Influence of exercise mode and osteogenic index on bone biomarker responses during short-term physical training

Abstract

Prescribing exercise based on intensity, frequency, and duration of loading may maximize osteogenic responses in bone, but a model of the osteogenic potential of exercise has not been established in humans. In rodents, an osteogenic index (OI) has been used to predict the osteogenic potential of exercise. The current study sought to determine whether aerobic, resistance, or combined aerobic and resistance exercise programs conducted over eight weeks and compared to a control group could produce changes in biochemical markers of bone turnover indicative of bone formation. We further sought to determine whether an OI could be calculated for each of these programs that would reflect observed biochemical changes. We collected serum biomarkers [bone-specific alkaline phosphatase (BAP), osteocalcin, tartrate-resistant acid phosphatase (TRAP), C-terminal telopeptide fragment of type I collagen (CTx), deoxypyridinoline (DPD), 25-hydroxy vitamin D (25(OH)D), and parathyroid hormone (PTH)] in 56 women (20.3 ± 1.8 years) before, during and after eight weeks of training. We also measured bone mineral density (BMD) at regional areas of interest using DXA and pQCT. Biomarkers of bone formation (BAP and osteocalcin) increased in the Resistance and Combined groups (p < 0.05), while biomarkers of bone resorption (TRAP and DPD) decreased and increased, respectively, after training (p < 0.05) in all groups. Small changes in volumetric and areal BMD (p < 0.05) were observed in the distal tibia in the Aerobic and Combined groups, respectively. Mean weekly OIs were 16.0 ± 1.9, 20.6 ± 2.2, and 36.9 ± 5.2 for the Resistance, Aerobic, and Combined groups, respectively. The calculated osteogenic potential of our programs did not correlate with the observed changes in biomarkers of bone turnover. The results of the present study demonstrate that participation in an eight week physical training program that incorporates a resistance component by previously inactive young women results in alterations in biomarkers of bone remodeling indicative of increased formation without substantial alterations in markers of resorption.

Introduction

The mechanisms underlying the osteogenic response to physical exercise have been investigated since the pioneering work of Hert and colleagues [1] nearly 40 years ago [2], [3], [4], [5]. This work established that long-term loading of bone has an anabolic effect, improving bone mass and strength. Experimental evidence since these first reports validated the concept that long-term (> 6 months) training programs consisting of high strain rates and high peak forces are more effective in maximizing osteogenic responses in bone than training with a large number of low-force repetitions [6], [7], [8], [9], [10].

Training interventions of longer duration (6–36 months) have consistently reported positive bone mineral density increases, whereas those of shorter durations (< 6 months) have failed to show similar adaptations [11]. Due to limitations in densitometry measurements, the length of the bone remodeling cycle and transient changes in bone mineralization during training it is difficult to quantify acute changes in bone adaptation [12], [13], [14]. However, acute unaccustomed bone loading has been associated with a metabolic uncoupling of bone formation and resorption favoring resorption thus leading to acute microdamage. Over time, the cumulative microdamage in the most extreme cases, can result in stress fractures [15], [16], [17]. For example, due to the unaccustomed activity, the incidence of stress fractures in female trainees in the U.S. Army is between 3.5% and 21% during its nine week basic combat training with most of these injuries observed within six weeks [18].

Since imaging limitations have precluded the ability to document the effects of short-term exercise interventions (< 6 months) on structural and mineral adaptations in bone, measures of serum biomarkers of bone turnover can be used to provide insight into bone remodeling cascades initiating osteogenic responses that accompany short-term exercise interventions. It is hypothesized that these changes are localized to areas of high strain, but the metabolic response may be systemic, and reflected by changes in markers of bone turnover. We are not aware of controlled studies that have evaluated the relationship between biochemical markers and bone metabolism and structural and mineral properties of bone subjected to specific types of short-term physical conditioning.

Unfortunately, the relationship between serum biomarkers of bone turnover and eventual structural and mineral adaptations indicative of bone growth is still not fully elucidated. The result is that it is difficult to predict what mode and volume of exercise are optimal for maximizing osteogenic responses in bone and minimizing resorptive responses, particularly in response to short-term interventions. Early site-specific changes in bone structural or mineral properties may occur prior to stress fracture injury. Indeed, several investigators have demonstrated that lower bone mineral density (BMD) is a risk factor for bone stress injuries, particularly in groups who participate in high volumes of running [19], [20], [21], [22]. However, no prospective studies have reported whether deleterious changes in BMD have been observed over the course of strenuous, short-term exercise programs.

Quantifying the osteogenic potential of an exercise regimen would be beneficial to assist in developing programs that could promote bone formation, particularly in a setting where preventable injuries are a significant burden in terms of cost, personnel and lost time (i.e. military basic combat training). While both the amount and intensity of physical stimuli influence subsequent bone tissue responses, it is known that the intensity of bone loading has greater relative importance in stimulating osteogenesis [11], [23]. Unfortunately, it is quite difficult to quantify exercise intensity of bone-loading forces.

Recently, however, Turner and Robling [23] have developed an equation known as the osteogenic index (OI) of exercise. The OI is defined as OI = I ⁎ LN(N + 1), where I is the intensity of exercise, LN is the natural logarithm, and N is the number of loading cycles produced by the exercise. This equation incorporates intensity of loading (i.e. the ground reaction force through the leg during running) and volume of exercise allowing for a reasonable prediction of the impact of an exercise program on bone mass and osteogenic potential. Studies in rats designed to optimize bone formation via incorporating exercise interventions with a high OI have shown that, despite marginal changes in structural and mineral parameters as assessed using DXA or pQCT, bone strength was significantly improved [23]. To date, there have been no studies which have related the OI to markers of bone turnover in humans.

The current investigation sought to determine whether three types of controlled, periodized eight week exercise programs could safely promote bone formation in young, healthy women. We hypothesized that serum markers of bone turnover would demonstrate changes indicative of bone formation in our training groups. We further sought to determine if early changes in bone adaptation might be evidenced through regional imaging analyses at fracture prone sites. A secondary aim was to determine whether the OI associated with a particular program was predictive of changes in biochemical markers. We hypothesized that subjects in the Combined exercise group with the highest predicted OI would show the greatest improvements in biomarkers indicative of positive bone remodeling as compared to subjects in the Control group or those in an exercise program consisting solely of aerobic or resistance exercise.