ANZCTR - Registration

Registration number

ACTRN12623000665639p

Ethics application status

Submitted, not yet approved

Date submitted

6/06/2023

Date registered

20/06/2023

Date last updated

20/06/2023

Date data sharing statement initially provided

20/06/2023

Type of registration

Prospectively registered

Titles & IDs

Public title

Ketogenic low carbohydrate, high fat diet in marathon runners

Scientific title

The effects of a ketogenic low carbohydrate, high fat diet on performance in endurance marathon runners: a randomized controlled trial

Secondary ID [1]

none

Universal Trial Number (UTN)

Trial acronym

Linked study record

Health condition

Health condition(s) or problem(s) studied:

Athletic performance nutrition

Body composition

Glycaemic control

Condition category

Condition code

Diet and Nutrition

Other diet and nutrition disorders

Physical Medicine / Rehabilitation

Other physical medicine / rehabilitation

Intervention/exposure

Study type

Interventional

Description of intervention(s) / exposure

Methodology
Single-blind randomized controlled trial. Competitive marathon runners will be randomized to their usual high carbohydrate diet (comparator) or a ketogenic-low carbohydrate, high fat (K-LCHF) diet (intervention) for 6 months. It will not be possible to blind participants to diet group allocation, but outcome assessors will be blinded to the primary outcome during the performance test. Baseline marathon performance will be assessed in the 2023 Sydney marathon. Post intervention performance will be assessed 12 months later in the 2024 Sydney marathon. This design allows for real-world running performance to be tested with practicable control for season of the year and racecourse effects on running time. After baseline marathon performance is measured, diet and training will be monitored for 6 months leading into the diet randomization (lead-in diet and training monitoring). Baseline marathon performance and baseline and pre randomization aerobic capacity (VO2max), sex, body composition, as well as training characteristics (overall load, low / high intensity training split) assessed during the lead-in, will be used to pair-match participants. Once matched, participants will be randomly assigned (blinded, random number generator method) to follow either their usual high carbohydrate diet (comparator) or a K-LCHF diet (intervention) for the 6 months leading into the 2024 Sydney marathon.

Diet intervention
Before being randomized to a dietary intervention group, diet will be assessed at baseline and at 1-month intervals (diet reporting for 2 weekdays and 1 weekend day on each occasion) over a 6-month period (lead-in diet monitoring) using the diet tracking app, EasyDietDiary (Australian based diet tracking App, password protected login access). This will be done to confirm that participants are following a high carbohydrate diet (i.e., following the Australian dietary guidelines) at baseline and during lead-in to the diet intervention. After randomization, participants in the K-LCHF diet intervention group will receive nutritional counselling and resources to assist in learning how to adhere to a K-LCHF diet. The diet intervention will be delivered online and will involve reviewing a series of modules that have been designed specifically for this study. Modules will cover practical knowledge and skills required to follow the K-LCHF diet, including acceptable food items, shopping and meal preparation suggestions presented as written text, figures, and short videos. A module covering training and competition fuelling strategies that adhere to the principles of a K-LCHF diet will also be provided. Reviewing the online learning modules will take approximately 4 hours over the first 2 weeks of the diet intervention. Learning materials will remain accessible at any time during the 6 month intervention. A ketogenic, low carbohydrate diet is a whole food, nutrient rich diet that avoids grains, starches, refined sugars, and most processed foods. Well-formulated ketogenic diets promote the consumption of meat, chicken, seafood, eggs, dairy, leafy and cruciferous vegetables, avocados, nuts, seeds, and berries. Nutritional counselling and monitoring will be overseen by Associate Professor Caryn Zinn, a Registered Dietitian with more than 15 years’ experience. Participants will be asked to use EasyDietDiary to record all food intake for 7 days every week for the first 4 weeks after randomization, followed by 3 days (2 weekdays and 1 weekend day) every month for the remainder of the study (i.e., weeks;1, 2, 3, 4, months; 2, 3, 4, 5, 6). The K-LCHF diet will be prescriptive with regards to macronutrient intake (70% fat, 20% protein, and the lesser of 20 g or 10% nett carbohydrates) but ad libitum for total calories plus or minus 20% of lead-in caloric intake. Participant EasyDietDiary diet information will be reviewed at the end of each diet monitoring assessment block and participants will be advised if any modifications to macronutrient composition, micronutrient requirements or total caloric intake is required. Suggested meal plans and meal recipes will be formulated by the study dietitian to meet subjects’ micronutrient requirements with special attention to sodium intake because higher sodium losses have been shown to occur during very low carbohydrate intake. Diets that are very low in carbohydrates stimulate the production of ketones (an alternate source of fuel when glucose is low) and ketones are considered a valuable biomarker demonstrating adherence to K-LCHF diets. Participants will self-monitor blood ketones throughout the study (blood finger prick sample) by measuring their blood ketone levels. Participants will be provided with a ketone monitor, ketone strips, alcohol wipes and lancets and asked to take replicate (i.e., two measures for increased reliability) blood ketone measurements once per month during the diet intervention.

Competition fuelling strategy
Participants will be asked to report their marathon race fuelling strategy i.e., the intervals, type, and quantify of fuel consumed during their marathon race. Participants randomized to the K-LCHF diet intervention group will be encouraged to follow the prescriptive diet macronutrient intake (70% fat, 20% protein, and the lesser of 20 g or 10% carbohydrates) on the day of the post intervention marathon race but it may be unreasonable for researchers to enforce this. While it is not unusual for some athletes (even those habitually consuming a standard high carbohydrate diet) to completely avoid carbohydrate consumption during marathon races, unless athletes have previously experimented using a low carbohydrate fuelling strategy, some individuals in this study may not be confident in limiting carbohydrate fuelling during their marathon. To be practical, it is reasonable to accept that an additional carbohydrate allowance during the marathon race, should athletes feel they require it, is in the best interests of the athletes and compatible with the research design. Habitual K-LCHF diet, not marathon race fuelling strategy is the independent variable in this study.

Training and athlete monitoring
We will use questionnaires within a secure password protected App, StackTeamApp (no cost to participants) to monitor athletes training and general health, including any adverse effects during the diet intervention. The App will assist in managing participant scheduling for attending lab testing sessions and sending in App and text reminders to complete research tasks. This approach is designed to minimize the burden on participants for reporting data. Training loads will be tracked throughout the study and reported as training hours, self-reported intensity splits, training pace, and estimated running distances per week. When available, participants can share training metrics measured using GPS tracking watches and measures of heart rate with researchers. The intervention will not include control over athlete training practices (load and structure). Participants will be asked to select and specify a marathon training plan of their choice at the start of the study and follow the same training plan as closely as circumstances allow, before both the 2023 and 2024 Sydney marathons. Participants in this study will be experienced marathon runners and can therefore select any training plan that is compatible with their personalized training approach. Imposing training requirements on experienced marathon runners could affect study outcomes in unexpected ways. For example, it may have unintended negative consequences for recovery, or injury risk, or participant adherence. By having participants report their training plan and carefully monitoring their training practices during the 6-month lead-in phase, and using this information to pair-match participants before randomization, we believe large deviations in training practices at the intervention group level are unlikely. Any differences will be comprehensively measured during the intervention and could be used to better understand variance in individual performances at the end of the trial. Adopting a ketogenic, low carbohydrate diet can be associated with side effects that are usually transient and mild. Common adverse effects, often referred to as “keto flu,” include light-headedness, generalized fatigue, difficulty exercising, poor sleep, and constipation. These typically resolve within days to weeks. Participants can report symptoms or raise concerns any time they feel the need using the StackTeamApp.

Intervention code [1]

Treatment: Other

Intervention code [2]

Lifestyle

Comparator / control treatment

Competitive marathon runners randomly allocated to the active control will be asked to follow their usual high carbohydrate diet (20% fat, 20% protein, and 60% carbohydrate) and usual training and competition schedule for the 6 months leading into the 2024 Sydney marathon.

Control group

Active

Outcomes

Primary outcome [1]

Marathon race time is the primary outcome for this clinical trial. This outcome will be assessed via audit of the official marathon results.

Timepoint [1]

Marathon race performances will be measured in a real-world setting of the 2023 (baseline) and 2024 (post intervention) September Sydney Marathon events. The Sydney marathon is a World Athletics Gold Label Marathon.

Secondary outcome [1]

Maximal aerobic oxygen consumption (VO2max) measured as absolute L/min and relative mL/kg/min will be assessed in a standard graded treadmill running test using indirect calorimetry breath-by-breath gas analysis (QUARK CPET; COSMED, Rome, Italy).

Timepoint [1]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [2]

Exercise fuel utilization (%fat vs %carbohydrate expressed as the respiratory quotient, a value between 0.7 and 1.0) using indirect calorimetry breath-by-breath gas analysis (QUARK CPET; COSMED, Rome, Italy) will be measured at four different running speeds corresponding to the speed of individually determined race pace and speeds above and below anaerobic threshold.

Timepoint [2]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [3]

Running economy (oxygen cost per km run per kg of body weight) using indirect calorimetry breath-by-breath gas analysis (QUARK CPET; COSMED, Rome, Italy) will be measured at four different running speeds corresponding to the speed of individually determined race pace and speeds above and below anaerobic threshold.

Timepoint [3]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [4]

Body weight measured using digital scales

Timepoint [4]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [5]

Whole-body composition including estimates of fat and lean mass will be measured using dual-energy X-ray absorptiometry (DXA; Medilink Medix DR, 2D-Fan beam, Montpellier, France).

Timepoint [5]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [6]

4-compartment model of body composition that combines DXA measurements with bioelectrical impedance measurements to provide an estimate of muscle protein (i.e. lean mass) that is independent of total body water.

Timepoint [6]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [7]

Comprehensive analysis of blood markers of interest. A finger prick blood sample (fingerstick method) will be collected in Lithium Heparin capillary tubes and used to perform a Lipid Panel analysis (measures total cholesterol (CHOL), high-density lipoprotein cholesterol (HDL), triglycerides (TRIG), low-density lipoprotein cholesterol (LDL), very low-density lipoprotein cholesterol (VLDL), alanine aminotransferase (ALT), aspartate aminotransferase (AST)) and a Comprehensive Metabolic Panel analysis (measures alanine aminotransferase (ALT), albumin, alkaline phosphatase (ALP), aspartate aminotransferase (AST), calcium, chloride, creatinine, glucose, potassium, sodium, total bilirubin, total carbon dioxide, total protein, and blood urea nitrogen (BUN)) using Piccolo Xpress blood analyser.

Timepoint [7]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [8]

Finger prick blood analysis during treadmill running tests. Blood glucose will be measured at rest, in recovery, and following each of the 4 exercise stages in running treadmill economy / VO2max test.

Timepoint [8]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [9]

Glucose kinetics measured using continuous glucose monitors. Abbott Freestyle Libre Pro IQ continuous glucose monitors will be attached to the back of the upper arm using the manufacturer's sensor applicator tool and instructions. Each sensor will be worn for up to 14 days. Sensors are scanned by a reader to wirelessly download the glucose data. Participants are not required to regularly download the data and the system also blinds participants to the glucose readings during the study.

Timepoint [9]

Assessed at baseline (in the 2 weeks leading up to and including the 2023 Sydney marathon), at diet randomization (in the 1 week leading up to diet randomization and during the first week after diet randomization at 6 months post-baseline), at post-completion of intervention (in the 2 weeks leading up to and including the 2024 Sydney marathon at 12 months post-baseline).

Secondary outcome [10]

Resting heart rate measured using a chest strap.

Timepoint [10]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [11]

Anaerobic threshold as a measure of intensity corresponding to a percentage of maximal aerobic capacity will be assessed in a standard graded treadmill running test using indirect calorimetry breath-by-breath gas analysis (QUARK CPET; COSMED, Rome, Italy).

Timepoint [11]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [12]

Finger prick blood analysis during treadmill running tests. Blood ketones will be measured at rest, in recovery, and following each of the 4 exercise stages in running treadmill economy / VO2max test.

Timepoint [12]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [13]

Finger prick blood analysis during treadmill running tests. Blood lactate will be measured at rest, in recovery, and following each of the 4 exercise stages in running treadmill economy / VO2max test.

Timepoint [13]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [14]

Resting blood pressure measured using an automatic sphygmomanometer.

Timepoint [14]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Secondary outcome [15]

Exercise heart rate measured using a chest strap.

Timepoint [15]

Assessed at baseline (at the time of the 2023 Sydney Marathon), at pre-diet randomization (6 months post-baseline), at 6 weeks post-diet randomization, at post-completion of intervention (12 months post-baseline).

Eligibility

Key inclusion criteria

Competitive marathon runners will be recruited from running clubs in the greater Sydney region in Australia.

To be considered ‘competitive’, runners must have previously competed in a marathon race with evidence of running a sub 4-hour marathon time (recorded within the past 3 years), be currently completing > 3 h/week training, and have > 2 years training experience.

At baseline, runners must be following Australian dietary guidelines for healthy eating which is considered a high carbohydrate diet.

Minimum age

18 Years

Maximum age

65 Years

Sex

Both males and females

Can healthy volunteers participate?

Yes

Key exclusion criteria

Unable or unwilling to be randomized to either diet group for the 6 month diet intervention.

Unable or unwilling to enter and compete in the 2023 and 2024 Blackmores Sydney marathon.

Unable to participate in their usual training and competition schedules due to an injury or chronic illness at the study baseline.

Currently, or within the previous 12 months, consuming a low carbohydrate, high fat diet.

Unable or unwilling to attend laboratory-based testing sessions at the university laboratories located in Blacktown, NSW.

Study design

Statistical methods / analysis

Sample Size
The required study sample size has been calculated by a biostatistician using the primary outcome measure of marathon running performance (time) for a RCT design with analysis of covariance. We obtained the finishing times for all runners in the 2022 Sydney marathon. We set an upper bound cut-point of eligible study participants needing a sub-4 hour finishing time. Only a small proportion (54 runners) of the total 3445 race finishers had a time less than 2 hours and 50 minutes so we set this as the lower bound time for runners we can reasonably expect to recruit to the study (this would not exclude runners with times faster than this from participating in the study). This produced an average running time of 3 hours and 35 minutes with a standard deviation of 17.5 minutes for the target population. In the 2022 Sydney marathon, 1228 runners had a finishing time of 4 hours or less (i.e., would meet our study inclusion criteria). Using publicly available data, we estimated that approximately 93% of these participants were Australian (1142 Australian runners), we could not however, determine how many reside within the greater Sydney region and would be able to participate in the study, but we expected this to be the majority. We assumed experienced competitive marathon runners would have a high pre-post-race time correlation. Assuming a pre-post correlation of 0.8 and a power of 0.8 and alpha of 0.05, we calculate that a total of 38 participants (19 per group) will allow us to detect a performance difference between groups of 10 minutes. A change of 10 minutes equates to a medium to large effect size. Diet intervention studies typically have a high loss to follow-up. We will assume a 30% attrition rate and over-recruit by an additional 6 participants per group for a total of 50 participants in the study. Based on the above, we estimated that using the sample size (n = 50) we are aiming to recruit approximately 5% of the eligible population.

Main analysis
IBM Statistics SPSS 24 (Illinois, Chicago, USA) will be used for statistical analyses. Independent sample t-tests or Mann Whitney U tests (where data are not normally distributed) will be used to determine differences between usual high carbohydrate, and K-LCKD diet groups at baseline, with the alpha level for significance set at P < 0.05. Effects for each group will be analysed using ANCOVA, with pre-intervention measures included as covariates. As a measure of effect size, partial eta-squared (np2) will be used. Effect sizes were determined as: np2 = 0.01 (small effect), np2 = 0.09 (medium effect), and np2 = 0.25 (large effect) (Cohen 1988). Paired samples t-tests or Wilcoxon signed ranks test (where data are not normally distributed) will be used to examine changes over time within groups, if ANCOVA P value is < 0.05.

Outliers
We will check all data for implausibly low or high measurements which will be excluded if they fall outside of ±4SD of the mean.

Missing data
We will assume missing covariate and moderator data to be missing at random or completely at random. We will impute missing values using multiple imputation by chained equations. Participants with missing primary outcome data will be excluded from main analyses.

Recruitment

Recruitment status

Not yet recruiting

Date of first participant enrolment

Anticipated

7/08/2023

Actual

Date of last participant enrolment

Anticipated

15/09/2023

Actual

Date of last data collection

Anticipated

30/09/2024

Actual

Sample size

Target

50

Accrual to date

Final

Recruitment in Australia

Recruitment state(s)

NSW

Funding & Sponsors

Funding source category [1]

University

Name [1]

Australian Catholic University

Country [1]

Australia

Primary sponsor type

University

Name

Australian Catholic University

Country

Australia

Secondary sponsor category [1]

None

Name [1]

Country [1]

Ethics approval

Ethics application status

Submitted, not yet approved

Ethics committee name [1]

ACU Human Research Ethics Committee (HREC)

Ethics committee address [1]

Australian Catholic University
North Sydney Campus
PO Box 968
NORTH SYDNEY, NSW 2059

Ethics committee country [1]

Australia

Date submitted for ethics approval [1]

06/04/2023

Approval date [1]

Ethics approval number [1]

Summary

Brief summary

Dietary macronutrient composition (amounts of fat and carbohydrate) can be manipulated to manage body composition and alter fuel availability and metabolic energy production for the optimisation of athletic performance. A high carbohydrate intake (>60% of daily energy intake) is generally recommended as an optimal diet and fuelling strategy for most athletic pursuits. However, there is growing interest in the potential performance benefits of high fat dietary approaches such as ketogenic-low carbohydrate high fat (K-LCHF, carbohydrate <10% of daily energy intake) diets for endurance-based activities. 
Performance in endurance-based activities is underpinned by a high maximal oxygen consumption (VO2MAX), the ability to sustain work at a high percentage of VO2MAX, and good economy or efficiency of movement (oxygen cost at a given velocity). Diet and competition fuelling choices are important determinants of endurance performance by directly or indirectly affecting these physiological traits, along with body composition, fuel substrate utilization (burning fat or glucose), recovery, limiting the negative effects of gastrointestinal symptoms (runners’ gut) from high carbohydrate intake, and the ability of athletes to avoid fatigue-inducing hypoglycaemia (colloquially know as ‘bonking’ or ‘hitting the wall’). 
Many athletes, coaches, and practitioners continue to experiment with K-LCHF diets to determine whether performance benefits exist. Despite the popularity of ketogenic low carbohydrate diets and some good evidence supporting their theoretical benefits for performance and health, we currently only have data from a few scientific studies. These studies are generally small (underpowered to draw conclusions) and are relatively short duration. In addition, participants in most previous studies of K-LCHF diets have self-selected their diet intervention group. This introduces confounders that could bias conclusions. Finally, very few studies of K-LCHF diets have used real-world measures of endurance performance because contrived race distances and intensities are typically easier to manage in research settings. 
To address the limitations in previous studies and the need to offer stakeholders high-quality research to guide evolving practices, we aim to conduct a randomized controlled trial to assess the impact of a chronic (6 month) ketogenic low carbohydrate, high fat diet in competitive marathon runners. Our primary endpoint will be real-world race performance as measured in the Sydney Marathon. Secondary endpoints will be measures of exercise metabolism, running economy, body composition, blood lipid profiles, and glucose kinetics during diet adaptation, training and competition.

Trial website

Public notes

Contacts

Principal investigator

Name

Dr Tim Hartwig

Address

Australian Catholic University
Level 2, 163-167 Albert Rd, Strathfield, NSW 2135
Edward Clancy Building, Office 640.1.29

Country

Australia

Phone

+61297014355

Email

[email protected]

Contact person for public queries

Name

Tim Hartwig

Address

Australian Catholic University
Level 2, 163-167 Albert Rd, Strathfield, NSW 2135
Edward Clancy Building, Office 640.1.29

Country

Australia

Phone

+61297014355

Email

[email protected]

Contact person for scientific queries

Name

Tim Hartwig

Address

Australian Catholic University
Level 2, 163-167 Albert Rd, Strathfield, NSW 2135
Edward Clancy Building, Office 640.1.29

Country

Australia

Phone

+61297014355

Email

[email protected]

Trial Review

Documents added manually

Documents added automatically