Evidence of muscular adaptations within four weeks of barbell training in women

Human Movement Science 45 (2016) 7–22

Contents lists available at ScienceDirect

Human Movement Science journal homepage: www.elsevier.com/locate/humov

Evidence of muscular adaptations within four weeks of barbell training in women Matt S. Stock a,⇑, Kendra D. Olinghouse a, Alexander S. Drusch b, Jacob A. Mota a, Jennah M. Hernandez a, Chibuzo C. Akalonu a, Brennan J. Thompson c a

Human Performance Laboratory, Texas Tech University, Lubbock, TX, USA Center for Rehabilitation Research, Texas Tech University Health Science Center, Lubbock, TX, USA c Neuromuscular Research Laboratory, Utah State University, Logan, UT, USA b

a r t i c l e

i n f o

Article history: Received 1 June 2015 Revised 3 November 2015 Accepted 4 November 2015

Keywords: Motor unit Resistance training Vastus lateralis

a b s t r a c t We investigated the time course of neuromuscular and hypertrophic adaptations associated with only four weeks of barbell squat and deadlift training. Forty-seven previously untrained women (mean ± SD, age = 21 ± 3 years) were randomly assigned to low volume training (n = 15), moderate volume training (n = 16), and control (n = 16) groups. The low and moderate volume training groups performed two and four sets, respectively, of five repetitions per exercise, twice a week. Testing was performed weekly, and included dual X-ray absorptiometry and vastus lateralis and rectus femoris B-mode ultrasonography. Bipolar surface electromyographic (EMG) signals were detected from the vastus lateralis and biceps femoris during isometric maximal voluntary contractions of the leg extensors. Significant increases in lean mass for the combined gynoid and leg regions for the low (+0.68 kg) and moderate volume (+0.47 kg) groups were demonstrated within three weeks. Small-to-moderate effect sizes were shown for leg lean mass, vastus lateralis thickness and pennation angle, and peak torque, but EMG amplitude was unaffected. These findings demonstrated rapid muscular adaptations in response to only eight sessions of back squat and deadlift training in women despite the absence of changes in agonist–antagonist EMG amplitude. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction Historically, investigators have considered skeletal muscle growth to be a slow process. Publications endorsed by both the National Strength and Conditioning Association (Baechle & Earle, 2008) and the American College of Sports Medicine (2009) state that the initial adaptations associated with resistance training in novices are controlled by neuromuscular factors. Neuromuscular adaptations associated with short-term training interventions have been relatively well documented (Aagaard, Simonsen, Andersen, Magnusson, & Dyhre-Poulsen, 2002; Carolan & Cafarelli, 1992; Moritani & deVries, 1979; Stock & Thompson, 2014; Vila-Cha, Falla, & Farina, 2010). Perhaps the most influential investigation was carried out by Moritani and deVries (1979), who were the first to examine the time course associated with neuromuscular and hypertrophic adaptations during resistance training, and did so with the use of monopolar surface electromyography (EMG), skinfolds, and circumference measurements. What was novel about their experimental approach was the sophisticated analysis of the linear ⇑ Corresponding author at: Texas Tech University, 3204 Main Street, Box 43011, Room 103C, Lubbock, TX 79409-3011, USA. E-mail address: [email protected] (M.S. Stock). http://dx.doi.org/10.1016/j.humov.2015.11.004 0167-9457/Ó 2015 Elsevier B.V. All rights reserved.

8

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

slope coefficient for the EMG amplitude versus isometric force relationship over time. Moritani and deVries (1979) concluded that the initial improvement in muscular strength was due to neuromuscular factors, and that hypertrophy did not become evident until the latter stages of the investigation. While the novelty of their techniques was admirable, one might contend that expressing neuromuscular and hypertrophic changes into mutually exclusive categories is simplistic and creates the impression that small-to-moderate changes in muscle size do not occur. In contrast to the concept that the rapid improvement in muscular strength is governed solely by neuromuscular adaptations, some authors have reported evidence of muscle hypertrophy in response to short-term training (Baroni et al., 2013; Boone, Stout, Beyer, Fukuda, & Hoffman, 2015; DeFreitas, Beck, Stock, Dillon, & Kasishke, 2011; Seynnes, de Boer, & Narici, 2007; Staron et al., 1994). However, making direct comparisons among these studies is complicated by the fact that each of them assessed dissimilar time courses, distinct training programs, and different methods for measuring muscle growth. Using muscle biopsy sampling, Staron et al. (1994) reported that short-term resistance training resulted in a gradual shift for the type IIx muscle fibers to the more oxidative type IIa, and this alteration occurred in the absence of increases in both fat-free mass and fiber cross-sectional area. Three investigations have utilized B-mode ultrasonography to examine hypertrophy of the superficial quadriceps femoris muscles in response to short-term training (Baroni et al., 2013; Boone et al., 2015; Seynnes et al., 2007). In general, these studies have demonstrated that within four weeks of training, the vastus lateralis and/or rectus femoris showed altered muscle architecture, as evidenced by increased thickness, pennation angle, or fascicle length (Baroni et al., 2013; Boone et al., 2015; Seynnes et al., 2007). Perhaps the most intriguing findings were reported by DeFreitas et al. (2011), who performed testing on a weekly basis using peripheral quantitative computed tomography. In that study, the training program involved three sessions per week of leg press and leg extension exercises. The subjects performed three sets for each exercise with external loads that allowed for muscular failure between 8 and 12 repetitions. These authors showed statistically significant increases in thigh muscle cross-sectional area after only two training sessions, with further increases each week. It is worth noting that none of the previously described short-term investigations have studied multiple training groups and compared responses among differing types of training programs (e.g., high volume versus high load training). If meaningful increases in muscle hypertrophy are attainable within only a few training sessions (as shown by DeFreitas et al. (2011)), it is unclear how this response may be optimized by manipulating various acute training program variables. The purpose of this investigation was to examine weekly neuromuscular and hypertrophic adaptations associated with only four weeks (two sessions per week) of barbell back squat and deadlift training in previously untrained women. The barbell squat and deadlift were of interest because they involve dozens of muscles and multiple joints through a large range of motion. A secondary purpose was to compare these responses between subjects exposed to low versus moderate volume training. We hypothesized that four weeks of barbell squat and deadlift training would result in increased isometric peak torque and surface EMG amplitude, as well as decreases in agonist–antagonist coactivation. In agreement with two recent studies (Baroni et al., 2013; DeFreitas et al., 2011), we further hypothesized that significant increases in lean mass, muscle thickness, and pennation angle would be evident within four weeks. We also postulated that moderate volume training would bring about better improvements than those demonstrated for low volume training. 2. Methods 2.1. Subjects Forty-seven women (mean ± SD age = 21 ± 3 years; body mass = 63.3 ± 11.0 kg; height = 162.1 ± 9.6 cm) who were not engaged in resistance training during the previous six months completed this study. Prior to participation, potential subjects were screened for health-related illness. Women were not able to participate if they were affected by neuromuscular or metabolic disease. Furthermore, women with recent musculoskeletal discomfort, pain, or injury were not able to participate. Contraceptive use was permitted as long as its usage had remained consistent over the previous three months, since a previous training study showed minimal influence on strength-related outcomes (Nichols, Hetzler, Villanueva, Stickley, & Kimura, 2008). This study and its procedures were approved by the university’s Human Research Protection Program. All subjects read, understood, and signed an informed consent form prior to participation. Each subject was randomly assigned to one of three groups: (1) low volume training (n = 15), (2) moderate volume training (n = 16), and (3) control (n = 16). The subjects were asked to refrain from resistance training outside of the study, but up to two hours physical activity (e.g., walking, light jogging, and cycling) per week were permitted. 2.2. Testing and training schedules The subjects participated in six separate testing sessions, the first of which served as a familiarization trial. The pretest occurred 48 h following the familiarization session. Testing commenced at the conclusion of each of the four training weeks. For the subjects assigned to one of the two training groups, each testing session occurred 72 h following the final training session of the week. In addition, the subjects in the training groups performed testing and training on the same day (testing followed by training). A concerted effort was made to test each subject at the same time of day for the six testing sessions (±one hour). Most of the subjects in the training groups visited the laboratory on Monday (training only) and Thursday

9

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

(testing followed by training), or Tuesday (training only) and Friday (testing followed by training). The subjects in the control group visited the laboratory once per week for testing. The testing procedures always occurred in the order listed below. Test–retest reliability statistics for each of our dependent variables were calculated a priori based on the calculations described by Weir (2005), and have been presented in Table 1. 2.3. Experimental procedures 2.3.1. Dual X-ray absorptiometry (DXA) The subjects completed one total body scan using the DXA (Lunar Prodigy Primo, GE Healthcare, Madison, WI) during each of the testing sessions. All scans were performed by a trained technician that had completed both university radiation training and a training session held by the device’s manufacturer. In accordance with the manufacturer’s recommendations, DXA quality assurance testing was performed every 48 h throughout the study. Each variable was analyzed with manufacturer-provided software (Lunar Radiation Body Composition, version 13.60, GE Healthcare, Wauwatosa, WI). Specific regions of interest were also determined with the manufacturer-provided software. In addition to total body measures (mass, lean mass, and fat mass), regional analyses were used to sum the lean mass values from the gynoid and legs. As the barbell squat and deadlift both rely heavily on hip extension, we hypothesized that training would primarily affect lean mass for both the gynoid and leg regions. Therefore, in addition to analyzing lean mass from the legs, we summed the gynoid and leg components to create an additional dependent variable. Each of the five DXA-related dependent variables has been reported in kg. 2.3.2. Ultrasonography Following each DXA scan, the subjects remained supine on the table for ultrasound measurements of the right thigh. Performing the DXA scans in the supine position prior to ultrasound measurements allowed for the redistribution of body fluids (Berg, Tedner, & Tesch, 1993). All assessments were performed with the right leg positioned in full extension. Ultrasound measurements were performed on the right vastus lateralis and rectus femoris muscles with a portable B-mode imaging device (GE Logiq e BT12, GE Healthcare, Milwaukee, WI) and a multi-frequency linear-array probe (12 L-RS, 5–13 MHz, 38.4-mm field of view; GE Healthcare, Wauwatosa, WI). Vastus lateralis images were taken over the belly of the muscle in the same location as that for the EMG sensor placement (see below). Rectus femoris images were taken at 50% of the line from the anterior spina iliaca superior to the superior border of the patella. Ultrasound settings were optimized (Frequency: 12 MHz, Gain: 50 dB, Dynamic Range: 72, and Depth: 5 cm) based on a recent study by Wells et al. (2014). For some of the subjects with excessive adipose tissue over the superficial quadriceps femoris muscles, image depth was increased to ensure visual clarity of the deep aponeurosis. The ultrasonography probe was covered with water-soluble transmission gel (Aquasonic 100 ultrasound transmission gel, Parker Laboratories, Inc., Fairfield, NJ) and oriented along the longitudinal axis of the muscle. Three images were taken at each site. The same investigator performed all of the ultrasound measurements. To ensure image location consistency, the skin over the muscles was lightly marked following the initial assessment, and was referenced during subsequent data collection trials. Images were digitized and analyzed with Image J software (version 1.46, National Institutes of Health, Bethesda, MD). Image analyses were performed by a separate investigator that was blinded to both subject identification and group membership. For both muscles, muscle thickness was calculated as the vertical distance between the borders of the superficial and deep aponeuroses (cm) at the left, center, and right portions of each image, thereby providing nine values for each testing session. The highest and lowest values were then removed from the dataset to minimize variability, and the mean of the remaining seven values was used as a measure of muscle thickness. For both muscles, pennation angle was calculated as the angle between the muscle fascicle and the deep aponeurosis as

Table 1 Test–retest reliability statistics for each of the dependent variables in this study. These data were collected from 18 women on two days separated by 48 h (±one hour). Each test was performed by the same investigator. Ultrasound images were analyzed by a separate investigator who was blinded to both subject identification and group membership. The values for the SEM and MD are presented in the same units of measurement as those for each mean ± SD. Dependent variable

Test

Re-test

p

95% CI

Cohen’s d

ICC2,1

SEM

SEM (%)

MD

Body mass (kg) Lean mass (kg) Gynoid & leg lean mass (kg) Leg lean mass (kg) Fat mass (kg) Vastus lateralis muscle thickness (cm) Rectus femoris muscle thickness (cm) Vastus lateralis pennation angle (°) Leg extensor MVC peak torque (Nm) Vastus lateralis EMG amplitude (lV RMS) Biceps femoris coactivation (%)

62.23 ± 8.91 37.77 ± 3.49 18.41 ± 1.63 12.80 ± 1.22 21.67 ± 8.30 2.07 ± 0.25 2.03 ± 0.23 18.7 ± 3.7 161.30 ± 22.51 152.69 ± 60.75 47.96 ± 31.33

62.26 ± 8.87 37.97 ± 3.40 18.38 ± 1.71 12.77 ± 1.26 21.66 ± 8.61 2.06 ± 0.28 2.12 ± 0.16 18.0 ± 2.4 152.58 ± 26.03 141.23 ± 64.31 40.95 ± 20.34

.867 .593 .814 .805 .935 .600 .091 .408 * .047 .207 .327

0.37 to 0.32 0.98–0.58 0.23–0.29 0.18 to 0.23 0.34 to 0.36 0.03 to 0.06 0.19 to 0.02 1.04 to 2.45 0.11–17.33 6.99 to 29.92 7.68 to 21.87

.01 .06 .02 .02 .01 .05 .45 .23 .36 .19 .27

.997 .900 .952 .946 .997 .940 .391 .370 .711 .818 .784

0.49 1.11 0.38 0.30 0.50 0.07 0.15 2.5 12.24 26.24 13.92

0.79 2.93 2.04 2.30 2.30 3.19 7.21 13.6 7.80 17.86 31.30

1.36 3.08 1.04 0.82 1.38 0.18 0.42 6.9 33.93 72.75 38.57

* Statistically significant decrease from test to re-test; 95% CI = 95% confidence interval for the mean difference; ICC2,1 = intraclass correlation coefficient, model 2,1; SEM = standard error of measurement; MD = minimal difference needed to be considered real.

10

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

performed in previous studies (Ando et al., 2014; Baroni et al., 2013). For each subject and time point, the best fascicle was selected for analysis. For several of the subjects with high levels of subcutaneous adipose tissue over the superior thigh, we had difficulty locating the borders of the deep rectus femoris aponeurosis. As a result, we were unable to consistently locate rectus femoris fascicles that could confidently be used for the determination of pennation angle for all subjects and data collection trials. Consequently, pennation angle has only been reported for the vastus lateralis muscle. Example pretest and posttest images for the vastus lateralis muscle have been displayed in Fig. 1. 2.3.3. Isometric peak torque testing Following the ultrasound assessments, the subjects were seated on an isokinetic dynamometer (Biodex System 3, Biodex Medical Systems, Shirley, NY) in accordance with the manufacturer’s instructions for testing of the knee joint. Straps were secured around the subjects’ hips and chest, and they were instructed to gently hold onto the handles throughout testing. The right knee joint was visually aligned with the input axis of the dynamometer. During each subject’s initial visit to the laboratory, the dynamometer’s settings were recorded to ensure consistency throughout the study. Isometric testing was performed for both the leg extensors and flexors, but the data from the biceps femoris was only used as means of normalizing its activity when it served as an antagonist. All data collection sessions began with a warm-up of submaximal isometric contractions of the leg extensors and flexors. Specifically, for both muscle groups, the subjects performed five, five second isometric contractions corresponding to 50% of their perceived maximum interspersed by five seconds of rest (i.e., five seconds ‘‘on,” five seconds ‘‘off”). Isometric testing was performed at knee joint angles of 60 and 30 degrees for the leg extensors and flexors, respectively (0 = full extension). Each maximal voluntary contraction (MVC) was five seconds in duration. During all MVC testing, the subjects were verbally encouraged to ‘‘push” or ‘‘pull” as hard and fast as possible. Two minutes of rest were allotted between each MVC. The torque signals were digitized at a sampling rate of 1926 Hz (a preset commercial hardware device frequency). Leg extensor peak torque (Nm) was the dependent variable, and was calculated as the mean of the highest 250 ms epoch during the plateau phase of the MVC. For each data collection trial, the MVC with the highest peak torque value was used for analysis. 2.3.4. Surface EMG measurements and signal processing Bipolar surface EMG signals were detected from the vastus lateralis and biceps femoris during each MVC. The signals were detected with two separate TrignoTM wireless EMG sensors (interelectrode distance = 10 mm [Delsys Inc., Natick, MA]) with a bandwidth of 20–450 Hz. The sensors were placed over the muscles in accordance with the Surface EMG for the NonInvasive Assessment of Muscles project (Hermens, Freriks, Disselhorst-Klug, & Rau, 2000). For the vastus lateralis, the sensor was placed at 2/3 on the line from the anterior spina iliaca superior to the lateral side of the patella. For the biceps femoris, the sensor was placed at 50% on the line between the ischial tuberosity and the lateral epicondyle of the tibia. As performed by Carolan and Cafarelli (1992), a custom-built foam pad was secured to the posterior thigh to ensure that the biceps femoris sensor did not come into contact with the seat of the isokinetic dynamometer. Prior to testing, the skin was gently shaved and cleansed with rubbing alcohol. We asked the subjects to refrain from placing lotion over their right thigh within 24 h prior to each testing session. During the warm-up contractions, an investigator visually inspected the EMG signals to ensure low baseline noise and minimal line interference (<1.0). The mean ± SD baseline noise values for the vastus lateralis and biceps femoris muscles were 10.7 ± 9.7 and 8.5 ± 4.7 lV root-mean-squared (RMS), respectively. The EMG signals from the two muscles were time synced with the torque signal, and were also digitized at a sampling rate of 1926 Hz. The EMG signals were stored in a personal computer (Dell Optiplex 755, Round Rock, TX) for subsequent analyses. All torque and EMG signal processing was performed using custom programs with LabVIEW programming software (LabVIEW 2012, National Instruments, Austin, TX). The vastus lateralis (agonist) and biceps femoris (antagonist) EMG signals from the leg extension MVCs were selected from the same 250 ms portion of the torque curve that encompassed the peak

Fig. 1. Example ultrasound images from the vastus lateralis muscle for one subject in the low volume training group. The images on the left and right are from the pretest and posttest, respectively. The horizontal, dotted lines correspond to the aponeuroses of the muscle. Muscle thickness was determined as the distance between the inferior border of the superficial aponeurosis and the superior border of the deep aponeurosis. Each angled line corresponds to the location of the most visible fascicle that clearly intersected the deep aponeurosis, and was used in the calculation of pennation angle.

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

11

torque value. The RMS value of each selected signal was calculated as a measure of EMG amplitude. When the vastus lateralis served as an agonist, absolute EMG amplitude values (lV RMS) were examined. Biceps femoris antagonist activity was reported as a percentage of the muscle’s EMG amplitude value when it served as an agonist (Carolan & Cafarelli, 1992). 2.3.5. Delayed onset muscle soreness Prior to each of the testing sessions (i.e., 72 h following the last training session), the subjects completed a Likert scale to indicate their current level of lower-body delayed onset muscle soreness. The scale had values of 0 and 10 that represented the complete absence of soreness and severe pain, respectively. 2.3.6. Dietary analyses The subjects in this investigation were asked to keep a detailed 28 day food log that required them to record all of the foods and beverages consumed. The subjects were instructed to be as specific as possible, including details such as item brand names and the method of food preparation. When feasible, the subjects were asked to measure their food and beverage quantities. The subjects were instructed to continue their normal ad libitum diet, and to keep their caloric intake consistent throughout the study. The subjects were asked to refrain from alcohol consumption. We also asked the subjects to keep their caffeine consumption (or lack thereof) consistent and to arrive for testing well hydrated. The subjects were frequently reminded to complete their food logs by members of the research team. Nonetheless, underreporting of dietary intake in women has been documented (Hutchesson, Rollo, Callister, & Collins, 2015), and was an expected limitation of this investigation. For each week, we analyzed data for two weekdays and one weekend day at random. The food entries for these three days were then averaged to provide an estimate of the food and beverage consumption within each week. Food logs were analyzed for estimated total calories, protein, carbohydrates, and fats with NutriBase Clinical Nutrition Manager software (version 7.18, CyberSoft, Inc., Phoenix, AZ). In the event that the nutrient content of a specific food was not available, USDA standard references were used as an estimate. Eight subjects did not provide a satisfactory level of detail when recording their food and beverage information. Thus, nutrition data has been presented for only 39 subjects. 3. Back squat and deadlift training The subjects in the two training groups trained at the university’s Human Performance Laboratory twice per week for four weeks. Both exercises were preceded by two warm-up sets using external loads between 15.9 and 25.0 kg. Excluding the warm-up sets, the low volume training group performed five repetitions of two sets per exercise; the moderate volume group performed an additional two sets per exercise. Three minutes of rest were allowed between each set. Prior to the initial training session, we took ample time to thoroughly explain how the back squat and deadlift were to be performed throughout the study. The subjects had the opportunity to watch demonstrations of the exercises being performed correctly, and also learned of mental cues to concentrate on. Once the subjects had a detailed understanding of how these exercises were to be performed, training commenced. All of the training for this study was closely supervised, and the subjects received consistent verbal instruction concerning their squat and deadlift technique. The subjects were taught how to correctly execute the Valsalva maneuver to increase intra-abdominal pressure, and its use was encouraged. Weight belts were not used, but the subjects were encouraged to chalk their hands prior to training. Each repetition was performed in a controlled but forceful manner with approximately two second concentric and eccentric phases. Back squats were always performed prior to deadlifts. Examples of the exercise technique used in this study have been displayed in Fig. 2. All of the back squats were performed inside of a power rack. Two spotters were present at all times (one on each side of the barbell). The safety pins of the rack were set at a height that was just below the barbell when the subject was at the bottom of the range of motion. The subjects were taught to perform the barbell back squat with their feet approximately shoulder width apart. The toes pointed slightly outward. The barbell was placed across the top of the trapezius muscle just below the seventh cervical vertebrae. The subjects were instructed to keep the musculature of the upper-back contracted throughout the range of motion. The hands were kept outside of the shoulders with a closed grip, the head remained in a neutral position, and the elbows were positioned behind the barbell to sustain the load. Each subject squatted slightly below the parallel position, which was attained when the greater trochanter of the femur reached a position level to the patella. The subjects were taught to perform the deadlift with their feet hip width apart. Their toes were pointed forward. The starting position involved the barbell making light contact with the shank while the tibias were aligned in a vertical position. The shoulders were slightly in front of the barbell. The arms were held just outside of the thighs, and the subjects were allowed to use either a pronated or an alternated grip, depending on personal preference. The subjects were instructed to maintain their cervical spine in a neutral position throughout the range of motion. Particular attention was paid to each subject’s lower back to ensure that all repetitions were performed with the lumbar spine in a rigid, extended position. The subjects were instructed to briefly stop between repetitions and not bounce the weight from the floor. To ensure that the range of motion was consistent across all subjects, bumper plates (45.0 cm diameter) were used for all deadlift repetitions. Given that the subjects were unfamiliar with the technical aspects of these exercises, we did not feel comfortable having them perform repetition maximum strength testing at the initiation of the study. In addition, our concern was that in novices, strength testing alone would be a sufficient stimulus for improving maximal force output. Thus, the assigned

12

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

Fig. 2. An example of the barbell back squat (top) and deadlift (bottom) technique used in this study. All back squats were performed inside of a power rack with two spotters present. The back squat shown in this figure was performed outside of the power rack for visual clarity.

external loads used in this study were not based off of the results from a maximal strength assessment. Instead, a nontraditional approach that was used with success in a previous study was replicated (Stock & Thompson, 2014). For each set, our goal was for the subjects to perform all five repetitions with both a high level of physical exertion and proper technique. During the initial warm-up sets, exercise technique was closely monitored. In the event that a subject was able to perform each repetition with proper technique for a set of five, 2.3–4.6 kg was added to the barbell for the subsequent set. Conversely, in the event that all five repetitions could not be performed or the exercise technique became compromised (e.g., rounding of the lumbar spine, squat depth above parallel, etc.), additional weight was not added to the barbell. A minimum of 2.3 kg was added to the barbell for both exercises during each training session. For the subjects in the low volume group, the mean ± SD loads used for the back squat and deadlift increased from 25.8 ± 8.7 and 33.7 ± 6.7 to 52.9 ± 13.1 and 63.6 ± 10.5 kg, respectively. For the subjects in the moderate volume group, the mean ± SD loads used for the back squat and deadlift increased from 25.3 ± 8.1 and 31.0 ± 7.9 to 47.9 ± 11.0 and 56.5 ± 6.5 kg, respectively. The mean training volume for both groups and exercises has been displayed in Table 2. 4. Statistical analyses The data for this investigation were analyzed with a variety of statistical approaches. First, for each group, weekly Cohen’s d effect size statistics were used to compare the changes from each time point to the values from the pretest. Effect size values of 0.2, 0.5, and 0.8 corresponded to small, moderate, and large differences, respectively (Cohen, 1988). Second, we evaluated the number of subjects that exceeded the minimal difference needed to be considered real (MD) statistic. As discussed by Weir (2005), all measurements performed in research are influenced by testing error. The primary utility of the standard error of measurement and MD statistics is their ability to conclude whether a particular change is above and beyond what would be normally expected due to error associated with a given measurement. Univariate scatterplots illustrating the greatest absolute change score (regardless of the testing occasion) for each subject were designed using the templates described by Weissgerber, Milic, Winham, and Garovic (2015). Furthermore, traditional mixed-factorial (group [moderate volume, low volume, control]  time [pretest, week one, week two, week three, posttest]) analyses of variance (ANOVAs) were used to examine mean differences. In the event of a significant group  time interaction, the data were further decomposed with

Low volume mean back squat external loads (kg) Training # 1 2 3 4 Mean 25.8 30.6 34.9 38.9 SD 8.7 10.0 11.1 11.5

5 42.8 11.3

6 47.1 12.4

7 50.0 13.1

8 52.9 13.1

Low volume deadlift mean Training # 1 Mean 33.7 SD 6.7

Moderate volume back squat external Training # 1 2 Mean 25.3 30.5 SD 8.1 10.4

loads (kg) 3 4 34.5 37.9 11.6 11.8

5 40.6 11.2

6 43.5 11.2

7 45.4 11.1

8 47.9 11.0

Moderate volume deadlift Training # 1 Mean 31.0 SD 7.9

Low volume mean back squat volume Training # 1 2 Mean 257.9 306.3 SD 87.4 100.1

(kg) 3 348.6 110.6

4 388.7 115.3

5 428.0 113.3

6 471.1 124.2

7 499.8 131.1

8 529.3 130.9

Low volume deadlift mean Training # 1 Mean 337.3 SD 66.9

(kg) 3 689.1 231.1

4 758.6 236.3

5 811.7 225.0

6 869.2 224.3

7 908.9 222.8

8 958.5 220.3

Moderate volume deadlift Training # 1 Mean 619.6 SD 158.6

7 55.0

8 55.2

Moderate volume back squat volume Training # 1 2 Mean 505.5 610.4 SD 162.0 208.6

Mean training volume for low volume relative to moderate volume (%) Back squat Training # 1 2 3 4 5 6 51.0 50.2 50.6 51.2 52.7 54.2

Deadlift Training #

1 54.4

external loads (kg) 2 3 38.6 43.5 6.9 7.9

4 48.9 8.7

5 53.9 9.9

6 56.3 9.8

7 60.3 10.7

8 63.6 10.5

4 44.9 9.3

5 49.0 8.4

6 51.8 8.1

7 53.7 7.4

8 56.5 6.5

volume (kg) 2 3 386.4 434.8 68.8 79.2

4 488.5 86.8

5 539.2 98.5

6 563.4 98.0

7 602.7 106.7

8 636.0 104.9

volume (kg) 2 3 750.1 839.4 189.1 195.0

4 898.9 185.3

5 979.8 168.7

6 1035.8 162.6

7 1074.0 147.9

8 1129.3 130.6

4 54.3

5 55.0

6 54.4

7 56.1

8 56.3

external loads (kg) 2 3 37.5 42.0 9.5 9.8

2 51.5

3 51.8

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

Table 2 The mean and SD external loads (kg), training volume (kg), and relative volume (%) for the low and moderate volume training groups. The displayed external loads represent the mean weight used during each training session across all sets. As stated in the Section 2, additional loads were progressively added throughout the study, both within and between training sessions. Mean training volume was calculated as: external load  number of sets (two or four)  number of repetitions (five). Due to minor differences in loading schemes, the low volume group trained with slightly heavier external loads. As displayed in the bottom row, however, the training volume for the low volume group was still roughly 55% of that for the moderate volume group. The increases in exercise training volume were extremely linear (lowest Pearson r = .985).

13

14

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

repeated measures ANOVAs (across time) and one-way ANOVAs (across group). In the event of main effects for group or time, marginal means were examined with Bonferroni pairwise comparisons. In addition, the partial eta squared (ή2) statistic was reported for each ANOVA. Values of 0.01, 0.06, and 0.14 correspond to small, medium, and large effect sizes, respectively. An alpha level of .05 was used to determine statistical significance for all analyses. Finally, true individual response differences were quantified by comparing the SDs of the pretest–posttest change scores (ignoring weeks one, two, and three) between training and control groups (Atkinson & Batterham, 2015; Hopkins, 2015). This was calculated as follows: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi SDR = SD2training SD2control . The SDR was then standardized by dividing it by the SD of all subject scores during the pretest. The thresholds for interpreting the standardized SDs were 0.1, 0.3, 0.6, 1.0, and 2.0 for small, moderate, large, very large, and extremely large, respectively (Hopkins, 2015). 5. Results 5.1. Body mass There were no meaningful group mean changes in body mass. Seven (46.7%) and eight (50.0%) subjects from the low and moderate volume groups, respectively, showed changes that exceeded the MD (±1.36 kg). The results from the two-way mixed-factorial ANOVA indicated that there was no group  time interaction (p = .731, ή2 = .022), as well as no main effects for group (p = .205, ή2 = .069) or time (p = .573, ή2 = .013). For both training groups, the pretest–posttest standardized SDR was small (0.13 and 0.05 for low and moderate volume, respectively). 5.2. Total body lean mass Both training groups showed small week-to-week effect sizes for the increase in lean body mass (low volume training Cohen’s d range = 0.04–0.17; moderate volume training Cohen’s d range = 0.04–0.20). None of the subjects demonstrated changes that exceeded the MD (±3.08 kg). The results from the two-way mixed-factorial ANOVA indicated that there was no group  time interaction (p = .505, ή2 = .040). There was also no main effect for group (p = .856, ή2 = .007). There was, however, a main effect for time (p = .002, ή2 = .096). The marginal mean Bonferroni pairwise comparisons indicated that lean body mass was significantly greater at week three (38.44 kg) compared to both the pretest (37.99 kg) and week one (37.93 kg). For both training groups, the SD for the pretest–posttest change score was less than that observed for the control group (i.e., negative SDR). 5.3. Gynoid + leg lean mass The pretest–posttest effect sizes for the training groups ranged from d = 0.09 to 0.34 (Fig. 3). The MD (±1.04 kg) was exceeded by eight (53.3%) and five (31.3%) subjects from the low and moderate volume groups, respectively (Fig. 4). Twelve of these thirteen subjects demonstrated increases that were above the MD, whereas one subject in the moderate volume group showed a meaningful decrease. One subject gained over 2.9 kg in less than four weeks. The results from the twoway mixed-factorial ANOVA indicated that there was a group  time interaction (p = .001, ή2 = .159). For the control group, the results from a repeated measures ANOVA indicated that the mean at the posttest (18.6 kg) was significantly less than the means for weeks two (18.9 kg; p = .029) and three (19.0 kg; p = .036). For the low volume group, significant increases above the pretest were shown at week two (+0.68 kg; p = .045) and the posttest (+0.67 kg; p = .042). Similarly, statistically significant increases were shown at week three (+0.47 kg; p = .038) and the posttest (+0.55 kg; p = .041) for the moderate volume group. The results from five separate one-way ANOVAs indicated that there were no between-group differences at any of the time points. For both training groups, the pretest–posttest standardized SDR was small (0.20 and 0.10 for low and moderate volume, respectively). 5.4. Leg lean mass As displayed in Fig. 3, both training groups showed small-to-moderate increases in leg lean mass. Interestingly, the low volume group showed slightly higher effect sizes than those for the moderate volume group (Cohen’s d ranges = 0.34–0.43 and 0.07–0.27, respectively). Twelve subjects showed changes that exceeded the MD (±0.82 kg, Fig. 4), one of which was a decrease in leg lean mass. The results from the two-way mixed-factorial ANOVA indicated that there was a group  time interaction (p = .001, ή2 = .165). Follow-up repeated measures ANOVAs indicated that: (1) the control group showed significantly less leg lean mass at the posttest (12.9 kg) compared to weeks two (13.2 kg) and three (13.3 kg) and (2) the increases in leg lean mass for the low volume and moderate volume groups were not significant at an alpha level of .05 (low volume pretest to posttest mean increase = 0.57 kg, p = .088; moderate volume pretest to posttest mean increase = 0.40 kg, p = .149). Although the control group had slightly greater leg lean mass at the beginning of the study (control, low volume, and moderate volume = 13.2, 12.7, and 12.7 kg, respectively) the results from five separate one-way ANOVAs indicated that there were no between-group differences at any of the time points. For the low and moderate volume training groups, the pretest–posttest standardized SDR was moderate (0.31) and small (0.16), respectively.

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

15

Fig. 3. Weekly effect size (top) and mean ± SD bar graphs (bottom) for gynoid + leg (left) and leg (right) lean body mass. Each effect size statistic corresponds to the change from the pretest. Negative effect sizes have been displayed for visual purposes only to represent a decrease in the mean values over time. For both training groups, the repeated measures ANOVAs indicated that there were significant increases in gynoid + leg lean mass. *Significantly greater than the corresponding pretest mean. àSignificantly less than weeks two and three (control group).

Fig. 4. Individual subject data points that represent the largest absolute change in gynoid + leg (left) and leg (right) lean mass, regardless of the testing session (i.e., weeks one-three and posttest). For each scatterplot, the data points on the left, middle, and right correspond to the control, low volume, and moderate volume groups, respectively. The black bars show the mean of each group. The red lines correspond to the minimal difference needed to be considered real, and the data points above or below each line surpass this threshold. The reader is asked to keep in mind that these are the greatest changes shown in the study, and therefore accentuate extremes for each subject. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

16

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

5.5. Fat mass Each group demonstrated very small reductions in fat mass (Cohen’s d range = 0.00–0.06). The low volume group showed a pretest to posttest fat mass mean decrease of 0.53 kg. Four (26.7%) and two (12.5%) subjects from the low and moderate volume groups, respectively, showed decreases in fat mass that exceeded the MD (±1.38 kg). The results from the two-way mixed-factorial ANOVA indicated that there was no group  time interaction (p = .795, ή2 = .021), as well as no main effects for group (p = .077, ή2 = .110) or time (p = .066, ή2 = .056). For both training groups, the pretest–posttest standardized SDR was small-to-trivial (0.15 and 0.04 for low and moderate volume, respectively). 5.6. Vastus lateralis muscle thickness An analysis of the week-to-week effect sizes indicated that the low volume group demonstrated small increases throughout the study (Cohen’s d range = 0.23–0.35, Fig. 5). Furthermore, the moderate volume group showed slightly higher effect sizes, with a consistent increase each week (Cohen’s d values compared to the pretest = 0.25, 0.27, 0.44, 0.48). The MD (±.18 cm) was equivalent to or exceeded by five (33.3%) and eight (50.0%) subjects from the low and moderate volume groups, respectively (Fig. 6). All five of the changes from the low volume group were increases in muscle thickness. Of the eight from the moderate volume group, one subject showed a decrease in muscle thickness that was equivalent to the MD. The results from the two-way mixed-factorial ANOVA indicated that there was no group  time interaction (p = .129, ή2 = .070). There was also no main effect for group (p = .805, ή2 = .010). There was, however, a main effect for time (p < .001, ή2 = .129). The results from the Bonferroni pairwise comparisons indicated that the vastus lateralis muscle thickness marginal means at weeks one (2.13 cm), three (2.17 cm), and the posttest (2.14 cm) were all greater than that for the pretest (2.07 cm). Despite these changes, for both training groups, the SD for the pretest–posttest change score was less than that observed for the control group (i.e., negative SDR). 5.7. Vastus lateralis pennation angle The largest effect sizes shown in this investigation were for the increase in vastus lateralis pennation angle for the low volume training group. This finding was in spite of moderate reliability statistics for this dependent variable (Table 1). As

Fig. 5. Weekly effect size (top) and mean ± SD bar graphs (bottom) for vastus lateralis muscle thickness (left) and pennation angle (right). Each effect size statistic corresponds to the change from the pretest. Negative effect sizes have been displayed for visual purposes only to represent a decrease in the mean values over time. For muscle thickness (left), there was a main effect for time. The marginal means at weeks one (2.13 cm), three (2.17 cm), and the posttest (2.14 cm) were all greater than the pretest marginal mean (2.07 cm).

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

17

Fig. 6. Individual subject data points that represent the largest absolute change in vastus lateralis muscle thickness (left) and pennation angle (right), regardless of the testing session (i.e., weeks one-three and posttest). For each scatterplot, the data points on the left, middle, and right correspond to the control, low volume, and moderate volume groups, respectively. The black bars show the mean of each group. The red lines correspond to the minimal difference needed to be considered real, and the data points above or below each line surpass this threshold. The reader is asked to keep in mind that these are the greatest changes shown in the study, and therefore accentuate extremes for each subject. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

shown in Fig. 5, the week-to-week effect size values were 0.19, 0.61, 0.71, and 0.67. The increases for the moderate volume group were slightly lower, with a pretest-to-posttest effect size of 0.57 (means of 18° and 20°, respectively). The mean pennation angle values for the control group were similar at each testing session (18°). Four subjects (12.9%) from the two training groups combined showed increases that exceeded the MD (±6.9°, Fig. 6). The results from the two-way mixedfactorial ANOVA indicated that there was no group  time interaction (p = .289, ή2 = .054). There was also no main effect for group (p = .266, ή2 = .060). There was, however, a main effect for time (p = .021, ή2 = .067). Although each of the marginal means showed increases over time, none of these differences were statistically significant. For the low volume training group, the SD for the pretest–posttest change score was less than that observed for the control group (i.e., negative SDR). In contrast, the pretest–posttest standardized SDR for the moderate volume group was moderate-to-large (0.85). 5.8. Rectus femoris muscle thickness None of the groups showed important changes in rectus femoris muscle thickness. The results from the two-way mixedfactorial ANOVA indicated that there was no group  time interaction (p = .807, ή2 = .023). In addition, there were no main effects for group (p = .358, ή2 = .046) or time (p = .828, ή2 = .007). For both training groups, the SD for the pretest–posttest change score was less than that observed for the control group (i.e., negative SDR). 5.9. Leg extensor MVC peak torque The results from the two-way mixed-factorial ANOVA indicated that there was no group  time interaction (p = .065, ή2 = .081). There was also no main effect for group (p = .121, ή2 = .091). There was, however, a main effect for time (p < .001, ή2 = .140). The results from the Bonferroni pairwise comparisons indicated that the leg extensor peak torque marginal mean at the posttest (167.0 Nm) was significantly greater than that for the pretest (155.1 Nm), week one (157.0 Nm), and week two (159.3 Nm). Despite the lack of a group  time interaction, both training groups demonstrated mean increases in the leg extensor peak torque values (Fig. 7). Specifically, when compared to the pretest, the low volume training group showed weekly effect sizes of d = 0.05, 0.14, 0.40, and 0.62 for weeks one, two, three, and the posttest, respectively. The moderate volume group showed slightly smaller changes (d = 0.02, 0.14, 0.16, and 0.40, respectively). For leg extensor MVC peak torque, the MD (±33.93 Nm) was exceeded by five (33.3%) and three (18.8%) subjects from the low and moderate volume groups, respectively (Fig. 8). Of the five from the low volume training group, three demonstrated increases and two showed decreases. Of the three from the moderate volume group, two subjects showed an increase and one subject displayed a decrease in peak torque. It should be noted that our test–retest reliability statistics showed a significance difference between trials for this dependent variable (p = .047). For both training groups, the SD for the pretest–posttest change score was less than that observed for the control group (i.e., negative SDR).

18

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

Fig. 7. Weekly effect size (top) and mean ± SD bar graphs (bottom) for isometric peak torque (left) and EMG amplitude for the vastus lateralis (right), both of which were obtained during an MVC of the right leg extensors. Each effect size statistic corresponds to the change from the pretest. For isometric peak torque, there was a main effect for time. The results from the Bonferroni pairwise comparisons indicated that the leg extensor peak torque marginal mean at the posttest (167.0 Nm) was significantly greater than that for the pretest (155.1 Nm), week one (157.0 Nm), and week two (159.3 Nm).

Fig. 8. Individual subject data points that represent the largest absolute change in isometric peak torque (left) and EMG amplitude for the vastus lateralis (right), regardless of the testing session (i.e., weeks one-three and posttest). For each scatterplot, the data points on the left, middle, and right correspond to the control, low volume, and moderate volume groups, respectively. The black bars show the mean of each group. The red lines correspond to the minimal difference needed to be considered real, and the data points above or below each line surpass this threshold. The reader is asked to keep in mind that these are the greatest changes shown in the study, and therefore accentuate extremes for each subject. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

19

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

5.10. Vastus lateralis EMG amplitude Each of the three groups showed small increases in vastus lateralis EMG amplitude over time (Fig. 7). The largest change was displayed for the low volume group (pretest–posttest Cohen’s d = 0.38). Four subjects from both the low (26.7%) and moderate (25.0%) volume groups demonstrated increases that exceeded the MD (±72.75 lV RMS, Fig. 8). The results from the two-way mixed-factorial ANOVA indicated that there was no group  time interaction (p = .487, ή2 = .040). In addition, there were no main effects for group (p = .989, ή2 < .001) or time (p = .112, ή2 = .044). The moderate volume training group showed a moderate-to-large pretest–posttest SDR (0.48), whereas the SDR for the low volume group was negative. 5.11. Biceps femoris coactivation The two training groups showed similar week-to-week effect sizes throughout the study that were each <.28. At the posttest, the control group showed an increase in coactivation that resulted in an effect size of d = 0.51. Four (26.7%) and three (18.8%) of the subjects from the low and moderate volume groups, respectively, showed changes that exceeded the MD (±38.57%). The results from the two-way mixed-factorial ANOVA indicated that there was no group  time interaction (p = .456, ή2 = .041). In addition, there were no main effects for group (p = .157, ή2 = .081) or time (p = .225, ή2 = .033). The low volume training group showed a moderate-to-large pretest–posttest SDR (0.56), whereas the SDR for the moderate volume group was negative. 5.12. Delayed onset muscle soreness The subjects reported very little muscle soreness. Careful evaluation of individual subject data showed that some of the subjects in the moderate volume training group reported slightly higher delayed onset muscle soreness than those in the low volume training group, but these differences were not noteworthy. For the low and moderate volume training groups, five (33.3%) and eight (50.0%) of the subjects, respectively, reported a complete absence of muscle soreness throughout the entire investigation. The results from the two-way mixed-factorial ANOVA indicated that there was no group  time interaction (p = .102, ή2 = .075) and no main effects for group (p = .298, ή2 = .054) or time (p = .893, ή2 = .005). 5.13. Dietary analyses Data concerning the self-reported dietary intake of 39 subjects have been displayed in Table 3. In general, the subjects in the low volume group consumed less total and relative calories, protein, carbohydrate, and fat. For total calories, there was no significant group  time interaction (p = .968, ή2 = .012) and no main effect for time (p = .877, ή2 = .006), but there was a

Table 3 Mean ± SD caloric and macronutrient contents for each group (n = 39). For daily energy intake, there was a significant main effect for group (moderate volume > low volume). Although the food logs for these 39 subjects were detailed and thorough, the values below are likely indicative of underreporting. Week one

Week two

Week three

Week four

1613.9 ± 450.3 1392.8 ± 386.2 1770.1 ± 438.2

1507.6 ± 534.7 1391.2 ± 616.2 1775.2 ± 463.0

1599.9 ± 583.6 1286.9 ± 524.7 1692.9 ± 408.9

1580.4 ± 524.0 1336.5 ± 388.9 1762.5 ± 454.4

Relative daily energy intake (kCal/kg body mass) Control (n = 14) 26.9 ± 7.3 Low volume (n = 13) 21.4 ± 7.7 Moderate volume (n = 12) 28.5 ± 7.2

25.1 ± 8.5 22.2 ± 13.5 29.3 ± 10.4

26.8 ± 9.5 20.3 ± 11.0 27.4 ± 7.5

26.7 ± 9.9 21.0 ± 9.5 28.5 ± 9.1

Relative daily protein intake (g/kg lean body mass) Control (n = 14) 1.55 ± 0.48 Low volume (n = 13) 1.36 ± 0.45 Moderate volume (n = 12) 1.89 ± 0.54

1.54 ± 0.73 1.45 ± 0.63 1.85 ± 0.62

1.60 ± 0.33 1.29 ± 0.66 1.87 ± 0.84

1.55 ± 0.58 1.42 ± 0.44 1.84 ± 0.72

Relative daily protein intake (g/kg body mass) Control (n = 14) 0.97 ± 0.33 Low volume (n = 13) 0.79 ± 0.34 Moderate volume (n = 12) 1.15 ± 0.39

0.98 ± 0.48 0.85 ± 0.45 1.14 ± 0.45

1.03 ± 0.52 0.76 ± 0.44 1.16 ± 0.56

0.99 ± 0.40 0.84 ± 0.35 1.13 ± 0.51

Relative daily carbohydrate intake (g/kg body mass) Control (n = 14) 3.65 ± 1.00 Low volume (n = 13) 2.77 ± 0.92 Moderate volume (n = 12) 3.09 ± 1.40

3.28 ± 1.04 2.75 ± 1.67 3.45 ± 1.10

3.51 ± 1.21 2.46 ± 1.45 3.05 ± 0.77

3.54 ± 1.38 2.53 ± 1.19 3.31 ± 0.84

Relative daily fat intake (g/kg body mass) Control (n = 14) Low volume (n = 13) Moderate volume (n = 12)

0.93 ± 0.40 0.90 ± 0.63 1.27 ± 0.69

1.07 ± 0.48 0.85 ± 0.50 1.19 ± 0.57

0.99 ± 0.44 0.85 ± 0.44 1.24 ± 0.57

Energy intake (kCal) Control (n = 14) Low volume (n = 13) Moderate volume (n = 12)

0.96 ± 0.37 0.84 ± 0.36 1.19 ± 0.50

20

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

main effect for group (p = .040, ή2 = .164). The results from the Bonferroni pairwise comparisons indicated that the moderate volume group consumed more total calories than the low volume group (p = .036; moderate and low volume marginal means = 1750.2 and 1351.8 kCal, respectively). For the data displayed in Table 3, there were no additional group  time interactions (p P .621, ή2 6 .039) and no main effects for group (p P .081, ή2 6 .130) or time (p P .772, ή2 6 .012). 6. Discussion The most important finding of this investigation was that resistance training resulted in statistically significant increases in gynoid + leg lean mass, with similar changes for low and moderate volume training (pretest–posttest increase = 0.67 and 0.55 kg, respectively). In addition, small-to-moderate increases were demonstrated for leg lean mass, vastus lateralis muscle thickness and pennation angle, and peak torque. Many of the dependent variables showed similar changes for the two training groups, which may be indicative of a minimal work threshold that was met by all of the training subjects. For each of these variables, the MD was exceeded by some, but not all, of the subjects. Our findings are in agreement with five investigations that have reported muscular adaptations within four weeks of initiating a resistance training program (Baroni et al., 2013; Boone et al., 2015; DeFreitas et al., 2011; Seynnes et al., 2007; Staron et al., 1994). Although none of them used DXA to examine changes in body composition in women, Rana et al. (2008) used air displacement plethysmography to assess increases in lean mass in response to six weeks of low velocity resistance training, traditional resistance training, and a resistance training program designed to improve muscular endurance. It was reported that all groups (including a control group) demonstrated significant improvements in body composition. Overall, when training interventions performed in women between eight and 20 weeks in duration are considered, it seems that 0.5–3.0 kg increases in lean mass are typical (Fleck & Kraemer, 2004). Thus, our findings seem to be in line with the changes seen in longer duration training studies, which may provide support for the notion of a fairly linear increase in muscle growth at the onset of a resistance training program. One of the most perplexing aspects of our DXA and ultrasound findings was that those in the low volume group self-reported fewer calories consumed than the moderate volume group, yet many of these subjects still showed important changes (Table 3). To our knowledge, the only previous short-term resistance training intervention to examine dietary patterns of research subjects was recently published by Boone et al. (2015). These authors randomly assigned 18 men to resistance training + protein (n = 9) and resistance training only (n = 9) groups. The subjects were asked to provide a three-day dietary recall at pretesting and posttesting. Unlike the present investigation, Boone et al. (2015) reported no significant between-group differences for total calories or macronutrient distribution, and concluded that short-term resistance training resulted in significant increases in muscle size in untrained young men, regardless of protein supplementation. The findings of the present study are in agreement with those described by Boone et al. (2015), and suggested that in untrained women beginning a resistance training program, increases in muscle size may be plausible irrespective of nutritional status. Our results demonstrated that the agonist–antagonist behavior for the leg extensors was largely unaffected by training. It should be noted that although many studies have reported changes in EMG amplitude following resistance training (Hakkinen & Komi, 1983; Hakkinen et al., 1998; Moritani & deVries, 1979; Stock & Thompson, 2014), others have not (Beck et al., 2007; Holtermann, Roeleveld, Vereijken, & Ettema, 2005). In addition to increased EMG amplitude for the agonist(s), decreases in coactivation are thought to be an important neuromuscular adaptation to resistance training. Carolan and Cafarelli (1992) first demonstrated that increased isometric leg extension force was brought about not by increases in EMG amplitude of the agonist, but decreases in biceps femoris activity. This was demonstrated with only one week of training. In the present study, while the exact reason for the lack of change in EMG amplitude for both the examined agonist and antagonist is unknown, the issue of training versus testing specificity is a reasonable explanation. Both back squats and deadlifts are dynamic movements that allow the leg and hip extensors to produce force through a large range of motion. Despite the fact that the isometric peak torque values showed small-to-moderate increases as a result of the training program (Fig. 7), it is likely that the subjects’ improvements were specific to the movement patterns used during training (Rutherford & Jones, 1986). One of the challenges associated with short-term training studies in which measures of muscle strength and size are both examined is the likelihood for data to be influenced by muscle soreness and edema. Changes in ultrasound echo intensity and muscle thickness have been used as indirect measures of muscle damage (Yin Lau, Blazevich, Newton, Xuan Wu, & Nosaka, 2015), and Warren, Lowe, and Armstrong (1999) concluded that the measurement of MVC force/torque is the best method for quantifying muscle injury. Since these tools are also used to examine muscle strength and growth, when designing studies aimed at investigating short-term changes, investigators must strike a balance between training in a manner that is conducive for optimizing adaptations while ensuring that the changes observed are not a reflection of the muscle’s recovery process. This issue was potentially exemplified in the study by DeFreitas et al. (2011). These authors reported that muscle cross-sectional area increased within one week of training, yet the increase in isometric MVC force was not significant until the fourth week. Although muscle soreness was not measured, DeFreitas et al. (2011) stated that many of their subjects complained of it until the third week of training. In addition to the differences in sex and exercise selection, in the present study, testing occurred 72 h following the previous training session, whereas DeFreitas et al. (2011) performed assessments 48 h following exercise. One of the reasons we compared low and moderate volume training programs was because we believed that our subjects would report substantial muscle soreness, thereby complicating data interpretation. Since force loss is

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

21

associated with the amount of eccentric work performed (Nosaka, Sakamoto, Newton, & Sacco, 2001), we expected the moderate volume group to demonstrate no change, or even a mean decrease, in isometric peak torque within the first week of training. While at least one study has compared increases in muscle size and strength for work-matched concentric and eccentric training (Moore, Young, & Phillips, 2012), we are unaware of investigations that have done so on a short-term basis. Future investigators that are interested in this topic should consider examining the time course of muscle growth for workmatched concentric versus eccentric training to tease out the relative influence of muscle hypertrophy and edema, the latter of which is associated only with lengthening contractions (Warren et al., 1999). To our understanding, the present study was the first to report SDR values in the examination of short-term resistance training adaptations. It was also only the third resistance training intervention to make interpretations based on the MD statistic (DeFreitas et al., 2011; Stock & Thompson, 2014). When data for both training groups were combined, the variables corresponding to body composition and muscle architecture (i.e., DXA and ultrasonography) showed changes that exceeded the MD for 25.2% of the subjects. This degree of improvement was very similar to that observe for the isometric strength and EMG assessments, both of which demonstrated meaningful changes in 25.1% of the subjects. Furthermore, for several (but not all) of the dependent variables, the SDR was 60, which is indicative of greater variability for the control group than that observed for the training group(s). According to recent reviews by Hopkins (2015) and Atkinson and Batterham (2015), an observed SDR that is 60 is indicative of high tester error and/or excessive random within-subject variation, and any corresponding mean change as a result of an intervention should be ignored. Therefore, in spite of the results from our traditional statistical approaches (i.e., ANOVAs and effect sizes), we concede that many of the changes observed may not have been above and beyond our laboratory’s testing error. This investigation had several noteworthy limitations that we are compelled to describe. First, as can be surmised by examining the values shown in Table 3, it is very likely that some of our subjects underreported their food and beverage consumption. We fully acknowledge that asking our subjects to record the precise details of all of their food and beverage consumption was a demanding request. Each approach for monitoring dietary intake has inherent advantages and disadvantages, and this remains a challenging issue for body composition researchers (Hutchesson et al., 2015; Schoenfeld, Aragon, Wilborn, Krieger, & Sonmez, 2014). We contend that outside of constant, 24 h monitoring and providing meals with known calories and macronutrient distribution, each approach to considering dietary patterns has the potential for error. Second, although each subject’s time of day for testing was consistent, our DXA scans were not performed in the fasted state and hydration was not measured prior to each data collection session. Although this was considered during the design of the study, it was decided that limiting the 200+ DXA scans to the morning hours only would have made scheduling difficult, since the subjects in the training groups performed testing prior to half of their training sessions (i.e., roughly 2.5 h per laboratory visit). Though our reliability statistics were calculated following DXA scans that were also not performed in the fasted state (Table 1), this decision introduced additional variability in our body composition measures (Nana, Slater, Hopkins, & Burke, 2012; Nana, Slater, Stewart, & Burke, 2015). Finally, although the subjects were asked to refrain from resistance training and participate in only light physical activity, we have no way of knowing whether they followed this request. In addition to measurement error, any one of these limitations may have been the cause of the statistically significant body composition changes demonstrated for the control group. In summary, these results demonstrated small-to-moderate lower-body muscular adaptations in response to only eight sessions of barbell squat and deadlift training in women. Similar findings were shown for subjects that trained with low and moderate volumes. While the most notable finding was a statistically significant increase in gynoid + leg lean mass for both training groups, main effects and small-to-moderate effect sizes were also demonstrated for vastus lateralis muscle thickness and pennation angle and peak torque. No significant changes in surface EMG amplitude were demonstrated for the vastus lateralis (agonist) and biceps femoris (antagonist) during MVCs. In general, the subjects in the present study reported relatively little delayed onset muscle soreness, which may be a confounding factor in short-term training studies that involve high eccentric forces (DeFreitas et al., 2011). Our findings may support the notion that there is a rapid, linear increase in muscular adaptations upon initiating a resistance training program in previously untrained women. Acknowledgments We wish to thank Carson Maher and Travis Newcomb for their help. Grant funding for this study was provided by the National Strength and Conditioning Association Foundation. The authors declare that they have no conflicts of interest. References Aagaard, P., Simonsen, E. B., Andersen, J. L., Magnusson, P., & Dyhre-Poulsen, P. (2002). Increased rate of force development and neural drive of human skeletal muscle following resistance training. Journal of Applied Physiology, 93, 1318–1326. American College of Sports Medicine (2009). American College of Sports Medicine position stand. Progression models in resistance training for healthy adults. Medicine and Science in Sports and Exercise, 41, 687–708. Ando, R., Taniguchi, K., Saito, A., Fujimiya, M., Katayose, M., & Akima, H. (2014). Validity of fascicle length estimation in the vastus lateralis and vastus intermedius using ultrasonography. Journal of Electromyography and Kinesiology, 24, 214–220. Atkinson, G., & Batterham, A. M. (2015). True and false interindividual differences in the physiological response to an intervention. Experimental Physiology, 100, 577–588. Baechle, T., & Earle, R. W. (2008). Essentials of strength training and conditioning (third ed.). Champaign, IL: Human Kinetics.

22

M.S. Stock et al. / Human Movement Science 45 (2016) 7–22

Baroni, B. M., Geremia, J. M., Rodrigues, R., De Azevedo Franke, R., Karamanidis, K., & Vaz, M. A. (2013). Muscle architecture adaptations to knee extensor eccentric training: Rectus femoris vs. vastus lateralis. Muscle and Nerve, 48, 498–506. Beck, T. W., Housh, T. J., Johnson, G. O., Weir, J. P., Cramer, J. T., Coburn, J. W., ... Mielke, M. (2007). Effects of two days of isokinetic training on strength and electromyographic amplitude in the agonist and antagonist muscles. The Journal of Strength & Conditioning Research, 21, 757–762. Berg, H. E., Tedner, B., & Tesch, P. A. (1993). Changes in lower limb muscle cross-sectional area and tissue fluid volume after transition from standing to supine. Acta Physiologica Scandinavica, 148, 379–385. Boone, C. H., Stout, J. R., Beyer, K. S., Fukuda, D. H., & Hoffman, J. R. (2015). Muscle strength and hypertrophy occur independently of protein supplementation during short-term resistance training in untrained men. Applied Physiology, Nutrition and Metabolism, 40, 797–802. Carolan, B., & Cafarelli, E. (1992). Adaptations in coactivation after isometric resistance training. Journal of Applied Physiology, 73, 911–917. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (second ed.). Hillsdale, NJ: Lawrence Erlbaum Associates. DeFreitas, J. M., Beck, T. W., Stock, M. S., Dillon, M. A., & Kasishke, P. R. 2nd, (2011). An examination of the time course of training-induced skeletal muscle hypertrophy. European Journal of Applied Physiology, 111, 2785–2790. Fleck, S. J., & Kraemer, W. J. (2004). Designing resistance training programs (third ed.). Champaign, IL: Human Kinetics. Hakkinen, K., Kallinen, M., Izquierdo, M., Jokelainen, K., Lassila, H., Malkia, E., ... Alen, M. (1998). Changes in agonist–antagonist EMG, muscle CSA, and force during strength training in middle-aged and older people. Journal of Applied Physiology, 84, 1341–1349. Hakkinen, K., & Komi, P. V. (1983). Electromyographic changes during strength training and detraining. Medicine and Science in Sports and Exercise, 15, 455–460. Hermens, H. J., Freriks, B., Disselhorst-Klug, C., & Rau, G. (2000). Development of recommendations for SEMG sensors and sensor placement procedures. Journal of Electromyography and Kinesiology, 10, 361–374. Holtermann, A., Roeleveld, K., Vereijken, B., & Ettema, G. (2005). Changes in agonist EMG activation level during MVC cannot explain early strength improvement. European Journal of Applied Physiology, 94, 593–601. Hopkins, W. G. (2015). Individual responses made easy. Journal of Applied Physiology, 118, 1444–1446. Hutchesson, M. J., Rollo, M. E., Callister, R., & Collins, C. E. (2015). Self-monitoring of dietary intake by young women: Online food records completed on computer or smartphone are as accurate as paper-based food records but more acceptable. Journal of the Academy of Nutrition and Dietetics, 115, 87–94. Moore, D. R., Young, M., & Phillips, S. M. (2012). Similar increases in muscle size and strength in young men after training with maximal shortening or lengthening contractions when matched for total work. European Journal of Applied Physiology, 112, 1587–1592. Moritani, T., & deVries, H. A. (1979). Neural factors versus hypertrophy in the time course of muscle strength gain. American Journal of Physical Medicine, 58, 115–130. Nana, A., Slater, G. J., Hopkins, W. G., & Burke, L. M. (2012). Effects of daily activities on dual-energy X-ray absorptiometry measurements of body composition in active people. Medicine and Science in Sports and Exercise, 44, 180–189. Nana, A., Slater, G. J., Stewart, A. D., & Burke, L. M. (2015). Methodology review: Using dual-energy X-ray absorptiometry (DXA) for the assessment of body composition in athletes and active people. International Journal of Sport Nutrition and Exercise Metabolism, 25, 198–215. Nichols, A. W., Hetzler, R. K., Villanueva, R. J., Stickley, C. D., & Kimura, I. F. (2008). Effects of combination oral contraceptives on strength development in women athletes. The Journal of Strength & Conditioning Research, 22, 1625–1632. Nosaka, K., Sakamoto, K., Newton, M., & Sacco, P. (2001). The repeated bout effect of reduced-load eccentric exercise on elbow flexor muscle damage. European Journal of Applied Physiology, 85, 34–40. Rana, S. R., Chleboun, G. S., Gilders, R. M., Hagerman, F. C., Herman, J. R., Hikida, R. S., ... Toma, K. (2008). Comparison of early phase adaptations for traditional strength and endurance, and low velocity resistance training programs in college-aged women. The Journal of Strength & Conditioning Research, 22, 119–127. Rutherford, O. M., & Jones, D. A. (1986). The role of learning and coordination in strength training. European Journal of Applied Physiology and Occupational Physiology, 55, 100–105. Schoenfeld, B. J., Aragon, A. A., Wilborn, C. D., Krieger, J. W., & Sonmez, G. T. (2014). Body composition changes associated with fasted versus non-fasted aerobic exercise. Journal of the International Society of Sports Nutrition, 11, 54. Seynnes, O. R., de Boer, M., & Narici, M. V. (2007). Early skeletal muscle hypertrophy and architectural changes in response to high-intensity resistance training. Journal of Applied Physiology, 102, 368–373. Staron, R. S., Karapondo, D. L., Kraemer, W. J., Fry, A. C., Gordon, S. E., Falkel, J. E., ... Hikida, R. S. (1994). Skeletal muscle adaptations during early phase of heavy-resistance training in men and women. Journal of Applied Physiology, 76, 1247–1255. Stock, M. S., & Thompson, B. J. (2014). Sex comparisons of strength and coactivation following ten weeks of deadlift training. Journal of Musculoskeletal and Neuronal Interactions, 14, 387–397. Vila-Cha, C., Falla, D., & Farina, D. (2010). Motor unit behavior during submaximal contractions following six weeks of either endurance or strength training. Journal of Applied Physiology, 109, 1455–1466. Warren, G. L., Lowe, D. A., & Armstrong, R. B. (1999). Measurement tools used in the study of eccentric contraction-induced injury. Sports Medicine (Auckland, New Zealand), 27, 43–59. Weir, J. P. (2005). Quantifying test–retest reliability using the intraclass correlation coefficient and the SEM. The Journal of Strength & Conditioning Research, 19, 231–240. Weissgerber, T. L., Milic, N. M., Winham, S. J., & Garovic, V. D. (2015). Beyond bar and line graphs: Time for a new data presentation paradigm. PLoS Biology, 13, e1002128. Wells, A. J., Fukuda, D. H., Hoffman, J. R., Gonzalez, A. M., Jajtner, A. R., Townsend, J. R., ... Stout, J. R. (2014). Vastus lateralis exhibits non-homogenous adaptation to resistance training. Muscle and Nerve, 50, 785–793. Yin Lau, W., Blazevich, A. J., Newton, M. J., Xuan Wu, S. S., & Nosaka, K. (2015). Reduced muscle lengthening during eccentric contractions as a mechanism underpinning the repeated bout effect. American Journal of Physiology: Regulatory, Integrative and Comparative Physiology, 15, R879–R886.