Mechanical loading of the skeleton by weight-bearing exercise is a key component of bone health throughout life (Santos et al., 2017; Warden et al., 2022). Non-weight-bearing exercise, such as cycling, has no strong benefits for bone health (Olmedillas et al., 2012). In fact, road-race cycling seems to be detrimental for bone tissue, as the greater part of elite road-race cyclists is characterized by low bone mineral density (BMD; Campion et al., 2010; Hilkens et al., 2023; Medelli et al., 2009). Importantly, low BMD in elite road-race cyclists is associated with higher fracture incidence during the active cycling career (Hilkens et al., 2023). Furthermore, low BMD at a young age is a predictor of osteoporosis at a more advanced age (Weaver et al., 2016). Altogether, this emphasizes the need to prevent and treat low BMD in elite road-race cyclists.
The cause of low BMD in elite road-race cyclists is likely multifactorial. Besides performing extreme amounts of non-weight-bearing cycling exercise (Hilkens et al., 2023; Plasqui et al., 2018), elite road-race cyclists are at risk of low energy availability (Keay et al., 2018) and generally have a low body weight (Hilkens et al., 2023), all of which are factors that have been linked to low BMD (Hilkens et al., 2023). Also, a history with little bone-specific exercise appears to predict low BMD in this population (Hilkens et al., 2023). In contrast, chronic exercise with a high mechanical load can promote BMD, at least in the general population (Howe et al., 2011; Kistler-Fischbacher et al., 2021; Kohrt et al., 2004; Santos et al., 2017; Zhao et al., 2014). Bone appears to respond optimally to high-intensity (i.e., strain magnitude) exercise that is applied rapidly (i.e., strain rate), and is dynamic, and intermittent with multidirectional loading patterns (Kohrt et al., 2004; Turner & Robling, 2005). In this regard, high-intensity resistance exercise training (i.e., ≥80% one-repetition maximum) and impact exercise (i.e., jumping, kicking, punching) are often considered the preferred exercise modes (Beck et al., 2017; Kohrt et al., 2004). As many elite road-race cyclists are reluctant to perform high-intensity resistance exercise training (Hoon et al., 2019), frequent, short high-impact exercise sessions could prove to form a feasible intervention to increase BMD (Hutson et al., 2021). Using this approach, a significant adaptive response of the bone may be accomplished in a time-efficient manner (Hutson et al., 2021), without large energy costs or undesired gains in body or muscle mass. Previous work has shown that a 9-month jumping exercise program improves BMD in adolescent recreational cyclists (Vlachopoulos et al., 2018). However, it is unknown whether such a high-impact training program is effective and feasible for elite-level male and female road-race cyclists, who will need to perform this intervention in addition to the typical high volumes of cycling training.
In addition to exercise, nutritional factors are also known to affect bone health (Weaver et al., 2016). Dietary collagen supplementation has recently emerged as a novel strategy to augment bone collagen synthesis with exercise (Shaw et al., 2017) and to increase BMD (König et al., 2018). This may be due to the fact that the structure and function of bone is dependent on its collagen-rich extracellular matrix, with Type I collagen accounting for approximately 95% of this matrix (Hart et al., 2020). Nevertheless, it remains to be established whether collagen supplementation combined with exercise can increase BMD in elite road-race cyclists. In the current study, we assessed the impact of combined jumping exercise and collagen supplementation on BMD in elite road-race cyclists. Furthermore, we examined the effect of this intervention on trabecular bone score (TBS) and serum bone turnover markers. We hypothesized that frequent short sessions of jumping exercise performed five times weekly, combined with the ingestion of 15 g dietary collagen, can improve BMD status during the off-season period in male and female elite road-race cyclists.
Methods
Study Design and Setting
In the current 18-week open-label, randomized trial with a parallel group design, participants were randomly allocated to either an intervention group (INT) or a no-treatment control group (CON). We employed a home-based exercise and nutritional intervention, conducted during the off-season period of elite road-race cycling between October 2021 and April 2022. The testing procedures took place at the Sport and Research Center of HAN University of Applied Sciences in Nijmegen, the Netherlands. The study was approved by the Medical Ethical Committee Zuyd, the Netherlands, and conducted in accordance with the standards for the use of human participants as outlined in the most recent version of the Declaration of Helsinki. The study was registered at the Netherlands Trial Registry (https://clinicaltrialregister.nl) as NL9770.
Participants
Forty-three elite road-race cyclists (16–35 years), including eight males and 35 females, were recruited from Dutch Continental and WorldTour cycling teams and from the national talent pool (i.e., TeamNL). Medication use known to affect bone metabolism, a recent fracture (<6 months before baseline measurements) or a current musculoskeletal injury, were set as exclusion criteria. Before inclusion in the study, all potential participants were informed about the nature and possible risks of the experiment through written information and an online presentation by the coordinating researcher.
Randomization and Blinding
A computer-generated randomization list was made by an independent researcher. Block randomization was used with block sizes of 2 and 4, stratified by sex. This strategy was chosen to avoid selection bias, and to balance the number of male and female cyclists allocated to CON and INT. The researcher responsible for screening allocated each eligible participant to the next available number on entry into the trial. Both the participants and researchers were not blinded for treatment allocation, as the current study was an open-label trial.
Intervention
Exercise Program
Participants in INT completed an 18-week home-based intervention, consisting of jumping exercise and collagen supplementation five times weekly. The exercise intervention comprised short sessions (∼5 min) of jumping activities, including multidirectional hopping and vertical jumping, designed to maximize the bone adaptive response based on bone loading characteristics (Beck et al., 2017; Kohrt et al., 2004; Turner & Robling, 2005). The hopping and bounding exercise sessions involved 10 sets of 15–25 repetitions, separated by 15 s of rest, while the vertical jumping sessions consisted of three to five sets of 10 maximal bilateral vertical jumps, separated by 20–30 s of rest. To reduce the risk of injury, participants initially performed the exercises three times a week for the first 2 weeks, with the volume and frequency gradually increasing to five times per week. The exercise selection and sequence were modified every 2 weeks to ensure novel stimuli and to enhance compliance. Participants were instructed not to integrate the jumping exercise into cycling or resistance exercise training, but rather to perform it separately from regular training for a few hours. Participants were acquainted with the basic jumping exercises during the first lab visit and were instructed on the jumping exercise sessions using online videos during the intervention. The full 18-week training program is available as Supplementary Material S1 (available online).
Supplementation
Immediately before every exercise session, participants ingested a commercially available collagen supplement (Vital Supply) that was tested for doping substances. A 15 g dose was used, containing 229 kJ (55 kcal), 13.5 g protein (hydrolyzed Type 1 collagen), 0 g carbohydrates, 0 g fat, 60 mg vitamin C, and 7.5 μg vitamin D. The collagen supplement was provided as powder in jars, including a 15 g scoop to ensure accurate dosing. Participants mixed the collagen powder with ∼150 ml of tap water, with the option of adding a noncaloric flavor. Compliance and potential adverse events were monitored weekly by a researcher who contacted the participants by phone.
Study Procedures
Cyclists interested in participating in the current study were invited for baseline measurements in our laboratory. Cyclists arrived in the morning after an overnight fast, with their written informed consent first obtained. After checking eligibility by means of a health questionnaire, baseline data were obtained. First, participants’ body mass and height were determined using a digital scale and a mobile stadiometer, respectively (Seca 770 and Seca 213i, Seca). Next, a venous blood sample was obtained, followed by the assessment of body composition, BMD, and bone mineral content (BMC) by dual-energy X-ray absorptiometry (DXA). Finally, participants allocated to INT were familiarized with the jumping exercises and received written and oral instructions regarding the exercise intervention and preparation of the collagen drink. The measurements at baseline were repeated 48 hr after completing the last jumping exercise session of the 18-week intervention period. All testing procedures were conducted in the morning and under the same standardized conditions (i.e., no exercise or alcohol 24 hr before testing).
During Weeks 8–10 of the intervention, habitual dietary intake was assessed during two cycling exercise training days using a web-based 24-hr recall system (Compl-eat, Wageningen University), as described previously (Wardenaar et al., 2015). Furthermore, during Weeks 5 and 14 of the intervention, participants allocated to INT completed an online questionnaire on the feasibility and acceptability of the intervention.
Study Outcomes
BMD, BMC, TBS, and Body Composition
Whole-body, lumbar spine (L1–L4), dual proximal femur (femoral neck and total hip) BMD (in grams per square centimeter), and BMC (in grams) were determined by DXA (Horizon, Hologic), according to the procedures recommended by The International Society for Clinical Densitometry (Shuhart et al., 2019), using the system’s software package (Apex, version 5.6.0.5, Hologic). In addition to BMD and BMC, TBS of the lumbar spine was determined using TBS iNnsight software (version 3.1.2, Medimaps), providing an indirect assessment of trabecular microarchitecture (Silva et al., 2014, 2015). Furthermore, measurements of whole body and regional body composition were performed, according to recommended standardized protocols (Nana et al., 2015), using the classic calibration algorithm (without NHANES correction). The DXA system was calibrated each morning before measurements took place. Measurements and analyses were performed by a single researcher. Precision of BMD measurements (coefficient of variation [CV]) within our lab are 0.93%, 0.84%, 1.26%, and 1.97% for the whole body, lumbar spine, total hip, and femoral neck, respectively. Measurement precision (CV) of fat mass and fat-free mass within our lab are 1.73% and 0.41%, respectively.
Serum Markers of Bone Turnover
Procollagen Type I N propeptide (P1NP) and carboxy-terminal cross-linking telopeptide of Type I collagen (CTX-I) were selected as markers of bone formation and bone resorption, respectively (Vasikaran et al., 2011). Blood samples were collected in serum separator tubes and were allowed to clot for ∼30 min and centrifuged at 1,000g for 15 min. Subsequently, aliquots of serum were stored at –80 °C, until further analysis. Intact PINP and CTX-I were measured using chemiluminescent immunometric assays on the IDS‐iSYS instrument (Immunodiagnostic Systems, PLC). Interassay precision of CTX-I and P1NP measurements (CV) were ≤12% and ≤4%, respectively.
Feasibility and Acceptability
A questionnaire was designed using a web-based tool (Qualtrics) to provide information on the feasibility and acceptability of the intervention. Participants responded to three questions on the relevance and importance of bone health for elite road-race cyclists, eight questions on the palatability and feasibility of the collagen drink, and 10 questions on the enjoyability and feasibility of the exercise sessions. Participants rated their answers on a 5-point Likert scale (1 = highly disagree, 5 = highly agree).
Sample Size and Statistical Analysis
Sample size was calculated using G*Power (version 3.1.9.2). Based on the effect of a comparable jumping intervention on BMD reported by Kato et al. (2006), we assumed no change in BMD in CON and an increase of ∼2.5 ± 0.5% in INT. With 0.8 power to detect a significant difference (p < .05, two-sided), a total of 34 participants were needed. To compensate for potential dropouts, and to involve all cyclists within the collaborating cycling teams, a total of 50 participants were initially approached. Forty-three cyclists were assessed for eligibility and participated in the study (Supplementary Material S2 [available online]).
Before hypothesis testing, data were examined for normality using the Shapiro–Wilk test. Nonnormally distributed variables were logarithmically transformed before analysis. Baseline characteristics were compared using independent sample t tests. The primary analyses were conducted on participants who completed the study per protocol. The impact of the intervention on BMD, BMC, TBS, and bone turnover was assessed by mixed-model analysis of variance with time (pre- and postintervention) as within-subject factor and treatment (CON vs. INT) as between-subject factor. The magnitude of the Time × Treatment effects were also examined by calculating partial eta squared (
Results
Participants and Compliance
The experimental period was completed by 36 (CON: n = 18, INT: n = 18) of the 43 participants initially allocated to the experimental conditions. In INT, four participants discontinued the intervention due to injuries or personal reasons. Only one injury was potentially related to the jumping intervention (ligament injury of the foot). In CON, three participants withdrew from the study, with two citing injuries and one withdrawing due to health issues (Supplementary Material S2 [available online], CONSORT flow diagram).
Except for TBS, no differences were observed at baseline between CON and INT in participants’ characteristics (Table 1), bone characteristics (Table 2), or serum markers of bone turnover (Table 3). In addition, no differences were observed in regular vitamin D supplementation, and participation in running and resistance training as part of the regular training program (Table 4). Cycling exercise training data were extracted from the TrainingPeaks software platform, indicating no differences between treatments in weekly training volume (Table 4). Overall, weekly cycling training volume increased from 5,742 kcal during the first half of the intervention period to 9,346 kcal during the second half of the intervention period. Although habitual dietary intake was assessed during the intervention period, the digital 24-hr recalls were not completed by all participants and food intake was very likely underreported. The data available indicated no differences in energy, calcium, or vitamin D intake between treatments (Table 4).
Participants’ Characteristics Before and After 18 Weeks of Either Combined Jumping Exercise and Collagen Supplementation (Intervention) or No Intervention (Control)
| Control (n = 18) | Intervention (n = 18) | p | Effect size ( | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Pre | Post | Change | Pre | Post | Change | Time | Treatment | Interaction | ||
| Age (years)a | 20 ± 2 | 22 ± 4 | ||||||||
| Body mass (kg) | 61.3 ± 8.9 | 62.4 ± 8.1 | 1.1 ± 1.9 | 63.7 ± 5.8 | 63.3 ± 6.1 | −0.4 ± 1.7 | .23 | .50 | .02 | .15 |
| Height (m)a | 1.72 ± 8.2 | 1.73 ± 0.08 | ||||||||
| BMI (kg/m2) | 20.5 ± 1.3 | 20.8 ± 1.1 | 0.3 ± 0.7 | 21.3 ± 1.3 | 21.1 ± 1.3 | −0.2 ± 0.7 | .56 | .22 | .03 | .14 |
| Lean mass (kg) | 46.3 ± 7.8 | 47.5 ± 7.5 | 1.2 ± 1.3 | 47.9 ± 4.8 | 49.0 ± 5.1 | 1.1 ± 1.1 | <.001 | .37 | .55 | .002 |
| Fat mass (kg) | 11.3 ± 2.6 | 11.3 ± 2.8 | 0 ± 1.4 | 12.0 ± 3.9 | 10.5 ± 3.7 | −1.5 ± 1.3 | .001 | .99 | <.01 | .24 |
| Fat mass (%) | 19.1 ± 4.1 | 18.6 ± 4.5 | −0.5 ± 1.9 | 19.2 ± 5.3 | 16.9 ± 5.1 | −2.3 ± 1.6 | <.001 | .63 | <.01 | .24 |
Note. Mixed-model analysis of variance was applied to assess the effect of the intervention on the dependent variables. Bolded values indicate p values < .05. No differences in participants’ characteristics were observed at baseline between groups (p > .05 for all). BMI = body mass index;
aChanges in age and height over time not relevant.
Bone Characteristics Before and After 18 Weeks of Either Combined Jumping Exercise and Collagen Supplementation (Intervention) or No Intervention (Control)
| Control (n = 18) | Intervention (n = 18) | p | Effect size ( | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Pre | Post | Change | Pre | Post | Change | Time | Treatment | Interaction | ||
| Hip BMC (g) | 34.17 ± 7.89 | 33.90 ± 7.84 | −0.27 ± 0.84 | 34.67 ± 4.80 | 35.11 ± 5.14 | 0.44 ± 0.97 | .57 | .70 | .03 | .14 |
| Hip BMD (g/cm2) | 0.889 ± 0.105 | 0.886 ± 0.103 | −0.003 ± 0.012 | 0.915 ± 0.081 | 0.921 ± 0.082 | 0.005 ± 0.014 | .57 | .03 | .08 | .09 |
| Hip z score | −0.5 ± 0.9 | −0.6 ± 0.9 | 0.0 ± 0.1 | −0.4 ± 0.6 | −0.3 ± 0.6 | 0.0 ± 0.1 | .62 | .39 | .15 | .06 |
| FN BMC (g) | 4.21 ± 0.80 | 4.16 ± 0.77 | −0.05 ± 0.18 | 4.27 ± 0.53 | 4.24 ± 0.56 | −0.02 ± 0.15 | .21 | .75 | .70 | .01 |
| FN BMD (g/cm2) | 0.789 ± 0.104 | 0.774 ± 0.095 | −0.015 ± 0.018 | 0.803 ± 0.058 | 0.809 ± 0.066 | 0.006 ± 0.021 | .18 | .39 | <.01 | .23 |
| FN z score | −0.6 ± 0.9 | −0.8 ± 0.8 | −0.1 ± 0.2 | −0.5 ± 0.5 | −0.5 ± 0.5 | 0.0 ± 0.2 | .15 | .43 | <.01 | .20 |
| LS BMC (g) | 57.49 ± 12.03 | 58.45 ± 12.73 | 0.96 ± 2.44 | 64.53 ± 10.30 | 65.25 ± 10.38 | 0.72 ± 1.49 | .02 | .08 | .72 | .004 |
| LS BMD (g/cm2) | 0.923 ± 0.111 | 0.928 ± 0.116 | 0.005 ± 0.021 | 0.981 ± 0.093 | 0.989 ± 0.090 | 0.008 ± 0.021 | .07 | .09 | .62 | .01 |
| LS z score | −1.1 ± 1.2 | 0.0 ± 0.2 | −0.1 ± 0.2 | −0.5 ± 0.8 | −0.5 ± 0.8 | 0.1 ± 0.2 | .30 | .11 | .70 | .01 |
| WB BMC (g) | 2,238.45 ± 399.27 | 2,273.73 ± 417.87 | 35.28 ± 54.29 | 2,364.99 ± 281.07 | 2,378.44 ± 290.27 | 13.45 ± 27.46 | <.01 | 0.33 | .14 | .06 |
| WB BMD (g/cm2) | 1.086 ± 0.079 | 1.093 ± 0.078 | 0.007 ± 0.020 | 1.127 ± 0.065 | 1.130 ± 0.067 | 0.003 ± 0.015 | .12 | .11 | .57 | .01 |
| WB z score | −0.5 ± 1.1 | −0.4 ± 1.3 | 0.0 ± 0.2 | 0.0 ± 0.8 | 0.1 ± 0.8 | 0.1 ± 0.3 | .25 | .14 | .70 | .01 |
| TBS of the lumbar spine | 1.38 ± 0.08 | 1.40 ± 0.09 | 0.02 ± 0.04 | 1.46 ± 0.08 | 1.47 ± 0.08 | 0.01 ± 0.02 | <.01 | <.01 | .33 | .03 |
Note. Mixed-model analysis of variance was applied to assess the effect of the intervention on the dependent variables. Bolded values indicate p values < .05. BMC = bone mineral content; BMD = bone mineral density; FN = femoral neck; LS = lumbar spine; WB = whole body; TBS = trabecular bone score;
Blood Markers of Bone Formation and Bone Resorption Before and After 18 Weeks of Either Combined Jumping Exercise and Collagen Supplementation (Intervention) or No Intervention (Control)
| Control (n = 18) | Intervention (n = 18) | p | Effect size ( | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Pre | Post | Change | Pre | Post | Change | Time | Treatment | Interaction | ||
| P1NP (ng/ml)a | 83.6 ± 24.8 | 71.4 ± 23.1 | −12.2 ± 18.9 | 82.8 ± 30.7 | 66.3 ± 30.6 | −16.6 ± 14.7 | <.001 | .50 | .22 | .04 |
| CTX-I (ng/ml) | 0.6 ± 0.2 | 0.5 ± 0.2 | −0.1 ± 0.2 | 0.6 ± 0.3 | 0.5 ± 0.3 | −0.0 ± 0.2 | .08 | .70 | .58 | .009 |
Note. Mixed-model analysis of variance was applied to assess the effect of the intervention on the dependent variables. Bolded value indicates p value < .05.
aNonnormally distributed data were logarithmically transformed before analysis.
Training Volume and Nutritional Intake in the INT and CON
| CON | INT | p | |
|---|---|---|---|
| Weekly resistance exercise, n (%a) | 13 (72) | 14 (78) | .70 |
| Weekly running exercise, n (%b) | 5 (28) | 8 (44) | .30 |
| Weekly cycling training volume (kcal)c | 7,714 ± 2,013 | 8,071 ± 2,249 | .66 |
| Weekly cycling training volume (hr)c | 14 ± 1 | 15 ± 2 | .32 |
| Daily energy intake (kcal)d | 1,937 ± 500 | 2,130 ± 527 | .42 |
| Daily calcium intake (mg)d | 1,354 ± 504 | 1,425 ± 445 | .75 |
| Daily vitamin D intake (μg)d | 3.7 ± 1.6 | 3.9 ± 2.4 | .84 |
| Daily vitamin D supplementation | 78% | 67% | .46 |
Note. INT = intervention group; CON = no-treatment control group.
aPercent of participants that performed 1–2 weekly sessions of resistance exercise. bPercent of participants that performed 1–2 weekly sessions of running exercise. cCON: n = 14, INT: n = 15. dCON: n = 10, INT: n = 9.
Compliance with the intervention decreased from 3.9 ± 0.9 sessions and supplements per week in the first 9 weeks (84% of sessions completed) to 3.2 ± 1.2 sessions and supplements per week in the last 9 weeks (55% of sessions completed; p < .01). The resulting compliance for the entire intervention was 3.5 ± 0.9 sessions and supplements per week (77% of total sessions completed). Only one participant had an average jumping frequency and supplement ingestion below 3 sessions per week (i.e., 2.4).
Body Composition
Body composition before and after the intervention is presented in Table 1, while individual responses are shown in Figure 1. Body mass increased in CON (from 61.3 ± 8.9 to 62.4 ± 8.1 kg), while remaining stable in INT (from 63.7 ± 5.8 to 63.3 ± 6.1 kg; Time × Treatment: p = .02; Table 1, Figure 1). Lean body mass increased to a similar extent (∼2.6%) in CON (from 46.3 ± 7.8 to 47.5 ± 7.5 kg) and INT (from 47.9 ± 4.8 to 49.0 ± 5.1 kg; time effect: p < .001, Time × Treatment: p = .78; Table 1, Figure 1). In contrast, fat mass was unchanged in CON (11.3 ± 2.6 to 11.3 ± 2.8 kg), while a decrease was noted in INT (12.0 ± 3.9 to 10.5 ± 3.7 kg; time effect: p = .001, Time × Treatment: p < .01; Table 1, Figure 1).
—Change in body mass (in kilograms) in CON (n = 18) and INT (n = 18) over the 18-week intervention period. The circles indicate individual cases (white = CON, black = INT), and the black bars indicate mean ± SD. *Mixed-model analysis of variance revealed a Time ×Treatment effect (p < .05). INT = intervention group; CON = no-treatment control group.
Citation: International Journal of Sport Nutrition and Exercise Metabolism 34, 1; 10.1123/ijsnem.2023-0080
Bone Characteristics
Of all participants, 19%, 25%, 39%, and 14% presented with low BMD (z score < −1) at the hip, femoral neck, lumbar spine, and whole body, respectively (Table 2). BMD of the femoral neck decreased in CON (from 0.789 ± 0.104 to 0.774 ± 0.095 g/cm2), while being preserved in INT (from 0.803 ± 0.058 to 0.809 ± 0.066 g/cm2; Time × Treatment: p < .01; Figure 2). A similar effect was observed for BMD of the total hip, although this interaction did not reach statistical significance (p = .08; Table 2). No Time × Treatment interactions were observed for BMD at the lumbar spine and whole-body level (Time × Treatment: p > .05 for all; Table 2). For BMC, a significant Time × Treatment interaction in favor of INT was detected for total hip (p = .03), but not for any other skeletal site. TBS increased from 1.38 ± 0.08 to 1.40 ± 0.09 in CON and from 1.46 ± 0.08 to 1.47 ± 0.08 in INT, respectively (time effect: p < .01), with no differences between treatments (Time × Treatment: p = .33; Table 2).
—Femoral neck BMD in CON (n = 18) and INT (n = 18). (a) Femoral neck BMD values before and after the 18-week intervention period. The circles with error bars represent mean ± SD, and the gray lines represent individual cases. (b) Absolute changes in femoral neck BMD over the 18-week intervention period. The circles indicate individual cases (white = CON, black = INT), and the black bars indicate mean ± SD. *Mixed-model analysis of variance revealed a Time × Treatment effect (p < .01). INT = intervention group; CON = no-treatment control group; BMD = bone mineral density.
Citation: International Journal of Sport Nutrition and Exercise Metabolism 34, 1; 10.1123/ijsnem.2023-0080
Bone Turnover Markers
Mean concentrations of serum bone turnover markers are presented in Table 3, while individual responses are shown in Figure 3. Serum P1NP concentrations decreased by ∼17% in both CON (from 83.6 ± 24.8 to 71.4 ± 23.1 ng/ml) and INT (from 82.8 ± 30.7 to 66.3 ± 30.6 ng/ml; time effect: p < .001, Time × Treatment: p = .22). Serum CTX-I concentrations tended to decrease over time (CON: 0.6 ± 0.2 to 0.5 ± 0.2 ng/ml, INT: 0.6 ± 0.3 to 0.5 ± 0.3 ng/ml; time effect: p = .08), with no differences between treatments (Time × Treatment: p = .58).
—Serum bone turnover markers in CON (n = 18) and INT (n = 18). (a) Serum P1NP values before and after the 18-week intervention period. The circles with error bars represent mean ± SD, and the gray lines represent individual cases. (b) Absolute changes in serum P1NP values over the 18-week intervention period. The circles indicate individual cases (white = CON, black = INT), and the black bars indicate mean ± SD. (c) Serum CTX-I values before and after the 18-week intervention period. (d) Absolute changes in serum CTX-I values over the 18-week intervention period. INT = intervention group; CON = no-treatment control group; P1NP = procollagen Type I N propeptide; CTX-I = carboxy-terminal cross-linking telopeptide of Type I collagen.
Citation: International Journal of Sport Nutrition and Exercise Metabolism 34, 1; 10.1123/ijsnem.2023-0080
Feasibility and Acceptability
Results of the questionnaires regarding the feasibility and acceptability of the intervention are presented in Supplementary Material S3 (available online). The questionnaires indicated that the collagen drink did not decrease appetite nor resulted in gastrointestinal complaints. However, the palatability of collagen powder mixed with only water was rated low, while the palatability of the collagen shake was considered neutral when prepared with added noncaloric sweetener. The jumping exercises were considered valuable and easy to execute. Still, enjoyment of the jump training decreased from the beginning of the intervention to the end of the intervention. Nonetheless, participants would be willing to continue the jump training if it improved BMD, with a frequency of three weekly jump sessions considered achievable.
Discussion
The present study demonstrated that an 18-week intervention of combined jumping exercise and collagen supplementation, implemented during the off-season period, had a positive impact on bone health at the hip region in elite road-race cyclists. However, the intervention did not exhibit any significant effects on whole body and lumbar spine BMD and BMC, as well as TBS of the lumbar spine. Furthermore, the intervention did not affect the change in serum bone turnover markers during the off-season period.
Mechanical loading of the skeleton is a primary factor in maintaining or increasing BMD in humans (Howe et al., 2011; Kohrt et al., 2004; Santos et al., 2017). However, it is challenging to add bone-specific physical activity on top of the already high volumes of cycling exercise in elite road-race cyclists. We used the off-season period to assess the impact of frequent, short sessions of high-impact exercise combined with collagen supplementation on BMD in elite road-race cyclists. Despite the substantial volume of cycling exercise training during the off-season period (∼15 hr/week), the compliance with the intervention was good, as 77% of the prescribed jumping exercise sessions were completed. During this period of only 18 weeks, the intervention beneficially affected BMD at the femoral neck. This finding is consistent with previous longer term interventions showing that 6–12 months of high-impact jumping exercise beneficially affects BMD at the femoral neck in both young and older individuals (Allison et al., 2013; Hartley et al., 2020; Kato et al., 2006). It should be noted, however, that the beneficial effect of the intervention in the current study could not be attributed to an increase in BMD, but rather the prevention of its decrease. In fact, we observed a 1.8% decline in femoral neck BMD in CON after as little as 18 weeks, while the intervention resulted in the preservation of femoral neck BMD (+0.7%). This finding slightly contrasts previous work by Barry and Kohrt (2008), who have shown a 1%–2% decline in BMD at several hip segments over a 9-month competitive season in nonelite competitive road-race cyclists, but without a further decline during the off-season period. It could be possible that the off-season period in elite cycling is characterized by a higher volume and intensity of cycling exercise training, when compared with nonelite cyclists, which may explain the decline of femoral neck BMD during the off-season period in the current study.
Despite the positive effect of the intervention on femoral neck BMD in the current study, there was no observed effect on lumbar spine and whole-body BMD. This finding is consistent with the principle of specificity, which states that bone adaptations only occur at skeletal sites exposed to a change in habitual loading (Kerr et al., 1996). As such, it is plausible that bone cells in the hip region received a higher osteogenic stimulus due to ground reaction and muscle contraction forces, compared with those in the upper body. Our findings are confirmed by two meta-analyses demonstrating that high-impact jumping exercise increases BMD at the femoral neck, but not at the lumbar spine in premenopausal women (Babatunde et al., 2012; Martyn-St James & Carroll, 2010). Nevertheless, the lumbar spine seems the most affected skeletal site in elite road-race cyclists (Hilkens et al., 2023), and many traumatic fractures in cycling occur at upper body skeletal sites. Therefore, future interventions to support cyclists’ bone health should consider impact exercises aimed at stimulating the bones of the upper body. In addition, our evaluation of the current intervention suggests that three weekly sessions of bone-specific exercise is an acceptable frequency for most elite road-race cyclists.
The current study employed the DXA-derived lumbar spine TBS as surrogate marker of bone microarchitecture (Silva et al., 2014). Although we observed an increase in lumbar spine TBS over the off-season period, this change was not affected by the intervention. This contrasts with previous work showing a positive effect on lumbar spine TBS following 20 weeks of resistance-type exercise training, albeit in older adults (Pinho et al., 2020; Ulvestad et al., 2021). It is worth noting that those studies included specific resistance exercises that load the lower back (i.e., squats, back extensions). Nevertheless, Vlachopoulos et al. (2018) reported beneficial effects on lumbar spine TBS after 9 months of exclusively countermovement jumps in adolescent (∼14 years) recreational cyclists. Hence, more specific lumbar spine loading exercises or a more prolonged jumping intervention may be required to positively affect lumbar spine TBS. Furthermore, high-resolution peripheral quantitative computed tomography methodology may be preferred to detect relevant changes in bone structure and bone strength.
In the present study, serum P1NP concentrations, a marker of bone formation, declined in both CON and INT. Serum CTX-I concentrations, a marker of bone resorption, also seemed to decrease over time, although this finding was not statistically significant. Collectively, these findings indicate a reduction of bone turnover during the off-season period. Nevertheless, this change over time was not affected by the intervention. The observed decline in P1NP during the off-season period is contrary to our expectations, as previous work has shown an increase in bone formation markers along with an increase in BMD following resistance training in middle-aged and older individuals (Menkes et al., 1993; Vincent & Braith, 2002). However, previous research has also reported improvements in bone properties alongside a decrease in bone formation markers (Hinton et al., 2015; Vlachopoulos et al., 2018). The off-season period complicates the interpretation of bone turnover markers, as changes in cycling training volume throughout this period may influence bone turnover. More frequent assessments of P1NP and CTX-I could have provided additional insights into the effect of our intervention on bone turnover.
We aimed to maximize the effect of the jumping exercise intervention on bone formation by supplementing collagen before each exercise session. As collagen supplementation was used as a co-intervention with exercise, it is impossible to separate the effect of collagen supplementation and jumping exercise on our study outcomes. Nonetheless, previous work has shown that collagen supplementation augments the increase in P1NP following exercise (Shaw et al., 2017), increases BMD (König et al., 2018), and further augments BMD when combined with calcium and vitamin D supplementation in postmenopausal women (Argyrou et al., 2020; Lampropoulou-Adamidou et al., 2022). Next to its potential benefits for bone health, collagen supplementation has also been shown to modulate body composition (Jendricke et al., 2019; Oertzen-Hagemann et al., 2019). Interestingly, we found a decline in fat mass after combined jumping exercise and collagen supplementation. Considering the limited energy expenditure by the jumping exercises, it seems more reasonable to attribute this effect to collagen supplementation rather than the short exercise bouts. While prior studies have shown that the combination of collagen supplementation and exercise can result in a reduction in fat mass (Jendricke et al., 2019; Zdzieblik et al., 2021), the physiological mechanism for such changes in body composition remains unclear. Further research is required to validate these findings and elucidate the underlying physiological mechanisms driving such changes.
The strengths of the current study include the population of elite road-race cyclists, as well as a comprehensive assessment of bone status following a novel intervention strategy that combined jumping exercise and collagen supplementation. Yet, certain limitations must be acknowledged. The intervention lacked blinding, and a placebo was not administered in the control group. Nevertheless, given the robustness and reliability of the BMD outcomes, the risk of bias is arguably low for these particular outcomes. Furthermore, due to the off-season availability of the elite cyclists, the intervention duration was limited to a maximum of 18 weeks, which may not have been sufficient to observe the full effect of the intervention on BMD. Additionally, the current study design did not allow for the separation of the effects of collagen supplementation and jumping exercise on bone health and/or body composition. Finally, the use of 2-day digital 24-hr recalls to assess habitual dietary intake was chosen to minimize participant burden, but resulted in unsatisfactory quality of the dietary intake data.
In conclusion, frequent short bouts of jumping exercise combined with collagen supplementation beneficially affects femoral neck BMD during the off-season period in elite road-race cyclists. Hence, this intervention seems a promising and feasible strategy to counteract the negative impact of professional cycling on bone health.
Acknowledgments
The authors are grateful to the cyclists who participated in this study and the support provided by the staff members of the cycling teams. The authors also thank Esther van der Burg for her help with data collection. Author Contributions: Study design: Hilkens, van Dijk. Data collection: Hilkens, van Schijndel, Weijer. Data analysis: Hilkens. Data interpretation and manuscript preparation: Hilkens, Decroix, Bons, van Loon, van Dijk. Data interpretation and discussion of the manuscript for important intellectual content and approved the final manuscript: All authors. Funding: The work of Hilkens and van Dijk on this topic is part of the Eat2Move project and sponsored by a grant from the Province of Gelderland, the Netherlands. The collagen supplements were provided free of charge by Niche4Health, he Netherlands. Conflict of Interest: No conflicts of interest, financial or otherwise, related to the current manuscript are declared by the authors. The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. Ethics Approval: This study was approved by Medical Ethical Committee Zuyd, the Netherlands, and conformed to the standards for the use of human participants as outlined in the most recent version of the Declaration of Helsinki. The study was registered at the Netherlands Trial Registry (https://clinicaltrialregister.nl) as NL9770.
References
Allison, S.J., Folland, J.P., Rennie, W.J., Summers, G.D., & Brooke-Wavell, K. (2013). High impact exercise increased femoral neck bone mineral density in older men: A randomised unilateral intervention. Bone, 53(2), 321–328. https://doi.org/10.1016/j.bone.2012.12.045
Argyrou, C., Karlafti, E., Lampropoulou-Adamidou, K., Tournis, S., Makris, K., Trovas, G., ... Triantafyllopoulos, I.K. (2020). Effect of calcium and vitamin D supplementation with and without collagen peptides on bone turnover in postmenopausal women with osteopenia. Journal of Musculoskeletal & Neuronal Interactions, 20(1), 12–17.
Babatunde, O.O., Forsyth, J.J., & Gidlow, C.J. (2012). A meta-analysis of brief high-impact exercises for enhancing bone health in premenopausal women. Osteoporosis International, 23(1), 109–119. https://doi.org/10.1007/s00198-011-1801-0
Barry, D.W., & Kohrt, W.M. (2008). BMD decreases over the course of a year in competitive male cyclists. Journal of Bone and Mineral Research, 23(4), 484–491.
Beck, B.R., Daly, R.M., Singh, M.A., & Taaffe, D.R. (2017). Exercise and Sports Science Australia (ESSA) position statement on exercise prescription for the prevention and management of osteoporosis. Journal of Science and Medicine in Sport, 20(5), 438–445. https://doi.org/10.1016/j.jsams.2016.10.001
Campion, F., Nevill, A.M., Karlsson, M., Lounana, J., Shabani, M., Fardellone, P., & Medelli, J. (2010). Bone status in professional cyclists. International Journal of Sports Medicine, 31(07), 511–515.
Hart, N.H., Newton, R.U., Tan, J., Rantalainen, T., Chivers, P., Siafarikas, A., & Nimphius, S. (2020). Biological basis of bone strength: Anatomy, physiology and measurement. Journal of Musculoskeletal & Neuronal Interactions, 20(3), 347–371.
Hartley, C., Folland, J.P., Kerslake, R., & Brooke-Wavell, K. (2020). High-impact exercise increased femoral neck bone density with no adverse effects on imaging markers of knee osteoarthritis in postmenopausal women. Journal of Bone and Mineral Research, 35(1), 53–63. https://doi.org/10.1002/jbmr.3867
Hilkens, L., van Schijndel, N., Weijer, V., Boerboom, M., van der Burg, E., Peters, V., ... van Dijk, J.W. (2023). Low bone mineral density and associated risk factors in elite cyclists at different stages of a professional cycling career. Medicine & Science in Sports & Exercise, 55(5), 957–965. https://doi.org/10.1249/mss.0000000000003113
Hinton, P.S., Nigh, P., & Thyfault, J. (2015). Effectiveness of resistance training or jumping-exercise to increase bone mineral density in men with low bone mass: A 12-month randomized, clinical trial. Bone, 79, 203–212. https://doi.org/10.1016/j.bone.2015.06.008
Hoon, M.W., Haakonssen, E.C., Menaspà, P., & Burke, L.M. (2019). Racing weight and resistance training: Perceptions and practices in trained male cyclists. The Physician and Sportsmedicine, 47(4), 421–426. https://doi.org/10.1080/00913847.2019.1607602
Howe, T.E., Shea, B., Dawson, L.J., Downie, F., Murray, A., Ross, C., ... Creed, G. (2011). Exercise for preventing and treating osteoporosis in postmenopausal women. Cochrane Database of Systematic Reviews, 6(7), Article CD000333. https://doi.org/10.1002/14651858.CD000333.pub2
Hutson, M.J., O’Donnell, E., Brooke-Wavell, K., Sale, C., & Blagrove, R.C. (2021). Effects of low energy availability on bone health in endurance athletes and high-impact exercise as a potential countermeasure: A narrative review. Sports Medicine, 51(3), 391–403. https://doi.org/10.1007/s40279-020-01396-4
Jendricke, P., Centner, C., Zdzieblik, D., Gollhofer, A., & Konig, D. (2019). Specific collagen peptides in combination with resistance training improve body composition and regional muscle strength in premenopausal women: A randomized controlled trial. Nutrients, 11(4), Article 892. https://doi.org/10.3390/nu11040892
Kato, T., Terashima, T., Yamashita, T., Hatanaka, Y., Honda, A., & Umemura, Y. (2006). Effect of low-repetition jump training on bone mineral density in young women. Journal of Applied Physiology, 100(3), 839–843. https://doi.org/10.1152/japplphysiol.00666.2005
Keay, N., Francis, G., & Hind, K. (2018). Low energy availability assessed by a sport-specific questionnaire and clinical interview indicative of bone health, endocrine profile and cycling performance in competitive male cyclists. BMJ Open Sport & Exercise Medicine, 4(1), Article 424.
Kerr, D., Morton, A., Dick, I., & Prince, R. (1996). Exercise effects on bone mass in postmenopausal women are site-specific and load-dependent. Journal of Bone and Mineral Research, 11(2), 218–225. https://doi.org/10.1002/jbmr.5650110211
Kistler-Fischbacher, M., Weeks, B.K., & Beck, B.R. (2021). The effect of exercise intensity on bone in postmenopausal women (part 2): A meta-analysis. Bone, 143, Article 115697. https://doi.org/10.1016/j.bone.2020.115697
Kohrt, W.M., Bloomfield, S.A., Little, K.D., Nelson, M.E., & Yingling, V.R. (2004). American college of sports medicine position stand: Physical activity and bone health. Medicine & Science in Sports & Exercise, 36(11), 1985–1996. https://doi.org/10.1249/01.mss.0000142662.21767.58
König, D., Oesser, S., Scharla, S., Zdzieblik, D., & Gollhofer, A. (2018). Specific collagen peptides improve bone mineral density and bone markers in postmenopausal women—A randomized controlled study. Nutrients, 10(1), Article 97. https://doi.org/10.3390/nu10010097
Lampropoulou-Adamidou, K., Karlafti, E., Argyrou, C., Makris, K., Trovas, G., Dontas, I.A., ... Triantafyllopoulos, I.K. (2022). Effect of calcium and vitamin D supplementation with and without collagen peptides on volumetric and areal bone mineral density, bone geometry and bone turnover in postmenopausal women with osteopenia. Journal of Clinical Densitometry, 25(3), 357–372. https://doi.org/10.1016/j.jocd.2021.11.011
Martyn-St James, M., & Carroll, S. (2010). Effects of different impact exercise modalities on bone mineral density in premenopausal women: A meta-analysis. Journal of Bone and Mineral Metabolism, 28(3), 251–267. https://doi.org/10.1007/s00774-009-0139-6
Medelli, J., Lounana, J., Menuet, J.J., Shabani, M., & Cordero-MacIntyre, Z. (2009). Is osteopenia a health risk in professional cyclists? Journal of Clinical Densitometry, 12(1), 28–34.
Menkes, A., Mazel, S., Redmond, R.A., Koffler, K., Libanati, C.R., Gundberg, C.M., ... Hurley, B.F. (1993). Strength training increases regional bone mineral density and bone remodeling in middle-aged and older men. Journal of Applied Physiology, 74(5), 2478–2484. https://doi.org/10.1152/jappl.1993.74.5.2478
Nana, A., Slater, G.J., Stewart, A.D., & Burke, L.M. (2015). Methodology review: Using dual-energy X-ray absorptiometry (DXA) for the assessment of body composition in athletes and active people. International Journal of Sport Nutrition and Exercise Metabolism, 25(2), 198–215. https://doi.org/10.1123/ijsnem.2013-0228
Oertzen-Hagemann, V., Kirmse, M., Eggers, B., Pfeiffer, K., Marcus, K., de Marees, M., & Platen, P. (2019). Effects of 12 weeks of hypertrophy resistance exercise training combined with collagen peptide supplementation on the skeletal muscle proteome in recreationally active men. Nutrients, 11(5), Article 1072. https://doi.org/10.3390/nu11051072
Olmedillas, H., González-Agüero, A., Moreno, L.A., Casajus, J.A., & Vicente-Rodríguez, G. (2012). Cycling and bone health: A systematic review. BMC Medicine, 10(1), Article 168. https://doi.org/10.1186/1741-7015-10-168
Pinho, J.P., Forner-Cordero, A., Rodrigues Pereira, R.M., Hernandez, A.J., Dorea, E.L., Mezencio, B., ... Amadio, A.C. (2020). A high-intensity exercise intervention improves older women lumbar spine and distal tibia bone microstructure and function: A 20-week randomized controlled trial. IEEE Journal of Translational Engineering in Health and Medicine, 8, Article 2100108. https://doi.org/10.1109/jtehm.2019.2963189
Plasqui, G., Rietjens, G., Lambriks, L., Wouters, L., & Saris, W.H. (2018). Energy expenditure during extreme endurance exercise: The Giro d’Italia. Medicine & Science in Sports & Exercise, 51(3), 568–574. https://doi.org/10.1249/MSS.0000000000001814
Santos, L., Elliott-Sale, K.J., & Sale, C. (2017). Exercise and bone health across the lifespan. Biogerontology, 18(6), 931–946. https://doi.org/10.1007/s10522-017-9732-6
Shaw, G., Lee-Barthel, A., Ross, M.L., Wang, B., & Baar, K. (2017). Vitamin C-enriched gelatin supplementation before intermittent activity augments collagen synthesis. The American Journal of Clinical Nutrition, 105(1), 136–143. https://doi.org/10.3945/ajcn.116.138594
Shuhart, C.R., Yeap, S.S., Anderson, P.A., Jankowski, L.G., Lewiecki, E.M., Morse, L.R., ... Shepherd, J.A. (2019). Executive summary of the 2019 ISCD position development conference on monitoring treatment, DXA cross-calibration and least significant change, spinal cord Injury, Peri-prosthetic and orthopedic bone health, transgender medicine, and pediatrics. Journal of Clinical Densitometry, 22(4), 453–471. https://doi.org/10.1016/j.jocd.2019.07.001
Silva, B.C., Broy, S.B., Boutroy, S., Schousboe, J.T., Shepherd, J.A., & Leslie, W.D. (2015). Fracture risk prediction by Non-BMD DXA measures: The 2015 ISCD Official Positions Part 2: Trabecular bone score. Journal of Clinical Densitometry, 18(3), 309–330. https://doi.org/10.1016/j.jocd.2015.06.008
Silva, B.C., Leslie, W.D., Resch, H., Lamy, O., Lesnyak, O., Binkley, N., ... Bilezikian, J.P. (2014). Trabecular bone score: A noninvasive analytical method based upon the DXA image. Journal of Bone and Mineral Research, 29(3), 518–530.
Turner, C.H., & Robling, A.G. (2005). Exercises for improving bone strength. British Journal of Sports Medicine, 39(4), 188–189. https://doi.org/10.1136/bjsm.2004.016923
Ulvestad, M., Godang, K., Durheim, M.T., Kongerud, J.S., Brit Lund, M., Bollerslev, J., & Edvardsen, E. (2021). Effect of high-intensity training on bone health and body composition in lung transplant recipients: A secondary analysis of a randomized controlled trial. Clinical Transplantation, 35(8), Article 14375. https://doi.org/10.1111/ctr.14375
Vasikaran, S., Eastell, R., Bruyère, O., Foldes, A.J., Garnero, P., Griesmacher, A., ... Kanis, J.A. (2011). Markers of bone turnover for the prediction of fracture risk and monitoring of osteoporosis treatment: A need for international reference standards. Osteoporosis International, 22(2), 391–420. https://doi.org/10.1007/s00198-010-1501-1
Vincent, K.R., & Braith, R.W. (2002). Resistance exercise and bone turnover in elderly men and women. Medicine & Science in Sports & Exercise, 34(1), 17–23. https://doi.org/10.1097/00005768-200201000-00004
Vlachopoulos, D., Barker, A.R., Ubago-Guisado, E., Williams, C.A., & Gracia-Marco, L. (2018). A 9-month jumping intervention to improve bone geometry in adolescent male athletes. Medicine & Science in Sports & Exercise, 50(12), 2544–2554. https://doi.org/10.1249/MSS.0000000000001719
Warden, S.J., Sventeckis, A.M., Surowiec, R.K., & Fuchs, R.K. (2022). Enhanced bone size, microarchitecture, and strength in female runners with a history of playing multidirectional sports. Medicine & Science in Sports & Exercise, 54(12), 2020–2030. https://doi.org/10.1249/mss.0000000000003016
Wardenaar, F.C., Steennis, J., Ceelen, I.J., Mensink, M., Witkamp, R., & de Vries, J.H. (2015). Validation of web-based, multiple 24-h recalls combined with nutritional supplement intake questionnaires against nitrogen excretions to determine protein intake in Dutch elite athletes. British Journal of Nutrition, 114(12), 2083–2092. https://doi.org/10.1017/s0007114515003839
Weaver, C.M., Gordon, C.M., Janz, K.F., Kalkwarf, H.J., Lappe, J.M., Lewis, R., ... Zemel, B.S. (2016). The National Osteoporosis Foundation’s position statement on peak bone mass development and lifestyle factors: A systematic review and implementation recommendations. Osteoporosis International, 27(4), 1281–1386. https://doi.org/10.1007/s00198-015-3440-3
Zdzieblik, D., Jendricke, P., Oesser, S., Gollhofer, A., & König, D. (2021). The influence of specific bioactive collagen peptides on body composition and muscle strength in middle-aged, untrained men: A randomized controlled trial. International Journal of Environment and Research Public Health, 18(9), Article 837. https://doi.org/10.3390/ijerph18094837
Zhao, R., Zhao, M., & Zhang, L. (2014). Efficiency of jumping exercise in improving bone mineral density among premenopausal women: A meta-analysis. Sports Medicine, 44(10), 1393–1402. https://doi.org/10.1007/s40279-014-0220-8
