The 4+2R Metabolic Diet was developed by Taiwanese physician Wang Ziyun (王姿允) and marketed as a phased dietary programme targeting gut microbiome remodelling and metabolic set-point recalibration. The protocol has attracted widespread attention in Taiwan via social media (Dcard, Instagram) and fitness communities. The author maintains a commercial website (4plus2r.com) and online courses, representing a potential conflict of interest. No peer-reviewed RCT directly testing the complete 4+2R protocol has been published to date.
| 階段 | 名稱 | 主要食物 | 目標 | 預設時長 |
|---|---|---|---|---|
| R1 | Clear / 清腸期 | 無糖豆漿、蛋白質粉、水 | 排除壞菌、啟動代謝 | ~1 週 |
| R2 | Renew / 植物期 | 多樣蔬菜、豆腐、菇類（無油、全素） | 養好菌、達到 −10% 體重 | 4–6 週 |
| R3 | Repair / 修復期 | R2 + 低脂肉類/海鮮 | 增肌、修復細胞 | 不固定 |
| R4 | Recode / 重編碼 | R3 + 優質澱粉（糙米、地瓜） | 增肌減脂、重設代謝定點 | 不固定 |
| R5 | Remember / 鞏固期 | 健康均衡飲食 | 穩定代謝 | 6–12 月 |
| R6 | Reset / 整合期 | 漸進回歸正常飲食 | 長期生活整合 | 6–12 月 |
四大守則（所有階段）：每日 4,000–6,000 cc 水分；全程禁乳製品；睡前 4 小時不進食；適時補充益生菌。
2.1 短期減重效果
Very-low-calorie diets (VLCD; 400–800 kcal/day) achieve rapid weight loss of 1–3 kg/week versus 0.2–0.5 kg/week for conventional low-calorie diets [7]. The R1 phase — unsweetened soy milk plus protein powder — approximates a protein-sparing modified fast (PSMF). Short-term efficacy is real; however, long-term (>12 months) weight loss is not superior to conventional low-calorie diets.
2.2 瘦體重與 RMR 下降（關鍵 RCT）
Magkos et al. 2021 (RCT, n=108, BMI 28–40, 8 weeks; PMID 34749132) compared VLCD delivering 52 g/day vs 77 g/day protein [1]. Both groups experienced identical lean mass (LM) loss: −6% from baseline and identical resting metabolic rate (RMR) reduction: −9–10% from baseline. LM contributed 27±2% of total weight loss in both arms (p=0.973). Despite the EFSA-recommended protein increase, higher protein content provided no protection.
| 研究 | 設計 | 主要效果 | GRADE |
|---|---|---|---|
| Magkos et al. 2021 [1] | RCT, n=108, 8 wk，52 vs 77 g/d protein | LM ↓6%；RMR ↓9–10%；兩蛋白質組無差異（p=0.155 / 0.389） | ⊕⊕⊕⊝ Moderate |
| Janssen et al. 2023 [2] | Review，Curr Opin Clin Nutr | 阻力運動為保護肌肉第一優先；蛋白質為輔助策略 | ⊕⊕⊕⊝ Moderate |
| Edwards-Hampton & Ard 2024 [7] | Review，Diabetes Obes Metab | VLCD 短期 1–3 kg/週；長期（>1 yr）優勢消失；限用於 BMI ≥30 + 共病者 | ⊕⊕⊕⊝ Moderate |
3.1 減重與代謝改善
Crosby et al. 2022 (RCT, n=219, BMI 28–40, 16 weeks; PMID 35452873) randomised participants to a low-fat vegan diet (~10% energy from fat) or usual diet [3]. The vegan group achieved −5.9 to −6.5 kg body weight and −3.9 to −4.3 kg fat mass. Increased legume intake was the single strongest food-group predictor of weight loss (Spearman r = −0.38, p < 0.0001); decreased meat/fish intake correlated with weight reduction (r = +0.43, p < 0.0001). Diet quality score (AHEI-2010) improved by 6.0 points, correlating with body weight, fat mass, and HOMA-IR reductions.
3.2 腸道菌相效應
A 2025 systematic review (Schoonakker et al., 36 trials) found that low-fat diets (<30% energy from fat) increased alpha-diversity, fecal SCFA, and Faecalibacterium prausnitzii [4]. Low-carbohydrate diets, by contrast, decreased Bifidobacterium and SCFA. A 2026 scoping review confirmed plant-based patterns shift microbiota toward Prevotella, Lachnospiraceae, and Bifidobacterium, creating anti-inflammatory profiles [5].
| 研究 | 設計 | 主要效果 | GRADE |
|---|---|---|---|
| Crosby et al. 2022 [3] | RCT, n=219, 16 wk | 體重 ↓5.9–6.5 kg；體脂 ↓3.9–4.3 kg；HOMA-IR↓；豆類 r=−0.38；肉↓ r=+0.43 | ⊕⊕⊕⊝ Moderate |
| Schoonakker et al. 2025 [4] | SR, 36 trials，Nutr Res Rev | 低脂飲食：alpha-diversity↑、SCFA↑、F. prausnitzii↑；低醣飲食：Bifidobacterium↓ | ⊕⊕⊝⊝ Low |
| van Kalkeren et al. 2026 [5] | Scoping Review，Adv Nutr | 植物性飲食→ Prevotella↑、Lachnospiraceae↑、Bifidobacterium↑；血糖與體組成改善（觀察性） | ⊕⊕⊝⊝ Low |
The 4+2R protocol prohibits all dairy products across all six phases. The proposed rationale — reducing gut dysbiosis and inflammation — lacks any direct RCT support. The systematic review literature consistently points in the opposite direction.
Rossi et al. 2016 (SR + meta-analysis, 24 prospective cohort studies) found no significant association between total dairy consumption and body weight change [8]. Participants with the highest dairy intake had 15% lower risk of abdominal obesity and 13% lower risk of overweight versus lowest intake. A 2024 dose-response meta-analysis (PMID 41206904) confirmed that full-fat dairy was also not associated with greater abdominal adiposity. Proposed mechanisms for dairy's neutral-to-beneficial metabolic effect include: calcium-mediated inhibition of lipogenesis and stimulation of faecal fat excretion; whey protein-induced satiety and thermogenesis; fermented dairy modulation of gut microbiota.
| 研究 | 設計 | 發現 | GRADE |
|---|---|---|---|
| Rossi et al. 2016 [8] | SR + Meta-analysis, 24 cohort studies | 乳製品攝取 vs 體重：無顯著關聯；高攝取腹部肥胖 ↓15%、過重 ↓13% | ⊕⊕⊝⊝ Low |
| 2024 Dose-Response Meta-analysis (PMID 41206904) | SR + Meta-analysis | 全脂乳品與腹部肥胖無正向關聯；整體效應中性至有利 | ⊕⊕⊝⊝ Low |
| 4+2R 全程禁乳之理論基礎 | （無 RCT） | 腸道菌相/發炎：純理論，無直接 RCT 支持；與現有 meta-analysis 方向相悖 | ⊕⊝⊝⊝ Very Low |
The R5–R6 "maintenance" phases rest on the hypothesis of a "metabolic set-point reset." The actual physiology is less optimistic: caloric restriction triggers adaptive thermogenesis, wherein resting metabolic rate decreases more than can be explained by changes in lean or fat mass alone.
Magkos et al. 2021 (n=108, 8 weeks VLCD) confirmed RMR reductions of 9–10% from baseline, irrespective of protein content — a magnitude inconsistent with mass loss alone [1]. The CALERIE 2 ancillary analysis (Sci Rep 2024) showed that 25% caloric restriction over 24 months caused sleeping energy expenditure to fall more at 12 months than predicted from any body-composition model. Neuroendocrine mediators — leptin ↓, ghrelin ↑, thyroid hormones ↓ — further drive compensatory hyperphagia and weight regain once restriction is lifted.
| 機制 / 議題 | 發現 | 主要來源 | GRADE |
|---|---|---|---|
| Adaptive thermogenesis | RMR ↓ 超過體重/體組成可解釋的幅度 | CALERIE 2 2024；Magkos 2021 [1] | ⊕⊕⊕⊝ Moderate |
| Leptin↓ / Ghrelin↑ | 熱量限制後食慾增加、飽足感下降，驅動復胖 | 多項 RCT | ⊕⊕⊕⊝ Moderate |
| VLCD vs LCD 長期體重差異 | 長期（>1 yr）：兩者無顯著差異 | Edwards-Hampton & Ard 2024 [7] | ⊕⊕⊕⊝ Moderate |
| 代謝定點重設（分階段飲食） | 無 RCT 支持；與代謝適應現象相悖 | （無直接證據） | ⊕⊝⊝⊝ Very Low |
The 4+2R theory positions gut microbiome remodelling as the central link between dietary phases and sustained weight loss. The current evidence base is less supportive.
A 2025 Cochrane systematic review (Fahim et al., 17 RCTs, n=838) evaluated probiotics, prebiotics, synbiotics, SCFAs, and FMT for obesity management [6]. Critical caveat: this Cochrane review covers only children and adolescents (0–19 years) — not adults, representing an indirectness limitation in the OE-cited evidence. Certainty of evidence was Very Low throughout. In adolescents, gut microbiome interventions showed little-to-no effect on BMI, body weight, or metabolic parameters. Prebiotics showed a small BMI reduction in one trial (n=41; MD −0.70, 95% CI −1.25 to −0.15) — a single low-powered study.
A 2025 systematic review in adults (PMID 41372103) confirmed that lifestyle changes alter gut microbiome composition, but evidence for a causal link between microbiome changes and long-term weight loss remains limited. The Firmicutes:Bacteroidetes ratio as an obesity biomarker is now considered inconsistent and unreliable across studies.
| 介入類型 | 研究 | 發現 | GRADE |
|---|---|---|---|
| 益生元/益生菌/合生元/SCFA/FMT | Fahim 2025 Cochrane [6] ⚠️ 兒科 | 多數 outcomes：無效益（Very Low certainty）；prebiotics → BMI ↓0.70（1 study, n=41） | ⊕⊝⊝⊝ Very Low |
| 飲食改變→菌相→體重維持（因果） | SR 2025 (PMID 41372103) | 菌相改變已確認；菌相→長期減重的因果鏈未確立 | ⊕⊝⊝⊝ Very Low |
| 低脂飲食→F. prausnitzii↑（腸道保護） | Schoonakker 2025 [4] | 方向一致；長期代謝臨床意義未知 | ⊕⊕⊝⊝ Low |
The protocol mandates 4,000–6,000 cc/day — approximately double standard adult recommendations (1,500–2,500 cc/day). Evidence for this practice is mixed, with the overall metabolic benefit being marginal and the risk of hyponatraemia being clinically relevant.
Drinking 500 ml of water acutely increases resting energy expenditure by ~30% for 30–90 minutes (sympathetic, osmosensitive mechanism). Pre-meal water (500 ml) can reduce energy intake or modestly augment weight loss in short-term trials. However, a 2024 systematic review and meta-analysis of RCTs (Nutrients) found no significant effect of water intake intervention on adiposity outcomes overall, with substantial heterogeneity across included trials.
| 效果面向 | 發現 | 設計 | GRADE |
|---|---|---|---|
| 急性熱量生成 | 500 ml → RMR ↑30% at 30–90 min（交感神經活化） | 小型 RCT | ⊕⊕⊝⊝ Low |
| 餐前飲水→能量攝入↓ | 500 ml 餐前攝取可減少能量攝入或增加減重 | 小型 RCT | ⊕⊕⊝⊝ Low |
| 對脂肪量的總體效應 | SR + MA（Nutrients 2024）：水分介入對脂肪量無顯著效益（異質性高） | SR of RCTs | ⊕⊕⊝⊝ Low（效益存疑） |
| 低鈉血症風險（4–6 L/d + 低鈉飲食） | 稀釋性低鈉血症：複合風險高 | 生理機制 + 臨床案例 | ⚠️ 安全警示 |
| 安全顧慮 | 機制 | 嚴重程度 | 臨床建議 |
|---|---|---|---|
| 瘦體重流失 | VLCD + 無阻力運動→肌肉萎縮、BMR↓9–10% | +++ | 加入阻力訓練（第一優先）[2] |
| 低鈉血症 | 高水分（>4 L/d）+ 低鹽全素飲食→稀釋效應 | +++ | 監測血鈉；老年人/SSRI 使用者高危 |
| 低血糖 | R1 極低卡 + 磺醯脲素/胰島素 | +++ | 用藥前須調整降糖藥劑量；醫師監督 |
| 腎功能惡化 | R1 高蛋白粉負荷 + CKD | +++ | CKD 患者絕對禁忌 |
| 鈣質不足 | 全程禁乳 + 無油（維生素 D 吸收↓） | ++ | 補充鈣 ≥1,000 mg/d + 維生素 D 800 IU/d |
| 脂溶性維生素缺乏 | 無油→A/D/E/K 吸收不良 | ++ | 每日適量健康油脂（≥1 tsp 橄欖油）或補充劑 |
| 骨密度下降 | 長期禁乳 + 低卡限制 | ++ | 定期 DXA；停經後女性高危 |
絕對禁忌：懷孕 / 哺乳、eGFR <45 的 CKD、嚴重骨質疏鬆（T-score ≤ −2.5）、第一型糖尿病。
相對禁忌（需醫療監督）：第二型糖尿病（胰島素或磺醯脲素）、高齡（≥70 歲）、飲食障礙病史、eGFR 45–59、甲狀腺功能低下。
| 聲稱 / 議題 | 效果值（N） | 研究設計 | GRADE |
|---|---|---|---|
| VLCD 短期減重（vs LCD） | ↑1–3 kg/wk（多項） | Meta-analysis of RCTs | ⊕⊕⊕⊝ Moderate |
| VLCD 長期體重優勢（>1 yr） | 無顯著差異 | Meta-analysis | ⊕⊕⊕⊝ Moderate（不支持） |
| VLCD→LM↓6%、RMR↓9–10% | LM −6%；RMR −9–10%（n=108） | RCT | ⊕⊕⊕⊝ Moderate |
| 蛋白質量↑保護瘦體重 | 52 vs 77 g/d：無差異（p=0.155） | RCT | ⊕⊕⊕⊝ Moderate（不支持） |
| 阻力運動為 VLCD 保肌首要策略 | 質性建議（專家共識） | Review | ⊕⊕⊕⊝ Moderate |
| 低脂植物性飲食→體重↓5.9–6.5 kg/16 wk | −5.9 to −6.5 kg（n=219） | RCT | ⊕⊕⊕⊝ Moderate |
| 低脂飲食→alpha-diversity↑、SCFA↑ | 方向一致，效果量異質（36 trials） | SR | ⊕⊕⊝⊝ Low |
| 菌相改變→長期體重維持（因果） | 未確立 | 觀察性 | ⊕⊝⊝⊝ Very Low |
| 禁乳製品→減重效益 | 不支持；高攝取腹部肥胖↓15% | SR + Meta-analysis | ⊕⊝⊝⊝ Very Low（不支持） |
| 代謝適應（adaptive thermogenesis） | RMR↓超過體組成預測 | RCT + Cohort | ⊕⊕⊕⊝ Moderate |
| 代謝定點重設（分階段飲食） | 無 RCT 支持 | — | ⊕⊝⊝⊝ Very Low |
| 高水分（4–6 L/d）→體脂↓ | SR：無顯著效益（RCTs） | SR of RCTs | ⊕⊕⊝⊝ Low（效益存疑） |
| 完整 4+2R 流程的臨床效益 | 無直接 RCT | — | ⊕⊝⊝⊝ Very Low |
The 4+2R Metabolic Diet combines elements drawn from VLCD literature, low-fat plant-based research, and gut microbiome theory into a commercially marketed, six-phase programme. The evidence-base assessment as of June 2026 is as follows.
- 部分組成有文獻支持：低脂植物性飲食減重（Moderate ⊕⊕⊕⊝）；VLCD 急速減重（Moderate）。
- 最薄弱的環節：「菌相重設→代謝定點改變→防止復胖」是整套理論核心，亦是最缺乏 RCT 支持之處（Very Low ⊕⊝⊝⊝）。
- 禁乳依據不成立：現有 meta-analysis 方向均與「禁乳有助減重」相反；長期禁乳需補充鈣質與維生素 D。
- 最重要的設計缺漏：無阻力運動指引。現有最強建議指出阻力運動是 VLCD 期間防止肌肉流失的第一優先，4+2R 對此完全缺失。
- 安全風險：R1/R2 複合風險（低鈉血症 + 瘦體重流失 + 鈣缺乏）在無醫療監督的自行執行下尤為危險；CKD、糖尿病用藥、懷孕者應完全避免。
- 商業背景：作者有商業利益；無獨立同儕審查 RCT；個案見證驅動社群流行。
臨床對話建議
- ✅ 認可「低脂高纖植物性飲食」的減重效益
- ➕ 強調阻力運動為必要配套（第一優先，非可選）
- ⚠️ 告知長期代謝適應的現實（停止後復胖風險高）
- 🦴 提醒補充鈣 ≥1,000 mg/d + 維生素 D（禁乳補償）
- 💧 說明 4–6 L/d 效益有限且有低鈉血症風險
- 🩺 建議在醫師 + 營養師協作下執行，自行執行風險較高
- 1Magkos F, Hjorth MF, Asping S, et al. A protein-supplemented very-low-calorie diet does not mitigate reductions in lean mass and resting metabolic rate in subjects with overweight or obesity: A randomized controlled trial. Clin Nutr. 2021;40(12):5726–5733. PMID 34749132
- 2Janssen TAH, Van Every DW, Phillips SM. The impact and utility of very low-calorie diets: the role of exercise and protein in preserving skeletal muscle mass. Curr Opin Clin Nutr Metab Care. 2023;26(6):521–527. PMID 37724991
- 3Crosby L, Rembert E, Levin S, et al. Changes in Food and Nutrient Intake and Diet Quality on a Low-Fat Vegan Diet Are Associated with Changes in Body Weight, Body Composition, and Insulin Sensitivity in Overweight Adults: A Randomized Clinical Trial. J Acad Nutr Diet. 2022;122(10):1922–1939. PMID 35452873
- 4Schoonakker MP, van Peet PG, van den Burg EL, et al. Impact of dietary carbohydrate, fat or protein restriction on the human gut microbiome: a systematic review. Nutr Res Rev. 2025;38(1):238–255. DOI 10.1017/S0954422424000131 (Cambridge Core 確認；PubMed 未收錄)
- 5van Kalkeren CAJ, Adam TC, Blaak EE. Gut Microbiome-Associated Effects of Plant-Based Diets on Glucose Homeostasis, Body Composition, and Cognitive Function: A Scoping Review. Adv Nutr. 2026;17(4):100610. DOI 10.1016/j.advnut.2026.100610
- 6Fahim SM, Huey SL, Palma Molina XE, et al. Gut microbiome-based interventions for the management of obesity in children and adolescents aged up to 19 years. Cochrane Database Syst Rev. 2025;7:CD015875. PMID 40637175 ⚠️ 研究對象：0–19 歲；不適直接外推成人
- 7Edwards-Hampton SA, Ard J. The latest evidence and clinical guidelines for use of meal replacements in very-low-calorie diets or low-calorie diets for the treatment of obesity. Diabetes Obes Metab. 2024;26(Suppl 4):28–38. DOI 10.1111/dom.15819
- 8Rossi AP, Fantin F, Zamboni G, et al. Consumption of Dairy Products in Relation to Changes in Anthropometric Variables in Adult Populations: A Systematic Review and Meta-Analysis of Cohort Studies. PLoS One. 2016;11(6):e0157461. PMID 27322888