Table 1.
Overview of the clinical trials (n=14)
| Observational studies | |||||
| Author, year | Aim | General characteristics at baseline | Diabetes characteristics at baseline | Study design | Major findings |
|
Wong et al, 2020 Canada [18] |
To explore the experience of individuals living with diabetes who have followed or follow the KD |
n=14 8 ♀/6 ♂ Caucasian BMI 31.5±5.1 kg/m2 |
T2D (n=11)/T1D (n=3) | Retrospective |
KD has been followed for 6 to 19 months Main motivation to start was improve blood glucose control or to reduce/stop taking diabetes medications, weight loss, and diabetes reversal Benefits such as improved glycemic control, weight loss, and satiety have been reported by participants The main challenges were the lack of support from health-care providers and information sources Most participants could continue the KD for the rest of their lives |
|
Webster et al., 2019 South Africa [19] |
To describe the foods and characteristics of a LCHF “lifestyle” that was sustainable and effective for certain T2D patients in a real-world setting |
n=28 14 ♀/14 ♂ BMI 30±6 kg/m2 |
T2D Median diabetes duration 7.4 years HbA1c 7.5 (6.5–9.5) % |
Prospective duration 15±2 months Adult who currently followed an LCHF diet for at least 6 months |
A majority of participants perceived reduced hunger and cravings LCHF diets are socially difficult to follow ↘ HbA1c (p<0.01), ↘ diabetes medications ↘ weight (p<0.001) Lack of support by their doctors which may involve suboptimal medical supervision |
| Interventional studies | |||||
| Author, year | General characteristics at baseline | Diabetes characteristics at baseline | Diet(s) | Duration of the intervention | Major findings |
|
Walton et al., 2019 USA [20] |
n=11 11 ♀ Caucasian BMI 36.3±1.4 kg/m2 |
T2D HbA1c 8.9±0.4% No medication |
VLCKD: CHO ~5% (≤30g/day), ~20–25% of protein, ~70–75% of fat. Adherence was monitored with weekly plasma ketones tests |
3 months |
↘HbA1c (p<0.0001) ↘Weight (p<0.0001) ↗ HDL-c (p<0.005), ↘TAG (p<0.005), ↘TAG: HDL-c ratio (p<0.005), ↔ LDL-c ↔ AST, ↔ ALT |
|
Romano et al., 2019 Italy [21] |
n=20 10 ♀/10 ♂ BMI 37.1±6.8 kg/m2 |
T2D Diabetes duration 5.9±1.7 years HbA1c 7.3±1.1% Medication: none (n=8), metformin (n=15), SU (n=5), insulin (10) |
VLCKD 5–10% of CHO (≤25g/day), 60–70% protein, 25–30% of fat | 2 months |
↘ HbA1c (p<0.0001), ↘ HOMA index (p<0.0001) ↘ Weight (p<0.001), ↘ BMI (p<0.001), ↘ WC (p<0.001), ↘ segmental (p<0.001), whole fat mass (p<0.001) ↘ AST (p<0.0001), ↘ ALT (p<0.0001) |
|
Myette-Côté et al., 2018 Canada [22] |
n=11 4 ♀/7 ♂ BMI 34.0±8.0 kg/m2 |
T2D Diabetes duration 6.4±4.3 years HbA1c 7.0±1.0% Medication: metformin (n=5), metformin + SU (n=2), metformin + GLP-1 (n=1), metformin + SU + DPP4 (n=1) |
3 energy-matched diets 1. Low-fat low-glycemic index diet ~55% CHO, 25% protein, 20% fat 2. VLCKD <10% CHO, 25% protein, ~65% fat 3. VLCKD + exercise <10% CHO, 25% protein, ~65% fat + 15 min of walking beginning 30 min after each meals |
4 days |
Compared to baseline: ↘ proinsulin for VLCKD (p = 0.001) and VLCKD + exercise (p=0.005), but not for low-fat low-glycemic index diet Compared to low-fat low-glycemic index diet ↘ mean glucose in the VLCKD with or without exercise (p≤0.001), ↔ time in hypoglycemia |
|
Barbosa et al., 2018 Germany [23] |
n=36 22 ♀/14 ♂ BMI 35.0±5.0 kg/m2 |
T2D HbA1c 8.9±0.4 % |
Hypocaloric VLCKD: caloric reduction of 1200–1500 kcal/day. 5–10% CHO (≤40g/day), 20–30% protein, 60–70% fat Hypocaloric low fat (LF) 1000–1200 kcal/day, 50% CHO, 20% protein, <30% fat |
3 weeks |
Compared to baseline, in the VLCKD group ↘ HbA1c in the VLCKD (p<0.001) ↘ weight (p<0.001), ↘ total body fat (p=0.001), ↘ visceral adipose tissue (p=0.024), ↘ intra hepatic fat (p=0.003) Compared to baseline, in the LF group ↘ weight (p<0.001), ↘ total body fat (p<0.001), ↘ visceral adipose tissue (p<0.001), ↘ intra hepatic fat (p<0.001) Compared to baseline, in both group ↘ T-Chol (p≤0.001), ↘ LDL-c (p≤0.004), ↘ TAG in both group (p≤0.042) |
|
Hallberg et al., 2018 USA [24] |
n=349 (262 in intervention) 175 ♀/87 ♂ BMI 40.4±8.8 kg/m2 18% African American |
T2D Diabetes duration 8.44±7.22% HbA1c 7.6±1.5 Medication: metformin (n=15), SU (n=62), insulin (10) |
Personalized CHO restriction, which provides 0.5–3.0 mmol/l of BHOB level in blood. Protein 1.5 g/kg of body weight Fat to satiety |
12 months |
↘ HbA1c (p<0.0001), ↘ FPG (p<0.0001), ↘ fasting insulin (p<0.0001), ↘ HOMA index (p<0.0001), ↘ diabetes medication, except metformin (↗) and GLP-1 (↔) ↘ weight (p<0.0001) ↘ TAG (p<0.0001), ↗ HDL-c (p<0.0001), ↘ LDL-c (p<0.0001), ↔ ApoB ↘ ALT (p<0.0001), AST (p<0.0001) |
|
Saslow et al., 2017 USA [25] |
n=34 BMI >25 kg/m2 |
T2D HbA1c >6.0% No insulin ≤3 glucose-lowering agents |
VLCKD: non-calorie-restricted diet, CHO 20–50 g/day (0.5–3.0 mmol/l of plasma BHOB level) Medium CHO, low fat: 500 kcal reduced/day, CHO 45–50%, low fat |
12 months |
Compare to control group, VLCKD ↘ HbA1c (p<0.007), ↔ fasting insulin, ↔ HOMA index ↘ medication (SU and DDP-4 inhibitors, p=0.005, metformin p=0.08) ↘ weight (p<0.001) ↔ TAG, ↔ HDL-c, ↔ LDL-c |
|
Saslow et al., 2017 USA [26] |
n=25 15 ♀/10 ♂ Different types of ethnic (60% Caucasian) |
T2D Mean diabetes duration ≈ 5 years Mean HbA1c ≈ 7.0% |
VLCKD: CHO 20–50 g/day “Create Your Plate” diet: low-fat diet, lean protein sources (1/4 of plate), and limited starchy (1/4 of plate) and non-starchy vegetables (1/2 plate) |
8 months |
Compared to the control group, VLCKD ↘ HbA1c (p=0.002) ↘ weight (p<0.001) ↘ TAG (p=0.01), ↔ HDL-c, ↔ LDL-c |
|
Wycherley et al., 2016 Australia [27] |
n=115 49 ♀/66 ♂ BMI 34.6±0.4 kg/m2 |
T2D HbA1c 7–10% |
Both diet was energy reduced VLCKD (n= 58): CHO 14%, 28% protein, 58% fat High CHO Diet (n=57): CHO 53%, 17% protein, 30% fat |
12 months |
In both group; no different effect of diet ↘ HbA1c (p<0.001) ↘ weight (p<0.001) |
|
Goday et al., 2016 Spain [28] |
n=89 58 ♀/31 ♂ BMI 33.1±1.6 kg/m2 |
T2D HbA1c 6.9±1.1% No insulin 80% glucose-lowering agents 20% lifestyle |
VLCKD (n=45), three stages: 600–800 kcal/day CHO <50 g, protein 0.8–1.2g/kg of ideal weight, fat (10 g of olive oil/day) Standard low-calorie diet (n=44) 500–1000 kcal, CHO 45–60%, protein 10–20%, fat <30% |
4 months |
Compared to baseline, in both group ↘ HOMA index (p≤0.001) ↘ BMI (p<0.0001), ↘ WC (p≤0.048) ↔ T-Chol, ↔ LDL-c, ↔ HDL-c Compared to baseline, in the VLCKD diet group: ↘ FPG (p<0.0001), ↘ HbA1c (p<0.0001), ↘ medication (p=0.0267) ↘ weight (p<0.0001) ↘ TAG (p<0.004) |
|
Tay et al., 2015 Australia [29] |
n=115 BMI 34.6±4.3 kg/m2 |
T2D HbA1c 7.3±1.1 % Duration of diabetes 8±6 years |
VLCKD (n= 57): hypocaloric. CHO 14% (50 g/d), protein 28%, fat 58% (10% saturated fat) HC diet (n=58): energy-matched, CHO 53%, protein 17%, fat 30% (10% saturated fat) + physical activity (60 min; 3 day/week) |
12 months |
Compared to baseline, in both group ↔ HbA1c, ↔ FPG ↔ BMI, ↔ WC Compared to HC diet group, VLCKD ↔ MAGE (p=0.09), ↘ CONGA-1 (p=0.003), ↘ CONGA-4 (p=0.02), ↔ time in euglycemia, ↔ time in hypoglycemia, ↘ medication (p=0.02) ↘ TAG (p=0.001), ↗ HDL-c (p=0.002) |
|
Hussain et al., 2012 Kuwait [30] |
n=363 277 ♀/86 ♂ BMI 37.3±0.3 kg/m2 |
Non-diabetic (261) and T2D (102) HbA1c 7.9±0.1% With the diet initiation, antidiabetic medications were decreased |
2 groups attributed on personal preferences. VLCKD (n=143), initial goal CHO ~ 20 g/day LCD (n=220), 2200 kcal/day |
6 months |
In the T2D subgroup and compared to baseline, in both group ↘ weight (p<0.0001), ↘ WC (p<0.0001) In the T2D subgroup and compared to LCD, VLCKD diet group ↘ TAG (p<0.0001), ↘ T-Chol (p<0.0001), ↘ LDL-c (p<0.0001), ↗ HDL-c (p<0.0001) |
|
Goldstein et al., 2011 Israel [31] |
n=52 27 ♀/25 ♂ 35–75 years BMI 30–39.9 kg/m2 |
T2D HbA1c >7% Diet or oral medication |
Stage 1 (4 weeks): Dietary Approaches to Stop Hypertension (DASH) diet with 20% kcal restriction Stage 2 (3 months): 1. Atkins diet no caloric restriction (n=26): <25g CHO/day for 6 weeks and after, ≤40g/day 2. ADA calorie-restricted diet (n=26): ♂ ≥1500 kcal/day, ♀ ≥1200 kcal/day. CHO 40–45%, protein 20%, 35g fibers |
3–12 months |
Compared to the end of stage 1, in both group ↘ HbA1c, ↔ FPG ↔ weight ↗ TAG (p=0.027), ↘ T-Chol (p=0.038), ↗ HDL-c (p=0.0026) |
Data are presented as mean±SD, or mean (range), or as n (%). p-value is indicated for changes (↗ or ↘) in brackets if provided by the authors.
Abbreviations: KD, ketogenic diet; T2D, type 2 diabetes mellitus; T1D, type 1 diabetes mellitus; HbA1c, glycated hemoglobin; FPG, fasting plasma glucose; HOMA, Homeostasis Model Assessment; SU, sulfonylurea; LCHF, low CHO high fat; VLCKD, very low calorie ketogenic diet; BMI, body mass index (kg/m2); WC, waist circumference; OW, overweight subjects; OBW, obese weight subjects; ADA, American Diabetes Association; CHO, CHO; AST, aspartate transaminase; ALT, alanine transaminase; T-Chol, total cholesterol; LDL-c, low-density lipoprotein cholesterol; HDL-c, high-density lipoprotein cholesterol; TAG, triacylglycerol; ApoB, apolipoprotein B; BHOB, beta-hydroxybutyrate; MAGE, mean amplitude of glycemic excursion; CONGA-1, continuous overall net glycemic action of observations 1 h apart; CONGA-4, continuous overall net glycemic action of observations 4 h apart