Dataset_Randomized Controlled Trial: Effect of isokinetic eccentric resistance training on strength, flexibility and muscle structure for the shoulder external rotator cuff muscles in male junior handball players

1. MethodsThis study is a randomized-controlled trial with an exercise-based intervention. Five days before and after the last training session, functional and structural parameters were examined for the dominant throwing shoulder using mDTI and an isokinetic dynamometer as described below. The participants were recruited locally in German handball pre-season in August 2022. The study can be found in the German Register for Clinical Trials (DRKS00028885). Study protocol was approved by the Leipzig University ethics committee (ethical approval number: 362/21-ek). The study was conducted in accordance with the relevant guidelines and regulations. Written informed consent was obtained from all subjects.1.1 SampleThe sample consisted of 29 professional male junior handball players (German male handball division A and U23) from a local elite handball club (18.0 ± 1.6 years; 186.8 ± 6.3 cm; 84.8 ± 11.3 kg). The participants were randomly assigned to one of the two groups according to their playing position. Participants had to show a healthy throwing shoulder. Therefore, exclusion criteria were regular medication that may affect the adaptability, muscle-nerve diseases, discomfort, pain or known lesions of the right shoulder. Further, they were told to stop with unregular physical activity or heavy training five days prior to data collection.1.2 Intervention group eccentric exerciseAltogether, 12 eccentric isokinetic strength training sessions for the external rotators were conducted within six weeks (twice weekly). The training was performed on a BTE Primus RS isokinetic dynamometer (Baltimore Therapeutic Equipment Company, Hanover, MD, USA). All subjects completed two familiarization sessions before intervention phase. The eccentric isokinetic training was incorporated into the regular athletic training routine, but replaced specific preventive shoulder exercises performed by the active control group. The eccentric training protocol was standardized based on relevant training determinants⁵⁵ as follows: 5x10 repetitions with 90 seconds rest in between, ±10 % load magnitude based on sets, four seconds of time under tension per repetition, 120° ROM, 30°/s movement velocity, supine position with 90° elbow flexion and neutral hand position, 48-76 hours rest between training sessions. After three weeks of intervention, the total ROM was shifted approximately 15° toward the internal ROM and the load magnitude was increased based on the load magnitude of the last three sessions.1.3 Control Group conventional trainingThe active control group performed common preventive exercises in athletic training as recommended by the German Handball Federation in the athletics concept⁵⁶ and shoulder MAPS- program.⁵⁷ Depending on the needs of the local handball club, the prevention exercises included different free-weight rotator cuff exercises such as shoulder abduction, front rowing, and shoulder rotation. The training load was adapted and equalized to the eccentric training exercise in terms of weeks and days of training, and sets and repetitions in each session.1.4 Isokinetic TestingFlexibility and strength tests were performed prior to MRI and on the IsoMed 2000 isokinetic dynamometer (D&amp;R Ferstl GmbH, Hemau, Germany) for the dominant throwing shoulder. As described for eccentric training, subjects were tested in the supine position with fixed shoulder-arm joints (90° shoulder abduction, 90° elbow flexion). To prevent shoulder elevation, the shoulder was also fixed ventrally. In terms of order, a 10-minute warm-up on the rowing machine was performed first. Then the isokinetic diagnostics began with active and passive stretching tests, followed by strength tests at 60°/s and then at 30°/s isokinetic velocity, always separated for concentric and eccentric movement. One minute of rest was allowed between tests.The active stretch test consisted of two familiarization trials and five consecutive test trials that dynamically alternated between internal and external rotation. When the subject was unable to continue the antagonist stretch produced by the agonist muscles, the maximum active ROM and the turning point were reached. Passive tests were performed at an isokinetic speed of 10°/s. The passive stretch consisted of ten trials of alternating internal and external rotation with the eyes closed. Maximum ROM was reached when subjects either reached 9 Nm or 100° of internal rotation and 140° of external rotation.The maximal strength tests were performed separately for concentric and eccentric, also alternating between internal and external rotation. After two familiarization trials, three repetitions were performed over an amplitude of 150° (60° internal rotation and 90° external rotation) at an isokinetic velocity of 60°/s and then 30°/s. All tests were performed in the same order and a 90s rest in-between. The position of each subject , the dynamometer arm and gravity correction as well as the settings were individualized at baseline and used for posttests. The data were recorded on the isokinetic dynamometer with a recording frequency of 200 Hz.1.5 Diffusion Tensor ImagingIn supine position MRI scans were performed using a 3-Tesla Siemens MAGNETOM Prisma scanner (Siemens Healthcare, Erlangen, Germany) and a shoulder coil (XL, 16-channel). Commercial T1-weighted (T1w) and diffusion-weighted imaging sequences (DWI) were used. The total scan time was twelve minutes. The volunteers lay in a head first supine position with the right arm in the neutral position and the hand supinated. For the T1w: sequences the following settings were acquired: repetition and echo time TR/TE = 492/20 ms, slice thickness = 0.7 mm, flip angle = 120°, field of view = 180 x 180 mm², matrix = 256x256 mm². Also, a commercial Siemens 2D echo planar DWI sequence was acquired with the following standard parameter settings: repetition and echo time TR/TE = 6100/69 ms, slice thickness = 4 mm, flip angle = 90°, field of view = 240 x 240 mm², matrix = 122 x 122 mm², 48 diffusion sampling directions with b = 400 s/mm².1.6 Data processingData from the functional tests using the isokinetic dynamometer, were performed using MATLAB v.R2022a (MathWorks. Natick. USA). Raw-data were filtered using a fifth order, zero-lag Butterworth low-pass filter at a cut-off frequency of 6 Hz.⁵⁸ Afterwards, the torque and angle data were cut according to repetition number and movement direction (internal or external rotation). From each repetition the acceleration and deceleration phase were cut leaving only the range with desired isokinetic speed. The maximum and average torque were calculated from the resulting torque-angle curves for each movement direction (internal/external rotation) and contraction mode (eccentric/ concentric) per participant. For further analysis, the torque and angle data were normalized to body mass.^59–61 In order to compare not only parametrized absolute values from the torque-angle curve, but also the curve characteristics between pre- and posttest, the torque and angle data were interpolated to 101 samples and 1SPM vector field analysis were performed.⁶² Flexibility analysis were based on the average the five trials of each active and passive stretching test. For these trials, a three-parametric e-function was fitted. This enabled analyses for submaximal ROM and maximal ROM. 1SPM was also used to analyze the passive torque-angle curve and its morphology changes over time. More details on the method and data processing can be found in the preliminary study to this paper.⁶³MRI data were processed using Mimics Materialise v.24.0 (Leuven, Belgium) for prior muscle segmentation and DSI Studio (v. 3 December 2021. http://dsi-studio.labsolver.org) for muscle tractography. The T1w images from the MRI were used for manual segmentation of the supraspinatus and infraspinatus muscles. After extraction of the segmented muscle volume of interest, the DWI were corrected for motion and eddy current distortion using DSI Studio's integrated FSL eddy current correction. The DWI were visually inspected by two independent raters. If artifacts or field of view did not allow full muscle analysis, subjects were excluded. Tractography was performed using specific stopping criteria defined in previous work.⁵³The deterministic fiber tracking is based on an RK4 algorithm. After tractography, DSI Studio was used to calculate the muscle FL and the FV of all muscle fascicles using the integrated statistics tool. This method is leaned on Vetter and colleagues.⁵³<br>