Effects of velocity loss during body mass prone-grip pull-up training on strength and endurance performance

EFFECTS OF VELOCITY LOSS DURING BODY MASS PRONE PULL-UP TRAINING ON STRENGTH AND ENDURANCE PERFORMANCE Running title: Effects of velocity loss during pull-up training Physical Performance and Sports Research Center, Universidad Pablo de Olavide, Seville, Spain Miguel Sánchez-Moreno1,2, Pedro Jesús Cornejo-Daza1, Juan José González-Badillo1, Fernando Pareja-Blanco1 1. Physical Performance and Sports Research Center, Pablo de Olavide University, Seville, Spain. 2. Department of Physical Education and Sport, University of Sevilla, Sevilla, Spain. Adress for correspondence: Miguel Sánchez-Moreno Department of Physical Education and Sport, University of Sevilla, C/Pirotecnia s/n, C.P. 41013, Seville, Spain. Email: [email protected] Tel + 34 651 9171 516, Fax: +34 954 348 659 EFFECTS OF VELOCITY LOSS DURING BODY MASS PRONE PULL-UP 1 TRAINING ON STRENGTH AND ENDURANCE PERFORMANCE 2 ABSTRACT 3 This study aimed to analyze the effects of two pull-up (PU) training programs that 4 differed in the magnitude of repetition velocity loss allowed in each set (25% velocity 5 loss “VL25” vs. 50% velocity loss “VL50”) on PU performance. Twenty-nine nine 6 strength-trained men (age = 26.1 ± 6.3 years, body mass = 74.2 ± 6.4 kg, 15.9 ± 4.9 PU 7 repetitions to failure) were randomly assigned to two groups: VL25 (n = 15) or VL50 (n 8 = 14) and followed an 8-week (16 sessions) velocity-based body mass (BM) prone PU 9 training program. Mean propulsive velocity (MPV) was monitored in all repetitions. 10 Assessments performed at Pre-training and Post-training included: estimated one11 repetition maximum (1RM); average MPV attained with all common external loads used 12 during Pre-training and Post-training testing (AVinc); peak MPV lifting one’s own BM 13 (MPVbest); maximum number of repetitions to failure lifting one’s own BM (MNR); and 14 average MPV corresponding to the same number of repetitions lifting one’s own BM 15 performed during Pre-training testing (AVMNR). VL25 attained significantly greater gains 16 than VL50 in all analyzed variables except in MNR. Additionally, VL25 improved 17 significantly (P<0.001) in all the evaluated variables while VL50 remained unchanged. 18 In conclusion, our results suggest that once a 25% velocity loss is achieved during PU 19 training, a further increase does not elicit further gains and can even blunt the 20 improvement in strength and endurance performance. 21 Keywords: velocity-based resistance training, training volume, movement velocity, 22 athletic performance, strength training 23 2 INTRODUCTION 24 Controlling and monitoring the training load undertaken by athletes during resistance 25 training (RT) is a complex process for strength and conditioning coaches. The interaction 26 between training intensity and volume produces what is termed a ‘level of effort’, which 27 is defined as the relationship between the repetitions completed in a set and those that 28 could potentially be performed (23). The indicators that have traditionally been used as 29 references for quantifying the RT load (one-repetition maximum, “1RM” and maximum 30 number of repetitions, “MNR” tests) present potential limitations, such as daily changes 31 in the actual 1RM. Therefore, it cannot be guaranteed that the relative loads (%1RM) 32 employed in each particular training session truly represent the scheduled ones. Another 33 limitation is that the MNR that can be performed with a given %1RM shows a great 34 variability between individuals (8,22). Hence, a given MNR does not necessarily 35 represent the same %1RM for every participant. 36 Velocity monitoring may provide a better quantification of the level of effort involved 37 during RT, together with a better monitoring of training effects (7,19,23). The validity of 38 the velocity-based training approach (VBT) is based on: i) the strong relationship 39 observed between movement velocity and %1RM in different exercises (7,15,24,25,28), 40 and ii) the relationship between the velocity loss induced in each set and the percentage 41 of repetitions actually performed in each set with respect to those that could be completed 42 (8,23). Hence, the velocity loss achieved in the set provides very accurate information 43 about the level of effort incurred in a set, in terms of the percentage of repetitions actually 44 performed with regard to the MNR (8). 45 The pull-up (PU) is a multi-joint upper-body exercise, which is considered a valid 46 measure of weight-related muscular strength (21,27). This exercise is commonly used in 47 sport disciplines that require upper-body pulling strength, such as canoeing (4), climbing 48 (9) and kayaking (16). Furthermore, it has traditionally been used as a physical fitness 49 testing tool to assess upper-body strength and endurance in a variety of populations 50 including the military, firefighters, and police officers (2). The PU performance is 51 generally scored by the MNR completed until muscular failure lifting subject’s own body 52 mass (BM), or by the value of 1RM. 53 One of the most popular practices for training in PU exercise is to perform repetitions 54 until muscular failure using one’s own BM. However, a recent meta-analysis 55 3 demonstrated that similar increases in muscular strength can be achieved with failure and 56 non-failure RT (3). To our knowledge, no study has analyzed the effect of different PU 57 training programs on 1RM and MNR in this exercise. A recent paper reported a close 58 relationship (r = -0.96) between relative load and movement velocity in PU, together with 59 a strong relationship (R2 = 0.88) between the velocity loss induced in a set and the 60 percentage of MNR performed (28). These findings allow us to estimate the percentage 61 of MNR that has been completed as soon as a given percentage of velocity loss is detected 62 during a PU set. A velocity loss of 25% in a PU set means that an individual has 63 completed ∼50% of the MNR, whereas a velocity loss of 50% corresponds to ∼85% of 64 the MNR, regardless of the total number of repetitions to failure that could be completed 65 (28). Pareja-Blanco et al. (19) compared the effects of two squat training programs that 66 differed in the velocity loss reached in each set: 20% vs. 40%. A velocity loss of 20% 67 (which corresponded to performing approximately 50% of MNR in squat exercise) 68 resulted in similar or even superior strength gains to a 40% velocity loss (close to muscle 69 failure in this exercise). However, to our knowledge, no previous study has analyzed the 70 effect of different velocity loss magnitudes on upper body exercises. Therefore, it is still 71 unknown whether it is possible to extrapolate findings from VBT training studies carried 72 out in lower body exercises to upper body exercises. Thus, in an attempt to gain further 73 insight into the adaptations brought about by different velocity losses during the set in 74 upper body exercises, we aimed to compare the effects of two PU training programs that 75 differed in the magnitude of repetition velocity loss allowed in each set (25% vs. 50%). 76 METHODS 77 Experimental Approach to the Problem 78 Subjects trained twice per week (72-96 h apart) over an 8-week period for a total of 16 79 sessions. The training program used only the prone PU exercise (Table 1). The two 80 groups trained with their own BM (without external loads) in each session but differed in 81 the maximum percent velocity loss reached in each set (25% vs. 50%). As soon as the 82 corresponding target velocity loss limit was exceeded, the set was terminated. Sessions 83 were performed in a research laboratory under the direct supervision of the investigators, 84 at the same time of day (±1 h) for each subject and under controlled environmental 85 conditions (20ºC and 65% humidity). Both groups were assessed on two occasions: before 86 and after the 8-week training intervention. Pre-training and Post-training testing sessions 87 4 took place in one session which comprised the PU loading tests up to 1RM and the 88 maximum number of repetitions to failure (MNR test) without added weight (performed 89 in that order, separated by a 5 min rest, and described later in detail). Any upper body pull 90 exercises were removed from the usual strength training during the experimental period 91 to avoid any additive effect caused by this type of exercise. 92 Subjects 93 Thirty-four strength-trained men (mean ± SD: age = 26.5 ± 6.3 years, BM = 74.3 ± 6.1 94 kg, height = 176.1 ± 5.3 cm) volunteered to take part in this study. Subjects had a training 95 background in PU exercise ranging from 2 to 4 years (2-3 sessions per week; 15.9 ± 4.9 96 PU repetitions to failure with BM). Subjects were randomly assigned to one of two 97 groups, which differed only in the magnitude of repetition velocity loss allowed in each 98 training set: 25% (VL25; n = 17) or 50% (VL50; n = 17). Only those subjects who 99 complied with at least 95% of all training sessions were included in the statistical 100 analyses. Five subjects withdrew from the study during the 8-week training period, one 101 of them due to injury and the rest because they missed training sessions. Thus, of the 34 102 initially enrolled subjects, twenty-nine subjects remained for statistical analysis (VL25, n 103 = 15, age = 26.7 ± 5.5 years, BM = 74.1 ± 4.7 kg, height = 175.8 ± 6.0 cm vs. VL50, n = 104 14, age = 24.8 ± 6.1 years, BM = 74.3 ± 8.1 kg, height = 176.1 ± 5.0 cm). Once informed 105 about the purpose, testing procedures and potential risks of the investigation, all subjects 106 gave their voluntary written consent to participate. The present investigation was 107 approved by the Research Ethics Committee of Pablo de Olavide University, and was 108 conducted in accordance with the Declaration of Helsinki. 109 Testing Procedures 110 All PU tests were performed on a standard stationary, horizontal bar (28 mm diameter). 111 To be counted as a complete PU repetition, the subject lifted had to lift his body from a 112 full-arm extension hanging position until his chin was above the bar. A self-selected width 113 with pronated grip (approximately 150% of the biacromial distance) was used throughout 114 the first testing session and this was recorded so that it could be repeated in the subsequent 115 testing sessions. During each repetition of both tests (progressive loading and MNR) and 116 all training sessions, the subjects were required to perform the eccentric phase in a 117 controlled manner and maintain a static position for ∼1 s at the end of this phase before 118 performing the concentric phase at maximal intended velocity upon hearing the 119 5 command. In addition, at the end of the eccentric phase, any possible horizontal 120 movements caused by this phase were eliminated by the researchers holding the subjects 121 by the ankles. All repetitions were recorded using a linear velocity transducer (T-Force 122 System, Ergotech, Murcia, Spain). This device has been found to be reliable (23). All 123 reported repetition velocities in this study corresponded to the mean propulsive velocity 124 (MPV) (26). The same warm-up protocol, which consisted of 5 minutes of jogging at a 125 self-selected easy pace, 5 minutes of joint mobilization exercises and one set of 3 PU 126 repetitions with no external load, was followed in all testing sessions. Strong verbal 127 encouragement was provided during all tests to motivate subjects to give maximal effort. 128 Progressive loading test 129 Individual load–velocity relationships and 1RM strength were determined using a 130 progressive loading test. The test-retest reliability of this relationship in the PU exercise 131 has been previously established (28). Subjects started without additional weight and the 132 load was gradually increased, initially in 5 kg increments until the attained MPV was 133 lower than 0.30 m·s-1, which represents at least 95% 1RM, so that 1RM could be 134 determined (28). Because subjects needed to lift their BM, 1RM was calculated as the 135 sum of the maximum weight lifted and the subject’s BM. Three repetitions were executed 136 when the MPV was higher than 0.75 m·s−1, two when the MPV was between 0.75 and 137 0.55 m·s−1, and only one when the MPV was less than 0.55 m·s−1. Inter-set rests were 3 138 min when the MPV was higher or equal than 0.55 m·s-1 and 4 min when the MPV was 139 less than 0.55 m·s−1. This resulted in a total of 6.5 ± 2.7 increasing loads performed by 140 each subject. Only the best repetition (fastest and executed correctly) at each load was 141 considered for subsequent analysis. To add additional weight, a specialized belt was used 142 which could be adjusted around the waist and allowed weights to be attached via a chain. 143 The cable from the linear velocity transducer was fixed to the back of the belt. The 144 following variables derived from this progressive loading test were used for analysis: a) 145 estimated 1RM value, which was calculated from the MPV attained against the heaviest 146 load of the test (>95%1RM), as follows: %1RM = -53.472 · MPV + 110.68 (R = -0.96; 147 SEE = 3.2% 1RM) (28); b) average MPV attained against all absolute loads common to 148 Pre and Post-tests (AVinc); and c) fastest MPV attained without additional weight 149 (MPVbest). The AVinc value was used in an attempt to analyze the extent to which the two 150 training interventions affected the PU load-velocity relationship (20). 151 6 Maximum number of repetitions test 152 During this test, subjects were required to complete the maximum number of repetitions 153 until muscular failure, lifting their own BM from a full-arm extension hanging position 154 (using the same width pronated grip and execution as in the progressive loading test) until 155 the chin was above the bar. The test was considered complete when the subject was not 156 able to raise the chin above the bar or when the subjects paused more than 2-3 s in the 157 extended position. Test-retest reliability has been previously reported elsewhere (29). The 158 following variables derived from this test were used for analysis: a) maximal number of 159 repetitions to failure (MNR); and b) average MPV attained against the same number of 160 repetitions to Pre-training and Post-training (AVMNR). This enabled assessment of the 161 changes in MPV corresponding to the MNR at Pre-training. 162 Resistance training program 163 The descriptive characteristics of the training program are presented in Table 1. Both 164 groups trained using only the BM prone PU exercise (no external load). The technical 165 execution was identical to that previously described in the Testing Procedures section. 166 The number of sets (progressed from 2 to 4) and inter-set recovery periods (3 min) were 167 kept identical for both groups in each training session. Instead of fixing a number of 168 repetitions before beginning the program, we set a target fatigue level (velocity loss). 169 Therefore, the groups differed in the degree of fatigue experienced during the exercise 170 sets, which was objectively quantified by the magnitude of velocity loss attained in each 171 set (25% vs. 50%) and, consequently, differed in the number of repetitions performed per 172 set (Table 1) and the total number of repetitions completed during the training program 173 (Fig. 1). During training, subjects received immediate velocity feedback from the 174 measurement system while being encouraged to perform each repetition at maximal 175 intended velocity. 176 ***Table 1 about here*** 177 Statistical analyses 178 Values are reported as mean ± standard deviation (SD). The normality of distribution of 179 the variables and the homogeneity of variance across groups were verified using the 180 Shapiro-Wilk test and Levene’s test, respectively. Data were analyzed using a repeated 181 measures ANCOVA (with baseline values as covariate) analysis with a Bonferroni post hoc 182 7 adjustment.. Additionally, ES were calculated using Hedge’s g on the pooled SD (10). 183 Probabilities were also calculated to establish whether the true (unknown) differences 184 were lower, similar or higher than the smallest worthwhile difference or change (0.2 x 185 between-subject SD) (Cohen, 1988). Quantitative chances of better or worse effects were 186 assessed qualitatively as follows: <1%, almost certainly not; 1-5%, very unlikely; 5-25%, 187 unlikely; 25-75%, possible; 75-95%, likely; 95-99%, very likely; and >99%, almost 188 certain. If the chances of obtaining beneficial/better or detrimental/worse were both >5%, 189 the true difference was assessed as unclear (1,12). Inferential statistics based on the 190 interpretation of magnitude of effects were calculated using a purpose-built spreadsheet 191 for the analysis of controlled trials (11). The rest of the statistical analyses were performed 192 using SPSS software version 18.0 (SPSS Inc., Chicago, IL). 193 RESULTS 194 No significant differences between groups were found at Pre for any of the variables 195 analyzed. The %1RM that represented participants’ BM at Pre did not differ between 196 groups (69.2 ± 7.6 vs. 66.3 ± 10.5 %1RM, for VL25 and VL50, respectively). No 197 significant changes were observed in BM for any group. The repetitions performed in 198 different velocity ranges by each group are shown in Fig. 1. The VL25 group trained at a 199 significantly faster mean velocity than the VL50 group (0.71 ± 0.11 vs. 0.56 ± 0.13 m·s200 1, respectively; P < 0.001), whereas VL50 performed more repetitions (P < 0.001) than 201 VL25 (556.3 ± 121.9 vs. 363.0 ± 84.6 repetitions). Furthermore, VL50 completed 202 significantly (P < 0.001) more repetitions at slow and moderate velocities (<0.6 m·s-1) 203 than VL25 (Fig. 1). The actual mean velocity loss of the entire training program (i.e. for 204 all sessions and all sets combined) was 26.3 ± 4.1% for VL25 vs. 50.5 ± 7.9% for VL50. 205 ***Figure 1 about here*** 206 Progressive loading test 207 Significant ‘group’ x ‘time’ interactions were observed for 1RM, AVinc and MPVbest 208 (Table 2). Significant differences between groups were observed in these 3 variables at 209 Post-training (Table 2). The VL50 group did not attain significant improvements in any 210 of these variables, whereas VL25 improved (P < 0.001) in 1RM, AVinc and MPVbest 211 (Table 2). Additionally, the VL25 group showed greater ESs for 1RM, AVinc and MPVbest 212 than VL50(Fig. 2). 213 8 ***Table 2 about here*** 214 ***Figure 2 about here*** 215 Test of maximum number of repetitions to failure 216 A significant ‘group’ x ‘time’ interaction was observed for AVMNR (Table 2). Only the 217 VL25 group attained significant increases both in MNR and AVMNR, whereas the VL50 218 group did not show significant improvements in any of these variables (Table 2). In 219 addition, VL25 showed greater ES compared to VL50 group on MNR and AVMNR (Fig. 220 2). 221 DISCUSSION 222 The main finding of this study was that training with a velocity loss of 25% (VL25) in 223 each set induced greater gains in strength (1RM as well as the velocity attained against 224 all loads) and muscular endurance performance (MNR as well as the velocity attained 225 against the same number of repetitions) than training with a velocity loss of 50% (VL50). 226 These results were observed despite the fact that the VL50 group performed significantly 227 more repetitions than VL25 (556 vs. 363 repetitions) during the training program. 228 Although both groups performed a similar number of repetitions at very high (>0.9 m·s229 1) and moderate velocities (0.6-0.7 m·s-1), VL25 completed significantly more repetitions 230 at high velocities (from 0.7-0.9 m·s-1) whereas VL50 completed significantly more 231 repetitions at slow velocities (0.6-0.3 m·s-1) (Fig. 1). These training programs resulted in 232 better strength and endurance adaptations in VL25 compared to VL50 over the 8-week 233 program. Therefore, setting a certain percentage of velocity loss during the training 234 program seems a plausible way to avoid performing unnecessarily slow and fatiguing 235 repetitions that may not contribute to the desired PU training effect. 236 The present findings also support previous studies that suggested the existence of an 237 inverted U-shaped relationship between training volume and performance increase 238 (5,6,14). In this regard, Pareja-Blanco et al. (19) observed that eight weeks of RT in squat 239 exercise with a velocity loss of 20% (which corresponded to performing approximately 240 50% of MNR) resulted in similar gains in performance compared to a velocity loss of 241 40% (close to muscle failure in this exercise), and even greater gains in high velocity 242 actions such as vertical jumps. In another previous study, a professional soccer team was 243 divided into two groups: one trained at a velocity loss of 15% and the other trained at a 244 Table 2. Changes in selected performance variables from preto post-training for each group Pre Post ES (90% CI) Percent changes of better/trivial/worse effect BM-VL25 (kg) 74.1 ± 4.7 74.1 ± 5.2 0.00 (-0.10 to 0.10) 0/100/0 Most Likely Trivial BM-VL50 (kg) 74.3 ± 8.1 73.8 ± 8.0 -0.05 (-0.11 to 0.01) 0/100/0 Most Likely Trivial 1RM-VL25 (kg) † 108.4 ± 10.4 114.3 ± 8.9*** Ұ 0.54 (0.33 to 0.75) 99/1/0 Very Likely + 1RM-VL50 (kg) 114.4 ± 20.8 115.2 ± 19.8 0.04 (-0.07 to 0.15) 1/99/0 Very Likely Trivial AVinc-VL25 (m·s-1) † 0.54 ± 0.07 0.63 ± 0.08*** Ұ 1.24 (0.79 to 1.69) 100/0/0 Most Likely + AVinc-VL50 (m·s-1) 0.57 ± 0.10 0.59 ± 0.08 0.20 (-0.12 to 0.51) 50/48/2 Possibly + MPVbest-VL25 (m·s-1) † 0.78 ± 0.14 0.89 ± 0.14*** Ұ 0.77 (0.52 to 1.02) 100/0/0 Most Likely + MPVbest-VL50 (m·s-1) 0.83 ± 0.20 0.84 ± 0.16 -0.05 (-0.23 to 0.33) 19/75/6 Unclear MNR-VL25 (rep) 15.6 ± 5.0 17.9 ± 3.9*** 0.43 (0.23 to 0.64) 97/3/0 Very Likely + MNR-VL50 (rep) 16.1 ± 5.0 17.1 ± 4.4 0.18 (-0.01 to 0.36) 41/59/0 Possibly Trivial AVMNR-VL25 (m·s-1) † 0.52 ± 0.08 0.63 ± 0.11*** 1.17 (0.59 to 1.76) 99/1/0 Very Likely + AVMNR-VL50 (m·s-1) 0.57 ± 0.11 0.58 ± 0.09 0.10 (-0.25 to 0.45) 31/61/8 Unclear Data are mean ± SD; ES = Effect Size within-group; CI = Confidence Interval. VL25: group that trained with a mean velocity loss of 25% in each set (n = 15); VL50: group that trained with a mean velocity loss of 50% in each set (n = 14); BM: body mass; 1RM: estimated one-repetition maximum pull up strength; AVinc: average MPV attained against absolute loads common to preand post-test in the pull up progressive loading test; MPVbest: fastest MPV attained without additional weight in the pull up progressive loading test; MNR: maximal number of repetitions to failure in the pull up exercise without additional weight; AVMNR: average MPV attained against the same number of repetitions to preand post-test in the pull up maximal number of repetitions test. Significant group x time interaction: † P < 0.05. Between-groups significant differences at Post-training Ұ P < 0.05. Intra-group significant differences from Preto Post-training: *** P < 0.001.