除了一些异常值和缺少营养数据的参与者后，最终的队列有近 75 000 人。通过监测住院情况确定结果。

参与者平均每天食用约 1.5 个花生（1.5 克）；该范围在四分位数 1 到 4.3 中没有。花生消费量较高的男性和女性往往更年轻、更健康、更活跃。他们还吃更多的鱼、水果和蔬菜。在中位随访 15 年期间，研究人员观察到 3599 次中风和 849 次缺血性心脏病事件（心肌梗塞或猝死）。在四分位数 2（风险比，0.82 [95% CI，0.72-0.93]）、四分位数（风险比，0.86 [95% CI，0.77-0.96]）和四分位数中，较高的花生摄入量与缺血性卒中风险降低相关4（风险比，0.80 [95% CI，0.71–0.90]）与四分位数 1 相比，在调整了许多协变量（包括蔬菜、水果和鱼类的摄入量）后。

花生有益作用的发现几乎令人难以置信，特别是当您考虑到花生的剂量在那些表现出益处的人中每天只有 0.7 到 4.3 颗花生时，而且这种作用并没有在四分位数范围内逐渐增加。然而，在变得过于怀疑之前，还有其他研究支持这些发现。Ikehara 等人⁵¹提醒我们对美国护士健康研究和健康专业随访研究的综合分析；每周食用花生≥2 次的参与者中风风险降低 10%。⁵³至少另一项大型队列研究也表明花生对中风风险有特定益处。⁵⁴相比之下，护士健康研究和卫生专业人员后续研究的结果并未显示花生对中风死亡率的影响，¹⁵并且对所有类型坚果的研究结果喜忧参半。^55–57

少量花生的任何有益效果可能会被等热量替代品放大。⁵⁸也就是说，吃花生的人可能会将花生作为整体健康饮食模式的一部分，在这种饮食模式中，花生替代了不太健康的食物。用其他膳食蛋白质来源替代红肉与降低中风风险有关。⁴⁸

观察性研究的结果，例如这里的花生研究，不支持中风预防专业指南中的 1 类建议；1 级需要具有心血管结局的 III 期随机临床试验。2021 年美国心脏协会预防卒中和短暂性脑缺血发作患者的卒中指南将地中海饮食列为“合理”的 2a 级建议⁵⁹; 它为他汀类药物治疗、抗血小板治疗、某些糖尿病药物、颈动脉手术和运动提供了更强的（1 类）建议。这种差异反映了缺乏关于健康饮食有效性的高质量随机临床试验证据。自七国研究以来，仅报告了 2 项关于饮食预防血管的充分研究：较早的里昂饮食心脏研究对冠心病⁶⁰和最近的 PREDIMED 研究 (Prevención con Dieta Mediterránea) 对高血压患者进行了研究。风险。⁶¹两项研究都表明饮食可以预防主要的心血管事件。在 PREDIMED 中，地中海式饮食，辅以混合坚果或橄榄油，将次要终点中风的风险降低了 39%（风险比，0.61 [95% CI，0.44–0.86]）。⁶¹

国家神经疾病和中风研究所和国家心肺和血液研究所目前都没有资助具有心血管终点的 III 期饮食试验。⁶²对中风后患者进行一项或多项试验似乎是一个令人信服的提议：这些患者患血管疾病的风险很高，饮食疗法价格低廉、完全安全、几乎没有禁忌症，并且可能对患者有吸引力。Ikehara 等人在研究中观察到的风险比⁵¹与一些已证实的中风二级预防疗法试验中的试验相似，并且规模大到足以在合理规模的随机临床试验中进行测试。然而，资助机构优先考虑使用地中海式饮食的试验可能更有意义，因为中风患者同时食用多种食物，并且在观察性工作和 PREDIMED 试验中地中海式饮食的治疗效果很大，并且需要较小的样本量。

为了对个别中风患者进行明确的 III 期饮食试验，美国国立卫生研究院和研究界需要克服方法学上的挑战，尤其是参与者的选择、实验和比较操作的选择、场地准备、操作的实施，并考虑处理交叉。⁶³这并不容易。然而，如果试验证实了对硬终点的治疗效果，生活方式治疗设施可能最终进入医疗保健领域，资金充足，如手术室、药房和影像中心。我预测很多人会在办理登机手续时提供花生。

披露无。

本文中表达的观点不一定是编辑或美国心脏协会的观点。

有关披露，请参阅第 3553 页。