Acute catabolic bone metabolism response to exercise in young and older adults: a narrative review

Abstract

Exercise is recommended for cardiometabolic benefits and to preserve or improve bone health, especially for older adults at increased risk of fracture. However, exercise interventions have modest benefits on areal bone mineral density (aBMD), and exercise can lead to bone loss in young athletes under certain conditions. In this narrative review, we discuss evidence for a disruption in calcium homeostasis during exercise that may diminish the skeletal benefits of exercise. Topics include 1) a general overview of the effects of exercise on aBMD; 2) discussion of the exercise-induced disruption in calcium homeostasis; 3) factors that influence the magnitude of the exercise-induced disruption in calcium homeostasis, including age, sex, and exercise mode, intensity, and duration; 4) oral calcium supplementation to minimize the exercise-induced disruption in calcium homeostasis; and 5) potential for exercise-induced increase in parathyroid hormone to be both catabolic and anabolic to bone.

1. INTRODUCTION

Low bone mass and osteoporosis affect an estimated 53.6 million U.S. adults aged 50 years or older,¹ increasing the risk of fracture and subsequent morbidity and mortality. One year after hip fracture, the mortality rate is 24% and 38% in older women and men, respectively.² Of those who survive, 50% are permanently disabled.³ Habitual weight-bearing exercise has traditionally been thought to benefit bone, in that the mechanical forces experienced during exercise can improve areal bone mineral density (aBMD) or slow age-related loss of bone.^4–6 Indeed, several organizations, including the American College of Sports Medicine (ACSM)⁷ and the International Osteoporosis Foundation (IOF), recommend regular weight-bearing exercise to reduce the risk of osteoporotic fracture. However, emerging data suggest that exercise can generate an acute catabolic bone response,^8–19 which may contribute to the bone loss observed in some studies of young athletes.^20–22

To provide a framework to evaluate how exercise impacts bone health, this narrative review will provide 1) a general overview of exercise on aBMD; 2) discussion of the exercise-induced disruption in calcium homeostasis; 3) factors that influence the magnitude of the exerciseinduced disruption in calcium homeostasis, including age, sex, and exercise mode, intensity, and duration; 4) oral calcium supplementation to minimize the exercise-induced disruption in calcium homeostasis; and 5) potential for exercise-induced increase in parathyroid hormone (PTH) to be both catabolic and anabolic to bone. These factors are important considerations for designing exercise interventions aimed at improving bone health.

2. EFFECTS OF EXERCISE ON AREAL BONE MINERAL DENSITY (aBMD)

Exercise is believed to enhance bone health and reduce risk for osteoporosis. Intervention studies of the effect of exercise on osteoporotic fracture risk are limited,²³ and fracture risk reduction from the intervention may have occurred due to a decline in injurious falls rather than improvements in aBMD or bone strength.²³ However, cross-sectional and observational studies have identified higher aBMD in people who exercise regularly and decreased osteoporotic fracture incidence in exercisers versus sedentary populations.^24,25 Furthermore, cross-sectional comparisons of active versus sedentary women support a protective effect of exercise for the reduction of osteoporotic fracture risk, mediated in part through higher aBMD.²⁶ For further information on the bone response to exercise training, consult relevant review articles.^5,6,27

Systematic reviews of exercise intervention trials to improve aBMD in older adults have consistently found that increases in aBMD were modest or non-significant,^28–34 although exercise may slow the rate of bone loss in postmenopausal women.^28–30 It should be noted that an improvement in aBMD is not necessarily needed for healthy older adults; often the goal of exercise is to maintain aBMD or slow the age-related decline in aBMD. Any increase in aBMD, or even the maintenance of aBMD with exercise, observed with exercise studies are still of potential clinical importance compared with the aBMD loss that is reported in the control groups.^28,35,36 However, this may not be sufficient for individuals with osteoporosis in whom improvements in aBMD are needed to reduce fracture risk.

The profound loss of BMD that occurs with disuse and immobilization suggests that skeletal loading through physical activity is critical for bone health and that exercise is important to maintain bone health in older adults. aBMD can decrease by 1% per month during bedrest^37,38 and by 1.6% per month during spaceflight.³⁷ Further, it is unclear if there is a recovery of aBMD following disuse; bedrest studies suggest full recovery does not occur. After 17 weeks of bedrest in healthy men, there was a 3.9% decrease in lumbar aBMD and a 3.6% decrease in aBMD at the femoral neck.³⁹ Six months after the resumption of normal activity, aBMD remained 3.4% and 3.6% below baseline at the lumbar spine and the femoral neck, respectively.³⁹ These data, albeit from relatively small studies, indicate that the absence of skeletal loading can result in significant, rapid declines in aBMD in healthy individuals that may not fully recovery when skeletal loading is restored. The rapid bone loss that occurs with immobilization could be especially problematic for older adults who may be more at-risk for bone loss and associated fracture.

Despite the importance of skeletal loading for maintaining aBMD in older adults, exercise has been associated with a loss of aBMD or bone mineral content (BMC) in a few studies of young athletes. Young competitive male cyclists followed over a year of training and competition had a decrease in hip aBMD of −1.8% over the competitive season.²⁰ There was a slight rebound in aBMD during the off-season, but the net loss over 12 months was −1.5%.²⁰ A follow-up study of young competitive female cyclists found a similar decrease in total hip aBMD of −1.4% over 12 months.²¹ The loss of bone over a competitive season has also been observed in young male NCAA Division I-A basketball players, who had a decrease in total body BMC of −6.1% and a decrease in leg BMC of −10.5%.²²

There may be other explanations for low aBMD or bone loss in young competitive athletes. For example, when exercise is combined with insufficient energy intake, subsequent declines in sex steroids can significantly impact bone health, particularly during vulnerable times in skeletal development (e.g., Relative Energy Deficiency in Sport; RED-S syndrome).⁴⁰ Although this component was not specifically evaluated in previous studies, when male NCAA basketball players received calcium supplementation at practice, there was a partial reversal in the BMC loss observed in the previous year.²² These data indicate that calcium homeostasis, in addition to adequate energy intake, likely plays a role in optimal bone health in young competitive athletes.

Overall, intervention trials demonstrate that benefits of exercise on aBMD in older adults are modest, but that bedrest or skeletal unloading can result in large decreases in aBMD over short periods of time. The loss of bone with bedrest highlights the importance of weight-bearing physical activity in maintenance of skeletal integrity, particularly for older adults. However, there is evidence that vigorous exercise is associated with a loss of aBMD in young athletes. Mechanisms for exercise-induced bone loss in young athletes and/or the modest benefits of exercise on bone in older adults are unknown, but the disruption in calcium homeostasis that occurs acutely in response to exercise may play a role.

3. DISRUPTION OF CALCIUM HOMEOSTASIS DURING EXERCISE

The modest improvements in aBMD in exercise intervention trials and the bone loss observed in some young athletes over a competitive season suggest exercise has the potential to stimulate both bone formation and resorption. The overarching hypotheses for our research in recent years is that 1) mechanical loading forces during exercise can generate an anabolic signal in bone; 2) metabolic factors during exercise can generate catabolic signals in bone; and 3) the balance between mechanical and metabolic factors determines whether exercise has a net anabolic or catabolic effect. Indeed, there are several purported mechanisms through which exercise (e.g., increase in stress hormones, cytokines) and exercise training (e.g., weight loss, suppression of sex hormones) can stimulate an increase in bone resorption. However, emerging evidence demonstrating an acute increase in bone resorption in response to a disruption in calcium homeostasis during exercise may also contribute.

For the purposes of this narrative review, and as has been used in previous research,^10,11,14,15 the disruption in calcium homeostasis is defined as the exercise-induced decrease in serum ionized calcium (iCa). The subsequent increase in parathyroid hormone (PTH) in response to the decrease in iCa then stimulates bone resorption (as estimated by bone resorption markers) to release calcium into the blood stream to defend serum iCa. Indeed, the exercise-induced increase in bone resorption in response to an acute bout of moderate and vigorous endurance exercise has been well-characterized. Table 1 summarizes available studies that have measured serum iCa, PTH, and c-telopeptide of type I collagen (CTX; a marker of bone resorption) from before to after exercise. These studies have found that an exercise bout lasting 1 hour or longer resulted in decreases in serum iCa of −1% to −6% from before to immediately after exercise (Table 1) in young^{8–11,13,16,19} or older adults^13,14 during weight-bearing^13,14,19 or weight-supported^8–11,13,16 activities. After adjusting for the exercise-related contraction in plasma volume, serum iCa content can decrease by approximately −1 mg/dL in the first 15 minutes of exercise in young adults,^10,11 representing a ~25% decrease in iCa, although there is a slight recovery during exercise that leads to a net overall iCa decrease of ~0.6–0.85 mg/dL from before to after exercise. In contrast, older adults have a steadier decline throughout the exercise bout and their lowest adjusted iCa concentration occurs at the end of exercise.^14,15 While earlier studies only reported the change in iCa from before to after exercise, ^8,12–14 several studies have demonstrated that the decrease in iCa can occur within the first 15 minutes of exercise.^9–11,14,19

Table 1.

Summary of studies that have reported serum ionized calcium (iCa), parathyroid hormone (PTH), and c-telopeptide of type I collagen (CTX) concentration responses following an exercise bout. Absolute change is from immediately before to after exercise. Significance is indicated when reported in the original research. Relative changes in iCa, PTH, and CTX have been estimated using unadjusted data (UNADJ) and plasma volume-adjusted data (ADJ), when available. If a study had multiple conditions or study arms, each condition or arm is listed separately.

Study	Participants	Exercise Stimulus	ΔiCa (mg/dL)	ΔPTH (pg/mL)	ΔCTX (ng/mL)
Barry 2011 (crossover) 8
Placebo	Men (n=20); 37±8 y	Simulated 35 km time trial (~60 min vigorous stationary cycling)	UNADJ: −0.24#	UNADJ: +74.0#	UNADJ: +0.23#
			ADJ: −0.60#	ADJ: +70.1#	ADJ: +0.18#
Calcium Before Exercise (1000 mg)	Men (n=20); 37±8 y	Simulated 35 km time trial (~60 min vigorous stationary cycling)	UNADJ: −0.28#	UNADJ: +55.8#	UNADJ: +0.17#
			ADJ: −0.64#	ADJ: +52.1#	ADJ: +0.11#
Calcium During Exercise (1000mg)	Men (n=20); 37±8 y	Simulated 35 km time trial (~60 min vigorous stationary cycling)	UNADJ: −0.20#	UNADJ: +58.0#	UNADJ: +0.28#
			ADJ: −0.64#	ADJ: +53.2#	ADJ: +0.22#
Haakonssen 2015 (crossover) 9
High Calcium Meal Before Exercise (~1350 mg)	Women (n=32); 24±4 y	60% of baseline maximal aerobic power for 80 min followed by a 10 min time trial	UNADJ: −0.16	UNADJ: --	UNADJ: --
			ADJ: −0.48	ADJ: +10.2	ADJ: −0.07
Low Calcium Meal Before Exercise (~46 mg)	Women (n=32); 24±4 y	60% of baseline maximal aerobic power for 80 min followed by a 10 min time trial	UNADJ: −0.16	UNADJ: --	UNADJ: --
			ADJ: −0.44	ADJ: +21.0	ADJ: +0.03
Kohrt 2018 (crossover) 10
Calcium Infusion	Men (n=11); 34±5 y	1 hour of stationary cycling at ~80% of HRmax	UNADJ: +0.21#	UNADJ: +10.5#	UNADJ: +0.07#
			ADJ: −0.04	ADJ: +8.3#	ADJ: +0.01#
Saline Infusion	Men (n=11); 34±5 y	1 hour of stationary cycling at ~80% of HRmax	UNADJ: −0.16#	UNADJ: +30.1#	UNADJ: +0.24#
			ADJ: −0.56	ADJ: +20.6#	ADJ: +0.19#
Kohrt 2019 (crossover) 11
Warm Condition	Men (n=12), Women (n=13); 33.5±5.5 y	1 hour of stationary cycling at 75–85% of HRmax	UNADJ: −0.07	UNADJ: +35.8*	UNADJ: +0.15*
			ADJ: −0.85*	ADJ: +28.8*	ADJ: +0.04*
Cool Condition	Men (n=12), Women (n=13); 33.5±5.5 y	1 hour of stationary cycling at 75–85% of HRmax	UNADJ: −0.07	UNADJ: +34.4*	UNADJ: +0.11*
			ADJ: −0.85*	ADJ: +28.2*	ADJ: +0.08*
Shea 2014 (crossover) 12
1000mg calcium total consumed before and during exercise (began 60 min before exercise)	Women (n=10); 61±4 y	1 hour of treadmill walking at 75–80% VO2peak	UNADJ: −0.17	UNADJ: +8.3	UNADJ: +0.01
			ADJ:	ADJ:	ADJ:
Placebo	Women (n=10); 61±4 y	1 hour of treadmill walking at 75–80% VO2peak	UNADJ: −0.27#	UNADJ: +26.1#	UNADJ: +0.06*
			ADJ: --	ADJ: --	ADJ: --
1000mg calcium total consumed before and during exercise (began 15 min before exercise)	Women (n=23); 61±4 y	1 hour of treadmill walking at 75–80% VO2peak	UNADJ: −0.19#	UNADJ: +30.58#	UNADJ: +0.05*
			ADJ: --	ADJ: --	ADJ: --
Placebo	Women (n=23); 61±4 y	1 hour of treadmill walking at 75–80% VO2peak	UNADJ: −0.29#	UNADJ: +28.3#	UNADJ: +0.06#
			ADJ: --	ADJ: --	ADJ: --
Sherk 2017 13
1000 mg calcium before exercise	Men (n=23); 35±7 y	Simulated 35km time trial (~60 min vigorous stationary cycling)	UNADJ: −0.14#	UNADJ: +49.4#	UNADJ: +0.16#
			ADJ: −0.70#	ADJ: +45.9#	ADJ: +0.12#
Placebo	Men (n=28); 37±8 y	Simulated 35km time trial (~60 min vigorous stationary cycling)	UNADJ: −0.25#	UNADJ: +72.3#	UNADJ: +0.17#
			ADJ: −0.91#	ADJ: +67.7#	ADJ: +0.12#
Wherry 2019 (crossover) 14
Warm Condition	Men (n=7), Women (n=5); 67±5 y	1 hour of treadmill walking at 70–80% HRmax	UNADJ: −0.19*	UNADJ: +17.2*	UNADJ: +0.09*
			ADJ: −0.24	ADJ: +16.9*	ADJ: +0.09*
Cool Condition	Men (n=7), Women (n=5); 67±5 y	1 hour of treadmill walking at 70–80% HRmax	UNADJ: −0.16*	UNADJ: +17.3*	UNADJ: +0.10*
			ADJ: −0.07	ADJ: +18.2*	ADJ: +0.10*
Wherry 2021 (crossover) 15
Calcium Infusion	Men (n=6), Women (n=6); 66±6 y	1 hour of treadmill walking at 70–80% HRmax	UNADJ: +0.07	UNADJ: +3.2	UNADJ: +0.01
			ADJ: +0.16	ADJ: +3.4	ADJ: +0.02
Saline Condition	Men (n=6), Women (n=6); 66±6 y	1 hour of treadmill walking at 70–80% HRmax	UNADJ: −0.15*	UNADJ: +11.3	UNADJ: +0.11*
			ADJ: +0.05	ADJ: +12.5	ADJ: +0.12

^*Within-group change from baseline p<0.05

^#Within-group change from baseline p<0.01.

Y = years; min = minutes; km = kilometers; HRmax = maximal heart rate.

The decrease in iCa stimulates a robust increase (20% to 200%) in unadjusted serum PTH that peaks about 15 minutes after exercise ends and is near the pre-exercise level by 60 minutes after exercise (Table 1).^8,10–14,16 The increase in unadjusted PTH is followed almost immediately by an increase in unadjusted CTX.^{8–14,17–19} The peak CTX level occurs approximately 60 minutes after exercise ends. ^9–11,14,15 The duration CTX remains elevated following exercise is unknown. However, CTX can remain elevated for at least 4 h^10,15 after an acute endurance exercise bout and may remain elevated for at least 24 hours after endurance exercise.¹⁷ It is unknown when CTX concentration returns to baseline following an exercise bout. The relationship between PTH and CTX was not changed by adjusting for plasma volume shifts.

The temporal relationship between changes in serum iCa, PTH, and CTX in response to endurance exercise indicates that the activation of bone resorption occurs in response to the decrease in iCa after only ~15 minutes of exercise (Figure 1). However, the mobilization of Ca from bone does not restore iCa to the pre-exercise level until after exercise ends.^10,11,14 Our original hypothesis was that the decrease in serum iCa during exercise was triggered by calcium loss in sweat, but two lines of evidence indicate this is not the case. First, the decline in serum iCa begins relatively early in exercise before substantial sweat loss has occurred. Second, changes in iCa, PTH and CTX are similar despite different amounts of dermal calcium loss. We determined this by having young adults exercise in cool and warm conditions to manipulate sweat rate. Despite sweat rate and estimated dermal calcium loss being ~50% greater in the warm condition than the cool condition, the changes in iCa, PTH, and CTX did not differ between conditions.¹¹ Similar results were obtained in a study of older adults who exercised in cool and warm conditions.¹⁴ It is possible that longer aerobic exercise (e.g., marathons, triathlons, century bicycle rides) result in a larger volume of sweat calcium losses than what was observed in these studies. If true, dermal calcium loss could exacerbate the acute activation of bone resorption and contribute to the bone loss observed in some elite athletes.^20,21 The mechanisms underlying the decrease in iCa during exercise, including the potential role of dermal calcium losses during longer duration exercise or the increased need for calcium by working muscle, require further investigation.

Figure 1.

Change in unadjusted serum ionized calcium (iCa, Panel A), parathyroid hormone (PTH, Panel B), and c-telopeptide of type I collagen (CTX, Panel C) during 1 hour of brisk walking and 1 hour of recovery in older adults (adapted from Wherry SJ et al., 2019; open circles, solid line), 1 hour of brisk walking and 1 hour of recovery in older recreationally active adults (adapted from Wherry SJ et al., 2021: open diamonds, dashed line), 1 hour of stationary cycling and 1 hour of recovery in young adults (adapted from Kohrt WM et al., 2019; closed triangles, dashed line), and 1 hour of stationary cycling and 1 hour of recovery in young men (adapted from Kohrt WM et al., 2018; closed squares, solid line).

The strongest evidence that the activation of bone resorption during exercise is caused by the decrease in serum iCa is from a “calcium clamp” experiment that prevented a decrease in serum iCa during exercise.¹⁰ Briefly, an intravenous (IV) infusion of calcium gluconate was started 15 min before exercise to raise serum iCa by ~0.2 mg/dL. Serum iCa was measured every 5 min during exercise and the calcium gluconate infusion rate was adjusted to prevent a decline in iCa. At a second exercise visit, procedures were duplicated with a volume-matched, half-normal saline infusion serving as the control condition. Under the control saline infusion, serum iCa began to decrease after 15 minutes of exercise, which triggered a ~2-fold increase in PTH and CTX. The calcium gluconate infusion prevented the decline in iCa during exercise and the PTH and CTX responses were attenuated by ~70%. The total amount of calcium infused from 15 min before to the end of exercise was 156±20 mg, 117±18 mg of which was infused during exercise alone. The calcium clamp experiment clearly demonstrated that the decrease in iCa is a major stimulus for the increases in PTH and CTX during exercise. Although the calcium clamp markedly attenuated the PTH and CTX responses, there were still small increases during exercise. When this experiment was replicated in older, recreationally active adults during 1 hour of brisk treadmill walking, results were similar, although a smaller amount of calcium was infused (134±16 mg total, 98±15 mg during exercise) to prevent the exercise-induced decrease in iCa.¹⁵

Additional factors could potentially explain the observed increase in PTH in the absence of the decrease of iCa. Metabolic acidosis has been shown to increase PTH secretion in dogs, ^41,42 rats,^43,44 and humans.^45,46 However, the relationship between pH and PTH has been primarily studied in the context of the metabolic acidosis that occurs with kidney disease, in which other metabolic derangements could contribute to PTH elevations.⁴⁷ It is unclear if exercise-induced metabolic acidosis similarly impacts PTH secretion in humans. Further, vigorous exercise could also induce bone microdamage that could elevate bone resorption,⁴⁸ but it is not clear if the induction of bone microdamage would occur quickly enough to explain the increase in CTX that occurs within 15–30 minutes of starting exercise. The potential contribution of bone microdamage to the CTX increases observed during exercise and recovery and should be explored in future research.

The cited research has also only studied the responses to acute exercise bouts; it is not known how the acute catabolic response to exercise may influence bone adaptations over time. To our knowledge, no study has integrated the measurement of the acute disruption in calcium homeostasis during exercise with changes in bone mass or strength over time (i.e., change in aBMD, volumetric BMD, bone architecture, or estimates of bone strength). It is important to study this to reconcile if and how the observed acute catabolic response to exercise may oppose the anabolic actions of exercise on bone. There are several possibilities: 1) the acute catabolic response to exercise is outweighed by anabolic actions over time (similar to pharmacologic therapies such as teriparatide) without subsequent loss of bone mass or strength; 2) the acute catabolic response to exercise opposes the anabolic mechanical actions of exercise, potentially leading to bone loss or less bone gained, but bone integrity is still enhanced; or 3) the acute catabolic response to exercise opposes the anabolic mechanical signals with subsequent deleterious skeletal integrity. It is possible that the exercise-induced increase in PTH may lead to increased bone formation (discussed in Section 6) and studies to date have only captured the initial catabolic bone response.

4. FACTORS THAT INFLUENCE THE ACUTE DISRUPTION OF CALCIUM HOMEOSTASIS DURING EXERCISE

In studies that characterized the iCa, PTH, and/or CTX responses to an acute endurance exercise bout, overall patterns of change have been similar across studies and labs, but the magnitude of change was variable.^{8–14,17–19} There are multiple possible factors that may have contributed to lab-to-lab differences, including age, mode of exercise, exercise duration, training status, exercise intensity, and sex.

4.1. Age.

The majority of studies have included young adults,^{8–11,13,16–19} predominantly men, and have reported robust increases in PTH and CTX following an acute endurance exercise bout. Studies of older adults^12,14 also found significant increases in PTH and CTX in response to endurance exercise, but the magnitudes of change were smaller than in young adults. For example, the change in unadjusted CTX from before exercise to peak concentration ranged from 0.11 to 0.28 ng/mL in young adults^{8–11,13,16–19} and 0.01 to 0.10 ng/mL in older adults,^12,14 which followed a similar, smaller decrease in the magnitude of change in iCa and PTH in older compared to young adults. It is not clear whether age, per se, is a determinant of the disruption in calcium homeostasis during exercise or whether age-related differences are attributable to other factors, such as mode of exercise or absolute exercise intensity (discussed below).

4.2. Mode of Exercise.

Cycling, a weight-supported exercise, has been the primary mode of exercise in several studies of young adults on this topic,^8–11,13,16 although a smaller number of studies have also used treadmill running,^17–19 which is a weight-bearing exercise. Older adults have been studied during treadmill walking.^12,14 To our knowledge, the effect of mode of endurance exercise (weight-bearing versus weight-supported) on the disruption of calcium homeostasis has not been evaluated using a within-subject design, but comparisons of mode of exercise across studies provide some insight. Studies of young adults that evaluated the PTH and CTX responses to stationary cycling^8–11,13,16 have generally reported greater magnitudes of change than studies that used treadmill running,^17–19 despite similar intensity and duration of exercise.

It is not clear how modes of endurance exercise other than stationary cycling and treadmill walking/running or non-endurance exercise (e.g., resistance training) influence the iCa, PTH, and CTX responses because they have not been thoroughly characterized. However, in a study of jumping exercise in postmenopausal women, performing 6 sets of 10 repetitions of different high ground reaction force jumps did not significantly increase CTX from before to after exercise; iCa and PTH were not captured.⁴⁹ When jumping exercise was performed to exhaustion in a study of young adults, there was a small, non-significant decrease in CTX from before to after exercise, and changes in iCa and PTH were not captured.⁵⁰ These limited data suggest that non-endurance exercise may not result in the same exercise-induced activation of bone resorption, but jumping exercise duration was short compared to the endurance exercise protocols and changes in iCa and PTH were not collected. Importantly, the modes and relative and absolute intensities of exercise used in these studies may not accurately represent typical exercise regimens performed by the majority of the public, recreationally active individuals, or those with osteoporosis.

4.3. Exercise Duration.

Exercise sessions have predominantly been ~1 hour in the studies that have investigated the disruption in calcium homeostasis in response to an acute exercise bout.^{8,10–14,17–19} A few studies used an exercise duration exceeding 1 hour, but these were at a lower exercise intensity (moderate intensity versus vigorous intensity).^8,16 The studies that included more than 1 hour of exercise reported a ~67–71% and ~8% increase in adjusted PTH and CTX, respectively, which are within range for relative changes in PTH and CTX for shorter duration, higher intensity exercise. The lack of a difference in the change in PTH and CTX, despite the longer exercise duration, may be due to the lower intensity of the exercise. There may be independent effects of both exercise intensity and duration on the disruption of calcium homeostasis, but this has not been investigated and should be a focus of future research.

4.4. Training Status.

Most of the research to date was conducted in trained participants who were accustomed to the prescribed exercise intensity and duration, including studies of older adults. When competitive athletes were included in the study population, study visits occurred in both the competitive season and the off-season, and there were no clear differences between study visits or participants that would indicate that results differ by season or training status.^{8,10,11,13,16} Further, one study investigated the role of training status by comparing the PTH and CTX responses to an exhaustive bout of treadmill running in recreationally active versus highly trained endurance athletes. There were no significant differences in the PTH and CTX responses between the groups, suggesting that exercise training status was not an important determinant of the response.¹⁷ However, because this was a cross-sectional comparison, results should be interpreted with caution, and the effects of exercise training should be investigated to determine if the acute catabolic response of bone to exercise (i.e., increase in CTX) changes over time.

4.5. Relative and Absolute Exercise Intensity.

For the purposes of this section, “exercise intensity” refers to the cardiovascular response to exercise, such as the percent of maximal heart rate (HRmax) or VO2peak. Exercise intensity is not quantified by impact, which could have different effects on the bone response to exercise. However, the role of exercise impact (i.e., weight-bearing versus weight-supported) is discussed in Section 4.2.

In general, most studies evaluated the effects of high intensity endurance exercise (i.e., >70% HRmax) on the disruption of calcium homeostasis,^{8,10–14,17–19} although some used a moderate intensity exercise.^9,16,18 High intensity exercise appears to result in greater increases in PTH and CTX than moderate intensity exercise when duration of exercise is the same. Running at 75% VO₂max resulted in a higher peak PTH concentration than running at 55% or 65% of VO2max in the same participants.¹⁸ Peak CTX concentration was also higher with high intensity exercise.¹⁸

Although relative exercise intensity exercise (i.e., % of HRmax or %VO₂max) was similar across several studies, this does not ensure that absolute exercise intensity was similar. This is most apparent in studies of young and older adults. In two studies, both young and older adults exercised at 75% to 85% of HRmax,^11,14 but the cycling exercise in young adults was approximately 176 watts¹⁰ and the walking exercise in older adults was approximately 3.4 mph at a 0% grade.¹⁴ As mentioned previously, the differences in magnitudes of change in iCa, PTH, and CTX between young and older adults may be due to the large differences in absolute exercise intensity rather than age. However, additional research is needed to investigate the intersections of exercise intensity, exercise mode, duration, and age.

4.6. Sex.

Most of the research on the iCa, PTH, and CTX responses to an acute endurance exercise bout has been in young men. Three studies included women and men,^11,14,15 and there was no apparent sex difference. However, sample sizes were relatively small and there are several reasons why the bone response to exercise may differ by sex. For example, differences in bone size or muscle size could potentially influence calcium utilization during exercise, although this has not been fully investigated. Secondly, differences in sex hormones may modify the skeletal response to exercise. Menstrual cycle phase affects exercise performance,^51,52 and estrogen plays an important role in regulating bone metabolism.⁵³ Similarly, hormonal contraceptive use⁵⁴ or menopausal status⁵⁵ may further augment sex differences in the calcium or bone response to exercise. In addition, RED-S is more common in women.⁴⁰ For these reasons, potential sex differences in the disruption of calcium homeostasis across the lifespan during exercise should be investigated in future studies.

4.7. Summary.

Several factors could influence the magnitudes of change in iCa, PTH, and CTX in response to exercise. If absolute exercise intensity is an important determinant of these responses, then isolating the effects of age and sex will be challenging. Exercise intensity appears to be an apparently important factor, but the distinction between relative and absolute intensity is not clear. Finally, there has been a paucity of research on the disruption of calcium homeostasis during exercise other than continuous cycling and walking/running (e.g., high intensity interval exercise, resistance exercise).

5. CALCIUM SUPPLEMENTATION TO MINIMIZE THE DISRUPTION IN CALCIUM HOMEOSTASIS DURING EXERCISE

The calcium clamp study demonstrated that intravenous calcium administration during exercise attenuates the acute decline in iCa and subsequent increase in bone resorption in response to exercise.¹⁰ Several studies used oral calcium supplementation before and/or during exercise and found a partial attenuation of the increases in PTH and CTX.^8,9,12,13 A study of young, cyclingtrained women measured the iCa, PTH, and CTX responses to two identical exercise bouts consisting of 80 min of moderate intensity cycling followed immediately by a 10-min time trial.⁹ Two hours before each exercise bout, participants consumed a meal with either low (46 mg) or high (1,356 mg) calcium content. iCa concentration was higher during all exercise timepoints following the high calcium meal compared to the low calcium meal. The high calcium meal also attenuated the PTH and CTX responses to exercise by ~60% compared to the low-Ca meal. This degree of attenuation was similar to that observed during the calcium clamp, indicating that oral calcium supplementation may be an effective strategy to reduce exercise-induced bone resorption. However, consuming a high calcium meal 2 hours before exercise is not necessarily practical for the general population.

Other studies used oral calcium timed closer to exercise to enhance the practical use of pre-exercise calcium supplementation.^8,12,13 These studies suggest that both formulation and timing of calcium supplementation influence the iCa, PTH, and CTX responses to exercise. Studies that compared the timing of oral supplementation prior to exercise (ranging from 15 minutes to 60 minutes prior to exercise) using an calcium-enriched beverage or a chewable supplement (~1000 mg) in young adults found a partial attenuation of the decrease in iCa (~34–56%) and the increase PTH (~10–70%) but limited to no effect on the increase in CTX.^8,12,13 In postmenopausal women, a calcium-enriched beverage attenuated the increases in PTH (~70%) and CTX (~83%) when supplementation began 60 minutes before exercise, but not when it began 15 minutes before exercise.¹² Collectively, these studies suggest that ~1,000 mg of oral calcium supplements must be consumed at least an hour before exercise to attenuate the disruption of calcium homeostasis during exercise.

6. PARADOXICAL ACTIONS OF PARATHYROID HORMONE (PTH)

Although calcium supplementation before exercise can attenuate the increases in PTH and CTX during exercise, it is premature to recommend pre-exercise calcium supplementation for this purpose. PTH has paradoxical effects on bone, such that intermittent increases in PTH stimulate bone formation,^56–58 but a chronic elevation in PTH (e.g., hyperparathyroidism) is catabolic to bone.^59,60 Indeed, two drugs used to treat osteoporosis stimulate the PTH receptor (teriparatide, abaloparatide). Whether PTH has catabolic or anabolic actions on bone is determined primarily by the length of time it is elevated rather than the magnitude of the increase, with the transition from anabolic to catabolic action occurring if PTH is elevated more than ~4 hours.⁶¹ In the acute exercise studies discussed above, PTH was elevated for less than 2 hours, so it is plausible that the exercise-induced increase in PTH has anabolic actions in bone. However, to the best of our knowledge, the role of PTH in regulating the anabolic response of bone to exercise has not been investigated.

Studies that used calcium supplementation before or during exercise to manipulate the PTH response have not found a PTH-mediated increase in a serum marker of bone formation.^9,10 However, it may take many hours or days to detect a change in a marker of bone formation, and the studies were not designed to capture that information.⁶² Pharmacokinetic studies of teriparatide and abaloparatide used to treat osteoporosis typically demonstrate an increase in bone formation markers after several weeks, although limited data shows an increase in P1NP within 24 hours of a dose.⁶³

For a more direct comparison to a single exercise bout, a single injection 40 μg of teriparatide in postmenopausal women increased CTX by 47% in the 4 hours following the injection (Figure 2).⁶⁴ In young men, CTX increased ~75% 3 hours following a single 20 μg dose of teriparatide.⁶⁵ These teriparatide-induced increases in CTX are in comparison to the ~100% increase in CTX in young and older adults following an hour of vigorous exercise.^10,15 Despite this acute increase in CTX, daily administration of teriparatide for 12 weeks has a robust anabolic effect on bone, as reflected by an increase in serum P1NP of ~100%.^66,67 If the exercise-induced increase in PTH occurs repeatedly with multiple exercise sessions, this may generate an anabolic response that is similar to teriparatide, but that has not yet been captured in studies to date. Research discussed previously that characterized the PTH and CTX responses to exercise have used only 1 to 3 exercise bouts,^{8–14,16–19} and the acute studies that included P1NP only collected data for up to 4 hours after exercise;^9,10 it may take longer-term exercise training, as well as longer sampling timelines, to detect changes in P1NP in response to exercise. Additional research is needed to understand the role of repeated exercise bouts and longer sampling timelines on PTH, CTX, and P1NP.

Figure 2.

Serum CTX response to 1) a single 40 μg injection of teriparatide in postmenopausal women (adapted from Tsai JN et al., 2015; closed circle, dashed line); 2) a single 20 μg injection of teriparatide in young men (adapted from Zikan V and Stepan JJ, 2008; closed diamond, solid line); 3) 60 minutes of vigorous cycling in young men (adapted from Kohrt WM et al., 2018; open triangle, dashed and dotted line); and 4) 60 minutes of brisk treadmill walking in older adults (adapted from Wherry SJ et al, 2021).

7. CONCLUSIONS

Exercise is believed to enhance bone mass but exercise intervention trials to improve aBMD have yielded only modest improvements. The robust increases in PTH and CTX during and after exercise suggest that exercise may acutely increase bone resorption. The effects of the acute increase in bone resorption on aBMD and bone strength over time or with exercise training are unknown. More research is needed to understand the effect of different modes of exercise, age, sex, exercise duration, and exercise intensity on the disruption of calcium homeostasis during exercise. Calcium supplementation can attenuate the exercise-induced increases in PTH and bone resorption, but the degree of attenuation is influenced by the timing and possibly the amount of supplementation. Because transient increases in PTH can have an anabolic effect on bone, it is premature to recommend pre-exercise calcium supplementation to attenuate the PTH response. It is possible that, like the anabolic osteoporosis medications, the catabolic actions (i.e., increased bone resorption) of exercise on bone are outweighed over time by the anabolic actions, which would not adversely impact bone mass or bone strength. Further research is needed to understand how the PTH and CTX responses to an acute exercise bout change with exercise training, including if the acute catabolic response to exercise influences chronic (potentially anabolic) changes in aBMD, bone strength, or bone structure. This will guide the determination of whether strategies to attenuate the exercise-induced increase in PTH should be implemented.

HIGHLIGHTS.

• Exercise can increase bone mass and prevent falls.

• There is an acute catabolic response of bone metabolism to exercise.

• Not yet known if exercise-induced increase in PTH is anabolic.