Embed Size (px)
Citation preview
Accepted Manuscript
A Systematic Review of High Intensity Progressive Resistance Strength Training ofthe Lower Limb Compared to Other Intensities of Strength Training in Older People
Melissa J. Raymond, B. Physio Rebecca E. Bramley-Tzerefos, B.P&O, B. Physio,MPH, MPA Kimberley J. Jeffs, MD, MBBS Adele Winter, B. App Sci (Physiotherapy),MPH Anne E. Holland, B. App Sci (Physiotherapy), Doctor of Philosophy
PII: S0003-9993(13)00201-3
DOI: 10.1016/j.apmr.2013.02.022
Reference: YAPMR 55379
To appear in: ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION
Received Date: 18 September 2012
Revised Date: 7 February 2013
Accepted Date: 15 February 2013
Please cite this article as: Raymond MJ, Bramley-Tzerefos RE, Jeffs KJ, Winter A, Holland AE, ASystematic Review of High Intensity Progressive Resistance Strength Training of the Lower LimbCompared to Other Intensities of Strength Training in Older People, ARCHIVES OF PHYSICALMEDICINE AND REHABILITATION (2013), doi: 10.1016/j.apmr.2013.02.022.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
1
Title Page 1
2
1) Running head: Review: HIPRST in older people 3
4
2) Title: A Systematic Review of High Intensity Progressive Resistance Strength Training of the 5
Lower Limb Compared to Other Intensities of Strength Training in Older People 6
7
3) Authors’ Names: 8
Melissa J. Raymond – B. Physio, 9
Rebecca E. Bramley-Tzerefos – B. P&O, B. Physio, MPH, MPA, 10
Kimberley J. Jeffs – MD, MBBS, 11
Adele Winter – B. App Sci (Physiotherapy), MPH, 12
Anne E. Holland – B. App Sci (Physiotherapy), Doctor of Philosophy 13
14
4) Authors’ Institutional Affiliations: From the Department of Physiotherapy Caulfield 15
Hospital, Alfred Health Melbourne, Victoria (Raymond, Winter), The Northern Hospital, 16
Northern Health Melbourne, Victoria (Bramley-Tzerefos, Jeffs), Department of Physiotherapy 17
Alfred Hospital, Alfred Health Melbourne, Victoria (Holland); and the Department of 18
Physiotherapy, La Trobe University Melbourne, Victoria (Raymond, Holland), Australia 19
20
5) Presentations: This material has been presented in abstract form to physiotherapists at the 21
Australian Physiotherapy Association (APA) Conference in Sydney September 2009, and in 22
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
2
further detail at APA Gerontology Physiotherapy Australia lecture evenings in February 1
(Camberwell, Melbourne) and April 2010 (Bendigo, Victoria). 2
3
6) Financial Support: This work has been supported by a Small Projects Grant from Caulfield 4
Research Trust, Alfred Health. 5
6
7) The Authors' Financial Disclosure: We certify that no party having a direct interest in the 7
results of the research supporting this article has or will confer a benefit on us or on any 8
organization with which we are associated AND, if applicable, we certify that all financial and 9
material support for this research (eg, NIH or NHS grants) and work are clearly identified in the 10
title page of the manuscript. 11
There is no conflict of interest. 12
13
8) Acknowledgements: Thankyou to Dr R Cassilhas and Dr Marco Tu’lio de Mello for 14
providing additional raw data for this review. The authors would also like to thank Karen Perkins 15
and the Physiotherapy Department at Caulfield Hospital for their ongoing support. 16
17
18
9) Corresponding Author: Melissa Raymond, c/o Physiotherapy Department, Caulfield 19
Hospital 260 Kooyong Road, Caulfield Victoria Australia 3162. Ph (business and home): +61 20
423 416 254, fax +613 9076 6369, email: [email protected] 21
22
10) Reprints: Reprints can be obtained from Melissa Raymond 23
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
1
A Systematic Review of High Intensity Progressive Resistance Strength Training of the Lower 1
Limb Compared to Other Intensities of Strength Training in Older People 2
3
Objective: To examine the effect of high intensity progressive resistance strength training 4
(PRST) on strength, function, mood, quality of life and adverse events compared to other 5
intensities in older people. 6
7
Data sources: Online databases were searched from their inception to July 2012. 8
9
Study Selection: Randomized controlled trials of high intensity PRST (HIPRST) of the lower 10
limb compared to other intensities of PRST in older people (mean age ≥ 65 years) were 11
identified. 12
13
Data Extraction: Two reviewers independently completed quality assessment using the PEDro 14
Scale and data extraction using a prepared checklist. 15
16
Data Synthesis: Twenty-one trials were included. Study quality was fair to moderate (PEDro 17
Scale range 3 to 7). Studies had small sample sizes (18 to 84) and participants were generally 18
healthy. 19
20
Meta-analyses revealed HIPRST improved lower limb strength greater than moderate and low 21
intensity PRST, SMD 0.79; 95% CI 0.40, 1.17 and 0.83; 95% CI -0.02, 1.68 respectively. 22
Studies where groups performed equivalent training volumes resulted in similar improvements in 23
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
2
leg strength, regardless of training intensity. Similar improvements were found across intensities 24
for functional performance and disability. The effect of intensity of PRST on mood was 25
inconsistent across studies. Adverse events were poorly reported, however, no correlation was 26
found between training intensity and severity of adverse event. 27
28
Conclusion(s): HIPRST improves lower limb strength more than lesser training intensities, 29
although it may not be required to improve functional performance. Training volume is also an 30
important variable. HIPRST appears to be a safe mode of exercise in older people. Further 31
research into its effects in older people with chronic health conditions across the care continuum 32
is required. 33
34
Key words: Resistance Training; Aged; Review [Publication Type] 35
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
3
Abbreviations List: 36
37
HIPRST – high intensity progressive resistance training 38
MIPRST – moderate intensity progressive resistance training 39
LIPRST – low intensity progressive resistance training 40
RM – repetition maximum 41
RCTs – randomized controlled trials 42
ADLs – activities of daily living 43
44
QOL - quality of life 45
SMD – standardized mean difference 46
WMD – weight mean difference 47
SD – standard deviation 48
CI – confidence interval 49
POMS – profile of mood states 50
GDS – geriatric depression scale 51
HRSD – Hamilton rating scale for depression 52
LL – lower limb 53
Ext – extension 54
Flex – flexion 55
56
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
4
Systematic Review of High Intensity Progressive Resistance Strength Training of the 57
Lower Limb Compared to Other Intensities of Strength Training in Older People 58
59
Loss of muscle mass and strength is common among older people, with a strong association 60
between reduced lower limb function and dependence in activities of daily living (ADLs)1-61
3_ENREF_1. Evidence demonstrates that exercise in older adults can minimize the effect of 62
functional decline2, 4-8, with progressive resistance strength training (PRST) a commonly 63
employed method. Progressive resistance strength training, where the resistance is progressed to 64
maintain intensity9, improves strength10 and functional ability11, 12 and can eliminate the need for 65
gait aids in older adults13. Moderate and high intensity PRST have also been shown to have a 66
positive impact on depression and quality of life3, 14. A previous meta-analysis suggests that high 67
intensity PRST (HIPRST) is better than lesser intensities for strength outcomes, however may 68
not be required for functional outcomes15. Training volume may also be an important variable, as 69
some trials have reported similar strength gains with low repetitions at high intensity and high 70
repetitions at low intensity16, 17. The risks associated with HIPRST may be greater than for lower 71
intensities, although the safety of HIPRST in this population is not well understood. At present 72
there is no consensus regarding the optimal training intensity for achieving improvements in 73
functional status, mood and quality of life while maintaining safety in older people. 74
75
Other reviews have compared PRST functional limitation and disability in older people without 76
examining the effect of intensity18 or the effect of intensity on these outcomes15. This review was 77
performed to summarize the evidence comparing high intensity to other intensities of PRST in 78
older people on a broader range of outcomes and to provide recommendations for clinical 79
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
5
practice and future research. The primary aims were to examine the effectiveness of HIPRST in 80
older people in improving strength, endurance and functional performance and assess its safety 81
compared with other intensities of PRST. Secondary aims were to examine the effect of HIPRST 82
on cognition, psychological status, quality of life (QOL), falls rate, power, flexibility, and 83
cardiovascular fitness in those who have undertaken such training. 84
85
METHODS 86
87
88
Selection Criteria 89
90
91
Only published randomized controlled trials comparing HIPRST with other intensities of PRST 92
were considered for inclusion. 93
94
Studies with participants’ mean age of 65 years or older were included, but excluded if any 95
participants were aged less than 60 years old. Participants were untrained in PRST. There were 96
no exclusions on the basis of health, gender, residence or setting of therapy. 97
98
There are no consistent criteria for maximal, high, moderate and low intensity strength training 99
in the literature2, 8, 19-23. Percentage of one repetition maximum (1RM) is often used to specify 100
intensity. This is the maximum weight a person can lift once only “before fatigue using good 101
form and technique,24” “performed primarily by the specified muscle groups without the use of 102
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
6
momentum or any changes in body position, other than those directly resulting from the 103
movement of the weight during the exercise motion25.” Although not specifically defined in a 104
previous systematic review18 many studies with 70-90%1RM were labelled as high intensity26-28. 105
Other papers have described moderate intensity as 50-70% 2, 3, 21 and low intensity as 50% or 106
lower11, 12, 21, 29-32. Therefore for the purposes of this systematic review HIPRST was defined as 107
70-89% of 1RM, maximal intensity was defined as 90%1RM or greater, moderate intensity 108
(MIPRST) as 50-69%1RM and low intensity (LIPRST) less than 50%1RM. 109
110
Studies were considered if the HIPRST program was land based and performed against any type 111
of resistance within the above %1RM ranges. All studies were required to have at least one 112
control group who undertook another intensity of strength training: higher, lower and/or variable. 113
Trials of HIPRST of the lower limb with or without additional upper limb, trunk or abdominal 114
strengthening were considered. Trials were excluded if the high intensity range was outside of 115
the defined %1RM; where the specified intensity was not defined as a percentage of 1RM; where 116
HIPRST was combined with other types of exercise other than a warm up and cool down (e.g. 117
aerobic exercise); where training was performed at different velocities such as power training at 118
high velocities; or where the comparison group did not include a second intensity of training (a 119
no training control group). 120
121
Strength, endurance, gait speed, functional limitation, power, torque, flexibility, VO2 Max, 122
cognition, psychological status and quality of life were analyzed as continuous variables. Falls 123
and adverse events were reported as dichotomous outcomes. Relative training volumes were 124
calculated (total repetitions x number of sets x relative load as %1RM22) where they were not 125
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
7
reported in studies. For each trial, these were reported as a percentage of the training volume 126
undertaken by HIPRST group participants. 127
128
Search Strategy 129
130
131
The following online databases were searched from the earliest date available until July 2012: 132
the Cochrane Central Register of Controlled Trials (CENTRAL), Medline, Embase, CINAHL, 133
AMED, Ageline and the Physiotherapy Evidence Database (PEDro). Reference lists from 134
relevant studies and review studies were hand searched for additional articles. No language 135
restrictions were applied. 136
137
Search terms (Appendix 1) were used in the search strategy in Medline and adapted for use in the 138
other databases. These terms were limited to randomized controlled trials. 139
140
Quality Assessment and Data Extraction 141
142
143
Data was extracted independently by two reviewers using a prepared checklist. Disagreements 144
were resolved by consensus. Extraction from each trial included: (1) characteristics of trial 145
participants (including age, gender and level of normal physical activity), and the trial’s 146
inclusion and exclusion criteria; (2) HIPRST and comparison PRST program details (including 147
intensity, repetitions and sets, duration, frequency, number of lower limb exercises, equipment 148
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
8
used); (3) type of outcome measure(s) and adverse events. Methodological quality assessment 149
was conducted independently by two reviewers (MR and RBT) using the PEDro Scale33. 150
Disagreements were resolved by consensus. Where unclear, authors of included studies were 151
approached via email and asked to provide details of missing data or clarification regarding 152
methods. 153
154
Data Analysis 155
156
157
Where studies were considered clinically homogeneous, data were pooled for meta-analyses 158
using Review Manager 5.1 software and weighted mean differences (WMD) with 95% 159
confidence intervals (95% CI) were calculated. A fixed or random effects model was used 160
depending on assessment of heterogeneity. To enable comparison between continuous variables 161
with different units, standardized mean differences (SMD) with 95% CI were calculated. Based 162
on Cohen’s conventional definition for SMD, values of 0.2 were considered small; 0.5 as 163
medium and 0.8 as large34. To assess the effects of our definitions of intensity on the review 164
outcomes, a sensitivity analysis was performed where only extreme values were included (high 165
versus low intensity). Where there were sufficient data, a second sensitivity analysis examined 166
whether the effects differed in higher quality trials (PEDro 6 or more) compared with lower 167
quality trials (PEDro 5 or less). 168
169
170
RESULTS 171
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
9
172
173
The search strategy returned a total of 912 citations. Details on study selection can be found in 174
Figure 1. Where a single RCT reported resulted in two or more publications, each report was 175
included if they investigated a separate outcome measure of interest. Eight of the included 176
reports fell into this category8, 12, 16, 32, 35-38. Multiple reports are referenced as the first published 177
report only. Twenty-one reports of 17 trials were eligible for inclusion (Figure 1), with a total of 178
830 enrolled participants and 724 participants’ data reported. 179
Table 1 summarizes details from included trials. Eight trials (10 reports) investigated high and 180
moderate intensity training only2, 3, 16, 20, 22, 35, 39, 40, six trials (seven reports) examined high and 181
low intensity only11, 12, 14, 17, 29, 30, one trial examined maximal, high and moderate21, one trial 182
(two reports) examined high, moderate and low intensity8, 32 and one trial reported on high versus 183
a variable intensity program41. Some trials included upper limb and trunk PRST as part of their 184
programs. 185
186
Methodological Quality Assessment 187
188
189
Most studies were of poor to moderate methodological quality and ranged from three to seven 190
out of ten on the PEDro Scale. No study involved blinded participants or therapists due to the 191
nature of the interventions. Two studies used blinded assessors12, 37 and one study employed 192
concealed allocation14. One study reported that groups were not of similar age at baseline38. The 193
majority of studies obtained at least 85% of data for a primary outcome2, 3, 12, 14, 20, 21, 29, 32, 35, 39-41 194
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
10
and six studies used intention to treat analysis3, 12, 20, 21, 35, 40. All studies reported between-group 195
statistical comparisons and point measures and measures of variability for at least one primary 196
outcome. 197
198
Participants 199
200
201
All participants were untrained in PRST. Group characteristics are summarized in Table 1. 202
203
Programs 204
205
206
Where reported, programs were carried out in community based training facilities14, 16 and local 207
fitness centres41. The length of training sessions was reported in only seven trials ranging from 208
approximately 4541 to 90 minutes40. Some programs included a warm-up2, 3, 12, 16, 22, 30, 32, 37, 40, 41 209
prior to PRST and a cool-down2, 14, 16, 30, 35, 40 after training. Where specified, these warm-up and 210
cool-down programs consisted of low intensity repetitions of the intended PRST exercise, low 211
intensity cardiovascular exercise, stretching and/or calisthenics. 212
213
Programs ranged from eight to 52 weeks (mean 20.5, SD 13.3), with participants attending two 214
to three times per week. 215
216
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
11
Outcome Measures 217
218
219
Lower Limb Strength 220
221
222
Table 2 summarizes the effect sizes of HIPRST versus other intensities on maximal lower limb 223
strength. Positive SMD favour HIPRST and negative SMD favour the comparison intensity. In 224
the trials that included leg or knee extension, the movements were not described in adequate 225
detail to determine the difference, if any, between these outcomes. 226
227
High versus low. Seven trials reported on high versus low intensity training11, 12, 14, 17, 29, 30, 32, 228
with SMD in favour of HIPRST ranging from 0.21 to 2.05 (Table 2). One trial revealed large 229
SMD in favour of HIPRST for knee extension and leg press after 24 weeks in sedentary male 230
participants8, 32. During one 12 month program, there was no difference between training 231
intensities for knee extension, knee flexion or leg press strength at 3 monthly time points, except 232
for knee extension strength at 3 months, which was greater in HIPRST participants17. Another 233
study11 reported maximal knee extension strength improved significantly with both intensities, 234
with a moderate but not statistically significant effect in favour of high intensity PRST (SMD 235
0.62; 95% CI -0.48, 1.71). Other trials found no difference between intensities in 1RM hip and 236
leg strength at 15 and 52 weeks30 nor for maximal eccentric or isometric quadriceps strength29. 237
Pooled data from four out of seven trials revealed a large SMD in favour of high intensity 238
training, but this just failed to reach statistical significance (SMD 0.83; 95% CI -0.02, 1.68). 239
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
12
These trials were of moderate to low methodological quality with PEDro scores rating 5 or less. 240
[Fig 2] 241
242
High versus moderate. Ten trials (13 reports) compared high versus moderate intensity PRST2, 3, 243
16, 20-22, 32, 35, 39, 40. The SMD tended to favour high intensity training, however, there was marked 244
variability with SMD ranging from 0 to 1.66 (Table 2) and a pooled SMD 0.79 (95% CI 0.40, 245
1.17, Figure 3). Studies were grouped by methodological quality for further analysis and 246
demonstrated consistency of results: lower quality studies with PEDro scores of 5 or less resulted 247
in SMD of 0.67 (95% CI 0.33, 1.02) and those of higher quality (PEDro 6 or more) resulted in 248
SMD of 0.80 (95% CI 0.39, 1.22). Four trials (5 reports) showed statistically significant 249
differences between intensities for leg extension, leg press, leg curl, overall lower body strength 250
and knee extension2, 3, 32, 35, 37 and three studies showed moderate to large SMD favouring high 251
intensity training for various lower limb strength outcomes without statistically significant 252
differences between groups (knee flexion and leg press)11, 35, 40. Three studies showed small 253
SMD without statistically significant differences between groups16, 20, 22, including no effect for 254
leg press strength16. Two studies reported insufficient data to analyse differences between 255
intensities21, 39. Pooled data from four trials revealed a WMD 14.17kg; 95% CI 9.86, 18.48 (leg 256
extension, leg press and knee extension). 257
258
High versus maximal. One trial reported on high compared to maximal training21 and tested 1RM 259
knee extensor strength at various speeds, from 60 - 180°/s. They reported a 7.3% to 11.2% in 260
strength for males and 2.3% to 15.2% increase for females across all intensities (high, maximal 261
and moderate) with no between-group comparison. 262
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
13
263
High versus variable. One study reported significant strength gains for knee extension with both 264
variable intensity and high intensity training, with no significant difference between intensity 265
types (SMD 0.64; 95% CI -0.14, 1.38)41. 266
267
Table 3 shows effect sizes for other non-strength related outcome measures. 268
269
Power and Torque 270
271
272
High, maximal and moderate intensity PRST improved isokinetic peak torque (p<0.05), with no 273
difference between high and moderate intensity training21. The SMDs favoured maximal 274
intensity over high intensity and ranged from -0.12 to -1.25. Another study reported statistically 275
significant improvements in isokinetic peak torque after high and moderate intensity PRST, with 276
high intensity training significantly better than moderate intensity on only one of the eight 277
outcomes35. A third study reported no statistically significant differences between high and low 278
intensity PRST for explosive strength measures including maximal rate of tension 279
development29_ENREF_53. When investigating peak power, one study reported HIPRST 280
resulted in significantly greater gains than MIPRST (SMD 0.69; 95% CI 0.02, 1.33)2. 281
282
Endurance 283
284
285
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
14
In residents of Aged Care Facilities, both high and moderate intensity training increased knee 286
extension endurance by 284% and 117% respectively, with significantly greater improvements 287
for high over moderate intensity PRST (SMD 1.71; 95% CI 0.42, 3.01)11_ENREF_12. In a 288
healthy older population, leg press endurance increased with both high and moderate intensity 289
training, however, there was no significant difference between intensities (SMD 0.14; 95% CI -290
0.44, 0.72)16. In this same population, treadmill time to exhaustion increased with both high and 291
moderate intensity training, with no significant difference between intensities (SMD 0.25; 95% 292
CI -0.34, 0.82)36. 293
294
Cardiovascular fitness 295
296
297
There was no clear benefit of high intensity training on VO2 Peak16, 32, 39, 41 or Submaximal VO241 298
over other intensities of training. Two studies found improvement with both high and moderate 299
intensity PRST with no significant difference between groups32, 36; a third study reported no 300
effect with either intensity of PRST39. Pooled data from two studies revealed a WMD 301
0.93mL/kg/min (95% CI -0.69, 2.55). When compared to variable intensity training, there was a 302
moderate but non-significant effect on Submaximal VO2 in favour of variable intensity (SMD 303
0.66; 95% CI -0.10, 1.42)41. 304
305
Flexibility 306
307
308
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
15
Two studies examined flexibility. One trial reported both HIPRST and MIPRST significantly 309
improved sit and reach with no difference between intensities. Mean improvements across 310
training groups were approximately 3.5-5cm: a 13% and 15.5% increase from baseline for high 311
and moderate intensity PRST respectively37. A second trial also reported significant 312
improvements with both HIPRST and MIPRST in flexibility. Knee flexion and hip extension 313
range improved with no difference between intensities. There were large and statistically 314
significant SMD with HIPRST over LIPRST for increasing knee flexion and hip extension range. 315
Across all intensities tested, the increase in range was 5º to 14º of knee flexion and 0.3º to 4.7° 316
for hip extension, with the largest gain of 4.7° of hip extension after HIPRST. There was no 317
improvement in hip flexion range for any intensity examined8. 318
319
Functional Performance 320
321
322
Both variable and high intensity PRST reduced perceived exertion while carrying an object with 323
a larger reduction observed in those who received variable resistance (SMD -0.77; 95% CI -1.55, 324
0.00). Relative muscle activation was calculated by dividing the integrated electromyography 325
measured during the weighted walking task by maximal elbow flexion. This was reduced only in 326
the variable intensity group, with a large SMD. Two studies reported reductions in stair climb 327
time with high versus moderate intensity PRST (SMD -0.34; 95% CI -1.16, 0.4937 and SMD -328
0.57; 95% CI -1.16, 0.0216). A small but non-significant improvement in stair climb power was 329
reported for high over low intensity PRST11. 330
331
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
16
One trial (two reports) examined an aggregate physical performance score which included sit to 332
stand time, habitual and maximal safe walking speeds, and stair climb time12, 38 and reported 333
similar improvements in performance with high and low intensity training with no difference 334
between intensities (p>0.05). The authors reported that those with the lowest scores at baseline 335
had the greatest improvements38, independent of training intensity. 336
337
There were significant improvements with both high and low intensity PRST for chair-rise time 338
with no difference between intensities (SMD -0.06; 95% CI -1.12, 1.00 respectively)11. 339
340
HIPRST improved 6 Minute Walk Test distance more than moderate intensity training, with a 341
moderate but not significant effect (SMD 0.65; 95% CI -0.44, 1.75)11. 342
343
Disability 344
345
346
Disability, as measured by the Health Assessment Questionnaire Disability Index, was reduced 347
by both high and low intensities with no difference between intensities11 (SMD 0.09; 95% CI -348
0.97, 1.15). 349
350
Falls 351
352
353
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
17
Falls rate was reported in one study in an older population with depression14. While there were 354
less falls per person in the HIPRST group when compared with LIPRST group (0.15 ± 0.37 and 355
0.28 ± 0.75 respectively), this was not statistically significant. 356
357
Hospitalisation 358
359
360
Number of days in hospital were also reported in an older population with depression over an 8-361
week period. These were 0.5 ± 0.2 days per person in the HIPRST group compared with none in 362
the LIPRST, reported to be not statistically significant14. 363
364
365
Quality of life 366
367
368
Quality of life (QOL) was reported in two studies using the Short Form-36 (SF-36)3, 14. There 369
was no improvement in QOL with moderate or high intensity training in any of the domains in 370
SF-36 reported (p>0.05)3. However, another study showed moderate improvements in the vitality 371
domain for high over low intensity training in an older population with depression (SMD 0.59; 372
95% CI -0.10, 1.25)14. In this study, both high and low intensity PRST were reported to improve 373
QOL across six of the eight health-related QOL domains: Physical Function, Role Physical, 374
Vitality, Social Function, Role Emotional and Mental Health (p values ranged <0.001 to 0.04). 375
376
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
18
Psychological Status 377
378
379
One study reported no improvement in self-rated depression (Geriatric Depression Scale – GDS) 380
with high or moderate intensity PRST in a sedentary, older male population (p>0.05)3. In an 381
older population with a diagnosis of depression, HIPRST participants experienced more than 382
twice the relative response of LIPRST participants in both GDS (58% ± 7% versus 23 ± 7%) and 383
Hamilton rating scale for depression (HRSD) (52% ± 7% versus 25% ± 8%)14. This was 61% of 384
HIPRST participants had a clinically important response (a 50% improvement in therapist-rated 385
scores), compared with 29% of LIPRST participants (p=0.05). The authors described a 386
significant inverse relationship between strength and self-reported depression levels with high 387
intensity exercise only14. 388
389
Mood was reported in two studies investigating moderate and high intensity PRST. One study 390
reported that both intensities improved overall mood scores (Profile of Mood States - POMS), 391
with no statistically significant difference between intensities(p>0.05)3. The second study 392
reported that MIPRST tended to improve the POMS-tension domain more than HIPRST (SMD 393
0.59; 95% CI -0.23, 1.41)20. In the same study, trait anxiety, as measured by the State and Trait 394
Anxiety Inventory, was significantly lessened with MIPRST compared to HIPRST (SMD 0.86; 395
95%CI 0.02, 1.71). 396
397
Cognition 398
399
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
19
400
Two studies investigated the effects of HIPRST and MIPRST on cognition. One reported 401
improvements in cognitive functioning for both high and moderate intensities in some areas of 402
cognitive testing (digit span forward, Corsi’s block-tapping task backwards, similarities, Rey-403
Osterreith complex figure immediate recall) with no difference between intensities (p>0.05)3. 404
This population had at least eight years of schooling and Mini-Mental State Exam (MMSE) score 405
of 24/30 or greater. The other study found no treatment effects for neurocognitive function in 406
healthy but sedentary older adults with at least a high school education20. 407
408
Program Duration 409
410
411
There was no strong link between duration of the HIPRST program and 1RM strength outcomes 412
(Figure 4). Some shorter programs (range 8 to 12 weeks) were sufficient to produce significantly 413
greater strength gains for HIPRST compared to lesser intensities12, 14, 37. 414
415
Training Volumes 416
417
418
Seven out of eight trials where the groups undertook similar training volumes (comparison 419
volume 75 to 125% of HIPRST volume) displayed similar lower limb strength gains, regardless 420
of training intensity2, 16, 17, 20, 22, 29, 30, 41 (Table 2). There was no clear link between the magnitude 421
of strength increase and training volume across studies; larger training volumes did not 422
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
20
necessarily result in larger strength gains. Conversely smaller training volumes did not result in 423
lesser strength gains. 424
425
426
Adherence 427
428
429
Where reported, drop outs ranged from 0%3, 21 to 44%17 of participants and adherence to training 430
ranged from 66%30_ENREF_26 to 100%3, 21 of sessions. Reasons for lack of adherence and drop 431
outs ranged from lack of interest14, 17, 22, personal reasons11, 14, 17, 29, 37, medical reasons unrelated 432
to the study11, 12, 16, 22, conflicts arising from project/job22, 39, commuting issues16, 39, pain or 433
injury12, 14, 16. On analysis of all reported data, adherence and drop outs did not differ according 434
to training intensity. 435
436
Adverse Events 437
438
439
Eight studies did not specifically state the absence or presence of adverse events3, 17, 20, 21, 37, 39-41. 440
Nine studies reported on the presence of adverse events2, 11, 12, 14, 16, 22, 29, 30, 32 however these were 441
generally not reported in detail. Two of these studies reported no training-induced adverse 442
events11, 22 and two reported injuries without detail2, 32. The remainder of studies12, 14, 16, 29, 30 443
reported adverse events that ranged from musculoskeletal discomfort to the exacerbation of 444
underlying medical conditions30 to a more serious event in one trial. This was an exacerbation of 445
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
21
chronic obstructive pulmonary disease (COPD) and myocardial infarction three days post 446
HIPRST12. Although independent monitors suggested that the exercise program could have 447
contributed, the authors stated that HIPRST remains safe for older people to undertake. 448
449
Where injuries were reported, rates of injury appeared comparable among high, moderate and 450
low intensity training. One trial reported no difference in adverse events between those 451
participating in various intensities of PRST and those not14. 452
453
Sensitivity analysis for definitions of intensity 454
455
456
Sensitivity analyses were limited by the small number of trials comparing extreme values of 457
intensity (i.e. high versus low). However, where data were available, the pattern of findings was 458
similar to when all three intensities were considered. Sensitivity analysis for the outcomes of 459
maximal strength showed moderate to large effects for HIPRST over LIPRST in 3 out of 7 trials, 460
with statistically significant differences in 3 other trials where raw data were unable to be 461
obtained. There were only two trials which had similar training volumes across high and low 462
intensities. These trials showed disparate effects, with one trial showing no significant benefit of 463
HIPRST over LIPRST (SMD of 0.4)42 and the other favouring HIPRST with a large effect (SMD 464
of 2.05)32. While only examined in single trials, depression14, QOL14 and flexibility8 465
demonstrated significantly larger effects when HIPRST was compared to LIPRST than when 466
HIPRST was compared to MIPRST, although raw data were not available for analyses on 467
depression and QOL. Adverse events and drop outs were similar across high and low intensity 468
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
22
training. Although not able to be performed on all outcomes in this meta-analysis, sensitivity 469
analyses support the between-group differences favouring high intensity training that were found 470
in the primary analyses. 471
472
473
474
DISCUSSION 475
476
477
This meta-analysis has shown that HIPRST demonstrates advantages over lower intensities of 478
PRST. However, trials with similar training volumes had similar gains in lower limb strength, 479
regardless of training intensity. Other outcomes such as functional performance improved 480
similarly across all intensities of PRST. Flexibility increased more with HIPRST and MIPRST 481
than LIPRST. The effects of HIPRST compared to other intensities of PRST on psychological 482
status remains unclear. 483
484
Although there were greater improvements in strength with high intensity training compared to 485
moderate and low intensity training (Figures 2 and 3), this was not consistent across all trials. 486
Training volume also appears to have an important effect on strength, with similar gains seen in 487
training groups who performed a similar volume of PRST, regardless of absolute intensity. These 488
results suggest that clinicians should consider whether there is a need for older participants to 489
undertake HIPRST if the primary goal is to improve strength. 490
491
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
23
Reduced flexibility may contribute to limited function43, 44 and falls45 in older people. Studies in 492
younger people have shown strength training does not impair flexibility; in some cases it may 493
improve it46 to levels comparable to static stretching47. In this review, both HIPRST and 494
MIPRST have been shown to increase flexibility to similar extents and were more beneficial than 495
LIPRST. It has been suggested that strengthening muscle groups may also improve range of 496
movement across the joints that these muscles span48. 497
498
Functional improvements have been demonstrated with HIPRST in older people in other studies, 499
with improvements in ability to stand from a chair6, 10, reduced need for walking aids10 500
_ENREF_65and self paced gait speed6_ENREF_66. Results from this review are consistent with 501
this literature, however, walking speed, functional performance and disability were found to 502
improve similarly with moderate and low intensity PRST, also supported by other meta-503
analyses15. Therefore high intensity may not be required to improve functional performance and 504
lesser intensities may suffice. 505
506
While there was no significant benefit in overall QOL from HIPRST compared to other 507
intensities, a trend for improvement was shown for components of QOL measures14. Mixed 508
results for other psychological outcomes were reported. Other research has reported HIPRST 509
programs having reduced depressive symptoms in those with both minor and major depression, 510
with the majority of exercisers no longer meeting the diagnostic criteria for depression after 10 511
weeks of exercise. Increased training intensity was also reported to predict the reduction in 512
depressive symptoms31. Previous systematic reviews on exercise for depression suggest exercise 513
may reduce depression in this population in the short term; however, more robust, larger, and 514
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
24
longer term randomized clinical trials are required to define the parameters around type of 515
exercise, intensity, duration of programs and target populations for optimal outcomes49, 50. 516
517
There was no obvious relationship between program duration and outcome, suggesting longer 518
programs are not required to achieve additional benefits. Shorter HIPRST programs of 12 weeks 519
demonstrated significantly greater gains in outcomes such as strength and endurance than lower 520
intensities of training. As positive results may be seen within two to three months of HIPRST, 521
this finding may encourage potential participants to undertake such a program, particularly 522
where time commitments may be a barrier. This is also positive for clinicians in community 523
based settings, where longer programs may be cost-prohibitive and unfeasible due to long 524
waiting lists. 525
526
Less than a third of the included studies reported on adverse events related to training; however 527
events were not reported in adequate detail for further analysis. Nearly half of the trials did not 528
report the presence or absence of adverse events and therefore it is plausible that under-reporting 529
occurred. The majority of events reported were minor musculoskeletal injuries and there were no 530
deaths due to high intensity training. There was no difference in rates of adverse events between 531
intensities of training. The single serious adverse event occurred two to three days after training 532
and it was unclear if this event occurred in the placebo or drug group12. Due to this period of 533
delay it would seem unlikely that strength training caused these events. HIPRST has been 534
reported to be safe in older people to the age of 96 years old10, 13. Prior to implementation, 535
medical screening of potential participants and supervised programs should be considered. 536
537
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
25
Study Limitations 538
539
540
This review had some limitations. While some of the effect sizes calculated were moderate to 541
large, many were not statistically significant. This may have been due to the lack of power in 542
studies due to small sample sizes. Caution must also be used when interpreting studies with low 543
methodological quality, as many studies did not use intention-to-treat analysis, concealed 544
allocation or blinded assessors, which reduced internal validity and may have overestimated the 545
effect of the interventions. Prior training status has been reported to be an important factor in 546
determining the most effective training intensity for optimal strength outcomes across the age 547
spectrum, where moderate intensity may result in better outcomes in the untrained and high 548
intensity may be more effective in those who are trained51. This effect was not considered in the 549
review. Meta-analyses were limited for outcomes other than strength and raw data were unable 550
to be obtained in many instances. As most participants were sedentary and previously untrained 551
in PRST, the findings from this review may have limited application to physically active or 552
trained older adults. 553
554
The literature in this area also had some limitations. The trials included in this systematic review 555
contained programs of PRST primarily run in the community with resistance machines. Only one 556
trial used elastic bands, which may be a low-cost equipment option in hospitals and community 557
settings. Although resistance bands have been shown to achieve comparable strength gains 558
compared to weight machines in a few studies52-54, to our knowledge, there are no similar 559
comparison studies in older people undertaking HIPRST. Outcomes other than strength were 560
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
26
reported in few studies. There were no trials on the effect of HIPRST in hospitalized older 561
people, where HIPRST may play an important part on falls rate, length of stay and health service 562
utilization. There is a need for further research into outcomes other than strength, in particular 563
with a focus on older people with various health conditions in different settings through the care 564
continuum. 565
566
CONCLUSIONS 567
568
569
High intensity PRST may improve strength more than lower intensities of strength training, 570
although training volume also has an important effect on the strength gains achieved. Training 571
intensity did not appear to impact greatly on outcomes other than strength. Long programs of 572
HIPRST are not required to see initial benefits nor to achieve additional improvements. Adverse 573
events related to strength training in older people are minor; however, screening prior to 574
implementing new programs is recommended to minimise participant discomfort. 575
576
Future research needs to establish whether HIPRST results in clinically meaningful 577
improvements in important outcomes such as falls, hospitalisation and use of health care 578
services. Larger trials are required, with particular emphasis on sub-groups of older people with 579
chronic diseases, those who are hospitalized and in residential aged care. 580
581
582
Acknowledgements 583
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
27
584
This work was supported by a Small Projects Research Grant from the Caulfield Research Trust, 585
Alfred Health. 586
Thank you to Dr Ricardo Cassilhas and Dr Marco Tulio de Mello who provided additional raw 587
data for this review. The authors would also like to thank Karen Perkins and the staff of the 588
Caulfield Hospital Physiotherapy Department for their ongoing support. 589
590
REFERENCES: 591
592
593
1. Guralnik J, Ferrucci L, Simonsick E, Salive M, Wallace R. Lower extremity function in 594
persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med. 595
1995;332:556-561. 596
2. Fatouros I, Kambas A, Katrabasas I, et al. Strength training and detraining effects on 597
muscular strength, anaerobic power, and mobility of inactive older men are intensity 598
dependent. BJSM online. Oct 2005;39(10):776-780. 599
3. Cassilhas RC, Viana VA, Grassmann V, et al. The impact of resistance exercise on the 600
cognitive function of the elderly. Med Sci Sports Exerc. Aug 2007;39(8):1401-1407. 601
4. Jones CT, Lowe AJ, MacGregor L, Brand CA. Randomised controlled trial of an exercise 602
intervention to reduce functional decline and health service utilisation in the hospitalised 603
elderly. Australasian Journal on Ageing. 2006;25(3):126-133. 604
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
28
5. Rosendahl E, Gustafson Y, Nordin E, et al. A randomized controlled trial of fall 605
prevention by a high-intensity functional exercise program for older people living in 606
residential care facilities. Aging Clin Exp Res. Feb 2008;20(1):67-75. 607
6. Rosendahl E, Lindelof N, Littbrand H, et al. High-intensity functional exercise program 608
and protein-enriched energy supplement for older persons dependent in activities of daily 609
living: a randomised controlled trial. Aust J Physiother. 2006;52(2):105-113. 610
7. Siebens H, Aronow H, Edwards D, Ghasemi Z. Randomized controlled trial of exercise 611
to improve outcomes of acute hospitalization in older adults. J Am Geriatr Soc. 612
2000;48(12):1545-1552. 613
8. Fatouros IG, Kambas A, Katrabasas I, et al. Resistance training and detraining effects on 614
flexibility performance in the elderly are intensity-dependent. J Strength Cond Res. Aug 615
2006;20(3):634-642. 616
9. Kraemer WJ, Adams K, Cafarelli E, et al. The American College of Sports Position 617
Stand: Progression models in resistance training for healthy adults. Med Sci Sports Exerc. 618
2002:364-380. 619
10. Fiatarone M, Marks E, Ryan N, Meredith C, Lipsitz L, Evans W. High-intensity strength 620
training in nonagenarians: effects on skeletal muscle. Journal - American Medical 621
Association. Jun 13 1990;263(22):3029-3034. 622
11. Seynnes O, Fiatarone Singh MA, Hue O, et al. Physiological and functional responses to 623
low-moderate versus high-intensity progressive resistance training in frail elders. J 624
Gerontol A Biol Sci Med Sci. May 2004;59(5):503-509. 625
12. Sullivan DH, Roberson PK, Johnson LE, et al. Effects of muscle strength training and 626
testosterone in frail elderly males. Med Sci Sports Exerc. Oct 2005;37(10):1664-1672. 627
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
29
13. Fiatarone MA, O'Neill EF, Ryan ND, et al. Exercise training and nutritional 628
supplementation for physical frailty in very elderly people. N Engl J Med. Jun 23 629
1994;330(25):1769-1775. 630
14. Singh NA, Stavrinos TM, Scarbek Y, et al. A randomized controlled trial of high versus 631
low intensity weight training versus general practitioner care for clinical depression in 632
older adults. J Gerontol A Biol Sci Med Sci. Jun 2005;60(6):768-776. 633
15. Steib S, Schoene D, Pfeifer K. Dose-response relationship of resistance training in older 634
adults: A meta-analysis. Medicine and Science in Sports and Exercise. May;42 (5):902-635
914. 636
16. Vincent KR, Braith RW, Feldman RA, et al. Resistance exercise and physical 637
performance in adults aged 60 to 83. J Am Geriatr Soc. Jun 2002;50(6):1100-1107. 638
17. Taaffe DR, Pruitt L, Pyka G, Guido D, Marcus R. Comparative effects of high- and low-639
intensity resistance training on thigh muscle strength, fiber area, and tissue composition 640
in elderly women. Clin Physiol. Jul 1996;16(4):381-392. 641
18. Latham NK, Bennett DA, Stretton CM, Anderson CS. Systematic review of progressive 642
resistance strength training in older adults. Journals of Gerontology: Series A: Biological 643
Sciences and Medical Sciences. 2004;59A(1):48-61. 644
19. Spruit M, Eterman R, Hellwig V, Janssen P, Wouters E, Uszko-Lencer N. Effects of 645
moderate-to-high intensity resistance training in patients with chronic heart failure. Heart 646
Lung and Circulation. 2009;95:1399-1408. 647
20. Tsutsumi T, Don BM, Zaichkowsky LD, Takenaka K, Oka K, Ohno T. Comparison of 648
high and moderate intensity of strength training on mood and anxiety in older adults. 649
Percept Mot Skills. Dec 1998;87(3 Pt 1):1003-1011. 650
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
30
21. Beneka A, Malliou P, Fatouros I, et al. Resistance training effects on muscular strength of 651
elderly are related to intensity and gender. J Sci Med Sport. Sep 2005;8(3):274-283. 652
22. Harris C, DeBeliso MA, Spitzer-Gibson TA, Adams KJ. Effect of resistance-training 653
intensity on strength-gain response in the older adult. Journal of Strength and 654
Conditioning Research. 2004;18(4):833-838. 655
23. Jan MH, Lin JJ, Liau JJ, Lin YF, Lin DH. Investigation of clinical effects of high- and 656
low-resistance training for patients with knee osteoarthritis: a randomized controlled trial. 657
Physical Therapy 2008 Apr;88(4):427-436. 2008. 658
24. ACSM Position Stand on The Recommended Quantity and Quality of Exercise for 659
Developing and Maintaining Cardiorespiratory and Muscular Fitness, and Flexibility in 660
Adults. . Med Sci Sports Exerc. 1998;30(6):975-991. 661
25. Charette SL, McEvoy L, Pyka G, et al. Muscle hypertrophy response to resistance 662
training in older women. J Appl Physiol. 1991;70:1912-1916. 663
26. Rall LC, Roubenoff R, Cannon JG, Abad LW, Dinarello CA, Meydani SN. Effects of 664
progressive resistance training on immune response in aging and chronic inflammation. 665
Med Sci Sports Exerc. Nov 1996;28(11):1356-1365. 666
27. Flynn MG, Fahlman M, Braun WA, et al. Effects of resistance training on selected 667
indexes of immune function in elderly women. J Appl Physiol. Jun 1999;86(6):1905-668
1913. 669
28. Hagerman FC, Walsh SJ, Staron RS, et al. Effects of high-intensity resistance training on 670
untrained older men. I. Strength, cardiovascular, and metabolic responses. J Gerontol A 671
Biol Sci Med Sci. Jul 2000;55(7):B336-346. 672
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
31
29. Hortobagyi T, Tunnel D, Moody J, Beam S, DeVita P. Low- or high-intensity strength 673
training partially restores impaired quadriceps force accuracy and steadiness in aged 674
adults. J Gerontol A Biol Sci Med Sci. 2001;56A(1):B38-B47. 675
30. Pruitt L, Taaffe DR, Marcus R. Effects of a one- year high-intensity versus low-intensity 676
resistance training program on bone mineral density in older women. J Bone Miner Res. 677
1995;10(11):1788-1795. 678
31. Singh NA, Clements KM, Fiatarone MA. Randomized controlled trial of progressive 679
resistance training in depressed elders. Journals of Gerontology: Series A: Biological 680
Sciences and Medical Sciences. 1997;52A(1):M27-M35. 681
32. Fatouros IG, Tournis S, Leontsini D, et al. Leptin and adiponectin responses in 682
overweight inactive elderly following resistance training and detraining are intensity 683
related. J Clin Endocrinol Metab. Nov 2005;90(11):5970-5977. 684
33. Maher C, Sherrington C, Herbert RD, Moseley AM, M. E. Reliability of the PEDro Scale 685
for rating quality of randomized controlled trials. Phys Ther. 2003;83 (8):713-721. 686
34. Cohen J. Statistical Power Analysis for the Behavioral Sciences. NY: Academic Press. 687
1969. 688
35. Kalapotharakos V, Michalopoulou M, Godolias G, Tokmakidis SP, Malliou P, 689
Gourgoulis V. The effects of high- and moderate- resistance training on muscle function 690
in the elderly. J Aging Phys Act. 2004;11:131-143. 691
36. Vincent KR, Braith RW, Feldman RA, et al. Improved cardiorespiratory endurance 692
following 6 months of resistance exercise in elderly men and women. Arch Intern Med. 693
Mar 25 2002;162(6):673-678. 694
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
32
37. Kalapotharakos VI, Michalopoulos M, Tokmakidis SP, et al. Effects of a heavy and a 695
moderate resistance training on functional performance in older adults. J Strength Cond 696
Res. Aug 2005;19(3):652-657. 697
38. Sullivan DH, Roberson PK, Smith ES, et al. Effects of muscle strength training and 698
megestrol acetate on strength, muscle mass, and function in frail older people. J Am 699
Geriatr Soc. Jan 2007;55(1):20-28. 700
39. Tsutsumi T, Don BM, Zaichkowsky LD, Delizonna LL. Physical fitness and 701
psychological benefits of strength training in community dwelling older adults. Appl 702
Human Sci. Nov 1997;16(6):257-266. 703
40. Willoughby DS, Pelsue SC. Muscle strength and qualitative myosin heavy chain isoform 704
mRNA expression in the elderly after moderate- and high-intensity weight training. J 705
Aging Phys Act. 1998;6:327-339. 706
41. Hunter GR, Wetzstein CJ, McLafferty CL, Jr., Zuckerman PA, Landers KA, Bamman 707
MM. High-resistance versus variable-resistance training in older adults. Med Sci Sports 708
Exerc. Oct 2001;33(10):1759-1764. 709
42. Hortobagyi T, Tunnel D, Moody J, Beam S, de Vita P. Low- or high-intensity strength 710
training partially restores impaired quadriceps force accuracy and steadiness in aged 711
adults. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences 712
2001 Jan;56(1):B38-B47. 2001. 713
43. Jette AM, Branch LG, Berlin J. Musculoskeletal impairments and physical disablement 714
among the aged. J Gerontol. 1990;45:M203-208. 715
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
33
44. Bergstrom G, Aniansson A, Bjelle A, Grimby G, Lundgren-Lindquist B, Svanborg A. 716
Functional consequences of joint impairment at age 79. Scand J Rehab Med. 717
1985;17(183-190). 718
45. Kerrigan DC, Lee LW, Collins JJ, Riley PO, Lipsitz LA. Reduced hip extension during 719
walking: healthy elderly and fallers versus young adults. Arch Phys Med Rehab. 720
2001;82(11):26-30. 721
46. Thrash K, Kelly B. Flexibility and Strength Training. The Journal of Strength & 722
Conditioning Research. 1987;1(4):74-75. 723
47. Morton SK, Whitehead JR, Brinkert RH, Caine DJ. Resistance Training vs. Static 724
Stretching: Effects on Flexibility and Strength. The Journal of Strength & Conditioning 725
Research. 2011;25(12):3391-3398. 726
48. Fatouros IG, Taxildaris K, Tokmakidis SP, et al. The effects of strength training, 727
cardiovascular training and their combination on flexibility of inactive older adults. Int J 728
Sports Med.23(5):112-119. 729
49. Mead G, Morley W, Campbell P, Greig C, McMurdo M, Lawlor D. Exercise for 730
depression. Cochrane Database of Systematic Reviews 2009(3):Art. No.: CD004366. . 731
50. Lawlor D, Hopker S. The effectiveness of exercise as an intervention in the management 732
of depression: systematic review and meta-regression analysis of randomised controlled 733
trials. Bmj. 2001(322):1-8. 734
51. Rhea MR, Alvar BA, Burkett LN, SD B. A meta-analysis to determine the dose response 735
for strength development. Med Sci Sports Exerc. 2003;35(3):456-464. 736
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
34
52. Vanbiervliet W, Pelissier J, Ledermann B, Kotzki N, Benaim C, Herisson C. Strength 737
training with elastic bands: Measure of its effects in cardiac rehabilitation after coronary 738
diseases. [French]. Ann Readapt Med Phys. 2003;46 (8):545-552. 739
53. Colado J, Triplettt T. Effects of a short-term resistance program using elastic bands 740
versus weight machines for sedentary middle aged women. J Strength Cond Res. 741
2008;22(5):1441-1448. 742
54. Andersen L, Andersen C, Mortensen O, Poulsen O, Bjornlund I, Zebis M. Muscle 743
activation and perceived loading during rehabilitation exercises: comparison of 744
dumbbells and elastic resistance. Phys Ther. 2010;90:538-549. 745
746
747
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
35
Figure 1. Flowchart showing the selection of studies for this systematic review. 748
749
Figure 2. High versus low intensity PRST for maximal lower limb strength. 750
751
Figure 3. High versus moderate intensity progressive resistance strength training for lower limb 752
strength. 753
754
Figure 4. The effect of program duration on lower limb 1RM strength outcomes. A SMD greater 755
than zero favours HIPRST. 756
757
758
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
APPENDICES Appendix 1. Terms Used in the Search Strategy Population Intervention Outcome exp Aged/
• Resistance Training/ • strength training.ti.ab. • progressive resistance.ti.ab. • high intensity • resistance exercise.ti.ab. • Exercise Therapy/ • Exercise • Weight Lifting
• repetition maximum.mp • 1RM.mp
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
25
Table 1: Summary of 21 Included Studies Comparison intensity:
(%1RM) Sets x reps
Study PEDro Score (/10)
Age Range (mean)
Participants Sample Size*
(final no.)
No. of LL
Ex
Frequency and
duration†
High intensity resistance (%1RM), Sets x reps
Maximal Moderate Low/Other
Adverse events
Outcomes‡
Beneka 2005
6 60 + (65)
Healthy sedentary
64 (64) 3 3 x week 16 weeks
70% 3 x 8 – 10
90% 3 x 4 – 6
50% 3 x 12 - 14
NA Not Reported
Strength
Cassilhas 2007 6 65 – 75
Male sedentary 62 (62) 2 3 x week 24 weeks
80% 2 x 8
NA 50% 2 x 8
NA Not Reported
Strength, cognition, QOL, depression
Fatouros, Kambas et al 2005
5 (71.2) Male, sedentary 59 (52) 3 3 x week 24 weeks
80 – 85% 2 – 3 x 6 – 8
50 – 55% 2 – 3 x 14 - 16
Reported, no details
Strength, power, functional performance, walking speed
Fatouros, Tournis 2005
5 65 – 78
Male, sedentary, overweight
57 (50) 3 3 x week 24 weeks
80 – 85% 2 - 3 x 8
NA
60-65% 2 – 3 x 10
45 – 50% 2 – 3 x 14
Reported, no details
Strength, cardiovascular fitness
Fatouros 2006 5 As above Strength, flexibility Harris 2004 4 61 – 85
(71.2) Healthy 76 (62) 3 2 x week
18 weeks 75% 3 x 9 84% 4 x 6
NA 67% 2 x 15
NA Nil Strength
Hortobagyi 2001 5 66 – 83 (72)
Healthy 30║ (27) 1 3 x week 10 weeks
80 % 5 x 4 – 6
40% 5 x 8 – 12
Yes Strength
Hunter 2001 4 61 – 77 (>65)
Healthy 30 (28) 2 3 x week 25 weeks
80% 2 x 10
NA NA Variable: 50%, 65%,
80%
Not Reported
Strength, perceived exertion, functional performance, cardiovascular fitness
Kalapotharakos 2004
6 60 - 74 Healthy sedentary
33(33) 2 3 x week 12 weeks
80% 3 x 8
60% 3 x 15
Not reported Strength
Kalapotharakos 2005
6 As above 35 (33) Not reported Strength, functional performance, flexibility walking speed
Pruitt 1995 4 65 – 79 Female, healthy 40 (26) 5 3 x week 52 weeks
80% 2 x 7
(warm up 40% 1 x 14)
NA 40% 3 x 14
Yes Strength
Seynnes 2004 4 73 – 95 (81.5)
Institutionalized 27 (22) 1 3 x week 10 weeks
80% 3 x 8
NA NA 40% 3 x 8
Nil Strength, endurance, 6MWT, disability
Singh 2005 6 60 – 85
Depression 60 (54) 3 3 x week 8 weeks
80% 3 x 8
NA NA 20% 3 x 8
Yes Strength, QOL, depression
Sullivan 2007 5 65 – 93 (79.4)
Recent functional decline
29 (24) 1 3 x week 12 weeks
80% 3 x 8
NA NA 10-20% 3 x 8
Yes Strength, functional performance
Sullivan 2005
7 65 – 93 (78.2)
Recent functional decline
71 (61) 3 3 x week 12 weeks
80% 3 x 8
NA NA 10-20% 3 x 8
Yes Strength, functional performance
Taaffe 1996 3 65 – 79 Healthy women 36 (25) 3 3 x week 52 weeks
80% 2 x 7
(warm up 40% 1 x 14)
NA NA 40% 3 x 14
Not reported Strength
Tsutsumi 1997 5 61-86 (68.6)
Healthy, sedentary
45 (42) 2 3 x week 12 weeks
75 – 85% 2 x 8 – 10
NA 55-65% 2 x 14-16
NA Not reported Strength, mood and anxiety, cognition, functional
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
26
performance Tsutsumi 1998 6
60-86 (68.5)
Women, healthy, sedentary
36 (36) 2 3 x week 12 weeks
75 – 85% 2 x 8 – 10
NA 55-65% 2 x 14-16
NA Not reported Strength, psychological measures
Vincent…Magyari et al 2002
4 60 – 83 (68.4)
Healthy 84 (62) 6 3 x week 26 weeks
80% 1 x 8
NA 50% 1 x 13
NA Yes Strength, endurance
Vincent,… Kalls et al 2002
As above
Cardiovascular fitness
Willoughby 1998 6 (69) Male, sedentary 18 (18) 3 3 x week 12 weeks
75 – 80% 3 x 8 – 10
NA 60 – 65% 3 x 15 – 20
NA Not reported Strength
No. = number; NA – not applicable; QOL = quality of life; LL = lower limb; Ex = exercises; 6MWT – 6 minute walk test. *Number of enrolled participants. †Total duration - some studies included a 1-4 week conditioning period which gradually increased %RM to the target level. ‡Outcomes of interest to this review. §Includes 1 week of pretraining/orientation – 1 set only ║Participants 60 years or older
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Table 2: Effects of HIPRST versus other intensities of PRST on Strength Outcomes
Study High Intensity Resistance % 1RM
Comparison Intensity 1 % 1RM
Comparison Training volume as a % of High Intensity group (%)*
Outcome†
Effect Size 95% Confidence Interval or p value
Comparison Intensity 2 % 1RM
Comparison Training volume as a % of High Intensity group (%)*
Effect Size
95% Confidence Interval
Beneka 2005 70 50 140 Knee extension ND 90 144 ND Cassilhas 2007‡ 80 50
63 Leg press
Leg curl 0.87 1.15
[0.22, 1.53] [0.47, 1.82]
NA
Fatouros, Tournis 2005 80 – 85 60 – 65 95 Leg extension 1.14 [0.30, 1.98] 45 – 50 101 2.05 [1.11, 2.98]
Fatouros 2006 As above As above As above Leg press 0.65 [-0.16, 1.42] 45 – 50 As above 1.90 [0.96, 2.73]
Fatouros, Kambas 2005 80 – 85 50 – 55 137 Leg press 1.66 [0.91, 2.41] NA
Harris 2004 75 67 100 Lower body strength§ 0.26 [-0.46, 0.99] 84 vs 67 100 0.10 [-0.58, 0.77] Hortobagyi 2001 80 40 100 Leg press
Max eccentric strength Isometric Concentric
0.40 0.11 0.03 0.49
[-0.54, 1.34] [-0.11, 0.33] [-0.89, 0.96] [-0.49, 1.39]
NA
Hunter 2001 80 50, 65, 80 82 Knee extension 0.62 [-0.14, 1.38] NA Kalapotharakos 2005 80 60 141 Lower body strength║ 1.00 [0.12, 1.88]
Kalapotharakos 2004 As above As above Knee extension Knee flexion
1.04 0.80
[0.16, 1.92] [-0.06, 1.65]
NA
Pruitt 1995 80 40 75 Hips¶ Legs#
0.21 0.21
[-0.81, 1.22] [-0.94, 1.36]
NA
Seynnes 2004 80 40 50 Knee extension 0.62 [-0.48, 1.71] NA
Singh 2005 80 20 25 Average strength** ND, RS p<0.0001 NA Sullivan 2007 80 10-20 25 Leg press ND, RNS p>0.05 NA Sullivan 2005 As above As above Leg press ND, RS p<0.05 NA Taaffe 1996 80 40 75 Leg press††
Knee extension††
Knee flexion††
ND, RNS ND, RS‡‡
ND, RNS
p>0.05 p<0.05 p>0.05
NA
Tsutsumi 1997 75 – 85 55 – 65 105 Leg strength (10RM) RNS RNS NA Tsutsumi 1998 75 – 85 55 – 65 125 Leg extension 0.23 [-0.57, 1.03] NA Vincent, Braith, Feldman, Magyari et al 2002
80 50 102 Leg press Leg curl Leg ext
0.00 0.32 0.29
[-0.58, 0.58] [-0.27, 0.90] [-0.29, 0.87]
NA
Willoughby 1998 75 – 80 60 – 65 162 Leg press 0.62 [-0.46,1.70] NA Note: A positive ES indicates HIPRST having larger strength gains than the comparison group. RM = Repetition Maximum; ND = no data available; NA = not applicable; RS = reported significant; RNS = reported
not significant.*Training volume each session as a percentage of high intensity training volume. See Methods for further details. †1RM where not specified. ‡Raw data (unpublished) obtained directly from authors. § Combined knee ext, leg press, leg curl. ║Combined knee extension and flexion. ¶Hip abduction plus hip adduction. #Leg press plus knee extension plus knee flexion. **Mean of percent change across 3 upper limb and 3 lower limb resistance exercises ††Outcome measure assessed at 3, 6, 9 and 12 months. ‡‡Significant at 3 months only.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPTTable 3: Effect Size Table for High Intensity Versus Other Intensity PRST – Various Outcomes
Study High Intensity Resistance
%1RM
Comparison Intensity 1
%1RM
Comparison Training
volume as a % of High Intensity
group (%)*
Outcome Effect Size [95% Confidence
Interval] or p value
Comparison Intensity 2
%1RM
Comparison Training
volume as a % of High Intensity
group (%)*
Effect Size [95% Confidence Interval]
Beneka 2005 70 50 140 Isokinetic Peak Torque Males: Knee ext 90° deg/sec
0.51 [-0.49, 1.51]
90% 144 -0.30 [-1.29, 0.69]
Females: Knee ext 90° deg/sec 0.71 [-0.31, 1.73] -0.71 [-1.71, 0.31] Males: Knee ext 60° deg/sec 0.63 [-0.38, 1.64] -0.34 [-1.33, 0.64] Females: Knee ext 60° deg/sec -0.10 [-1.08, 0.88] -1.18 [-2.27, -0.09]
Cassilhas 2007 80 50 63 Quality of life ND, RNS NA Depression – self rated (GDS) ND, RNS Mood ND, RNS Cognition ND, RNS
Fatouros, Tournis 2005
80 – 85 60 – 65 95 VO2 Max 0.56 [-0.23, 1.34] 45 – 50 101 0.50 [-0.28, 1.29]
Fatouros 2006 80 – 85 60 – 65 As above Flexibility – knee flexion Flexibility – hip flexion Flexibility – hip extension
0.10 [-0.67, 0.87] 0.04 [-0.73, 0.82] 0.29 [-0.48, 1.07]
45 – 50 As above 0.88 [0.10, 1.66] 0.06 [-0.68, 0.80] 2.93 [1.82, 4.04]
Fatouros, Kambas 2005
80 – 85 50 – 55 137 Peak power Mean power
0.69 [0.02, 1.33] 0.38 [-0.27, 1.01]
NA
Timed Up and Go Test -0.07 [-0.70, 0.57] Walking speed -0.05 [-0.68, 0.59] Stepping up -0.18 [-0.82, 0.46] Stepping down -0.21 [-0.85, 0.42]
Hortobagyi 2001 80 40 100 Explosive Strength: Max rate of tension development
0.27 [-0.66, 1.20]
Hunter 2001 80 50, 65, 80 82 Performing functional tasks: Perceived exertion
0.77 [0.00, 1.55]
NA
Submaximal VO2 0.66 [-0.10, 1.42] Relative muscle activation* 1.23 [0.41, 2.04]
Kalapotharakos 2004 80 60 141 Isokinetic Peak Torque Left knee ext 60° deg/sec Right knee ext 60° deg/sec Left knee flex 60° deg/sec Right knee flex 60° deg/sec Left knee ext 180° deg/sec Right knee ext 180° deg/sec Left knee flex 180° deg/sec Right knee flex 180° deg/sec
-0.06 [-0.93, 0.82] 0.26 [-0.62, 1.14] -0.12 [-1.00, 0.76] 0.18 [-0.70, 1.05] 0.30 [-0.58, 1.17] 0.05 [-0.83, 0.92] 0.15 [-0.73, 1.03] 0.14 [-0.74, 1.02]
NA
Kalapotharakos 2005 80 60 As above Flexibility - sit and reach Walking Speed Chair Rise Stair Climb
-0.24 [-1.06, 0.58] 0.12 [-0.70, 0.94] 0.12 [-0.69, 0.95] -0.34 [-1.16, 0.49]
NA
Seynnes 2004 80 40 50 Endurance 1.71 [0.42, 3.01] NA Chair rise time -0.06 [-1.12, 0.99] Stair climb power 0.29 [-0.78,1.35] 6 minute walk distance 0.65 [-0.44, 1.75] Disability (HAQ – DI) 0.09 [-0.97, 1.15]
Singh 2005 80 20 25 Quality of life – vitality 0.59 [-0.10, 1.25] NA Depression Therapist rated (HRSD)
Self-rated (GDS) -0.15 [-0.82, 0.51] -0.16 [-0.83, 0.50]
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPTSullivan 2005 80 20 25 Aggregate physical score (sit to stand,
stair climb, safe and maximal gait speed)
ND, RNS NA
Sullivan 2007 80 20 As above Aggregate physical score (as above) ND NA
Tsutsumi 1997 75 – 85 55 – 65 105 Physical activity efficacy: Walking
-0.25 [-0.99, 0.50]
NA
Stair climb 0.40 [-0.35, 1.15] Physical Self-efficacy Scale:
PPA
0.37 [-0.38, 1.12]
PSPC -0.54 [-1.29, 0.22] Cognition
Mental arithmetic
0.48 [-0.28, 1.23]
Mirror drawing time
0.26 [-0.48, 1.01]
dots -0.14 [-0.88, 0.60] errors -0.97 [-1.77, -0.16] VO2 Max ND, RNS
Tsutsumi 1998 75 – 85 55 – 65 125 POMS tension 0.59 [-0.23, 1.41] NA POMS vigor -0.27 [-1.07, 0.54] State Anxiety (STAI) 0.27 [-0.53, 1.08] Trait Anxiety (STAI) 0.86 [0.02, 1.71] VO2 Max ND, RNS
Vincent, Braith, Feldman, Magyari et al 2002
80 50 102 Muscular endurance (leg) Stair climb time
0.14 [-0.44, 0.72] -0.57 [-1.16, 0.02]
NA
Vincent, Braith, Feldman, Kalls et al 2002
80 50 As above VO2 Peak Endurance – treadmill time
-0.06 [-0.63, 0.52] 0.25 [-0.34, 0.82]
NA
Note: Improvements in depression, stair climbing time, stepping up/down, disability, anxiety, Mirror drawing errors (cognition), perceived exertion are indicated by a negative effect size. Improvements in all other outcomes are indicated by a positive effect size. Max = maximal; Ext = extensors; Flex = flexors; HAQ DI = Health Assessment Questionnaire Disability Index subscale (French version); HRSD – Hamilton Rating Scale of Depression; GDS = Geriatric Depression Scale. POMS = Profile of Mood States; STAI = State-Trait Anxiety Inventory; ND = No Data; RNS = reported not significant; PPA = Perceived Physical Ability; PSPC = Physical Self-Presentation Confidence. *as calculated by dividing IEMG of the weight load while walking by maximal elbow flexion
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Records identified through database searching
(n = 1263)
Additional records identified through other sources
(n = 11 )
Records after duplicates removed (n = 912)
Total records excluded (n = 890) Reasons:
- Not land-based lower limb PRST (527)
- Combined therapies, compared different types of therapies or did not compare intensities of PRST (339)
- No outcomes of interest (8) - Did not meet age criteria (5) - Did not meet high intensity criteria
(9) - Not an RCT (1) - Systematic review (1)
Abstract and Full-text articles assessed for eligibility (n = 216)
Studies included in systematic review
(n = 21)
Figure 1. Flowchart showing the selection of studies for this systematic review.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Study or Subgroup
Fatouros, Tournis 2005Hortobagyi 2001Pruitt 1995Seynnes 2004
Total (95% CI)
Heterogeneity: Tau² = 0.50; Chi² = 8.82, df = 3 (P = 0.03); I² = 66%Test for overall effect: Z = 1.91 (P = 0.06)
Mean
102.21,19365.611.3
SD
14.6320
14.32.828
Total
14988
39
Mean
75.51,05062.79.7
SD
10.4360
11.91.715
Total
14976
36
Weight
26.0%26.0%24.6%23.4%
100.0%
IV, Random, 95% CI
2.05 [1.11, 2.98]0.40 [-0.54, 1.34]0.21 [-0.81, 1.22]0.62 [-0.48, 1.71]
0.83 [-0.02, 1.68]
HIPRST LIPRST Std. Mean Difference Std. Mean DifferenceIV, Random, 95% CI
-2 -1 0 1 2Favours LIPRST Favours HIPRST
Figure 2. High versus low intensity PRST for maximal lower limb strength.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Study or Subgroup
Cassilhas 2007Fatouros 2005Fatouros, Tournis 2005Harris 2004Kalapotharakos 2004Tsutsumi 1998Vincent...Magyari 2002Willoughby 1998
Total (95% CI)
Heterogeneity: Tau² = 0.15; Chi² = 14.00, df = 7 (P = 0.05); I² = 50%Test for overall effect: Z = 3.99 (P < 0.0001)
Mean
131.591.7
102.2244.0443.1230.68347.11.88
SD
18.728.2
14.671.6711.498.271670.32
Total
202014131112227
119
Mean
110.5376.885.9
222.7231.5428.73305.61.63
SD
17.079.4
12.983.4710.03
81140.43
Total
191812171212247
121
Weight
14.2%13.0%11.6%13.4%11.0%12.1%16.2%8.5%
100.0%
IV, Random, 95% CI
1.15 [0.46, 1.83]1.66 [0.91, 2.41]1.14 [0.30, 1.98]
0.26 [-0.46, 0.99]1.04 [0.16, 1.92]
0.23 [-0.57, 1.03]0.29 [-0.29, 0.87]0.62 [-0.46, 1.70]
0.79 [0.40, 1.17]
HIPRST MIPRST Std. Mean Difference Std. Mean DifferenceIV, Random, 95% CI
-2 -1 0 1 2Favours MIPRST Favours HIPRST
Figure 3. High versus moderate intensity PRST for lower limb strength.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
Figure 4. The effect of program duration on lower limb 1RM strength outcomes.
