Intermittent fasting vs daily calorie restriction for type 2 diabetes prevention: a review of human findings

REVIEW ARTICLE Intermittent fasting vs daily calorie restriction for type 2 diabetes prevention: a review of human findings ADRIENNE R. BARNOSKY, KRISTIN K. HODDY, TERRY G. UNTERMAN, and KRISTA A. VARADY CHICAGO, ILL

Intermittent fasting (IF) regimens have gained considerable popularity in recent years, as some people find these diets easier to follow than traditional calorie restriction (CR) approaches. IF involves restricting energy intake on 1–3 d/wk, and eating freely on the nonrestriction days. Alternate day fasting (ADF) is a subclass of IF, which consists of a ‘‘fast day’’ (75% energy restriction) alternating with a ‘‘feed day’’ (ad libitum food consumption). Recent findings suggest that IF and ADF are equally as effective as CR for weight loss and cardioprotection. What remains unclear, however, is whether IF/ADF elicits comparable improvements in diabetes risk indicators, when compared with CR. Accordingly, the goal of this review was to compare the effects of IF and ADF with daily CR on body weight, fasting glucose, fasting insulin, and insulin sensitivity in overweight and obese adults. Results reveal superior decreases in body weight by CR vs IF/ADF regimens, yet comparable reductions in visceral fat mass, fasting insulin, and insulin resistance. None of the interventions produced clinically meaningful reductions in glucose concentrations. Taken together, these preliminary findings show promise for the use of IF and ADF as alternatives to CR for weight loss and type 2 diabetes risk reduction in overweight and obese populations, but more research is required before solid conclusions can be reached. (Translational Research 2014;-:1–10) Abbreviations: ADF ¼ Alternate day fasting; BMI ¼ Body mass index; CR ¼ Calorie restriction; HOMA-IR ¼ Homeostatic model assessment-insulin resistance; IF ¼ Intermittent fasting

INTRODUCTION

From the Division of Endocrinology, Department of Medicine, University of Illinois at Chicago, Chicago, Ill; Department of Kinesiology and Nutrition, University of Illinois at Chicago, Chicago, Ill. Submitted for publication January 15, 2014; revision submitted May 5, 2014; accepted for publication May 8, 2014. Reprint requests: Krista A. Varady, Department of Kinesiology and Nutrition, University of Illinois at Chicago, 1919 West Taylor Street, Room 506F, Chicago, IL 60612; e-mail: [email protected]. 1931-5244/$ - see front matter Ó 2014 Mosby, Inc. All rights reserved. http://dx.doi.org/10.1016/j.trsl.2014.05.013

A

t present, 35% of adults older than 20 years in the United States have prediabetes.1 If no lifestyle changes are made to improve health, 15%–30% of these individuals will develop type 2 diabetes within 5 years.1 A key strategy to prevent the progression of prediabetes to type 2 diabetes is weight loss.2 Accumulating evidence suggests that even modest weight loss (5%–7% of initial weight) helps to improve several diabetes risk parameters, including fasting glucose, insulin, and insulin sensitivity.3,4 Daily calorie restriction (CR) regimens are still the most common diet strategies implemented for weight loss.5 CR regimens involve reducing energy intake 1

2

Barnosky et al

every day by 20%–50% of needs.5 Although CR is effective for weight loss in some individuals, many people find this type of dieting difficult, as it requires vigilant calorie counting on a daily basis.6 People also grow frustrated with this diet, as they are never able to eat freely throughout the day. In light of these issues with CR, another approach termed intermittent fasting (IF) has shown promise in achieving weight loss goals.7 IF differs from CR, in that it only requires an individual to restrict energy 1–3 d/wk, and allows for ad libitum food consumption on the nonrestriction days.7 Alternate day fasting (ADF) is a subclass of IF, which consists of a ‘‘fast day’’ (75% energy restriction) alternating with a ‘‘feed day’’ (ad libitum food consumption). Recent reviews suggest that IF and ADF are equally as effective as CR for weight loss cardioprotection.7,8 What has yet to be elucidated, however, is whether IF and ADF elicit comparable improvements in diabetes risk indicators, when compared with CR. Accordingly, the goal of this review was to compare the effects of IF and ADF with daily CR on body weight, fasting glucose, fasting insulin, and insulin sensitivity in overweight and obese adults. METHODS

We performed a systematic search in MEDLINE PubMed using the following search strings: (1) ‘‘intermittent fasting and weight loss,’’ (2) ‘‘alternate day fasting and weight’’ or ‘‘alternate day calorie restriction,’’ (3) ‘‘calorie restriction and weight loss and insulin,’’ (4) ‘‘caloric restriction and weight loss and obesity,’’ and (5) ‘‘calorie restriction and metabolic syndrome.’’ Two reviewers (A.B. and K.H.) separately screened the abstracts for inclusion and exclusion. Full text articles were retrieved from all abstracts that were potentially relevant and were reviewed independently by the 2 researchers. The comprehensive literature search revealed 108 articles under the umbrella category of IF and 4945 articles in the category of CR. Articles that were excluded if they did not meet the inclusion criteria, were review articles, editorials, letters, comments, or conferences proceedings. References of the retrieved articles were also screened for additional studies. Inclusion criteria were as follows: (1) randomized control trials and nonrandomized trials, (2) total sample size $8 subjects, (3) primary endpoints of body weight and one or more relevant diabetes risk parameter, (4) average daily energy restriction ,50% (to exclude very low calorie diets that result in muscle wasting9), (5) trial duration between 3 and 24 weeks, (6) male and female subjects, (7) age between 25 and 75 years, (8) body mass index (BMI) between 25 and 40 kg/m2, (9) nonsmokers (because of the effects of

Translational Research - 2014

smoking on lipid metabolism),10 (10) sedentary or moderately active individuals, and (11) articles published after 2003. We chose 2003 as a cutoff date because all the IF studies found were published within this time frame, and we wanted to use the same time frame for CR studies. Exclusion criteria were as follows: (1) cohort and observational studies; (2) trials that combined CR/IF with supplements, pharmacologic substances, or exercise; (3) diabetic; and (4) very active individuals or athletes. Ten CR trials and 9 IF trials were found that matched these criteria. None of the papers retrieved implemented intention to treat analyses. BODY WEIGHT AND VISCERAL FAT MASS

Obesity is a well-established risk factor for the development of type 2 diabetes. Findings from the Nurses’ Health Study demonstrate a 100-fold increase in diabetes risk over 14 years in those with a BMI .35 kg/m2 compared with normal weight individuals.11 At least one contributing factor to insulin resistance that occurs in obesity is the decrease in insulin-mediated peripheral glucose uptake.12 Weight loss results in substantial reductions in insulin resistance, with every 1 kg lost associated with a 16% reduction in estimated risk of developing diabetes.2 The distribution of excess fat mass also contributes to the risk for metabolic derangements.13 In 1947, the concept of regional fat distribution having different physiological and metabolic effects was first introduced by Vague.14 Over the subsequent decades, it has been shown that visceral obesity has a stronger correlation with a risk for the development of diabetes, hypertension, hyperlipidemia, hepatic steatosis, and coronary artery disease compared with that of a gluteoemoral fat distribution.13 The presence of visceral obesity has also been shown to have a strong inverse relationship with insulin sensitivity.13 Evaluation of glucose disposal rates by euglycemic insulin clamps and visceral adipose tissue by the computed tomography technique, illustrated an inverse association.15 Thus a higher visceral fat content is correlated with lower insulin sensitivity.15 Weight loss has been shown to decrease both visceral fat and improve markers of insulin sensitivity.16 IF: effects on body weight and visceral fat mass. Body weight changes were assessed in 2 IF studies17,18 and 7 ADF studies19-25 (Table I). Findings from these trials demonstrate 3%–8% reductions in body weight after 3–24 weeks of treatment. Providing food to subjects on the fast day appears to be a key factor in determining greatest weight loss. For instance, the most pronounced weight loss was seen in a study performed by Johnson et al,21 where ADF subjects were provided with a 320–380 kcal meal replacement shake on each fast day.

Table I. Intermittent fasting: effect on body weight and type 2 diabetes risk parameters

Intermittent fasting studies Klempel et al17

Harvie et al18

Subjects

Trial length

Intervention

n 5 54, F 48 6 2 y Obese Prediabetic

8 wk 1.1 d/wk 100% CR, 6 d/wk 30% CR-liquid diet 2.1 d/wk 100% CR, 6 d/wk 30% CR-food diet Food provided n 5 53, F 24 wk 1.2 d/wk 75% CR, 5 d/wk ad libitum 30–45 y Overweight Food not provided Obese

1.Y4%* 2.Y3%*

1.Y4* 2.Y3*

WC 1.Y7%* 2.Y5%*

—

DXA 1.Y1 2.0

1.Y3%* 2.Y2%

1.Y21%* 2.Y13%

HOMA-IR 1.Y23%* 2.Y12%

1.20%

1.Y7%*

1.Y6*

WC 1.Y6%*

BIA 1.Y3%*

BIA 1.Y1*

1.Y2%

1.Y29%*

HOMA-IR 1.Y27%*

1.50%

1.Y3%*

—

—

DXA 1.Y1%*

DXA 1.Y1*

1.Y1%

1.Y57%*

—

1.35%

1.Y7%*

1.Y3

WC 1.Y6%*

DXA 1.Y1%

DXA 1.Y2

1.Y6%

—

—

1.40%

1.Y8%*

1.Y9*

—

—

—

1.[6%

1.Y37%*

HOMA-IR 1.Y33%*

1.35%

1.Y6%*

1.Y6*

WC 1.Y4%*

BIA 1.0%

BIA 1.0

1.Y4%*

1.Y20%*

HOMA-IR 1.Y19%*

1.35% 2.35%

1.Y5%* 2.Y4%*

1.Y4* 2.Y4*

WC 1.Y7%*

DXA 1.[1% 2.[1%

DXA 1.[1 2.[1

1.Y2%

—

—

1.35% 2.0%

1.Y4%† 2.0%

1.Y3† 2.0

WC 1.Y4%† 2.Y1%

BIA 1.Y1 2.0%

BIA 1.Y2% 2.0%

1.Y3% 2.[3%

1.Y11%† 2.[1%

HOMA-IR 1.Y9%† 2.[2%

Barnosky et al 3

Alternate day fasting studies Heilbronn et al19 n 5 16, MF 3 wk 1. Fast day: 100% 23–53 y CR, feed day: Overweight ad libitum Food not provided Eshghinia and n 5 15, F 6 wk 1. Fast day: 70% CR, Mohammadzadeh20 34 6 6 y feed day: ad libitum Obese Food not provided Johnson et al21 n 5 10, MF 8 wk 1. Fast day: 80% CR, Age NR feed day: ad libitum Food provided on the Obese fast day Varady et al22 n 5 16, MF 8 wk 1. Fast day: 75% CR, 46 6 2 y feed day: ad libitum Obese Food provided on the Prediabetic fast day Klempel et al23 n 5 32, F 8 wk 1. Fast day: 75% CR, 42 6 2 y feed day: ad Obese libitum-HF 2. Fast day: 75% CR, feed day: ad libitum-LF Food provided on the fast day Bhutani et al24 n 5 32, MF 12 wk 1. Fast day: 75% CR, 43 6 3 y feed day: ad libitum Obese 2. Control: ad libitum fed every day Food provided on the fast day (week 1–4 only)

1.40% 2.40%

Translational Research Volume -, Number -

Reference

Insulin resistance/ Average Fasting Fasting sensitivity prescribed Body weight Body Visceral fat Lean mass Lean glucose insulin restriction/d (% change) weight (kg) (% change) (% change) mass (kg) (% change) (% change) (% change)

(Continued )

Abbreviations: BIA, bioelectrical impedance analysis; CR, calorie restriction; DXA, dual-energy X-ray absorptiometry; F, female; HF, high-fat diet; HOMA-IR, homeostatic model of assessment for insulin resistance; LF, low-fat diet; M, male; NR, not reported; WC, waist circumference. All data reported are for subjects who completed the entire trial. *Post-treatment value significantly different from baseline (P , 0.05). † Significantly different from the control group (P , 0.05). Prescribed daily restriction estimated assuming 0% restriction (ie, 100% intake) on the ad libitum feed days.

HOMA-IR 1.Y28%† 2.[2% 1.Y31%† 2.[2% 1.Y6%† 2.[1% DXA 1.Y2 2.0% DXA 1.Y3% 2.Y1% WC 1.Y6%† 2.Y1% 1. Y5† 2.Y1 1.Y7%† 2.Y1% 1.35% 2.0% n 5 32, MF 12 wk 1. Fast day: 75% CR, 47 6 4 y feed day: ad libitum Normal wt 2. Control: ad libitum Overweight fed every day Prediabetic Food provided on the fast day Varady et al25

Intervention Trial length Subjects Reference

Table I. (Continued )

Translational Research - 2014

Barnosky et al

Insulin Average Fasting Fasting resistance/ prescribed Body weight Body Visceral fat Lean mass Lean glucose insulin sensitivity restriction/d (% change) weight (kg) (% change) (% change) mass (kg) (% change) (% change) (% change)

4

After 8 weeks of treatment, subjects lost 8% of body weight.21 Comparable decreases in body weight (6%–7%) were also noted in the other 8-week ADF studies that provided food on the fast day.22,23,25 An exception to this rule is the ADF study by Bhutani et al.24 In this 12-week trial, fast day food was provided, but only a 4% weight loss was observed.24 This limited weight loss may be explained by the fact that food was only provided for the first 4 weeks of the study,24 and not for the entire duration of trial. Another factor that appears to impact degree of weight loss is the number of fast days per week. Not surprisingly, a faster rate of weight loss was observed in the ADF studies,19-25 which required subjects to fast 3–4 d/wk, compared with the IF studies,17,18 which required subjects to only fast 1–2 d/wk. On average, ADF appears to produce a 0.75 kg weekly reduction in body weight, whereas IF produces a 0.25 kg weekly weight loss. As such, clinicians may want to recommend ADF to their patients who are eager to lose weight more rapidly, and IF to patients who would prefer to lose weight at a slower pace. Visceral fat changes were assessed in 2 IF studies17,18 and 5 ADF studies.20,22-25 Results reveal 4%–7% reductions in visceral fat after 6–24 weeks of treatment. In most studies, the percentage of visceral fat loss closely paralleled the percentage of weight loss. For example, in the ADF study by Varady et al,22 a 6% decrease in visceral fat was observed corresponding to 7% weight loss. Bhutani et al24 had a similar study design, however, only a 4% decrease in visceral fat was observed corresponding to a 4% weight loss. Similar reductions in visceral fat were also demonstrated in IF studies (3%–7%),17,18 suggesting that IF and ADF produce comparable decreases in this body composition parameter. It should be noted, however, that visceral fat was assessed indirectly in each of these trials by measuring waist circumference. Thus, these studies are limited, in that actual kilogram decreases in visceral fat were not determined. Future studies of IF and ADF should therefore strive to use more robust techniques, such as magnetic resonance imaging, to measure actual kilogram changes in visceral fat mass. CR: effects on body weight and visceral fat mass. Body weight changes were assessed in 10 CR trials16,18,26-33 (Table II). Findings from these studies demonstrate a 4%–14% reduction in weight after 6–24 weeks of treatment. The greatest weight loss was observed in the trials with the largest average weekly caloric restriction.16,27,33 In a study by Larsen-Meyer et al,33 overweight participants were randomized into 1 of 3 groups: (1) 50% CR every day, (2) 25% CR every day, or (3) a control group with ad libitum feeding every day. After 24 weeks of treatment, participants in the

Reference

Xydakis et al16

Trussardi Fayh et al26

Svendsen et al27

Mollard et al28

Clifton et al29

Subjects

n 5 80, MF 47 6 1 y Obese n 5 35, MF 30 6 6 y Obese Prediabetic n 5 17, F 25 6 3 y Overweight n 5 40, MF 46 6 1 y Obese n 5 62, F 47 6 10 y Obese Prediabetic

Trial length

6 wk

6 wk

Intervention

1.50% CR-highprotein diet Food provided 1.25% CR Food not provided

8 wk

1.50% CR-highprotein diet Food not provided 8 wk 1.25% CR-highfiber diet Food provided 12 wk 1.25% CR-LF diet 2.25% CR-high MUFA lowprotein diet Food provided 12 wk 1.25% CR-liquid diet 2.25% CR-food diet Food provided in the liquid diet group 12 wk 1.25% CR every day Food not provided

Agueda et al31

n 5 78, F 37 6 7 y Obese n 5 157, MF 12 wk 1.30% CR-low 38 6 7 y energy density diet Obese 2.30% CR-low Prediabetic glycemic index diet 3.30% CR-portion control diet Food not provided n 5 48, MF 24 wk 1.25% CR every day 25–50 y 2.50% CR every day until 15% body Overweight weight lost 3. Control: ad libitum fed every day All food provided for weeks 1–12 and 22–24

Melanson et al32

Larson-Meyer et al33

1.50%

1.Y7%*

1.Y18*

WC 1.Y8%*

—

—

1.Y15%*

1.Y65%*

HOMA-IR 1.Y70%*

1.25%

1.Y5%*

1.Y4*

WC 1.Y4%*

—

—

1.Y3%

1.Y11%*

HOMA-IR 1.Y20%*

1.50%

1.Y11%*

1.Y8%*

1.Y35%*

HOMA-IR 1.Y32%*

1.25%

1.Y1%

1.Y1

WC 1.Y2%*

—

—

1.0%

1.[2%

HOMA-IR 1.[1%

1.25% 2.25%

1.Y11%* 2.Y8%*

1.Y10* 2.Y9*

—

—

—

1.Y1%* 2.Y2%*

1.Y26%* 2.Y41%*

—

1.25% 2.25%

1.Y8%* 2.Y5%*

1.Y8* 2.Y5*

WC 1.Y5%* 2.Y4%*

BIA 1.Y1%* 2.Y3%*

BIA 1.Y1* 2.Y1*

1.0% 2.[1%

1.Y24%* 2.Y25%

HOMA-IR 1.Y25%* 2.Y12%

1.25%

1.Y9%*

Y 8*

WC 1.Y7%*

DXA 1.Y5%*

DXA 1.Y3*

1.Y2%*

1.Y27%*

HOMA-IR 1.Y30%*

1.30% 2.30% 3.30%

1.Y5%* 2.Y4%* 3.Y4%*

1.Y4* 2.Y3* 3.Y4*

—

— — —

1.[1 2.Y1 3.0

1.[1% 2.[1% 3.Y1%

1.Y16%* 2.Y8%* 3.Y12%*

HOMA-IR 1.Y17%* 2.Y7%* 3.Y17%*

1.25% 2.50% 3. No CR

1.Y10%† 2.Y14%† 3.0%

1.Y8† 2.Y11† 3.0

CT 1.Y28%† 2.Y38%† 3.Y3%

DXA 1.Y5%† 2.6%† 3.0%

DXA 1.Y3† 2.Y3† 3.0

1.Y1% 2.[1% 3.[2%

1.Y29%† 2.Y15%† 3.[2%

OGTT-IS 1.[27%* 2.[52%* 3.[11%

WC 1.Y22%*

5

n 5 40, MF 65 6 8 y Obese

Visceral fat (% change)

Insulin Fasting Fasting resistance/ Lean mass Lean glucose insulin sensitivity (% change) mass (kg) (% change) (% change) (% change)

Barnosky et al

De Luis et al30

Average prescribed Body weight Body restriction/d (% change) weight (kg)

Translational Research Volume -, Number -

Table II. Daily CR: effect on body weight and type 2 diabetes risk parameters

(Continued )

Abbreviations: BIA, bioelectrical impedance analysis; CR, calorie restriction; DXA, dual-energy x-ray absorptiometry; F, female; HOMA-IR, homeostasis model of assessment for insulin resistance; LF, low-fat diet; M, male; MUFA, monounsaturated fatty acid; OGTT-IS, oral glucose tolerance test for insulin sensitivity; WC, waist circumference. All data reported are for subjects who completed the entire trial. *Post-treatment value significantly different from baseline (P , 0.05). † Significantly different from control group (P , 0.05).

HOMA-IR 1.Y19%* 1.Y15%* 1.Y2% 1.Y1* 1.Y2%* Waist circumference 1.Y4%* 1.Y6* 1.Y5%* 1.25% n 5 54, F 30–45 y Overweight Obese Harvie et al18

24 wk 1.25% CR every day No food provided

Visceral fat (% change) Subjects Reference

Trial length

Intervention

Average prescribed Body weight Body restriction/d (% change) weight (kg)

Table II. (Continued )

Translational Research - 2014

Barnosky et al

Insulin Fasting Fasting resistance/ Lean mass Lean glucose insulin sensitivity (% change) mass (kg) (% change) (% change) (% change)

6

50% CR group had greater weight loss (14%), when compared with the 25% CR group (10% weight loss).33 A faster rate of weight loss was also noted in the other studies which implemented 50% CR16,27 when compared with the 25% CR trials.18,26,28-32 The background macronutrient composition of the CR diets, however, did not seem to have any impact on rate of weight loss. This is evidenced in the trial by Melanson et al,32 which compared the effects of a 30% CR-low energy density diet with that of a 30% CRlow glycemic index diet. After 12 weeks of treatment, both groups lost similar amounts of weight (4%–5%) despite differences in the background macronutrient profiles.32 It was also noted that the rate of weight loss by CR is similar in overweight and obese individuals, and males and females, when the same degree of CR is applied. Similar amounts of weight loss were also noted in older30 vs younger adult subjects.29 For instance in the study by De Luis et al,30 elderly men and women experienced comparable weight loss after 12 weeks of 25% CR, when compared with middleaged adults undergoing a similar intervention.29 Thus, CR appears to be effective for weight loss independent of BMI class, sex, and age. Visceral fat changes were assessed in 8 CR studies.16,18,26-31,33 After 6–24 weeks of diet, 2%–38% reductions in visceral fat mass were observed. Similar to what was seen in IF studies, percentage of visceral fat loss generally paralleled the percentage of weight loss. For example, the greatest decrease in visceral fat (38%) was observed in the study by Larson-Meyer et al,33 which implemented a 50% CR protocol for 24 weeks. This degree of visceral fat loss corresponded to a 14% weight loss.33 Within the same study, a 28% reduction in visceral fat was seen in the 25% CR group with a 10% weight loss.33 Moreover, Svendsen et al27 showed a 22% decrease in visceral fat, corresponding to an 11% reduction in body weight. Taken together, it would appear as although greater degrees of energy restriction produce the most optimal changes in body weight and visceral fat mass. GLUCOSE AND INSULIN

Individuals are categorized as having ‘‘prediabetes’’ when (1) fasting glucose falls between 100 and 125 mg/dL, (2) plasma glucose falls between 140 and 199 mg/dL 2 after an oral glucose tolerance test, or (3) hemoglobin A1c falls between 5.7% and 6.4%.34 Lifestyle modification, namely dietary changes and exercise with the goal of weight loss, are commonly used as the first line therapy. Randomized, controlled trials have shown that with intensive dietary counseling and increased physical activity, type 2 diabetes can be

Translational Research Volume -, Number -

prevented in high-risk individuals with prediabetes.35,36 As expected, these studies have shown decreases in fasting glucose levels and improvements in glucose tolerance. The Finnish Diabetes Prevention Study found a 58% reduction in the overall incidence of diabetes in the intensive lifestyle group compared with that of controls with similar benefit seen in other studies.35-37 IF: effects on fasting glucose and insulin levels. Changes in fasting glucose were assessed in 2 IF studies17,18 and 7 ADF studies19-25 (Table I). However, the present discussion will be limited to studies that recruited prediabetic individuals,17,22,25 as it may not be scientifically valid to assess the effect of these diets on glucose and insulin levels of normoglycemic subjects. Results from these trials demonstrate consistent yet minor decreases (3%–6%) in fasting glucose after 8–12 weeks of treatment. The greatest decrease in glucose was observed in the study by Varady et al.25 In this trial,25 participants were randomized to an ADF group (35% average daily CR) vs a control group (no restriction). After 12 weeks of diet, a 6% reduction in glucose concentrations was observed in the ADF group relative to controls.25 Reductions in glucose concentrations (3%–4%) were also observed in the 8-week ADF study by Varady et al22 and the 8-week IF study by Klempel et al.17 These less pronounced decreases in glucose are most likely because of the shorter intervention period imposed by these 2 studies, that is 8 weeks17,22 vs the 12-week intervention implemented by Varady et al.25 Fasting insulin levels were assessed in 2 IF studies17,18 and 5 ADF studies.19,21,22,24,25 However, as mentioned previously, only studies that recruited prediabetic individuals will be discussed here.17,22,25 In these trials, decreases in fasting insulin ranged from 20% to 31% after 8–12 weeks of treatment.17,22,25 Similar reductions in fasting insulin were seen with ADF (20%)22 and IF (21%)17 after 8 weeks of diet. Remarkably, these comparable decreases occurred despite the greater number of fasting days implemented by the ADF study (3–4 fast days)22 vs the IF study (1 fast day).17 However, it should be noted that, overall, both interventions prescribed the same level of energy restriction (35%–40%).17,22 Thus, degree of restriction may be a stronger predictor of insulin lowering when compared with the number of fasting days. The greatest decreases in insulin concentrations (31%) were noted in the study by Varady et al.25 These superior reductions are most likely the result of the longer intervention period (12 weeks) used by this trial.25 CR: effects on fasting glucose and insulin levels. Fasting glucose levels were assessed in all the CR studies reviewed here16,18,26-33 (Table II). Similar to IF and

Barnosky et al

7

ADF, we will limit our discussion for glucose to those trials that recruited prediabetic subjects.26,29,32 Results from these studies indicate that 6–12 weeks of 25%–30% CR has virtually no effect on fasting glucose concentrations.26,29,32 Moreover, modulating the background nutrient composition of the diet (via a high monounsaturated fat, low fat, or glycemic index diet) also does not appear to impact fasting glucose levels.29,32 It will be of interest in future trials to examine whether improvements in glucose can be seen with greater energy restriction or longer treatment durations in this population group. Fasting insulin levels were assessed in 10 CR studies.16,18,26-33 However, our discussion will be limited to those trials that recruited prediabetic subjects only.26,29,32 Findings from these trials demonstrate fairly consistent reductions in insulin levels, ranging from 11% to 41% after 6–12 weeks of treatment.26,29,32 The most pronounced reductions in insulin (41%) were noted in the study by Clifton et al,29 which implemented a 25% CR-high monounsaturated fat-low-protein diet intervention for 12 weeks. In a separate arm of this study,29 less pronounced reductions in insulin (26%) were observed in response to a 25% CR-low-fat diet. Thus, a background diet that is high monounsaturated fat may produce more optimal changes in insulin compared with a diet that is low in fat. The degree to which insulin is lowered may also be related to amount of weight loss. For instance, the study by Clifton et al29 observed the greatest degree of weight loss (8%–11%) when compared with the studies by Trussardi Fayh et al26 (5%) and Melanson et al32 (4%–5%). Thus, there may be a dose-response relationship between weight loss and insulin lowering by CR in prediabetic individuals. INSULIN SENSITIVITY

Insulin resistance is seen in virtually all patients with type 2 diabetes and occurs early in the disease, before overt diabetes is diagnosed. Both a decrease in insulin sensitivity and insulin deficiency are thought to contribute to type 2 diabetes. Interventions directed at reducing body weight have shown promise for improving insulin sensitivity, and have also been shown to delay or prevent onset of type 2 diabetes.38 IF: effects on insulin sensitivity. Changes in insulin sensitivity were assessed in 2 IF studies17,18 and 4 ADF studies21,22,24,25 (Table I). Results from these trials demonstrate consistent improvements in insulin sensitivity after 3–24 weeks of treatment in normoglycemic and prediabetic subjects. The primary method of measuring insulin sensitivity was homeostatic model assessment (HOMA-IR). In reviewing these results, it would appear as although the greatest

8

Barnosky et al

improvements in insulin sensitivity occurred with the highest amount of weight loss. For instance, in the ADF study by Johnson et al,21 subjects experienced the greatest degree of weight loss (8%), which corresponded to the largest decline in HOMA-IR (33%). In the ADF trial by Bhutani et al,24 body weight decreased by 4% from baseline, which produced moderate reductions in HOMA-IR (9%). Impressive reductions in insulin resistance were also noted by IF. For instance, 23% decreases in HOMA-IR were noted after 8 weeks of fasting 1 d/wk with a liquid diet.17 Moreover, 27% decreases in HOMA-IR were observed after 24 weeks of fasting 2 d/wk.18 Weight loss in these 2 IF trials was 4%17 and 7%,18 respectively. Although reductions in body weight appear to be an important factor in determining improvements in insulin sensitivity, no relationship between visceral fat mass and insulin sensitivity could be established from the studies reviewed here. This is most likely because of the small number of studies, and because visceral fat mass was only measured indirectly via waist circumference. Implementing techniques that can quantify actual kilogram decreases in visceral fat mass (ie, magnetic resonance imaging), may help clarify the relationship between abdominal fat reductions and improvements in insulin sensitivity in future IF and ADF trials. CR: effects on insulin sensitivity. Modulations in insulin sensitivity by CR were assessed in 9 studies16,18,26-28,30-33 (Table II). Substantial improvements in insulin sensitivity after 6–24 weeks were noted in all but one CR study.28 The primary method for measuring insulin sensitivity was HOMA-IR, with the exception of the study by Larson-Meyer et al,33 which implemented an oral glucose tolerance test. The most important factor in determining improvement in insulin sensitivity appears to be the degree of energy restriction imposed. For instance, in the study by Xydakis et al,16 a 6-week 50% CR-high-protein diet resulted in a 70% decrease in HOMA-IR. Similarly, Svendsen et al27 demonstrated a 32% decrease in insulin resistance after 8 weeks of a high restriction regimen (50% CR combined with a high-protein background diet). Findings from 25% CR studies also demonstrated reductions in insulin resistance,18,26,30,31 although these effects were less pronounced than the 50% CR studies. For example, Harvie et al18 implemented a 25% CR protocol in overweight and obese subjects for 24 weeks, and a 15% decrease in HOMA-IR was observed. Likewise, Trussardi Fayh et al26 prescribed a 25% CR diet, and HOMA-IR decreased by 20% from baseline. Although the higher energy restriction diets (50% CR) seem to show a greater decrease in HOMA-IR, sustainability of this type of diet is likely to be challenging. This should be

Translational Research - 2014

considered when prescribing CR protocols to patients in a clinical setting. Unfortunately, none of these studies included a follow-up period to investigate what happens when the individual ceases the diet altogether. Whether insulin resistance rebounds more quickly after stopping a 50% CR protocol, vs a 25% protocol, warrants further investigation, as this may help in determining which level of energy restriction should be prescribed for longer-term success. SUMMARY OF FINDINGS: EFFECTIVENESS OF IF VS CR FOR TYPE 2 DIABETES PREVENTION Body weight and visceral fat mass. Findings from the present review indicate that CR produces slightly superior weight loss when compared with IF/ADF after similar durations of treatment. For instance, after 3–24 weeks of IF or ADF, 3%–8% reductions in body weight were observed. As for CR, 6–24 weeks of diet produced reductions ranging from 4% to14%. Not surprisingly, greater degrees of energy restriction and longer treatment durations produced larger reductions in body weight. IF, ADF, and CR appear to be effective for reducing body weight in men and women, older and younger adults, and prediabetic individuals. Similar decreases in visceral fat mass were also noted by all 3 interventions, and the degree to which visceral fat mass was reduced paralleled the degree of weight loss. Glucose and insulin. The impact of IF, ADF, and CR on fasting glucose concentrations in prediabetic subjects was variable. Although IF and ADF studies demonstrated minor decreases in glucose (3%–6% from baseline), CR studies general report no effect after 6–12 weeks of diet. Fasting insulin, on the other hand, was highly responsive to all 3 interventions. In general, insulin concentrations were reduced by 20%–31% after 8–12 weeks of IF and ADF, and by 11%–41% after 6–12 weeks of CR. Reductions in insulin concentrations by IF, ADF, and CR appeared to be most strongly related to the degree the of imposed restriction and amount of weight loss. Insulin sensitivity. Consistent improvements in insulin sensitivity were noted by all 3 interventions after 3–24 weeks of treatment. These improvements occurred in prediabetic subjects and subjects with normal fasting glucose values. The degree to which insulin sensitivity was improved appeared to be most strongly related to the degree of energy restriction and amount of total weight loss. This observation is supported by other studies in this field.39,40 For instance, in a study by Wing et al,39 subjects were randomized to either a 400 kcal/d group or a 1000 kcal/d group, with the goal of losing 11% of baseline body weight in both groups. Results reveal that those individuals in the 400 kcal/d

Translational Research Volume -, Number -

group had lower fasting glucose and increased insulin sensitivity when compared with the 1000 kcal/d group, despite the same weight loss.39 Moreover, in an early study by Henry et al,40 improvements in glycemic control were noted within 3 days of starting a hypocaloric diet, suggesting that dietary restriction can affect glycemia even before actual weight loss occurs. Limitations. This review has a number of limitations. Firstly, the protocols, interventions, and populations between studies are quite heterogeneous. This heterogeneity makes it difficult to draw clear conclusions from the data as a whole, and needs to be taken into consideration when interpreting the present findings. Secondly, the number of studies that have been conducted in the IF and ADF field are very limited. Thus, it is not possible to make clinical recommendations as to the efficacy of these dietary restriction protocols for use by the general public. Thirdly, we were not able to report weight loss as change in mass per week, because most of these studies only reported baseline and posttreatment body weight values. Including these data would have offered an indication of trends in weight loss, that is whether weight loss was linear or greater at the beginning of the trial, which would have been of value. CONCLUSIONS

In sum, IF, ADF, and CR regimens appear to be effective for reducing body weight, although CR may result in slightly greater weight loss. As for visceral fat mass, and fasting insulin and insulin sensitivity, the effect of IF, ADF, and CR on these diabetic risk parameters appears comparable. Whether these regimens are effective for glucose lowering remains uncertain, and warrants further investigation. Although these preliminary findings show promise for the use of IF and ADF as alternatives to CR for weight loss and type 2 diabetes risk reduction, clear conclusions cannot be drawn because of the limited number of studies published in this field. Much work remains to be done to understand these diet strategies fully. ACKNOWLEDGMENTS

This work was funded by grants from the National Institutes of Health (NIDDK T32DK080674, NHLBI 1R01HL106228-01). Conflict of interests: None. REFERENCES

[1]. Diabetes Facts. National Center for Chronic Disease Prevention and Health Promotion Division of Diabetes Translation. 2013. [2]. Hamman RF, Wing RR, Edelstein SL, et al. Effect of weight loss with lifestyle intervention on risk of diabetes. Diabetes Care 2006;29:2102–7.

Barnosky et al

9

[3]. Andersen CJ, Fernandez ML. Dietary strategies to reduce metabolic syndrome. Rev Endocr Metab Disord 2013;14: 241–54. [4]. Wycherley TP, Moran LJ, Clifton PM, Noakes M, Brinkworth GD. Effects of energy-restricted high-protein, low-fat compared with standard-protein, low-fat diets: a metaanalysis of randomized controlled trials. Am J Clin Nutr 2012; 96:1281–98. [5]. Omodei D, Fontana L. Calorie restriction and prevention of ageassociated chronic disease. FEBS Lett 2011;585:1537–42. [6]. Das SK, Gilhooly CH, Golden JK, et al. Long-term effects of 2 energy-restricted diets differing in glycemic load on dietary adherence, body composition, and metabolism in CALERIE: a 1-y randomized controlled trial. Am J Clin Nutr 2007;85: 1023–30. [7]. Varady KA. Intermittent versus daily calorie restriction: which diet regimen is more effective for weight loss? Obes Rev 2011;12:e593–601. [8]. Varady KA, Hellerstein MK. Alternate-day fasting and chronic disease prevention: a review of human and animal trials. Am J Clin Nutr 2007;86:7–13. [9]. Bryner RW, Ullrich IH, Sauers J, et al. Effects of resistance vs. aerobic training combined with an 800 calorie liquid diet on lean body mass and resting metabolic rate. J Am Coll Nutr 1999;18: 115–21. [10]. Kabagambe EK, Ordovas JM, Tsai MY, et al. Smoking, inflammatory patterns and postprandial hypertriglyceridemia. Atherosclerosis 2009;203:633–9. [11]. Colditz GA, Willett WC, Rotnitzky A, Manson JE. Weight gain as a risk factor for clinical diabetes mellitus in women. Ann Intern Med 1995;122:481–6. [12]. Ye J. Mechanisms of insulin resistance in obesity. Front Med 2013;7:14–24. [13]. Lee MJ, Wu Y, Fried SK. Adipose tissue heterogeneity: implication of depot differences in adipose tissue for obesity complications. Mol Aspects Med 2013;34:1–11. [14]. Vague J. La differenciation sexuelle, facteur determinant des formes de l’obesity. Presse Med 1947;30:339–40. [15]. Banerji MA, Lebowitz J, Chaiken RL, Gordon D, Kral JG, Lebovitz HE. Relationship of visceral adipose tissue and glucose disposal is independent of sex in black NIDDM subjects. Am J Physiol 1997;273:E425–32. [16]. Xydakis AM, Case CC, Jones PH, et al. Adiponectin, inflammation, and the expression of the metabolic syndrome in obese individuals: the impact of rapid weight loss through caloric restriction. J Clin Endocrinol Metab 2004;89:2697–703. [17]. Klempel MC, Kroeger CM, Bhutani S, Trepanowski JF, Varady KA. Intermittent fasting combined with calorie restriction is effective for weight loss and cardio-protection in obese women. Nutr J 2012;11:98. [18]. Harvie MN, Pegington M, Mattson MP, et al. The effects of intermittent or continuous energy restriction on weight loss and metabolic disease risk markers: a randomized trial in young overweight women. Int J Obes (Lond) 2011;35:714–27. [19]. Heilbronn LK, Smith SR, Martin CK, Anton SD, Ravussin E. Alternate-day fasting in nonobese subjects: effects on body weight, body composition, and energy metabolism. Am J Clin Nutr 2005;81:69–73. [20]. Eshghinia S, Mohammadzadeh F. The effects of modified alternate-day fasting diet on weight loss and CAD risk factors in overweight and obese women. J Diabetes Metab Disord 2013;12:4. [21]. Johnson JB, Summer W, Cutler RG, et al. Alternate day calorie restriction improves clinical findings and reduces markers of

10

[22].

[23].

[24].

[25].

[26].

[27].

[28].

[29].

[30].

Barnosky et al

oxidative stress and inflammation in overweight adults with moderate asthma. Free Radic Biol Med 2007;42:665–74. Varady KA, Bhutani S, Church EC, Klempel MC. Short-term modified alternate-day fasting: a novel dietary strategy for weight loss and cardioprotection in obese adults. Am J Clin Nutr 2009;90:1138–43. Klempel MC, Kroeger CM, Varady KA. Alternate day fasting (ADF) with a high-fat diet produces similar weight loss and cardio-protection as ADF with a low-fat diet. Metabolism 2013;62:137–43. Bhutani S, Klempel MC, Kroeger CM, Trepanowski JF, Varady KA. Alternate day fasting and endurance exercise combine to reduce body weight and favorably alter plasma lipids in obese humans. Obesity (Silver Spring) 2013;21:1370–9. Varady KA, Bhutani S, Klempel MC, et al. Alternate day fasting for weight loss in normal weight and overweight subjects: a randomized controlled trial. Nutr J 2013;12:146. Trussardi Fayh AP, Lopes AL, Fernandes PR, ReischakOliveira A, Friedman R. Impact of weight loss with or without exercise on abdominal fat and insulin resistance in obese individuals: a randomised clinical trial. Br J Nutr 2013;110:486–92. Svendsen PF, Jensen FK, Holst JJ, Haugaard SB, Nilas L, Madsbad S. The effect of a very low calorie diet on insulin sensitivity, beta cell function, insulin clearance, incretin hormone secretion, androgen levels and body composition in obese young women. Scand J Clin Lab Invest 2012;72:410–9. Mollard RC, Luhovyy BL, Panahi S, Nunez M, Hanley A, Anderson GH. Regular consumption of pulses for 8 weeks reduces metabolic syndrome risk factors in overweight and obese adults. Br J Nutr 2012;108(Suppl 1):S111–22. Clifton PM, Noakes M, Keogh JB. Very low-fat (12%) and high monounsaturated fat (35%) diets do not differentially affect abdominal fat loss in overweight, nondiabetic women. J Nutr 2004;134:1741–5. de Luis DA, Izaola O, Garcia Alonso M, Aller R, Cabezas G, de la Fuente B. Effect of a commercial hypocaloric diet in weight loss and post surgical morbidities in obese patients with chronic arthropathy, a randomized clinical trial. Eur Rev Med Pharmacol Sci 2012;16:1814–20.

Translational Research - 2014

[31]. Agueda M, Lasa A, Simon E, Ares R, Larrarte E, Labayen I. Association of circulating visfatin concentrations with insulin resistance and low-grade inflammation after dietary energy restriction in Spanish obese non-diabetic women: role of body composition changes. Nutr Metab Cardiovasc Dis 2012;22: 208–14. [32]. Melanson KJ, Summers A, Nguyen V, et al. Body composition, dietary composition, and components of metabolic syndrome in overweight and obese adults after a 12-week trial on dietary treatments focused on portion control, energy density, or glycemic index. Nutr J 2012;11:57. [33]. Larson-Meyer DE, Heilbronn LK, Redman LM, et al. Effect of calorie restriction with or without exercise on insulin sensitivity, beta-cell function, fat cell size, and ectopic lipid in overweight subjects. Diabetes Care 2006;29:1337–44. [34]. Association AD. Diagnosing diabetes and learning about prediabetes. Diabetes basics. Centers for Disease Control Publication 2013. [35]. Pan XR, Li GW, Hu YH, et al. Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance. The Da Qing IGT and Diabetes Study. Diabetes Care 1997;20: 537–44. [36]. Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343–50. [37]. Eriksson KF, Lindgarde F. Prevention of type 2 (non-insulindependent) diabetes mellitus by diet and physical exercise. The 6-year Malmo feasibility study. Diabetologia 1991;34:891–8. [38]. Wilding JP. The importance of weight management in type 2 diabetes mellitus. Int J Clin Pract 2014;68:682–91. [39]. Wing RR, Blair EH, Bononi P, Marcus MD, Watanabe R, Bergman RN. Caloric restriction per se is a significant factor in improvements in glycemic control and insulin sensitivity during weight loss in obese NIDDM patients. Diabetes Care 1994; 17:30–6. [40]. Henry RR, Scheaffer L, Olefsky JM. Glycemic effects of intensive caloric restriction and isocaloric refeeding in noninsulindependent diabetes mellitus. J Clin Endocrinol Metab 1985; 61:917–25.