Experimental Gerontology 48 (2013) 1351–1361 Contents lists available at ScienceDirect Experimental Gerontology journal homepage: www.elsevier.com/locate/expgero Strength training at high versus low external resistance in older adults: Effects on muscle volume, muscle strength, and force–velocity characteristics Evelien Van Roie a,⁎, Christophe Delecluse a, Walter Coudyzer b, Steven Boonen c, Ivan Bautmans d a Physical Activity, Sports and Health Research Group, Department of Kinesiology, Faculty of Kinesiology and Rehabilitation Sciences, Katholieke Universiteit Leuven, Tervuursevest 101, 3001 Leuven, Belgium b Radiology Section, Department of Morphology and Medical Imaging, Faculty of Medicine, Katholieke Universiteit Leuven, Herestraat 49, 3000 Leuven, Belgium c Leuven University Center for Metabolic Bone Disease and Division of Geriatric Medicine, Faculty of Medicine, Katholieke Universiteit Leuven, Herestraat 49, 3000 Leuven, Belgium d Gerontology Department and Frailty in Ageing Research Group, Vrije Universiteit Brussell (VUB), Laarbeeklaan 103, 1090 Brussels, Belgium a r t i c l e i n f o Article history: Received 13 May 2013 Received in revised form 18 July 2013 Accepted 21 August 2013 Available online 30 August 2013 Section Editor: Christiaan Leeuwenburgh Keywords: Muscle hypertrophy Resistance training Training load Muscle fatigue Elderly a b s t r a c t Muscle adaptations can be induced by high-resistance exercise. Despite being potentially more suitable for older adults, low-resistance exercise protocols have been less investigated. We compared the effects of high- and low-resistance training on muscle volume, muscle strength, and force–velocity characteristics. Fifty-six older adults were randomly assigned to 12 weeks of leg press and leg extension training at either HIGH (2 × 10–15 repetitions at 80% of one repetition maximum (1RM)), LOW (1 × 80–100 repetitions at 20% of 1RM), or LOW+ (1 × 60 repetitions at 20% of 1RM, followed by 1 × 10–20 repetitions at 40% of 1RM). All protocols ended with muscle failure. Leg press and leg extension of 1RM were measured at baseline and post intervention and before the first training session in weeks 5 and 9. At baseline and post intervention, muscle volume (MV) was measured by CT-scan. A Biodex dynamometer evaluated knee extensor static peak torque in different knee angles (PTstat90°, PTstat120°, PTstat150°), dynamic peak torque at different speeds (PTdyn60°s−1, PTdyn180°s−1, PTdyn240°s−1), and speed of movement at 20% (S20), 40% (S40), and 60% (S60) of PTstat90°. HIGH and LOW+ resulted in greater improvements in 1RM strength than LOW (p b 0.05). These differences were already apparent after week 5. Similar gains were found between groups in MV, PTstat, PTdyn60°s−1, and PTdyn180°s−1. No changes were reported in speed of movement. HIGH tended to improve PTdyn240°s−1 more than LOW or LOW+ (p = 0.064). In conclusion, high- and low-resistance exercises ending with muscle failure may be similarly effective for hypertrophy. Highresistance training led to a higher increase in 1RM strength than low-resistance training (20% of 1RM), but this difference disappeared when using a mixed low-resistance protocol in which the resistance was intensified within a single exercise set (40% of 1RM). Our findings support the need for more research on low-resistance programs in older age, in particular long-term training studies and studies focusing on residual effects after training cessation. © 2013 Elsevier Inc. All rights reserved. 1. Introduction Human aging is characterized by a progressive decline in skeletal muscle mass, accompanied by marked decreases in muscle strength and muscular function. These losses can have a significant impact on a person's ability to independently perform activities of daily life (Abellan van Kan, 2009; Marcell, 2003). It is clear that effective interventions are needed to prevent or reverse these losses in older adults. ⁎ Corresponding author at: Tervuursevest 101, 3001 Leuven, Belgium. Tel.: +32 16 379431; fax: +32 16 329197. E-mail addresses: evelien.vanroie@faber.kuleuven.be (E. Van Roie), christophe.delecluse@faber.kuleuven.be (C. Delecluse), walter.coudyzer@gmail.com (W. Coudyzer), steven.boonen@uzleuven.be (S. Boonen), ivan.bautmans@vub.ac.be (I. Bautmans). 0531-5565/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.exger.2013.08.010 For optimal muscle growth, strengthening exercise at moderate to high external resistance (70–85% of the one repetition maximum (1RM)) is recommended (American College of Sports Medicine, 2006). However, practitioners remain reluctant to prescribe exercises that challenge the musculoskeletal system of older adults at high external resistances (Abellan van Kan et al., 2009). Interestingly, numerous studies suggest that low-resistance exercise, with vascular occlusion, can induce gains in muscle volume and strength equivalent to those observed after training at higher resistances (Karabulut et al., 2010; Loenneke et al., 2012; Takarada et al., 2000). In line with these and other (Taaffe et al., 1996; Tanimoto and Ishii, 2006; Van Roie et al., 2013) findings, Mitchell et al. (2012) provided evidence that lifting low resistances to the point of momentary muscle fatigue (failure) leads to hypertrophy and strength gains roughly equivalent to those achieved with conventional high-resistance training. This would suggest 1352 E. Van Roie et al. / Experimental Gerontology 48 (2013) 1351–1361 that the use of high external resistances may not be a necessity to elicit muscle hypertrophy. Rather than having to be exposed to high external resistances, achieving maximal effort might be of greater importance for gains in muscle mass and strength (Carpinelli, 2008; Goto et al., 2005; Rooney et al., 1994; Schott et al., 1995). Maximal effort is typically reached when performing a series of repetitions to the point of momentary muscle fatigue. This maximal effort might be needed to maximize motor unit recruitment and thus to enhance the hypertrophic response (Carpinelli, 2008; Goto et al., 2005). In previous research, endurance-type strength training regimens have been compared to traditional high-resistance training. Effects on muscle mass and muscle strength, however, remain inconclusive (Anderson and Kearney, 1982; Campos et al., 2002; Holm et al., 2008; Taaffe et al., 1996). Remarkably, most endurance-type protocols restricted the number of repetitions per set to about 20 or 30. Takarada and Ishii (2002) suggested that, in these protocols, the interset rest period should be kept short (~30 s) to reduce metabolite clearance, which in turn creates the need for additional motor unit recruitment in subsequent sets. One can expect that high-repetition exercise protocols (≥60 repetitions) would also be effective to maximize motor unit recruitment. However, to date, virtually no studies have focused on the effects of such high-repetition protocols. Another interesting approach to further optimize motor unit recruitment might be to vary training resistance. Training resistance can vary not only over the course of a training period (periodization), but also from set to set within a single training session (Tan, 1999). A previous study by Goto et al. (2004) suggested that adding a single set of exercise at 50% of 1RM to a strength-type regimen at 90% of 1RM may optimize strength adaptation. A recent study tested a strength training protocol in young adults, in which a highly fatiguing protocol of 60 repetitions at 20–25% of 1RM was immediately followed (no rest) by a set of 10 repetitions at 40% of 1RM. This mixed low-resistance exercise protocol showed interesting benefits on dynamic strength and speed of movement of the knee extensors (Van Roie et al., 2013). Especially in older adults, these muscle parameters are of major importance in activities of daily living (Sayers et al., 2005; Van Roie et al., 2011). In this study, the purpose was to compare the effects of low-resistance training at high repetitions with traditional high-resistance training at low repetitions (HIGH) on muscle volume, muscle strength, and force– velocity characteristics in older adults. To further investigate the beneficial effect of varying training resistance within a single training session, two low-resistance exercise protocols were created: one high-repetition low-resistance protocol (LOW) in which external resistance was kept constant within one session, and also one mixed high-repetition lowresistance protocol (LOW+) in which resistance was increased after 60 repetitions. All training protocols were designed to end with maximal effort (i.e. muscle fatigue and failure to continue the exercise). If performing repetitions to the point of momentary muscle fatigue is sufficient to reach (near) maximal motor unit recruitment, and thus to activate type II muscle fibers, all exercise protocols would be expected to be effective in improving muscle volume, muscle strength, and force–velocity characteristics. However, it might be possible that, in addition to muscle fatigue, a mechanical stimulus is needed to activate type II muscle fibers, especially when training at very low external resistances (protocols with many repetitions). If that is the case, increasing the resistance in LOW+ at the end of a high-repetition protocol should allow this group to obtain better effects than LOW on muscle volume, muscle strength, and force–velocity characteristics. 2. Methods 2.1. Study participants Community-dwelling adults aged 60 and older were locally recruited through advertisements and oral communications for inclusion in a 12-week resistance training program. Exclusion criteria were current participation in structured endurance exercise and/or participation in resistance exercise in the last 6 months prior to the study, knee or hip problems, unstable cardiovascular disease, neuromuscular disease, and acute hernia. A flowchart of the study is provided in Fig. 1. Fifty-six older men and women were randomly assigned to one of three training interventions: traditional high-resistance training (HIGH), low-resistance training (LOW), or mixed low-resistance training (LOW +). Randomization was stratified for gender, age, and baseline isometric knee extension strength. The study was approved by the University's Human Ethics Committee in accordance with the declaration of Helsinki. All participants provided written informed consent. 2.2. Resistance training protocol The exercise sessions were organized at a local fitness and health center over a period of 12 weeks. Baseline and post intervention measurements were performed from January to March 2012 and from April to June 2012, respectively. After an initial familiarization session, in which proper lifting technique was demonstrated and practiced for each of the exercises, participants exercised three times weekly on nonconsecutive days for 12 weeks (total of 36 sessions). Exercise equipment included leg press and leg extension (Technogym, Gambettola, Italy). Training programs were based on previous research (Van Roie et al., 2013). Each training session started with a 10 min warm-up on a cycle ergometer (Technogym Bike Excite, Gambettola, Italy) or on a treadmill (walking pace) (Technogym Run Excite, Gambettola, Italy). Participants were instructed to perform all exercises at a moderate speed, i.e. 2 s for each concentric and 3 s for each eccentric action. Between the exercises, a rest period of at least 2 min was provided. The IsoControl (Technogym, Gambettola, Italy) provided feedback on the number of repetitions, the movement speed, the rest period between sets, and the range of motion during exercise. All training sessions were closely supervised by a qualified fitness instructor, and participants were verbally encouraged to continue the exercises until failure (i.e. inability to perform more repetitions due to local muscle fatigue). Immediately after each individual exercise, participants graded their level of perceived exertion on the OMNIResistance Exercise Scale of Perceived Exertion (scale from 0 to 10) (Robertson et al., 2003). Exercise protocols were initially designed to be approximately equal in volume (% resistance × repetitions) (Fig. 2). The protocol used in HIGH was based on ACSM's guidelines for resistance training (American College of Sports Medicine, 2006). These guidelines recommend performing at least one set to the point of failure for healthy individuals. In HIGH, the external resistance was initially set at 80% of 1RM. To ensure that maximal effort would be reached at the end of each set, participants were instructed to perform at least 10 to 15 repetitions. Two sets were performed with one minute of rest between sets. In LOW, the external resistance was initially set at about 20% of 1RM. Participants were instructed to perform 1 set of 80 to 100 repetitions. In LOW +, participants were instructed to complete a fatiguing protocol of 60 repetitions at about 20% of 1RM. Immediately afterwards (no rest), external resistance was increased to about 40% of 1RM and participants were instructed to perform 10 to 20 additional repetitions. In all groups, participants were encouraged to continue the exercise if maximal effort was not achieved after the prescribed number of repetitions. External resistance was adjusted if participants performed repetitions beyond the prescribed training zone, as well as if the rate of perceived exertion dropped below 6. This strategy was used to ensure that maximal effort was reached at the end of each exercise set, as this may be necessary to optimize muscular adaptations (Goto et al., 2005; Rooney et al., 1994; Schott et al., 1995). 1353 E. Van Roie et al. / Experimental Gerontology 48 (2013) 1351–1361 89 responded to advertisement Excluded (n = 33): - Declined informed consent (n= 19) - Current participation in exercise (n = 3) - Unstable cardiovascular disease (n = 4) - Acute hernia (n = 4) - Hip (n = 1) and knee (n = 1) problems - Planned surgery (n = 1) Randomized and completed pretests (n = 56) Allocated to HIGH Allocated to LOW Allocated to LOW+ (n = 18, 8m, 10f) (n = 19, 9m, 10f) (n = 19, 9m, 10f) Completed posttests Completed posttests Completed posttests (n = 18) (n = 19) (n = 19) Fig. 1. Flowchart of the study. Average training volume per exercise session was calculated for each participant as: Xn i¼1 ðnumber of repetitions leg press  %1RMÞi þ ðnumber of repetitions leg extension  %1RMÞi n with n = total number of exercise sessions performed during the 12week training intervention. This equation used relative data on training resistance (training load as a percentage of 1RM) instead of absolute data (in kg), as the latter can be biased by a number of confounding factors. Amongst others, confounding factors that influence training load (kg) in individuals may be gender, body weight, strength level, and muscle volume. 2.3. Outcome measures 2.3.1. Muscle volume At baseline and post intervention, four 2.5 mm-thick axial slices were measured in the middle of the right upper leg by a computed tomography scan (Siemens SOMATOM Definition Flash, Forchheim, Germany). The axial images were obtained at the midpoint of the distance between the medial edge of the greater trochanter and the intercondyloid fossa of the femur. The four slices were then put together as one 10 mm-thick slice and the software program Volume (Siemens) calculated the overall muscle volume (in cm3) for this 10 mm-thick slice. Standard Hounsfield Units ranges for skeletal muscle (0–100) were used to segment muscle tissue area. Corrections were made for bone marrow. All measurements were performed by one expert radiologist in the university hospital. 2.3.2. One repetition maximum and local muscular endurance One repetition maximum was evaluated every 4 weeks: at baseline (pre), before the first training session in week 5 and week 9, and after 12 weeks of training (post). Training volume on the test sessions in week 5 and week 9 was reduced to only the leg extension exercise. Local muscular endurance was assessed at baseline and post intervention. The assessment of leg press 1RM started with a warm-up set of 8 repetitions at approximately 50% of the estimated 1RM, followed by another set of 3 repetitions at 70% of the estimated 1RM. Subsequent lifts (ranging from 3 to 5) were single repetitions with progressively heavier resistances until failure. A rest period of 1 to 5 min was allotted between each attempt to ensure recovery. The heaviest successful lift was determined as 1RM. Leg extension 1RM was estimated through a logarithmic regression formula (Van Roie et al., 2013). In this testing procedure, participants had to complete a maximum number of repetitions at a resistance 1354 E. Van Roie et al. / Experimental Gerontology 48 (2013) 1351–1361 that was selected by the supervisor (5 to 15 repetitions were targeted). Using the formula, the percentage of 1RM was estimated based on the number of repetitions, and subsequently, the 1RM was derived. This method was chosen over a direct method, as it was considered less time consuming. After determining leg extension 1RM, the local muscular endurance test was assessed, consisting of completing a maximum number of repetitions at 60% of 1RM (until failure) (Campos et al., 2002). 2.3.3. Force–velocity characteristics Force–velocity characteristics of the knee extensors were measured on a Biodex Medical System 3® dynamometer (Biodex Medical Systems, Shirley, NY). Tests were performed unilaterally on the right side, unless there was a medical contraindication. Participants were seated on a backwardly-inclined (5°) chair and secured with safety belts across the upper leg of the test side, the hips, and the shoulders. The rotational axis of the dynamometer was aligned with the transversal knee-joint axis and was attached to the tibia with a length-adjustable lever arm. The position of the rotational axis and the chair, as well as the length of the lever arm were identical at pre- and posttest. The protocol consisted of three standardized tests: an isometric test, a speed of movement test, and an isokinetic test. These three tests, as described below, were consecutively performed with about 2 min of rest between tests. After a first completion of the three tests and a rest period of 5 min, the protocol was repeated for a second time. The best performance for each parameter was reported. First, static knee extension strength was obtained at different knee joint angles: 120°, 90°, and 150°. Maximal voluntary isometric contractions over a 5-second period were performed twice at each angle, separated by a 15-second rest interval. The static peak torque (Nm) recorded at these knee joint angles was reported as PTstat120°, PTstat90°, and PTstat150°, respectively. Second, three ballistic speed of movement tests for the knee extensors were performed, separated by a 20-second rest interval. Participants extended the lower leg twice as quickly as possible from a knee joint angle of 90° to 160°. Resistances were 20%, 40%, and 60% of the individual static peak torque (90°). The maximum speed (°s−1) recorded at these resistances was reported as S20, S40, and S60, respectively. Third, three maximal isokinetic extension–flexion movements were performed at an angular velocity of 60°s− 1, followed by five movements at both 180°s− 1 and 240°s− 1. A 20-second rest period was provided between the sequences. The dynamic peak torque (Nm) of the knee extensors, irrespective of knee angle, was recorded as PTdyn60°s− 1, PTdyn180°s− 1, and PTdyn240°s− 1, respectively. 2.3.4. Functional performance The modified physical performance test (mPPT) (Brown et al., 2000) was executed at baseline in order to document the overall functional performance level of our study sample. The mPPT consists of nine functional items: (i) Romberg test for balance, (ii) chair sit-to-stand test, (iii) lifting a book from waist height to a shelf at shoulder level, (iv) putting on and taking off a coat, (v) picking up a penny from the floor, (vi) turning 360°, (vii) walking 15 m, (viii) ascending one flight of stairs, and (ix) climbing four flights of stairs. The score of each item ranged from 0 (the inability to complete the task) to 4 (the highest level of performance), with a summary performance score (mPPTscore) of maximum 36 points. To analyze the effects of the intervention on functional performance, the following tests were performed at baseline and post intervention: 6-minute walk test, maximal gait speed test, 30-second chair sit-tostand test, 5-repetition chair sit-to-stand test, and timed up-and-go test. The 6-minute walk test was performed over a walking course of 20 m (American Thoracic Society, 2002). Participants walked up and down the course at a fast but comfortable pace, and the 6-minute walk distance (in meters) was reported. To measure maximal gait speed, participants were asked to walk 7.5 m as quickly as possible without running. The test was performed twice, and the best result (in meters per second) was used. The 5-repetition and 30-second chair sit-to-stand tests were performed using a standard chair without arm rests (McCarthy et al., 2004). Participants crossed both arms against the chest, started from a seated position (upper back against seat), stood up to full extension and sat down again (upper back against seat). The time required (in seconds) to perform 5 chair stands was evaluated in the 5-repetition chair sit-to-stand test. This test was performed twice, using the best result in further analyses. In the 30-second chair sit-to-stand test, the number of successful repetitions was counted over a 30-second period. Due to the exhausting nature of this test, it was performed only once. In the timed up-and-go test, the time (in seconds) required for the participant to stand up from a standard armchair, walk a distance of 3 m, turn, walk back and sit down again was measured (Bischoff et al., 2003; Podsiadlo and Richardson, 1991). The test was performed as quickly as possible, however, without running. The best result from two trials was used. 2.4. Statistical analyses Data were initially analyzed for normality with a Shapiro–Wilk test. The following variables were not normally distributed: muscle volume, leg press 1RM, static peak torque (at 90°), speed of movement (at 40% of PTstat90°), dynamic peak torque (at 60°s−1, 180°s−1, 240°s−1), mPPTscore, and 5-repetition chair sit-to-stand test. For all normally distributed variables, one-way analysis of variance with Bonferroni post hoc testing was used to test for baseline differences between groups. Within-group changes from baseline to post were analyzed with paired T-tests. To assess between-group differences in changes over time, linear mixed-model analysis with an unstructured covariance structure was used, with time as repeated factor and group as fixed factor. To account for differences in average training volume, this variable was introduced as a covariate in the mixed model analyses. Post hoc analyses were conducted to determine differences in changes between groups. For all non-normally distributed variables, Kruskal–Wallis tests were used to search for baseline differences between groups. Time-effects from baseline to post intervention were analyzed with Friedman tests, and within-group changes from baseline to post were analyzed with Wilcoxon-signed rank tests. Percent changes from baseline to post were calculated for each individual and were divided by the average training volume to account for differences in training volume. These variables were then used in Kruskal–Wallis tests to determine differences in changes between groups. Only when significance was revealed with the Kruskal–Wallis test, Mann–Whitney U tests were used as post hoc tests. Pearson's correlation coefficient was determined to assess associations between percent changes in muscle parameters and percent changes in functional performance measures. All statistical tests were executed with SPSS software version 19 (SPSS Inc., Chicago, IL). Level of significance was set at p b 0.05. 3. Results 3.1. Baseline characteristics and training adherence No side effects of the intervention were reported in any of the groups. All subjects completed the study. However, two Biodex measurements (one baseline for HIGH, one post for LOW+) failed due to a lack of compliance of the participants with the test instructions and were excluded from analysis. Overall adherence (number of training sessions attended as a percentage of the total number of training sessions) to the training program was 95.7% in HIGH, 95.8% in LOW, and 95.3% in LOW+, with no significant differences between groups. 1355 E. Van Roie et al. / Experimental Gerontology 48 (2013) 1351–1361 Moreover, none of the participants' characteristics (Table 1) nor any of the outcome variables differed between groups at baseline (all p N 0.05). At baseline, the overall summary performance score on the mPPT was 35.34 ± 1.16. 3.2. Training volume Training variables are listed in Table 2. The average training volume accomplished during each session was calculated for all groups. Although training programs were designed to be approximately equal in volume, the average training volume per session on the leg extension exercise revealed a significantly higher volume in LOW compared to HIGH and LOW+ (p b 0.05). Average training volume (leg press and leg extension summed) was therefore used as a covariate in the mixed model analyses and in the Kruskal–Wallis tests. 3.3. Outcome measures 3.3.1. Muscle volume Muscle volume of the upper leg increased significantly over time, with no difference between HIGH (+3.2 ± 3.7%, p = 0.003), LOW (+2.4 ± 2.7%, p = 0.002), and LOW+ (+2.6 ± 3.8%, p = 0.016) (Table 3 and Fig. 3A). 3.3.2. One repetition maximum and local muscular endurance All training groups showed a significant increase from baseline to post in both leg press 1RM and leg extension 1RM (all p b 0.05). With regard to leg press 1RM, a significant time by group interaction effect was observed from baseline to post (p = 0.002), with post hoc tests revealing that both HIGH (+46.2% ± 32.3%) and LOW+ (+39.2 ± 20.7%) increased significantly more than LOW (+23.1% ± 20.7%) (p = 0.001 and p = 0.006, respectively) (Table 3 and Fig. 3B). From baseline to week 5, a higher increase was found in HIGH (p b 0.001) and LOW+ (p = 0.002) compared to LOW (Fig. 4). For leg extension 1RM, linear mixed model analysis revealed a time by group interaction effect from baseline to post (p = 0.003), with HIGH (+30.0 ± 11.5%) and LOW+ (+29.7 ± 19.8%) improving significantly more than LOW (+19.2 ± 5.3%) (p = 0.001 and p = 0.011, respectively) (Table 3). From baseline to week 5, LOW showed less of an increase than both LOW+ (p = 0.007) and HIGH (p = 0.040). From week 5 to week 9, HIGH tended to increase more than HIGH 10-15 1-min 10-15 reps reps 80% 1RM LOW 80% 1RM LOW (p = 0.067). From week 9 to post, HIGH increased more than both LOW (p = 0.011) and LOW+ (p = 0.045) (Fig. 5). Local muscular endurance increased significantly in LOW (+ 19.8 ± 29.8%, p = 0.021) and LOW + (+ 16.3 ± 20.6%, p = 0.008). However, no time by group interaction effect was found for this variable (p = 0.770) (Table 3). 3.3.3. Force–velocity characteristics No differences were observed between groups for changes in static peak torque (Fig. 3C), speed of movement, or dynamic peak torque at 60°s−1 and at 180°s−1 (Table 3, all p N 0.05). Static peak torque at all knee joint angles increased significantly in all groups, although only a trend was found for PTstat90° in HIGH (p = 0.084) (Fig. 6). Speed of movement (S20, S40, and S60) did not change from baseline to post in any of the groups. Dynamic peak torque at 60°s−1 tended to increase in LOW only (p = 0.051). Dynamic peak torque at 180°s−1 increased significantly in HIGH (p = 0.011) and LOW (p = 0.018). At 240°s−1, HIGH was the only group with a significant increase in dynamic peak torque, although LOW also tended to show an increase (p = 0.064). A trend towards a significant time by group interaction effect was found for dynamic peak torque at 240°s−1 (p = 0.064) (Table 3), with HIGH (+6.9 ± 8.3) improving more than LOW+ (+1.4 ± 6.4, p = 0.044) and LOW (+2.5 ± 6.0, p = 0.041). 3.3.4. Functional performance A trend towards a time by group interaction effect was only observed for maximal gait speed (p = 0.051), with HIGH improving significantly more than LOW+ (p = 0.023) and tending to improve more than LOW (p = 0.051). However, for 6-minute walk distance, 30-second and 5-repetition chair sit-to-stand tests, and timed up-andgo test, no time by group interaction effect was revealed (all p N 0.05) (Table 4). HIGH showed a significant improvement for 6-minute walk distance (p = 0.029), 30-second chair sit-to-stand test (p b 0.001), and 5-repetition chair sit-to-stand test (p = 0.001), and a trend to improvement for timed up-and-go test (p = 0.076). In LOW, a significant improvement was only seen for 5-repetition chair sit-to-stand test (p b 0.001), and a trend for improvement for 30-second chair sit-tostand test (p = 0.050). LOW+ showed a significant improvement for 30-second chair sit-to-stand test (p b 0.001) and 5-repetition chair sitto-stand test (p = 0.001), and a trend to improvement for 6-minute walk distance (p = 0.060) and timed up-and-go test (p = 0.063). Most Pearson's correlation coefficients between percent changes in muscle parameters and changes in functional performance did not reach statistical significance. Only the change in PTstat90° was positively correlated with the change in 30-second chair sit-to-stand test (r = 0.30, p = 0.030) and negatively correlated with the change in timed up-and-go test (r = −0.41, p = 0.002). 4. Discussion In this study, we investigated whether high-repetition low-resistance exercise protocols (LOW and LOW+) would be similarly effective in 80-100 reps Table 1 Participants' characteristics at baseline (mean ± SD). 20% 1RM LOW+ 60 reps 20% 1RM 10-20 reps 40% 1RM Fig. 2. Protocols for exercise training on leg press and leg extension. HIGH LOW+ LOW p (n = 18: 8 m, 10f) (n = 19: 9 m, 10f) (n = 19: 9 m, 10f) Age (y) Weight (kg) Height (m) BMI (kg/m2) mPPT-score 67.72 71.58 1.67 25.54 35.61 ± ± ± ± ± 4.28 11.33 0.08 2.82 0.70 67.43 76.63 1.66 27.57 35.32 ± ± ± ± ± 5.90 12.10 0.08 3.08 1.06 68.76 75.58 1.65 27.62 35.11 ± ± ± ± ± 4.96 14.77 0.09 4.12 1.56 0.701a 0.460a 0.769a 0.117a 0.616b HIGH = high-resistance training; LOW+ = mixed low-resistance training; LOW = low-resistance training; mPPT = modified physical performance test. a Results of one-way analysis of variance between baseline group means. b Results of Kruskal–Wallis test. 1356 E. Van Roie et al. / Experimental Gerontology 48 (2013) 1351–1361 Table 2 Training variables (mean ± SD). Repetitions per set Leg press set 1 Leg press set 2 Leg extension set 1 Leg extension set 2 Leg press set 1 Leg press set 2 Leg extension set 1 Leg extension set 2 Leg press Leg extension Leg press + leg extension Resistance (% of 1RM) Training volume HIGH (n = 18: 8 m, 10f) LOW+ (n = 19: 9 m, 10f) LOW (n = 19: 9 m, 10f) 16.1 ± 15.1 ± 13.6 ± 12.2 ± 87.2 ± 87.0 ± 76.2 ± 76.1 ± 27.4 ± 19.7 ± 47.2 ± 60.1 20.8 60.1 16.6 29.9 50.3 24.2 45.0 28.6 22.1 50.8 91.6 ± – 87.8 ± – 35.5 ± – 30.0 ± – 32.2 ± 26.3 ± 58.7 ± 2.4 1.8 1.5 1.1 9.5 9.3 2.7 2.6 5.8 2.3 6.9 ± ± ± ± ± ± ± ± ± ± ± 0.1 7.1 0.3 2.7 5.0 7.2 5.2 6.6 6.2 6.2 11.3 9.7 9.3 8.4 3.5 6.7 5.3a,b 9.0a,b HIGH = high-resistance training; LOW+ = mixed low-resistance training; LOW = low-resistance training; 1RM = one repetition maximum. n ∑ ðnumber of repetitions×%1RMÞi , with n = total number of exercise sessions performed during the 12-week training intervention. Training volume = i¼1 n a Different from HIGH (p b 0.05, results of one-way analysis of variance with Bonferroni post hoc testing). b Different from LOW+ (p b 0.05, results of one-way analysis of variance with Bonferroni post hoc testing). achieving gains in muscle volume, muscle strength, force–velocity characteristics, and functional performance as traditional high-resistance training (HIGH). This question is of great importance for older adults, as muscle volume and strength decline with aging and since practitioners remain skeptical about applying high-resistance exercise in this population. In one of the low-resistance exercise protocols used in this study (LOW+), external resistance was increased after a fatiguing protocol of 60 repetitions in order to examine the beneficial effect of intensifying the training resistance at the end of a single exercise set. Interestingly - despite doing a smaller volume of work- compared to LOW, HIGH and LOW+ achieved the same or greater improvements on muscle volume, muscle strength, and functional performance. In particular, all groups demonstrated similar gains in muscle volume (2.4–3.2%), irrespective of the resistance used during training. These results are in line with previous research showing that lowresistance exercise, as long as maximal effort is reached, can induce comparable hypertrophic responses as high-resistance exercise (Mitchell et al., 2012; Taaffe et al., 1996; Takarada et al., 2000). Table 3 Mean and SD for baseline (pre-) and posttest and % change (±SD) for muscle volume, one repetition maximum, and force–velocity characteristics of the knee extensors for the three intervention groups. HIGH (n = 18) 3 MV (cm ) 1RMLP (kg) 1RMLE (kg) END (reps) PTstat90° (Nm) PTstat120° (Nm) PTstat150° (Nm) S20 (°s−1) S40 (°s−1) S60 (°s−1) PTdyn60°s−1 (Nm) PTdyn180°s−1 (Nm) PTdyn240°s−1 (Nm) Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post Mean SD 150.2 154.5 99.8 137.2 30.0 38.8 16.4 17.8 162.5 166.1 143.5 156.3 74.2 79.0 370.0 370.3 260.7 263.2 172.6 183.6 126.7 131.4 80.3 83.3 71.2 74.6 39.0 38.1 56.4 64.9 6.0 7.5 3.8 2.5 55.0 51.1 39.9 44.7 19.8 22.0 52.0 48.1 33.1 41.9 29.8 34.0 37.8 37.9 25.8 26.1 21.7 22.4 LOW+ (n = 19) % 3.2 ± 3.7d 46.2 ± 32.3d,e 30.0 ± 11.5d,e 14.6 ± 31.3 5.5 ± 8.7 11.8 ± 7.3d 8.6 ± 9.8d 0.0 ± 3.6 −0.5 ± 6.6 5.1 ± 12.3 5.7 ± 11.8 6.1 ± 7.9d 6.9 ± 8.3d Mean SD 155.0 158.4 96.2 130.9 29.5 37.5 16.6 18.9 158.3 168.7 140.2 154.4 71.0 75.9 367.8 364.1 264.8 258.8 182.1 180.8 127.4 130.1 81.1 83.7 73.7 75.4 38.8 37.6 40.7 51.3 6.7 6.5 4.7 5.4 54.0 56.4 40.0 42.6 18.9 19.3 48.3 47.3 42.4 45.2 45.8 53.4 35.3 37.8 22.9 24.2 21.4 22.6 LOW (n = 19) % 2.6 ± 3.8d 39.2 ± 20.7d,e 29.7 ± 19.8d,e 16.3 ± 20.6d 6.8 ± 9.3d 10.1 ± 9.2d 7.7 ± 11.9d −0.6 ± 5.5 −1.3 ± 12.0 1.0 ± 30.2 1.9 ± 9.5 2.6 ± 5.0 1.4 ± 6.4 Mean SD 156.6 160.3 110.0 132.2 31.0 36.9 17.1 19.8 158.7 164.1 138.0 150.4 68.9 76.9 360.5 366.7 254.3 261.0 179.8 189.9 128.9 134.3 81.9 84.9 72.2 74.2 36.0 36.8 56.3 68.3 8.4 10.2 3.3 3.8 56.5 59.5 44.3 47.7 22.5 27.2 49.3 45.3 40.7 32.1 36.8 31.5 44.6 45.7 28.8 29.4 24.5 25.9 p-Values % Time Time × group 2.4 ± 2.7d b0.001b 0.570c 23.1 ± 20.7d b0.001b 0.002c 19.2 ± 5.3d b0.001a 0.003a 19.8 ± 29.8d b0.001a 0.770a 3.3 ± 5.7d b0.001b 0.202c 9.5 ± 10.5d b0.001a 0.749a 11.4 ± 14.8d b0.001a 0.634a 2.2 ± 8.3 0.751a 0.338a 4.0 ± 12.5 0.276b 0.326c 9.1 ± 28.3 0.229a 0.430a 4.8 ± 9.1 0.074b 0.620c b0.001b 0.319c 0.006b 0.064c 3.8 ± 6.5d 2.5 ± 6.0 HIGH = high-resistance training; LOW + = mixed low-resistance training; LOW = low-resistance training; MV = muscle volume; 1RMLP = leg press one repetition maximum; 1RMLE = leg extension one repetition maximum; END = muscular endurance (number of repetitions at 60% of 1RMLE); PTstatx = static (isometric) peak torque at knee angle of x°; Sx = speed of movement at x% of PTstat90°; PTdynx = dynamic (isokinetic) peak torque at x°s−1. a Results of linear mixed models analyses, time × group effect corrected for average training volume. b Results of Friedman test. c Results of Kruskal–Wallis test corrected for average training volume; significance level p b 0.05. d Significant change from pre to post (p b 0.05). e Significant difference with LOW (p b 0.05). 1357 A 10 5 0 -5 -10 HIGH LOW+ LOW % change in 1RM leg press 160 B a p = 0.001 b p = 0.006 160 150 140 a p < 0.001 b p = 0.002 HIGH 130 LOW+ LOW 120 110 100 0 pre week 5 week 9 post Time Fig. 4. Leg press one repetition maximum (mean ± standard error) for high-resistance training group (HIGH), low-resistance training group (LOW), and mixed low-resistance training group (LOW+). 1RM was measured at baseline (pre), before the first training session at weeks 5 and 9, and post intervention. Baseline 1RM was equated as 100%. a Difference in change between HIGH and LOW (p b 0.05). bDifference in change between LOW+ and LOW (p b 0.05). 80 60 40 20 0 HIGH LOW+ LOW* 35 % change in static peak torque (120°) 170 100 C 30 25 20 each exercise session. However, within the scope of this study we were not able to assess the contribution of central and peripheral factors in the occurrence of muscle failure in our participants. Another training variable that needs to be taken into account when studying the hypertrophic response to resistance exercise is the volume of work (% resistance x repetitions) performed during training (Krieger, 2010; Marx et al., 2001; Mitchell et al., 2012). In this study, exercise protocols were designed to be approximately equal in training volume. However, in LOW, the external resistance (35.5 ± 8.4% of 1RM for leg press and 30.0 ± 3.5% of 1RM for leg extension) used to reach muscle fatigue within the prescribed number of repetitions was higher than 15 10 5 0 -5 -10 HIGH LOW+ LOW Fig. 3. Plot of the individual percent changes (baseline to post) in muscle volume (A), 1RM leg press (B), and static peak torque (knee joint angle of 120°) (C) for high-resistance training group (HIGH), low-resistance training group (LOW), and mixed low-resistance training group (LOW+). *Significant difference with HIGH and LOW+ (p b 0.05). Henneman's size principle of motor unit recruitment indicates that, when a submaximal contraction is sustained, initially recruited motor units will fatigue, creating the need to additionally activate larger motor units. When the exercise is repeated to the point of muscle failure, (near) maximal motor unit recruitment will occur, regardless of the external resistance used (Carpinelli, 2008; Henneman, 1957). Thus, activation of a similar amount of muscle fibers can be expected when training with either high or low resistances until muscle failure, clarifying the equivalent extent of hypertrophy found in our three training groups. However, it cannot be excluded that differences in central and peripheral fatigue might have occurred between the three protocols, leading to different levels of muscle fiber activation at the end of Percentage one repetition maximum leg extension % change in muscle volume 15 Percentage one repetition maximum leg press E. Van Roie et al. / Experimental Gerontology 48 (2013) 1351–1361 140 a p = 0.001 b p = 0.011 a p = 0.011 c p = 0.045 130 a p = 0.040 b p = 0.007 120 HIGH LOW+ LOW 110 100 0 pre week 5 week 9 post Time Fig. 5. Leg extension one repetition maximum (mean ± standard error) for highresistance training group (HIGH), low-resistance training group (LOW), and mixed lowresistance training group (LOW+). 1RM was measured at baseline (pre), before the first training session at weeks 5 and 9, and post intervention. Baseline 1RM was equated as 100%. aDifference in change between HIGH and LOW (p b 0.05). bDifference in change between LOW+ and LOW (p b 0.05). cDifference in change between HIGH and LOW+ (p b 0.05). 1358 E. Van Roie et al. / Experimental Gerontology 48 (2013) 1351–1361 * * * * * * * * Fig. 6. Box plot representing percent changes in static peak torque of the knee extensors at different knee joint angles for high-resistance training group (HIGH), low-resistance training group (LOW), and mixed low-resistance training group (LOW+). *Significant change (p b 0.05). the initially aimed resistance of about 20% of 1RM. It cannot be excluded that (improvements of) strength-endurance capacity of the muscle might have played a role here. This phenomenon led to a higher average training volume for LOW than initially anticipated. On the leg extension, average training volume was significantly higher in LOW than in both HIGH and LOW+. This difference in training volume between groups was probably a consequence of the strategy used to adjust training resistance over the training period. Another strategy could have been to use a fixed number of repetitions at a predetermined resistance (% 1RM) in order to exactly match training volumes between groups. However, maximal effort would not have been reached in all participants when using the latter strategy. This could potentially lead to smaller muscular gains (Goto et al., 2005; Rooney et al., 1994; Schott et al., 1995). To ascertain that this difference did not interfere with the exercise-induced effects, all analyses were corrected for average training volume as confounding factor. Confirming our results on hypertrophy, basic strength gains obtained from the Biodex dynamometer, including static strength and dynamic strength at low speed (60°s−1), did not differ between the three training groups. These training-induced gains, ranging from 3.3 to 11.8% and from 1.9 to 5.7% for static and dynamic strength parameters, respectively, were rather low compared to gains previously reported in older adults (Cannon and Marino, 2010; Hortobagyi et al., 2001). Although none of the subjects had been participating in resistance exercise prior to the start of the intervention, they were all healthy, well-functioning and rather active older adults. Gains would probably have been greater if weaker and more sedentary older adults were included in the study. As typically found in previous research, gains in 1RM strength exceeded gains obtained from the Biodex dynamometer, probably because of neuromuscular adaptations specific to the trained movement. These 1RM strength gains were comparable to gains in previous studies (Cannon and Marino, 2010; Hortobagyi et al., 2001; Lemmer et al., 2000; Sayers and Gibson, 2010; Slivka et al., 2008). It should be noted, however, that training resistance had an impact on 1RM strength. HIGH clearly demonstrated a larger gain in 1RM than LOW on both the leg press and leg extension exercise. Previous work on the impact of different external resistances focused on 1RM strength of the trained movement. These studies agree with our findings and suggest that the use of high external resistances is a prerequisite for maximizing gains in 1RM (Anderson and Kearney, 1982; Campos et al., 2002; Holm et al., 2008; Mitchell et al., 2012). Overall, our data support the “strength-endurance continuum” of DeLorme (1945) suggesting that low-repetition high-resistance training favors strength adaptations and high-repetition low-resistance training increases muscular endurance. However, a unique aspect of the current study was the design of a low-resistance exercise protocol, in which training resistance varied within a single set (LOW +). More specifically, after a fatiguing protocol of 60 repetitions, additional repetitions at a higher resistance were performed until failure. Importantly, training resistance always remained low (≤ 50% of 1RM). Interestingly, no differences were found in 1RM gains between HIGH and LOW +. Nevertheless, differences in 1RM gains between LOW + and LOW did occur. These differences might be linked to the supplementary mechanical stimulus in LOW +, created by increasing resistance after a highly repetitive fatiguing exercise protocol. However, it might as well be the case that differences in voluntary muscle activation at fatigue between LOW and LOW + account for the difference in 1RM gain. There are studies suggesting that sustained contractions at low resistance tend to cause more central fatigue, while protocols at higher resistances tend to induce more peripheral fatigue (or fatigue within the muscle itself) (Neyroud et al., 2012; Yoon et al., 2007). However, these studies investigated sustained isometric contractions, contrary to repetitive dynamic contractions as in our training protocol. Kay et al. (2000) and Babault et al. (2006) already stated that neuromuscular fatigue appears to develop differently, depending on the muscular action modes. Thus, the underlying mechanism of this difference in 1RM Table 4 Mean and SD for baseline (pre-) and posttest and % change (±SD) for functional performance for the three intervention groups. HIGH (n = 18) 6MWD (m) GSmax (ms−1) 30-s STS (reps) 5× STS (s) TUG (s) Pre Post Pre Post Pre Post Pre Post Pre Post Mean SD 595.6 627.7 1.83 2.05 15.8 17.1 9.1 8.1 5.8 5.6 65.8 72.0 0.29 0.30 2.5 2.5 1.7 1.0 0.8 0.7 LOW+ (n = 19) % 5.7 ± 9.7d 12.6 ± 11.9d 8.6 ± 8.0d −10.0 ± 8.5d −2.9 ± 7.7 Mean SD 564.1 588.5 1.82 1.86 15.0 16.3 9.5 8.4 6.3 5.9 92.5 87.2 0.33 0.27 2.1 1.8 1.1 0.9 0.8 0.7 LOW (n = 19) % 5.1 ± 11.4 3.7 ± 12.4 9.1 ± 9.2d −11.5 ± 10.0d −4.9 ± 11.4 Mean SD 553.0 570.2 1.93 1.97 15.7 16.4 9.5 8.3 6.0 5.8 71.3 70.0 0.29 0.39 2.3 2.1 1.8 1.1 1.1 1.2 p-Values % Time Time × group 3.5 ± 9.1 0.001a 0.871a 2.0 ± 13.3 0.001a 0.051a 5.0 ± 9.0 b0.001a 0.239a −10.6 ± 10.0d b0.001b 0.756c 0.003a 0.570a −3.2 ± 8.8 HIGH = high-resistance training; LOW+ = mixed low-resistance training; LOW = low-resistance training; 6MWD = 6-minute walk distance; GSmax = maximal gait speed over 7.5 m; 30-s STS = 30-second chair sit-to-stand test; 5× STS = 5-repetitions chair sit-to-stand test; TUG = timed up-and-go test. a Results of linear mixed models analyses, time × group effect corrected for average training volume. b Results of Friedman test. c Results of Kruskal–Wallis test corrected for average training volume; significance level p b 0.05. d Significant change from pre to post (p b 0.05). E. Van Roie et al. / Experimental Gerontology 48 (2013) 1351–1361 gain remains poorly understood and should be investigated more into detail in further research. What we did find was that differences in 1RM gains between LOW and both HIGH and LOW + were already apparent after 4 weeks of training. Therefore, they were probably due to coordinative and neuromuscular adaptations specific to the movement that was trained. Regarding local muscular endurance, our results are in line with the “strength-endurance continuum” as well (DeLorme, 1945), since significant improvements were only found in LOW and LOW+. Although no significant increase was found in HIGH, linear mixed model analysis did not reveal a time by group interaction effect. However, the training protocols in LOW and LOW+ used resistances that were notably lower than the resistance applied in the local muscular endurance test. In addition to muscle volume, muscle strength, and muscular endurance, strength training regimens in older adults should focus on muscle power and speed of movement. Muscle power, i.e. the product of force and velocity, appears to be a key component in everyday function (Cuoco et al., 2004). A recent study in younger adults, in which a similar protocol was used as LOW+, showed promising results on speed of movement at different resistances, even though training was performed at a moderate speed (Van Roie et al., 2013). More specifically, gains in speed of movement were demonstrated after 9 weeks of mixed lowresistance exercise, whereas no gains were found after high-resistance exercise. The current study, however, failed to confirm these findings in older adults. No increases in speed of movement were found in any of the groups. Two possible explanations may have contributed to this disparity. First, speed of movement during training, although moderate in both studies, was different. The older subjects in this investigation trained at a moderate speed of 2 s for each concentric and 3 s for each eccentric action. As the range of motion on each exercise covers about 70°, concentric actions were performed at about 35°s−1. In the study with young adults, concentric actions were performed in only 1 s, and thus at about 70°s−1, i.e. twice as fast as in the study with older adults. It can be argued that a resistance training protocol using higher speeds is likely to attain greater gains in speed of movement (Fielding et al., 2002; Fleck and Kraemer, 2004; Sayers and Gibson, 2010). However, we chose a training speed that is more commonly used in practice because of its safety and effectiveness (Delmonico et al., 2005; Hortobagyi et al., 2001). It represents a similar speed as the one recommended by the visual feedback system of the training devices used in this study, which is often used in fitness centers (IsoControl, Technogym). Second, literature on age-related changes to central activation is mixed, with many studies indicating no effect of age (Klass et al., 2007; Knight and Kamen, 2001; Lanza et al., 2004; Roos et al., 1999). However, an article by Stevens et al. (2003) indicated that there may be a meaningful deficit in voluntary muscle activation in the knee extensors of older adults. Although further research is needed and no definite conclusions can be drawn, it seems possible that older adults might not be able to activate type II muscle fibers as easily as young adults. Combined with the fact that aging is accompanied by a selective atrophy and denervation of type II muscle fibers (Manini and Clark, 2012), it seems plausible that older adults show less effect on speed-related parameters. Noteworthy is that speed of movement tests were performed relative to the individual isometric strength. Thus, if a subject improved isometric strength, higher resistances were used in the speed of movement test. This method was chosen in order to measure actual gains in velocity, independent of increases in strength. Velocity-related force production can also be measured using an isokinetic testing approach. Although this is a standardized procedure, it does not simulate natural body movements. However, to extend our data on force-velocity characteristics, we included isokinetic strength at high speeds (180°s−1 and 240°s−1) in our testing protocol. A time effect was found, indicating that dynamic peak torque at high speeds increased after training, similarly as in dynamic peak torque at 60°s−1. For increases in dynamic peak torque at 240°s−1, HIGH did appear to be 1359 beneficial, but the underlying mechanism remains unclear. Older adults might experience some difficulty in activating type II muscle fibers, and the use of high resistances might facilitate the activation of these fibers, leading to better performances on high-speed strength tests. Improvements in muscle volume and muscle strength do not automatically result in improved functional performance (Miszko et al., 2003). In line with this concept, we found hardly any significant relationships between gains in muscle volume and strength and changes in functional performance. It should be noted that the older adults in this study were already well-functioning before the start of the intervention. Only one subject was considered ‘mildly frail’ (mPPT-score of 31), while all other subjects could be categorized as ‘not frail’ (mPPTscore N 31) (Brown et al., 2000). Nevertheless, improvements in functional performance were demonstrated after 12 weeks of training. All groups performed better on both sit-to-stand tests after training, with no differences between groups. On the 6-minute walk test and the timed up-and-go test, only HIGH and LOW+ improved, even though not significantly different from LOW. Only for maximal gait speed, the gains in HIGH tended to exceed those in LOW and LOW+. So it seems that strengthening exercise at high- and low-resistances may be similarly effective in improving functional performance. Some limitations of this study need to be considered. First, although our findings suggest that differences in 1RM gains following HIGH and LOW+ compared to LOW might be related to neuromuscular adaptations, assessment of these adaptations was not within the scope of this study. Second, our data lack information on the acute cardiovascular responses of these training protocols, another research area of interest. Third, the findings in this study point to the value of high-repetition low-resistance exercise protocols in older adults. It should be noted, however, that training sessions were closely supervised by a personal trainer, who motivated subjects to continue the exercise until muscle failure. Because of inter-individual differences in strength-endurance capacity, it is difficult to predict the optimal training resistance. Highrepetition low-resistance exercise protocols might therefore ask for more guidance and fine-tuning in the starting phase. Fourth, dietary control and adequate consumption of energy and proteins are especially relevant when studying the older population. Delaying protein intake after a program of resistance exercise training can significantly impact training induced adaptations (Hartman et al., 2007). However, food intake was not monitored during the intervention. An inadequate protein consumption could be part of the explanation for the modest improvements on muscle volume and on static and dynamic peak torques. Fifth, an estimate was used for measuring leg extension 1RM, as this method was considered less time consuming. Using this estimate might have led to an overestimation of the 1RM and can thus be considered a limitation of the study. It can be argued that this estimate for leg extension 1RM seems to recapitulate local muscular endurance measured on the same equipment. However, the average number of repetitions performed during both tests significantly differed (10.7 for 1RM versus 16.7 for endurance; p b 0.001). In addition, no significant correlation was found between leg extension 1RM (kg) and local muscular endurance (number of repetitions) (baseline r = −0.041; post r = 0.150; p N 0.05). Likewise, no significant correlation was found between percent changes from baseline to post for 1RM and local muscular endurance (r = −0.153; p = 0.260). These data seem to suggest that the two leg extension tests measure different aspects of muscular performance. In conclusion, this study showed that strengthening exercise at either high or low external resistance can be similarly effective in increasing muscle volume and basic strength, as long as maximal effort is achieved during training. The unique aspect of the current study was the design of a highly repetitive protocol using even lower resistances than in previous research (LOW) (Hortobagyi et al., 2001; Tsutsumi et al., 1997; Vincent et al., 2002). Moreover, we designed a mixed low-resistance exercise protocol (LOW+) to investigate the added value of intensifying the resistance at the end of such a high- 1360 E. Van Roie et al. / Experimental Gerontology 48 (2013) 1351–1361 repetition low-resistance exercise protocol. Differences that did appear between groups were specific to the trained movement. Confirming previous research, high-resistance training led to a higher increase in 1RM strength than low-resistance training. However, when using a mixed low-resistance protocol in which the resistance was increased after a fatiguing protocol of 60 repetitions, this difference in 1RM gain disappeared. Although the underlying mechanism remains poorly understood, it seems possible that neuromuscular adaptations, specific to the trained movement, are triggered by a mechanical stimulus. This mechanical stimulus can be created either by using high resistances or by increasing resistance after a highly repetitive fatiguing protocol. However, it remains unclear whether differences in muscle activation or in stress/strain occurred between the different protocols. Long-term training studies as well as studies focusing on residual effects after training cessation are needed to confirm the current findings and to further clarify differences between these training approaches. Further research should additionally investigate the underlying mechanisms accounting for variations in the effectiveness of high- and low-resistance training protocols. Conflict of interest The authors have no conflicts of interests. Acknowledgments We would like to acknowledge the Research Foundation Flanders, Belgium (FWO-Vlaanderen), as E. Van Roie is a Ph.D. Fellow of FWO. Also, the authors are very grateful to Sports and Health Center Respiro (Rotselaar, Belgium) for the use of the training equipment. References Abellan van Kan, G., 2009. Epidemiology and consequences of sarcopenia. J. Nutr. Health Aging 13, 708–712. Abellan van Kan, G., Andre, E., Bischoff Ferrari, H.A., Boirie, Y., Onder, G., Pahor, M., Ritz, P., Rolland, Y., Sampaio, C., Studenski, S., Visser, M., Vellas, B., 2009. Carla task force on sarcopenia: propositions for clinical trials. J. Nutr. Health Aging 13, 700–707. American College of Sports Medicine, 2006. ACSM's Guidelines for Exercise Testing and Prescription, 7th ed. Lippincott Williams & Wilkins, Philadelphia, Pa. American Thoracic Society, 2002. ATS statement: guidelines for the six-minute walk test. Am. J. Respir. Crit. Care Med. 166, 111–117. Anderson, T., Kearney, J.T., 1982. Effects of three resistance training programs on muscular strength and absolute and relative endurance. Res. Q. Exerc. Sport 53, 1–7. Babault, N., Desbrosses, K., Fabre, M.S., Michaut, A., Pousson, M., 2006. Neuromuscular fatigue development during maximal concentric and isometric knee extensions. J. Appl. Physiol. 100, 780–785. Bischoff, H.A., Stahelin, H.B., Monsch, A.U., Iversen, M.D., Weyh, A., von Dechend, M., Akos, R., Conzelmann, M., Dick, W., Theiler, R., 2003. Identifying a cut-off point for normal mobility: a comparison of the timed ‘up and go’ test in community-dwelling and institutionalised elderly women. Age Ageing 32, 315–320. Brown, M., Sinacore, D.R., Binder, E.F., Kohrt, W.M., 2000. Physical and performance measures for the identification of mild to moderate frailty. J. Gerontol. A Biol. Sci. Med. Sci. 55, M350–M355. Campos, G.E., Luecke, T.J., Wendeln, H.K., Toma, K., Hagerman, F.C., Murray, T.F., Ragg, K.E., Ratamess, N.A., Kraemer, W.J., Staron, R.S., 2002. Muscular adaptations in response to three different resistance-training regimens: specificity of repetition maximum training zones. Eur. J. Appl. Physiol. 88, 50–60. Cannon, J., Marino, F.E., 2010. Early-phase neuromuscular adaptations to high- and low-volume resistance training in untrained young and older women. J. Sports Sci. 1–10. Carpinelli, R.N., 2008. The size principle and a critical analysis of the unsubstantiated heavier-is-better recommendation for resistance training. J. Exerc. Sci. Fit. 6, 67–86. Cuoco, A., Callahan, D.M., Sayers, S., Frontera, W.R., Bean, J., Fielding, R.A., 2004. Impact of muscle power and force on gait speed in disabled older men and women. J. Gerontol. A Biol. Sci. Med. Sci. 59, 1200–1206. Delmonico, M.J., Kostek, M.C., Doldo, N.A., Hand, B.D., Bailey, J.A., Rabon-Stith, K.M., Conway, J.M., Carignan, C.R., Lang, J., Hurley, B.F., 2005. Effects of moderate-velocity strength training on peak muscle power and movement velocity: do women respond differently than men? J. Appl. Physiol. 99, 1712–1718. DeLorme, T.L., 1945. Restoration of muscle power by heavy resistance exercise. J. Bone Joint Surg. Br. 645–667. Fielding, R.A., LeBrasseur, N.K., Cuoco, A., Bean, J., Mizer, K., Fiatarone Singh, M.A., 2002. High-velocity resistance training increases skeletal muscle peak power in older women. J. Am. Geriatr. Soc. 50, 655–662. Fleck, S.J., Kraemer, W.J., 2004. Designing Resistance Training Programs, 3rd ed. Human Kinetics, Champaign, IL. Goto, K., Nagasawa, M., Yanagisawa, O., Kizuka, T., Ishii, N., Takamatsu, K., 2004. Muscular adaptations to combinations of high- and low-intensity resistance exercises. J. Strength Cond. Res. 18, 730–737. Goto, K., Ishii, N., Kizuka, T., Takamatsu, K., 2005. The impact of metabolic stress on hormonal responses and muscular adaptations. Med. Sci. Sports Exerc. 37, 955–963. Hartman, J.W., Tang, J.E., Wilkinson, S.B., Tarnopolsky, M.A., Lawrence, R.L., Fullerton, A.V., Phillips, S.M., 2007. Consumption of fat-free fluid milk after resistance exercise promotes greater lean mass accretion than does consumption of soy or carbohydrate in young, novice, male weightlifters. Am. J. Clin. Nutr. 86, 373–381. Henneman, E., 1957. Relation between size of neurons and their susceptibility to discharge. Science 126, 1345–1347. Holm, L., Reitelseder, S., Pedersen, T.G., Doessing, S., Petersen, S.G., Flyvbjerg, A., Andersen, J.L., Aagaard, P., Kjaer, M., 2008. Changes in muscle size and MHC composition in response to resistance exercise with heavy and light loading intensity. J. Appl. Physiol. 105, 1454–1461. Hortobagyi, T., Tunnel, D., Moody, J., Beam, S., DeVita, P., 2001. Low- or high-intensity strength training partially restores impaired quadriceps force accuracy and steadiness in aged adults. J. Gerontol. A Biol. Sci. Med. Sci. 56, B38–B47. Karabulut, M., Abe, T., Sato, Y., Bemben, M.G., 2010. The effects of low-intensity resistance training with vascular restriction on leg muscle strength in older men. Eur. J. Appl. Physiol. 108, 147–155. Kay, D., St Clair, G.A., Mitchell, M.J., Lambert, M.I., Noakes, T.D., 2000. Different neuromuscular recruitment patterns during eccentric, concentric and isometric contractions. J. Electromyogr. Kinesiol. 10, 425–431. Klass, M., Baudry, S., Duchateau, J., 2007. Voluntary activation during maximal contraction with advancing age: a brief review. Eur. J. Appl. Physiol. 100, 543–551. Knight, C.A., Kamen, G., 2001. Adaptations in muscular activation of the knee extensor muscles with strength training in young and older adults. J. Electromyogr. Kinesiol. 11, 405–412. Krieger, J.W., 2010. Single vs. multiple sets of resistance exercise for muscle hypertrophy: a meta-analysis. J. Strength Cond. Res. 24, 1150–1159. Lanza, I.R., Russ, D.W., Kent-Braun, J.A., 2004. Age-related enhancement of fatigue resistance is evident in men during both isometric and dynamic tasks. J. Appl. Physiol. 97, 967–975. Lemmer, J.T., Hurlbut, D.E., Martel, G.F., Tracy, B.L., Ivey, F.M., Metter, E.J., Fozard, J.L., Fleg, J.L., Hurley, B.F., 2000. Age and gender responses to strength training and detraining. Med. Sci. Sports Exerc. 32, 1505–1512. Loenneke, J.P., Wilson, J.M., Marin, P.J., Zourdos, M.C., Bemben, M.G., 2012. Low intensity blood flow restriction training: a meta-analysis. Eur. J. Appl. Physiol. 112, 1849–1859. Manini, T.M., Clark, B.C., 2012. Dynapenia and aging: an update. J. Gerontol. A Biol. Sci. Med. Sci. 67 (1), 28–40. Marcell, T.J., 2003. Sarcopenia: causes, consequences, and preventions. J. Gerontol. A Biol. Sci. Med. Sci. 58, M911–M916. Marx, J.O., Ratamess, N.A., Nindl, B.C., Gotshalk, L.A., Volek, J.S., Dohi, K., Bush, J.A., Gomez, A.L., Mazzetti, S.A., Fleck, S.J., Hakkinen, K., Newton, R.U., Kraemer, W.J., 2001. Lowvolume circuit versus high-volume periodized resistance training in women. Med. Sci. Sports Exerc. 33, 635–643. McCarthy, E.K., Horvat, M.A., Holtsberg, P.A., Wisenbaker, J.M., 2004. Repeated chair stands as a measure of lower limb strength in sexagenarian women. J. Gerontol. A Biol. Sci Med. Sci. 59, 1207–1212. Miszko, T.A., Cress, M.E., Slade, J.M., Covey, C.J., Agrawal, S.K., Doerr, C.E., 2003. Effect of strength and power training on physical function in community-dwelling older adults. J. Gerontol. A Biol. Sci. Med. Sci. 58, 171–175. Mitchell, C.J., Churchward-Venne, T.A., West, D.W., Burd, N.A., Breen, L., Baker, S.K., Phillips, S.M., 2012. Resistance exercise load does not determine training-mediated hypertrophic gains in young men. J. Appl. Physiol. 113, 71–77. Neyroud, D., Maffiuletti, N.A., Kayser, B., Place, N., 2012. Mechanisms of fatigue and task failure induced by sustained submaximal contractions. Med. Sci. Sports Exerc. 44, 1243–1251. Podsiadlo, D., Richardson, S., 1991. The timed “Up & Go”: a test of basic functional mobility for frail elderly persons. J. Am. Geriatr. Soc. 39, 142–148. Robertson, R.J., Goss, F.L., Rutkowski, J., Lenz, B., Dixon, C., Timmer, J., Frazee, K., Dube, J., Andreacci, J., 2003. Concurrent validation of the OMNI perceived exertion scale for resistance exercise. Med. Sci. Sports Exerc. 35, 333–341. Rooney, K.J., Herbert, R.D., Balnave, R.J., 1994. Fatigue contributes to the strength training stimulus. Med. Sci. Sports Exerc. 26, 1160–1164. Roos, M.R., Rice, C.L., Connelly, D.M., Vandervoort, A.A., 1999. Quadriceps muscle strength, contractile properties, and motor unit firing rates in young and old men. Muscle Nerve 22, 1094–1103. Sayers, S.P., Gibson, K., 2010. A comparison of high-speed power training and traditional slow-speed resistance training in older men and women. J. Strength Cond. Res. 24, 3369–3380. Sayers, S.P., Guralnik, J.M., Thombs, L.A., Fielding, R.A., 2005. Effect of leg muscle contraction velocity on functional performance in older men and women. J. Am. Geriatr. Soc. 53, 467–471. Schott, J., McCully, K., Rutherford, O.M., 1995. The role of metabolites in strength training. II. Short versus long isometric contractions. Eur. J. Appl. Physiol. Occup. Physiol. 71, 337–341. Slivka, D., Raue, U., Hollon, C., Minchev, K., Trappe, S., 2008. Single muscle fiber adaptations to resistance training in old (N80 yr) men: evidence for limited skeletal muscle plasticity. Am. J. Physiol. Regul. Integr. Comp. Physiol. 295, R273–R280. Stevens, J.E., Stackhouse, S.K., Binder-Macleod, S.A., Snyder-Mackler, L., 2003. Are voluntary muscle activation deficits in older adults meaningful? Muscle Nerve 27, 99–101. Taaffe, D.R., Pruitt, L., Pyka, G., Guido, D., Marcus, R., 1996. Comparative effects of highand low-intensity resistance training on thigh muscle strength, fiber area, and tissue composition in elderly women. Clin. Physiol. 16, 381–392. E. Van Roie et al. / Experimental Gerontology 48 (2013) 1351–1361 Takarada, Y., Ishii, N., 2002. Effects of low-intensity resistance exercise with short interset rest period on muscular function in middle-aged women. J. Strength Cond. Res. 16, 123–128. Takarada, Y., Takazawa, H., Sato, Y., Takebayashi, S., Tanaka, Y., Ishii, N., 2000. Effects of resistance exercise combined with moderate vascular occlusion on muscular function in humans. J. Appl. Physiol. 88, 2097–2106. Tan, B., 1999. Manipulating resistance training program variables to optimize maximum strength in men: a review. J. Strength Cond. Res. 13, 289–304. Tanimoto, M., Ishii, N., 2006. Effects of low-intensity resistance exercise with slow movement and tonic force generation on muscular function in young men. J. Appl. Physiol. 100, 1150–1157. Tsutsumi, T., Don, B.M., Zaichkowsky, L.D., Delizonna, L.L., 1997. Physical fitness and psychological benefits of strength training in community dwelling older adults. Appl. Human Sci. 16, 257–266. 1361 Van Roie, E., Verschueren, S.M., Boonen, S., Bogaerts, A., Kennis, E., Coudyzer, W., Delecluse, C., 2011. Force-velocity characteristics of the knee extensors: an indication of the risk for physical frailty in elderly women. Arch. Phys. Med. Rehabil. 92, 1827–1832. Van Roie, E., Bautmans, I., Boonen, S., Coudyzer, W., Kennis, E., Delecluse, C., 2013. Impact of external resistance and maximal effort on force–velocity characteristics of the knee extensors during strengthening exercise: a randomized controlled experiment. J. Strength Cond. Res. 27, 1118–1127. Vincent, K.R., Braith, R.W., Feldman, R.A., Magyari, P.M., Cutler, R.B., Persin, S.A., Lennon, S.L., Gabr, A.H., Lowenthal, D.T., 2002. Resistance exercise and physical performance in adults aged 60 to 83. J. Am. Geriatr. Soc. 50, 1100–1107. Yoon, T., Schlinder, D.B., Griffith, E.E., Hunter, S.K., 2007. Mechanisms of fatigue differ after low- and high-force fatiguing contractions in men and women. Muscle Nerve 36, 515–524.