European Journal of Clinical Nutrition (2022) 76:65–73

https://doi.org/10.1038/s41430-021-00909-2

ARTICLE

Metabolism and Metabolomics

Effects of intermittent (5:2) or continuous energy restriction on basal

and postprandial metabolism: a randomised study in normal-
weight, young participants
1,2,3 1,2,3
Yangfan Gao ●
Kostas Tsintzas ●
Ian A. Macdonald1,2,3 Sally M. Cordon
●
1,2,3 ●
Moira A. Taylor 1,2

Received: 23 July 2020 / Revised: 1 March 2021 / Accepted: 19 March 2021 / Published online: 26 May 2021
© The Author(s) 2021. This article is published with open access

Abstract
Background/objectives Intermittent energy restriction (IER) may overcome poor long-term adherence with continuous
energy restriction (CER), for weight reduction. We compared the effects of IER with CER for fasting and postprandial
metabolism and appetite in metabolically healthy participants, in whom excess weight would not confound intrinsic
metabolic differences.
1234567890();,:
1234567890();,:

Subjects/methods In a 2-week randomised, parallel trial, 16 young, healthy-weight participants were assigned to either CER
(20% below estimated energy requirements (EER)) or 5:2 IER (70% below EER on 2 non-consecutive days; 5 days at EER,
per week). Metabolic and appetite regulation markers were assessed before and for 3 h after a liquid breakfast; followed by
an ad libitum lunch; pre- and post-intervention.
Results Weight loss was similar in both groups: −2.5 (95% CI, −3.4, −1.6) kg for 5:2 IER vs. −2.3 (−2.9, −1.7) kg for
CER. There were no differences between groups for postprandial incremental area under the curve for serum insulin, blood
glucose or subjective appetite ratings. Compared with CER, 5:2 IER led to a reduction in fasting blood glucose con-
centrations (treatment-by-time interaction, P = 0.018, η2p = 0.14). Similarly, compared with CER, there were beneficial
changes in fasting composite appetite scores after 5:2 IER (treatment-by-time interaction, P = 0.0003, η2p = 0.35).
Conclusions There were no significant differences in postprandial insulinaemic, glycaemic or appetite responses between
treatments. However, 5:2 IER resulted in greater improvements in fasting blood glucose, and beneficial changes in fasting
subjective appetite ratings.

Introduction premature mortality [6]. Body weight (BW) reduction of

5–10% is recommended to improve metabolic function [7],
Obesity increases the risk of cardiovascular diseases [1], using continuous energy restriction (CER) [8]. However,
T2DM [2], dyslipidaemia [3], dementia [4], cancers [5] and metabolic adaptations in appetite and energy utilisation [9–
12] and poor compliance [13–16] limit success. 5:2 inter-
mittent energy restriction (5:2 IER) is a popular alternative
involving 2 days of energy restriction per week interspersed
Supplementary information The online version contains with unrestricted, or less restricted periods [17].
supplementary material available at https://doi.org/10.1038/s41430-
In people with overweight or obesity, compared with
021-00909-2.
CER, 5:2 IER, with 2, non-consecutive energy-restricted
* Moira A. Taylor days/week, resulted in comparable reductions in BW, fat
moira.taylor@nottingham.ac.uk mass and fat-free mass, and improved glycated Hb levels
1 and fasting metabolic parameters, over 3, 6 or 12 months
Division of Physiology, Pharmacology and Neuroscience, School
of Life Sciences, University of Nottingham, Queen’s Medical [18–22]. Energy restriction on 2 consecutive days/week,
Centre, Nottingham, UK compared with CER, has demonstrated greater reductions in
2
National Institute for Health Research (NIHR) Nottingham fasting insulin concentration, insulin resistance (homo-
Biomedical Research Centre, Nottingham, UK eostasis model assessment of insulin resistance, HOMA-IR)
3
MRC Versus Arthritis Centre for Musculoskeletal Ageing and fat mass over 3 or 6 months, with comparable energy
Research, Nottingham, UK deficits and weight loss [16, 23]. As a consequence of

Content courtesy of Springer Nature, terms of use apply. Rights reserved

66 Y. Gao et al.

modern lifestyles, potentially much of the 24-h period is Study design

typified by an absorptive state; however, there are limited
data on the impact of 5:2 IER on postprandial metabolism. A randomised, parallel-armed design compared 5:2 IER with
In one study, postprandial TAG decreased with 5:2 IER but CER, for 2 weeks. With 5:2 IER, food was provided for
increased with CER [24]. Differences were attributed to the 5 days to meet the participants’ estimated energy requirements
shortened absorptive period during the 2 consecutive (EER) and for 2 non-consecutive days as a 70% energy-
energy-restricted days. restricted diet. For the CER, food was restricted on all days
This study thus aims to compare both the fasting and (20% energy restriction), resulting in the same overall energy
postprandial metabolic and appetitive effects of 2 weeks restriction in the two diets. Normal dietary and physical
of 5:2 IER (with non-consecutive energy-restricted days) activity patterns were maintained during a 1-week, pre-study
and CER. Participants were healthy, young, and of nor- period commencing at the beginning of the 2nd week of the
mal weight to minimise the potential impact on treatment menstrual follicular phase, in women. Waking free-living
differences, of metabolic defects associated with over- physical activity level (PAL, daily step counts) was estimated
weight and obesity. All food and beverages were pro- using a pedometer (Omron Walking Style One 2.1, Omron,
vided to improve compliance. The main objective of this USA). Following a pre-intervention laboratory visit and a 7-
study was to compare the incremental area under the day gap (to ensure post-intervention measurements were also
curve (iAUC) for serum insulin, concurrently with other conducted during the 2nd week of the follicular phase for
metabolic and appetite responses, for 3 h after a stan- women) the experimenter randomly assigned the participants
dardised liquid breakfast following 2 weeks of CER and to a treatment group using computer-generated randomisation
5:2 IER. (https://www.randomizer.org/) and the intervention period
began. On Day 7 of each 2-week intervention, participants
attended the laboratory in the morning, fasting, for BW
Methods assessment and subcutaneous continuous glucose monitoring
device (CGM) attachment. Glucose values were reviewed
Participants across 24 h, the day hours and the night hours, for the mean,
maximum, minimum, iAUC, SD and %CV. Whole-day gly-
Sixteen healthy men and women (20–35 years, BMI caemic profile was plotted by calculating the mean of each of
20–25 kg/m2, waist circumference < 94 cm for men and the 6 days of measurement (Days 8–13) at a timepoint, for a
<80 cm for women) were recruited from the University of participant, and then calculating the mean at the timepoint for
Nottingham and Queen’s Medical Centre. They were the six participants in each treatment group.
moderately physically active (International Physical
Activity Questionnaire (IPAQ)) [25] with a stable weight Dietary intervention periods
(±2 kg) over the previous 3 months. Exclusion criteria are
shown in Supplementary Information 1. Participants Energy provision was based on estimated basal metabolic
attended five visits (1 × 10 min, 1 × 30 min, 1 × 1 h and rate [30] and daily PAL [31, 32] adjusted for the deficit
2 × 6 h) from November 2018 to May 2019 (Supplemen- required. The 70% energy restriction applied on Days 1, 4,
tary Fig. 1). The study was approved by the University of 8 and 11 of 5:2 IER (Supplementary Table 1). The 3-day
Nottingham, Faculty of Medicine and Health Sciences menu, included three meals/day, had identical food items
Research Ethics Committee (Reference Number 121- (same proportions but with adjusted quantities to meet
1809), conformed to the World Medical Association individual EER and the protocol), meal times and daily
Declaration of Helsinki and was registered at Clin- distribution of energy, between groups (Supplementary
icalTrials.gov PRS: NCT04138160. Table 2). Total energy from carbohydrate, fat and protein
was 50%, 31% and 19%, respectively (Supplementary
Screening Information 2). Home deliveries or participant collection
from the laboratory occurred weekly. Participants were free-
BW, height, waist and hip circumferences, heart rate (HR), living but instructed to follow the prescribed meal plan.
systolic and diastolic blood pressure (BP) were measured Submission to the experimenter, in real time, of a photo-
and a blood sample obtained for routine investigation. graph of each meal was required prior to eating; leftovers,
Skinfold thicknesses (four sites) were measured to estimate or additional food eaten, were logged. Interstitial glucose
body fat percentage [26, 27]. A Personal Information and was monitored by a Medtronic MiniMed iPro™2 (North-
General Health form, the Beck Depression Inventory [28], ridge, CA, USA) CGM system (Days 8–13) with four
the Eating Attitudes Test-26 [29] and the IPAQ were calibration capillary finger-prick glucose measures per day
completed. (ACCU-CHEK® Performa Blood Glucose Meter, Roche,

Content courtesy of Springer Nature, terms of use apply. Rights reserved

Effects of intermittent (5:2) or continuous energy restriction on basal and postprandial metabolism: a. . . 67

Germany) and the pattern of glucose excursions was used to USA) and the remaining sample was used to obtain serum or
verify the meal times. plasma. Serum insulin was determined by the human radio-
immunoassay kit (HI-14K; Merck Millipore, MA, USA) [40],
Laboratory-visit protocol pre- and post-intervention plasma FFA concentrations using a kit (NEFA-HR-2, Wako,
Germany) run on a photometric auto-analyser (ABX Pentra
Fasting appetite and metabolic measurements were made at 400, Horiba Ltd., France) as were serum total cholesterol,
~8:30 a.m., following an overnight fast, and for 3 h, fol- HDL-cholesterol, LDL-cholesterol and TAG concentrations
lowing a standardised liquid breakfast (meal tolerance test, with reagents from Horiba Medical (France).
MTT), consumed at ~9:30 a.m. (Ensure® Compact, Abbott
Nutritional Ireland, 10 kJ/mL, vanilla flavour; 42 kJ/kg BW). Statistical analyses
Baseline resting energy expenditure (REE) and respira-
tory exchange ratio (RER) were measured using a Quark Prism 8 software (GraphPad Software Inc., CA, USA) was
CPET open-circuit metabolic cart (Cosmed, Quark CPFT, used for data entry and analyses (statistical significance
Rome, Italy) (mid 15 min of each 20 min measurement) and accepted as P < 0.05). Values in the text, figures and tables are
the abbreviated Weir equation [33]. Postprandial REE was mean and SD unless otherwise stated. Data were tested for
measured (20 min in every hour) for 3 h after the MTT, normality using the D’Agostino–Pearson normality test.
enabling calculation of diet-induced thermogenesis (DIT). iAUC and decremental area under the curve (dAUC) of the
Two baseline arterialised venous blood samples were postprandial measurements were calculated using the trape-
obtained from a retrograde 20-G cannula (Ohmeda, Swe- zoid rule. Unpaired Student’s t tests compared single data
den) placed in a dorsal hand vein, using a warm-air box points between the two groups. Two-factor Mixed Model
(55 °C) [34], for mean fasting whole-blood glucose, serum ANOVA (between groups factor: treatment, 5:2 IER vs. CER;
insulin, plasma FFA, serum TAG, total cholesterol, HDL- within groups factor: sampling time; pre vs. post each 2-week
cholesterol and LDL-cholesterol. Further arterialised sam- energy restricted intervention) evaluated intervention effects;
ples were obtained every 10 min for glucose and every when significant treatment, time or treatment × time interac-
20 min for insulin, FFA and TAG concentrations, for 3 h tion effects were observed, post hoc comparisons (where
after the MTT. The HOMA-IR was used to quantify fasting relevant) were explored using paired or unpaired Student’s t
insulin sensitivity and pancreatic β-cell function [35] and test. Effect size was estimated as partial eta squared (η2p),
the Matsuda index, over 180 min, for evaluation of insulin which is the standard method giving the proportion of var-
sensitivity after the MTT, reflecting hepatic and peripheral iance associated with two-way ANOVA analysis.
tissue postprandial sensitivity to insulin [36]. In response to limited previous studies on postprandial
Subjective appetite ratings were obtained at baseline and responses to 5:2 IER or 5:2 intermittent fasting, the power
every 20 min for 3 h after the MTT and every 20 min for an analysis was based on an intermittent fasting study using
hour after the ad libitum test meal using visual analogue glucose disposal as an index of postprandial insulin sensi-
scales [37]. The composite appetite score (CAS) was cal- tivity [41]. The statistical power analysis indicated that eight
culated as follows [38]:] participants were required per group to detect a 16%
improvement in postprandial whole-body insulin sensitivity
CAS ¼ ½hunger þ desire to eat þ prospective food consumption with a power of 80% at a significance level of P < 0.05.
þ ð100  fullnessÞ þ ð100  satisfactionÞ=5 The primary outcome of this study was the iAUC for
insulin for 3 h after a standardised liquid breakfast. Secondary
A higher CAS value is indicative of greater motivation to outcomes were changes in postprandial glucose, FFA, TAG,
eat or less feeling of satiety. CAS and Matsuda index, energy intake of the ad libitum test
Three hours after the MTT, a ~430 g portion of a pasta meal and whole-day glycaemic profiles for the average of 6
based test meal was served at ≥82 °C, as a standardised consecutive days. Exploratory outcomes included fasting
measure of ad libitum food intake [39] (composition: blood glucose, fasting ratings of subjective appetite, fasting
620 kJ/100 g with 49.8% carbohydrate, 15.4% protein and REE, DIT and fasting and postprandial RER responses.
34.8% fat) to participants instructed to eat until they felt
‘comfortably full’. The bowl was repeatedly topped up,
when about two thirds had been consumed. A final blood Results
sample was taken 1 h after the ingestion of the ad libitum
test meal, for determination of glucose concentration. Participant characteristics
Immediately after each sampling, whole-blood glucose
was analysed with a Yellow SpringsTM glucose and lactate Eight participants per group completed the study (CONSORT
analyser (YSI 2300 STAT PLUS, Yellow Springs Inc., Ohio, diagram in Supplementary Fig. 2) with characteristics at

Content courtesy of Springer Nature, terms of use apply. Rights reserved

68 Y. Gao et al.

Table 1 Anthropometric, physiological characteristics and blood measurements before (pre) and after (post) each interventiona.
5:2 IER (n = 8) CER (n = 8) 5:2 IER vs.
CERb
Pre Post Pre Post
Mean ± SD Mean ± SD Mean ± SD Mean ± SD

Body weight, kg 63.4 ± 14.0 60.9 ± 13.2 66.1 ± 11.2 63.9 ± 11.0 NS
Waist, cm 71.1 ± 8.7 68.7 ± 8.1 75.2 ± 6.7 70.8 ± 5.2 NS
Hip, cm 96.3 ± 4.7 93.8 ± 6.1 96.7 ± 7.2 93.6 ± 7.1 NS
Heart rate, BPM 68 ± 6 62 ± 6 60 ± 8 60 ± 7 <0.05
DBP, mmHg 65 ± 2 65 ± 5 63 ± 8 63 ± 5 NS
SBP, mmHg 110 ± 11 108 ± 11 104 ± 13 103 ± 10 NS
Fasting glucose, mmol/L 4.39 ± 0.28 4.07 ± 0.23 4.42 ± 0.16 4.43 ± 0.17 <0.05
Fasting insulin, mIU/L 9.40 ± 3.06 7.78 ± 2.87 7.98 ± 1.54 6.43 ± 1.78 NS
Fasting FFA, mmol/L 0.46 ± 0.20 0.43 ± 0.10 0.43 ± 0.16 0.45 ± 0.11 NS
Fasting TAG, mmol/L 0.80 ± 0.37 0.64 ± 0.28 0.68 ± 0.26 0.60 ± 0.16 NS
Fasting total cholesterol, mmol/L 3.64 ± 0.37 3.35 ± 0.45 3.76 ± 0.63 3.62 ± 0.54 NS
Fasting LDL-C, mmol/L 2.03 ± 0.23 1.87 ± 0.41 2.15 ± 0.42 2.06 ± 0.43 NS
Fasting HDL-C, mmol/L 1.17 ± 0.30 1.06 ± 0.25 1.20 ± 0.21 1.16 ± 0.16 NS
Glucose iAUC, mmol/L × 180 min 261 ± 124 334 ± 173 222 ± 76 247 ± 65 NS
Insulin iAUC, mIU/L × 180 min 8633 ± 4016 7419 ± 2983 6270 ± 1721 6330 ± 2436 NS
TAG iAUC, mmol/L × 180 min 62 ± 60 36 ± 22 42 ± 32 29 ± 22 NS
FFA dAUC, mmol/L × 180 min −54 ± 30 −46 ± 10 −52 ± 22 −53 ± 17 NS
Fasting CAS, mm 83 ± 4 63 ± 14 73 ± 11 83 ± 12 <0.001
CAS dAUC (180 min after −5570 ± 2549 −5853 ± 1592 −5355 ± 2093 −6112 ± 2241 NS
the MTT)
CAS dAUC (60 min after the ad −4944 ± 1248 −3850 ± 1546 −4472 ± 1079 −4580 ± 1274 NS
libitum test meal)
Daily step count (habitual levels 9650 ± 2771 9780 ± 3227 9725 ± 2444 9481 ± 1869 NS
and levels during interventions)
5:2 IER 5:2 intermittent energy restriction, CER continuous energy restriction, NS not statistically significant, DBP diastolic blood pressure, SBP
systolic blood pressure, iAUC incremental area under the curve, dAUC decremental area under the curve, CAS composite appetite score, MTT meal
tolerance test (a standardised liquid breakfast).
a
Blood measurements include fasting blood measurements, postprandial incremental area under the curve (Glucose, Insulin and TAG) and
decremental area under the curve (FFA) for blood measurements over 180 min after consumption of the liquid breakfast.
b
Two-way ANOVA, treatment-by-time interaction.

screening, detailed in Supplementary Table 3, showing fasting levels within 180 min (Fig. 1a). There was no sig-
matching between groups for age, sex and anthropometrics. nificant treatment-by-time interactions between groups (P
5:2 IER and CER were equally effective for weight loss = 0.31, two-way ANOVA, η2p = 0.01) or main effects of
(mean weight change: −2.5 (SD 1.0) kg 5:2 IER, −2.3 (SD treatment or time in iAUC for insulin over 180 min post-
0.7) kg CER after 2-week intervention; −1.3 (SD 1.2) kg 5:2 prandially (Table 1).
IER, −1.1 (SD 0.5) kg CER after 1-week intervention) and
had comparable reductions in waist and hip circumferences. Fasting and postprandial whole-blood glucose
There was a significant treatment-by-time interaction for responses
fasting HR (P < 0.05, two-way ANOVA). Both groups had
similar habitual daily steps, which were comparable to during Whole-blood glucose concentrations increased rapidly after
the intervention, as was fasting BP (Table 1). the MTT and remained above fasting levels for 180 min
(Fig. 1b). There were no significant treatment-by-time
Postprandial serum insulin responses interactions or main effects of treatment or time in iAUC for
whole-blood glucose (Table 1). However, as an exploratory
Serum insulin rapidly increased and peaked after finding, there was a significant treatment-by-time interac-
20–40 min, before steadily declining, but did not return to tion between the 5:2 IER and CER regimens for fasting

Content courtesy of Springer Nature, terms of use apply. Rights reserved

Effects of intermittent (5:2) or continuous energy restriction on basal and postprandial metabolism: a. . . 69

Fig. 1 Metabolic response to the liquid breakfast and ad libitum measurement, 5:2 IER 5:2 intermittent energy restriction, CER con-
meal. Serum insulin (a), whole-blood glucose (b) following the liquid tinuous energy restriction, FFA free fatty acid, TAG triglycerides,
breakfast (both parameters) and the ad libitum meal (glucose only); HOMA-IR homoeostasis model assessment of insulin resistance. a–d
insulin sensitivity assessed using the HOMA-IR (c) and Matsuda index Values are means with their standard errors. a–d n = 8 per group.
(d). Pre- pre-intervention measurement, Post- post-intervention

blood glucose (P = 0.02, two-way ANOVA, η2p = 0.14); Fasting and postprandial CAS responses
with post-intervention values significantly lower after 5:2
IER than CER (P < 0.01) (Table 1). There were no significant treatment-by-time interactions in
CAS dAUCs over 3 h after the MTT or 1 h after the ad
Whole-body insulin sensitivity libitum test meal between the two groups (Fig. 2 and Table
1). Interestingly, fasting CAS showed significant treatment-
There was no significant treatment-by-time interaction by-time interaction between the two groups (P < 0.001)
between groups in HOMA-IR (Fig. 1c); but there was a (Table 1) with fasting CAS decreasing after the 5:2 IER and
tendency to a decrease in both groups over the 2-week increasing after the CER treatment.
intervention period (P = 0.052, main effect of time, two-
way ANOVA, η2p = 0.15). Similarly, no significant Energy intake of the ad libitum test meal
treatment-by-time interaction between groups was detected
in the Matsuda index (over 180 min) (Fig. 1d). However, There was no treatment-by-time interaction or a main effect
the Matsuda index increased in both groups over the 2-week of treatment or time for ad libitum test meal energy intake:
interventions (P = 0.02, main effect of time, two-way 2253 (SD 956), 1798 (SD 704), 1898 (SD 462) and 1883
ANOVA, η2p = 0.08), indicating improvement in whole- (SD 857) kJ in the pre- and post-5:2 IER and the pre- and
body insulin sensitivity. post-CER visits, respectively. No interaction or main effects
of eating duration or eating speed were identified between
Postprandial FFA and TAG responses groups.

A reduction in plasma FFA concentrations occurred in both Free-living continuous glucose monitoring
groups, following the MTT (Supplementary Fig. 3a and
Table 1). TAG concentrations steadily increased after meal Whole-day glycaemic profiles, with each timepoint repre-
consumption (Supplementary Fig. 3b and Table 1). There senting the mean of 6 consecutive days (Days 8–13) per
were no significant treatment-by-time interactions or main participant, averaged over the participants, by treatment, are
effects of treatment or time between groups in AUCs for presented in Fig. 3 (restricted intake on Days 8 and 11 for
plasma FFA and serum TAG. the 5:2 IER group). Shaded error bands (SEM) indicate the

Content courtesy of Springer Nature, terms of use apply. Rights reserved

70 Y. Gao et al.

100 CER pre

inter-individual variation for the mean of 6 days. Both
CER post

80 5:2 IER pre groups exhibit sharp elevations in interstitial glucose con-
5:2 IER post centrations following consumption of the three main meals,
Distance (mm)

60 peaking after ~1 h and returning to within the fasting range

40
before the next meal. Responses were similar between the
two groups (Table 2).
20

0
Resting energy expenditure (indirect calorimetry
BL 0 20 40 60 80 100 120 140 160 180 210 230 250 270 data)
Time (min)

Fig. 2 Mean composite appetite score (CAS). 5:2 IER pre/post-5:2 There was no significant difference between groups in
intermittent energy restriction pre-intervention/post-intervention mea-
surement, CER pre/post continuous energy restriction pre-intervention/
fasting REE although a significant main effect of time was
post-intervention measurement. Values are means with their standard observed (P < 0.05, two-way ANOVA), which showed a
errors. n = 8 per group. reduction by ~0.26 kJ/min in 5:2 IER and by 0.08 kJ/min in

5:2 IER
7
CER
Interstitial Glucose mmol/L

0
57

57

57

57

57

57

57

57

57

57

57

7
:5

:5

:5

:5

:5

:5

:5

:5

:5

:5

:5

:5

:5
0:

1:

2:

3:

4:

5:

6:

7:

8:

9:

:
10

11

12

13

14

15

16

17

18

19

20

21

22

23
Time-point (hr:min)

Fig. 3 Whole-day glycaemic profiles calculated as mean interstitial are means with their standard errors. SEM for the 5:2 IER is the light
glucose concentrations of 6 consecutive days (Days 8–13) for six shading and for the CER is the dark shading. n = 6 per group. Black
participants of the 5:2 IER (○) and CER (●). 5:2 IER 5:2 inter- arrows at 08:00, 13:00 and 18:00 indicate when the breakfast, lunch
mittent energy restriction, CER continuous energy restriction. Values and dinner started, respectively.

Table 2 Indices derived from 24 h continuous glucose monitoring during Days 8–13 of both interventions.
Full 24 h Day hours (0700–2359) Night hour (2400–0659)
5:2 IER (n = 6) CER (n = 6) P a
5:2 IER (n = 6) CER (n = 6) Pa
5:2 IER (n = 6) CER (n = 6) Pa
Valid days Days 8–13 Days 8–13 Days 8–13 Days 8–13 Days 8–13 Days 8–13
Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD Mean ± SD

Mean glucose, mmol/L 4.8 ± 0.3 5.0 ± 0.3 NS 5.0 ± 0.4 5.1 ± 0.3 NS 4.5 ± 0.3 4.6 ± 0.3 NS
Maximum, mmol/L 6.0 ± 0.6 6.3 ± 0.6 NS 6.0 ± 0.6 6.3 ± 0.6 NS 4.7 ± 0.4 4.8 ± 0.3 NS
Minimum, mmol/L 4.3 ± 0.4 4.4 ± 0.4 NS 4.4 ± 0.4 4.4 ± 0.4 NS 4.4 ± 0.3 4.5 ± 0.3 NS
SD 0.4 ± 0.2 0.5 ± 0.1 NS 0.4 ± 0.2 0.5 ± 0.2 NS 0.1 ± 0.0 0.1 ± 0.0 NS
% CV 8.4 ± 3.2 9.4 ± 3.1 NS 8.0 ± 3.0 9.1 ± 3.8 NS 1.6 ± 0.6 1.6 ± 0.6 NS
5:2 IER 5:2 intermittent energy restriction, CER continuous energy restriction, NS not statistically significant, % CV percentage coefficient of
variation.
a
Unpaired Student’s t test.

Content courtesy of Springer Nature, terms of use apply. Rights reserved

Effects of intermittent (5:2) or continuous energy restriction on basal and postprandial metabolism: a. . . 71

CER (Supplementary Fig. 4a). DIT (iAUC for the entire common with a study in which participants had obesity; but
postprandial REE response) showed no significant interac- in contrast to another considering participants with normal
tion or treatment or time effect between diet groups (two- weight [43, 44]. In participants with T2DM and overweight
way ANOVA) (Supplementary Fig. 4b); nor did either or obesity, both groups reported comparable reductions in
fasting or postprandial RER responses differ between appetite and increased feelings of fullness and satisfaction
groups (Supplementary Fig. 4c). [18]. However, participants were taking hypoglycaemic
drugs, potentially impacting on glycaemia and appetite.
Hyperphagia was not seen in either group at the ad
Discussion libitum lunch (energy intake: 20% of their EER), but
decreased fasting CAS post-5:2 IER was not associated
The present study demonstrates that under free-living con- with a significantly decreased energy intake at this meal;
ditions, 2 weeks of 5:2 IER fails to improve postprandial although a numerical decrease was observed with post-5:2
insulin iAUC response (primary outcome) compared with IER intervention. An accumulatively greater difference in
CER in healthy young men and women. Nor were post- total energy intake might occur across all meals of the day.
prandial glucose, FFA, TAG, CAS or energy intake at the However, subjective and objective measures of appetite
ad libitum test meal (secondary outcomes) different. 5:2 may not always align [45], especially in relatively short-
IER resulted in a decrease in fasting glucose and fasting term interventions.
CAS compared with CER (exploratory outcome); con- The present study extends the investigations of inter-
current weight loss was comparable (3–4% of original BW). mittent fasting [41, 46] and other metabolic challenges
The absence of differences in postprandial insulinaemic [39, 47] by considering both fasting and postprandial state
response (with small effect size) and postprandial gly- in young, normal-weight participants in whom effects on
caemic, FFA and TAG responses to MTT suggests that metabolism and appetite are not confounded by the meta-
metabolic differences (e.g., fasting insulin and insulin bolic abnormalities associated with excess BW. Importantly
resistance) observed between these diets in previous, longer when studying meal pattern and metabolism, and in contrast
trials [16, 23] cannot be explained by changes in post- to previous studies, food was provided over the intervention
prandial responses. However, lower fasting glucose fol- period and compliance monitored by CGM. Three major
lowing 5:2 IER in our current study (with large effect size) excursions in daily glucose profiles were seen in the CGM,
has been observed previously [42]. Further investigation is coinciding with prescribed meal times, reflecting good
warranted, of improved hepatic insulin sensitivity, to elu- dietary adherence, facilitated by food provision. This
cidate the mechanisms underlying the observed differences counters the potential limitation of the participants being
in fasting blood glucose levels. free-living [48, 49], as did monitoring activity by using step
Whole-body postprandial insulin sensitivity (Matsuda counts. The balanced sex ratio of participants may help to
index) improved equally (with intermediate effect size), offset any sex bias, hence making this study more repre-
along with a strong tendency for increased fasting insulin sentative of the general population. In this study, pre-
sensitivity (HOMA-IR) (with large effect size) in both intervention metabolic measurements may be confounded
groups, which were probably attributable to equivalent by the dinner consumed the evening before. A standardised
weight loss. Greater reductions in fasting insulin con- dinner, prior to each measurement day, may help to mini-
centration and insulin resistance (HOMA-IR) were reported mise the effects of diverse dietary characteristics of the
following 5:2 IER compared with CER in two similar stu- dinner on fasting and postprandial substrates metabolism
dies [16, 23], however, 2-day energy restriction was on [50]. A longer study period would confirm stability of the
consecutive days and participants had overweight or obe- effects noted.
sity. Importantly, in our and the previous studies, assess- This study shows that 5:2 IER is not superior in
ments were made at least 3 days after the last energy- improving postprandial insulin and glucose responses
restricted day with 5:2 IER; favourable metabolic effects of compared with CER, suggesting that these are not key to
5:2 IER appear not to be due to an acute carry over effect of metabolic differences in longer-term intervention studies.
greater negative energy balance during the last energy- Interestingly, the 5:2 IER regimen led to potentially bene-
restricted day with 5:2 IER. ficial changes in fasting blood glucose and fasting sub-
Although no significant differences in postprandial CAS jective appetite scores, which were independent of weight
response were observed between interventions, beneficial loss, over a 2-week period. This was consistent with similar
changes in fasting CAS (with large effect size) and indivi- studies conducted in participants with overweight or obe-
dual subjective appetite and satiety ratings were observed in sity. Further studies are required to establish mechanisms of
the 5:2 IER but not CER group. These differences might be action and to examine long-term adherence, safety and
expected to support long-term compliance and are in efficacy of intermittent fasting.

Content courtesy of Springer Nature, terms of use apply. Rights reserved

72 Y. Gao et al.

Acknowledgements The authors thank staff at the David Greenfield mortality: individual-participant-data meta-analysis of 239 pro-
Human Physiology Unit for assistance during the trial and the parti- spective studies in four continents. Lancet. 2016;388:776–86.
cipants and Andrew Wilhelmsen and Chloe Monnier for proofreading. https://linkinghub.elsevier.com/retrieve/pii/S0140673616301751.
The study was funded by the University of Nottingham. 7. Jensen MD, Ryan DH, Apovian CM, Ard JD, Comuzzie AG,
Donato KA, et al. 2013 AHA/ACC/TOS guideline for the man-
Author contributions Designed the research: YG, IAM, MAT and KT; agement of overweight and obesity in adults. Circula-
conducted research and analysed the data: YG; conducted blood tion.2014;129(25 Suppl 2):S102–38. http://circ.ahajournals.org/
analysis: YG and SMC; wrote the paper: YG, MAT, KT and IAM. All lookup/doi/10.1161/01.cir.0000437739.71477.ee.
authors read and approved the final version of the paper. 8. National Institute for Health and Care Excellence. Weight man-
agement: lifestyle services for overweight or obese adults. 2014.
Available from: https://www.nice.org.uk/guidance/ph53/chapter/
Compliance with ethical standards 1-Recommendations.
9. Hall KD, Heymsfield SB, Kemnitz JW, Klein S, Schoeller DA,
Conflict of interest The authors have no direct financial or personal Speakman JR. Energy balance and its components: implications
competing interests to declare. Full list of potential conflicts for IAM over for body weight regulation. Am J Clin Nutr. 2012;95:989–94.
the past 5 years: Scientific Advisory Boards: Nestle Research, Novo- https://academic.oup.com/ajcn/article/95/4/989/4576902.
zymes, AIJN (European Fruit Juice Consortium), ILSI Europe Dietary 10. Westerterp KR. Exercise, energy balance and body composition.
Carbohydrate Task Force, Mars Inc., Waltham Pet Healthcare Research Eur J Clin Nutr. 2018;72:1246–50. http://www.nature.com/a
Institute. Government Committees: Scientific Advisory Committee on rticles/s41430-018-0180-4.
Nutrition-including Working Group with NHS-E and Diabetes UK on 11. Müller MJ, Enderle J, Pourhassan M, Braun W, Eggeling B,
High Fat Diets in Diabetes management. MRC Nutrition Grants panel. Lagerpusch M, et al. Metabolic adaptation to caloric restriction
Editorial duties: Joint Editor of International Journal of Obesity. and subsequent refeeding: the Minnesota Starvation Experiment
revisited. Am J Clin Nutr. 2015;102:807–19. https://academic.
Publisher’s note Springer Nature remains neutral with regard to oup.com/ajcn/article/102/4/807/4564599.
jurisdictional claims in published maps and institutional affiliations. 12. Martin CK, Das SK, Lindblad L, Racette SB, McCrory MA,
Weiss EP, et al. Effect of calorie restriction on the free-living
physical activity levels of nonobese humans: results of three
Open Access This article is licensed under a Creative Commons
randomized trials. J Appl Physiol. 2011;110:956–63. https://www.
Attribution 4.0 International License, which permits use, sharing,
physiology.org/doi/10.1152/japplphysiol.00846.2009.
adaptation, distribution and reproduction in any medium or format, as
13. Dansinger ML, Gleason JA, Griffith JL, Selker HP, Schaefer EJ.
long as you give appropriate credit to the original author(s) and the
Comparison of the Atkins, Ornish, weight watchers, and Zone
source, provide a link to the Creative Commons license, and indicate if
diets for weight loss and heart disease risk reduction. JAMA.
changes were made. The images or other third party material in this
2005;293:43. http://jama.jamanetwork.com/article.aspx?doi=10.
article are included in the article’s Creative Commons license, unless
1001/jama.293.1.43.
indicated otherwise in a credit line to the material. If material is not
14. Middleton KR, Anton SD, Perri MG. Long-term adherence to
included in the article’s Creative Commons license and your intended
health behavior change. Am J Lifestyle Med. 2013;7:395–404.
use is not permitted by statutory regulation or exceeds the permitted
http://journals.sagepub.com/doi/10.1177/1559827613488867.
use, you will need to obtain permission directly from the copyright
15. Varady KA, Hellerstein MK. Alternate-day fasting and chronic
holder. To view a copy of this license, visit http://creativecommons.
disease prevention: a review of human and animal trials. Am J
org/licenses/by/4.0/.
Clin Nutr. 2007;86:7–13. https://academic.oup.com/ajcn/article/
86/1/7/4633143.
16. Harvie MN, Pegington M, Mattson MP, Frystyk J, Dillon B,
References Evans G, et al. The effects of intermittent or continuous energy
restriction on weight loss and metabolic disease risk markers: a
1. Lavie CJ, Milani RV. Ventura HO, Obesity and cardiovascular randomized trial in young overweight women. Int J Obes.
disease. Risk factor, paradox, and impact of weight loss. J Am 2011;35:714–27. http://www.nature.com/articles/ijo2010171.
Coll Cardiol. 2009;53:1925–32. https://linkinghub.elsevier.com/ 17. Mattson MP, Longo VD, Harvie M. Impact of intermittent fasting
retrieve/pii/S0735109709007463. on health and disease processes. Ageing Res Rev. 2017;39:46–58.
2. Kahn SE, Hull RL, Utzschneider KM. Mechanisms linking obesity https://linkinghub.elsevier.com/retrieve/pii/S1568163716302513.
to insulin resistance and type 2 diabetes. Nature. 2006;444:840–6. 18. Carter S, Clifton PM, Keogh JB. The effects of intermittent
http://www.nature.com/articles/nature05482. compared to continuous energy restriction on glycaemic control in
3. Klop B, Elte J, Cabezas M. Dyslipidemia in obesity: mechanisms type 2 diabetes; a pragmatic pilot trial. Diabetes Res Clin Pract.
and potential targets. Nutrients. 2013;5:1218–40. http://www. 2016;122:106–12. https://linkinghub.elsevier.com/retrieve/pii/
mdpi.com/2072-6643/5/4/1218. S0168822716307768.
4. Kivipelto M, Ngandu T, Fratiglioni L, Viitanen M, Kåreholt I, Win- 19. Conley M, Le Fevre L, Haywood C, Proietto J. Is two days of
blad B, et al. Obesity and vascular risk factors at midlife and the risk intermittent energy restriction per week a feasible weight loss
of dementia and Alzheimer disease. Arch Neurol. 2005;62:1556–60. approach in obese males? A randomised pilot study. Nutr Diet.
http://archneur.jamanetwork.com/article.aspx?doi=10.1001/archneur. 2018;75:65–72. http://doi.wiley.com/10.1111/1747-0080.12372.
62.10.1556. 20. Schübel R, Nattenmüller J, Sookthai D, Nonnenmacher T, Graf
5. Bhaskaran K, Douglas I, Forbes H, Dos-Santos-Silva I, Leon ME, Riedl L, et al. Effects of intermittent and continuous calorie
DA, Smeeth L. Body-mass index and risk of 22 specific can- restriction on body weight and metabolism over 50 wk: a rando-
cers: a population-based cohort study of 5·24 million UK mized controlled trial. Am J Clin Nutr. 2018;108:933–45. https://a
adults. Lancet. 2014;384:755–65. https://linkinghub.elsevier. cademic.oup.com/ajcn/article/108/5/933/5201451.
com/retrieve/pii/S0140673614608928. 21. Sundfør TM, Svendsen M, Tonstad S. Effect of intermittent versus
6. Di Angelantonio E, Bhupathiraju SN, Wormser D, Gao P, continuous energy restriction on weight loss, maintenance and
Kaptoge S, de Gonzalez AB, et al. Body-mass index and all-cause cardiometabolic risk: a randomized 1-year trial. Nutr Metab

Content courtesy of Springer Nature, terms of use apply. Rights reserved

Effects of intermittent (5:2) or continuous energy restriction on basal and postprandial metabolism: a. . . 73

Cardiovasc Dis. 2018;28:698–706. https://linkinghub.elsevier. insulin clamp. Diabetes Care. 1999;22:1462–70. http://care.dia
com/retrieve/pii/S0939475318301005. betesjournals.org/cgi/doi/10.2337/diacare.22.9.1462.
22. Carter S, Clifton PM, Keogh JB. Effect of intermittent compared 37. Flint A, Raben A, Blundell J, Astrup A. Reproducibility, power
with continuous energy restricted diet on glycemic control in and validity of visual analogue scales in assessment of appetite
patients with type 2 diabetes. JAMA Netw Open. 2018;1:e180756. sensations in single test meal studies. Int J Obes. 2000;24:38–48.
http://jamanetworkopen.jamanetwork.com/article.aspx?doi=10. http://www.nature.com/articles/0801083.
1001/jamanetworkopen.2018.0756. 38. Stubbs RJ, Hughes DA, Johnstone AM, Rowley E, Reid C, Elia
23. Harvie M, Wright C, Pegington M, McMullan D, Mitchell E, M, et al. The use of visual analogue scales to assess motivation to
Martin B, et al. The effect of intermittent energy and carbohy- eat in human subjects: a review of their reliability and validity
drate restriction v. daily energy restriction on weight loss and with an evaluation of new hand-held computerized systems for
metabolic disease risk markers in overweight women. Br J Nutr. temporal tracking of appetite ratings. Br J Nutr. 2000;84:405–15.
2013;110:1534–47. https://www.cambridge.org/core/product/ 39. Alhussain MH, Macdonald IA, Taylor MA. Irregular meal-pattern
identifier/S0007114513000792/type/journal_article. effects on energy expenditure, metabolism, and appetite regula-
24. Antoni R, Johnston KL, Collins AL, Robertson MD. Intermittent v. tion: a randomized controlled trial in healthy normal-weight
continuous energy restriction: differential effects on postprandial women. Am J Clin Nutr. 2016;104:21–32. https://academic.oup.
glucose and lipid metabolism following matched weight loss in com/ajcn/article/104/1/21/4633920.
overweight/obese participants. Br J Nutr. 2018;119:507–16. 40. Yalow RS, Berson SA. Immunoassay of endogenous plasma
https://www.cambridge.org/core/product/identifier/ insulin in man. Obes Res. 1996;4:583–600. http://doi.wiley.com/
S0007114517003890/type/journal_article. 10.1002/j.1550-8528.1996.tb00274.x.
25. Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, 41. Halberg N, Henriksen M, Söderhamn N, Stallknecht B, Ploug T,
Ainsworth BE, et al. International Physical Activity Ques- Schjerling P, et al. Effect of intermittent fasting and refeeding on
tionnaire: 12-country reliability and validity. Med Sci Sports insulin action in healthy men. J Appl Physiol. 2005;99:2128–36.
Exerc. 2003;35:1381–95. http://journals.lww.com/00005768- http://www.physiology.org/doi/10.1152/japplphysiol.00683.2005.
200308000-00020. 42. Cho Y, Hong N, Kim K, Cho S, Lee M, Lee Y, et al. The effec-
26. Durnin JVGA, Rahaman MM. The assessment of the amount of tiveness of intermittent fasting to reduce body mass index and glu-
fat in the human body from measurements of skinfold thickness. cose metabolism: a systematic review and meta-analysis. J Clin Med.
Br J Nutr. 1967;21:681–9. https://www.cambridge.org/core/ 2019;8:1645. https://www.mdpi.com/2077-0383/8/10/1645.
product/identifier/S0007114567000728/type/journal_article. 43. Varady KA, Bhutani S, Klempel MC, Kroeger CM, Trepanowski
27. Durnin JVGA, Womersley J. Body fat assessed from total body JF, Haus JM, et al. Alternate day fasting for weight loss in normal
density and its estimation from skinfold thickness: measurements weight and overweight subjects: a randomized controlled trial.
on 481 men and women aged from 16 to 72 years. Br J Nutr. Nutr J. 2013;12:146. http://nutritionj.biomedcentral.com/articles/
1974;32:77–97. http://www.journals.cambridge.org/abstract_ 10.1186/1475-2891-12-146.
S0007114574000614. 44. Klempel MC, Bhutani S, Fitzgibbon M, Freels S, Varady KA.
28. Beck AT. An inventory for measuring depression. Arch Gen Dietary and physical activity adaptations to alternate day modified
Psychiatry. 1961;4:561. http://archpsyc.jamanetwork.com/article. fasting: implications for optimal weight loss. Nutr J. 2010;9:35.
aspx?doi=10.1001/archpsyc.1961.01710120031004. http://nutritionj.biomedcentral.com/articles/10.1186/1475-2891-9-35.
29. Garner DM, Olmsted MP, Bohr Y, Garfinkel PE. The Eating 45. Blundell J, de Graaf C, Hulshof T, Jebb S, Livingstone B, Lluch
Attitudes Test: psychometric features and clinical correlates. A, et al. Appetite control: methodological aspects of the evalua-
Psychol Med. 1982;12:871–8. https://www.cambridge.org/core/ tion of foods. Obes Rev. 2010;11:251–70. http://doi.wiley.com/
product/identifier/S0033291700049163/type/journal_article. 10.1111/j.1467-789X.2010.00714.x.
30. Henry C. Basal metabolic rate studies in humans: measurement 46. Jones R, Pabla P, Mallinson J, Nixon A, Taylor T, Bennett A,
and development of new equations. Public Health Nutr. et al. Two weeks of early time-restricted feeding (eTRF) improves
2005;8:1133–52. https://www.cambridge.org/core/product/ skeletal muscle insulin and anabolic sensitivity in healthy men.
identifier/S1368980005001394/type/journal_article. Am J Clin Nutr. 2020:1–14. https://academic.oup.com/ajcn/adva
31. Scientific Advisory Committee on Nutrition (SACN). Dietary nce-article/doi/10.1093/ajcn/nqaa192/5878409.
reference values for energy. London: TSO; 2011. 47. Krug S, Kastenmüller G, Stückler F, Rist MJ, Skurk T, Sailer M,
32. Department of Health. Dietary reference values for food energy et al. The dynamic range of the human metabolome revealed by
and nutrients for the United Kingdom. London: HMSO; 1991. challenges. FASEB J. 2012;26:2607–19. https://onlinelibrary.
33. Weir JBdV. New methods for calculating metabolic rate with wiley.com/doi/abs/10.1096/fj.11-198093.
special reference to protein metabolism. J Physiol. 1949;109:1–9. 48. Subar AF, Freedman LS, Tooze JA, Kirkpatrick SI, Boushey C,
http://doi.wiley.com/10.1113/jphysiol.1949.sp004363. Neuhouser ML, et al. Addressing current criticism regarding the
34. McGuire EAH, Helderman JH, Tobin JD, Andres R, Berman M. value of self-report dietary data. J Nutr. 2015;145:2639–45.
Effects of arterial versus venous sampling on analysis of glucose https://academic.oup.com/jn/article/145/12/2639/4585705.
kinetics in man. J Appl Physiol. 1976;41:565–73. https://www. 49. Rynders CA, Thomas EA, Zaman A, Pan Z, Catenacci VA,
physiology.org/doi/10.1152/jappl.1976.41.4.565. Melanson EL. Effectiveness of intermittent fasting and time-
35. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher restricted feeding compared to continuous energy restriction for
DF, Turner RC. Homeostasis model assessment: insulin resistance weight loss. Nutrients. 2019;11:2442. https://www.mdpi.com/
and beta-cell function from fasting plasma glucose and insulin 2072-6643/11/10/2442.
concentrations in man. Diabetologia. 1985;28:412–9. http://link. 50. Robertson MD, Henderson RA, Vist GE, Rumsey RDE. Extended
springer.com/10.1007/BF00280883. effects of evening meal carbohydrate-to-fat ratio on fasting and
36. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from postprandial substrate metabolism. Am J Clin Nutr. 2002;75:505–10.
oral glucose tolerance testing: comparison with the euglycemic https://academic.oup.com/ajcn/article/75/3/505/4689343.

Content courtesy of Springer Nature, terms of use apply. Rights reserved

Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

onlineservice@springernature.com