4月

2026

改訂日：2026年7月16日

動脈硬化のプラークを縮小させることが証明された唯一の食事法（地中海式ダイエットではない！）

この記事の使い方

医療上の免責事項：この記事は教育目的のものであり、医学的な助言ではありません。個別の指導については、必ずかかりつけの医師にご相談ください。.

読みやすい

本当に心の詰まりを取り除くことができるのか？心疾患を逆転させる意外な科学

長い間、ほとんどの医師や患者は、心臓病は一方通行の道だと信じていました。一度動脈がプラークで詰まり始めると、歯垢 , 、目標は単に衰えを遅らせることだけでした。それは古い車の錆の問題のように扱われていました。錆を完全に取り除くことはできず、上から塗装を重ねるか、広がるのを急ピッチで食い止めるしかありません。それはまるで止められない老化の勢いのように感じられました。しかし、もし心臓病が一生の宣告でなかったとしたらどうでしょうか？あなたの動脈が次のようなものであると想像してみてください。 庭用ホース . 毎日そのホースに油脂や汚れを流し込み続けていけば、やがて詰まってしまいます。ほとんどの治療法は、わずかに残った隙間に無理やり水を通そうとしたり、高価な道具を使って汚れに穴を開けたりすることに焦点を当てています。しかし、新しい科学によれば、汚れを一切流し込むのをやめれば、体にはホース自体をきれいにする驚くべき能力が備わっていることが分かっています。これはと呼ばれています “臨床的退行” または単に、, 反転. 最近の研究によって、私たちは「臨床の分かれ道」に到達したことが示されている。心疾患を予防したいだけなのか、あるいはすでに起きてしまったダメージを積極的に修復したいのかによって、口にするものの選択はまったく変わってくる。.

ポイント1：動脈にとっての「マジックナンバー」は10%です

心臓の健康の世界では、多くの人が「低脂肪」について語りますが、それが何を意味するのかを定義することはめったにありません。心臓病を真に逆転させるために、科学者たちは “量的脂肪パラダイム.” これは、食べる脂肪の種類と同じくらい、脂肪の総量も重要であるということを意味しています。これに関する最も有名な証拠は、 オーニッシュ・ライフスタイル心臓治験 . 本研究では、中等度から重度の心疾患患者を対象に、総脂肪摂取量を以下に抑えるプログラムが実施された 1日の摂取カロリーの10% . 彼らは、追加の油、ナッツ、アボカドを含まない、ホールフードプラントベース（全植物性食品）の食事を摂りました。.実践する科学その結果は画期的なものでした。「対照群」――標準的な医学的アドバイスに従った人々――では動脈の詰まりが進行したのに対し、実験群では実際にプラークが縮小しました。データによると、10%未満の脂肪を摂取したグループでは、動脈の詰まりが 40.7% から 37.3% へ 5年以上経過した。一方、「普通の」アドバイスに従っていた人々は、血管の詰まりが 41.3% ～ 51.9% . しかし、それは単なる画面上の数字ではなく、彼らがどう感じているかということでした。病気を逆転させたグループには、 72%の減少狭心症 （胸痛）であったのに対し、もう一方のグループでは36%の減少にとどまった。おそらく最も衝撃的だったのは「何もしないことによるリスク」であり、標準治療を受けた人々には 2.47のリスク比 主要な心臓発作. これは、彼らが約2.5倍の確率で心臓発作 10%高脂肪食を摂取したグループと比較して、手術を必要とするケースが少なかった。この研究で指摘されているように、この超低脂肪の閾値は、「それ以下のレベルになると、動脈組織が修復状態に入るという生物学的下限」を生み出すようであり、これは「より高い食事脂肪摂取量では達成できない状態」である。”それが重要な理由 つまり、10%レベルを下回って体内の「老廃物」が排出されれば、体は生存のための闘いをやめ、損傷の修復を始めます。それは、まるで体がようやく一息ついて、動脈に「清掃部隊」を送り込む余裕ができたようなものです。.

テイクアウト2：「健康的な」油でさえ、血管を一時的に麻痺させることがある

私たちはよく〜だと言われるオリーブオイル「心臓に良い」とされています。バターよりはマシかもしれませんが、科学的調査によると、血管には意外な影響があることが分かっています。油を使った食事をとってから約3〜4時間後、血管には一時的なある現象が起きます “麻痺”医師はこれを「」と呼ばれるもので測定します 血流依存性血管拡張反応 . これは基本的に、血流を通すために血管が傘のようにどれだけうまく広がるかを調べるテストです。ロバート・ヴォーゲル博士による有名な研究では、オリーブオイルを含む1回の食事が、この拡張を約 31% . 要するに、その油が血管を「硬く」し、血液を効率よく送り出す能力を低下させるのです。.実践する科学この影響は「悪玉」脂肪酸だけに限定されるわけではありません。精製されたオリーブオイルでさえ、加工され孤立した脂肪であるため、この機能低下を引き起こします。しかし、 “マトリックス” 食品において――つまり自然界における脂肪の包まれ方は――非常に重要です。その研究の参加者が食べたとき 殻付きくるみ , 血管は柔軟で健康なままでした。なぜでしょうか？くるみが次のような成分を含むホールフードだからです。 アルギニン . アルギニンは、体が〜を作り出すために使用する特別な構成要素です 一酸化窒素 , 血管にリラックスして広がるよう伝える「奇跡の分子」、精製油にはこの保護の「マトリックス」が欠けています。.それが重要な理由心臓病を改善しようとする場合、目標は血管を一日中100%の状態に保つことです。食事に精製油を加えることは、食事をするたびに血液循環の妨げとなるようなものです。単離された油ではなく、自然のままの食品を選ぶことで、「血管」の柔軟性と反応性を維持することができます。.

テイクアウト3：有名な「地中海式ダイエット」の研究には隠された欠陥があった

健康ニュースをご覧になっている方なら、おそらく地中海式ダイエット心臓の健康におけるゴールドスタンダードです。これは主に、次のような大規模な研究に基づいています。 PREDIMED . 地中海式食生活は確かに標準的な欧米食よりも優れていますが、PREDIMED研究には、ほとんどの人が知らない「舞台裏」の重大な問題がいくつかありました。.実践する科学まず、その研究は、実際には地中海式食生活と本当の意味での低脂肪食を比較したわけではありませんでした。その研究における「低脂肪」グループは、実際には 摂取カロリーの37%が脂肪由来である ！心臓病の回復法の世界では、37%は脂肪分が非常に多いとされています。したがって、この研究は地中海式食事法が高脂肪の西洋式食事法よりも優れていることを確かに証明しましたが、<10%の回復食よりも優れていることを証明したわけではありません。また、研究の実施方法には重大な公平性の問題もありました。地中海式食事群には 無料の食べ物 （オリーブオイルを週に1リットル飲むようなもの）と、集中的なパーソナルコーチングでした。「低脂肪」グループには基本的に、ただ パンフレット 最初の3年間は放置され、自分で何とかしなければならなかった。この「“不均衡な介入”その結果をもたらした唯一の理由が食事だと言い切ることを非常に困難にしている。さらに、約 参加者のうち21% 適切にグループに割り当てられておらず、試験が早期に中止されたため、結果が実際よりも過剰に印象的なものに見えることがよくあります。.それが重要な理由ここでの教訓は単純です。地中海式ダイエットは 防ぐこと 健康な人の問題。それはまるで優れた保険のようだ。しかし、すでに心臓病を患っており、反転それに対して、地中海式ダイエットは、超低脂肪のホールフードプラントベース（植物性食品中心）のアプローチほど、「血管の大掃除」のような効果を示していません。.

ポイント4：世界で最も長寿な人々の「85%」という炭水化物の秘密

私たちは「炭水化物恐怖症」の世界に生きているが、世界で最も健康な人々の歴史は異なる物語を語っている。例えば伝統的な 沖縄の人々 . 1970年以前、彼らの心臓病の発生率は地球上で最も低い部類に入り、約 8分の1の料金 アメリカ合衆国で。.実践する科学彼らの秘訣は何だったのか？彼らの食事は基本的に “高性能燃料” 彼らの食生活は 85% 炭水化物 , その大部分は栄養価の高い紫芋由来のものでした。彼らの脂肪摂取量は非常に少なく、わずか約 総カロリーのうち6% . 彼らは精製された砂糖や白パンを食べておらず、全粒のデンプン質の植物を食べていました。これは偶然ではありません。環境が変わったときに何が起きたのかを正確に見ることができます。1970年以降、沖縄の人々がより西洋型の食生活へと移行するにつれて、脂肪の摂取量は約 27% . ほぼ同時期に、彼らの心疾患の発症率が急上昇し、その名高い長寿も失われ始めた。.それが重要な理由この「自然実験」は、ほぼ完全なホールフードの植物性炭水化物からなる食生活でも、人体が健康を維持し、動脈を完全にきれいに保つことができることを示している。脂肪の摂取量を非常に低く抑えている限り、体は非常に健康な状態を保つ。ホールフードの植物由来の炭水化物は敵ではなく、多くの場合、問題を引き起こしているのは追加された脂肪である。.

テイクアウト5：なぜ「アポB」こそがあなたの心臓にとって唯一重要な数値なのか

血液検査をするとき、おそらく「“総コレステロール”または“LDL.しかし、科学は現在、次のような別の指標を示しています。 アポリポ蛋白B 最も重要な数字として.実践する科学動脈をハイウェイだと考えてみてください。コレステロールしかし貨物は アポリポ蛋白B 「配送トラック」が荷物を運んでいるようなものです。詰まりを引き起こす可能性のあるすべての微粒子には、正確に1つのApoB分子が存在しています。それはまるで 粘着フック それにより、粒子があなたの内部に閉じ込められるようになる動脈壁。ハイウェイにトラック（ApoB値）が増えれば増えるほど、「衝突事故」すなわち心臓発作を起こす確率が高くなります。大規模な分析によると、 UKバイオバンク アポBこそが心筋梗塞を独自に予測する唯一の脂質マーカーであることを突き止めました。しかし、なぜそれがあなたの夕食の皿と結びつくのか、その「理由」がここにあります： 飽和脂肪酸 （肉類、乳製品、一部の油に含まれる）は、実際には肝臓がこれらのトラックを処理する能力を停止させてしまう。それらは「下方制御」する LDL受容体 , それは肝臓の掃除機のようなものです。これらの受容体がオフになると、ApoBというトラックは血液中に留まり続け、ぐるぐると回りながら詰まりを引き起こすのを待つのです。.それが重要な理由 <10%>の脂肪中心の食事法がこれほど効果的なのは、これらの「掃除機」の働きを再び活性化させるからです。食事から摂取する脂肪が肝臓に過剰に流れ込むのを止めると、肝臓は血液中からこれらの「ApoBトラック」をはるかに速いペースで除去し始めます。まるで、肝臓が滞留した倉庫から、高速の物流センターへと生まれ変わったかのようです。.

テイクアウト6：「臨床の分かれ道」— 自分の進む道を選ぶ

私たちは今、個人の健康上の目標に基づいて、どのように食事を摂るべきかについての明確な違いに達しました。これは “臨床上の分岐点”

予防の道： If you are currently healthy and want to stay that way, a Mediterranean-style diet is a robust and well-supported choice. It includes plenty of fruits, veggies, and legumes, with fat coming from whole nuts and seeds.
The Reversal Path: If you already have documented heart disease and your goal is to shrink existing plaque, the “moderation” of the Mediterranean diet may not be enough. The evidence points toward the ultra-low-fat (<10%), no-oil, whole-food plant-based path.実践する科学 Switching from a standard diet to this <10% path is, as scientists say, “analogous to pharmacological dose escalation.” That is a fancy way of saying it’s like increasing the dosage of a powerful medicine. You are moving from a “lifestyle choice” to a “medical-grade treatment” using food.This isn’t just a fringe theory. Even a former President of the American College of Cardiology stated that プラークの逆転 has only been proven with four methods:

The Ornish diet.
The Esselstyn diet.
High-dose Atorvastatin (a medication).
High-dose Rosuvastatin (a medication).Notice that the Mediterranean diet, while healthy, is not on that list for reversal.

Conclusion: A Simple Choice for a Complex Heart

The science of heart disease is complex, but the solution for our bodies can be remarkably simple. Our arteries are not destined to clog as we age. Instead, our bodies are built with an incredible capacity to heal, provided we stop the constant “insult” of high-fat meals and refined oils.When we lower the fat in our diet to below 10%, we aren’t just eating “healthy”—we are shifting our biology into a reparative state. We are clearing the “delivery trucks” from our blood and allowing our vessels to stay flexible and open. We are giving the “garden hose” a chance to clear itself out.If you knew your kitchen held the key to cleaning your “inner pipes,” would you change what’s on your plate today? The choice between just managing a disease and actively reversing it is now in your hands. Your body is ready to heal; it’s just waiting for you to give it the right environment to start.

ディープダイブ

The Dietary Architecture of Heart Disease Reversal:

Optimal Fat Thresholds and Food-Quality Parameters for Prevention and Clinical Regression

All citations: peer-reviewed primary literature | Vancouver/ICMJE format

抄録

Background: The clinical management of 冠動脈疾患 (CAD) has historically emphasized pharmacological risk reduction and interventional 血行再建術. However, a substantial body of peer-reviewed evidence demonstrates that the dietary macronutrient architecture can move the therapeutic needle from disease stabilization toward active プラーク退縮. This review synthesises 治験 data, population longevity studies, and vascular mechanistic research to define optimal dietary thresholds for 一次予防 and secondary regression of established CAD.

Methods: A システマティックレビュー of peer-reviewed primary literature was conducted using PubMed-indexed sources including ランダム化比較試験 (RCTs), large 前向きコホート studies, and mechanistic investigations. All non-peer-reviewed sources have been excluded and replaced with primary literature. The two principal dietary paradigms examined are: (1) the quantitative fat-restriction model (total fat <10% of calories, whole-food plant-based, no added oils — Ornish/Esselstyn protocols); and (2) the qualitative fat-substitution model (total fat 30-45%, Mediterranean pattern, replacement of saturated with 不飽和脂肪酸).

Results: The ultra-low-fat whole-food plant-based (WFPB) diet is the only dietary protocol to have demonstrated angiographically confirmed 歯垢 regression in peer-reviewed RCT data spanning five years (Ornish et al., JAMA 1998; PMID 9863851). The Mediterranean dietary model provides robust primary prevention benefits but has not demonstrated equivalent 血管造影上の退縮. The primary RCT evidence base for the Mediterranean model — the PREDIMED trial — is subject to significant methodological limitations, including ランダム化 failures in 21% of participants, early trial termination that likely inflated effect sizes, and a control arm that never achieved genuinely low-fat status. The CORDIOPREV trial confirms Mediterranean superiority over a modestly low-fat (~28-30%) comparator but does not test the <10% threshold.

Conclusion: Evidence supports a ‘clinical fork in the road’ model: a Mediterranean dietary pattern is appropriate for primary prevention in the general population; however, for patients with established CAD seeking documented plaque regression, the <10% fat WFPB protocol is the only dietary intervention with angiographic proof of efficacy in prospective clinical data.

1. はじめに

The clinical management of 心血管疾患 (CVD) has historically operated under a paradigm of pharmacological risk management and interventional symptom mitigation. This paradigm, while life-saving in acute presentations, leaves the underlying atherogenic process unresolved. A substantial body of peer-reviewed evidence suggests that the dietary macronutrient architecture — the specific distribution of fat, 炭水化物, 、およびタンパク質, along with the qualitative source of dietary lipids — can produce physiological states conducive not merely to disease stabilisation but to the clinical regression of established atherosclerotic plaque.(1,2,5)

This review investigates two competing nutritional philosophies: (1) the ‘Fat Quantity’ restriction model, pioneered by Ornish and Esselstyn, which targets total dietary fat at <10% of calories through a no-oil WFPB protocol; and (2) the ‘Fat Quality’ substitution model, supported by the AHA and embodied in the Mediterranean dietary pattern, which permits total fat intakes of 30-45% provided saturated and トランス脂肪酸 are replaced by poly- and monounsaturated alternatives. By synthesizing RCT data, population longevity studies, and vascular mechanistic research — and by rigorously evaluating the methodological limitations of each paradigm’s primary evidence base — this review defines the optimal dietary thresholds for both primary prevention and secondary disease regression.

A central finding is that the PREDIMED trial — the most frequently cited RCT for the Mediterranean paradigm — carries multiple serious methodological concerns that substantially reduce the certainty of its conclusions. These are reviewed in detail in Section 5. Investigators and clinicians citing PREDIMED as definitive evidence of Mediterranean dietary superiority over a genuinely low-fat comparator should be aware of these limitations.

2. The Quantitative Fat Paradigm: Ultra-Low-Fat WFPB Protocols for Clinical Reversal

The hypothesis that intensive total dietary fat reduction can induce regression of established atherosclerotic plaque is supported by the observation that populations with extremely low fat intake demonstrate near-complete absence of coronary 動脈 disease.(13,14) Ornish and Esselstyn independently translated this epidemiological observation into clinical protocols that challenge the conventional view of CAD as a progressive, irreversible condition.(1,2,3,4,5)

2.1 The Lifestyle Heart Trial — Randomized Controlled Evidence for Angiographic Regression

そのライフスタイル心臓トライアル (LHT) is the peer-reviewed RCT that most directly demonstrates dietary 反転 of coronary artery disease. Patients with moderate-to-severe angiographically documented CAD were randomized to an experimental program centered on a <10% fat ベジタリアン食, 有酸素運動, stress management, and 喫煙 cessation.(1,2) The dietary component prohibited all animal products except egg whites and non-fat dairy, and excluded added oils, avocados, and nuts to maintain the fat threshold.

The one-year results demonstrated that mean percent diameter stenosis decreased from 40.7% to 38.5% in the experimental group, while the control group — following standard care — saw progression from 41.3% to 42.3% (p < 0.001).(2) These specific figures derive from the JAMA 1998 five-year paper (PMID 9863851), which reports a restricted sub-cohort with complete angiographic follow-up. The original 1990 Lancet paper (PMID 1973470) reports the full randomized cohort with different figures: 40.0% to 37.8% in the experimental group and 42.7% to 46.1% in the control group — directionally consistent and statistically significant in both analyses.(1)

Table 1. Longitudinal Outcomes of the Lifestyle Heart Trial (1-Year and 5-Year Data)

Clinical Parameter	Experimental (<10% Fat WFPB)	Control (Standard Care)	Statistical Significance
Diameter 狭窄 — Baseline	40.7%	41.3%	N/A
Diameter Stenosis — 1 Year	38.5%	42.3%	p < 0.001
Diameter Stenosis — 5 Years	37.3%	51.9%	p < 0.001
Relative Change	7.9% Regression	27.7% Progression	p = 0.001
狭心症 Frequency	72% Decrease	36% Decrease	p < 0.05
Risk Ratio — Any Cardiac Event	1.0 (Reference)	2.47 (95% CI: 1.48-4.20)	Significant

Source: Ornish D et al. JAMA. 1998;280(23):2001-7 (PMID: 9863851). Data from restricted sub-cohort with complete 5-year angiographic follow-up. All figures verified against published primary source.

The five-year data show that the control group experienced a risk ratio of 2.47 (95% CI: 1.48-4.20) for major 心臓発作 relative to the experimental group, confirming that standard-care ‘moderation’ was insufficient to halt disease progression in this high-risk cohort.(2) Angina frequency declined by 72% in the experimental group versus 36% in the control arm.(2) These data support the hypothesis that the <10% fat threshold functions as a biological floor below which arterial tissue enters a reparative state not achievable at higher dietary fat intakes.

2.2 The Esselstyn Cleveland Clinic Cohort

Esselstyn’s 2014 観察研究 of 198 patients with significant CAD (J Fam Pract, PMID: 25198208) provides the largest dataset for the no-oil WFPB protocol in an advanced-disease population.(5) The cohort included patients who had already failed conventional treatments, including multiple revascularization procedures. Among the 177 patients adherent to the protocol over a mean of 3.7 years, one cardiovascular event occurred (event rate 0.6%). Among the 21 non-adherent patients, 13 events occurred (62% event rate).(5) Follow-up angiography documented 動脈硬化 reversal in 22% of adherent participants.(5) PET imaging evidence of improved myocardial perfusion under an equivalent dietary protocol was provided by Gould et al. (JAMA 1995, PMID: 7674504).(29)

An important methodological limitation of the Esselstyn cohorts is the concurrent use of cholesterol-lowering medication in most patients, making it impossible to isolate the dietary contribution from the pharmacological contribution to the observed lipid reductions and event-rate data. The magnitude of the adherent-vs-non-adherent event-rate disparity, however — with both groups receiving access to similar pharmacological support — argues for a substantial independent dietary effect beyond what medication alone would produce.(5)

3. The Qualitative Fat Paradigm: Fat Substitution and the Mediterranean Model

A contrasting paradigm, supported by the AHA and embodied in the Mediterranean dietary pattern, argues that total fat quantity is less determinative than fat quality. This ‘substitution logic’ focuses on replacing saturated fatty acids (SFA) and trans-fatty acids (TFA) with polyunsaturated (PUFA) and monounsaturated (MUFA) fats, typically achieving total fat intakes of 30-45% of calories.(24,31)

3.1 The North Karelia Project — Population-Level Evidence for Fat Substitution

その North Karelia Project provides one of the most compelling population-level demonstrations of the substitution model. By systematically shifting the Finnish population from saturated dairy fat toward rapeseed (canola) oil and increasing vegetable intake, the multi-decade public health intervention achieved an approximately 84% reduction in CHD mortality among men aged 35-64 over 40 years, as reported by Jousilahti et al. in Global Heart (PMID: 27242088).(11) Butter use on bread declined from over 90% of the population in 1972 to less than 5% by 2012, as documented by Vartiainen et al. (PMID: 27242084).(12)

Table 2. The North Karelia Project — Population-Level Impact of Fat Substitution (1972-2012)

パラメータ	1972 Baseline	2012 Outcome	Impact
Butter Usage on Bread	>90% of population	<5% of population	Substantial SFA reduction
Primary Fat Source	Saturated (dairy/butter)	Unsaturated (rapeseed/veg)	Substitution logic demonstrated
Serum コレステロール	Markedly elevated	Significant reduction	Largest CVD mortality driver
CHD Mortality (men 35-64)	Highest globally	~84% reduction	Population-level reversal

Sources: Jousilahti P et al. Glob Heart. 2016;11(2):207-12 (PMID: 27242088); Vartiainen E et al. Glob Heart. 2016;11(2):179-84 (PMID: 27242084); Puska P et al. Glob Heart. 2016;11(2):173-8 (PMID: 27242083).

3.2 The PREDIMED Trial — Critical Evaluation of the Mediterranean Diet’s Primary RCT Evidence

The PREDIMED trial is the most frequently cited RCT supporting the Mediterranean dietary pattern for primary cardiovascular prevention. Comparing a ‘low-fat’ control diet to two Mediterranean arms (EVOO-supplemented and nut-supplemented) in 7,447 high-risk participants, it reported approximately 30% reduction in major cardiovascular events in the Mediterranean groups.(6) However, as detailed in Section 5 of this review, PREDIMED carries multiple serious methodological limitations that substantially reduce the certainty of its conclusions. The reader is directed to Table 4 for a systematic analysis of these concerns.

A fundamental limitation for the central debate of this review is that the PREDIMED control arm never achieved genuinely low-fat status: mean fat intake in the control group reached 37% of calories by study end — far above any accepted definition of low-fat diet and entirely incomparable to the <10% fat threshold used in plaque-reversal research.(6,7) PREDIMED therefore demonstrates that among two high-fat dietary patterns, the Mediterranean is superior; it does not demonstrate Mediterranean superiority over a truly low-fat or WFPB comparator.

3.3 The Lyon Diet Heart Study — Secondary Prevention Evidence

そのリヨン地中海食心臓研究 provides more methodologically robust evidence for the Mediterranean paradigm’s secondary prevention capacity.(8,9) This RCT demonstrated a 73% reduction in cardiac death and non-fatal 心筋梗塞 over 27 months (adjusted RR 0.27, 95% CI 0.12-0.59, p = 0.001) in post-infarction patients following an ALA-enriched Mediterranean pattern versus a standard post-infarction diet, and was stopped early due to the magnitude of observed benefit.(9) Unlike PREDIMED, the Lyon trial does not suffer from the randomization and control-group contamination issues reviewed in Section 5, making its findings more methodologically credible.

The Lyon study does not, however, resolve the central debate: it compared a 地中海式ダイエット against a standard Western post-infarction diet, not against a WFPB or ultra-low-fat protocol. Its findings demonstrate that a Mediterranean dietary pattern substantially outperforms the standard Western diet for secondary prevention — a conclusion well-supported by the evidence. The findings do not address whether the Mediterranean pattern is equivalent to or superior to the WFPB protocol for plaque regression.

4. Population Studies and the Longevity Paradox

Two contrasting long-lived populations — the traditional Okinawans and the Mediterranean ブルーゾーン of Ikaria and Sardinia — offer informative natural experiments in dietary architecture and cardiovascular longevity.

Table 3. Comparative Macronutrient Distributions in Longevity Populations

Nutrient	Traditional Okinawa (Pre-1970)	Mediterranean Blue Zones (Ikaria / Sardinia)	Modern Western Diet
Total Fat (% kcal)	6%	30-45%	35-40%
Saturated Fat (% kcal)	2%	8-10%	12-15%
Carbohydrates (% kcal)	85%	40-45%	45-50%
Main Fat Source	Soy / seaweed	Olive oil / nuts	Butter / lard / refined oils
CAD Mortality (Relative)	~1/8 of US rate	Very low	高

Sources: Willcox DC et al. J Am Coll Nutr. 2009;28(Suppl):500S-516S (PMID: 20234038); Trichopoulou A et al. N Engl J Med. 2003;348(26):2599-608 (PMID: 12826634); Poulain M et al. Exp Gerontol. 2004;39(9):1423-9 (PMID: 15489066).

4.1 The Okinawa Model — Ultra-Low-Fat as a Cardiovascular Longevity Architecture

The traditional pre-1970 沖縄の食文化 — characterized by Willcox et al. (J Am Coll Nutr 2009, PMID: 20234038) as 6% fat / 85% carbohydrate / 9% protein, with the nutrient-dense sweet potato as the primary caloric staple — was associated with CAD mortality approximately one-eighth of contemporaneous US rates and the highest life expectancy in the world.(13,14) When Okinawans transitioned to a Westernised diet — raising total fat to approximately 27% — cardiovascular mortality increased substantially and their longevity ranking declined.(14) This secular transition provides a natural experiment in which rising total fat intake, rising SFA intake, and declining whole-plant food density occurred simultaneously with rising cardiovascular risk, supporting the quantitative fat-restriction hypothesis.

4.2 The Mediterranean Blue Zone Paradox and Its Implications

The Blue Zones of Ikaria and Sardinia consume diets with total fat often exceeding 35%, yet exhibit world-leading longevity and low cardiovascular disease rates.(15,17) This creates an apparent paradox: if ultra-low-fat intake was the critical variable in Okinawa, how do Mediterranean populations with far higher fat intake achieve comparable longevity?

Resolution of this paradox lies in the distinction between total fat quantity and fat quality, combined with the observation that both populations share the following structural commonalities: high whole-plant food density, minimal processed food intake, minimal refined sugar consumption, and minimal animal protein.(15,16,17) The critical distinction for the cardiovascular disease comparison specifically is that direct epidemiological data indicate that CAD mortality was even lower in traditional Okinawa than in Mediterranean Blue Zones — suggesting that, while both architectures protect against CVD relative to the Western diet, the ultra-low-fat WFPB pattern may confer a quantitatively superior degree of coronary arterial protection.(13,14)

5. Critical Evaluation of the PREDIMED Trial — A Systematic Methodological Analysis

Given PREDIMED’s foundational role as the primary RCT evidence base for the Mediterranean dietary paradigm in cardiovascular prevention, a rigorous evaluation of its methodological integrity is essential. The following analysis draws on the 撤回 notice and the republished paper (Estruch et al., N Engl J Med 2018, PMID: 29897866) and the peer-reviewed methodological critique by Agarwal and Ioannidis (BMJ 2019, PMID: 30733217).(6,7)

Table 4. PREDIMED Trial — Systematic Methodological Concerns

Methodological Concern	Description	Evidence-Based Assessment
Randomisation failures	21% of participants (n=1,588) were not properly individually randomised: household allocation, clinic-level assignment, and undocumented deviations at multiple sites	Retraction and republication confirmed in NEJM 2018. Directional results unchanged but certainty of evidence substantially downgraded
Trial stopped early	Halted after 4.8 of 6 planned years following interim benefit signal; early stopping systematically inflates effect sizes in dietary trials	Agarwal and Ioannidis (BMJ 2019, PMID 30733217): early stopping ‘was inappropriate given the revised results’
Control group not truly low-fat	Control arm reached 37% total fat at study end; ‘low-fat’ instruction was pamphlet-only for first 3 years. True low-fat is defined as <10% of calories	Ioannidis: ‘PREDIMED tested 41% vs 37% fat — a comparison of two high-fat dietary patterns, not Mediterranean vs truly low-fat’
Imbalanced intervention intensity	Mediterranean arms received free food (1L EVOO/week or 30g/day nuts) and intensive quarterly counselling from study outset; control group received pamphlets only for first 3 years	Differential attention and food provision confounds dietary attribution; benefits may partly reflect non-dietary effects
Stroke-dominant 複合エンドポイント	Primary composite endpoint significance driven by 脳卒中 reduction; MI and CVD mortality individually did not differ significantly between groups	NICE systematic review: PREDIMED at ‘serious’ risk of bias for individual CVD outcomes; ‘low or very low quality’ data for mortality endpoints
Secondary publication proliferation	Over 267 secondary analyses published using the PREDIMED database; multiple implausible observational claims reported (e.g., nut frequency predicting 全因死亡率)	‘Multiple contradictions between data reported across PREDIMED publications suggest a more generic problem with the trial’s quality’ (Agarwal & Ioannidis, BMJ 2019)

Sources: Estruch R et al. N Engl J Med. 2018;378(25):e34 (PMID: 29897866); Agarwal A, Ioannidis JPA. BMJ. 2019;364:l341 (PMID: 30733217).

Each of these concerns, taken individually, might be considered manageable within the context of a large, real-world dietary RCT. Taken collectively, they substantially reduce the evidentiary weight that can be assigned to PREDIMED as definitive evidence of Mediterranean dietary superiority over a genuinely low-fat comparator. Agarwal and Ioannidis concluded that ‘multiple contradictions between data reported across PREDIMED publications suggest a more generic problem with the trial’s quality’ and that ‘republication may not solve multiple problems that remain’.(7)

Importantly, these concerns were not raised in hindsight following the retraction: the NICE systematic review of PREDIMED — conducted independently and prior to the 2018 retraction — had already rated the trial at ‘serious’ risk of bias for individual CVD outcomes, with data classified as ‘low or very low quality’ for mortality endpoints.(7) This pre-retraction assessment is highly significant and argues strongly against interpreting PREDIMED as providing definitive evidence that the Mediterranean diet outperforms a genuinely low-fat dietary approach for hard cardiovascular endpoints.

The Lyon Diet Heart Study, which is free of PREDIMED’s randomization and control-group concerns and demonstrated a 73% reduction in cardiac death and non-fatal MI in a secondary-prevention population, remains the strongest single-trial dietary secondary-prevention RCT result in the peer-reviewed literature.(8,9)

6. Clinical Mechanisms: ApoB Lipoproteins, Endothelial Nitric Oxide, and Plaque Biology

6.1 Atherogenesis and ApoB-Containing Lipoproteins

Modern preventive cardiology identifies アポリポ蛋白 Bアポリポ蛋白B) as the primary causal marker of atherogenic risk. Each VLDL, IDL, 、および LDL particle contains exactly one アポB-100 molecule, meaning that plasma ApoB concentration directly quantifies the total number of potentially 動脈硬化惹起性粒子 in circulation.(21) Because 動脈プラーク initiation depends on particle-endothelium interaction probability, ApoB is a more causally precise predictor of myocardial infarction than traditional LDL-C concentration. This was confirmed in the UKバイオバンク analysis by Marston et al. (JAMA Cardiol 2022, PMID: 34773460), which demonstrated ApoB as the only lipid parameter independently associated with MI after full adjustment in a cohort of 389,529 individuals.(22)

Dietary saturated fatty acids reduce hepatic LDL受容体 expression through a mechanism of intracellular cholesterol redistribution, impairing receptor-mediated ApoB particle clearance from circulation.(23) This was established mechanistically by Woollett, Spady, and Dietschy (J Lipid Res 1992, PMID: 1552235) and confirmed in the 60-trial メタ分析 by Mensink et al. (Am J Clin Nutr 2003, PMID: 12716665).(23,24) The ultra-low-fat WFPB model achieves LDL-C and ApoB reductions through two simultaneous pathways: marked reduction in total SFA input, and an increase in the fractional catabolic rate of ApoB — the liver clears atherogenic particles substantially faster when total dietary lipid input is minimal.(1,2)

6.2 HDL Function and Reverse Cholesterol Transport

The PREDIMED sub-study by Hernaez et al. (Circulation 2017, PMID: 28193797) provides peer-reviewed evidence that an EVOO-enriched Mediterranean diet significantly improves the functional capacity of HDL for cholesterol efflux, independent of HDL-C concentration.(25) This represents a genuine mechanistic advantage of the Mediterranean model: the qualitative functionality of HDL particles is enhanced. Proponents of the ultra-low-fat model acknowledge this finding but argue that the absolute incoming cholesterol burden on a <10% fat WFPB diet is so dramatically reduced that net plaque cholesterol flux — and therefore the rate of regression — is superior even without equivalent enhancement of individual HDL particle function.(1,2,35) This specific mechanistic debate has not been resolved by any head-to-head trial measuring net plaque cholesterol balance under both conditions.

6.3 Refined Oils, Post-Prandial Endothelial Impairment, and Nitric Oxide Bioavailability

Flow-mediated dilation (FMD) of the brachial artery is the validated clinical measure of endothelial 一酸化窒素 production capacity and is a prospective predictor of cardiovascular events. Vogel, Corretti, and Plotnick (J Am Coll Cardiol 2000, PMID: 11079642) demonstrated that a single high-fat meal containing olive oil reduced brachial artery FMD by approximately 31%, an impairment comparable in magnitude to that produced by a high-saturated-fat fast-food meal.(18) A systematic review and meta-analysis by Fewkes et al. (Am J Clin Nutr 2022, PMID: 35665799) of 131 dietary trials confirmed that high-fat meals consistently and reproducibly impair post-prandial 血管内皮機能 across multiple fat types.(20)

In contrast, Cortes et al. (J Am Coll Cardiol 2006, PMID: 17045905) demonstrated that walnuts — containing equivalent fat calories to the olive-oil meal — preserved or improved post-prandial FMD, attributable to the アルギニン content (a direct nitric oxide precursor) and オメガ3脂肪酸 present in the whole-food matrix but absent in refined oil.(19) Ros et al. (Circulation 2004, PMID: 15037535) confirmed sustained FMD improvement with regular walnut consumption in 高コレステロール血症 subjects.(33)

Table 5. Post-Prandial Endothelial Response to Various Fat Sources (Flow-Mediated Dilation)

Fat Source / Meal	FMD Impact (3-4 h Post-Meal)	Mechanistic Explanation
Olive oil (refined)	~31% decrease in brachial artery dilation	Post-prandial lipemia; 酸化ストレス; NO suppression
Canola oil	Insignificant impairment	Higher ALA (omega-3) content partially protective
Salmon / fish oil	Minimal to no impairment	Protective EPA/DHA + nitric oxide synergy
Walnuts (whole food)	Preservation or improvement of FMD	Arginine (NO precursor) + omega-3s; arginine not present in refined oil
Olive oil + salad / vinegar	Significantly buffered impairment	Polyphenols そして antioxidants in vegetables neutralise post-prandial lipid stress
High-SFA meal (butter/salami)	Severe impairment	Direct pro-inflammatory endothelial insult; severe NO suppression

Sources: Vogel RA et al. J Am Coll Cardiol. 2000;36(5):1455-60 (PMID: 11079642); Cortes B et al. J Am Coll Cardiol. 2006;48(8):1666-71 (PMID: 17045905); Ros E et al. Circulation. 2004;109(13):1609-14 (PMID: 15037535); Fewkes JJ et al. Am J Clin Nutr. 2022;116(3):699-729 (PMID: 35665799).

The clinical implication of these FMD findings is direct: the Esselstyn protocol’s elimination of all refined oil — including EVOO — removes the post-prandial endothelial stressor across all meal occasions, allowing the 内皮 to maintain consistently higher nitric oxide bioavailability throughout the day. On a Mediterranean diet, endothelial function is impaired during the post-prandial window after oil-containing meals, even if fasting endothelial measurements remain favorable.(5,18,35) Vogel’s interpretation was that the benefit of the Mediterranean diet resides not in the olive oil itself but in the antioxidant-rich vegetables and vinegar consumed alongside it, which partially buffer the post-prandial endothelial stress.(18)

7. Evidence-Based Synthesis: The WFPB Protocol as the Only Proven Plaque-Reversal Strategy

7.1 The Uniqueness of the Angiographic Regression Evidence Base

The ultra-low-fat WFPB diet is, as of this review, the only dietary protocol with angiographically confirmed plaque regression in peer-reviewed RCT data with multi-year follow-up. The オーニッシュ・ライフスタイル心臓治験 documented a mean 7.9% relative regression in coronary stenosis over five years in the experimental group versus 27.7% relative progression in the control group (p < 0.001), with a 2.47-fold higher cardiac event rate in the control arm.(2) No comparable angiographic coronary regression data exist for the Mediterranean dietary pattern in a controlled prospective trial.

そのコディオプレブ trial (Delgado-Lista et al., Lancet 2022, PMID: 35525255) documented significantly reduced progression and modest regression of carotid 中内膜厚 (IMT-CC) on the Mediterranean diet compared to a low-fat comparator at 5-7 years, alongside reductions in carotid plaque height.(26,27) These are meaningful surrogate findings. However, the CORDIOPREV low-fat comparator used approximately 28-30% fat — not the <10% WFPB threshold — and the trial used carotid IMT rather than coronary angiography as the primary surrogate endpoint, making direct comparison with the Ornish angiographic regression data methodologically impermissible.(26)

The clinical position of the former President of the American College of Cardiology is directly relevant here: in peer-reviewed commentary, Williams stated that plaque regression in patients with angiographically documented CAD ‘has been shown with only four methods with an acceptable risk/benefit ratio’ — the Ornish diet, the Esselstyn diet, high-dose アトルバスタチン, and high-dose ロスバスタチン. The Mediterranean dietary pattern is not among the four.(35)

7.2 Re-evaluation of the Mediterranean Diet’s Evidence Hierarchy

The Mediterranean dietary pattern retains a robust evidence base for primary cardiovascular prevention, supported by the Lyon Diet Heart Study’s 73% event reduction in secondary prevention (PMID: 9989963), the North Karelia Project’s 84% CHD mortality reduction over 40 years (PMID: 27242088), the prospective cohort data from Trichopoulou et al. (NEJM 2003, PMID: 12826634), and the umbrella meta-analysis by Dinu et al. covering 57 meta-analyses (Eur J Clin Nutr 2018, PMID: 28488692).(9,11,15,31) This collective body of evidence is substantial and supports a genuine cardiovascular prevention role.

However, PREDIMED — the cornerstone RCT of the Mediterranean paradigm for primary prevention — cannot be considered a methodologically definitive trial given the randomization failures in 21% of participants, the early stopping that inflated effect sizes, the control arm that never achieved low-fat status, the imbalanced intervention intensity, and the NICE pre-retraction assessment of ‘serious risk of bias’.(6,7) Drawing conclusions about Mediterranean superiority over a genuinely low-fat diet from PREDIMED data is not epistemically justified.

7.3 Critical Limitations of the Ultra-Low-Fat Evidence Base

Intellectual integrity requires equally rigorous scrutiny of the WFPB evidence. The Lifestyle Heart Trial randomized only 48 patients, of whom 35 had complete five-year angiographic follow-up — a small sample with limited generalizability.(1,2) The program combined diet with exercise, stress management, and social support, making dietary isolation from the RCT data alone methodologically impossible.(2) The Esselstyn 2014 cohort is observational, not randomized, and most patients received concurrent cholesterol-lowering pharmacotherapy, limiting causal attribution to diet alone.(5)

These limitations do not negate the clinical significance of the findings but appropriately bound the certainty of the causal claims. The field urgently needs a large, adequately powered (n >= 400 per arm), multi-arm RCT directly comparing a <10% fat WFPB protocol against a Mediterranean diet against optimized medical therapy, using coronary CTA plaque quantification as the primary imaging endpoint. The ディスコCT randomized trial (Henzel et al., JACC Cardiovasc Imaging 2021, PMID: 33341413) offers a modern methodological template for intensive lifestyle intervention with coronary CTA imaging outcomes.(34)

8. Matrix of Consensus and Points of Evidence-Based Divergence

Table 6. Matrix of Consensus and Evidence-Based Divergence in Cardiovascular Nutrition

Dietary Factor	Area of Consensus	Area of Evidence-Based Divergence
Dietary Fibr	Universally recognized as critical for LDL-C and ApoB clearance via bile acid sequestration; both camps endorse high fibre intake	No conflict; higher intake universally endorsed by both paradigms
Protein Source	Both emphasize legumes, soy, and plant protein; agreement that processed meat is atherogenic	Reversal camp excludes fish entirely; Mediterranean camp endorses oily fish for omega-3 content
Refined Carbohydrates	Both identify refined grains and 添加糖 as independent cardiovascular risk drivers	No conflict; both restrict refined starches and added sugars
Saturated Fat	Both agree SFA raises LDL-C via hepatic LDL-receptor downregulation and should be <7-10% of calories	Direction identical; reversal camp targets <2% SFA as part of total fat <10%; Mediterranean camp targets <7-8%
Total Fat %	Agreement that the Western average of 35-40% total fat is atherogenic and warrants reduction	Reversal camp: <10% required for plaque regression; Mediterranean camp: 30-45% acceptable for primary prevention
Refined Oils	Both agree EVOO is vastly superior to butter and trans fats as a fat source; neither endorses partially hydrogenated oils	Reversal camp: zero added refined oil (all impair FMD in post-prandial window); Mediterranean camp: EVOO liberally endorsed
Nuts and Seeds	Population cohort data consistently show whole-nut consumption is associated with reduced cardiovascular events; both camps acknowledge this	Esselstyn camp restricts nuts in advanced CAD to maintain <10% total fat; Ornish and Mediterranean camps permit moderate nut intake

Synthesised from: Ornish D et al. JAMA. 1998;280(23):2001-7; Esselstyn CB Jr et al. J Fam Pract. 2014;63(7):356-364b; Estruch R et al. N Engl J Med. 2018;378(25):e34; Mensink RP et al. Am J Clin Nutr. 2003;77(5):1146-55.

9. Evidence-Based Conclusions and Clinical Recommendations

9.1 Primary Prevention — The Mediterranean Pattern as a Sufficient Dietary Architecture

For individuals without established coronary artery disease, a Mediterranean-style dietary pattern rich in whole fruits, vegetables, legumes, and whole grains — with fats derived primarily from whole nuts, seeds, and extra-virgin olive oil — represents the best-evidenced dietary intervention for primary cardiovascular prevention from the available peer-reviewed literature. Total fat should range from 25-35% of calories with saturated fat strictly limited to <7%, and refined carbohydrates minimized.(15,24,31) This recommendation is supported by a convergent body of RCT data (Lyon Diet Heart Study), prospective cohort studies (Trichopoulou et al.), population intervention data (North Karelia Project), and meta-analytic evidence (Dinu et al.), acknowledging that PREDIMED — while directionally supportive — should not be considered methodologically definitive.(9,11,15,31)

9.2 Secondary Prevention and Plaque Reversal — The WFPB Protocol as the Only Proven Architecture

For patients with established coronary artery disease who seek biological plaque regression, the current peer-reviewed evidence base supports the ultra-low-fat (<10% of calories) WFPB diet without added oils as the only dietary intervention with angiographic proof of efficacy over multi-year follow-up.(1,2,5) This recommendation is consistent with the clinical position articulated by the former President of the American College of Cardiology, who identified the Ornish and Esselstyn protocols as two of only four approaches — alongside high-dose atorvastatin and rosuvastatin — with an acceptable risk/benefit ratio for documented angiographic plaque regression.(35)

The biochemical target thresholds associated with documented regression — total serum cholesterol below 150 mg/dL and LDL-C below 70-85 mg/dL — are difficult to achieve consistently through diet alone on a Mediterranean pattern without significant pharmacological assistance.(1,2,3,4) The ultra-low-fat WFPB protocol achieves these targets through simultaneous mechanisms: markedly reduced dietary lipid input; increased ApoB fractional catabolic rate; elevated biliary cholesterol excretion via high plant-fiber intake; and restoration of endothelial nitric oxide bioavailability through elimination of post-prandial lipid stress from refined oils.(1,2,5,18,21)

Importantly, these two dietary architectures need not be considered mutually exclusive across a patient’s disease trajectory. A Mediterranean pattern may appropriately serve as a primary prevention strategy in early-to-middle adulthood; transition to a stricter WFPB no-oil protocol at the time of a CAD diagnosis represents a clinically justified intensification analogous to pharmacological dose escalation in response to established disease burden.

9.3 Recommendations for Future Research

The existing evidence base contains a critical gap: no adequately powered RCT has directly compared a <10% fat WFPB protocol against a Mediterranean diet against optimised medical therapy using hard imaging endpoints (coronary CTA plaque quantification or 血管内超音波検査). Such a trial, powered for a minimum of three years with coronary プラーク体積 as the primary endpoint, would resolve the central clinical question of this review and would provide the definitive data that both paradigms currently lack.

参考文献

透明性に関する注記： この記事はAIツールの支援を受けて作成されました。最終的なコンテンツは著者によって慎重に確認および編集されており、その正確性については著者が責任を負います。提供される情報は教育目的のみのものであり、医学的なアドバイスを構成するものではありません。.