Table 3.
Randomised Controlled Trials of Plant-Based Diets and Lipid Profiles.
| Reference | Country | Population (n) | Sex | Age (Years) | Intervention (n) | Study Length/Design | Outcomes | * Results | Significance |
|---|---|---|---|---|---|---|---|---|---|
| Acharya et al. [42] | USA | Overweight/obese (143) | M/F | LOV-D: 45.2; STD-D: 43.5 | LOV-D (64) vs. STD-D (79) | 6 months (parallel) | TC, LDL-C, HDL-C, TGs, | Changes from baseline (%): LOV-D: TC: −4.7, LDL-C: −6.1, HDL-C: −5.5, TGs: −3.8. STD-D: TC: −1.2, LDL-C: −4.2, HDL-C: −3.0, TGs: −1.26 | Both diets lowered lipid outcomes from baseline, but differences between diets were non-significant (p > 0.05) |
| Ågren et al. [34] | Finland | Rheumatoid arthritis (29) | M/F | VG: 49.0; NVD: 53.0 | VG (16) vs. NVD (13) | 3 months (parallel) | TC, LDL-C, HDL-C, TGs | TC: −0.94; LDL-C: −0.74; HDL-C: −0.16; TGs: −0.11 | p < 0.001 for TC and LDL-C; p > 0.05 (ns) for HDL-C and TGs |
| Barnard et al. [26] | USA | Healthy pre-menopausal women (35) | F | All: 36.1 | LFVG vs. usual diet + placebo pill | 5 menstrual cycles for each arm (crossover) | TC, LDL-C, HDL-C, VLDL-C, TGs | TC: −0.54; LDL-C: −0.3; HDL-C: −0.2; VLDL-C: +0.08; TGs: +0.18 | p < 0.001 for all but TGs (p < 0.01) |
| Barnard et al. [29] | USA | T2DM (99) | M/F | LFVG: 56.7; ADA: 54.6 | LFVG (49) vs. ADA-recommended diet (50) | 22 weeks (parallel) | TC, non-HDL-C, LDL-C, HDL-C, VLDL-C, TGs | ITT analysis: TC: −0.09; non-HDL-C: −0.05; LDL-C: −0.03; HDL-C: −0.05; VLDL-C: +0.03; TGs: −0.04; Medication-change-adjusted analysis: TC: −0.38; non-HDL-C: −0.29; LDL-C: −0.31; HDL-C: −0.08; VLDL-C: +0.01; TGs: +0.01 | ns (p > 0.05) difference between groups for all outcomes in ITT analysis; significantly lower TC (p = 0.01), non-HDL-C (p = 0.05) and LDL-C (p = 0.02) in analyses adjusted for medication changes. |
| Barnard et al. [30] | USA | T2DM (99) | M/F | LFVG: 56.7; ADA: 54.6 | LFVG (49) vs. ADA-recommended diet (50) | 74 weeks (parallel) | TC, non-HDL-C, LDL-C, HDL-C, VLDL-C, TGs | ITT analysis: TC: −0.18; non-HDL-C: −0.21; LDL-C: −0.11; HDL-C: +0.01; VLDL-C: −0.02; TGs: −0.29; Medication-change-adjusted analysis: TC: −0.35; non-HDL-C: −0.35; LDL-C: −0.26; HDL-C: −0.01; VLDL-C: −0.05; TGs: −0.32 | ns (p > 0.05) difference between groups for all outcomes in ITT analysis; significantly lower TC (p = 0.01), non-HDL-C (p = 0.02) and LDL-C (p = 0.03) in analyses adjusted for medication changes. |
| Barnard et al. [31] | USA | T2DM (45) | M/F | LFVG: 61.0; portion-controlled: 61.0 | LFVG (21) vs. portion-controlled group (24) | 20 weeks (parallel) | TC, LDL-C, HDL-C, TGs | TC: + 0.21; LDL-C: +0.02; HDL-C: +0.03; TGs: +0.52 | ns (p > 0.05) difference between groups for all outcomes |
| Barnard et al. [28] | USA | Overweight (62) | M/F | LFVG: 58.3; MD: 56.6 | LFVG (30) vs. MD (32) | 36 weeks: 16 weeks × 2 (crossover) with a 4-week washout in between | TC, LDL-C, HDL-C, TGs, VLDL-C | TC: −0.29; LDL-C: −0.28; HDL-C: −0.11; TGs: +0.23; VLDL-C: +0.11 | Treatment effect: p = 0.04 for TC and LDL-C; p = 0.009 for HDL-C; p = 0.01 for TGs and VLDL-C |
| Burke et al. [41] | USA | Overweight/obese (176) | M/F | LOV-D: 45.4; STD-D: 43.3 | LOV-D (90) vs. STD-D (96) | 18 months: 12-month intervention, 6-month maintenance phase (parallel) | TC, TGs | Changes given in %: STD-D baseline to 18 months (preference group yes/no): TC: −1.4/+2.5; TGs: +1.0/−6.7; LOV-D (preference group yes/no): TC: +1.0/−0.1; TGs: +8.6/−5.5 | ns (p > 0.05) difference between groups for all outcomes |
| Cooper et al. [44] | USA | Healthy (15) | M/F | All: 28.0 | LOV vs. typical USA diet | 6 weeks: 3 weeks × 2 (crossover) | TC, LDL-C, HDL-C, TGs | TC: −0.52; LDL-C: −0.41; HDL-C: −0.10; TGs: −0.02 | p < 0.05 for TC; p < 0.025 for LDL-C; ns (p > 0.05) for other outcomes |
| Djekic et al. [46] | Sweden | Overweight (31) | M/F | LOV: 67.0; NVD: 68.0 | Isocaloric LOV (16) vs. NVD (15) [both adhering to Nordic Recommendations] | 12 weeks: 4 weeks × 2 (crossover) with a 4-week washout in between | TC, LDL-C, HDL-C, TGs | TC: −0.13; LDL-C: −0.10; HDL-C: −0.03; TGs: +0.06 | p = 0.01 for TC, p = 0.02 for LDL-C; ns (p > 0.05) for all other outcomes |
| Elkan et al. [35] | Sweden | Rheumatoid arthritis (66) | M/F | VG: 50.0; NVD: 50.8 | VG gluten-free (38) vs. NVD (28) | 12 months (parallel) | TC, LDL-C, HDL-C, TGs | TC: −1.2; LDL-C: −1.1; HDL-C: 0.0; TGs: 0.0 | p < 0.001 for LDL-C; no significance test reported for difference between diet groups for all other outcomes |
| Ferdowsian et al. [27] | USA | Overweight and T2DM (113) | M/F | 21 to 65 | LFVG (68) vs. usual-diet control (45) | 22 weeks (parallel) | TC, LDL-C, HDL-C, TGs | TC: −0.21; LDL-C: −0.08; HDL-C: −0.10; TGs: −0.20 | p = 0.002 for HDL-C; ns (p > 0.05) for all other outcomes |
| Gardner et al. [49] | USA | Healthy and overweight (120) | M/F | LFLOV: M: 48.0 & F: 48.0; LFD: M: 49.0 & F: 46.0 | LFLOV (59) vs. Eucaloric LFD (61) | 4 weeks (parallel) | TC, LDL-C, HDL-C, TGs | TC: −0.22, LDL-C: −0.18, HDL-C: −0.04; TGs: −0.01 | Lower TC (p = 0.01) and LDL-C (p = 0.02); ns differences for HDL-C and TGs |
| Gonciulea and Sellmeyer [47] | USA | Overweight and pre-menopausal (173) | F | APD: 62.7; DPD: 64.5; NSPD: 62.2; SPD: 64.6 | Energy- and protein-matched APD vs. DPD vs. NSPD vs. SPD | 6 weeks (parallel) | TC, LDL-C, HDL-C, TGs | SPD vs. APD: TC: −0.56; LDL-C: −0.43; HDL-C: −0.14; TGs: +0.06; SPD vs. DPD: TC: −0.77; LDL-C: −0.69; HDL-C: −0.16; TGs: +0.14; NSPD vs. APD: TC: −0.35; LDL-C: −0.26; HDL-C: −0.09; TGs: +0.05; NSPD vs. DPD: TC: −0.56; LDL-C: −0.49; HDL-C: −0.11; TGs: +0.13 | SPD vs. APD: p < 0.001 for TC and LDL-C, p = 0.008 for HDL-C; SPD vs. DPD: p < 0.001 for TC and LDL-C, p = 0.003 for HDL-C; NSPD vs. APD: p = 0.02 for TC, p = 0.04 for HDL-C; NSPD vs. DPD: p = 0.003 for TC, p = 0.005 for LDL-C, p = 0.05 for HDL-C; all other results ns (p > 0.05) |
| Hall et al. [37] | USA | Overweight (20) | M/F | All: 29.9 | LFPBD vs. ABKD | 4 weeks: 2 weeks × 2 (crossover) | TC, LDL-C, HDL-C, TGs | TC: −1.11; LDL-C: −0.72; HDL-C: −0.25; TGs: +0.34 | p < 0.001 for all |
| Hunt et al. [45] | USA | Healthy (21) | F | All: 33.2 | LOV vs. NVD | 8 weeks: 4 weeks × 2 (crossover) | TC, LDL-C, HDL-C, TGs | TC: −0.37; LDL-C: −0.25; HDL-C: −0.14; TGs: +0.06 | p = 0.001 for TC and LDL-C; p = 0.05 for HDL-C; p > 0.05 (ns) for TGs |
| Jenkins et al. [38] | Canada | Hyperlipidaemic (34) | M/F | All: 58.4 | Statin vs. Portfolio Diet vs. low saturated fat control diet | 3 × 1 month (crossover) intervention periods with a 2-to-6-week washout period between | TC, LDL-C, HDL-C, TGs | TC: −1.12; LDL-C: −0.99; HDL-C: +0.04; TGs: −0.38 | p < 0.005 for TC and LDL-C; ns (p > 0.05) for HDL-C and TGs; results were non-significantly different for all included outcomes |
| Jenkins et al. [39] | Canada | Overweight and hyperlipidaemia (44) | M/F | LCPBD: 56.1; LFLOV: 57.8 | LCPBD (22) vs. LFLOV (22) | 1-month parallel, metabolically controlled study | TC, LDL-C, HDL-C, TGs | LCPBD: TC: −1.34; LDL-C: −0.96; HDL-C: −0.05; TGs: −0.86; LFLOV: TC: −0.83; LDL-C: −0.57; HDL-C: −0.08; TGs: −0.45 | LCPBD had significantly lower TC (p = 0.001), LDL-C (p = 0.002), and TGs (p = 0.02) vs. LFLOV; ns (p > 0.05) changes in HDL-C between groups |
| Jenkins et al. [40] | Canada | Overweight and hyperlipidaemia (39) | M/F | LCPBD: 57.6; LFLOV: 55.3 | LCPBD (20) vs. LFLOV (19) | 6 months (parallel) | TC, LDL-C, HDL-C, TGs | LCPBD: TC: −0.66; LDL-C: −0.47; HDL-C: +0.04; TGs: −0.73; LFLOV: TC: −0.26; LDL-C: 0.00; HDL-C: −0.01; TGs: −0.45 | LCPBD had significantly lower TC (p < 0.001), LDL-C (p < 0.001), and TGs (p = 0.005) vs. LFLOV; ns (p > 0.05) changes in HDL-C between groups |
| Kahleova et al. [23] | USA | Overweight (222) | M/F | LFVG: 53.0; Control: 57.0 | LFVG (117) vs. usual diet control (105) | 16 weeks (parallel) | TC, LDL-C, HDL-C, TGs | TC: −0.6; LDL-C: −0.5; HDL-C: −0.01; TGs +0.20 | p < 0.001 for TC and LDL-C; p = 0.02 for TGs; ns difference for HDL-C |
| Ling et al. [36] | Finland | Healthy (18) | M/F | VG: 48.0; NVD: 37.5 | Uncooked VG (including fermented foods) vs. mixed NVD | 4 weeks (parallel) | TC, LDL-C, HDL-C, TGs | TC: −0.77; LDL-C: −0.74; HDL-C: −0.09; TGs: −0.31 | No significance tests were conducted between groups. The VG diet significantly lowered TC (p < 0.001), LDL-C (p < 0.001), HDL-C (p < 0.01), and TGs (p < 0.05) vs. baseline values. |
| Mishra et al. [24] | USA | Overweight and T2DM (291) | M/F | LFVG: 44.3; Control: 46.1 | LFVG (142) vs. usual-diet control (149) | 18 weeks (parallel) | TC, LDL-C, HDL-C, TGs | TC: −0.21; LDL-C: −0.19; HDL-C: −0.07; TGs: +0.13 | p < 0.01 for TC, LDL-C, and HDL-C; p < 0.05 for TGs |
| Nicholson et al. [32] | USA | T2DM (11) | M/F | LFVG: 51; Conventional LFD: 60 | LFVG (7) vs. conventional LFD (4) | 12 weeks (parallel) | TC, HDL-C, TGs | TC: 0.00; HDL-C: −0.18; TGs: +0.19 | p < 0.05 for HDL-C, ns (p > 0.05) for TC and TGs |
| Shah et al. [33] | USA | Coronary artery disease (100) | M/F | VG: 63.0; AHA: 59.5 | VG (50) vs. AHA-recommended diet (50) | 8 weeks (parallel) | TC, non-HDL-C; LDL-C, HDL-C, TGs | TC: −0.13; non-HDL-C: 0.00; LDL-C: −0.21; TGs: +0.11 | ns (p > 0.0015) differences between groups for all outcomes using linear regression analysis (Bonferroni correction applied) |
| Sofi et al. [43] | Italy | Overweight/obese with elevated TC or LDL-C or TGs or glucose (118) | M/F | LCLOV: 49.5; LCMD: 52.0 | Isocaloric hypocaloric LCLOV vs. LCMD | 6 months: 3 months × 2 (crossover) | TC, LDL-C, HDL-C, TGs | TC: −0.14; LDL-C: −0.24 mmol/L; HDL-C: −0.03; TGs: +0.14 | p ≤ 0.01 for LDL-C and TGs; ns (p > 0.05) for other outcomes |
| Soroka et al. [48] | Israel | Chronic renal failure (9) | M/F | 30 to 85 | Soya-based vegetarian low-protein diet vs. animal-based low-protein diet | 12 months: 6 months × 2 (crossover) | TC, LDL-C, HDL-C, TGs | TC: −0.03; LDL-C: −0.10; HDL-C: −0.07; TGs: +0.56 | ns (p > 0.05) for all comparisons |
| Wright et al. [25] | New Zealand | Overweight/obese with comorbidities (49) | M/F | All: 56.0 | LFVG (25) vs. control (normal GP care; 24) | 6 months (parallel) | TC, LDL-C, HDL-C, TGs | TC: −0.5; LDL-C: −0.4; HDL-C: −0.2; TGs: +0.2; Excluding dropouts: LFVG vs. control for TC: −0.56 | p = 0.001 for HDL-C; ns (p > 0.05) for all other differences in outcomes; p = 0.05 for differences in TC excluding dropouts |
Abbreviations: ABKD: animal-based ketogenic diet; ADA: American Diabetes Association; AHA: American Heart Association; APD: animal protein diet; DPD: dairy protein diet; F: female; GP: general practitioner; HDL-C: high-density lipoprotein cholesterol; ITT: intention to treat; LCLOV: low-calorie lacto-ovo-vegetarian diet; LCMD: low-calorie Mediterranean diet; LCPBD: low-carbohydrate plant-based diet; LDL-C: low-density lipoprotein cholesterol; LFD: low-fat diet; LFLOV: low-fat lacto-ovo-vegetarian diet; LFPBD: low-fat plant-based diet; LFVG: low-fat vegan diet; LOV-D: calorie- and fat-restricted lacto-ovo-vegetarian diet; LOV: lacto-ovo-vegetarian diet; M: male; MD: Mediterranean diet; NVD: non-vegetarian diet; ns: non-significant; NSPD: non-soy plant protein diet; PBD: plant-based diet; SPD: soy protein diet; STD-D: standard calorie- and fat-restricted diet; T2DM: type 2 diabetes mellitus; TC: total cholesterol; TGs: triglycerides; VG: vegan diet; VLDL-C: very-low-density lipoprotein cholesterol. * Results are presented as the difference between interventions (PBD vs. comparison) for all trials except for Jenkins et al. [39,40]. All lipid measurements are given as mmol/L unless otherwise stated. Age is reported as mean, median, or range.