Impact Loading and Nutrition in Cyclists: A Clinical Intervention Examining Energy Availability and Bone Mineral Density
- Regina Hammond1, Margaret M. Harris1, Craig C. Elder1, Nanna L. Meyer,
- FACSM1, 2 ; Thesis directed by Assistant Professor Nanna L. Meyer
- University of Colorado | Colorado Springs, CO
- United States Olympic Committee, Colorado Springs, CO
ACSM ANNUAL CONFERENCE
This thesis and research was presented during the Energy Balance thematic poster session at the 60th American College of Sports Medicine Annual Meeting, June 1, 2013. The abstract was published in Medicine and Science in Sports and Exercise, Volume 45:5
My Role and Original Contributions
I entered this graduate research program with more than a decade of prior professional experience, along with practical experience as an endurance athlete, triathlon coach, and early user of TrainingPeaks for athlete programming. That background allowed me to approach the study not only as a graduate researcher, but also as someone who understood how training data, nutrition, athlete behavior, and project execution had to work together.
I conceived and developed the study’s individualized, periodized-nutrition system. At a time when this type of integrated nutrition-planning workflow was not yet common practice, I created a method for translating completed training data into athlete-specific energy and fueling recommendations.
My Contributions Included
- Developing the calculations used to estimate exercise energy expenditure and energy availability.
- Reviewing completed TrainingPeaks workouts alongside heart-rate data, SRM power files, training duration, intensity, distance, and kilojoule expenditure.
- Creating periodized calorie targets for easy, moderate, and hard training days.
- Designing individualized meal plans based on each cyclist’s energy availability, fat-free mass, training demands, dietary intake, and nutrition-related laboratory findings.
- Developing athlete-specific fueling recommendations for before, during, and after training.
- Incorporating menstrual-cycle tracking data when evaluating female athletes.
- Analyzing longitudinal biochemical data, including 25(OH)D and IGF-1, along with pre- and post-intervention CBC, testosterone, estradiol, ferritin, and cortisol results.
- Integrating biochemical findings with DXA measurements, body-composition data, resting metabolic rate, dietary intake, training load, and restrained-eating assessments.
- Creating study calendars, schedules, participant workflows, and organizational systems needed to coordinate data collection and maintain continuity across research phases.
This combination of study design, athlete monitoring, biochemical analysis, individualized nutrition planning, and research-project management was central to implementing the intervention and translating the collected data into practical recommendations for competitive cyclists.
RESEARCH CONTEXT: ENERGY AVAILABILITY BEFORE RED-S
This research was conducted before the International Olympic Committee introduced the Relative Energy Deficiency in Sport—or RED-S—framework. At that time, the primary energy-availability thresholds were based on Anne B. Loucks’s controlled research in regularly menstruating women.
Loucks examined energy availability at 10, 20, 30, and 45 kcal/kg FFM/day and demonstrated disruption of reproductive and metabolic hormone function below approximately 30 kcal/kg FFM/day. These values formed the scientific basis of our nutrition intervention and were applied to both the female and male cyclists.
One of the study’s most significant findings was that the male cyclists demonstrated greater energy restriction than the female cyclists. This was determined by comparing average weekly energy intake with estimated total energy expenditure and calculating energy availability relative to fat-free mass.
The findings provided early evidence that low energy availability and its potential effects on bone and metabolic health were not limited to female athletes.
The study preceded the IOC’s 2014 introduction of RED-S, which expanded upon the Female Athlete Triad model and formally recognized that male athletes can also experience the physiological and performance consequences of inadequate energy availability.
INTRODUCTION
Areal bone mineral density (aBMD) is low in endurance athletes, especially in cyclists. At the time of this study, emerging longitudinal evidence also indicated that cyclists could lose bone mass over the course of a competitive racing season.
PURPOSE
To assess changes in areal bone mineral density (aBMD) over four months in elite cyclists and examine those changes in relation to energy availability, training volume, and biochemical markers associated with bone metabolism.
The four-month study included:
- Assessment of training volume and exercise energy expenditure
- Calculation and monitoring of energy availability
- Individualized nutrition planning
- Impact-loading through a progressive jumping program
- Assessment of nutrient status and biochemical markers affecting bone metabolism
- Baseline and post-intervention measurement of aBMD
METHODS
METHODS
Subjects:
-Six elite male (42±8y; ht: 1.77±0.05m; wt: 71.8±5.7kg; %body fat [BF]: 11.6±3.2; FFM: 63.65±4.02kg) road and mountain bike cyclists who rode a minimum of 15 hrs per week
-Four elite female (39±10y ht: 1.65±0.06m; wt: 62.75±12.84; %BF: 25.0±9.3; FFM: 46.5±4.29kg) road and mountain bike cyclists who rode a minimum of 15 hrs per week
Qualifying Criteria: a BMD Z-score of <0 (DXA) measured on a Lunar iDXA, were recruited from Front Range towns near Denver, Colorado, including: Colorado Springs, Boulder, and Castle Rock.
INTERVENTION
INTERVENTION
The intervention consisted of:
- A progressive jumping program (using a Vertec) with 10 maximal vertical jumps performed 5 days/week for 4 months
- Custom calculated meal plan recommendations to increase energy availability > 30 kcal⋅kgFFM-1⋅d-1, and
- Recommendations for calcium and vitamin D supplementation based on blood draw results at start of study.
- Vertical jump height and aBMD were assessed at baseline and post-intervention.
- Restrained eating (RE) was assessed once using the Three Factor Eating Questionnaire
- Energy Availability was calculated monthly as data from training sessions were analyzed. Workouts within Training Peaks was analzyed based on Heart Rate, Power Output, Distance, and kJ.
- Venous blood samples were collected at Baseline, Month 1 & Month 4:
- 25(OH)D
- IGF-1
- Venous blood samples were collected Pre and Post intervention:
- CBC, Total Testosterone, Total Estradiol, Serum Ferritin, Cortisol
RES
RESULTS
RESULTS
Repeated measures ANOVA revealed no significant differences in wt, %BF, FFM, or vertical jump from pre to post-intervention overall and by gender.
Overall, aBMD did not change significantly (whole body: -0.5±1.1%; spine: -0.2±2.2%; femur -0.5±1.1%; left femur -0.7±1.5%; right femur -0.3±1.24%) nor were there differences by gender, although females seemed to gain aBMD, while males seemed to lose aBMD.
Despite nutritional intervention, average EA remained low (EA; males 21.2±15.97 kcal⋅kgFFM-1⋅d-1; females 26.1±7.7 kcal⋅kgFFM-1⋅d-1) and RE high (males 12±4; females 14±5).
No significant differences were detected for serum 25(OH) vitamin D at baseline, 2 and 4 months (pre: 39.1±7.47ng/mL, 2-month: 41.0±5.45 ng/mL; 4-months: 35.6±9.99ng/mL).
CONCLUSION
CONCLUSION
In conclusion, data show that a 4-month clinical intervention using high-impact jumping and nutrition is, for the most part, insufficient to prevent bone loss in cyclists.
Changes in Bone Mineral Density
GE Lunar iDXA Measure Locations: Whole Body, Lumbar Spine, Proximal Femur, Right/Left Femoral Neck, BMC (Bone Mineral Content), Fat Free Mass, Lean Tissue Mass, Body Fat Percentage
Among cyclists who improved vertical jump height, bone mineral density remained relatively stable across most regions over the four-month intervention. Females showed small increases in several regions, while males showed small declines; however, none of these changes were statistically significant.
Intervention: Custom Meal Plans
Individualized meal plans were developed following a comprehensive review of each athlete’s completed TrainingPeaks workouts, heart-rate and SRM power files, dietary intake, and—for female athletes—menstrual-cycle tracking data.
Each Meal Plan Included:
- 6 plans for breakfast, lunch, dinner
- Total kcal/day for meals based on 30 kcal*kgFFM
- 2 plans for easy, moderate, hard training days
- Custom fueling guidelines created for each athlete specific to before, during, after training based on duration, intensity, and mode of exercise
Table 3: Individualized calorie recommendations changed according to each athlete’s training intensity. Comparing actual intake with estimated needs revealed substantial differences between athletes, with several cyclists remaining below their recommended intake or target energy availability despite individualized meal planning.
- Periodized Calorie Goals according to Intensity of Training (duration, RPE, Power, HR), [Easy, Moderate, Hard]
- Actual Intake (AI)
- Goal Energy Availability (EA)
- Difference: Actual Intake – Energy Availability
- Difference: Recommended Intake – Actual Intake
Jumping Study Phase 2: Individualized Nutrition Intervention
Example of a periodized nutrition plan developed for one cyclist. Training type, duration, heart rate, resting metabolic rate, and estimated exercise energy expenditure were used to establish calorie targets and meal-and-snack recommendations for recovery, easy, moderate, and hard training days.
Figure 6. Average Energy Availability per Subject per Month
- Energy availablity energy balance in elite cyclists
- Calculations for Energy Availability
EA = energy intake – exercise energy expenditure
REE = 500 +22*kgFFM*PAL/1440
EEE = [(500+22*kgFFM/1440)*MET*minutes of exercise)]- (REE*minutes of exercise)
RMR measured via indirect calorimetry
MRI = based on EA of 30 kcal*kgFFM
Outcomes
- Average energy availability remained below the study target of 30 kcal/kg FFM/day during every month of the intervention.Average energy availability remained below the study target of 30 kcal/kg FFM/day
- Individual results varied considerably: some athletes periodically achieved adequate energy availability, while others continued to demonstrate significant underfueling
Serum 25-Hydroxyvitamin D [25(OH)D]
Serum 25(OH)D was measured at baseline, midway through the intervention, and after its completion. Vitamin D supplementation recommendations were individualized according to baseline laboratory results.
Mean serum 25(OH)D increased slightly from 39.1 ng/mL at baseline to 41.0 ng/mL at two months, then declined to 35.6 ng/mL at four months. These changes were not statistically significant. The final value remained above the study’s insufficiency threshold of 30 ng/mL but below the 40 ng/mL threshold identified as sufficient. Results should be interpreted in the context of the smaller post-intervention sample, which decreased from 10 to six participants.
KEY SCIENTIFIC FINDINGS
At the time, and to our knowledge, this was the first longitudinal study of bone mineral density in competitive male and female cyclists to combine three intervention strategies: high-impact bone loading, individualized nutrition planning to improve energy availability, and calcium plus vitamin D supplementation.
The intervention did not produce a statistically significant improvement in BMD. Females showed small, non-significant increases in six of the eight measured regions, while males generally demonstrated greater declines. Improvements in vertical jump height and adherence to the jumping protocol were not consistently associated with improvements in BMD.
Calcium and vitamin D remained important components of the investigation because of their roles in bone metabolism. Previous research in male cyclists documented significant calcium losses through sweat and lower baseline BMD among athletes with greater dermal calcium losses. In the current study, calcium and vitamin D recommendations were individualized according to laboratory results. Mean serum 25(OH)D did not change significantly during the intervention, although five cyclists demonstrated small, non-significant increases in spinal BMD.
Energy availability also remained below the study target despite individualized nutrition intervention. Male cyclists demonstrated greater energy restriction than female cyclists when average weekly intake was compared with estimated total energy expenditure. The inability to recover BMD—particularly among the male cyclists—suggested that high-impact exercise and supplementation alone might be insufficient when inadequate energy availability persisted.
Because only 10 cyclists completed both baseline and post-intervention DXA measurements, the study may have been underpowered to detect statistically significant changes. Nevertheless, the findings identified important relationships among cycling volume, energy availability, biochemical markers, calcium and vitamin D status, and skeletal health that warranted further investigation.
