Journal of Back and Musculoskeletal Rehabilitation 26 (2013) 427–435 427

DOI 10.3233/BMR-130402
IOS Press

The effects of strengthening and high-impact

exercises on bone metabolism and quality of
life in postmenopausal women: A randomized
controlled trial
Hande Basat, Sina Esmaeilzadeh∗ and Nurten Eskiyurt
Department of Physical Medicine and Rehabilitation, Istanbul Faculty of Medicine, Istanbul University, Istanbul,
Turkey

Abstract.
BACKGROUND AND OBJECTIVE: The aim of this prospective, randomized controlled 6-month interventional trial was to
investigate the effects of strengthening and high-impact exercise training on bone mineral density (BMD), bone turnover markers
and health-related quality of life (HRQoL) in postmenopausal women.
MATERIAL AND METHODS: Forty-two eligible postmenopausal women with osteopenia who referred to our outpatient
clinics were included in the study and allocated equally to three groups receiving strengthening exercise, high-impact exercise
or no exercise (control). The supervised training program consisted of a one-hour exercise session three times a week for six
months. BMD at the L1–L4 and femoral neck, serum bone turnover markers (osteocalcin, OC; N-telopeptides of type I collagen,
NTx) and HRQoL were measured at the beginning and sixth months.
RESULTS: There was a significant increase in the BMD at the lumbar spine (p = 0.017) and femoral neck (p = 0.013) in the
high-impact group compared to the strengthening and control groups. Serum OC (p = 0.033) increased, and NTx (p = 0.034)
decreased significantly only in the high-impact group. HRQoL improved significantly in both training groups (p < 0.001).
CONCLUSIONS: Our data suggest that 6-month supervised high-impact exercise training can be effective in prevention of
bone loss at lumbar spine and femoral neck. These data also indicate that both supervised training programs improve HRQoL in
postmenopausal women.

Keywords: Postmenopausal women, exercise, bone metabolism, quality of life

1. Introduction fect 200 million women around the world [2]. An os-
teoporotic fracture occurs every three seconds and a
Osteoporosis is defined as a progressive systemic vertebral fracture every 22 seconds [3]. Recent evi-
dence indicates that the incidence of osteoporosis and
skeletal disease characterized by low bone mass and
osteoporotic fractures has been elevated due to the in-
micro-architectural deterioration of bone tissue, with
creased life expectancy, as well as sedentary lifestyle
a consequent increase in bone fragility and suscepti- and poor nutritional habits [3–5]. As we know, osteo-
bility to fracture [1]. Osteoporosis is estimated to af- porotic fractures are associated with chronic pain, lim-
itations on physical functioning and psychosocial im-
pairment. For all the above reasons, prevention and
∗ Corresponding author: Sina Esmaeilzadeh, MD, Department
management of osteoporosis and its related fractures
of Physical Medicine and Rehabilitation, Istanbul Faculty of
Medicine, Millet Cad, Capa, Fatih, 34093, Istanbul, Turkey. Tel.: are of particular importance [5–11].
+90 212 4142000 33295; Fax: +90 212 6311167; E-mail: sinabox@ Many studies indicate that mechanical loading has
gmail.com. positive effects on the skeleton. Physical activities

ISSN 1053-8127/13/$27.50 
c 2013 – IOS Press and the authors. All rights reserved
428 H. Basat et al. / The effects of strengthening and high-impact exercises on bone metabolism and quality of life

and exercise trainings induce an optimal stimulus for education; participant exclusion criteria were as fol-
skeletal development [11]. Several studies indicate lows: (i) presence of surgical menopause, (ii) his-
that various exercise training programs can be asso- tory of osteoporotic fracture at any site, (iii) pres-
ciated with improvements in bone mass and can re- ence of any osteoporotic fractures in the dorsal and/or
duce fall-related fractures due to improvement in mus- lumbar spine radiographies, (iv) any biochemical ab-
cular strength [12–18]. There is also evidence suggest- normalities related to bone and mineral metabolism,
ing that exercise improves functional status and health- (v) prior or current use of any medication that al-
related quality of life (HRQoL) in postmenopausal ter bone metabolism (e.g. corticosteroids, thiazide di-
women [19]. However, data regarding the impact of uretics, anti-convulsants, anticoagulants, lithium), (vi)
different types of the exercise training on bone mass use of any medication for osteoporosis treatment in
and HRQoL are inconsistent. the last year (e.g. bisphosphonate, calcitonin), (vii)
The aim of this study was (i) to investigate the ef- smoking and alcohol consumption, (viii) presence of
fects of 6-month supervised strengthening and high- any disease that affects bone metabolism and leads to
impact exercise training programs on bone mineral secondary osteoporosis (e.g. hyperthyroidism, hyper-
density (BMD) and bone turnover markers in post- parathyroidism), (ix) participation in any training pro-
menopausal women with osteopenia; (ii) to determine grams that encompass strengthening or high-impact
the impact of different types of exercise training on exercises at least two days per week, (x) presence of
HRQoL; (iii) and finally to compare the effectiveness any condition or systemic disease that complicates the
of two different exercise training programs to each participation in exercise training programs (e.g. severe
other. osteoarthritis in hip or knee joints, severe cardiovas-
cular disease, chronic obstructive pulmonary disease,
cerebrovascular accidents, malignancies), (xi) pres-
2. Materials and methods ence of acute herniated disc or spondylolisthesis, (xii)
history of hip or knee replacement surgery.
2.1. Study design By the order of application to the outpatient clin-
ics, the eligible participants were randomly allocated
This study was a prospective, randomized, con- to either of three groups using the computer-generated
trolled 6-month interventional trial, which was con- random numbers: strengthening training group (n =
ducted between November 2006 and May 2008 at the 14), high-impact training group (n = 14) or control
Department of Physical Medicine and Rehabilitation in (no training) group (n = 14). Participant flow chart
the Faculty Hospital. The study protocol was approved and study profile are shown in Fig. 1. Demographic
by the Local Ethics Committee of the Faculty in con- characteristics and past medical history of the partici-
formity with the Declaration of Helsinki. All partici- pants (age, weight, height, marital and menstrual sta-
pants provided written informed consent before partic- tus, number of children, educational level and job) was
ipating in the study. obtained at baseline assessment.

2.2. Participants 2.3. Outcome measures

We studied 221 consecutive postmenopausal women 2.3.1. Bone mineral density

referred to the “Osteoporosis Diagnosis and Manage- In this study, BMD (g/cm2 ) at the L1 –L4 spine
ment Unit” at the Physical Medicine and Rehabili- and femoral neck was measured by dual-energy X-
tation Department; among these, a total of 42 post- ray absorptiometry (DXA) with a GE Lunar DPX-NT
menopausal osteopenic women who met the eligibil- bone densitometer (GE Lunar Radiation Corp, Madi-
ity criteria (see below) were included in the study. The son, WI, USA) by a certified technician. Subject posi-
inclusion criteria were as follows: (i) postmenopausal tions for anterior-posterior lumbar spine and left femur
women aged 40–70 years, (ii) experienced natural scans were standardized according to manufacturer’s
menopause between 40–50 years of age, (iii) absence instructions. Throughout the study, the calibration of
of menstruation for at least one year prior to the study the densitometer was performed in daily and monthly
enrolment, (iv) T-scores of BMD at lumbar spine (L1 – intervals against a standard calibration block supplied
L4 ) and/or femoral neck between −1.0 and −2.5 and by the manufacturer. Scan analysis was performed at
(v) presence of at least 5 years of primary school baseline and 6-month with the same densitometer. The
 H. Basat et al. / The effects of strengthening and high-impact exercises on bone metabolism and quality of life 429

Postmenopausal Women; n=221

Eligibility Criteria

Eligible Participants; n=42

Randomization

Strengthening Group; n=14 High-impact Group; n=14 Control Group; n=14

Baseline Measurement
BMD (L1-L4 spine & femoral neck), Serum Markers (OC & NTx), Quality of Life (QUALEFFO)

Intervention

Strengthening Training High-impact Training Not Trained

Cal (1200 mg/d), Vit D (800 IU/d) Cal (1200 mg/d), Vit D (800 IU/d) Cal (1200 mg/d), Vit D (800 IU/d)

n=1, follow-up failure

n=2, attendance failure n=2, follow-up failure
n=2, attendance failure

6-Month Measurement
BMD (L1-L4 spine & femoral neck), Serum Markers (OC & NTx), Quality of Life (QUALEFFO)

Strengthening Group; n=11 High-impact Group; n=12 Control Group; n=12

Statistical Analysis; n=35

Fig. 1. Participant flow and study profile. BMD, bone mineral density; OC, osteocalcin; NTx, N-telopeptides of type I collagen; QUALEFFO,
Quality of Life Questionnaire of the European Foundation for Osteoporosis.

baseline and 6-month outcome assessments were per- orosis-41” (QUALEFFO-41). The QUALEFFO-41 is
formed within two months prior to the beginning and a disease-specific self-administered questionnaire for
one month after the end of the training program, re- patients with postmenopausal osteoporosis. It is com-
spectively. posed of five domains: pain, physical function, social
function, general health perception and mental func-
2.3.2. Bone turnover markers tion (mood). These five domains can be evaluated sep-
After an overnight fast, venous blood samples were arately or as a combined score. In this study, we pre-
collected from the right or left antecubital veins at ferred the total score ranging from 0 (the best) to 100
baseline and 6-month of the study. All samples were (the worst). All participants completed QUALEFFO-
stored at −70◦ C. At the end of the study, all frozen 41 at baseline and 6-month assessments. The reliabil-
samples were analyzed in the same laboratory (see bel- ity and validity of the QUALEFFO-41 in Turkish have
ow) by the same person to avoid interpretational errors. been previously established [20].
Serum osteocalcin (OC) and serum N-telopeptides of
type 1 collagen (NTx) levels, as a biochemical marker 2.4. Interventions
of bone formation and resorption, were measured by
TM
enzyme immunoassay with a “EDI Osteocalcin (1– All participants were advised to continue normal
43/49) Specific ELISA Kit” (Li StarFish S.r.l., Milan, daily living and maintain their current physical activ-
Italy) and “Osteomark R
NTx Serum Kit” (Wampole ity levels throughout the study. The patients who allo-
Laboratories, Princeton, NJ, USA), respectively. All cated to the control group were asked not to partici-
assays were carried out at the Department of Medical pate in any exercise training programs. All patients re-
Biochemistry in the Faculty Hospital. ceived a daily oral supplement of calcium (1200 mg)
and vitamin D (800 IU) throughout the study to ensure
2.3.3. Quality of life adequate intake of calcium and vitamin D.
HRQoL was assessed using the “Quality of Life The participants in the training groups were pre-
Questionnaire of the European Foundation for Osteop- scribed strengthening or high-impact exercises. Exer-
430 H. Basat et al. / The effects of strengthening and high-impact exercises on bone metabolism and quality of life

Table 1
Details of intervention programs
Intervention/content Dose or intensity Duration SG HG CG
Oral supplementation
•Calcium 1200 mg/day + + +
•Vitamin D 800 IU/day + + +
Normal physical activity
•Activity of daily living + + +
Warm-up period
•Bicycling 50 W/10 min 10 min + + −
•Walking in place, static stretching exercises for the muscles 1 set/10 repetition for each exercise 5 min + + −
of spine, upper and lower limbs
Strengthening exercises
•Abdominal isometrics 1 set/10 repetition for each exercise 30–40 min + − −
•Hip extensor, flexor, abductor and adductor isometrics + − −
•Knee extensor and flexor isometrics + − −
•Push-up + − −
•Cat-camel + − −
High-impact exercises
•Jump rope 10 jumps/day, +5 jumps/week, maximum 10 min − + −
50 jumps/day
Cool-down period
•Relaxation + + −
•Stretching exercises + + −
SG, strengthening group; HG, high-impact group; CG: control group.

cise programs consisted of three sessions a week for six tive data were presented as mean (standard deviation)
months. Each exercise session lasted 60 minutes and and median (range). The homogeneity between the
supervised by the investigators at the Department of three groups was assessed by the Kruskal-Wallis test.
Physical Medicine and Rehabilitation. Attendance in To compare the posttest – pretest changes within each
both training groups was recorded by the investigators group, we used the Wilcoxon signed-rank test for two
and was analyzed at the end of the 6-month follow-up related samples. Improvement percent [(posttest group
period. Patients who participated in at least 60% of the mean – pretest group mean)/(pretest group mean) ×
total target number of exercise training sessions were 100] was calculated for each group and compared by
considered as compliant and included in the analysis. the Kruskal-Wallis test. A p-value of < 0.05 was con-
In each exercise session, the participants were first sidered as the significance level. The pair-wise com-
trained to warm up for 15 minutes then they were parisons were determined using the Mann-Whitney
subject to a 45–60 minute isometric strengthening or U tests and the significance level for multiple com-
high-impact exercises. Finally, they were trained to parison test was defined as 0.017 using the Bonfer-
cool down for 10 minutes. The strengthening or high- roni correction (P -value = 0.05/number of pair-wise
impact exercise training programs complied with the comparisons). All analyses were conducted using the
American College of Sports Medicine (ACSM) recom- SPSS software, version 17.0 for Windows R
(Statis-
mendations [21]. Participants who allocated to the con- tical Package for the Social Sciences, Chicago, IL,
trol group did not participate in any specified training USA).
program throughout the study. Details of the exercise
training programs are described in Table 1.
3. Results
2.5. Data analysis
3.1. Participant characteristics
The normality of distribution was determined by the
Shapiro-Wilk test. Since there was not a normal dis- Of the 42 postmenopausal women who started the
tribution in all variables (except for femoral neck) at study, a total of seven participants withdrew from the
the baseline and/or follow-up measures and the sample study (see Fig. 1) either due to a failure to participate
size of each group was very small, we preferred to use in the follow-up assessments (one in the strengthening
nonparametric tests in all statistical analyses to avoid and two in the control groups) or due to a low rate of
any bias in the interpretation of the results. Descrip- attendance (< 60%) in the exercise sessions over six
 H. Basat et al. / The effects of strengthening and high-impact exercises on bone metabolism and quality of life 431

Table 2
Homogeneity of demographic and clinical variables between three groups at baseline
Strengthening (n = 11) High-impact (n = 12) Control (n = 12) P†
Age (years) Mean ± SD 55.9±4.9 55.6±2.9 56.2±4.0 0.763
Median (min-max) 54.0 (50–66) 56.0 (49–59) 56.5 (49–61)
BMI (kg/m2 ) Mean ± SD 25.0±4.7 26.4±3.5 27.5±3.7 0.166
Median (min-max) 23.7 (19.9–33.5) 25.5 (22.6–32.8) 27.7 (21.6–34.2)
Age at menarche (years) Mean ± SD 13.3±0.5 13.1±0.5 12.8±0.7 0.142
Median (min-max) 13.0 (13–14) 13.0 (12–14) 13.0 (12–14)
Age at menopause (years) Mean ± SD 47.2±2.0 47.4±2.3 48.2±1.9 0.447
Median (min-max) 46.0 (45–50) 48.0 (43–50) 49.0 (45–50)
Years since menopause Mean ± SD 6.0±3.6 6.5±2.0 6.2±3.3 0.721
Median (min-max) 5.0 (2–14) 7.0 (3–10) 6.5 (2–13)
n (%) n (%) n (%)
Marital status Single 0 (0.0) 2 (16.7) 0 (0.0) 0.139
Married 11 (100) 10 (83.3) 12 (100)
Education level Primary school 4 (36.4) 4 (33.3) 5 (41.7) 0.777
High school 2 (18.2) 6 (50.0) 3 (25.0)
University 5 (45.5) 2 (16.7) 4 (33.3)
Job Housewife 5 (45.5) 6 (50.0) 7 (58.3) 0.949
Civil servant 0 (0.0) 0 (0.0) 1 (8.3)
Retired 5 (45.5) 4 (33.3) 1 (8.3)
Other 1 (9.1) 2 (16.7) 3 (25.0)
Number of children 0 0 (0.0) 1 (8.3) 0 (0.0) 0.229
1 1 (9.1) 5 (41.7) 3 (25.0)
2 8 (72.7) 4 (33.3) 6 (50.0)
3 1 (9.1) 2 (16.7) 3 (25.0)
4 1 (9.1) 0 (0.0) 0 (0.0)
SD, Standard Deviation; BMI, Body Mass Index. † The Kruskal-Wallis’s test (α = 0.05).

months (two in the strengthening and two in the high- was a decline in the control group (−1.0%). In betw-
impact groups). There was no meaningful difference een-groups comparison, there was a significant (p =
in the attendance at exercise sessions between the two 0.013) difference between the three groups; and, in the
exercise groups (p > 0.05). At the end of study, the pair-wise comparison, the only significant difference
withdrawn participants were not included in the statis- was between the high-impact and control groups (p =
tical analysis. Demographic and clinical characteristics 0.006).
of the participants which may potentially affect bone As seen in Table 4, over six months, there was
metabolism and outcome measures are shown in Ta- a significant increase in the serum OC levels in
ble 2 and Table 3. There was no significant difference the strengthening (+31.1%), as well as in the high-
regarding the baseline characteristics between the three impact (+30.2%) groups; and, a non-significant in-
groups (p > 0.05). crease (+4.7%) in the control group. When the groups
were compared, there was a significant (p = 0.033) dif-
3.2. Outcome measures ference between the three groups. The pair-wise com-
parison revealed a significant difference between the
The BMD values at the L1 –L4 spine increased both high-impact and control groups (p = 0.012).
in the strengthening (+1.3%) and the high-impact While the NTx levels increased in the control group
(+0.5%) groups; by contrast, there was a significant (+8.9), there was a significant decrease in the NTx
decrease (−2.5%) in the control group over six months levels in both strengthening (−12.7%) and the high-
(Table 4). In between-groups comparison, there was a impact (−25.7%) groups over six months. The change
significant difference between the study groups (p = in the serum NTx levels were significant between the
0.017). The pair-wise comparisons indicated signifi- three groups (p = 0.034) and the pair-wise compar-
cant differences between the high-impact and control isons indicated that there was a significant difference
groups (p = 0.006). between the high-impact and control groups (p =
Participants of the strengthening (+1.6%) and the 0.012).
high-impact (+1.2%) groups had an increase at the The QUALEFFO-41 scores were decreased in all
femoral neck BMD over six months (Table 4). There groups over six months; however, these changes were
432 H. Basat et al. / The effects of strengthening and high-impact exercises on bone metabolism and quality of life

Table 3
Homogeneity of outcome variables between three groups at baseline
Strengthening (n = 11) High-impact (n = 12) Control (n = 12) P†
L1 –L4 spine BMD (gr/cm2 ) Mean ± SD 0.913±0.022 0.907±0.036 0.934±0.038 0.134
Median (min-max) 0.910 (0.886–0.956) 0.913 (0.843–0.967) 0.940 (0.861–1.00)
Femoral neck BMD (gr/cm2 ) Mean ± SD 0.829±0.082 0.828±0.087 0.849±0.071 0.786
Median (min-max) 0.805 (0.698–0.994) 0.836 (0.719–1.01) 0.821 (0.763–0.980)
OC (ng/ml) Mean ± SD 13.6±2.4 13.6±6.0 14.2±3.4 0.470
Median (min-max) 12.8 (9.6–18.2) 11.4 (6.8–26.1) 15.9 (6.7–17.7)
NTx (nmol BCE/L) Mean ± SD 31.7±6.9 27.8±7.2 27.8±5.2 0.221
Median (min-max) 31.7 (21.4–43.1) 27.8 (19.6–43.2) 27.1 (19.1–41.7)
QUALEFFO (0–100) Mean ± SD 30.1±10.0 38.1±19.2 28.7±7.2 0.515
Median (min-max) 27.4 (20.3–56.8) 34.7 (6.2–76.5) 30.1 (17.6–39.8)
SD, Standard Deviation; BMD, Bone Mineral Density; OC, Osteocalcin; NTx, N-telopeptides of Type 1 Collagen; QUALEFFO, Quality of Life
Questionnaire of the European Foundation for Osteoporosis. ‡ The Kruskal-Wallis’s test (α = 0.05).

Table 4
Changes in BMD, serum markers and health-related quality of life by training group from the baseline to sixth month
Outcomes Measures Mean±SD Changes Between-groups comparisons
Baseline 6-month (%) Kruskal-Wallis Mann-Whitney U
p Pairwise p
L1 –L4 Spine BMD (gr/cm2 ) SG 0.913±0.022 0.925±0.039 +1.3 0.017∗∗ SG vs CG 0.032
HG 0.907±0.036 0.911±0.038 +0.5 HG vs CG 0.006∗∗∗
CG 0.934±0.038 0.911±0.048 −2.5∗ SG vs HG 0.976
Femoral Neck BMD (gr/cm2 ) SG 0.829±0.082 0.842±0.083 +1.6 0.013∗∗ SG vs CG 0.019
HG 0.828±0.087 0.837±0.087 +1.2∗ HG vs CG 0.006∗∗
CG 0.849±0.071 0.840±0.069 −1.0 SG vs HG 0.880
Serum OC (ng/ml) SG 13.6±2.4 17.4±2.6 +31.1∗ 0.033∗∗ SG vs CG 0.051
HG 13.6±6.0 16.6±5.4 +30.2∗ HG vs CG 0.012∗∗∗
CG 14.2±3.4 14.1±3.6 +4.7 SG vs HG 0.880
Serum NTx (nmol BCE/L) SG 31.7±6.9 26.9±5.6 −12.7∗ 0.034∗∗ SG vs CG 0.169
HG 27.8±7.2 20.3±6.1 −25.7∗ HG vs CG 0.012∗∗∗
CG 27.8±5.2 29.6±8.9 +8.9 SG vs HG 0.190
QUALEFFO-41 (0–100) SG 30.1±10.0 22.5±7.31 −24.6∗ <0.001∗∗ SG vs CG <0.001∗∗∗
HG 38.1±19.2 24.6±10.0 −33.0∗ HG vs CG <0.001∗∗∗
CG 28.7±7.2 28.0±8.5 −3.7 SG vs HG 0.104
BMD, bone mineral density; OC, osteocalcin; NTx, N-telopeptides of type I collagen; QUALEFFO, quality of life questionnaire of the European
Foundation for osteoporosis; SG, strengthening group; HG, high-impact group; CG, control group. ∗ Statistically significant changes in the
within-groups comparison by Wilcoxon Signed-rank test (α = 0.05). ∗∗ Statistically significant changes in the between-groups comparison by
Kruskal-Wallis test (α = 0.05). ∗∗∗ Statistically significant in the pair-wise comparisons by Mann-Whitney U tests with Bonferroni correction
(p < 0.017).

more pronounced in the high-impact (−33.0%) and impact exercise training program increased BMD at
the strengthening (−24.6%) groups compared to the the lumbar spine and femoral neck regions; more-
controls (−3.7%). When the groups were compared, over, the high-impact training increased bone forma-
there was a significant (p < 0.001) difference between tion and decreased bone resorption over six months
the three groups. The pair-wise analysis indicated that in postmenopausal women with osteopenia. Our data
both the strengthening (p < 0.001) and the high-impact also suggested that both strengthening and high-impact
(p < 0.001) exercises improved HRQoL compared to training programs increased the HRQoL in the partici-
the controls (Table 4). pants.

4.2. Bone mineral density

4. Discussion
In this study, there was an increase in the BMD
4.1. Overall effectiveness values at the lumbar in both training groups over six
months; and, only the difference between the high-
This prospective randomized controlled interven- impact and control groups was statistically signifi-
tional trial demonstrated that the supervised high- cant. This finding is important, because it suggests
 H. Basat et al. / The effects of strengthening and high-impact exercises on bone metabolism and quality of life 433

that training programs, especially high-impact exer- the resorption and formation processes at the same
cises increase BMD or at least decrease the rate of time (coupling phenomenon). Under normal condi-
bone loss during the postmenopausal period. To some tions, bone resorption/formation is tightly coupled,
extent, these findings were similar to those observed however, in osteoporotic or osteopenic bone; there is
in other studies such as the EFOPS (Erlangen Fitness an uncoupling in the process of bone turnover [33].
Osteoporosis Prevention Study) which suggested that Therefore, it was possible that, the changes in bone
weight-bearing and strengthening exercises in early turnover markers in the current study were the reflec-
postmenopausal osteopenic women increased BMD at tion of an uncoupled bone resorption/formation pro-
the lumbar spine by 1.3% over the 14 months com- cess in response to exercise. There is an inconsistency
pared with a 1.2% decrease in the controls [22]. When in the literature regarding the changes in the markers
the third- and forth-year results of the EFPOS were of bone turnover. In some traditional studies, OC was
reviewed, BMD at the lumbar spine continued to in- considered to be markers of bone formation and any
crease in the training group in contrast to the con- results that suggested there was an increase in serum
trol group [23–25]. Chien et al. [26] reported a statis- OC levels interpreted as improvement in the BMD val-
tically non-significant increase in the exercise group ues, however, more recent studies [34] have suggested
and significant decrease in the control group on BMD that serum OC is a marker of bone turnover rather than
at the L2 –L4 after the six months high-impact train- bone formation. Bemben et al. [28] and Danz et al. [35]
ing program. Nelson et al. [18] also reported that 12- reported that exercise programs increased serum OC
month strengthening training program increased BMD levels compared to controls. On the contrary, Shibata
at the lumbar spine. However, other studies reported et al. [36] demonstrated that walking and high-impact
that strengthening and/or high-impact training pro- exercises did not increase OC levels. Similarly, no
grams did not increase BMD at the lumbar spine [17, changes in serum OC in the second-year results of the
27,28]. EFOPS was reported [37]. In terms of bone resorp-
In the current study, we found that six months high- tion makers, while Yamazaki et al. [38] reported that
impact training increased BMD at the femoral neck re-
walking exercises decreased urine NTx levels in post-
gion compared to the strengthening training and the
menopausal osteopenic women after the 12 months ex-
controls. This may reflect the effect of jumping exer-
ercise training, Shibata et al. [36] demonstrated that
cises such as Jump rope, which causes highly mechan-
walking and high-impact exercises did not change NTx
ical loading on femoral neck. These results are consis-
levels significantly. Moreover, no significant changes
tent with those of Chien et al. [26] and Vainionpää et
in serum NTx in the second-year results of the EFOPS
al. [27], which reported that the high-impact training
was reported [38].
programs were effective for improving femoral neck
BMD. Additionally, systematic review of randomized
4.4. Health-related quality of life
controlled trials of exercise for osteoporotic women re-
vealed that only the high-impact training was effec-
In addition to positive findings in preventing of bone
tive for increasing femoral neck BMD [12]. In a re-
loss, both training programs improved the HRQoL in
cent meta-analysis of 6 randomized controlled trials,
postmenopausal women. However, there was no sig-
Babatunde et al. [29] reported that, brief high-impact
nificant difference between the two different exercise
exercises improve BMD at the hip but not at the lum-
bar spine. In contrast, findings of Bassey et al. [30] and training programs to each other. These results are con-
Korpelainen et al. [31] revealed that high-impact exer- sistent with findings of Angin et al. [39] and Liu-
cises did not increase BMD at the femoral neck. While Ambrose et al. [40], which reported that training pro-
Nelson et al. [18] reported that 12-month strengthening grams were effective in improving the HRQoL. In a
training program increased BMD at the femoral neck systematic review and meta-analysis in 2009, Li et
in postmenopausal women, Chilibeck et al. [32] and al. [19] revealed that exercise improves the HRQoL for
Bemben et al. [28] did not find any improvement on those with osteoporosis. However, Carter et al. [41] did
BMD at the femoral neck after 6 or 12 months training, not find any significant change in HRQoL after the 20-
respectively. week training program.

4.3. Bone turnover markers 4.5. Limitations and strength of the study

It is known that, bone turnover markers reflect the Our study had several limitations, and its positive
general process of bone turnover, which contains both or negative findings must be interpreted with atten-
434 H. Basat et al. / The effects of strengthening and high-impact exercises on bone metabolism and quality of life

tion. First and most notable, the sample size of the [2] Lane NE. Epidemiology, etiology, and diagnosis of osteo-
study was too small to detect a statistically significant porosis. Am J Obstet Gynecol. 2006; 194(2 Suppl): S3-11.
[3] Johnell O, Kanis JA. An estimate of the worldwide prevalence
difference in some of the within- or between-groups and disability associated with osteoporotic fractures. Osteo-
comparisons, especially in the comparison of differ- poros Int. 2006; 17(12): 1726-33.
ence between training programs. Second, the follow- [4] Holroyd C, Cooper C, Dennison E. Epidemiology of osteo-
up period of our study was six months, which was not porosis. Best Pract Res Clin Endocrinol Metab. 2008; 22(5):
671-85.
long enough to observe the exact effects of exercise on [5] Chapurlat R. Epidemiology of osteoporosis. J Soc Biol. 2008;
BMD or bone turnover markers. Third, the relatively 202(4): 251-5.
low compliance rate to the training program was a con- [6] Cummings SR, Melton LJ. Epidemiology and outcomes of
cern in this study due to the large number of the super- osteoporotic fractures. Lancet. 2002; 359(9319): 1761-7.
[7] Jahelka B, Dorner T, Terkula R, Quittan M, Bröll H, Erlacher
vised sessions over the course of study. Despite these L. Health-related quality of life in patients with osteopenia or
limitations, our study had a number of strengths. First, osteoporosis with and without fractures in a geriatric rehabil-
in our study, all training sessions were supervised by itation department. Wien Med Wochenschr. 2009; 159(9-10):
the investigators (instead of home-based), which mini- 235-40.
[8] Hallberg I, Rosenqvist AM, Kartous L, Löfman O, Wahlström
mizes variability in training sessions. Second, analysis O, Toss G. Health-related quality of life after osteoporotic
of bone density together with serum bone markers en- fractures. Osteoporos Int. 2004; 15(10): 834-41.
abled us to better detect the bone response to training. [9] Delaney MF, LeBoff MS. Metabolic bone disease. In: Ruddy
Finally, in addition to examining the general effects of S, Harris ED, Sledge CB, editors. Kelley’s textbook of
rheumatology. 6th ed. Philadelphia: Saunders; 2001. p. 1635-
the exercise training in postmenopausal women, this 52.
study compared the effects of two different types of ex- [10] Kleerekoper M, Avioli LV. Evaluation and treatment of post-
ercise (strengthening and high impact) with each other menopausal osteoporosis. In: Favus MJ, editor. Primer on the
metabolic bone diseases and disorders of mineral metabolism,
and additionally, investigated the effects of training
2nd ed. New York: Raven Press; 1993. p. 223-9.
programs on HRQoL. [11] Shaw JM, Witzke KA. Exercise for skeletal health and osteo-
porosis prevention. In: Roitman JL, editor. ACSM’s resource
manual for guidelines for exercise testing and prescription.
Baltimore: Williams and Wilkins; 1998. P. 288-93.
5. Conclusions
[12] Wallace BA, Cumming RG. Systematic review of randomized
trials of the effect of exercise on bone mass in pre- and post-
These findings indicate that six months supervised menopausal women. Calcif Tissue Int. 2000; 67(1): 10-8.
high-impact training can be effective in prevention of [13] Wolff I, van Croonenborg JJ, Kemper HC, Kostense PJ, Twisk
JW. The effect of exercise training programs on bone mass: a
bone loss at the lumbar spine and femoral neck in post- meta-analysis of published controlled trials in pre- and post-
menopausal women. Moreover, the HRQoL can be im- menopausal women. Osteoporos Int. 1999; 9(1): 1-12.
proved with both strengthening and high-impact exer- [14] National Osteoporosis Foundation. Physicians Guide to Pre-
cises. Although these results are encouraging, large- vention and treatment of Osteoporosis. Washington, DC:
Copyright NOF; 2008.
scale interventional studies with longer follow-up pe- [15] Brown JP, Fortier M, Frame H, Lalonde A, Papaioannou A,
riods are still necessary to confirm these results. Senikas V, Yuen CK. Osteoporosis Guidelines Committee.
Canadian Consensus Conference on osteoporosis, 2006 up-
date. J Obstet Gynaecol Can. 2006; 28(2 Suppl 1): S95-112.
[16] Bergström I, Landgren B, Brinck J, Freyschuss B. Physical
Acknowledgments training preserves bone mineral density in postmenopausal
women with forearm fractures and low bone mineral density.
This work was supported in part by funding from Osteoporos Int. 2008; 19(2): 177-83.
the Scientific Research Projects Unit of the Istanbul [17] Kerr D, Ackland T, Maslen B, Morton A, Prince R. Resistance
training over 2 years increases bone mass in calcium-replete
University (Ref. T-793/27122005). The authors of the postmenopausal women. J Bone Miner Res. 2001; 16(1): 175-
study just had control on the study design, data collec- 81.
tion and analysis, also on interpretations of results or [18] Nelson ME, Fiatarone MA, Morganti CM, Trice I, Greenberg
writing of the paper. RA, Evans WJ. Effects of high-intensity strength training on
multiple risk factors for osteoporotic fractures. A randomized
controlled trial. JAMA. 1994; 272(24): 1909-14.
[19] Li WC, Chen YC, Yang RS, Tsauo JY. Effects of exer-
References cise programmes on quality of life in osteoporotic and os-
teopenic postmenopausal women: a systematic review and
[1] Consensus development conference: diagnosis, prophylaxis meta-analysis. Clin Rehabil. 2009; 23(10): 888-96.
and treatment of osteoporosis. Am J Med. 1993; 94(6): 646- [20] Kocyigit H, Gulseren S, Erol A, Hizli N, Memis A. The reli-
50. ability and validity of the Turkish version of Quality of Life
 H. Basat et al. / The effects of strengthening and high-impact exercises on bone metabolism and quality of life 435

Questionnaire of the European Foundation for Osteoporosis bone mineral density in elderly women with low BMD: a
(QUALEFFO). Clin Rheumatol. 2003; 22(1): 18-23. population-based randomized controlled 30-month interven-
[21] Kohrt WM, Bloomfield SA, Little KD, Nelson ME, Yingling tion. Osteoporos Int. 2006; 17(1): 109-18.
VR. American College of Sports Medicine. American College [32] Chilibeck PD, Davison KS, Whiting SJ, Suzuki Y, Janzen CL,
of Sports Medicine Position Stand: physical activity and bone Peloso P. The effect of strength training combined with bis-
health. Med Sci Sports Exerc. 2004; 36(11): 1985-96. phosphonate (etidronate) therapy on bone mineral, lean tis-
[22] Kemmler W, Engelke K, Weineck J, Hensen J, Kalender WA. sue, and fat mass in postmenopausal women. Can J Physiol
The erlangen fitness osteoporosis prevention study: a con- Pharmacol. 2002; 80(10): 941-50.
trolled exercise trial in early postmenopausal women with low [33] Seibel MJ. Clinical application of biochemical markers of
bone density-first-year results. Arch Phys Med Rehabil. 2003; bone turnover. Arq Bras Endocrinol Metabol. 2006; 50(4):
84(5): 673-82. 603-20.
[23] Kemmler W, Engelke K, von Stengel S, Weineck J, Lauber D, [34] Azinge EC, Bolarin DM. Osteocalcin and bone-specific alka-
Kalender WA. Long-term four-year exercise has a positive ef- line phosphatase in sickle cell haemoglobinopathies. Niger J
fect on menopausal risk factors: the Erlangen Fitness Osteo- Physiol Sci. 2006; 21(1-2): 21-5.
porosis Prevention Study. J Strength Cond Res. 2007; 21(1): [35] Danz AM, Zittermann A, Schiedermaier U, Klein K, Hötzel
232-9. D, Schönau E. The effect of a specific strength-development
[24] Engelke K, Kemmler W, Lauber D, Beeskow C, Pintag exercise on bone mineral density in perimenopausal and post-
R, Kalender WA. Exercise maintains bone density at spine menopausal women. J Womens Health. 1998; 7(6): 701-9.
and hip EFOPS: a 3-year longitudinal study in early post- [36] Shibata Y, Ohsawa I, Watanabe T, Miura T, Sato Y. Ef-
menopausal women. Osteoporos Int. 2006; 17(1): 133-42. fects of physical training on bone mineral density and bone
[25] Kemmler W, von Stengel S, Weineck J, Lauber D, Kalender metabolism. J Physiol Anthropol Appl Human Sci. 2003;
W, Engelke K. Exercise effects on menopausal risk factors 22(4): 203-8.
of early postmenopausal women: 3-yr Erlangen fitness osteo- [37] Kemmler W, Lauber D, Weineck J, Hensen J, Kalender W, En-
porosis prevention study results. Med Sci Sports Exerc. 2005; gelke K. Benefits of 2 years of intense exercise on bone den-
37(2): 194-203. sity, physical fitness, and blood lipids in early postmenopausal
[26] Chien MY, Wu YT, Hsu AT, Yang RS, Lai JS. Efficacy osteopenic women: results of the Erlangen Fitness Osteo-
of a 24-week aerobic exercise program for osteopenic post- porosis Prevention Study (EFOPS). Arch Intern Med. 2004;
menopausal women. Calcif Tissue Int. 2000; 67(6): 443-8. 164(10): 1084-91.
[27] Vainionpää A, Korpelainen R, Leppäluoto J, Jämsä T. Effects [38] Yamazaki S, Ichimura S, Iwamoto J, Takeda T, Toyama
of high-impact exercise on bone mineral density: a random- Y. Effect of walking exercise on bone metabolism in post-
ized controlled trial in premenopausal women. Osteoporos menopausal women with osteopenia/osteoporosis. J Bone
Int. 2005; 16(2): 191-7. Miner Metab. 2004; 22(5): 500-8.
[28] Bemben DA, Fetters NL, Bemben MG, Nabavi N, Koh ET. [39] Angin E, Erden Z. The effect of group exercise on post-
Musculoskeletal responses to high- and low-intensity resis- menopausal osteoporosis and osteopenia. Acta Orthop Trau-
tance training in early postmenopausal women. Med Sci matol Turc. 2009; 43(4): 343-50.
Sports Exerc. 2000; 32(11): 1949-57. [40] Liu-Ambrose TY, Khan KM, Eng JJ, Lord SR, Lentle B,
[29] Babatunde OO, Forsyth JJ, Gidlow CJ. A meta-analysis of McKay HA. Both resistance and agility training reduce back
brief high-impact exercises for enhancing bone health in pre- pain and improve health-related quality of life in older women
menopausal women. Osteoporos Int. 2012; 23(1): 109-19. with low bone mass. Osteoporos Int. 2005; 16(11): 1321-9.
[30] Bassey EJ, Rothwell MC, Littlewood JJ, Pye DW. Pre- and [41] Carter ND, Khan KM, McKay HA, Petit MA, Waterman
postmenopausal women have different bone mineral density C, Heinonen A, et al. Community-based exercise program
responses to the same high-impact exercise. J Bone Miner reduces risk factors for falls in 65- to 75-year-old women
Res. 1998; 13(12): 1805-13. with osteoporosis: randomized controlled trial. CMAJ. 2002;
[31] Korpelainen R, Keinänen-Kiukaanniemi S, Heikkinen J, 167(9): 997-1004.
Väänänen K, Korpelainen J. Effect of impact exercise on