Calcif Tissue Int (2015) 96:490–499

DOI 10.1007/s00223-015-9976-6

ORIGINAL RESEARCH

Low-Load Very High-Repetition Resistance Training Attenuates

Bone Loss at the Lumbar Spine in Active Post-menopausal
Women
Vaughan P. Nicholson • Mark R. McKean •
Gary J. Slater • Ava Kerr • Brendan J. Burkett

Received: 7 November 2014 / Accepted: 27 February 2015 / Published online: 14 March 2015
Ó Springer Science+Business Media New York 2015

Abstract This study determined the effect of 6 months of Introduction

low-load very high-repetition resistance training on bone
mineral density (BMD) and body composition in nonos- Age-related reductions in bone mineral density (BMD)
teoporotic middle-aged and older women. Fifty healthy, occur at most skeletal sites after the third or fourth decade
active community-dwelling women aged 56–75 years took in men and women [1] and such reductions are typically
part in the two-group, repeated-measures randomized accelerated in women after menopause [2, 3]. Fortunately,
controlled trial. Participants either undertook 6 months of it is now well established that resistance training can
low-load very high-repetition resistance training in the maintain and improve BMD in post-menopausal women
form of BodyPumpTM or served as control participants. [4–6]. Typically, high-intensity (or high-load) training in-
Outcome measures included BMD at the lumbar spine, hip, terventions utilizing training loads over 75–80 % of one-
and total body; total fat mass; fat-free soft tissue mass and repetition maximum (1RM) have been more successful at
maximal isotonic strength. Significant group-by-time in- improving BMD when compared to low-load training
teractions were found for lumbar spine BMD and maximal (\50 % 1RM) interventions [7, 8], though the low-load
strength in favor of the BodyPumpTM group. No favorable interventions assessed have also used relatively low
effects were found for hip BMD, total body BMD, total fat repetition ranges. Interestingly, when training volumes are
mass, or fat-free soft tissue mass. Three participants equivalent both the low- and high-load training may be
withdrew from the intervention group due to injury or fear similarly effective at improving BMD [5].
of injury associated with training. Under the conditions Low-load high-repetition resistance training has been
used in this research, low-load very high-repetition resis- shown to improve maximal strength, balance, and func-
tance training is effective at attenuating losses in lumbar tional task performance in older populations [9, 10] but to
spine BMD compared to controls in healthy, active women date there has been no research exploring its effect on
aged over 55 years but did not influence hip and total body BMD when very low loads (\30 % 1RM) and very high
BMD or fat mass and fat-free soft tissue mass. repetitions ([60) are used. Animal bone loading ex-
periments have demonstrated that increasing repetitions at
Keywords Bone mass
Exercise
Strength
submaximal loads increases the bone response compared
Middle-age
Elderly with low repetitions [11, 12] suggesting that high cycle
numbers compensate for low strain magnitudes. Addition-
ally, low-load high-repetition training with vascular re-
striction produces similar changes in markers of bone
turnover as high-load training [13]. As such, low-load very
V. P. Nicholson (&)
M. R. McKean
G. J. Slater
A. Kerr
high-repetition ([60 repetitions) resistance training may
B. J. Burkett provide an alternative to high-load resistance training in
School of Health and Sport Sciences, University of the Sunshine
middle-aged and older adults. Moreover, low-load very
Coast, 90 Sippy Downs Drive, Sippy Downs, QLD 4556,
Australia high-repetition resistance programs such as BodyPumpTM
e-mail: Vaughan.Nicholson@research.usc.edu.au are available in over 14,000 fitness facilities globally [14].

123
V. P. Nicholson et al.: Bone density and low-load very high repetition resistance training 491

This pre-choreographed group class uses light weights and size of 25 per group to identify an expected difference of
very high (80–100) repetitions for each exercise so it 1 % between groups for lumbar spine BMD with a standard
provides a platform to assess the effect of low-load very deviation of 0.1 g/cm2 [4, 8]. To account for a 15 % at-
high-repetition resistance training and as recent trends trition rate a sample size of 29 per group was required.
indicate that a growing number of over 55 s are under- Participants undertook all testing at university testing fa-
taking fitness center-based activities [15], the apparent ef- cilities at baseline and after 6 months. Participants were
fectiveness of such activities warrants examination. allocated to either the intervention group (PUMP) or con-
Furthermore, social factors are a key motivator for exercise trol group (CON) on a 1:1 ratio using a computer-generated
participation in middle-aged and older adults [16] so par- random number list after baseline data collection.
ticipation in a group class may promote greater compliance PUMP participants were instructed to attend two
among participants and provide prolonged benefits. BodyPumpTM classes per week for 6 months. Each class
The aim of this study was to assess whether 6 months of was approximately 50 min in duration and included exer-
low-load very high-repetition resistance training in the cises such as squats, lunges, and chest press, utilizing light
form of BodyPumpTM would improve BMD and fat-free weights and a very high number of repetitions. Body-
soft tissue mass in a group of healthy, active females aged PumpTM release 83 was used for the duration of the pro-
over 55 years. It was hypothesized that BMD at the lumbar gram (Table 1). The pre-choreographed class is separated
spine and hip would be better preserved in the intervention into ten tracks that last up to 6 min each. Each track fo-
group compared with the control group and that fat-free cuses on a particular movement pattern such as squats or a
soft tissue mass would increase after 6 months of group particular muscle group such as biceps or triceps. All
training. Additionally it was hypothesized that maximal classes were conducted at a local fitness facility where
dynamic strength would increase following such an PUMP participants were provided with complimentary
intervention. access. Classes were exclusively attended by project par-
ticipants and all classes were instructed by experienced
group fitness instructors who were not associated with
Methods testing or recruitment of participants. The first 4 weeks of
the intervention were used to ensure if participants com-
Participants pleted all exercises with appropriate technique. From week
5 onwards all classes were instructed at a level that one
Apparently healthy women aged between 55 and 75 years would expect to encounter if they took part in a Body-
were invited to take part in the study. All participants were PumpTM class at any fitness center providing such classes.
recruited through local advertising and an adult education The weights lifted for each exercise were self-selected but
facility. All participants were physically active non-fallers were guided by general recommendations provided by the
who had not undertaken formal resistance training in the class instructor. Participants recorded the weight lifted for
previous year. All women were at least 5 years post- the selected exercises during each session and these records
menopause and provided details of medical history and were later used to determine changes in load over the
current medication use. Exclusion criteria included: acute course of the program (Table 2). Control (CON) par-
or terminal illness, myocardial infarction in the past ticipants did not undergo any training and were instructed
6 months, recent low impact fracture, osteoporosis, use of to maintain their current level of physical activity for the
hormone replacement therapy (HRT), and other medica- duration of the study.
tions known to affect bone metabolism in the previous
2 years, or any condition that would interfere with mod- Measurements
erate intensity exercise participation. A total of 50 women
aged 58–75 years took part in the intervention after pro- Participants were tested on two occasions: the first
viding informed consent conforming to the Declaration of assessment was conducted prior to the beginning of train-
Helsinki, approved by the Human Research Ethics Com- ing and the second assessment was conducted immediately
mittee of the university. after the 6-month intervention.

Study Design and Intervention Body composition

A two-group, repeated-measures, randomized control trial Body composition assessments were performed using dual-
was used to investigate the effects of 6 months of Body- energy X-ray absorptiometry (DXA). Participants were
PumpTM training. A priori power calculation with power overnight fasted and had not undergone any exercise in the
set at 0.8 and an alpha of 0.05 identified a required sample morning before measurements. Participants wore light

123
492 V. P. Nicholson et al.: Bone density and low-load very high repetition resistance training

Table 1 General characteristics Track number—name Exercise Approximate repetitions

of a BodyPumpTM session
(release 83) 1—Warm up Deadlift 24
Dead row 6
Upright row to overhead press 4
Overhead press 16
Squats 4
Lunges 8 (per leg)
Bicep curl 16
2—Squats Squats 108
3—Chest Chest press 94
4—Back Dead row 41
Dead lift 15
Clean & press 12
Power press 7
5—Triceps Triceps extension 22
Triceps press 52
Standing overhead extension 30
6—Biceps Bicep curl 68
7—Lunges Squat 50
Jump squat 16
Pulse jump squat 16
Lunge 25 (per leg)
8—Shoulders Push-up 36
Standing rear deltoid raise 8
Standing side raise 18
Upright row 8
Overhead press 16
9—Core Crunch 40
Hover 14
10—Cool down Stretch/mobility

Table 2 Weights used for squats and chest press during the 6-month was assessed by a wall-mounted stadiometer to the nearest
training intervention
0.001 m (Table 3).
Squat Chest press BMD of the lumbar spine (anteroposterior L2–4) and
Mean weight ± SD week 1 2.9 ± 1.9 kg 2.6 ± 1.1 kg hip (femoral neck, total hip, and trochanter) were assessed
Mean weight ± SD week 5 5.5 ± 3.9 kg 4.6 ± 5.2 kg
following the established protocols using pencil beam
Mean weight ± SD week 13 8.0 ± 3.1 kg 5.6 ± 4.2 kg
DXA (Lunar DPX Pro, GE Healthcare, UK) with analysis
Mean weight ± SD week 26 8.2 ± 3.1 kg 6.0 ± 4.4 kg
performed using enCoreTM software (GE Healthcare, UK).
Total body BMD, fat mass, and fat-free soft tissue mass
% 1RM ± SD week 1 3.7 ± 2.2 % 11.8 ± 6.3 %
were assessed by a whole-body scan. The DXA was
% 1RM ± SD week 5 7.0 ± 2.7 %a 20.7 ± 9.2 %c
a,b calibrated with phantoms as per the manufacturer’s
% 1RM ± SD week 13 10.0 ± 3.6 % 24.8 ± 10.1 %c
a,b guidelines each day before measurement. The short-term
% 1RM ± SD week 26 10.6 ± 3.8 % 27.4 ± 9.9 %c
coefficient of variance (CV) for BMD and body composi-
a
Significantly (p \ 0.001) more than week 1 tion in our laboratory ranged from 0.5 to 1.5 % and 0.6 to
b
Significantly (p \ 0.05) more than week 5 2.2 %, respectively. For the whole-body scan, the scanning
c
Significantly (p \ 0.05) more than week 1 mode was automatically chosen by the DXA machine with
scans for all but two participants (thick mode) scanned in
sports clothes with all jewelery and metal objects removed the standard mode. Subjects were centrally aligned in the
before each scan. Body weight was measured to the nearest scanning area for all scans. All scans were performed and
0.1 kg on electronic scales (Tanita, Japan) and body height analyzed by a single trained and licensed technician who

123
V. P. Nicholson et al.: Bone density and low-load very high repetition resistance training 493

Table 3 Baseline participant Pump (N = 28) Control (N = 29)

characteristics
Age (years) 66.0 ± 4.1 65.6 ± 4.7
Height (cm) 164.4 ± 4.4 163.3 ± 5.5
Values reported as mean ± SD Weight (kg)a 70.6 ± 9.1 66.8 ± 10.7
a
Significant difference BMI (kg/m2) 26.0 ± 3.2 24.5 ± 2.9
(p \ 0.05) between groups at Number of prescribed medications 1.5 ± 1.5 1.3 ± 1.1
baseline

was blinded to group allocation. The scans were analyzed general linear model (GLM) was used to determine group
automatically by the software but regions of interest were (PUMP, CON) and time (baseline, 6 months) effects on
subsequently confirmed by the technician. primary outcomes. Baseline values for any measures that
were different between groups at baseline were used as a
Strength, Diet, and Activity Assessment covariate in the GLM. Partial eta squared was used to
determine the effective size for each outcome variable.
Assessment of 1RM was conducted for the incline leg press When the GLM revealed significant interaction (time 9
(Calgym, Australia) and Smith machine bench press (Elite, group), paired samples t tests were used to assess differ-
Australia). A familiarization session was held ap- ences between initial and final values for each group. A
proximately 1 week prior to baseline 1RM testing to ensure repeated-measures ANOVA with pairwise comparisons
correct lifting technique using submaximal loads. Testing and Bonferroni correction was used to determine differ-
using established protocols [17] commenced after a light ences between the amounts of weight lifted (in terms of
cycling warm-up with leg press assessed prior to Smith percentage 1RM) for squats and chest press at four time
machine bench press. Squat 1RM was subsequently pre- points (week 1, week 5, week 13, and week 26) for PUMP.
dicted from the leg press assessment [18] and the amount Data were analyzed using both per-protocol (C85 %
of weight lifted during the program was monitored for both compliance) and intention-to-treat (ITT). As there were no
squats and chest press. significant differences between the two analyses, results
Dietary intake was assessed from self-reported 3-day presented are for the ITT data. Analyses were performed
food records completed at baseline and follow-up. Par- using IBM SPSS (version 21). Statistical significance was
ticipants were instructed to record the type and amount of set at p \ 0.05.
all food, drink, and supplements consumed over three
consecutive days (2 week days and 1 weekend day). All
records were entered into FoodWorks 7 (Xyris Software, Results
Australia) and daily consumption of total energy (kJ),
protein (g/kg body weight), and calcium (mg) were ana- PUMP Attendance, Adverse Events, and Training
lyzed. Energy expenditure derived from exercise and
physical activity was estimated by a 7-day activity diary. A Fifty participants aged between 58 and 75 years (PUMP
metabolic equivalent value was assigned to each activity n = 24, age = 66 ± 4.4 years, CON n = 26, age = 66 ±
and was used to determine the average amount of energy 4.5 years) completed all baseline and follow-up testing
used for exercise/planned physical activity (including with four participants lost to follow-up from PUMP and
BodyPump classes) during the program for both groups three lost to follow-up from CON (Fig. 1). Three of the
[19]. four participants lost to follow-up from PUMP were due to
fear of injury or injury associated with the intervention.
Statistical Analyses One participant (aged 64) withdrew due to intermittent
neck pain associated with shoulder press and chest press
All data are reported as mean and standard deviation (SD). and although lifting modifications were implemented the
Primary outcomes were changes from baseline in BMD, fat participant was unable to continue the program beyond
mass, and fat-free soft tissue mass in response to the 8 weeks. A further participant (aged 60) reported an ex-
6-month intervention. Secondary outcomes included acerbation of a previous knee complaint with squats and
6-month changes from baseline in maximal strength, di- lunges that did not improve following program modifica-
etary intake, energy expenditure, and changes in weight tions and ongoing physiotherapy treatment. A further par-
lifted during each class during the PUMP intervention. ticipant (aged 63) withdrew after 4 weeks of training due to
Potential differences between groups at baseline were fear of a neck or back injury associated with lifting. Mean
assessed by independent t tests. A repeated-measures PUMP attendance was 48 (±12) classes over 26 weeks

123
494 V. P. Nicholson et al.: Bone density and low-load very high repetition resistance training

Fig. 1 Project flow chart

Assessed for eligibility (n=66)

Enrolment

Excluded (n= 9)
Not meeng criteria (n= 7)
Declined to parcipate (n= 2)

Completed Baseline tesng

(n= 57)

Randomizaon &
Allocated to intervenon (n=28) Allocaon Allocated to control (n=29)

Follow-Up
Lost to follow-up (n= 4)
Lost to follow-up (n= 3)
Family commitments (n=1), neck pain (n=1), knee pain (n=1),
Fractured scaphoid (n=1, cycling injury),
fear of injury from intervenon (n=1)
fractured rib (n=1), knee surgery (n=1)

Final Analysis

Analyzed (n= 24) Analyzed (n= 26)

(range = 16–52 classes) resulting in an attendance com- (t (25) = -0.67, p = 0.51). There were no group-by-time
pliance of 89 %. There was an increase in the amount of interactions on femoral neck, total hip, or trochanter BMD.
weight lifted during PUMP (in terms of percentage 1RM)
for squats from week 1 to week 5 (p \ 0.001) and week 5 The Effect of PUMP on Body Composition and Body
to 13 (p = 0.004). There was no significant difference in Weight
the amount of weight lifted between weeks 13 and 26 for
squats (p = 0.76). Significant increases were only evident There were no baseline group differences in total body fat
for chest press from week 1 to 5 (p = 0.001). There were mass or fat-free soft tissue mass. There were no significant
no evident increases from week 5 to 13 (p = 0.15), or group-by-time interactions or time effects on total fat mass
week 13 to 26 (p = 0.25) (Table 2). or total fat-free soft tissue mass (Table 4). There was no
group-by-time interaction for body weight but there was a
The Effect of PUMP on BMD significant time effect (F (1, 49) = 40.25, p \ 0.001) re-
lated to significant reductions from baseline to follow-up in
There were no group differences in any BMD measure at both PUMP (t (23) = 4.85, p \ 0.001) and CON (t (25) =
baseline. There was a significant group-by-time interaction 4.09, p \ 0.001).
for lumbar spine BMD (F (1, 49) = 8.58, p = 0.005)
(Table 4). Paired samples t tests demonstrated a non- The Effect of PUMP on Maximal Strength
significant increase for PUMP (t (23) = 1.61, p = 0.121)
and a significant reduction in BMD for CON (t (25) = There were no group differences in leg press or Smith
2.53, p = 0.018) in lumbar BMD. There was a significant machine bench press at baseline. There were significant
group-by-time interaction for total body BMD (F (1, 49) = group-by-time interactions for both leg press (F (1, 48) =
4.07, p = 0.049) with paired samples t tests showing a 17.56, p \ 0.001) and Smith machine bench press
significant reduction in BMD for PUMP (t (23) = (F (1, 48) = 11.03, p = 0.002) (Table 4). Paired t tests
2.30, p = 0.031) and a non-significant change for CON demonstrated a significant increase for PUMP in both leg

123
 Table 4 Pre- and post-intervention values for BMD, body composition, maximal strength, dietary intake, and weekly energy expenditure
Outcome measure Pump (N = 24) Control (N = 26) Group-by-time Partial g2 for
interaction p group-by-time
Baseline Follow-up Mean percentage Baseline Follow-up Mean percentage
change change

Lumbar BMD (g/cm2) 1.088 ± 0.121 1.099 ± 0.122 1.01 ± 3.24 1.098 ± 0.136 1.075 ± 0.143 -2.09 ± 4.15 0.005 0.149
Femoral neck BMD (g/cm2) 0.875 ± 0.113 0.876 ± 0.118 0.11 ± 2.78 0.857 ± 0.108 0.848 ± 0.136 -1.05 ± 1.80 0.781 0.002
Total hip BMD (g/cm2) 0.937 ± 0.099 0.935 ± 0.102 -0.21 ± 2.13 0.937 ± 0.099 0.909 ± 0.124 -2.99 ± 4.12 0.244 0.048
Trochanter BMD (g/cm2) 0.762 ± 0.089 0.758 ± 0.088 -0.52 ± 3.61 0.747 ± 0.107 0.746 ± 0.119 -0.13 ± 2.08 0.618 0.006
2
Total body BMD (g/cm ) 1.059 ± 0.235 1.053 ± 0.235 -0.57 ± 1.37 1.087 ± 0.095 1.089 ± 0.096 0.18 ± 1.24 0.049 0.077
Total fat mass (kg) 28.19 ± 9.40 27.71 ± 8.75 -1.70 ± 5.32 24.21 ± 7.83 23.79 ± 7.66 -1.73 ± 5.65 0.905 0.000
Total fat-free soft tissue mass (kg) 37.50 ± 4.81 36.71 ± 3.86 -2.11 ± 2.75 36.07 ± 3.36 36.17 ± 3.02 0.28 ± 2.62 0.352 0.018
Body weight (kg) 70.3 ± 9.5 68.6 ± 10.0 -2.4 ± 2.7 64.4 ± 9.8 63.1 ± 9.5 -2.1 ± 2.3 0.448 0.012
1RM leg press (kg) 132 ± 28 147 ± 31 12 ± 11 123 ± 25 122 ± 25 -0.81 ± 12 \0.001 0.268
1RM Smith bench press (kg) 22 ± 5 25 ± 5 14 ± 17 23 ± 4 23 ± 4 0.0 ± 11 0.002 0.187
Energy (kJ) 7419 ± 1729 6435 ± 1468 -13 ± 28 7710 ± 2474 6904 ± 1321 -10 ± 29 0.761 0.002
Protein (g/kg body weight) 1.32 ± 0.31 1.15 ± 0.25 -13 ± 27 1.48 ± 0.62 1.28 ± 0.51 -14 ± 36 0.892 0.000
Calcium (mg) 929 ± 249 770 ± 256 -17 ± 37 854 ± 349 800 ± 390 -6 ± 39 0.548 0.009
V. P. Nicholson et al.: Bone density and low-load very high repetition resistance training

Weekly METs 32 ± 21 41 ± 19 30 ± 22 36 ± 24 37 ± 25 3±9% 0.027 0.096

Values presented as mean ± SD
BMD bone mineral density, 1RM one-repetition maximum, METs metabolic equivalents
495

123
496 V. P. Nicholson et al.: Bone density and low-load very high repetition resistance training

press (t (23) = 6.19, p \ 0.001) and bench press women in that study [5] were on HRT, and although a
(t (23) = 4.56, p \ 0.001). subgroup analysis was performed to compare HRT and
non-HRT participants there was no control group for direct
Dietary Intake and Energy Expenditure comparison. There were no evident training effects at any
hip site in this study which is in contrast to many previous
There were no differences between groups at baseline for resistance training studies in post-menopausal women [6,
dietary intake or energy expenditure. There were no sig- 7] that have typically identified a maintenance or im-
nificant group-by-time interactions for any dietary intake provement in BMD at least at one hip site compared to
measure. There were however significant time effects for controls. The results of previous low-load resistance
total energy (F (1, 44) = 5.99, p = 0.019), protein (F (1, training are disparate as one study reported no improve-
44) = 6.48, p = 0.015), and calcium intake (F (1, ment in BMD at the hip after 6 months of training [8] while
44) = 4.72, p = 0.036) that indicated a reduction in intake another reported improvements at the total hip and tro-
for all three measures between baseline and follow-up for chanter but no change at the femoral neck [5]. The con-
both groups. There was a significant group-by-time inter- flicting results of these two studies are likely related to the
action for energy expenditure (Table 4) due to a significant large difference in training volume as participants com-
increase in energy expenditure for PUMP (t (23) = 3.71, pleted just one set of 13 repetitions at 50 % 1RM [8]
p = 0.001) and a non-significant change for CON compared to three sets of 16 repetitions at 40 % 1RM [5].
(t (25) = 0.05, p = 0.64). The lack of improvement or maintenance in BMD ob-
served at the hip and total body in this study are likely due
to a number of factors including high baseline BMD levels,
Discussion the lack of progressive overload in the program, the exer-
cises used in the program, and low calcium intake of par-
We assessed the impact of a widely available pre-chore- ticipants. The hip BMD values in this cohort at baseline
ographed low-load very high-repetition resistance training were generally higher than age-matched reference values
program (PUMP) on BMD and body composition in fe- [22] which suggest that the current training protocol was
males aged over 55 years. There was a significant group- not enough to generate further improvement in those with
by-time interaction for lumbar spine BMD that was due to normal BMD. There was also minimal progression in the
a non-significant increase in the PUMP group and a sig- weight lifted during the PUMP intervention, particularly in
nificant reduction in the CON group. There were neither the second half of the program. The lack of progression
any significant interactions for femoral neck, total hip, or was largely due to reported apprehension and fear of injury
trochanter BMD, nor were there were any changes for fat associated with lifting heavier weights. This overall lack of
mass or fat-free soft tissue mass following training. Total progression would have reduced the potential for os-
body BMD reduced for PUMP following training while teogenic outcomes as strain thresholds would not have
maximal strength as determined by 1RM assessment in- been repeatedly exceeded [23]. A number of other studies
creased significantly following PUMP. that have reported improvements in BMD at the hip have
It was hypothesized that BMD would be maintained in included exercises such as hip abduction and hip flexion
the PUMP group following 6 months of low-load very that load the hip from various angles [5, 7] and produce
high-repetition training. This hypothesis was only partially strong muscular contractions of the gluteus medius and
supported by the significant reduction in lumbar spine psoas major, respectively, which attach directly to the hip.
BMD in the control group and the small non-significant In contrast, this intervention was a generic whole-body
improvement in PUMP. It should be noted that the small program that did not specifically target the hip. The exer-
improvement of 0.01 g/cm2 observed for PUMP is less cises used in this study that load the hip (namely squats and
than the smallest detectable difference for the lumbar spine lunges) would have provided a mostly compressive load
[20]. Other low-load protocols have also typically failed to and produced strong contractions of the gluteus maximus
demonstrate substantial positive changes in lumbar spine [24] which does not attach directly onto the hip. Although
BMD in similar aged cohorts [4, 8, 21]. To our knowledge, compressive loading provided by the squat [25], and for-
only one study has reported positive changes in lumbar ward leaning exercises such as deadlifts and dead rows
spine BMD following a low-load resistance training pro- appeared to be beneficial for lumbar spine BMD in this
gram [5]. In that study, participants aged 55–74 years study, the lack of variation in loading torques may have
trained for 40 weeks at either 40 % (3 9 16 reps) or 80 % limited the osteogenic effect at the hip. The relatively low
(3 9 8 reps) 1RM and all groups achieved similar gains in levels of calcium intake in this cohort may have also in-
lumbar spine BMD although gains tended to be higher for fluenced BMD. Although caution is warranted when de-
men than women. It should be noted that the majority of termining calcium intake from 3-day food records, the

123
V. P. Nicholson et al.: Bone density and low-load very high repetition resistance training 497

mean calcium intake was below 1000 mg for the majority throughout the intervention. As shown in Table 2, par-
of participants and it is recommended that adults over ticipants trained above 20 % of 1RM for chest press during
51 years intake 1000–1200 mg of calcium per day [26] as the majority of the program while squats were performed at
the benefits of calcium supplementation on BMD are well approximately 10 % of predicted 1RM throughout. Recent
established. Furthermore, a significant reduction in calcium research demonstrated that older adults were able to train at
intake occurred between baseline and follow-up which approximately 20 % 1RM for 80–100 repetitions when
could have limited positive changes in BMD. using leg press and leg extension [9]. The results of this
Our hypothesis that fat-free soft tissue mass would in- study suggest that similar repetition ranges at equivalent
crease in the PUMP group was not supported by the results training loads are achievable for chest press but perhaps
herein. It appears that training with loads less than 30 % due to the relative complexity of the task, it is likely that
1RM twice per week does not promote improvements in more modest training loads are required to enable com-
fat-free mass. The overall training volume may have been pletion of very high repetitions with squatting.
too low [27] to promote changes and although hypertrophic The potential benefits of any exercise program need to
changes have been seen following low-load resistance be considered in association with the potential risk of in-
training [9, 28], to our knowledge there have been no re- jury for that program. Three participants (10 %) withdrew
ports of improvements in total body fat-free mass in older from training due to injury or fear of injury associated with
adults following low-load training. Dietary factors may the program. This rate of training-related injury (or fear of
have also limited potential changes in body composition. injury) is higher than other high-repetition resistance
For example, the distribution, quality [29], and individual training programs [9] and high-load training programs [36,
dose [30] of protein consumed throughout the day can in- 37] of similar durations. The comparably higher rate of
fluence muscle protein synthesis in older adults but was not training-related issues may be associated with the more
monitored. The reduction in total energy consumption may complex lifting patterns during BodyPumpTM compared to
have limited the ability to accrue fat-free soft tissue mass other training studies that have utilized machine weights.
[31] during resistance training. Although there were no Another potential issue is the limited direct supervision
evident reductions in fat mass, both groups had significant achieved with one instructor.
reductions in body weight over the course of the program. A number of study limitations require attention. Firstly,
Although caution is warranted when determining total en- the participants in this study were healthy, moderately
ergy intake from food records as under-reporting is com- active women without osteoporosis so the effect of such
mon [32], the weight loss is largely attributed to the training on sedentary adults or those with low bone density
reduction in energy intake in both the groups and the sig- cannot be determined. Calcium intake was monitored by
nificant increase in activity among PUMP participants. food records and although both groups reported similar
One-repetition maximum strength increased in the levels of calcium intake, no supplementation was provided
PUMP group for both leg press and Smith machine bench to ensure the attainment of recommended levels of calci-
press. There was a 12 % improvement for leg press and um. The relatively small study size is likely to have af-
14 % for bench press. These gains are modest, particularly fected the ability to detect small changes between groups
for leg press when compared to previous low-load resis- should they exist. The study duration of 6 months is a
tance training programs of similar durations [9, 33]. The further limitation as a bone remodeling cycle can take up to
modest gains are largely due to the lack of load progression 9 months to complete. The program utilized herein was a
over the course of the program and potentially due to general whole-body low-load resistance training program
participants not training to voluntary muscle failure which and as such did not specifically target the lumbar spine or
is imperative when training with light loads [34]. hip which will have limited the likelihood of providing
In this study the amount of weights lifted for squats and specific loading at these sites. Finally, the loads lifted
chest press was monitored to determine changes in self- during the BodyPumpTM intervention were not progres-
selected loads and to ascertain whether progressive loading sively increased which would have limited the osteogenic
occurred. BodyPumpTM classes utilize a low-load very and hypertrophic effects of such training.
high-repetition principle but exercises are not strictly per- In conclusion, this study provides the first evidence that
formed to fatigue or momentary muscular failure which low-load resistance training performed with a very high
may limit strength gains [28] and relative strain on bones number of repetitions can limit reductions in lumbar spine
[35]. The progression of load was monitored by the class BMD in active post-menopausal women. This form of
instructors but each individual self-selected their weights training does not have evident impacts on hip BMD or fat
during the program. As demonstrated by the limited in- mass and fat-free soft tissue mass. Although other estab-
creases in squat and chest press load in the latter part of the lished forms of resistance training may provide greater
program it is clear that progressive overload did not occur BMD and body composition benefits to untrained

123
498 V. P. Nicholson et al.: Bone density and low-load very high repetition resistance training

middle-aged and older women, these findings suggest that middle-aged and older adults. J Sci Med Sport. doi:10.1016/j.
resistance training of very low loads can limit BMD de- jsams.2014.07.018
11. Cullen DM, Smith RT, Akhter MP (2001) Bone-loading response
clines if the training volume is adequate. Undertaking low- varies with strain magnitude and cycle number. J Appl Physiol
load high-repetition resistance training in this manner may 91(5):1971–1976
be more attractive to older adults who are fearful of 12. McDonald F, Yettram AL, MacLeod K (1994) The response of
training with heavy loads or those that prefer a group bone to external loading regimens. Med Eng Phys 16(5):384–397
13. Karabulut M, Bemben DA, Sherk VD, Anderson MA, Abe T,
training environment. However, due to the relatively Bemben MG (2011) Effects of high-intensity resistance training
complex lifting movements involved with BodyPumpTM and low-intensity resistance training with vascular restriction on
and the reported pain/injuries associated with training it is bone markers in older men. Eur J Appl Physiol 111(8):1659–1667
unlikely to be suitable for a number of individuals. Further 14. LMI (2013) Les Mills International. http://www.lesmills.com/.
2014
work should assess the effectiveness of such training over a 15. IHRSA (2011) IHRSA International Report. International Health,
longer period of time with a greater emphasis on indi- Raquet and Sportsclub Association, Boston
vidualized progressive overload and more specific loading 16. Kolt GS, Driver RP, Giles LC (2004) Why older Australians
of the hip and lumbar spine. participate in exercise and sport. J Aging Phys Act 12(2):185
17. Kraemer WJ, Fry AC (1995) Strength testing: development and
evaluation of methodology. In: Maud P, Nieman C (eds) Fitness
Acknowledgments This research was supported by PhD funding and sports medicine: a health-related approach, 3rd edn. Bull
provided by The Australian Fitness Network. Publishing, Palo Alto
18. Simpson SR, Rozenek R, Garhammer J, Lacourse M, Storer T
Conflict of interest Vaughan Nicholson, Mark McKean, Gary (1997) Comparison of one repetition maximums between free
Slater, Ava Kerr, and Brendan Burkett declare that they have no weight and universal machine exercises. J Strength Cond Res
conflict of interest. 11(2):103–106
19. Ainsworth BE, Haskell WL, Whitt MC, Irwin ML, Swartz AM,
Human and Animal rights and Informed Consent All participants Strath SJ, O’Brien WL, Bassett DR Jr, Schmitz KH, Emplain-
provided informed consent and all procedures performed were in accor- court PO (2000) Compendium of physical activities: an update of
dance with the ethical standards of the the Human Research Ethics com- activity codes and MET intensities. Med Sci Sports Exerc 32(9
mittee of the University and with the Declaration of Helsinki. Suppl):S498
20. El Maghraoui A, Do Santos Zounon AA, Jroundi I, Nouijai A,
Ghazi M, Achemlal L, Bezza A, Tazi MA, Abouqual R (2005)
References Reproducibility of bone mineral density measurements using dual
X-ray absorptiometry in daily clinical practice. Osteoporos Int
1. Melton LJ III, Khosla S, Atkinson EJ, O’Connor MK, O’Fallon WM, 16(12):1742–1748
Riggs BL (2000) Cross-sectional versus longitudinal evaluation of 21. Pruitt LA, Taaffe DR, Marcus R (1995) Effects of a one-year high-
bone loss in men and women. Osteoporos Int 11(7):592–599 intensity versus low-intensity resistance training program on bone
2. Warming L, Hassager C, Christiansen C (2002) Changes in bone mineral density in older women. J Bone Miner Res 10(11):1788–1795
mineral density with age in men and women: a longitudinal study. 22. Looker AC, Wahner HW, Dunn WL, Calvo MS, Harris TB,
Osteoporos Int 13(2):105–112 Heyse SP, Johnston CC Jr, Lindsay R (1998) Updated data on
3. Recker R, Lappe J, Davies K, Heaney R (2000) Characterization proximal femur bone mineral levels of US adults. Osteoporos Int
of perimenopausal bone loss: a prospective study. J Bone Miner 8(5):468–490
Res 15(10):1965–1973 23. Rubin CT, Lanyon LE (1985) Regulation of bone mass by me-
4. Kerr D, Ackland T, Maslen B, Morton A, Prince R (2001) Resis- chanical strain magnitude. Calcif Tissue Int 37(4):411–417
tance training over 2 years increases bone mass in calcium-replete 24. Caterisano A, Moss RE, Pellinger TK, Woodruff K, Lewis VC,
postmenopausal women. J Bone Miner Res 16(1):175–181 Booth W, Khadra T (2002) The effect of back squat depth on the
5. Bemben DA, Bemben MG (2011) Dose–response effect of EMG activity of 4 superficial hip and thigh muscles. J Strength
40 weeks of resistance training on bone mineral density in older Cond Res 16(3):428–432
adults. Osteoporos Int 22(1):179–186 25. Cappozzo A, Felici F, Figura F, Gazzani F (1985) Lumbar spine
6. Bocalini DS, Serra AJ, Dos Santos L, Murad N, Levy RF (2009) loading during half-squat exercises. Med Sci Sports Exerc 17(5):
Strength training preserves the bone mineral density of post- 613–620
menopausal women without hormone replacement therapy. 26. Ross AC, Manson JAE, Abrams SA, Aloia JF, Brannon PM,
J Aging Health 21(3):519–527 Clinton SK, Durazo-Arvizu RA, Gallagher JC, Gallo RL, Jones G
7. Kerr D, Morton A, Dick I, Prince R (1996) Exercise effects on (2011) The 2011 report on dietary reference intakes for calcium
bone mass in postmenopausal women are site-specific and load- and vitamin D from the Institute of Medicine: what clinicians
dependent. J Bone Miner Res 11(2):218–225 need to know. J Clin Endocrinol Metab 96(1):53–58
8. Vincent KR, Braith RW (2002) Resistance exercise and bone 27. Peterson MD, Sen A, Gordon PM (2011) Influence of resistance
turnover in elderly men and women. Med Sci Sports Exerc exercise on lean body mass in aging adults: a meta-analysis. Med
34(1):17–23 Sci Sports Exerc 43(2):249–258
9. Van Roie E, Delecluse C, Coudyzer W, Boonen S, Bautmans I 28. Mitchell CJ, Churchward-Venne TA, West DW, Burd NA, Breen
(2013) Strength training at high versus low external resistance in L, Baker SK, Phillips SM (2012) Resistance exercise load does
older adults: effects on muscle volume, muscle strength, and not determine training-mediated hypertrophic gains in young
force–velocity characteristics. Exp Gerontol 48(11):1351–1361 men. J Appl Physiol 113(1):71–77
10. Nicholson VP, McKean MR, Burkett BJ (2014) Low-load high- 29. Paddon-Jones D, Rasmussen BB (2009) Dietary protein recom-
repetition resistance training improves strength and gait speed in mendations and the prevention of sarcopenia: protein, amino acid

123
V. P. Nicholson et al.: Bone density and low-load very high repetition resistance training 499

metabolism and therapy. Curr Opin Clin Nutr Metab Care training in early postmenopausal women. Med Sci Sports Exerc
12(1):86–90 32(11):1949–1957
30. Pennings B, Groen B, de Lange A, Gijsen AP, Zorenc AH, 34. Carpinelli RN (2008) The size principle and a critical analysis of
Senden JMG, van Loon LJC (2012) Amino acid absorption and the unsubstantiated heavier-is-better recommendation for resis-
subsequent muscle protein accretion following graded intakes of tance training. Journal Exerc Sci Fit 6(2):67–86
whey protein in elderly men. Am J Physiol Endocrinol Metab 35. Yoshikawa T, Mori S, Santiesteban AJ, Sun TC, Hafstad E, Chen
302(8):E992–E999 J, Burr DB (1994) The effects of muscle fatigue on bone strain.
31. Pasiakos SM, Vislocky LM, Carbone JW, Altieri N, Konopelski J Exp Biol 188(1):217–233
K, Freake HC, Anderson JM, Ferrando AA, Wolfe RR, Ro- 36. Henwood TR, Riek S, Taaffe DR (2008) Strength versus muscle
driguez NR (2010) Acute energy deprivation affects skeletal power-specific resistance training in community-dwelling older
muscle protein synthesis and associated intracellular signaling adults. J Gerontol Series A 63(1):83–91
proteins in physically active adults. J Nutr 140(4):745–751 37. Maddalozzo GF, Snow CM (2000) High intensity resistance
32. Hill RJ, Davies PSW (2001) The validity of self-reported energy training: effects on bone in older men and women. Calcif Tissue
intake as determined using the doubly labelled water technique. Int 66(6):399–404
Br J Nutr 85(4):415–430
33. Bemben DA, Fetters NL, Bemben MG, Nabavi N, Koh ET (2000)
Musculoskeletal responses to high- and low-intensity resistance

123