Nursing Research (Due 24 hours)

Article2-Quantitative.pdf

Home >Nursing homework help >Nursing Research (Due 24 hours)

Benefits of a Paleolithic diet with and without supervised exercise on fat mass, insulin sensitivity, and glycemic control: a randomized controlled trial in individuals with type 2 diabetes

Julia Otten1*†

Andreas Stomby1†

Maria Waling2

Andreas Isaksson3

Anna Tellström1

Lillemor Lundin-Olsson4

Søren Brage5

Mats Ryberg1

Michael Svensson3

Tommy Olsson1

1Department of Public Health and Clinical Medicine, Medicine, Umeå University, Umeå, Sweden 2Department of Food and Nutrition, Umeå University, Umeå, Sweden 3Department of Community Medicine and Rehabilitation, Sports Medicine Unit, Umeå University, Umeå, Sweden 4Department of Community Medicine and Rehabilitation, Physiotherapy, Umeå University, Umeå, Sweden 5MRC Epidemiology Unit, University of Cambridge, Cambridge, UK

*Correspondence to: Julia Otten, Department of Public Health and Clinical Medicine, Umeå University, SE-90185 Umeå, Sweden. E-mail: [email protected]

†The authors Julia Otten and Andreas Stomby have contributed equally to this paper.

Abstract

Background Means to reduce future risk for cardiovascular disease in subjects with type 2 diabetes are urgently needed.

Methods Thirty-two patients with type 2 diabetes (age 59±8 years) followed a Paleolithic diet for 12 weeks. Participants were randomized to either standard care exercise recommendations (PD) or 1-h supervised exercise sessions (aerobic exercise and resistance training) three times per week (PD-EX).

Results For thewithin group analyses, fatmass decreased by 5.7 kg (IQR:�6.6, �4.1; p<0.001) in the PD group and by 6.7 kg (�8.2,�5.3; p<0.001) in the PD- EX group. Insulin sensitivity (HOMA-IR) improved by 45% in the PD (p<0.001) and PD-EX (p<0.001) groups. HbA1c decreased by 0.9% (�1.2,�0.6; p<0.001) in the PD group and 1.1% (�1.7, �0.7; p<0.01) in the PD-EX group. Leptin decreased by 62% (p<0.001) in the PD group and 42% (p<0.001) in the PD-EX group. Maximum oxygen uptake increased by 0.2 L/min (0.0, 0.3) in the PD-EX group, and remained unchanged in the PD group (p<0.01 for the differ- ence between intervention groups). Male participants decreased lean mass by 2.6 kg (�3.6,�1.3) in the PD group and by 1.2 kg (�1.3, 1.0) in the PD-EX group (p<0.05 for the difference between intervention groups).

Conclusions A Paleolithic diet improves fat mass and metabolic balance in- cluding insulin sensitivity, glycemic control, and leptin in subjects with type 2 diabetes. Supervised exercise training may not enhance the effects on these outcomes, but preserves lean mass in men and increases cardiovascular fitness. Copyright © 2016 John Wiley & Sons, Ltd.

Keywords type 2 diabetes; Paleolithic diet; diet intervention; exercise; glycosyl- ated haemoglobin A; insulin sensitivity; leptin

Abbreviations LDL, low density lipoprotein; HDL, high density lipoprotein; HOMA-IR, homeostatic model assessment of insulin resistance; NEFAs, non- esterified fatty acids; PD, Paleolithic diet and general exercise recommendations; PD-EX, Paleolithic diet with 3-h supervised exercise training per week; QUICKI, quantitative insulin sensitivity check index; VO2max, maximal oxygen uptake

Introduction

Among patients with diabetes, cardiovascular disease is the primary cause of death [1]. Thus, in this population, it is imperative to counteract cardiovascular

RESEARCH ARTICLE

Received: 27 November 2015 Revised: 6 May 2016 Accepted: 10 May 2016

DIABETES/METABOLISM RESEARCH AND REVIEWS Diabetes Metab Res Rev 2017; 33: e2828. Published online 30 June 2016 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/dmrr.2828

risk factors, such as hyperglycemia, high blood pressure, and dyslipidemia through diet, exercise, and drug treatment [2]. Earlier studies suggested a Paleolithic diet had powerful beneficial metabolic effects on obesity, as well as in type 2 diabetes [3,4]. This diet emphasizes a high intake of vegeta- bles, fruit, nuts, eggs, fish, and lean meat, while excluding refined sugar, salt, legumes, dairy products, and grains.

Additional metabolic effects beyond diet may be achieved with structured exercise interventions [5,6]. The combination of diet interventions with energy restric- tions and resistance training or aerobic exercise is benefi- cial for body composition in non-diabetic subjects [7]. In subjects with type 2 diabetes, the combination of aerobic exercise with resistance training lowers HbA1c levels more than either exercise modality separately [8]. To the best of our knowledge, studies on Paleolithic diet combined with resistance training and aerobic exercise have not been per- formed in subjects with type 2 diabetes.

In the present study, subjects with type 2 diabetes con- sumed a Paleolithic diet for 12 weeks, with or without super- vised aerobic exercise and resistance training. Our hypothesis was that exercise training would improve the beneficial ef- fects of a Paleolithic diet on fat mass and metabolic balance including insulin sensitivity, glycemic control, and leptin.

Materials and methods

Study design

We conducted a randomized controlled trial with two arms: Paleolithic diet and standard care exercise recom- mendations (PD) and Paleolithic diet with 1-h supervised exercise sessions three times per week (PD-EX). In a sec- ondary analysis, we included a non-randomized observa- tional group as a reference.

Participants of the randomized controlled trial

Subjects were recruited from the greater Umeå area of Northern Sweden through advertisements in local newspa- pers and posters at Umeå University Hospital. Recruitment began in 2012, and the study was completed in June 2014.We included individuals diagnosed with type 2 diabe- tes within the past 10 years, who had a BMI of 25–40 kg/m2

and were weight stable (i.e.<5%weight loss) for 6 months before study inclusion. Eligible males were 30–70 years old, while women were included after menopause and up to 70 years of age. All participants had HbA1c values between 6.5% and 10.8% (47–94 mmol/mol), and were using life- style modification and/or metformin for diabetes

treatment. Exclusion criteria were treatment with anti- diabetic drugs other than metformin, use of beta-blockers, blood pressure>160/100 mmHg, macroalbuminuria, heart disease, and being a smoker. Because we aimed to study the effect of exercise on sedentary individuals, we ex- cluded those who reported more than 30 min of moderate physical activity 5 days per week or resistance trainingmore than once every other week during the past 6 months. Of 261 volunteers who were interested in participating, 32 met the inclusion criteria and were randomized into the PD or PD-EX groups (Figure 1). All participants provided written informed consent. The study protocol was in accord with the Helsinki declaration, and was approved by the Re- gional Ethical Review Board, Umeå, Sweden.

Randomization and blinding

Participants were assigned to the PD and PD-EX groups using biassed coin minimization with an allocation ratio of 1:1 [9]. To minimize marginal imbalance based on the prognostic factors (sex and BMI above/below 30), par- ticipants were sequentially allocated using a base proba- bility of 0.9. The computer program MinimPy was used for treatment allocation [10]. Randomization was con- ducted after the baseline examinations. The study was single-blinded, such that group allocation was unknown to all staff that performed examinations and dietary counselling. Additionally, the statistician who randomized the participants and the research assistant who informed the participants of the randomization outcome were not involved in data collection or data analysis. The study was unblinded after the analysis of results.

Diet intervention

Both randomized groups (PD and PD-EX) were intro- duced to the Paleolithic diet after baseline examinations, and were instructed to follow the diet until all study mea- surements were completed. The diet was based on con- suming lean meat, fish, seafood, eggs, vegetables, fruits, berries, and nuts. Cereals, dairy products, legumes, re- fined fats, refined sugars, and salt were excluded with the exception of canned fish and cold cuts like ham. The diet was consumed ad libitum, with restrictions of the fol- lowing: eggs (1–2/day but a maximum of 5/week), pota- toes (1 medium sized/day), dried fruit (130 g/day), and nuts (60 g/day). Rapeseed or olive oil (maximum 15 g/day) and small amounts of honey and vinegar were allowed as flavouring in cooking. Participants were instructed to drink mainly still water. Coffee and tea were restricted to a maximum of 300 g/day, and red wine to a maximum of one glass/week.

2 of 11 J. Otten et al.

Each group participated separately in five group ses- sions held by a trained dietician at the Department of Food and Nutrition, Umeå University, Sweden. The first two meetings were held during the first 2 weeks, and the following meetings took place once a month. The par- ticipants received information about the diet and cooked food and were given recipes. Between the meetings, the participants could contact the dietician, who held the meetings by e-mail or phone.

Exercise intervention

Prior to randomization, both intervention groups received exercise recommendations based on the current guidelines for patients with type 2 diabetes. Thus, all study partici- pants were advised to performmoderate exercise (e.g. brisk walking) for at least 30 min every day. The PD-EX group underwent a program comprising a combination of aerobic exercise and resistance training in 1-h sessions three times weekly at the Sports Medicine unit at Umeå University. The exercise sessions were performed on weekdays, with at least 1 day of rest between sessions. They were super- vised by experienced personal trainers with bachelor’s de- grees in Sports Medicine. The training protocol had a progressive design in accord with the guidelines of the American College of Sports Medicine [11].

All exercise sessions started with aerobic exercise. The first session of each week consisted of low-intensity aerobic

training at 70% of the maximum heart rate on a cross- trainer (Monark Prime, XT 50, Vansbro, Sweden). The second session of the week consisted of ten high-intensity sprint intervals at 100% of the maximal workload on a cycle-ergometer (Monark, Ergomedic 839E, Vansbro, Swe- den), with low-intensity cycling between the sprints. The third session of each week comprised six moderate-intensity 5-min intervals between 45 and 60% of maximal workload on a cycle-ergometer. The duration/workload of the inter- vals increased every other week.When necessary, the inten- sity of the aerobic exercise sessions was adjusted in accord with the participant’s performance.

After the aerobic exercise, the sessions progressed to re- sistance training with both upper and lower body exer- cises, including leg presses, seated leg extensions, leg curls, hip raises, flat and incline bench presses, seated rows, dumbbell rows, lat pull-downs, shoulder raises, back extensions, burpees, sit-ups, step-ups, and wall ball shots. At each training session, the participant performed three to five of the aforementioned resistance exercises, with 10–15 repetitions and two to four sets. Once partici- pants could complete all repetitions, the workload was in- creased for the following session.

Measurements

At baseline and at 12 weeks, dietary intake was assessed using a 4-day self-reported weighed food record. Each

Figure 1. CONSORT flow diagram

Paleolithic Diet in Type 2 Diabetes 3 of 11

4-day food record period included one or two weekend days. Participants were instructed to weigh all food, beverages, and leftovers. Any uncertainties regarding the food records were clarified during meetings, via e-mail, or by phone. A trained dietician converted the reported food intake into estimated energy and nutrient intake using the nutritional analysis software Dietist XP version 3.2 (Kost och Näringsdata AB, Bromma, Sweden), based on the Swedish National Food Administration’s food database.

Participants were examined at baseline and after 12 weeks of the intervention by experienced physicians and nurses at the Clinical Research Center at Umeå Univer- sity Hospital, Umeå, Sweden. Resting energy expenditure was measured using indirect calorimetry (Datex-Ohmeda Deltatrac II; Datex-Ohmeda Inc., Madison, WI, USA) and adjusted by subtracting 5% during 8 h of sleep. Daily physi- cal activity energy expenditure over a 7-day period was estimated using data from a combined accelerometer and heart rate monitor (Actiheart®; CamNtech Ltd., Cambridge, UK) [12], modelled as described previously [13–15]. Diet-induced thermogenesis was fixed at 10% of the total energy expenditure. Total energy expenditure was calculated as the sum of the resting energy expenditure and physical activity expenditure plus 10%.

Fat mass (i.e. the primary outcome) and lean mass were analysed by dual-energy X-ray absorptiometry (Lunar Prodigy X-ray Tube Housing Assembly, Brand BX-1 L, Model 8743; GE Medical Systems, Madison, WI, USA). The participants were weighed on a digital cali- brated scale, wearing light clothing. Height was measured with a calibrated height-measuring gauge. Waist circum- ference was assessed with a measuring tape placed midway between the lowest rib and iliac crest during gentle exhalation. The abdomen height was measured at the umbilicus level with the participant lying down with straight legs.

An automated blood pressure meter (Boso Medicus, Bosch, Germany) was used to measure systolic and diastolic blood pressure from the right arm with the participant in a sitting position. Measurements were made twice at 2-min intervals, after 5 min of rest. Fasting venous blood samples were collected from patients in the intervention groups for analysis of HbA1c, serum insulin, serum cholesterol, high density lipoprotein (HDL), serum triglycerides, and plasma high-sensitivity C-reactive protein at the Department for Clinical Chemistry, Umeå University Hospital. We analysed fasting glucose from a capillary sample (HemoCue 201 RT; Radiometer Medical Aps, Brønshøj, Denmark). Aliquots of plasma were immediately stored at �80 °C for analysis of non-esterified fatty acids (NEFAs), adiponectin, and leptin after study completion. The NEFAs were analysed with NEFA-HR2 (Wako Chemicals, Neuss, Germany), adiponectin with the Human Adiponectin ELISA Kit, and

leptin with the Human Leptin ELISA Kit, both from Merck Millipore (Darmstadt, Germany). Insulin sensitivity was calculated as follows: homeostatic model assessment of insulin resistance (HOMA-IR)=(fasting glucose× fasting insulin) / 22.5 and the revised quantitative insulin sensitiv- ity check index (Revised QUICKI)=1/ (log fasting glucose + log fasting insulin+ log NEFA) [16,17]. The low density lipoprotein (LDL) was calculated as follows: (serum cholesterol� serum HDL� serum triglycerides) / 2.2. The maximal oxygen uptake (VO2max) and maximal workload were measured via cardiopulmonary exercise test at the De- partment of Clinical Physiology, Umeå University Hospital, Umeå, Sweden.

Observational group

For a secondary analysis, we recruited an observational group by advertisement in local newspapers and among those who were excluded from the intervention because of a lack of time, beta-blocker use, and cardiovascular disease. Nine individuals were included in the observa- tional group, one of whom could not attend the assess- ments at the end of the intervention period because of illness. Fasting glucose, fasting insulin, HbA1c, leptin, adiponectin, and blood lipids were analysed from venous blood samples. Body composition, weight, blood pressure, dietary intake, and physical activity energy expenditure were examined as described above.

Sample size and statistical analysis

The primary outcome in this study was the change in fat mass. Based on previous results from a similar study [18], we calculated that 13 individuals in each interven- tion group would be sufficient to detect a significant difference (p<0.05) with 80% power. Because several variables had a skewed distribution, the Wilcoxon rank- sum test was used to compare groups. All data were reported as medians with the interquartile range. The pri- mary analysis compared treatment effects (change from 0 to 12 weeks) between the PD and PD-EX groups. The change over time within each intervention group was determined using the Wilcoxon signed-rank test. The sec- ondary analysis compared the treatment effect (change from 0 to 12 weeks) in each intervention group with the observational group. A two-sided p value of <0.05 was considered statistically significant. All statistical analyses were performed using R version 3.1.1, a language and environment for statistical computing (R Foundation for Statistical Computing, Vienna, Austria).

4 of 11 J. Otten et al.

Results

Subject characteristics

The participants’ baseline characteristics are presented in Tables 1 and 2. The randomized groups did not differ in age, sex, BMI, or diabetes duration. The PD-EX group had higher fasting glucose and HDL levels than the PD group. During the course of the study, one participant in the PD group stopped his metformin treatment, two par- ticipants in the PD group stopped their blood pressure medication, and one participant in the PD group started antihypertensive treatment.

Compliance with the diet and supervised exercise program

Dietary intake did not differ between the groups at base- line and 12 weeks, except for a higher fibre intake in the PD-group at 12 weeks (Table 3). Both groups increased their relative intake of protein and their intake of mono- unsaturated and polyunsaturated fatty acids. Both groups lowered their intake of carbohydrates and saturated fatty acids. The reduction of sodium intake was only significant in the PD-EX group. Nine of the 14 participants in the PD- EX group completed the 36 exercise sessions according to the study protocol. The remaining five participants com- pleted between 27 and 35 workouts during the study pe- riod. The participants in the PD-EX group increased the cumulative weight load (weight× repetitions× sets) with the leg press during one exercise session from 1350 kg (900–1800) to 3000 kg (2700–4000) after 12 weeks.

Energy balance

At baseline, energy intake (kcal/day) and total energy ex- penditure (kcal/day) did not differ between groups (Table 2). Baseline energy intake in the PD group was 1112 kcal/day (�1434, �609) less than the total energy expenditure. In the PD-EX group, baseline energy intake was 1340 kcal/day (�1909, �778) less than the total en- ergy expenditure. Energy intake decreased in both groups during the intervention (Table 2). Furthermore, total en- ergy expenditure decreased in the PD group (p<0.05), but remained stable in the PD-EX-group (p=0.17, Table 2). This was caused by a decrease in the resting en- ergy expenditure, while the physical activity energy ex- penditure was unchanged. At the end of the study, the PD group reported an energy intake that was 1245 kcal/day (�1480, �905) less than the total energy expenditure; while, the energy intake of the PD-EX group was 1657 kcal/day (�2533, �881) less than total energy expenditure.

Body composition

Fat mass decreased during the study in both the PD and PD-EX groups (Figure 2). Both groups also showed de- creases in body weight, abdominal height, and waist cir- cumference, without differences between intervention groups (Table 2). Male participants decreased their waist circumference more in the PD-group compared to the PD- EX group (p<0.05, Supplementary Table S2). Males in the PD-EX group retained more lean mass than males in the PD-group (p<0.05, Supplementary Table S2).

Glucose metabolism

Insulin sensitivity and glycemic control improved in both groups, without a difference between groups. The HOMA-IR and revised QUICKI improved in both interven- tion groups (Figure 2, Table 2), and the HbA1c decreased during the study in both the PD group (19%) and the PD- EX group (20%, Table 2).

Cardiovascular fitness

Resting heart rate decreased more in the PD-EX group than the PD group (Table 2). The VO2max and the ergom- eter cycling workload increased during the study in the PD-EX group, but not in the PD group (Figure 2).

Table 1. Participants’ baseline characteristics

Paleolithic diet (n=15)

Paleolithic diet + exercise (n=14)

Age (years) 60 (53–64) 61 (58–66) Men/women (n) 10/5 9/5 Diabetes duration (years) 3 (1–5) 5.5 (1–8) BMI (kg/m2) 31.4 (29.4–33.1) 31.7 (29.2–35.4) Diabetes treatment (n) Diet only 5 4 Metformin 10 10 Other treatment (n) ACEI/ARB 9 10 Diuretic 6 5 Calcium-channel blocker 4 5 Statin 6 8 Antiplatelet drug 2 3 Other 8 2

Data are reported as the median (interquartile range). Abbreviations: ACEI, ACE inhibitor; ARB, angiotensin receptor blocker.

Paleolithic Diet in Type 2 Diabetes 5 of 11

Table 2. Energy balance, body composition, and cardiovascular risk factors during 12 weeks of intervention

Paleolithic diet (n=15) Paleolithic diet + exercise (n=14)

Energy balance Energy intake (kcal/day) Baseline 2022 (1583–2268) 1595 (1428–2257) Change 0–12 weeks �291 (�587, �66)## �530 (�863, �157)###

Physical activity energy expenditure (kcal/day) Baseline 1022 (904–1319) 997 (806–1568) Change 0–12 weeks �28 (�208, 30) �18 (�368, 340)

Physical activity energy expenditure (kcal/kg/day) Baseline 11.6 (10.6–13.0) 10.1 (9.2–16.6) Change 0–12 weeks 0.1 (�1.5, 2.1) 0.6 (�2.9, 4.8)

Resting energy expenditure (kcal/day) Baseline 1620 (1463–1719) 1709 (1319–1883) Change 0–12 weeks �120 (�157, �71)### �89 (�164, �44)##

Resting energy expenditure (kcal/kg/day) Baseline 17.0 (16.4–17.6) 16.7 (15.7–18.6) Change 0–12 weeks 0.0 (�0.4, 0.8) 0.5 (�0.3, 1.4)

Total energy expenditure (kcal/day) Baseline 2995 (2754–3356) 2960 (2433–3855) Change 0–12 weeks �227 (�307, �91)# �312 (�562, 122)

Weight (kg) Baseline 90.0 (83.3–100.8) 97.3 (83.9–110.3) Change 0–12 weeks �7.1 (�9.7, �6.3)### �7.1 (�8.7, �6.2)###

Body composition Body fat (%) Baseline 37.8 (33.1–40.8) 37.7 (34.7–43.1) Change 0–12 weeks �3.5 (�4.4, �2.6)### �4.1 (�5.8, �3.4)###

Lean mass (kg) Baseline 56.5 (49.0–63.8) 61.0 (44.1–66.8) Change 0–12 weeks �1.4 (�3.3, �1.2)### �1.2 (�1.4, �0.2)

Waist circumference (cm) Baseline 111 (105–116) 108 (104–115) Change 0–12 weeks �9 (�12, �7)### �8 (�10, �7)###

Abdominal height (cm) Baseline 27.2 (25.0–29.0) 26.5 (23.0–29.8) Change 0–12 weeks �3.6 (�4.7, �2.5)### �3.0 (�4.3, �1.1)###

Glucose metabolism HbA1c (%) Baseline 7.1 (6.5–7.3) 7.3 (6.8–7.6) Change 0–12 weeks �0.9 (�1.2, �0.6)### �1.1 (�1.7, �0.7)##

HbA1c (mmol/mol) Baseline 54 (48–57) 57 (51–60) Change 0–12 weeks �10 (�13, �6)### �12 (�19, �8)##

Fasting glucose (mmol/L) Baseline 8.0 (7.2–8.4) 8.9 (7.9–10.5)* Change 0–12 weeks �0.9 (�1.7, �0.2)# �2.0 (�3.2, �1.1)##

Fasting insulin (mIU/L) Baseline 23 (15–30) 16 (11–20) Change 0–12 weeks �8 (�16, �3)## �4 (�8, �2)###

Revised QUICKI Baseline 0.223 (0.192–0.227) 0.207 (0.196–0.224) Change 0–12 weeks 0.027 (0.004, 0.054)## 0.041 (0.031, 0.054)###

Cardiovascular fitness VO2max (mL/kg/min) Baseline 23.4 (21.5–27.0) 22.5 (21.0–25.2) Change 0–12 weeks 1.9 (0.6, 2.9)### 3.3 (2.7, 6.2)###*

Resting heart rate (bpm) Baseline 70 (65–78) 72 (66–77) Change 0–12 weeks �3 (�8, 1) �11 (�12, �7)#*

Blood pressure Systolic (mmHg) Baseline 135 (127–148) 132 (122–143) Change 0–12 weeks �17 (�24, 0)## �11 (�14, �7)###

Diastolic (mmHg) Baseline 86 (82–94) 82 (74–91) Change 0–12 weeks �9 (�15, �6)### �10 (�13, �7)###

(Continues)

6 of 11 J. Otten et al.

Blood pressure and blood lipids

Blood pressure decreased during the study in both interven- tion groups without any group difference: systolic, 13% in PD and 8% in PD-EX; diastolic, 10% in PD and 12% in PD- EX (Table 2). Triglycerides decreased in both study groups between baseline and 12 weeks; while, the HDL, LDL, and NEFA levels remained unchanged throughout the interven- tion (Table 2).

Adipokines

Leptin decreased in both the PD group (62%) and the PD- EX group (42%) (Table 2). Adiponectin increased in the PD group (8%) compared with the PD-EX group (Table 2).

Intervention groups (PD and PD-EX) versus the observational group

There were no significant differences in baseline characteris- tics between the observational group and the PD and PD-EX intervention groups (Supplementary Table S1). Compared with the observational group, the PD and PD-EX groups de- creased their total energy intakes and intakes of carbohy- drates and saturated fatty acids, but increased their relative

intake of protein and monounsaturated fatty acids during the 12 weeks of intervention (Supplementary Tables S2 and S3). The intervention groups improved fat mass (p<0.001), HOMA-IR, fasting insulin, HbA1c, systolic and diastolic blood pressure, and leptin comparedwith the obser- vational group (Supplementary Table S2). Triglycerides did not decrease in the intervention group compared with the observational group (Supplementary Table S2).

Discussion

Twelve weeks on a Paleolithic diet improved fat mass and metabolic balance including insulin sensitivity, glycemic control and leptin among individuals with type 2 diabetes. The addition of resistance training and aerobic exercise un- der observation increased cardiovascular fitness, without further improvements in fat mass or glycemic control. The observed effects of the Paleolithic diet were substantial. The lowering of HbA1c by 0.9% units was an effect size sim- ilar to that reported withmetformin in type 2 diabetes [19]. A previous study demonstrated that the Paleolithic diet re- duced HbA1c by 0.4%units more than a conventional diabe- tes diet [4]. The UK prospective diabetes study stated that a 1% unit improvement of HbA1c reduces microvascular com- plications by 37% and reduces diabetes-related death by 21% [20]. Thus, if sustained over time, the improvement

Table 2. (continued)

Paleolithic diet (n=15) Paleolithic diet + exercise (n=14)

Blood lipids Total cholesterol (mmol/L) Baseline 4.2 (3.4–4.7) 4.3 (4.1–4.9) Change 0–12 weeks �0.3 (�0.6, 0.1) �0.6 (�0.6, �0.4)##

Triglycerides (mmol/L) Baseline 2.1 (1.4–2.9) 1.7 (1.1–2.4) Change 0–12 weeks �0.6 (�1.5, �0.2)## �0.5 (�1.0, �0.2)###

HDL (mmol/L) Baseline 0.85 (0.81–0.99) 1.09 (0.98–1.21)** Change 0–12 weeks �0.01 (�0.08, 0.05) 0.01 (�0.03, 0.07)

LDL (mmol/L) Baseline 2.1 (1.8–2.7) 2.4 (2.0–3.0) Change 0–12 weeks �0.1 (�0.4, 0.2) �0.1 (�0.5, 0.1)

NEFA (μmol/L) Baseline 599 (560–756) 826 (670–922) Change 0–12 weeks 26 (�12, 173) �56 (�128, 117)

Adipokines Leptin (ng/mL) Baseline 13.8 (6.4–26.5) 13.3 (7.2–16.7) Change 0–12 weeks �8.5 (�12.2, �2.6)### �5.6 (�9.4, �3.5)###

Adiponectin (ng/mL) Baseline 4685 (2942–5939) 4864 (4004–6268) Change 0–12 weeks 379 (213, 717)## 5 (�256, 282)*

High-sensitivity CRP (mg/L) Baseline 1.2 (0.7–1.9) 1.5 (0.7–2.5) Change 0–12 weeks �0.4 (�1.1, 0.0) �0.4 (�0.9, 0.0)##

Data are reported as the median (interquartile range); *p< 0.05, **p< 0.01 between the Paleolithic diet group and the Paleolithic diet+ exercise group; #p< 0.05, ##p< 0.01, ###p< 0.001 for the change over time from baseline to 12 weeks within the group. Abbreviation: CRP, C-reactive protein.

Paleolithic Diet in Type 2 Diabetes 7 of 11

in glycemic control will provide large benefits in terms of morbidity and mortality.

These powerful observed effects of the Paleolithic diet may be explained by altered dietary patterns. The partici- pants reported reduced intakes of carbohydrates and satu- rated fatty acids, with relatively higher intakes of protein, as well as monounsaturated and polyunsaturated fatty

acids. The reduction of carbohydrates with a high glycemic index may be an important part of the beneficial effects of this diet [21]. Furthermore, increased intake of monounsat- urated fat may reduce postprandial hyperglycemia [22].

Supervised training with aerobic exercise in combina- tion with resistance training did not improve glycemic control and insulin sensitivity beyond the improvements observed with the Paleolithic diet alone. This was unex- pected, as exercise training has previously been shown to improve glycemic control substantially, particularly when combining resistance training with aerobic exercise for more than 150 min per week [6,23,24]. Among indi- viduals with type 2 diabetes, structured exercise interven- tions reduce HbA1c levels by about 0.6% units without weight changes [6]. In contrast, the addition of resistance or aerobic exercise to short-term (16 weeks) dietary inter- ventions with a calorie-restricted high-protein diet or a very low-calorie diet had limited additive effects on insu- lin sensitivity and glycemic control in patients with type 2 diabetes [18,25]. Notably, our diet recommendations were given ad libitum without any restrictions of caloric intake. However, it is possible that the catabolic state caused by decreased energy intake may have masked any potential effects of exercise on glycemic control and insulin sensitivity. Moreover, several study participants were using metformin or statins on a daily basis. These drugs may blunt the positive effects of exercise [26–28].

Importantly, cardiovascular fitness improved signifi- cantly in the PD-EX group compared with the PD group. Low cardiorespiratory fitness is a strong risk factor for all-cause mortality, independent of glycemic status and other cardiovascular risk factors [29,30]. A large cohort study showed that an increase of 1.44 mL/kg/min in VO2max (equivalent to a 1-min increase in the Balke pro- tocol treadmill time) corresponded to a 7.9% reduction in overall mortality [31]. Using these data, the presently ob- served increase of 3.3 mL/kg/min in VO2max would lead to an 18% reduction in all-cause mortality if changes can be sustained over time. Furthermore, it is of major interest to study if tissue-specific insulin sensitivity is influenced differently between groups. We demonstrated earlier that weight reduction by a Paleolithic diet had a profound effect on liver insulin sensitivity, while peripheral insulin sensitivity was unaltered [32]. The exercise intervention would be expected to add an increased muscular (peripheral) sensitivity. This is of interest because insulin resistance in skeletal muscle can play a key role in the de- velopment of metabolic complications in obesity-related disorders, including type 2 diabetes [33].

The combination of aerobic and resistance training is known to preserve or even increase lean mass during diet intervention [34]. In our study male participants in the PD-EX group lost less lean mass compared to males in the PD group. This difference was not significant if men

Table 3. Dietary intake

Paleolithic diet (n=14)

Paleolithic diet + exercise (n=13)

Protein (g/day) Baseline 83 (72–99) 77 (67–106) 12 weeks 96 (80–111) 79 (58–100)

Carbohydrate (g/day) Baseline 200 (160–262) 169 (152–197) 12 weeks 127 (93–158)### 77 (71–102)###

Total fat (g/day) Baseline 88 (66–102) 67 (48–94) 12 weeks 71 (56–97) 61 (47–79)

Protein (E%) Baseline 17 (15–19) 18 (17–20) 12 weeks 24 (19–27)### 26 (22–29)###

Carbohydrate (E%) Baseline 41 (37–45) 42 (33–48) 12 weeks 31 (24–39)## 27 (24–29)###

Total fat (E%) Baseline 39 (37–40) 34 (31–41) 12 weeks 42 (37–48) 45 (37–47)#

Saturated fatty acids (E%) Baseline 15.3 (13.4–16.8) 13.3 (12.1–17.1) 12 weeks 9.6 (7.9–11.6)### 8.8 (8.7–10.7)###

Monounsaturated fatty acids (E%) Baseline 14 (14–17) 12 (11–15) 12 weeks 20 (16–24)# 23 (20–24)##

Polyunsaturated fatty acids (E%) Baseline 5.0 (4.8–6.4) 5.4 (4.0–6.3) 12 weeks 7.9 (6.6–8.7)# 8.4 (6.8–9.7)##

Saturated fatty acids (g/day) Baseline 34 (24–44) 27 (20–35) 12 weeks 15 (12–23)### 14 (10–17)###

Monounsaturated fatty acids (g/day) Baseline 31 (26–38) 27 (17–36) 12 weeks 36 (25–52) 28 (23–40)

Polyunsaturated fatty acids (g/day) Baseline 12 (10–13) 10 (7–14) 12 weeks 15 (10–19) 13 (7–16)

Omega-3 fatty acids (g/day) Baseline 2.3 (2.0–3.1) 2.2 (1.4–2.9) 12 weeks 2.4 (1.3–4.6) 2.7 (1.3–2.9)

Omega-6 fatty acids (g/day) Baseline 10.9 (8.6–12.0) 8.0 (6.7–10.8) 12 weeks 11.8 (7.5–16.1) 10.6 (6.1–13.8)

Dietary cholesterol (mg/day) Baseline 315 (213–364) 324 (206–526) 12 weeks 531 (390–686)## 510 (399–632)

Dietary fibre (g/day) Baseline 21 (18–26) 20 (18–22) 12 weeks 23 (15–30) 14 (13–17)##*

Sodium (mg/day) Baseline 3051 (2610–3863) 3003 (2449–4097) 12 weeks 2119 (1745–2843) 1789 (1223–2786)#

Data are reported as the median (interquartile range); *p< 0.05 between the Paleolithic diet group and the Paleolithic diet+ exer- rcise group; #p< 0.05, ##p< 0.01, ###p< 0.001 for the change over time from baseline to 12 weeks within the group.

8 of 11 J. Otten et al.

and women were analysed together. Notably, a recent study showed that 12 weeks of exercise increased lean mass in obese males, but not in women [35].

In line with earlier studies, another beneficial effect of the Paleolithic diet is a major reduction in blood pressure [3,4,32]. The combination of weight reduction and re- duced sodium intake may be important for this effect. The reduced triglyceride levels, in both study groups, are also consistent with earlier studies. Previous studies showed that a Paleolithic diet decreased triglycerides even more than a consensus diet [3,4], but exercise under su- pervision did not significantly improved blood lipids [6].

Leptin levels decreased 62% in the PD group and 42% in the PD-EX group. Compared to other diet interventions this is a powerful reduction relative to the weight loss of 7.1 kg [36,37]. This is in line with a study of individuals with is- chemic heart disease where a Paleolithic diet for 12 weeks reduced leptin more relative to the amount of weight loss than a Mediterranean-like diet [38]. These beneficial ef- fects on leptin levels are of major importance because hyperleptinemia increases inflammation [39,40] and is an independent risk factor for cardiovascular events [41–43].

The increased adiponectin levels in the PD group may relate to weight loss, with increased protein intake as a contributing factor [44,45]. The unaltered hormone

levels in the PD-EX group may indicate the increased plasma volume because of physical activity [46].

In a secondary analysis, we compared the non- randomized observational group with the intervention groups. The intervention groups improved their anthropo- metric status and metabolic balance versus the observa- tional group, except for triglycerides, which improved in the observational group. Notably, the observational group was not randomized and therefore, we cannot guarantee equal distribution of confounding factors between the inter- vention groups and the observational group. Furthermore, some participants in the observational group suffered from cardiovascular disease, used beta-blockers, and did not have the time to participate in the interventions.

A strength of the present study is that energy intake was validated with objectively measured total energy expendi- ture. Differences between reported energy intake and mea- sured total energy expenditure at baselinemight be because of undereating, underreporting, overestimation of physical activity energy expenditure, and/or increased physical ac- tivity during the measurement period. Participants may have started changing their dietary intake and physical ac- tivity during the baseline-measuring period, even though they were not introduced to the intervention part of the study until baseline measurements were finished. Despite

Figure 2. Fat mass (a), insulin sensitivity (b), and cardiovascular fitness (c and d) during 12 weeks following either a Paleolithic diet with a supervised exercise program (PD-EX) or a Paleolithic diet combined with general exercise recommendations (PD). Boxes rep- resent medians and IQRs, whiskers represent the most extreme values besides outliers, and filled circles represent outliers (>1.5 IQR); **p< 0.01, ***p< 0.001

Paleolithic Diet in Type 2 Diabetes 9 of 11

the validation of energy intake, it is a weakness of the study that it is not known to what degree the participants actually followed the Paleolithic diet.

Based on our results, we conclude that the Paleolithic diet is a powerful tool to improve fatmass andmetabolic balance including insulin sensitivity, glycemic control, and leptin in individuals with type 2 diabetes. Supervised exercise train- ing did not provide additional effects on these outcomes, but preserved lean mass in men and increased cardiovascu- lar fitness. Detailed analyses of tissue-specific effects of these interventions, including putative effects on hepatic versusmuscle insulin sensitivity, are of further interest.

Acknowledgements

The authors gratefully acknowledge the study participants. We thank the research nurses Inger Arnesjö, Katarina Iselid, Lena Uddståhl, Camilla Ring, and Liv-Helene Bergman for their skillful technical assistance. We thank Marie Eriksson of the Department of Statistics, Umeå University, Sweden, for the randomization and Magnus Hedström of the Heart Centre, Umeå University Hospital, Sweden for planning and interpretation of the cardio- pulmonary exercise test. We thank laboratory technician Kristina Eriksson for adiponectin and leptin analysis and Kate Westgate (MRC Epidemiology Unit, University of Cambridge, UK) for assis- tance with physical activity data processing.

This study was supported by grants from the Swedish Heart and Lung Foundation (20120450), King Gustav V and Queen Vic- toria’s Foundation, The Swedish Diabetes Research Foundation (2014-096), the County Council of Västerbotten (VLL-460481), and Umeå University, Sweden.

Author contributions

JO and AS designed the study, recruited participants, collected the data, performed the statistical analysis, and wrote the manuscript. MW designed the study, imple- mented the dietary intervention, and analysed the data. AI conducted the exercise intervention and analysed the data. AT conducted the dietary intervention and analysed the data. LLO and MS designed the study and interpreted the data. SB analysed and interpreted the Actiheart data. MR designed the study, recruited participants, collected the data, and edited the manuscript. TO designed the study, interpreted the data, and wrote the manuscript. All authors actively participated in revising the article and approved the final version. JO is the guarantor of this work and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Conflicts of interest

The authors declare that there is no duality of interest as- sociated with this manuscript.

Clinical trial registration number

ClinicalTrials.gov NCT01513798

References

1. Gregg EW, Cheng YJ, Saydah S, et al. Trends in death rates among U.S. adults with and without diabetes between 1997 and 2006: findings from the Na- tional Health Interview Survey. Diabetes Care 2012; 35(6): 1252–7.

2. Authors/Task Force M, Ryden L, Grant PJ, et al. ESC Guidelines on diabetes, pre-diabetes, and cardiovascular dis- eases developed in collaboration with the EASD: the Task Force on diabetes, pre-diabetes, and cardiovascular dis- eases of the European Society of Car- diology (ESC) and developed in collaboration with the European Associ- ation for the Study of Diabetes (EASD). Eur Heart J 2013; 34(39): 3035–87.

3. Mellberg C, Sandberg S, Ryberg M, et al. Long-term effects of a Palaeolithic-type diet in obese postmenopausal women: a 2-year randomized trial. Eur J Clin Nutr 2014; 68: 350–7.

4. Jonsson T, Granfeldt Y, Ahren B, et al. Beneficial effects of a Paleolithic diet on cardiovascular risk factors in type 2

diabetes: a randomized cross-over pilot study. Cardiovasc Diabetol 2009; 8: 35.

5. Snowling NJ, Hopkins WG. Effects of dif- ferent modes of exercise training on glu- cose control and risk factors for complications in type 2 diabetic pa- tients: a meta-analysis. Diabetes Care 2006; 29(11): 2518–27.

6. Thomas DE, Elliott EJ, Naughton GA. Exer- cise for type 2 diabetes mellitus. Cochrane Database Syst Rev 2006; 3: CD002968.

7. Layman DK, Evans E, Baum JI, Seyler J, Erickson DJ, Boileau RA. Dietary protein and exercise have additive effects on body composition during weight loss in adult women. J Nutr 2005; 135(8): 1903–10.

8. Church TS, Blair SN, Cocreham S, et al. Effects of aerobic and resistance training on hemoglobin A1c levels in patients with type 2 diabetes: a randomized controlled trial. JAMA 2010; 304(20): 2253–62.

9. Pocock SJ, Simon R. Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Bio- metrics 1975; 31(1): 103–15.

10. Saghaei M, Saghaei S. Implementation of an open-source customizable minimi- zation program for allocation of patients to parallel groups in clinical trials. J Biomed Sci Eng 2011; 4: 734–9.

11. Garber CE, Blissmer B, Deschenes MR, et al. American College of Sports Medicine position stand. Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise. Med Sci Sports Exerc2011;43(7): 1334–59.

12. Brage S, Brage N, Franks PW, Ekelund U, Wareham NJ. Reliability and validity of the combined heart rate and move- ment sensor Actiheart. Eur J Clin Nutr 2005; 59(4): 561–70.

13. Brage S, Ekelund U, Brage N, et al. Hier- archy of individual calibration levels for heart rate and accelerometry to measure physical activity. J Appl Physiol (1985) 2007; 103(2): 682–92.

14. Stegle O, Fallert SV, MacKay DJ, Brage S. Gaussian process robust regression

10 of 11 J. Otten et al.

http://ClinicalTrials.gov

for noisy heart rate data. IEEE Trans Biomed Eng 2008; 55(9): 2143–51.

15. Brage S,Westgate K,Wijndaele K, Godinho J, Griffin S, Wareham N (2013) Evaluation of a method for minimizing diurnal infor- mation bias in objective sensor data. In: 3rd International Conference on Ambula- tory Monitoring of Physical Activity and Movement, Amherst, Massachusetts, USA

16. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Ho- meostasis model assessment: insulin re- sistance and beta-cell function from fasting plasma glucose and insulin con- centrations in man. Diabetologia 1985; 28(7): 412–9.

17. Perseghin G, Caumo A, Caloni M, Testolin G, Luzi L. Incorporation of the fasting plasma FFA concentration into QUICKI improves its association with in- sulin sensitivity in nonobese individuals. J Clin Endocrinol Metab 2001; 86(10): 4776–81.

18. Wycherley TP, Noakes M, Clifton PM, Cleanthous X, Keogh JB, Brinkworth GD. A high-protein diet with resistance exercise training improves weight loss and body composition in overweight and obese patients with type 2 diabetes. Diabetes Care 2010; 33(5): 969–76.

19. Campbell IW, Howlett HC. Worldwide experience of metformin as an effective glucose-lowering agent: a meta- analysis. Diabetes Metab Rev 1995; 11 (Suppl 1): S57–62.

20. Stratton IM, Adler AI, Neil HA, et al. Asso- ciation of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective ob- servational study. BMJ 2000; 321 (7258): 405–12.

21. Manheimer EW, van Zuuren EJ, Fedorowicz Z, Pijl H. Paleolithic nutri- tion for metabolic syndrome: systematic review and meta-analysis. Am J Clin Nutr 2015; 102(4): 922–32.

22. O’Keefe JH, Gheewala NM, O’Keefe JO. Dietary strategies for improving post- prandial glucose, lipids, inflammation, and cardiovascular health. J Am Coll Cardiol 2008; 51(3): 249–55.

23. Sigal RJ, Kenny GP, Wasserman DH, Castaneda-Sceppa C, Physical activity/ exercise and type 2 diabetes. Diabetes Care 2004; 27(10): 2518–39.

24. Sigal RJ, Kenny GP, Boule NG, et al. Ef- fects of aerobic training, resistance train- ing, or both on glycemic control in type

2 diabetes: a randomized trial. Ann In- tern Med 2007; 147(6): 357–69.

25. Snel M, Gastaldelli A, Ouwens DM, et al. Effects of adding exercise to a 16-week very low-calorie diet in obese, insulin- dependent type 2 diabetes mellitus pa- tients. J Clin Endocrinol Metab 2012; 97 (7): 2512–20.

26. Malin SK, Gerber R, Chipkin SR, Braun B. Independent and combined effects of exercise training and metformin on insu- lin sensitivity in individuals with predia- betes. Diabetes Care 2012; 35(1): 131–6.

27. Sharoff CG, Hagobian TA, Malin SK, et al. Combining short-term metformin treat- ment and one bout of exercise does not in- crease insulin action in insulin-resistant individuals.Am J Physiol Endocrinol Metab 2010; 298(4): E815–23.

28. Mikus CR, Boyle LJ, Borengasser SJ, et al. Simvastatin impairs exercise train- ing adaptations. J Am Coll Cardiol 2013; 62(8): 709–14.

29. Kohl HW, Gordon NF, Villegas JA, Blair SN. Cardiorespiratory fitness, glycemic status, and mortality risk in men. Diabe- tes Care 1992; 15(2): 184–92.

30. Wei M, Gibbons LW, Kampert JB, Nichaman MZ, Blair SN. Low cardiorespi- ratory fitness and physical inactivity as predictors of mortality in men with type 2 diabetes. Ann Intern Med 2000; 132 (8): 605–11.

31. Blair SN, Kohl HW 3rd, Barlow CE, Paffenbarger RS Jr, Gibbons LW, Macera CA. Changes in physical fitness and all- cause mortality. A prospective study of healthy and unhealthy men. JAMA 1995; 273(14): 1093–8.

32. RybergM, Sandberg S,Mellberg C, et al.A Palaeolithic-type diet causes strong tissue- specific effects on ectopic fat deposition in obese postmenopausal women. J Intern Med 2013; 274(1): 67–76.

33. Shulman GI. Ectopic fat in insulin resis- tance, dyslipidemia, and cardiometabolic disease. N Engl J Med 2014; 371(12): 1131–41.

34. Ghroubi S, Elleuch H, Chikh T, Kaffel N, Abid M, Elleuch MH. Physical training combined with dietary measures in the treatment of adult obesity. A comparison of two protocols. Ann Phys Rehabil Med 2009; 52(5): 394–413.

35. Sanal E, Ardic F, Kirac S. Effects of aer- obic or combined aerobic resistance ex- ercise on body composition in overweight and obese adults: gender

differences. A randomized intervention study. Eur J Phys Rehabil Med 2013; 49 (1): 1–11.

36. Cummings DE, Weigle DS, Frayo RS, et al. Plasma ghrelin levels after diet- induced weight loss or gastric bypass surgery. N Engl J Med 2002; 346(21): 1623–30.

37. Sumithran P, Prendergast LA, Delbridge E, et al. Long-term persistence of hor- monal adaptations to weight loss. N Engl J Med 2011; 365(17): 1597–604.

38. Jonsson T, Granfeldt Y, Erlanson- Albertsson C, Ahren B, Lindeberg S. A paleolithic diet is more satiating per cal- orie than a mediterranean-like diet in in- dividuals with ischemic heart disease. Nutr Metab (Lond) 2010; 7: 85.

39. Loffreda S, Yang SQ, Lin HZ, et al. Leptin regulates proinflammatory immune re- sponses. FASEB J 1998; 12(1): 57–65.

40. Yamagishi SI, Edelstein D, Du XL, Kaneda Y, Guzman M, Brownlee M. Leptin in- duces mitochondrial superoxide produc- tion and monocyte chemoattractant protein-1 expression in aortic endothelial cells by increasing fatty acid oxidation via protein kinase A. J Biol Chem 2001; 276 (27): 25096–100.

41. Soderberg S, Ahren B, Stegmayr B, et al. Leptin is a risk marker for first-ever hem- orrhagic stroke in a population-based co- hort. Stroke 1999; 30(2): 328–37.

42. Soderberg S, Stegmayr B, Stenlund H, et al. Leptin, but not adiponectin, pre- dicts stroke in males. J Intern Med 2004; 256(2): 128–36.

43. Soderberg S, Ahren B, Jansson JH, et al. Leptin is associated with increased risk of myocardial infarction. J Intern Med 1999; 246(4): 409–18.

44. Belalcazar LM, Lang W, Haffner SM, et al. Adiponectin and the mediation of HDL-cholesterol change with improved lifestyle: the Look AHEAD Study. J Lipid Res 2012; 53(12): 2726–33.

45. Kitabchi AE, McDaniel KA, Wan JY, et al. Effects of high-protein versus high- carbohydrate diets on markers of beta- cell function, oxidative stress, lipid per- oxidation, proinflammatory cytokines, and adipokines in obese, premenopausal women without diabetes: a randomized controlled trial. Diabetes Care 2013; 36 (7): 1919–25.

46. Convertino VA. Blood volume response to physical activity and inactivity. Am J Med Sci 2007; 334(1): 72–9.

Supporting information

Supporting information may be found in the online version of this article.

Paleolithic Diet in Type 2 Diabetes 11 of 11