Resistance Training Is Associated With Higher Lumbar Spine and Hip Bone Mineral Density in Competitive Male Cyclists

Abstract

Mathis, SL, and Caputo, JL. Resistance training is associated with higher lumbar spine and hip bone mineral density in competitive male cyclists. J Strength Cond Res 32(1): 274–279, 2018—In contrast to other athletic activities, competitive cycling is associated with low areal bone mineral density (BMD). This investigation evaluated training and dietary factors associated with BMD in a sample of competitive male cyclists. Lumbar spine, total hip, femoral neck, and femoral trochanter BMD were measured with dual X-ray absorptiometry in 40 cyclists at the start of a cycling season. Participants reported weekly hours of cycling and weight training, years of competitive cycling experience, and completed a 1-day dietary recall to allow estimation of dietary calcium intake. Multivariable linear regression analyses revealed that weight training was associated with higher BMD of the lumbar spine (β = 0.001, t = 2.88, p = 0.01), hip (β = 0.002, t = 4.95, p < 0.001), femoral neck (β = 0.002, t = 5.31, p < 0.001), and femoral trochanter (β = 0.002, t = 4.31, p < 0.001). No other factor was a predictor of preseason BMD in this sample. These data emphasize the need for competitive male cyclists to participate in weight training to maintain or increase bone mass of the lumbar spine and hip.

Introduction

Physical activity is considered a modality to improve bone health and to decrease the risk of osteoporosis (^7,13). However, competitive cycling is associated with low bone mineral density (BMD; ^8,19,22). In fact, osteopenia is considered a health risk in professional and amateur male cyclists (^8,17,18,23).

Studies documenting low BMD in cyclists are numerous (^{6,16,17,20,23,28}). Although athletes typically have higher BMD values than sedentary individuals (⁴), BMD of cyclists is lower than that of sedentary controls in the lumbar spine and total hip (^20,26). Reports of lower BMD in competitive cyclists compared with age-matched controls are found in adult male cyclists (²⁶), professional cyclists (^6,17), male master cyclists with years of experience exclusively in cycling training (²⁰), and in postpubertal boy cyclists (^21,24). A trend of bone loss across a single competitive season in male (^3,15) and female (²⁵) road cyclists has also been reported.

Bone mass is determined by 2 counteracting metabolic processes, formation and resorption. A balance in the ratio of bone deposition to bone resorption leads to maintenance of BMD, whereas disruption to this balance may lead to low BMD. Untreated, low BMD may lead to osteopenia and eventually osteoporosis. Osteoporosis is characterized by low bone mineral content and a compromised microarchitecture of the bone, leading to an increased susceptibility to fracture (¹⁴). Bone mineralization is enhanced by dietary consumption of calcium to maintain calcium homeostasis in the blood and by engaging in resistance and weight bearing activities (^1,2).

Low BMD in cyclists is particularly troubling because of the high prevalence of falls in competitive cycling. The purpose of this study was to identify factors significantly associated with BMD in competitive male cyclists. It was hypothesized that an increase in time spent training on a bicycle would be associated with lower BMD of the lumbar spine, total hip, femoral neck, and femoral trochanter. It was also hypothesized that an increase in weekly minutes spent lifting weights and a higher calcium intake would be associated with higher BMD of the lumbar spine, total hip, femoral neck, and femoral trochanter.

Methods

Experimental Approach to the Problem

It is well established that male endurance cyclists represent a population with an increased risk of low bone mass (¹⁹). In this study, factors that may be significantly associated with lumbar spine and hip BMD in competitive male cyclists were measured (¹⁶). Specifically, age, body mass index (BMI), and dietary calcium intake were obtained. Training specific factors assessed were cycling training load and weekly participation in weight training and run training (¹⁶). Because BMD can decrease across a single competitive season (^3,15,25), the number of years racing and training on a bike was also measured. Multivariable linear regression models were applied to determine the relationship of these factors to BMD in this sample of competitive male cyclists.

Subjects

Amateur cyclists (N = 40) were recruited for participation in this study by contacting male competitors at the beginning of a road bicycle racing season. Cyclists were 42.7 ± 9.4 years (range 31–69 years) and represented state, regional, and national competitors (including 1 former Olympian). Inclusion criteria consisted of cyclists older than 30 years, who trained regularly, and had at least 2 years of cycle-specific training experience. Exclusion criteria included the use of medication known to affect the endocrine system and bone metabolism and/or a history of endocrine disorders. Before data collection, Middle Tennessee State University Institutional Review Board approval was obtained. Participants were informed of the benefits and risks involved in study participation before signing an institution-approved informed consent document.

Procedures

Eligible participants reported to the University Exercise Science Laboratory within 2 weeks of the start of the racing season. On arrival, cyclists completed a self-report questionnaire consisting of questions pertaining to the number of weekly hours of cycling training, weekly minutes of weight training, and number of years of competitive cycling experience on a survey. Before the laboratory visit, cyclists completed a 1-day dietary recall, performed on a day representative of the cyclist's typical diet, in order that calcium intake could be estimated. Verbal and written instructions were provided to cyclists to assist them in an accurately determining portion size. Food intake, including vitamin supplements, was analyzed for daily calcium (mg·d⁻¹) by the principle investigator using the US Department of Agriculture website (http://www.mypyramid.gov).

Cyclists were provided with hospital scrubs and removed shoes for all laboratory measurements. Height was measured to the nearest 0.5 cm using a stadiometer (SECA model 222; seca GmbH & Co. KG, Hamburg, Germany) while standing on level ground with heels together and mass evenly distributed. After voiding, body mass was measured to the nearest 0.1 kg using a digital scale (SECA model 770; Vogel & Halke, Hamburg, Germany). These values were used to calculate BMI. Last, bone density scans were then completed in randomized order by alternating hip or spine scans first.

Bone Mineral Density

The most widely validated technique to measure BMD is dual energy X-ray absorptiometry (DXA). Lumbar spine (L1–L4), total hip, femoral neck, and trochanteric region BMD were measured using a Hologic Discovery QDR series DXA (Hologic, Bedford, MA, USA) at the start of a competitive cycling season. Standard manufacturer protocols were followed and scans were conducted and analyzed by a licensed technician. The DXA was calibrated using a spine phantom before data collection, assuring that all calibrations fell within appropriate ranges. Data were recorded as g·cm⁻² Z-scores, and T-scores. Femoral neck and trochanter scans were performed on the cyclists nondominant leg, described as the opposite leg used to hypothetically kick a ball. The hip scan included the femoral neck and the trochanteric region. To avoid false images on the scan, the cyclists wore hospital scrubs during the scans and removed all metal objects from their bodies.

Statistical Analyses

Anthropometric measurements, participant demographics, and BMD were reported as mean and SD. The assumptions required for parametric analyses were examined for all continuous variables using the Shapiro-Wilk test of normality. Because data are nonparametric, Spearman rank correlations were computed to determine correlations among age, BMI, calcium intake, cycle training, weight training, run training, years of cycling experience, number of races per year, and the 4 BMD sites. Separate bivariate linear regression analyses were performed to examine the relationship between all variables and BMD for each site. To assess the association between cyclist characteristics and BMD of lumbar spine, total hip, femoral neck, and femoral trochanter, independent variables that were significant at p ≤ 0.05 in the bivariate analyses or were considered to be relevant to the outcome measures were entered into multivariable linear regression models. Statistical significance was set at 0.05 for all analyses. Data were analyzed using Stata statistical analysis software (Version 14.0; StataCorp, College Station, TX, USA).

Results

A total of 40 cyclists participated in the study. Participant characteristics and training characteristics are presented in Table 1. Bone density measurements are presented in Table 2. Mean BMD of the lumbar spine, hip, femoral neck, and femoral trochanter were 1.028 g·cm⁻² (SD = 0.120), 0.965 g·cm⁻² (SD = 0.111), 0.807 g·cm⁻² (SD = 0.114), and 0.734 g·cm⁻² (SD = 0.096), respectively. The mean T-scores of the lumbar spine, total hip, femoral neck, and femoral trochanter of the cyclists aged 50 years or older were −0.8, −0.9, −1.4, and −0.5, respectively. For reference, the criterion used to classify low bone density, the precursor to osteoporosis, is a T-score that lies between −2.5 and −1.0, and osteoporosis is classified as a T-score less than −2.5 for men 50 years of age or older. Of the cyclists younger than 50 years, the lumbar spine BMD Z-scores ranged from −2.8 to 2.0. Also, 16% of cyclists' lumbar spine BMD met the criteria for “below the expected range for age” according to the International Society for Clinical Densitometry.

Table 1.: Baseline descriptive characteristics among male cyclists (N = 40).*

Table 2.: Baseline bone density among male cyclists (N = 40).

Spearman correlations are presented in Table 3. A strong positive correlation was noted between the number of minutes of weekly weight training and BMD of lumbar spine, total hip, femoral neck, and femoral trochanter. There was a negative correlation between the number of years of cycling experience and BMD of the femoral neck. The number of minutes spent running was not correlated with any BMD site measured. There was also a negative correlation between age and the number of minutes of weekly weight training.

Table 3.: Correlations between bone mineral density (BMD) and cyclist characteristics (N = 40).*

Bivariate linear regression analyses revealed that an increase in age was associated with lower BMD of total hip (β = −0.004, t = −2.27, p = 0.03, R² = 0.09) and femoral neck (β = −0.004, t = −2.18, p = 0.04, R² = 0.09). Calcium intake approached a significant positive association with higher BMD of the lumbar spine (β = 0.00007, t = 1.93, p = 0.06, R² = 0.07). Years of cycling experience was associated with lower BMD of the femoral neck (β = −0.005, t = −2.44, p = 0.02, R² = 0.11). The number of minutes spent lifting weights weekly was associated with a higher BMD of the lumbar spine (β = 0.002, t = 4.14, p < 0.001, R² = 0.29), total hip (β = 0.002, t = 5.85, p < 0.001, R² = 0.46), femoral neck (β = 0.002, t = 6.83, p < 0.001, R² = 0.56), and femoral trochanter (β = 0.001, t = 4.86, p < 0.001, R² = 0.37).

The variables significant at a level of 0.05 in the bivariate linear regression analyses were entered into the multivariable models. Age was included in the model because BMD decreases as age increases. Years of cycling experience was not included in the multivariable model because it was significantly correlated with age. Multivariable linear regression analyses of age, calcium intake, weekly minutes cycling training, and weekly minutes of weight training with the 4 BMD sites revealed that weight training was the only variable significantly and positively associated with higher BMD of the lumbar spine (β = 0.001, t = 2.88, p = 0.007, R² = 0.25), total hip (β = 0.002, t = 4.95, p < 0.001, R² = 0.47), femoral neck (β = 0.002, t = 5.31, p < 0.001, R² = 0.53), and femoral trochanter (β = 0.002, t = 4.31, p < 0.001, R² = 0.32). As the number of weekly minutes of weight training increases, the BMD of these 4 sites will also increase. See Table 4.

Table 4.: Multivariable linear regression analyses of age, calcium intake, cycling training, and weight training with bone mineral density (BMD) sites.*

Discussion

Hypothesized causes of low BMD in male cyclists include the lack of ground reaction force associated with the non–weight bearing activity, hormonal contributions, and dermal calcium loss through hours of excessive sweating during training and racing (^2,3,9,11,12). The discovery of pathways to improve bone health in this population is especially important because of the high prevalence of falls. Although fracture rate in cyclists is not well documented, falls are likely to occur. A fall at a high speed could cause any bone to fracture, but cyclists with low BMD may have much more serious injuries. The nature of the sport of competitive road cycling coupled with low BMD in the participants places these athletes at risk for fracture (²⁶). Therefore, there is a need to increase bone density in this population. The main finding of this study was that weight training is associated with a higher BMD of the lumbar spine, hip, femoral neck, and femoral trochanter. Secondary findings include the following: (a) Dietary calcium intake may present a minor contribution to BMD of the lumbar spine; (b) The volume of weekly bicycling was not associated with a lower BMD. Although these data lend evidence to support these conclusions, it is important to note that the limitations of the self-report assessment tools used to measure calcium intake and training volume. Future research with more robust measurements of dietary contributions to bone health is needed in this population. From these findings, it can be conjectured that cyclists who participate in weight training may offset the deleterious effects on BMD that occurs with time spent in a non–weight bearing activity.

The rate of bone loss that naturally occurs with aging may have been accelerated within this sample. To illustrate this, 8 of the 13 cyclists approaching or older than 50 years (older than 45 years) had BMD T-scores below −1.0. Also, the cyclists in the current sample reported participation in a range of 8–40 races per season while training 8–17 hours per week. The combination of a relatively high training load with the effect of a non–weight bearing activity on bone loss places importance on the discovery of a method to offset this trend. Also, the location of bone loss is especially alarming considering the impact to the femoral neck when one falls off of a bicycle. This finding highlights the importance of bone density screening for cyclists while including a training regimen to improve bone density to offset the potential negative effect of competitive cycling.

Daily dietary calcium intake only accounted for a small contribution to BMD of the lumbar spine. This finding confirms the results of Barry and Kohrt (³) and Mathis et al. (¹⁵) who found that the intake of a calcium supplement did not improve BMD. Barry and Kohrt (³) found that a calcium supplement provided at mealtime did not affect BMD of the same sites. Mathis et al. (¹⁵) found that a calcium supplement ingested during training and racing also did not improve bone health. In both of these studies, the authors reported a decline in BMD over the course of a competitive season. It has also been reported that cyclists do not consume enough calories, especially carbohydrates, to offset the energy expenditure of training and racing (²⁷). Cyclists reportedly restrained caloric intake to control body weight. Calcium-rich foods are also calorically dense and may be eliminated from the diet for weight management. Cyclists who chronically restrict food intake are at risk of suppressing bone formation that may further harm bone health (¹⁰). Conversely, an immediate effect of calcium provided during exercise attenuated the rise in biomarkers of bone turnover during a cycling trial (¹). Further investigation on the effects of increased dietary and/or supplemental calcium on BMD is needed with competitive cyclists.

An important finding from the current sample is the positive relationship between the number of minutes spent weight training and BMD of the lumbar spine, total hip, femoral neck, and femoral trochanter, respectively. Anecdotally, cyclists do not engage in weight bearing activity such as running or weight lifting, as it has been shown to have little influence on increasing aerobic power output during cycling (⁵). Cyclists reported training on the bicycle up to 17 hours per week. Anecdotally, recovery time is typically spent in a seated or reclining position, further compounding time spent without bone loading. In another investigation, cyclists who engaged in weight lifting throughout the year (2–6 months) still had lumbar spine BMD lower than controls matched by age and weight-lifting activity (²⁶). Although there is a lack of research involving competitive cyclists in a weight training intervention aimed to increase bone density, data from the current study suggest potential benefits. Therefore, it is recommended that athletes in non–weight bearing sports, such as bicycling, should lift weights consistently throughout the year to maintain or increase BMD.

Characteristics of a cyclist's typical training regimen were discovered in the correlation analyses. A negative correlation between weight training and years of cycling experience was noted. This negative correlation indicates that experienced cyclists may need encouragement to engage in weight training. There was a negative correlation between run training and years of cycling, as well as run training and the number of races per year. As weight bearing exercise may improve bone health, cyclists should be encouraged to engage in running during the off-season to offset the harmful effect of cycling during the competitive season. Last, a negative correlation between years of cycling experience and BMD of the femoral neck was found. However, this relationship may be confounded by the effect of increasing age on BMD in general.

Limitations to the study include reliance on the self-report assessments of cyclists weekly cycling training, weight training, and run training. The number of races per season was also a self-reported estimate based on the cyclists plan for the upcoming racing season. Calcium intake was based on self-report dietary recall for a 24-hour period and analyzed using the US Department of Agriculture website and may not be representative of the cyclists’ actual calcium intake. The reporting of dietary data did not include information on energy availability, a dietary factor relevant to cyclists because of the potential risk of developing low energy availability that is known to negatively affect bone mass. The current data lend evidence to support the need to confirm current findings suggesting that dietary calcium intake, energy availability, and other nutrients such as vitamin D will promote bone health.

Practical Applications

Although this analysis is not indicative of a cause of declining BMD in male cyclists, it provides information to coaches and health care providers on the characteristics of cyclists typical training regimen that may affect bone health. Coaches should pay close attention to the occurrence and prevalence of bone fractures and encourage male cyclists to monitor BMD throughout their racing career. Precautions should be taken by cyclists to determine the current state of their bone health. Health care professionals and coaches are encouraged to prescribe a weight-training regimen to male competitive cyclists to help protect their bone health.