Laboratorijska študija / Laboratory study M • • • v • a • a • m* v • Primerjava največjega navora med izokmeticmma dinamometroma SMM iMoment in Biodex System Pro 4 Peak Torque Comparison between SMM iMoment and Biodex System Pro 4 Isokinetic Dynamometers Avtor / Author Matjaž Vogrin1,2,3, Teja Ličen1, Nina Greiner1, Miloš Kalc1 Ustanova / Institute "'Univerza v Mariboru, Medicinska fakulteta, Inštitut za športno medicino, Maribor, Slovenija; 2Univerzitetni klinični center Maribor, Klinika za kirurgijo, Oddelek za ortopedijo, Maribor, Slovenija; 3Univerza v Mariboru, Medicinska fakulteta, Katedra za ortopedijo, Maribor, Slovenija; 'University of Maribor, Faculty of Medicine, Institute of Sports Medicine Maribor, Slovenia; 2University Medical Centre Maribor, Division of Surgery, Department of Orthopaedics, Maribor, Slovenia; 3University of Maribor, Faculty of Medicine, Department of Orthopedics, Maribor, Slovenia; Ključne besede: zanesljivost, iztegovalke kolena, upogibalke kolena, kot pr največjem navoru, največji navor Key words: Inter-machine reliability, Knee extensors, Knee flexors, Angle of peak torque, Peak torque Članek prispel / Received 1. 7. 2019 Članek sprejet / Accepted 17. 9. 2020 Naslov za dopisovanje / Correspondence Miloš Kalc, Univerza v Mariboru, Medicinska fakulteta, Inštitut za športno medicino, Taborska ulica 8, 2000 Maribor, Slovenija Telefon +386 40680264 E-pošta: milos.kalc@ism-mb.si Izvleček Namen: Primerjati meritev največjih navorov pri koncentričnem mišičnem krčenju iztega in upogiba kolena med izokinetičnima dinamometroma SMM iMoment in Biodex System Pro 4. Metode: Petindvajset študentov (14 moških, 11 žensk, starost 24.35 (1.41) let) je sodelovalo v študiji. Na dveh loče-nih obiskih smo izmerili največji navor (PT) in kot pri največjem navoru (APT) iztegovalk in upogibalk kolenskega sklepa pri hitrosti 60 °/s and 180 °/s. Parni t-test, medrazredni korelacijski koeficient (ICC), standardna napaka meritve (SEM), koeficient variacije (CVmed) in Bland Altman metodo smo uporabili za preverjanje razlik, relativne in absolutne zanesljivosti, ter ujemanje meritev med napravami. Rezultati: Zaznali smo statistično značilne razlike med napravama v PT med upogibom kolena pri hitrosti 60 °/s (p = 0.027). ICC vrednosti so pokazale viso- Abstract Purpose: To compare peak torque of knee extensors and flexors during muscle contractions between the SMM iMoment and Biodex System Pro 4 isokinetic dynamometers (inter-machine reliability). Methods: Twenty-five students (14 men and 11 women, age 24.35 (1.41) years) took part in a crossover study. Peak torque (PT) and angle of peak torque (APT) for knee extensors and flexors were assessed at the velocity of 60°/s and 180°/s on two visits. Paired t-test, intraclass correlation coefficient (ICC), standard error of measurement (SEM), coefficient of variation (CV-med) and Bland-Altman plots were used to determine significant differences, relative and absolute reliability and agreement between devices. Results: There were statistically significant differences in PT between machines at the velocity of 60°/s for the right i acta medico-biotechnica 2020; 13 (2): 46-54 Laboratorijska študija / Laboratory study ko do zelo visoko relativno zanesljivost pri nalogah izvedenih z levo nogo (0.84 - 0.93) in srednjo do zelo visoko zanesljivost pri nalogah izvedehih z desno nogo (0.79 - 0.91). V primerjavi z drugimi sorodnimi študijami, so vrednosti absolutne zanesljivosti razkrile večjo variabilnost podatkov (CVmed: 4.84% - 9.36%, SEM: 9.62 Nm - 26.11 Nm). Bland Altman metoda je pokazala dobro primerljivost pri nalogah izvedenih z levo nogo in rahlo slabšo primerljivost pri nalogah izvedenih z desno nogo (4-6 Nm). Ugotovili smo statistično značilne razlike med napravama v APT pri vseh pogojih merjenja. Zaključek: Vrednosti največjega navora mišic iztegovalk in upogibalk kolena so bile podobne na obeh napravah. Poudariti gre večjo razpršenost rezultatov izmerjenih s pomočjo naprave iMoment. leg knee flexion (p = 0.027). ICC values revealed high to very high relative reliability for the left leg (0.84 - 0.93) and moderate to very high for the right leg (0.79 - 0.91). Values of absolute reliability revealed more substantial differences between dynamometers compared to other similar studies (CVmed: 4.84% - 9.36%, SEM: 9.62 Nm - 26.11 Nm). Bland-Altman plots revealed no specific bias in exercise performed with the left leg and small bias (4-6 Nm) in exercises performed with the right leg. There were significant differences in APT between dynamometers in all conditions. Conclusion: Mean peak torque values of knee extensors and flexors were similar between machines; however, data collected using the iMoment dynamometer were substantially more variable. INTRODUCTION Isokinetic dynamometry is a widely accepted procedure in clinical, rehabilitation and research environments. The isokinetic device allows the assessment of joint and muscle maximal eccentric (ECC), concentric (CON) and isometric (ISO) strength under controlled constant velocities throughout specified range of motion. Even though dynamometers are developed to measure muscle strength in several joints (e.g., shoulder, torso, knee, ankle, etc.), the most measured muscles in practice, are the knee extensor and flexor muscles (1). As well as measuring muscle strength, the isokinetic dynamometers allow us to conduct a proper progressive rehabilitation protocol, such as hamstring and knee injuries (2). It is a recognised diagnostic tool that helps make more accurate decisions about patient progress, readiness and return-to-play evaluation (3). Some authors have suggested that preseason screening of unilateral and bilateral strength imbalance in healthy athletes can allow medical staff to identify the ones at a higher risk of incurring lower limb injuries during training or competition (4). Because there are many isokinetic machines present on the market, it is critical to establish the validity and reliability of these measuring devices. This allows confident assessment of the instrument output data in clinical and experimental environments. Existing devices are regularly updated and upgraded in the isokinetic market. New devices are also introduced. So far, the market is led mainly by the Biodex brand, which has a long history of producing valid and reliable isokinetic dynamometers and is considered a gold standard (5,6). However, other brands can produce reproducible and valid dynamometers like Cybex Norm (now Humac Norm), IsoMed 2000, REV9000 and iSAM 9000 (7, 8, 2, 9). In general, the results of these studies indicate good to excellent intra-machine reliability (repeatability of data within machine) and inter-machine (comparison of data between machines) reliability for measuring maximum knee strength performance (8). A newer, less known isokinetic dynamometer on the market is the iMoment dynamometer. The SMM Company (SMM d. o. o., Maribor, Slovenia) has developed the device in a joint venture with members of the Faculty of Sport at the University of Ljubljana. At the time of writing this paper, only two working prototypes have so far been developed and extensively tested at two important locations in Slovenia (Faculty of Sport, Ljubljana and Faculty of Medicine, Maribor). The manufacturer wanted to develop a modern, robust and reliable device. Excellent intra-machine repeatability and reliability for the iMoment peak torque has already been proven (10). Even though the device has been developed as a versatile platform capable of measuring i acta medico-biotechnica 2020; 13 (2): 46-54 Laboratorijska študija / Laboratory study muscle torque in several joints and body positions, only the knee flexion-extension and shoulder internal-external rotation measurements were fully functional at the time of writing this article. Machine settings (sitting position, dynamometer positioning and lever arm velocity) are fully automatized using controlled electric motors. This can be user-controlled on the main console of the software. It also allows storing patient positions and protocol settings (pictures of both dynamometers in Figure 1). This study aimed to compare peak torque measurements (inter-machine reliability) between the SMM iMoment and the Biodex System Pro 4 isokinetic dynamometers for knee extensor and flexors. The tests were conducted at velocities of 60°/s and 180°/s during concentric muscle contractions. Figure 1. Pictures of both machines and differences between fixation systems: a) Biodex System Pro 4, a1) chest fixation, a2) pelvis fixation, a3) thigh fixation, a4) lever arm fixation and b) SMM iMoment, b1) torso fixation, b2) thigh fixation, b3) lever arm fixation. MATERIALS AND METHODS Twenty-five healthy students (14 men and 11 women, age 24.35 (1.41) years) attending the Faculty of Medicine at the University of Maribor without any known history in locomotor or nerve injuries were recruited for the study. The volunteers were acquainted with the experimental procedure and their voluntary cooperation was confirmed by written consent on the day of the first testing. Subjects were instructed to maintain their daily exercise routine but were asked not to perform vigorous exercise 48 hours before the testing session. All procedures were in accordance with the latest version of the Helsinki declaration and were approved by the Research Ethics Committee at the University Medical Centre Maribor (UKC-MB-KME-14/19). Two isokinetic dynamometers were used in this study: iMoment isokinetic device and the System Pro 4 (Biodex Medical Systems, Shirley, NY, USA). The inter-machine reliability was assessed by performing knee extension and flexion exercises on both legs at separate velocities of 60°/s and 180°/s. A crossover experimental design was constructed to allow assessment of inter-machine reliability. All the subjects repeated the measurements twice, on two different visits. Each visit comprised the assessment on one machine (iMoment and Biodex respectively). There were at least 7 days rest between visits in order to minimise potential learning effects, fatiguing or potentiation effects of the neuromuscular system. The order of devices was randomised across subjects. Before data collection, the machines were calibrated according to the manufacturer instructions. The Biodex dynamometer calibration is automatic during machine start up. The calibration procedure for the iMoment dynamometer requires manual operation. The iMoment dynamometer was manually calibrated at the beginning of each measuring day. At the beginning of each visit, the subjects carried out a standardised 10-minute warm-up which consisted of stepping on a 30 cm high box at a frequency of 0.5 Hz. After the warm-up, the first of the two legs were selected and the subjects were positioned on the isokinetic device. The positioning was carefully carried out following the manufacturer instructions, that is, the knee joint axis was aligned with the centre of the machine rotational axis. The appropriate hip, thigh and chest straps were tightened in order to secure the body position and to minimise hip and knee movement. Biodex and iMoment fixation straps differ in position, width and fixation mechanism (Figure 1). However, the dynamometer lever arm was fixed at a similar length for both machines. The ankle stabilisation strap was fixed just above the ankle joint (5 cm above the external malleolus). The knee flexion and extension protocols were carried out in 60° range, between 90° and 30° angle (where the fully extended knee represented 0°). i acta medico-biotechnica 2020; 13 (2): 46-54 Laboratorijska študija / Laboratory study After aligning the subject to the correct position, a short familiarisation routine was carried out with the machine. The researcher verbally explained the experimental procedure and an additional warm-up consisting of 15 repetitions of knee extension and flexion at a velocity of 60°/s was conducted asking the subject to progressively increase the torque per repetition. Three maximum concentric contractions were executed in knee extension and flexion at a velocity of 60°/s. After a 2-minute rest, three maximum contractions were then executed in knee extension and flexion at a velocity of 180°/s. After a five-minute break, the subjects were repositioned on the machine to repeat the same procedure on the opposite leg. The experiment was carried out by an experienced researcher, who has taken more than 300 measurements on each machine. The gravitational moment (GET) was determined by the iMoment and Biodex start up procedure by measuring the leg weight at a position of 30° knee extension. Torque data were automatically GET corrected by individual machine software and used in further analysis. The repetition with the highest PT from each velocity (60°/s and 180°/s) and leg (left and right) were further analysed. Data were checked for normality of distribution using the Shapiro-Wilk test and by visual inspection of Q -Q plots. Paired Student's t-tests (Biodex vs iMoment) were used to assess statistical differences in PT and angle of peak torque (APT) between machines. All statistical analyses were performed using the R statistical software (11). P-values lower than 0.05 were considered statistically significant. The relative magnitude of the association between measures was assessed using the intraclass correlation coefficient (ICC 2, 1). To the best of our knowledge there is no consensus for which correlation values for use in reliability studies (12). For this reason, we adopted the classification proposed by Sole et al., in which correlations of 0.50 - 0.69 are "moderate", 0.70 - 0.89 are "high" and higher than 0.90 are "very high" (13). Absolute reliability (variability of the score from trial to trial in original units) was assessed using the standard error of measurement (SEM) and the coefficient of variability as proposed by Bardis et al. (1):The coefficient of variation was calculated by the following formula: where di is the difference between two results in each individual and xi is the mean of the two results. Median (CVmed), 10th and 90th percentiles CV% were calculated in order to obtain the coefficient of variation 80% central range (CV80%). CVmed and CV80% represents the central range to give information about the distribution of the variation (14). The standard error of measurement (SEM) was calculated with the formula: CV% = 100^(1/2 )di2 xi, The SEM is a calculation of how much measured test scores are spread around a "true" score, and it uses the same units as the test (peak torque in Nm in this study) (16). Bland-Altman plots and "lines of equality" were visually inspected and evaluated in order to define the agreement between the two measurements (15). Table 1. Mean and standard deviation of peak torque (Nm) Peak Toique Nm (SD) Left Peak Toique Nm (SD) Right Biodex iMoment P Biodex iMoment P 60% Extension 192.1 (55.8) 194.8 (61.6) 0.835 191.3 (48.5) 189.2 (59.5) 0.081 Flexion 100.9 (29.7) 101.6 (34.7) 0.191 100.8 (28.8) 99.9 (28.9) 0.027* 180°/s Extension 128.4 (45.7) 129.2 (47.4) 0.848 129.5 (43.5) 128.9 (40.9) 0.421 Flexion 71.2 (23.8) 72.8 (24.3) 0.920 72.7 (23.9) 70.9 (19.8) 0.161 *Statistically significant differences. i acta medico-biotechnica 2020; 13 (2): 46-54 Laboratorijska študija / Laboratory study RESULTS Means, standard deviations and paired Student's t-test p - values for knee extension and flexion PT at 60 °/s and 180 °/s velocities are shown in Table 1. There were statistically significant differences in PT between machines at the velocity of 60°/s for the right leg knee flexion exercise (p = 0.027). There were no significant differences in PT between machines at a velocity of 180 °/s. ICCs involving all assessments performed with the left leg were close to or higher than 0.90, indicating a high to very high relative reproducibility (Table 2). ICC coefficients on the right leg were slightly lower (compared to the left leg) for all measurements, reaching only moderate reproducibility in the knee extension at 180°/s velocity (Table 2). Coefficient of variation (CVmed) for the PT parameter of knee extensors and flexors showed moderate differences between dynamometers (see Table 2). Table 2. Relative (ICC) and absolute (CV and SEM) reliability for peak torque data Peak Toique Left Peak Toique Right ICC (95% CI) CVmed (CV80%) SEM ICC (95% CI) CVmed (CV80%) SEM 60% Extension 0.835 (0.722 - 0.930) 9.36 (1.86 - 19.80) 23.70 0.847 (0.700 - 0.925) 6.79 (0.99 - 18.60) 26.11 Flexion 0.929 (0.778 - 0.945) 7.01 (1.86 - 15.10) 10.44 0.911 (0.805 - 0.959) 7.40 (1.17 - 13.30) 9.62 180°/s Extension 0.859 (0.681 - 0.918) 6.69 (1.24 - 23.20) 18.18 0.788 (0.600 - 0.894) 4.84 (1.04 - 27.30) 24.52 Flexion 0.889 (0.856 - 0.966) 8.41 (4.03 - 23.70) 11.31 0.892 (0.785 - 0.948) 6.02 (2.00 - 17.40) 9.70 ICC = intraclass correlation coefficient; CI = confidence interval, CV= coefficient of variation (in %); CV80% = CV 80% range from 10th to 90th percentile; SEM = standard error of measurement (in Nm). A random relationship was observed between the individual differences for Biodex and iMoment assessments, as shown in Bland-Altman plots, which represents the differences against mean calculatins between machines at 60°/ s (left sides of Figure 2 and Figure 3) and 180°/s (left sides of Figure 4 and Figure 5), respectively. There was no explicit bias in PT between exercise performed with the left leg, but small bias was present in exercise performed with the right leg at both velocities (4-6 Nm less for iMoment measurements). In addition, line of equality demonstrates similarities between dynamometers at 60°/s (right sides of Figure 2 and Figure 3) and 180°/s (right sides of Figure 4 and Figure 5), respectively. 50 75 100 125 150 Mean Peak Torque (Nm) Figure 2. Difference against mean for left leg PT at 60°/s for a) knee extension, b) knee flexion; and line of equality between two dynamometers c) knee extension, d) knee flexion (Biodex - iMoment). i acta medico-biotechnica 2020; 13 (2): 46-54 Laboratorijska študija / Laboratory study Figure 3. Difference against mean for right leg PT at 60°/s for a) knee extension, b) knee flexion; and line of equality between two dynamometers c) knee extension, d) knee flexion (Biodex - iMoment). Figure 4. Difference against mean for left leg PT at 180°/s for a) knee extension, b) knee flexion; and line of equality between two dynamometers c) knee extension, d) knee flexion (Biodex - iMoment). 100 Mean b c 20e E ------ 0 1 10. •