Inconsistency in shoulder arthrometers for measuring glenohumeral joint laxity: a systematic review

Citation: Gomes, E.; Andrade, R.; Valente, C.; Santos, J.V.; Nunes, J.; Carvalho, Ó.; Correlo, V.M.; Silva, F.S.; Oliveira, J.M.; Reis, R.L.; et al. Inconsistency in Shoulder Arthrometers for Measuring Glenohumeral Joint Laxity: A Systematic Review. Bioengineering 2023,10, 799. https://doi.org/ 10.3390/bioengineering10070799 Academic Editors: Cheng-Kung Cheng, Huiwu Li and Huizhi Wang Received: 2 June 2023 Revised: 27 June 2023 Accepted: 28 June 2023 Published: 4 July 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). bioengineering Systematic Review Inconsistency in Shoulder Arthrometers for Measuring Glenohumeral Joint Laxity: A Systematic Review Eluana Gomes 1, Renato Andrade 1,2,3, Cristina Valente 1,2, J. Victor Santos 4, Jóni Nunes 1,5,6, Óscar Carvalho 4,7 , Vitor M. Correlo 8,9,10 , Filipe S. Silva 4,7, J. Miguel Oliveira 8,9, Rui L. Reis 8,9 and João Espregueira-Mendes 1,2,6,8,9,* 1Clínica Espregueira—FIFA Medical Centre of Excellence, 4350-415 Porto, Portugal 2Dom Henrique Research Centre, 4350-415 Porto, Portugal 3 Porto Biomechanics Laboratory (LABIOMEP), Faculty of Sports, University of Porto, 4200-450 Porto, Portugal 4Centre for Microelectromechanical Systems (CMEMS-UMINHO), Campus Azurém, University of Minho, 4800-058 Guimarães, Portugal 5Serviço de Ortopedia e Traumatologia do Hospital de Santa Maria Maior, 4750-333 Barcelos, Portugal 6School of Medicine, University of Minho, 4710-057 Braga, Portugal 7LABBELS Associate Laboratory, University of Minho, 4800-058 Guimarães, Portugal 8ICVS/3B’s–PT Government Associate Laboratory, 4805-017 Guimarães, Portugal 93B’s Research Group, I3Bs—Research Institute on Biomaterials, Biodegradables and Biomimetics Headquarters of the European Institute of Excellence on Tissue Engineering and Regenerative Medicine, AvePark, Parque de Ciência E Tecnologia, University of Minho, Zona Industrial da Gandra, Barco, 4805-017 Guimarães, Portugal 10 Pro2B, Consultoria e Gestão de Projetos, AvePark—Parque de Ciência e Tecnologia, Zona Industrial da Gandra, Barco, 4805-017 Guimarães, Portugal *Correspondence: [email protected] Abstract: There is no consensus on how to measure shoulder joint laxity and results reported in the literature are not well systematized for the available shoulder arthrometer devices. This systematic review aims to summarize the results of currently available shoulder arthrometers for measuring glenohumeral laxity in individuals with healthy or injured shoulders. Searches were conducted on the PubMed, EMBASE, and Web of Science databases to identify studies that measure glenohumeral laxity with arthrometer-assisted assessment. The mean and standard deviations of the laxity measurement from each study were compared based on the type of population and arthrometer used. Data were organized according to the testing characteristics. A total of 23 studies were included and comprised 1162 shoulders. Populations were divided into 401 healthy individuals, 278 athletes with asymptomatic shoulder, and 134 individuals with symptomatic shoulder. Sensors were the most used method for measuring glenohumeral laxity and stiffness. Most arthrometers applied an external force to the humeral head or superior humerus by a manual-assisted mechanism. Glenohumeral laxity and stiffness were mostly assessed in the sagittal plane. There is substantial heterogeneity in glenohumeral laxity values that is mostly related to the arthrometer used and the testing conditions. This variability can lead to inconsistent results and influence the diagnosis and treatment decision-making. Keywords: shoulder; glenohumeral; arthrometer; laxity; stiffness 1. Introduction Every year, shoulder dislocations occur on an average of 5–40 per 100,000 individuals in the general population [ 1 – 5 ]. Athletes and those that engage in sports activities are more prone to shoulder dislocation [ 6 – 10 ]. This is due to both the high repetitive loads involved in sports and traumatic events. A shoulder dislocation can damage the joint stabilizers and cause laxity. After a first dislocation, these individuals are more prone to redislocation events [ 4 , 11 ]. After repeated episodes of dislocation, these individuals can develop joint instability [ 4 , 12 ], which can result in persistent long-term pain and functional limitations. Bioengineering 2023,10, 799. https://doi.org/10.3390/bioengineering10070799 https://www.mdpi.com/journal/bioengineering Bioengineering 2023,10, 799 2 of 16 Therefore, an early diagnosis of shoulder instability is crucial in order to implement early treatment and secondary prevention strategies. The diagnosis of shoulder instability is mostly based on clinical history and manual glenohumeral (GH) laxity tests [ 13 ] that are combined with imaging scans to assess the structural integrity of musculoskeletal structures [ 14 ]. Despite manual tests (apprehension and relocation tests) displaying high specificity, they only show suboptimal sensitivity, thus being non-optimal for identifying those with shoulder instability with high diagnostic accuracy [ 15 , 16 ]. Moreover, manual tests result in variable findings (e.g., inter-rater reliability) dependent on the experience, skill, and sensibility of the examiner. These tests can only subjectively evaluate the degree of shoulder instability and are unable to provide an accurate measurement of joint laxity. To overcome the limitations of validity and replicability of manual laxity testing, a manifold of arthrometers have been developed to measure GH laxity. Joint arthrometers apply an external force to the joint with the aim of emulating the manual testing. Shoulder arthrometers are similar in concept to other arthrometers [ 16 ] for different joints that are already on the market (e.g., Telos, KT-1000/200). These devices can provide objective quantification and precise measurements of joint laxity leading to a more accurate diagnosis. Notwithstanding, the structure and use of these arthrometers can become heterogenous, i.e., inconsistent force application and patient positioning, and may lead to variable results and inconclusive findings. The use of arthrometers is important for clinical practice to provide a more precise estimate of GH joint laxity and to diagnose the presence and severity of shoulder instability. Clinical and arthrometric data on GH laxity would help clinicians to reach more accurate diagnosis and make an informed and adequate treatment planning (either conservative or surgical interventions). Scientific literature reporting shoulder arthrometry is still sparce and scattered, and there is no available source that systematizes these data, which can lead to inconsistent implementation of shoulder arthrometry and misinterpretation of their results. There is thus a clear need to systematize the scientific literature of the results of shoulder arthrometers for measuring GH laxity for consistent and reliable use of shoulder arthrometry. Our goal was to systematize the results of currently available shoulder arthrometers for measuring GH laxity in individuals with a healthy or injured shoulder. The purpose of this systematic review was thus to summarize state-ofthe-art of shoulder laxity measurement using arthrometry and to compare the results across the different available devices and between injured and healthy shoulders. This summary of current evidence can guide researchers and clinicians on how to use arthrometers for measuring shoulder laxity and compare their results with available data according to the different characteristics of the shoulder condition(s), arthrometer used, and method of measurement. 2. Materials and Methods This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement [ 17 ]. The protocol was a priori registered in the PROSPERO database under the number CRD42023404088. 2.1. Eligibility Criteria The eligibility criteria are framed according to the Participants, Intervention/Exposure, Comparison, Outcome and Study design (PICOS) strategy. As the eligible population, we included studies comprising both male and female individuals that were healthy (asymptomatic) and those presenting symptomatic shoulder conditions (shoulder pain and shoulder instability, among other conditions). Studies including individuals after shoulder stabilization surgery or rehabilitation were deemed eligible if they had the outcome of interest (GH laxity). Under exposure and the outcome of interest, studies that made an arthrometer-assisted measurement of GH laxity were also considered. GH laxity was defined as the objective quantification of joint displacement (mm) or joint stiffness (N/mm). Arthrometers must in- Bioengineering 2023,10, 799 3 of 16 clude a system that measures joint laxity (either visually, as an external connected device or software, or with concomitant imaging methods). The external application of load had to be applied locally at the shoulder joint by a direct mechanical actuator (activated manually or by an electrical system). Studies where the load was applied via free weights with a pulley system were excluded. A comparator group of exposure was not compulsory, but studies comparing shoulder arthrometry with other methods were included. Studies involving different methods under the implementation of shoulder arthrometer evaluation (e.g., different populations, patient positioning, applied loads, and methods of measurement) were also considered. We included all laboratory or clinical trials (from randomized controlled trials to case series) that allowed for the evaluation of the outcome of interest. Letters, editorials, conference abstracts, cadaveric and animal studies, case studies, commentaries, and reviews were excluded. Due to unavailable translation resources, only included studies written in English were analyzed. 2.2. Search Strategy Computerized searches were conducted in the PubMed, EMBASE and Web of Science databases up to 3 May 2023. The full search strategy for each database is reported in Table S1. The reference lists of the relevant reviews and of the included studies were screened for additional potentially eligible studies not identified via the database searches. 2.3. Study Selection All records were exported to EndNote X7 (Thomson and Reuters, Philadelphia, PA, USA); duplicates were removed using the software’s “duplicates” tool and then confirmed manually to check for any missing duplicate records. Two authors (E.G. and R.A.) independently scanned all titles and abstracts, and then revised the full texts of all potentially eligible studies. Disagreements were resolved by a third author (C.V.). 2.4. Data Collection and Extraction All data related to the study characteristics, arthrometers, and their outcomes were extracted in duplicate by two authors (E.G. and R.A.). Disagreements were resolved by consensus. We used an excel spreadsheet to record data, including: (i) Study characteristics (year and region); (ii) characteristics of the included population (number of individuals and shoulders, percentage of males/females, mean age, height, and body weight) and their clinical condition (asymptomatic or symptomatic); (iii) description of the arthrometer and testing conditions (name of the device, method of force application, amount of load, direction of force, patient positioning, shoulder fixation, laxity measurement system, and procedure) and their validity and reliability outcomes; (iv) outcome measures (laxity and stiffness). 2.5. Data Management Studies that included overlapping populations but that presented different outcomes were merged into a cluster of studies. The population characteristics were collected for each study (as means and standard deviations), but then summarized using proportions and pooled means, as well as standard deviations weighted to the sample size. When studies reported data for both shoulders or subgroups by sex, we collected and reported the outcomes of both shoulders/sexes separately (when available). When reported, the mean difference between shoulders was also collected. When studies presented data for the same outcome using different measuring methods (e.g., radiography and ultrasound) and varying loads or different patient positioning, the data from both methods in data synthesis were reported separately. However, when data were reported for the same population under the same testing conditions (e.g., for test–retest purposes) from the same study but from different trial reports, the data were combined using pooled means and standard deviations. Bioengineering 2023,10, 799 4 of 16 2.6. Risk of Bias The risk of bias was judged using the Risk of Bias Assessment tool for Non-randomized Studies (RoBANS) [ 18 ]. The RoBANS is a validated tool to assess the risk of bias of nonrandomized studies, comprising six domains of bias: (i) The selection of participants, (ii) confounding variables, (iii) measurement of exposure, (iv) blinding of outcome assessment, (v) incomplete outcome data, and (vi) selective outcome reporting (Table S2). Each domain was judged as low risk of bias, high risk of bias, or unclear. Risk of bias was judged at the outcome level for laxity and stiffness. Two authors (E.G. and R.A.) made all judgements, and disagreements were resolved by a third author (C.V.). 2.7. Data Synthesis Data were stratified according to population characteristics into three subgroups: (i) Healthy individuals with asymptomatic shoulders, (ii) athletes with asymptomatic shoulders, and (iii) individuals with injured shoulders. Data pooling for joint laxity and stiffness was not attempted due to the wide heterogeneity across the studies’ populations, arthrometers, and testing methods. In addition to stratification based on population characteristics, we also stratified the data based on the device used and we organized data in regard to testing characteristics (e.g., measurement system, amount of load, shoulder positioning, and shoulder being assessed). Laxity data were then plotted into figures for visual display of the anterior (PA), posterior (AP), inferior, and global laxity. Stiffness was not reliable to plot into a figure due to many overlapping slopes, and it was thus reported for each individual study. 3. Results The database and hand-searches yielded 2614 titles and abstracts. After removing the duplicates, the full texts of the 1435 relevant studies were analyzed according to the eligibility criteria. A total of 24 trial reports from 23 studies [ 19 – 41 ] met the eligibility criteria and were included in this systematic review (Figure 1). 3.1. Risk of Bias Nearly one-fourth of the studies (k = 9; 23%) were judged as having a high risk of selection bias due to the selection of participants, including healthy individuals and athletes with asymptomatic shoulders but reporting shoulder pain or a previously diagnosed shoulder injury. More than half of the studies (k = 13; 57%) were judged as having a high risk of selection bias due to uncontrolled confounding variables, mostly due to unreported sex and unbalanced shoulder dominance. Three-quarters of the studies (k = 15; 75%) were judged as having a high risk of performance bias for their measurement of laxity, but none for measuring stiffness. Likewise, most of the studies were also judged as having a high risk of detection bias for their laxity (k = 13; 65%) and stiffness (k = 4; 80%) measurements. Incomplete data outcome was not a concern, with no study being judged as having a high risk of attrition bias. Only one study [ 32 ] was judged as having a high risk of selective reporting, but as no study registered an a priori protocol, this domain should be viewed with some concerns for all studies. The judgment of risk of bias for each included study and domain is provided in Figure S1. 3.2. Population Characteristics A total of 1162 shoulders from 813 individuals with a mean age of 23.0 ± 3.7 years were included for analysis (Table 1). Among them, 401 were voluntary individuals with asymptomatic shoulders, 278 were athletes with asymptomatic shoulders, and 134 were individuals with injured shoulders. Table S3 details the population characteristics for each study. Bioengineering 2023,10, 799 5 of 16 Bioengineering 2023, 10, x FOR PEER REVIEW 5 of 16 Figure 1. PRISMA 2020 flow diagram for new systematic reviews that included searches of databases, registers, and other sources. 3.1. Risk of Bias Nearly one-fourth of the studies (k = 9; 23%) were judged as having a high risk of selection bias due to the selection of participants, including healthy individuals and athletes with asymptomatic shoulders but reporting shoulder pain or a previously diagnosed shoulder injury. More than half of the studies (k = 13; 57%) were judged as having a high risk of selection bias due to uncontrolled confounding variables, mostly due to unreported sex and unbalanced shoulder dominance. Three-quarters of the studies (k = 15; 75%) were judged as having a high risk of performance bias for their measurement of laxity, but none for measuring stiffness. Likewise, most of the studies were also judged as having a high risk of detection bias for their laxity (k = 13; 65%) and stiffness (k = 4; 80%) measurements. Incomplete data outcome was not a concern, with no study being judged as having a high risk of attrition bias. Only one study [32] was judged as having a high risk of selective reporting, but as no study registered an a priori protocol, this domain should be viewed with some concerns for all studies. The judgment of risk of bias for each included study and domain is provided in Figure S1. 3.2. Population Characteristics A total of 1162 shoulders from 813 individuals with a mean age of 23.0 ± 3.7 years were included for analysis (Table 1). Among them, 401 were voluntary individuals with asymptomatic shoulders, 278 were athletes with asymptomatic shoulders, and 134 were individuals with injured shoulders. Table S3details the population characteristics for each study. Figure 1. PRISMA 2020 flow diagram for new systematic reviews that included searches of databases, registers, and other sources. Table 1. Summary of the population characteristics of the included studies. Variable Asymptomatic Shoulders Injured Shoulders Total Sample Healthy Individuals Athletes K (population) 401 278 134 813 N (shoulders) 570 455 137 1162 Sex (M/F) 196/161 103/41 71/36 370/238 Age (years) 21.2 ±7.2 21.5 ±2.9 24.4 ±7.7 23.0 ±3.7 Weight (kg) 72.9 ±6.1 83.6 ±13.2 NR 79.3 ±12.1 Height (cm) 171.7 ±4.3 182.0 ±7.7 NR 177.8 ±8.5 NR: No reported. 3.3. Device Characteristics Six different shoulder arthrometers were reported in the literature (Table S4). The most reported arthrometers were the Telos + LigMaster ™ (six studies) [ 27 , 28 , 30 , 34 – 36 ] and the customized instrumented shoulder arthrometer (six studies) [ 20 – 23 , 39 , 40 ], followed by the Telos (five studies) [ 24 – 26 , 29 , 37 ]. The remaining arthrometers included a shoulder adaptation of the KT-1000/2000 (three studies) [ 31 , 38 , 41 ], the Donjoy ® Laxity Tester (two studies) [ 32 , 33 ], and a custom-designed robotic device (one study) [ 19 ]. The validity and reliability data from these arthrometers are described in Table S5. Bioengineering 2023,10, 799 6 of 16 3.4. Characteristics of Evaluation and Method of Measurement Sensors were the most commonly used method for measuring GH laxity and stiffness (15 studies). The sensors measured the force-induced position changes at the glenohumeral joint, which were used calculate the joint laxity and/or stiffness (12 studies) [ 20 – 23 , 27 , 28 , 30 , 34 – 36 , 39 , 40 ], while the other devices visually displayed the amount of laxity on a screen (three studies) [ 31 , 38 , 41 ]. Apart from sensors, other studies used stress imaging either with radiography or US devices (five studies) [ 24 – 26 , 29 , 37 ], a visual scale (two studies) [32,33], or a digital motion controller (one study) [19]. Force was applied with a controlled manual instrumented-assisted mechanism, with only one study using an electromechanical system [ 19 ]. The direction of force was usually in the sagittal plane (AP or PA), with only three studies applying an inferior-directed force [ 19 , 22 , 23 ]. The amount of load applied during the testing procedures was heterogenous across the studies. The load applied ranged from 10 to 150 N. Three studies used a progressive application of load (0–100, 0–134, or 10–80 N) and two studies applied force until the capsular endpoint [22,23]. The shoulder was usually positioned at 90 ◦ of abduction in the scapular plane (15 studies) [ 19 , 24 – 31 , 34 – 38 , 41 ], in either external rotation (12 studies) or neutral rotation (four studies). Other studies positioned the shoulder at 20 ◦ of abduction with neutral rotation (nine studies) [ 20 – 23 , 32 , 33 , 39 – 41 ]. The individuals were either lying in a supine position (modified KT-1000/2000 and modified custom-designed robotic device) or seated in a chair (customized instrumented shoulder arthrometer, Telos GA-II/E, Telos, and Donjoy ® ). Methods of fixation varied considerably across the studies and devices, and are detailed in Table S4. The measurement methods for GH laxity showed heterogeneity across the included studies (Table S4). Imaging methods measured the bone displacements (distance between the center or posterior humeral head to the glenoid) to calculate the joint laxity. When using sensors, laxity was calculated as the distance between two sensors placed at the humeral head and acromion (sagittal displacement) or between the humeral head and lateral epicondyle of the distal humerus (inferior displacement). While in most studies it was measured to total displacement, other studies restricted the displacement to the distance between the inflection point until the data point at the highest load. Stiffness was always calculated as the slope of the linear portion of the force–displacement curve. The KT1000/2000 estimated the humeral head displacement with a single sensor placed at this anatomical point, but without any other reference point. The Donjoy ® arthrometer used a visual-instrumented scale on the spring balance to measure the sagittal displacement of the humeral head. 3.5. Laxity and Stiffness Values The GH laxity was analyzed across asymptomatic healthy individuals, asymptomatic athletes, and individuals with injured shoulders. Overall, there was a wide heterogeneity in laxity values across and within subgroups of individuals with asymptomatic shoulders, especially when compared between devices. The laxity and stiffness data for each study and testing condition are detailed in Table S5. Asymptomatic healthy individuals showed varying laxity values, ranging from 0.7 to 27.72 mm for PA, 1 to 21.75 mm for AP (Figure 2a), and 0.6 and 2.1 mm for global translation (Figure 2b). Stiffness ranged from 16.3 to 16.7 N/mm for PA and 1.51 to 15.7 N/mm for inferior (Table 2). Only one study [ 22 ] assessed inferior laxity with a 13.9 mm displacement, while another study assessed AP stiffness with a similar slope of 15.4 N/mm [23]. Bioengineering 2023,10, 799 7 of 16 Bioengineering 2023, 10, x FOR PEER REVIEW 8 of 16 Figure 2. Glenohumeral laxity measurement outcomes of healthy individuals with asymptomatic shoulders: (a) Anterior, posterior, and inferior translation [20-23,25,26,30,31,37-41]; (b) global sagittal translation [32]. Figure 2. Glenohumeral laxity measurement outcomes of healthy individuals with asymptomatic shoulders: ( a ) Anterior, posterior, and inferior translation [ 20 – 23 , 25 , 26 , 30 , 31 , 37 – 41 ]; ( b ) global sagittal translation [32]. Bioengineering 2023,10, 799 8 of 16 Asymptomatic athletes also showed varying laxity values, ranging from 1.4 to 12.57 mm for PA, 4.82 to 12.71 mm for AP (Figure 3a), and 7.81 to 24.92 mm for global translation (Figure 3b). When comparing the dominant/throwing arm against the contralateral, there was no significant difference in the mean laxity values. The mean values of stiffness also showed a large range from 7.77 to 16.6 N/mm for PA and 8 to 15.3 N/mm for AP (Table 2). Table 2. Stiffness of the included studies. Population Arthrometer Studies Amount of Load Device Evaluated Arms PA AP Inferior Healthy Individuals Custom-designed robotic device Azarsa et al. (2021) [19]10–80 N Digital motion controller + software Right arm NR NR 1.51 Customized instrumented shoulder arthrometer Borsa et al. (2001,2002) [22,23]NR Sensor Nondominant arms 16.7 15.4 15.7 Borsa et al. (2000) [21]0–134 N Sensor Bilateral arms—male 20.5 NR NR Borsa et al. (2000) [21]0–134 N Sensor Bilateral arms—female 16.3 NR NR Healthy Athletes Telos + Ligmaster Crawford & Sauers (2006) [28]15 dN Sensor Throwing arm—neutral rotation 8.05 8.00 NR Crawford & Sauers (2006) [28]15 dN Sensor Nonthrowing arm—neutral rotation 7.77 8.05 NR Crawford & Sauers (2006) [28]15 dN Sensor Throwing arm—external rotation 10.87 NR NR Crawford & Sauers (2006) [28]15 dN Sensor Nonthrowing arm—external rotation 10.24 NR NR Borsa et al. (2006) [27]15 dN Sensor Throwing arm 16.6 15.1 NR Borsa et al. (2006) [27]15 dN Sensor Nonthrowing arm 16.2 15.3 NR Injured shoulders displayed a more uniform pattern of laxity, with mean values ranging from 2 to 3.4 mm for PA, 3.0 to 5.42 mm for AP (Figure 4a), and 2.8 to 11.9 mm for global translation (Figure 4b). None of the included studies assessed the stiffness of individuals with injured shoulders. Bioengineering 2023,10, 799 9 of 16 Bioengineering 2023, 10, x FOR PEER REVIEW 9 of 16 Asymptomatic athletes also showed varying laxity values, ranging from 1.4 to 12.57 mm for PA, 4.82 to 12.71 mm for AP (Figure 3a), and 7.81 to 24.92 mm for global translation (Figure 3b). When comparing the dominant/throwing arm against the contralateral, there was no significant difference in the mean laxity values. The mean values of stiffness also showed a large range from 7.77 to 16.6 N/mm for PA and 8 to 15.3 N/mm for AP (Table 2). Figure 3. Glenohumeral laxity outcomes of athletes with asymptomatic shoulders: (a) Anterior and posterior translation [24,26,28-30,34-36]; (b) global sagittal translation [24,28]. Injured shoulders displayed a more uniform pattern of laxity, with mean values ranging from 2 to 3.4 mm for PA, 3.0 to 5.42 mm for AP (Figure 4a), and 2.8 to 11.9 mm for global translation (Figure 4b). None of the included studies assessed the stiffness of individuals with injured shoulders. Figure 3. Glenohumeral laxity outcomes of athletes with asymptomatic shoulders: ( a ) Anterior and posterior translation [24,26,28–30,34–36]; (b) global sagittal translation [24,28]. Bioengineering 2023,10, 799 16 of 16 67. Andrade, R.; Pereira, R.; Leal, A.; Pereira, B.; Boas, J.P.V.; van Dijk, C.N.; Espregueira-Mendes, J. Development of a medical device compatible with MRI/CT to measure ankle joint laxity: The Porto Ankle Testing Device. Porto Biomed. J. 2021 ,6, e122. [CrossRef] 68. Espregueira-Mendes, J.; Andrade, R.; Leal, A.; Pereira, H.; Skaf, A.; Rodrigues-Gomes, S.; Oliveira, J.M.; Reis, R.L.; Pereira, R. Global rotation has high sensitivity in ACL lesions within stress MRI. Knee Surg. Sports Traumatol. Arthrosc. 2016 ,25, 2993–3003. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.