Table 2.
Type 2 diabetes remission (%) and mean weight loss (kg) from baseline according to different dietary regimens/patterns
| Authors, yr (study) | Design | Diet INT | CON arm | Analysis and dropout during INT | T2D remission | Weight change (kg or %) | Risk of biasa | Funding | ||
|---|---|---|---|---|---|---|---|---|---|---|
| INT | CON | INT | CON | |||||||
| TDR | ||||||||||
| Taheri et al. 2020 (DIADEM-I) [53] | RCT |
TDR 3.3–3.4 MJ/d (800–820 kcal) for 12 wk, then food re-introduction over 12 wk (n = 70) |
Usual care: no diet (n = 77) |
ITT Dropout: INT = 15/70 (21%); CON = 10/77 (13%) |
1 yr: 61% (43/70) | 1 yr: 12% (9/77) | 1 yr: −12.0 | 1 yr: −4.0 | Low | Qatar National Research Fund |
| Lean et al. 2018 and 2019 (DiRECT) [51, 89] | RCT |
TDR 3.5–3.6 MJ/d (825–853 kcal) for 12 wk, then food re-introduction over 2–8 wk. (n = 149) |
Usual care: no diet (n = 149) |
ITT Dropout 1 yr: INT = 32/149 (21%); CON = 0/149 Dropout 1–2 yr: INT = 16; CON = 0 |
1 yr: 46% (68/149) 2 yr: 36% (53/149) |
1 yr: 4% (6/149) 2 yr: 3% (5/149) |
1 yr: −10.0 2 yr: −7.6 |
1 yr: −1.0 2 yr: −2.3 |
Low | Diabetes UK |
| Bynoe et al. 2020 [102] | Single arm |
TDR 3.2 MJ/d (760 kcal) for 8 wk, then food re-introduction over 4 wk (n = 25) |
N/A |
ITT Dropout:1/25 at 8 mo |
8 wk: 60% (15/25) 8 mo: 36%b (9/25) |
N/A |
8 wk: −10.1 8 mo: −8.2 |
N/A | Critical | A grant from Virgin Unite |
| Steven et al. 2016 [103] | Single arm |
TDR 2.6–2.9 MJ/d (624–700 kcal) for 8 wk, then food re-introduction over 2 wk. (n = 30) |
N/A |
ITT Dropout: 1 at 1 wk due to not meeting weight loss target |
8–10 wk: 40% (12/30) 8 mo: 43% (13/30) |
N/A |
8–10 wk: −14.2 6 mo: −13.3 |
N/A | Critical | NIHR Newcastle |
| Formula meal replacement | ||||||||||
| Gregg et al. 2012 (Look AHEAD) [49] | RCT |
Liquid meal replacement to achieve goal of 5.0–7.5 MJ/d (1200–1800 kcal) with two meal replacements during 0–20 wk and then one meal replacement thereafter (n = 2262) |
Usual care: diabetes support and education; no diet (n = 2241) |
ITT: ancillary analysis Dropout 1 yr: INT = 74/2570 (3%); CON = 112/2575 (4%) |
1 yr: 11.5% (247/2157) 2 yr: 10.4% (218/2090) 3 yr: 8.7% (181/2083) 4 yr: 7.3% (150/2056) |
1 yr: 2.0% (43/2170) 2 yr: 2.3% (48/2101) 3 yr: 2.2% (46/2085) 4 yr: 2.0% (41/2042) |
1 yr: −8.6% | 1 yr: −0.7% | Some concerns | US Department of Health and Human Services and NIH |
| Mottalib et al. 2015 (Why WAIT) [57] | Single arm |
Liquid meal replacement for breakfast and lunch to achieve goal of 5.0–7.5 MJ/d (1200–1800 kcal), 40% CHO, 30% fat, 30% protein (n = 126) |
N/A |
ITT: ancillary analysis Dropout: 38/126 (30%) at 1 yr |
1 yr: 3.2%c (4/126) | N/A | 1 yr: −7.2 in those achieving remission | N/A | Critical | See footnoted |
| Mediterranean diets and LFDs | ||||||||||
| Gutierrez-Mariscal et al. 2021 [54] | RCT |
Mediterranean diet No E restriction (n = 80) |
LFD No E restriction (n = 103) |
Complete case analysis in subset of people with CHD with T2D in original trial. Ancillary analysis | 5 yr: 41.3% (33/80) | 5 yr: 38.8% (40/103) | 5 yr: −1.16 | 5 yr: −1.4 | Some concerns | See footnotee |
| Esposito et al. 2014 [59] | RCT |
Mediterranean diet E restriction Women: 6.3 MJ/d (1500 kcal) Men: 7.5 MJ/d (1800 kcal) (n = 108) |
LFD E restriction Women: 6.3 MJ/d (1500 kcal) Men: 7.5 MJ/d (1800 kcal) (n = 107) |
ITT: ancillary analysis Dropout 1 yr: INT = 10/108 (9%); CON = 10/107 (9%) |
1 yr: 14.7% (15/102) 2 yr: 10.6% (9/85) 3 yr: 9.7% (7/72) 4 yr: 7.7% (4/52) 5 yr: 5.9% (2/34) 6 yr: 5.0% (1/20) |
1 yr: 4.1% (4/97) 2 yr: 4.7% (3/64) 3 yr: 4.0% (2/50) 4 yr: 2.9% (1/35) 5 yr: 0 6 yr: 0 |
1 yr: −6.2 | 1 yr: −4.2 | Some concerns | Second University of Naples |
| Mollentze et al. 2019 [52] | Pilot RCT |
LFDf E restriction, mainly vegetables and soups (n = 9) |
Usual care: diet advice (n = 9) |
ITT No dropout |
3 mo: NR 6 mo: 22.2% (2/9) |
3 mo: NR 6 mo: 0% |
3 mo: −9.0% 6 mo: −9.6% |
3 mo: −1.9% 6 mo: −1.5% |
High | Mr Christo Strydom, South Africa |
| Sarathi et al. 2017 [104] | Single arm |
LFD 6.3 MJ/d (1500 kcal) (n = 32) |
N/A |
ITT No dropout |
1 yr: 75.0% (24/32) 2 yr: 68.8% (22/32) |
N/A | NR | N/A | Critical | No funding |
| Dave et al. 2019 [105] | Single arm |
LFD (ADA dietg) (n = 45) |
N/A |
ITT Dropout: 4 at 5y |
1 yr: 71.1% (32/45) 5 yr: 42.2%h (19/45) |
N/A |
1 yr: −7.6 5 yr: −6.4 |
N/A | Critical | No funding |
| Ketogenic diet | ||||||||||
| Hallberg et al. 2018 and Athinarayanan et al. 2019 (VIRTA) [50, 55] | Non-RCT |
VLCKD CHO <30 g/d to achieve ketosis, 1.5 g/kg protein per d, 3–5 servings of non-starchy vegetables, multivitamin, vitamin D3 and n-3 fatty acids supplements No E restriction advised (n = 262) |
Usual care: local medical provider and education (n = 87) |
ITT: ancillary analysis Dropout 1 yr: INT = 44/262 (17%); CON = 9/87 (10%) Dropout 1–2 yr: INT = 24; CON = 10 |
1 yr: 19.8%i (52/262) 2 yr: 17.6% (46/262) |
1 yr: NR 2 yr: 2.3% (2/87) |
1 yr: −13.8 2 yr: −11.9 |
1 yr: +0.6 2 yr: +1.3 |
Serious | Virta Health |
| VLED | ||||||||||
| Umphonsathien et al. 2019 [56] | Single arm |
VLED 8 wk 2.5 MJ/d (600 kcal) food-based diet, then food re-introduction over 4 wk (n = 20) |
N/A |
ITT Dropout: 1 during run-in |
8 wk: 75% (15/20) 12 wk: 75% (15/20) |
N/A |
8 wk: NR 12 wk: −9.5 |
N/A | Critical | Prasert Prasarttong-Osoth Research Fund |
| Thomas and Shamanna, 2018 [60] | Single arm |
VLED 1 wk 2.9 MJ/d (700 kcal) food-based on diet, then advice diet for ideal body weight (n = 9) |
N/A |
ITT Dropout: 1 after completing E restriction phase |
1 yr: 22.2%j (2/9) | N/A | 1 yr: −4.2 | N/A | Critical | NR |
Remissions in Gregg et al. 2012 [49 and Esposito et al. 2014 [59] are prevalence estimates with raw cases/denominators.
aCochrane Risk of Bias tool version 2 for RCT, and Risk Of Bias In Non-randomised Studies – of Interventions for non-RCT and single-arm intervention
bITT analysis was calculated from nine participants, who had fasting plasma glucose <7 mmol/l and no medication, in a total of 25 participants. For completer analysis, remission rate was 37.5% calculated from nine out of 24 completers at 8 months
cITT analysis calculated from four out of 126 participants who had HbA1c< 48 mmol/mol (<6.5%) and no medication at 1 year. For completer analysis, remission rate was 4.6% calculated from 52 out of 88 completers
dWhy WAIT programme received contributions from Novartis Medical Nutrition (currently Nestlé HealthCare Nutrition) and LifeScan.
eMinisterio de Economia y Competitividad & the Instituto de Salud Carlos III of Spain, the Directorate General for Assessment and Promotion of Research and the European Union's (EU's) European Regional Development Fund
fSee ESM Table 15 for details
gDiet according to the recommendation of the ADA [66]
hITT analysis was calculated from 19 participants who achieved remission in a total of 45 participants. For completer analysis, remission rate was 46.3% calculated from available data at 12 months (19 out of 41 completers)
iITT analysis calculated from 52 out of 262 participants in the intervention group who had HbA1c< 48 mmol/mol (<6.5%) and no medication at 1 year. For completer analysis, remission rate was 26% calculated from available data at 12 months (52 out of 204 completers)
jITT analysis calculated from two participants who had HbA1c< 48 mmol/mol (<6.5%) and no medication at 1 year, in a total of nine participants. For completer analysis, remission rate was 25% calculated from available data at 12 months (two out of eight completers)
CON, control; d, day; DIADEM-I, Diabetes Intervention Accentuating Diet and Enhancing Metabolism-I; E, energy; INT, intervention; Look AHEAD, Action for Health in Diabetes; mo, month; N/A, not applicable; NIH, National Institutes of Health; NIHR, National Institute for Health Research; NR, not reported; T2D, type 2 diabetes; TDR, total diet replacement; VIRTA, Virta Health Corp; VLCKD, very low-carbohydrate ketogenic diet; Why WAIT, Weight Achievement and Intensive Treatment; wk, week; yr, year