Calcif Tissue Int (2003) 73:350–355 DOI: 10.1007/s00223-002-0003-3

Calcified Tissue International Ó 2003 Springer-Verlag New York Inc.

The Duration of Exercise as a Regulator of Bone Turnover K. M. Karlsson,1 C. Karlsson,1 H. G. Ahlborg,1 O¨. Valdimarsson,1,2 S. Ljunghall2 1 2

Department of Orthopaedics, Malmo University Hospital, SE-205 02 Malmo, Sweden Department of Medicine, University Hospital, Uppsala, Sweden

Received: 11 February 2003 / Accepted: 20 February 2003 / Online publication: 24 July 2003

Abstract. The relationship between duration of exercise and serum remodeling markers of bone turnover was evaluated by osteocalcin (OC), carboxy-terminal propeptide of type I collagen (PICP), total and bone-specific alkaline phosphatase (ALP) and carboxyterminal cross-linked telopeptide of type I collagen (ICTP) in 24 male premier league soccer players exercising 12 hours/ week (range 8–18), 19 third league players exercising 8 hours/week (range 3–18) and 20 sixth league players exercising 6 hours/week (range 2–10). Twenty-seven volunteers served as controls. Forty-six former male soccer players (mean age 38 years, range 19–47), mean 15 years older than the current players, were compared with 41 matched controls. Data is presented as mean ± SEM. Active male players had 18 ± 4% higher OC, 37 ± 9% higher bone ALP and 36 ± 7% higher ICTP than controls (all P < 0.01). There were no differences in remodeling markers within the three groups of active players but each group had higher OC and ICTP than controls (both P < 0.05). Former players had no difference in bone remodeling markers compared to matched controls, but 39 ± 4% lower OC and 69 ± 8% lower ICTP than active players (both P < 0.001). Duration of activity was correlated with bone ALP and ICTP (both r = 0.3, P < 0.05) in individuals exercising 6 hours/week or less. No correlation was found in those exercising above this level. It seems as if the bone turnover, evaluated by serum bone remodeling markers, adapts to the current activity needed to maintain bone strength, and a duration of exercise above that level seems to confer no additional benefits. Key words: Athletes — Bone turnover — Bone markers — Bone metabolism — Exercise — Former athletes

Exercise may contribute to the prevention of osteoporosis by increasing peak bone mineral density (BMD), reducing age-related bone loss or restoring bone already lost in the elderly [1, 2]. As it is the intensity, rather than the duration of exercise, that appears to be a major

Correspondence to: K. M. Karlsson; E-mail: [email protected]

predictor of BMD [3, 4], high impact loading rather than endurance exercise should be recommended when trying to increase the BMD [5, 6]. Soccer exercise, an activity including both endurance and high impact loaded activities, is known to have 10–20% higher BMD in weight-loaded regions compared with controls [7, 8]. However, the duration of soccer exercise needed to achieve a given peak BMD is poorly defined [3, 4, 7–9]. To evaluate the importance of the duration of exercise, serum remodeling markers of bone turnover could be used to estimate bone formation and bone resorption, as changes in the duration of exercise confers changes in the serum remodeling markers within weeks [10–14]. The question is of public health interest when trying to define the duration of activity above which no further biological important benefits in bone formation (and BMD) are reached. Based on our previous work, evaluating BMD in soccer players on different duration of exercise and in active and former soccer players [7–9], we hypothesized that bone turnover, evaluated by serum remodeling markers of bone turnover, is higher in male soccer players than in controls but not in former male soccer players. We asked the following: (1) Do male soccer players have higher levels of serum remodeling markers of bone turnover than controls? (2) Do soccer players with longer duration of activity have higher level of remodeling markers than those with shorter duration of activity? (3) Do former male soccer players have a lower level of serum remodeling markers than active soccer players, but no different than controls?

Material and Methods We studied 63 active male soccer players, mean age 23 years (range 18–32), exercising at different levels for 2–18 hours/ week. All had been on a competitive level for at least 5 years and with constant activity during the 6 weeks proceeding the study. Twenty-four individuals with a mean age of 23 years (range 18–28) were professional athletes playing in the premier national league: 19 with a mean age of 22 years (range 18–28)

K. M. Karlsson et al.: Exercise and Bone Turnover

were playing in the 3rd national league and 20 with a mean age of 24 years (range 18–32) were playing in the 6th national league. The premier league players exercised a mean of 12 hours/week (range 8–18), the 3rd league players a mean of 8 hours/week (range 3–14) and the 6th league players a mean of 6 hours/week (range 2–10). The training consisted mainly of weight-loaded activities such as running with constant changes in directions, jumping and ball activities, including soccer. Activities using external weights were mainly used during the pre-seasonal training. During season, 1 to 2 competitive matches were played every week. Twenty-seven age- and gender-matched, healthy ambulant volunteers, with a mean age of 23 years (range 19–29), participating in mainly weight–loaded recreational activities for a mean of 3 hours/week (range 0–8) served as controls. Forty-six former national league soccer players with a mean age of 38 years (range 19–47), were also included. During their active career, they exercised a mean of 9 hours/week (range 4– 21) for a mean of 8 years (range 1–25). They were now retired from soccer for a mean of 9 years (range 1–25). The former soccer players currently exercised in mainly recreational weight-bearing sports for a mean of 3 hours/week (range 0– 10). Forty-one age- and gender-matched, healthy ambulant volunteers with a mean age of 38 years (range 20–76), participating in mainly weight-bearing recreational activities for a mean of 2 hours/week (range 0–10), served as controls. The former soccer players were further divided in to those who retired 0–5 years ago, and those who retired more than 5 years ago, in order to evaluate whether there were any beneficial effects on bone turnover remaining the first years after retirement. These two groups were also compared with age-matched controls. Bone turnover in the soccer players and controls was evaluated by serum samples of bone-remodeling markers, collected before noon without fasting, with a minimum of 24 hours without exercise. Osteocalcin (OC) was measured by radioimmunoassay (RIA) (CIS Bio InternationalÒ, Oris Industries, Gif-Sur-Yvette, Cedex, France) (intra- and interassay variation 2.7 and 5.5% respectively); carboxy-terminal propeptide of type I collagen (PICP) by a commercially available RIA kit (Orion DiagnosticaÒ, Espoo, Finland) (intra- and interassay variation 1.8 and 6.7% respectively); carboxyterminal cross-linked telopeptide of type I collagen (ICTP) by a commercially available RIA kit (Orion DiagnosticaÒ, Espoo, Finland) (intra- and interassay variation 4.8 and 5.4% respectively). The samples were allowed to clot for 2 hours, then centrifuged in room temperature, and the serum aliquots were stored in )70°C until assayed. The cases and controls were analyzed after randomly mixing their samples during the same batch assay. Total alkaline phosphatase (tALP) (with intraand interassay variation 2.11 and 1.94%, respectively), calcium (Ca) (intra- and interassay variation 1.29 and 1.96%, respectively), creatinine (creat) and albumin (alb) were measured spectrophotometrically as part of the clinical routine. OC, PICP, total ALP, bone ALP was used to evaluate bone formation and ICTP were used to evaluate bone resorption. Weight was measured by an electrical scale, and height by a manual scale. A questionnaire was used in both cases and controls documenting present and past exercise (hours per week), duration of retirement, occupation, alcohol and tobacco use and nutritional habits. Differences between the groups were determined by multiple analysis of variance (MANOVA). If statistical significance was found, Student’s t test between means were used to further evaluate the groups. Linear regression analyses were used to compare the markers with the duration of activity in the active soccer players and controls exercising up to 6 hours/ week or above. An ANCOVA was used to adjust for differences in age when comparing active and retired soccer players. Z scores, the number of standard deviations above or below the age predicted mean, were derived by linear regression using data in the controls. The Chi square test was used to evaluate differences in life-style factors. The results are pre-

351

sented as mean ± SEM, or mean and range, unless otherwise stated.

Results

Active male soccer players had higher OC (18 ± 4%), bone ALP (37 ± 9%) and ICTP (36 ± 7%), all P < 0.01, respectively, compared to controls. Also PICP (11 ± 4%) and total ALP (9 ± 5%) were higher, a difference not reaching statistical significance compared to controls (Fig. 1). Soccer players within the first, 3rd or 6th league all had higher OC and ICTP compared to controls (Table 1). Active soccer players with the longest duration of activity (mean 12 hours/week) had no higher level of serum remodeling markers of bone turnover compared to soccer players within the middle (mean 8 hours/week) or the shortest (mean 6 hours/week) duration of activity (Table 1) Former soccer players had no different level of serum remodeling markers of bone turnover compared to agematched controls (Table 1). The level of serum remodeling markers of bone turnover was lower in former soccer players than in active soccer players, expressed as Z scores in OC ) 0.2 ± 0.1 SD versus 0.5 ± 0.1 SD and in ICTP ) 0.5 ± 0.1 SD versus 0.7 ± 0.1 SD, both P < 0.001, respectively (Fig. 1). The differences remained after adjusting for differences in age (OC; P < 0.01 and ICTP; P = 0.06). Duration of exercise correlated with bone ALP (r = 0.35, P < 0.05) and ICTP (r = 0.33, P < 0.05) within the range of exercise of up to 6 hours per week (the mean duration of exercise in the 6th league players). Bone ALP increased by 13% and ICTP increased by 10% across every hour increase in duration of activity in those exercising up to 6 hours/week. No correlation was observed in those exercising above 6 hours/week. Social class differed when comparing active soccer players and age- and gender-matched controls (blue collar workers 51% versus 13%, respectively, P < 0.01). The proportion of smokers was no different (3% versus 5%) and none admitted to a daily intake of alcohol. None of the players excluded anything from their diets although one control subject was vegetarian (NS). The proportion of coffee drinkers was less in the active soccer players than among the controls (35% versus 82%, P < 0.001). Lifestyle factors differed minimally in the former soccer players and age- and gender-matched controls. Social class did not differ (blue collar workers 21% versus 17%, respectively), the proportion of smokers did not differ (18% versus 16% respectively). None of the active players and one control subject admitted to daily intakes of alcohol (NS). None of the former soccer players excluded anything from their diets although one control subject was vegetarian (NS) and the proportion of coffee

13.1 199.1 3.7 1.8 5.8 2.50 47.0 98.8

Biochemical markers OC (ng/ml) PICP (lg/l) Total ALP (lg/L) Bone ALP (lg/L) ICTP (lg/L) Ca (mg/L) Alb (g/L) Creat (mg/mL)

± ± ± ± ± ± ± ±

± ± ± ±

0.7a 10.7 0.3 0.2 0.4b 0.02b 0.4c 2.2e

0.6 1.1 1.5 0.5c,d,f

12.8 196.5 3.2 1.7 6.1 2.49 47.3 98.0

± ± ± ± ± ± ± ±

± ± ± ±

0.7a 14.5 0.2 0.2 0.6 0.2 0.4c 2.3

0.7 0.8 1.3 0.7c,g

league

19 22.0 180.5 76.6 7.7 —

3

rd

13.0 213.8 3.6 2.1 5.9 2.50 46.9 92.7

20 23.9 180.6 77.9 6.1 — ± ± ± ± ± ± ± ±

± ± ± ±

0.7a 14.7 0.3 0.2b 0.5b 0.02b 0.6c 1.9c

0.9 1.0 1.7 0.5c

6 league

th

10.9 182.2 3.2 1.4 4.3 2.59 51.2 103.1

27 23.3 180.1 75.8 2.5 — ± ± ± ± ± ± ± ±

± ± ± ±

Controls

0.6 19.5 0.2 0.1 0.3 0.02 0.7 2.0

1.1 1.2 2.0 0.5

9.7 178.8 2.9 — 3.6 2.45 46.5 91.9

23 34.5 180.5 79.2 3.1 3.0 1.6 1.3 1.8 0.7 0.4

± ± ± ±

0.2 0.02c 0.9 2.0b

± 0.3 ± 14.2 ± 0.1

± ± ± ± ±

Former players

10.2 165.6 2.7 — 4.1 2.57 49.1 102.6

38 34.9 179.5 77.1 2.3 —

1.9 1.0 1.7 0.4

± ± ± ±

0.2 0.02 0.9 3.0

± 0.5 ± 13.6 ± 0.1

± ± ± ±

Controls

8.9 153.2 2.9 — 3.4 2.46 46.0 95.6

1.0 1.0 1.7 0.4 1.1

± ± ± ±

0.1 0.02c 0.4b 2.8

± 0.6 ± 15.3 ± 0.2

23 41.3 ± 177.4 ± 81.9± 2.8 ± 13.3 ±

Former players

Former male soccer players and controls

9.2 149.2 2.9 — 3.7 2.57 48.7 102.8

34 40.7 179.4 79.6 2.3 —

2.4 1.2 2.1 0.5

± ± ± ±

0.1 0.01 0.8 2.6

± 0.4 ± 6.3 ± 0.1

± ± ± ±

Controls

Several of the controls are used both in comparison with soccer players with a retirement of 0–5 years and a retirement of more than years retirement. Data presented as mean ± SEM a p < 0.05, b P < 0.01, c P < 0.001 comparing 1st, 3rd and former players and controls. d p < 0.001 comparing 1st and 3rd league players e p < 0.05, f P < 0.001 comparing 1st and 6th league players g p < 0.05 comparing 3rd and 6th league players

24 22.6 180.8 77.4 11.9 —

Numbers Age (yrs) Height (cm) Weight (kg) Activity (hrs/week) Retirement /yrs)

1 league

st

Active male soccer players and controls

Table 1. Biochemical markers in active soccer players, former soccer players retired for 0–5 years, former soccer players retired more than 5 years, and in age- and gendermatched controls

352 K. M. Karlsson et al.: Exercise and Bone Turnover

K. M. Karlsson et al.: Exercise and Bone Turnover

353

Fig. 1. Z score serum-osteocalcin (OC), carboxy-terminal propeptide of type I collagen (PICP), total and bone-specific ALP and ICTP in 63 active male soccer players (mean age 23 years, range 18–32) and 46 retired male soccer players (mean age 38 years, range 19–47).

drinkers was no different among the former soccer players and the controls (89% versus 88%, respectively). Discussion

Active male soccer players had higher levels of serum bone remodeling markers than controls. There was an increase in serum remodeling markers with increased duration of exercise up to 6 hours/week, but no increase with further increased duration of activity. Former male soccer players had lower levels of serum remodeling markers of bone turnover than active players but no difference compared to controls. Since this is a crosssectional analysis we acknowledge the risk of selection bias to explain the results. However, lifestyle factors, nutrition, smoking and alcohol habits differed minimally when comparing the groups. Discrepancies in food intake preceding the test could also be a confounding factor, but as both cases and controls had eaten only a light breakfast and none were fasting, it seems less probable that the group discrepancies in bone remodeling markers could be explained by discrepancies in food intake. The present data support earlier studies in male weightlifters, reporting higher bone turnover in active athletes than in controls, but not in former lifters [12]. However, weight lifting is often confounded by the use of anabolic steroids, a drug usually not used among soccer players (none admitted to any use of anabolic steroids). Furthermore, our data differ from reports by Brahm et al. [15] who found 18% lower PICP and 22% lower ICTP in 30 long-distance runners compared to controls. These discrepancies could be due to endurance exercise specifically, as in running, and endurance exer-

cise mixed with impact-loaded activities, as in soccer, where the athletes both run and jump, and leed to not only discrepancies in BMD, but also in the serum remodeling markers of bone turnover [3, 16]. Different kind of exercise may lead to differences in bone turnover, as reported by Woitge et al. [17] in an 8-week prospective study evaluating anaerobic or aerobic activity. Aerobic exercise diminished the bone resorption markers but did not affect the bone formation markers whereas anaerobic activity resulted in an overall accelerated bone turnover, as evaluated by serum bone remodeling markers [17]. Evaluated in the same active soccer players as in the present study, it has previously, been reported that professional first league male soccer players had no higher BMD than 6th league players but higher BMD than controls [8]. The observation in the current study, that there are no differences in serum remodeling markers of bone formation or bone resorption when comparing male soccer players exercising on mean 12, 8 and 6 hours/week, respectively, supports the notion that soccer exercise above 6 hours/week does not produce higher BMD than in those exercising 6 hours/week. Although we have no measure of the exercise intensity, it is likely that the intensity, as well as the duration of exercise, was higher in the professional soccer players. In spite of this, no higher level of remodeling markers was found in the professional players compared to the 6th league players. It could also be that the exercise intensity, as well as the duration of activity, reaches a level where further increase does not produce higher bone formation or BMD [8]. The present study also reports a correlation between present duration of activity and both serum remodeling

354

markers of bone formation and bone resorption in individuals exercising up to 6 hours/week, but not in those exercising above. The data support previous observations that report BMD to correlate with the present duration of activity in soccer players exercising up to 6 hours/week, but not above [8]. It seems that 6 hours/ week of soccer exercise is sufficient to produce the required increase in bone strength, and that longer periods of exercise do not correlate with higher bone formation or BMD [8]. The hypothesis that the skeleton will adapt to the current activity needed to maintain the bone strength needed, and a duration of exercise above which there is no additional benefit, is supported in two other studies. MacDougall et al. [18] reported an increase in tibia BMD in runners who run up to 20 miles/week and Granheden et al. [19] found a limit of 12 hours/week in weight lifters. In neither study was longer duration of activity followed by higher BMD. None of the cited studies evaluated serum remodeling markers of bone turnover. From the level of bone formation markers and bone resorption markers, it is difficult to predict the BMD. There is no linearity between formation and resorption markers, so that 1 SD increase in both formation and resorption markers is not necessarily followed by an unchanged BMD level. Additionally, as reported in this study, all markers of bone formation were no higher in active soccer players than in controls, possibly because different formation markers reflect different stages in the osteoblastic life cycle. Furthermore, we evaluated the soccer players in a steady state after years of training. It is possible that previously during their career, while building a skeleton with high BMD, there was a relatively higher bone formation than bone resorption. Serum remodeling markers of bone turnover were no higher in former male soccer players than in matched controls while both the bone formation markers and the bone resorption markers were lower than in active soccer players. The literature infers that athletes who reduce the duration of activity seem to reach a new level of serum remodeling markers of bone turnover week to months after changed activity level [10, 11, 13, 14, 20]. Previous studies have reported a gradual decrease in the residually high BMD in former soccer players compared to controls so that no benefits were seen after age 65 years [7]. It seems that changes in both bone formation and bone resorption in former male soccer players compared to active players, evaluated by serum bone remodeling markers, are responsible for higher diminution in BMD across age than in controls [7, 9, 21, 22]. As the discrepancies when comparing active and former soccer players remained, also after adjusting for age, it seems probable that the discrepancies in bone turnover were the result of different training history. In summary, soccer exercise in men that exceeds 6 hours per week does not appear to have additional

K. M. Karlsson et al.: Exercise and Bone Turnover

benefits in bone turnover, as evaluated by serum remodeling markers. After an active career, serum remodeling markers of bone turnover are no different in former male soccer players and controls but lower in former soccer players than in active players. Based on these cross-sectional data, it seems as if bone turnover, evaluated by serum bone remodeling markers, adapts to the current activity needed to maintain bone strength, and a duration of exercise above that level does not seem to confer additional benefits. Acknowledgements. Financial support was obtained from the University Hospital Foundations, Centrum for Sports Medical Research (CIF), the Swedish Society of Medicine, the Swedish Society of Medical Research. References 1. Kannus P, Haapasalo H, Sankelo M, Sievanen H, Pasanen M, Heinonen A, Oja P, Vuori I (1995) Effect of starting age of physical activity on bone mass in the dominant arm of tennis and squash players. Ann Intern Med 123:27–31 2. Bass S, Pearce G, Bradney M, Hendrich E, Delmas PD, Harding A, Seeman E (1998) Exercise before puberty may confer residual benefits in bone density in adulthood: studies in active prepubertal and retired female gymnasts. J Bone Miner Res 13:500–570 3. Block JE, Friedlander AL, Brooks GA, Steiger P, Stubbs HA, Genant HK (1989) Determinants of bone density among athletes engaged in weight-bearing and nonweight-bearing activity. J Appl Physiol 67:1100–1105 4. Fehling PC, Alekel L, Clasey J, Rector A, Stillman RJ (1995) A comparison of bone mineral densities among female athletes in impact loading and active loading sports. Bone 17:205–210 5. Lanyon LE, Rubin CT (1984) Static vs dynamic loads as an influence on bone remodeling. J Biomech 17:897–905 6. Lanyon LE (1992) Control of bone architecture by functional load bearing. J Bone Miner Res 7(suppl 2):S369– S375 7. Karlsson MK, Linden C, Karlsson C, Johnell O, Obrant K, Seeman E (2000) Exercise during growth and bone mineral density and fractures in old age. Lancet 355:469–470 8. Karlsson MK, Magnusson H, Karlsson C, Seeman E (2001) The duration of exercise as a regulator of bone mass. Bone 28:128–132 9. Karlsson MK, Johnell O, Obrant KJ (1995) Is bone mineral density advantage maintained long-term in previous weight lifters? Calcif Tissue Int 57:325–328 10. Welsh L, Rutherford OM, James I, Crowley C, Comer M, Wolman R (1997) The acute effects of exercise on bone turnover. Int J Sports Med 18:247–251 11. Thorsen K, Kristoffersson A, Hultdin J, Lorentzon R (1997) Effects of moderate endurance exercise on calcium, parathyroid hormone, and markers of bone metabolism in young women. Calcif Tissue Int 60:16–20 12. Karlsson MK, Vergnaud P, Delmas PD, Obrant KJ (1995) Indicators of bone formation in weight lifters. Calcif Tissue Int 56:177–180 13. Brahm H, Piehl-Aulin K, Ljunghall S (1996) Biochemical markers of bone metabolism during distance running in healthy, regularly exercising men and women. Scand J Med Sci Sports 6:26–30 14. Rong H, Berg U, Torring O, Sundberg CJ, Granberg B, Bucht E (1997) Effect of acute endurance and strength exercise on circulating calcium-regulating hormones and bone markers in young healthy males. Scand J Med Sci Sports 7:152–159

K. M. Karlsson et al.: Exercise and Bone Turnover

15. Brahm H, Strom H, Piehl-Aulin K, Mallmin H, Ljunghall S (1997) Bone metabolism in endurance trained athletes: a comparison to population-based controls based on DXA, SXA, quantitative ultrasound, and biochemical markers. Calcif Tissue Int 61:448–454 16. Heinonen A, Oja P, Kannus P, Sievanen H, Manttari A, Vuori I (1993) Bone mineral density of female athletes in different sports. Bone Miner 23:1–14 17. Woitge HW, Friedmann B, Suttner S, Farahmand I, Muller M, Schmidt-Gayk H, Baertsch P, Ziegler R, Seibel MJ (1998) Changes in bone turnover induced by aerobic and anaerobic exercise in young males [see comments]. J Bone Miner Res 13:1797–1804 18. MacDougall JD, Webber CE, Martin J, Ormerod S, Chesley A, Younglai EV, Gordon CL, Blimkie CJ (1992) Relationship among running mileage, bone density, and

355

19. 20.

21. 22.

serum testosterone in male runners. J Appl Physiol 73:1165–1170 Granheden T (1988) Short- and long-term effects of vigorous physical activity on bone mineral in the human spine. Extreme spinal loadings (thesis). Gothenburg, Sweden Brahm H, Piehl-Aulin K, Ljunghall S (1997) Bone metabolism during exercise and recovery: the influence of plasma volume and physical fitness. Calcif Tissue Int 61:192–198 Karlsson MK, Johnell O, Obrant KJ (1993) Bone mineral density in professional ballet dancers. Bone Miner 21:163– 169 Karlsson MK, Hasserius R, Obrant KJ (1996) Bone mineral density in athletes during and after career: a comparison between loaded and unloaded skeletal regions. Calcif Tissue Int 59:245–248