The Impact of Tennis on the Bone Mineral Density in Adult Males

Κoronas¹V., Μavvidis² Α., Κoronas¹ Κ., Kousis¹ P.
Aristotle University Thessaloniki, Department of Physical Education and Sports Science.
Division of Physical Activity and recreation.

¹ Department of Physical Education and Sports Science, Aristotle University Thessaloniki, Greece.
² Department of Physical Education and Sports, Democritus University of Thrace,
Komotini, Greece

Correspondance:
Dr K. Koronas
Lecturer of Tennis,
Department of Physical Education and Sports Science,
Aristotle University of Thessaloniki, Greece
3 Adramitiou Str.
Thessaloniki, 56728
Greece
Tel. 003231617291
E-mail: kkoronas@phed.auth.gr

Abstract

The present study evaluates the effects of tennis on the bone mineral density of the lumbar of the vertebral column and the femoral neck in middle-aged males between 40 and 50 years of age. The study sample comprised of 40 middle-aged men divided into two groups: the experimental group was made up of 20 participants who play tennis for a hobby on a regular basis for over ten years and for at least three hours a week and the control group of 20 participants who were not involved in any form of sport or exercise whatsoever. The age range of the study sample chronologically covered the entire ten-year span (40-50 years). The participants’ mean age, weight, height, and body mass index (bmi) was: 45.5±2.9years, 84.8±9.4 kilogrammes, 176.55±6.3 centimetres, and 27.1745±2.3 respectively. The bone mineral density was measured using the D. X .A. (Dual X-ray Absorptiometry) method, in the regions of the lumbar of the vertebral column (L1 – L4) and the femoral neck. The results showed that there were significant positive differences in the bone mineral density of the middle-aged males in the experimental group.
Keywords: tennis, bone mineral density, lumbar, femoral neck.

Introduction

Mechanical loading produced during most forms of physical activity places considerable strain on the bones, which provides the best stimulus for the functional adjustment of the osteal tissue (Nichols et al., 1995; Humphries et al., 2000; Wagert, 2002). This has been demonstrated by considerable research conducted to study the impact of exercise on the formation of bone mineral density. Comparison studies between athletes and non-athletes found that the former presented a higher bone mineral density in certain bone regions such as the vertebral column, the femoral neck and the dominant upper limb (Biewener & Bertram, 1992; Drinkwater, 1994; Heaney, 1996; Mpakas, 2001). Similarly, Lazaridis et al. (2009) showed that a 12-week physical low-impact exercise program significantly improved the bone density levels in femor bone and lumbar spine in elderly subjects with Parkinson disease. Consequently, there appears to be a strong positive correlation between exercise and bone mineral density. By contrast, a decrease in physical activity of either a part or all of the body very quickly leads to loss in bone mineral density and an increase in the frequency of osteoporosis and related fractures. The danger of fractures due to osteoporosis can be greatly reduced if the peak bone mineral density is increased by the age of thirty and from then on is maintained at satisfactory levels through physical exercise (Abazides, 1997). Independent studies have reached the conclusion that individuals who live sedentary lives continue to ‘lose’ bone, while those who are energetic and remain active have a slight but crucial advantage (Nelson, 1994).

In comparative studies by Vuοri (1996) and Wolman et al (1991) conducted on sports which produce partial aggravation, such as tennis or squash, the x-rays of the players’ upper limbs, showed that the bones in the dominant upper limb were bigger and denser than those of the other arm. This discrepancy is largely due to the greater load applied to those bones caused by the particularity of these sports. Tennis is, therefore, effective as an exertion since it offers the opportunity to apply strength and alternating tension to the musculoskeletal system, which is much more useful than a mere application of strength, enabling bone mineral density to be developed and/or maintained at a satisfactory level (Dock et al., 1997). In addition, Wolman et al (1991) confirm the viewpoint that athletes who execute medium to high intensity sporting activities such as gymnastics, weight lifting and the various jumping events, were found to have a higher bone mineral density than those who do cycling or swimming.

Exercise is more effective when done in a preventative capacity from a young age in order to attain the highest possible peak bone density and is continued into mature age, which ensures a person has healthy bones all their life (Lyritis, 1998). Exercise also plays a double role in the prevention of osteoporosis: a) by increasing the bone mineral density up until and during adolescence (lengthening of the bone mainly under the impact of the gender hormones); and b) in aiming for the acquisition of the peak bone mineral density up until the age of thirty in both males and females. Furthermore, exercise, has therapeutic qualities since not only does it preserve the bone mass in adults, but it also slows down the rate of bone mineral density loss that comes with age, and this helps protect from falls in middle and especially old age. In addition, there are multiple benefits that exercise has as a precautionary measure to safeguard against osteoporosis. It reduces the loss of bone mineral density while increasing not only muscle mass but also muscle strength and resilience. It helps to rectify any defects in posture, enhance flexibility, improve balance, decrease the risk of falls, reduce fatigue and alleviate pain (Dionisiotis, 2005). Last but not least, exercise has been shown to strengthen the cardio-respiratory system, to produce both physical and mental benefits, to be especially beneficial in maintaining and improving health, and without doubt contributes to a better quality of life (Dionisiotis, 2005; Mavrovouniotis et al., 2008; Pikoula et al., 2005).

The aim of the present study is to examine the effects of tennis on the bone mineral density of middle-aged males who play on a regular basis, as well as to ascertain if through tennis satisfactory levels of bone mineral density can be maintained for the rest of the players’ lives.
Methods

Design
For the research study a quantitative approach was used (Kromrey, 2000) where one group of participants plays tennis, as a form of exercise while the other group takes no exercise at all. The bone mineral density of the lumbar of the vertebral column and the femoral neck of the two groups was compared. The participants of both groups had corresponding ages, which covered every year within the ten-year span of 40-50 year-olds.

Participants
The study sample comprised a total of 40 middle-aged men, whose mean age was 45.5±2.9 years. The participants were divided into two groups: the experimental group comprising of 20 individuals who were regular tennis players; and the control group comprising of 20 individuals who did no exercise at all. Table 1 records the basic characteristics of the total study sample, while Table 2 refers to the anthropometric characteristics of each subject individually for each group. No significant difference between the two groups was observed (Tables 1, 2). Finally, Table 3 shows the values of the experimental group regarding exercise.

In choosing the sample, the specific criteria that were taken into consideration in order to achieve a relatively homogeneous, comparable sample included age, weight, health (history of fractures), and occupation (sedentary or active). The basic prerequisite for the experimental group was that participants played tennis as a hobby for over ten years, for at least three hours a week but did no other form of exercise. The only precondition for those in the control group was that they did no exercise whatsoever (Tables 1, 2, 3). All participants were informed of the nature and purpose of the study before giving oral consent to participate.

Table 1: Total sample values for the basic characteristics.

	N	Minimum	Maximum	Mean	Std. Deviation
age	40	41.00	50.00	45.5000	2.90887
weight	40	65.00	104.00	84.8000	9.40594
height	40	165.00	190.00	176.5500	6.32030
bmi	40	22.22	32.37	27.1745	2.35183

Table 2: Anthropometric characteristics of tennis players and non-tennis players for each variable.

	Research Groups	N	Mean	Std. Deviation	T
age	Experimental group	20	45.5000	2.94690
	Control group	20	45.5000	2.94690	0.000
weight	Experimental group	20	86.4500	8.30013
	Control group	20	83.1500	10.34294	1.113
height	Experimental group	20	177.9000	5.73906
	Control group	20	175.2000	6.72466	1.366
bmi	Experimental group	20	27.2945	2.05318
	Control group	20	27.0546	2.66598	0.319

Table 3: Experimental Group values regarding exercise.

	N	Minimum	Maximum	Mean	Std. Deviation
years of playing tennis	20	13.00	25.00	18.0500	2.64525
hours of practice per week	20	3.00	15.00	6.5000	2.85620

Measures
The participants’ height was determined using a special mobile grading bar and their body weight was measured on the Bilance Sulus (Milano) Scales. An orthopedic surgeon conducted clinical examinations in order to ascertain the condition of each participant’s musculoskeletal system prior to the measuring procedure. The method used for measuring the bone mineral density was D.X.A (Dual X-ray - Absorptiometry) in the regions of the lumbar of the vertebral column (L1 – L4) and the femoral neck. The time required to measure each region was five (5) minutes. The D.X.A method has high objectivity, reliability and validity, whose in vitro accuracy is estimated as being from 0.5 to 2% and in vitro precision at 1% (Lyritis, 1991), while the absorbed radiation is approximately 1 mrem per region concerned or 1 uSV in comparison to a chest x-ray which is 50 uSV (Latsos, 1998).

The device used was a “Hologic 1000 Q.D.R.” which has a comparative database of 10.000 individuals with normal values (North American Caucasians). The North American normal data values were applied as they are more closely related to those of the Greek population (Molyvda – Athanassopoulou, 1997). Prior to each measurement, the device was calibrated and corrected to ensure accuracy by using a model of the lumbar of the vertebral column made especially for this purpose. The measurements were taken in the morning always by the same specialist machine operator. Raw and weighted values (T- score and Z-score) were recorded for each region according to the sample norms, which the device had in its memory bank.

As indications of reliability and validity the sample norms in the device are considered in accordance with the international data of the World Health Organisation (WHO) on the percentage of loss of bone mineral density in relation to Standard Deviation. The initial values of bone mineral density are comparable to the international norms (1.000 ± 0.12 gr/cm²) and follow natural distribution. The values of the measured regions were automatically compared to the T–score, (i.e., the Standard Deviation of bone mineral density above or below the mean bone mineral density of young individuals) with the Z–score which refers to of Standard Deviation of bone mineral density above or below the mean bone mineral density of individuals of the same age. High negative Τ-score values indicate the risk of fracture. Most of the Absorptiometry is determined by the Τ-score, i.e., the mean value for young adults (WHO, 1994). According to data from the World Health Organisation (WHO, 1994) a decrease in bone mineral density that is greater than 2.5 standard deviation from the mean value of the peak bone mineral density of young individuals (corresponding 1 S.D. = 10% B.M.D) constitutes a diagnostic criterion of osteoporosis. Recent findings demonstrate that the same diagnostic criterion (Τ-score ≤ -2.5) used for post-menopausal women can also be applied to men (Trovas et al. 1997, Selby et al. 2000).

Statistical Analysis
Group values were tested for significant differences with the Univariate Analysis of Covariance (ANCOVA). In each test the analysis of covariance with the Covariance of age, weight and height of the individuals. The value F is given of the degrees of freedom and Eta-Squared. There is a great effect when the latter is over 0.138 Bortz & Döring (2002). The 12^th version of SPSS (2003) was used for all the statistical analyses.

Results

As can be seen in Table 2 and in accordance with the T-test, there is no significant difference in the basic characteristics of the two groups. This provides a good precondition for the participants’ comparability of bone mineral density. The correlation between the hours and years of exercise, weight and height are presented in Table 4. Besides the high correlations anticipated, e.g. between weight and height, two negative correlations were observed between age, bone mineral index (bmi) and hours/week of exercise, that is, as the hours of exercise decrease, weight and/or bmi increase. In addition, weight and height (positively correlated) correspondingly affect the bone mineral density of the lumbar (Table 4). More specifically, it is observed that there is a significant increase in the bone mineral density in the lumbar of the vertebral column of athletes, i.e. where F (1;35) is 19.145** in the initial values, 18.961** in comparison to the Τ-score %, 19.415** in the Τ-score deviation, 18.506** in comparison to the Ζ-score %, and 18.882** in the Ζ-score deviation. The results for the bone mineral density of the femoral neck (presented in Table 6) are much more significant because an even bigger increase in the bone mineral density is observed, i.e. where F (1;35) 29.271** in the initial values, 34.881** in comparison to the Τ-score %, 35.031** in the Τ-score deviation, 34.162** in comparison to the Ζ-score % and 33.703** in the Ζ-score deviation, a finding which is particularly significant since following the appearance of osteopenia or osteoporosis in this region most of the related fractures in old age are incurred.

Table 4: Correlations of co-variants and dependent variables of the experimental group.

	Age	Weight	Height	BMI	Years Playing tennis	Hours Per week	Bone mineral density lumbar
Weight	-0.233
Height	-0.159	0.629**
BMI	-0.171	0.766**	-0.015
Years Playing tennis	0.294	0.148	0.011	0.179
Hours Per week	-0.494*	-0.234	0.170	-0.460*	-0.212
Bone mineral density lumbar	0.030	0.107	0.302	-0.104	-0.241	0.229
Bone mineral density femoral neck	-0.009	0.349*	0.362*	0.154	0.252	-0.009	0.643**

Cells contain zero-order (Pearson) correlations by N=40, df=38.
Cells referring to years and hours of playing tennis by N=20, df=18.
p<.05, ** p<.01

Table 5 presents the comparative results of the two groups in regards to the bone mineral density of the lumbar of the vertebral column. In all cases the experimental group presents significantly higher values than the control group. The effect is very high (** p<0.01).

Table 5: Values of bone mineral density of the lumbar of the study groups.

	Research Groups	Mean	Std. Deviation	F(1;35)	Eta-Squared
Bone mineral density lumbar (L1-L4) Initial values	Experimental group	1.21710	0.151202
	Control group	1.01395	0.117617	19.145**	0.354
	Total	1.11553	0.168699
Bone mineral density lumbar (L1-L4) T-score, % Comparison with young adults	Experimental group	112.4500	15.3266
	Control group	92.9000	10.2086	18.961**	0.351
	Total	102.6750	16.29596
Bone mineral density lumbar (L1-L4) T-score%, deviation	Experimental group	1.2105	1.44045
	Control group	-0.7000	1.06648	19.415**	0.357
	Total	0.2552	1.58141
Bone mineral density lumbar (L1-L4) Z-score, % Comparison with individuals of the same age	Experimental group	115.4500	16.54173
	Control group	95.0000	10.78986	18.506**	0.346
	Total	105.2250	17.24111
Bone mineral density lumbar (L1-L4) Z-score %, deviation	Experimental group	1.4670	1.53338
	Control group	-0.4765	1.04438	18.882**	0.350
	Total	0.4952	1.62646

F: F–value oft ANCOVA test, comparison of players – non-players
(1.35): Degrees of freedom
* p<.05 ** p<.01

Table 5 shows that tennis players surpassed the non-players in all the dependent variables; meaning that the former exceeded the latter not only in the initial values but also in the comparisons (Τ-score, Ζ-score), as well as in the deviation.
There are likewise important differences in the bone mineral density of the femoral neck, with the experimental group presenting significantly higher values than the control group. The extent of the effect here is also very high (** p<0.01) (Table 6).

Table 6: Values in the bone mineral density of the femoral neck of the study groups

	Research Groups	Mean	Std. Deviation	F(1;35)	Eta-Squared
Bone mineral density femoral neck Initial values	Experimental group	1.32535	0.151779
	Control group	1.06170	0.129590	29.271**	0.455
	Total	1.19353	0.192945
Bone mineral density femoral neck T-score, % Comparison with young adults	Experimental group	126.0500	14.2875
	Control group	99.3000	11.94769	34.881**	0.499
	Total	112.6750	18.78951
Bone mineral density femoral neck T-score, % deviation	Experimental group	2.1340	1.17288
	Control group	-0.0610	0.98118	35.031**	0.500
	Total	1.0365	1.54097
Bone mineral density femoral neck Z-score, %. Comparison with individuals of the same age	Experimental group	134.5000	15.26089
	Control group	106.4500	12.76292	34.162**	0.494
	Total	120.4750	19.86362
Bone mineral density femoral neck Z-score, % deviation	Experimental group	2.6390	1.17844
	Control group	0.4940	0.97386	33.703**	0.491
	Total	1.5665	1.52261

F: F–value of ANCOVA test, comparison of players – non-players
(1.35): Degrees of freedom
* p<.05 ** p<.01

Discussion
Not only does exercise produce a strong impact in increasing bone mass, but it also has a highly beneficial effect on the skeletal system as a whole. It is especially effective when implemented during childhood and adolescence, when the skeleton is developing (Georgiou, 1998). More specifically, the results of physical activity on the growing skeleton are the growth of new osteal tissue on the subperiostiel surface and an increase in bone diameter as a measure to adjust to aggravation. The bone responds as much to an increase as to a decrease in mechanical loading (Abazides, 1997). A determining factor in maintaining bone mineral density at satisfactory levels in middle-age is to attain a high peak bone mineral density when the person is young, which for males peaks at the age of 30 and for women at around 25 (Lyritis, 1998). The achievement of peak bone mineral density depends on various factors, of which exercise is one. The more calories are consumed due to exercise in the second and third decades of a person’s life, the higher the peak bone mineral density that is achieved (Lyritis, 1998). Many studies point to an increase of 15% in the bone mineral density of both the hip and the lumbar in people who exercise in comparison to individuals, who despite being in good health, lead a sedentary life (Wolman et al., 1991). In order for physical exercise to have a maximum beneficial effect in not only increasing but also maintaining the bone mineral density at satisfactory levels, it must begin at adolescence, be constant, regular and consist of a combination of the highest possible loading and aerobic gymnastics (Trovas, 2004). As a sport, tennis offers this blend of physical activity which aids in developing and/or maintaining bone mass at a satisfactory level, while further providing the ability to apply strength with alternating tension on the musculoskeletal system; a fact which has been shown to be much more beneficial than the mere application of strength (Avioli, 1997).

The sample age range (40 to 50 years of age) was chosen since after the age of thirty and every decade thereof a natural loss of bone mineral density ranging from 3-5% sets in (Yiatzidis et al., 1995). Research data clearly demonstrates that tennis has a positive correlation on the bone mineral density of people who exercise. Our research findings showed that tennis players presented a statistically significant superiority in all the bone mineral density indicators of both regions over those who do not exercise.

Lastly, reference should be made to the fact that many hours of exercise do not appear to be significantly associated with the indicators of bone mineral density. Nevertheless, the conclusion can be drawn that at least three hours a week of playing tennis is considered sufficient for the satisfactory development and maintenance of bone mineral density. Similar results were found in other research concerning women and the time needed on physical activity (Mpakas, 1996). For this reason, it is important that the recommended duration of physical activity should be specified, in order for there to be a beneficial outcome rather than negative consequences on the bone mineral density of those who exercise.

References

Abazides, G. (1997). Exercise and bone mass in young people. Ostoun, 8(4): 247-250 (In Greek).
Avioli, L.V. (1997). Clinical Manual on Osteoporosis (2^nd Ed.). Science Press, London.
Biewener, Α., & Bertram, J. E. (1992). Mechanical loading and bone growth in vivo. In: B. K. Hall (ed.): Bone (pp 1 – 36). CRC Press, Boca Raton Finnland.
Bortz, J., & Döring, N. (2002). Forschungsmethoden und Evaluation für Human- und Sozialwissenschaftler. 3. Uberarb Auflage: Springer, New York.
Georgiou V., (1998) Exercise and bone mass. Ostoun, 9(4): 213-217 (In Greek).
Dionysiotis, E J. (2005). The role of exercise on osteoporosis. Physical Education and Osteoporosis. Health of the Skeleton, (4)2: 55-59 (In Greek).
Dock, J. E., James, C., Henderson, N. K., & Price, R. I. (1997). Exercise and bone mineral density in mature female athletes. Medicine and Science in Sport Exercise, 29: 291-296.
Drinkwater, B. L. (1994). Does physical activity play a role in preventing osteoporosis? Res Q Exerc Sport, 65:197.
Heaney, R.P. (1995). Nutrition and bone mass. Physical Medicine and Rehabilitation Clinics of North America, 6(3): 551-566.
Humphries, B., Newton, R., Bronks, R., Marshall, S., McBride, J., & Triplet- McBride, T. (2000). Effect of exercise intensity on bone density, strength, and calcium turnover in older women. Medicine and Science in Sport Exercise, 32(6): 1043-1050.
Kromrey, H. (2000). Empirische Sozialforschung. (9 korr. Auflage), Leske & Budrich, Opladen.
Latsos, G. N. (1998) Basic principles on density measurement. Ostoun, 9(4): 235-240 (In Greek).
Lazaridis S., Kitsios A., Ioannidis T., & Zaggelidis G. (2009). The effect of a 12-week low-impact exercise program for elderly people with early Parkinson’s disease in bone mineral density. Medicina Sportiva, 17: 1054 – 1058.
Lyritis, G. P. (1991) Bone measurement density. Ostoun, 2(2): 71-76.
Lyritis, G.P. (1998). Metabolic bone diseases and disturbances in mineral metabolism. Press Line Publications, Athens (In Greek).
Mavrovouniotis, F., Argiriadou, Eir., & Papaioannou, Chr. (2008). Greek traditional dances and quality of old people’s life. Journal of Bodywork and Movement Therapies. doi:10.1016/j.jbmt.2008.11.005
Molyvda–Athanassopoulou, A. (1997). Bone density measurement. Greek population norms with DEXA. ZITI Publications, Thessoloniki (In Greek).
Mpakas, H.E. (1996). Physical Medicine and Rehabilitation. Osteoporosis, Athens.
Mpakas, H. E. (2001). Osteoporosis. Prevention, Recovery, Rehabilitation (2^nd Ed.). Siokis Medical Publishers, Thessaloniki (In Greek).
Nelson, D.A., Kleerekoper, M., & Peterson, E. L. (1994). Reversal of vertebral deformities in osteoporosis. Measurement errors or “rebound”? J. Bone Miner Res, 9:977.
Nichols, D. L., Sanborn, C. F., Bonnick, S. L., Gench, B., & DiMarco, N. (1995). Relationship of regional body composition to bone mineral density in college females. Medicine and Science in Sports Exercise 27(2): 178 -182.
Pikoula, I., Mavrovouniotis, F., & Argiriadou, Eir. (2007). Dance and health in middle-aged women. In Proccedings of the 3^rd FORUM of Physical Education, May 19, Democritus University of Thrace, Komotini, Greece (In Greek).
Trovas, G. P., & Lyritis, G. P. (1997). Aetiology in male osteoporosis. JBMR, 12 Suppl p: S 370
Trovas, G. (2004). Male osteoporosis. Hylonome Publishing, Athens (In Greek).
Selby, P. L, Davies, M., & Adams, J. E, (2000). Do men and women fracture bones at similar bone densities. Osteoporosis Int., 11: 153-157.
Vuοri, I. (1996). Peak bone mass and physical activity: A short review. Nutrition Review, 54: 511-414.
Wagert, P., Littbrand, H., Johansson, A., Nord-strom, P., & Gustafson, Y. (2002). Jumping exercises with and without raloxifene treatment in healthy elderly women. Journal of Bone and Mineral Metabolism, 20: 376-382.
Wolman, R.L., Fanlman, L., Clark, P., Hesp, R., & Harries, M. G. (1991). Different training patterns and bone mineral density of the femoral shaft in elite female athletes. Ann Rheum Dis, 50:487-489.
World Health Organisation (1994). Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. W.H.O. Technical report series 43, Genova.
Yiatzidis, A., Magiasis, B., Apostolopoulou, N., & Lyritis, G. (1995). Study of risk factors for the development of low peak bone density in Greek women aged 17-25.Ostoun, 6(1):10-16.