TYPE Original Research PUBLISHED 07 February 2023 DOI 10.3389/fphys.2023.986872 Measuring direct and indirect OPEN ACCESS tendon parameters to characterize EDITED BY Emiliano Cè, University of Milan, Italy the proximal tendinous complex of REVIEWED BY Niccolò Gori, the rectus femoris in football and Federazione Italiana Rugby, Italy Piotr Karauda, Medical University of Lodz, Poland futsal players *CORRESPONDENCE Sandra Mechó, Sandra Mechó 1,2*, Raquel Lisbona Ortega 3, Ricard Pruna 4, mechomeca@gmail.com Lexa Nescolarde Selva 5, Jordi Morillas Pérez 6, SPECIALTY SECTION This article was submitted to Medical Alfonso Rodríguez-Baeza 7, Javier Martínez Agea 2 and Physics and Imaging, Ricard Pérez-Andrés 3 a section of the journal 1 Frontiers in Physiology Autonomous University of Barcelona, Barcelona, Spain, 2Department of Radiology Hospital de Barcelona, SCIAS, Barcelona, Spain, 3Department of Radiology, Hospital Germans Trias i Pujol, Badalona, Spain, 4FC RECEIVED 05 July 2022 Barcelona Medical Services, Sant Joan Despí, Spain, 5Department of Electronic Engineering, Universitat ACCEPTED 13 January 2023 Politècnica de Catalunya, Barcelona, Spain, 6Unit of Intensive Care Medicine, Hospital de Barcelona, SCIAS, PUBLISHED 07 February 2023 Barcelona, Spain, 7Department of Morphological Sciences (Human Anatomy and Embryology Unit), Faculty CITATION of Medicine, Universitat Autònoma de Barcelona, Bellaterra, Spain Mechó S, Lisbona Ortega R, Pruna R, Nescolarde Selva L, Morillas Pérez J, Rodríguez-Baeza A, Martínez Agea J and Pérez-Andrés R (2023), Measuring direct and indirect tendon parameters to Objective: To present unprecedented radiological parameters that characterize the characterize the proximal tendinous angle between the direct and indirect tendons of the proximal rectus femoris (RF) complex of the rectus femoris in football and futsal players. and its inclinations and to evaluate the population variability according to Front. Physiol. 14:986872. demographic variables. doi: 10.3389/fphys.2023.986872 Materials and methods: From September 2019 to July 2021, using MRI multiplanar COPYRIGHT © 2023 Mechó, Lisbona Ortega, Pruna, reconstructions of the proximal thigh/hip, two blinded radiologists measured the Nescolarde Selva, Morillas Pérez, direct and indirect tendon angle and the inclination of each tendon in different Rodríguez-Baeza, Martínez Agea and planes. The intra- and inter-observer agreements were assessed with Bland–Altman Pérez-Andrés. This is an open-access article distributed under the terms of the analysis and intraclass correlation coefficient (ICC). The correlations between Creative Commons Attribution License radiological parameters and demographic variables were evaluated using linear (CC BY). The use, distribution or regression, Student’s t-test, and analysis of variance. reproduction in other forums is permitted, provided the original author(s) and the Results: We performed 112 thigh/hip MRI scans on 91 football players of different age, copyright owner(s) are credited and that the original publication in this journal is gender, and disciplines (football and futsal). For observer 1 (the reference), the mean cited, in accordance with accepted direct and indirect tendon angle was 56.74° ± 9.37, the mean indirect tendon slope academic practice. No use, distribution or was −7.90° ± 7.49, and the mean direct tendon slope was 22.16° ± 5.88. The three reproduction is permitted which does not comply with these terms. measurements showed inter- and intra-observer agreement (mean differences ~0). No correlation was observed between age and the parameters. Likewise, no statistically significant differences were found for gender, dominant limb, examined limb, and sport. Conclusion: There is an inter- and intra-observer agreement in the measurements of the direct and indirect tendon angle and the inclination of each tendon. There is population variability in the proximal tendinous complex unrelated to demographic factors. These results allow further detection of morphological patterns that represent a risk factor for lesions in the RF in professional football and futsal players and other sports. KEYWORDS MRI, rectus femoris, proximal tendinous complex, direct tendon and indirect tendon, DT-IT angle Frontiers in Physiology 01 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 Introduction described in the medical literature (Renstrom PA, 1992; Hasselman CT et al., 1995; Bianchi S et al., 2002; Bordalo-Rodrigues M and Muscle injury remains a diagnostic, therapeutic, and preventive Rosenberg ZS, 2005; Ouellette H et al., 2006; Tubbs RS, 2006; challenge in sports medicine. More than a third of the injuries suffered Gyftopoulos S et al., 2008; Balius R et al., 2009; Gamradt SC et al., by athletes annually are muscle injuries (Ekstrand J et al., 2011; 2009; Kassarjian A et al., 2012; Ryan JM et al., 2014; Mariluis and Guermazi A et al., 2017; Isern-Kebschull J et al., 2020). Among CupitoMamone, 2015; Moraux A et al., 2015), but the angle formed by quadriceps muscles, the rectus femoris (RF) is the most commonly the heads to converge on the CT and their inclination have not been injured and the third most common site of lower extremity muscle previously studied. Considering all the aforementioned factors, a injury (Entwisle T et al., 2019). Indeed, a study on professional female detailed characterization of all elements forming the proximal MTJ football players characterized the quadriceps muscle complex as the of the RF and their relationship would be of interest to detect possible most often injured (Larruskain et al., 2018). patterns with a high risk of injury and to reinforce injury prevention. The RF is a biarticular, fusiform muscle that tends to carry out Therefore, this study aims to present unprecedented radiological eccentric contractions with a high percentage of fast-twitch (type II) parameters to characterize the direction of the DT and IT tendons fibers. This explains why the RF is the quadriceps muscle more when converging on the CT, in the sagittal plane (i.e., the DT-IT angle), frequently injured in repetitive kicking and sprinting sports and to characterize the inclinations of the DT and IT in different planes. It (Hughes C et al., 1995; Morelli V and Weaver V., 2005; Kassarjian also aims to evaluate the population variability of these PTC structures and A et al., 2012; Mendiguchia J et al., 2013; Mariluis and CupitoMamone, their variability according to demographic factors. 2015; Entwisle T et al., 2019). The myotendinous junction (MTJ) of the central septum is the most common site of RF injury in football (Balius R et al., 2009; Materials and methods Mendiguchia J et al., 2013; Mariluis and CupitoMamone, 2015) and is associated, particularly proximal injuries, with long rehabilitation and Patients delayed return-to-play times (Cross TM et al., 2004; Ochard J et al., 2005; Balius R et al., 2009; Mendiguchia J et al., 2013; Brukner P and The Clinical Research Ethics Committee of the Sports Administration Connell D, 2015). The central septum belongs to the proximal MTJ of of Catalonia (Spain) approved this prospective study. Written informed the RF. As its origin is complex, the term proximal tendon complex consent was obtained from all participants included in this study. (PTC) will be applied from here onward. The PTC consists of two From September 2019 to July 2021, 100 patients or volunteers tendons: the direct or anterior tendon (DT) and the indirect or were initially included to perform hip or proximal thigh MR reflected tendon (IT). Sometimes, there is a third tendon, examinations with a 3D PD sequence added to the standard originating in the IT and attached to the structures that delimit it protocol. All participants were football players of different (Tubbs RS, 2006). Ultimately, these tendons converge into a conjoined disciplines (association football and futsal) and categories (junior, tendon (CT). The tendinous origin of the RF has been widely young, and professional teams). FIGURE 1 Flowchart of the participants in the study. DT, direct tendon; FSE, fast spin echo; IT, indirect tendon; PD, proton density; PTC, proximal tendon complex. Frontiers in Physiology 02 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 FIGURE 2 Multiplanar reconstruction. (A) Chosen axial 3D proton density section to orientate the sagittal plane following the lateral margin of the anterior portion of the indirect tendon (white arrowhead). (B) Chosen coronal 3D proton density section to orientate the sagittal plane following the lateral margin of the direct tendon (white arrow). (C) Resulting oblique sagittal plane. FIGURE 3 Measurement of the indirect tendon slope. (A) Sagittal 3D proton density with the reference line to obtain the batch of the oblique axial reconstruction following the orientation of the indirect tendon. (B) Median section of the oblique axial reconstruction of the indirect tendon with measurement of its inclination (indirect tendon anterolateral margin/anteroposterior axis). The portion of the CT is excluded. In this case, the merging of the indirect tendon with the CT in the axial plane is more lateral than the acetabular insertion of the indirect tendon, and the inclination is negative (−18.7°). FIGURE 4 Double oblique sagittal 3D proton density sections. (A) Lateral section of the indirect tendon with the cranial margin of the IT (indirect tendon line of the angle). (B) Middle line section of the direct tendon with the anterior margin of the tendon draughted (direct tendon line of the angle) and (C) resulting direct–indirect tendon angle. Frontiers in Physiology 03 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 TABLE 1 Baseline demographic and epidemiological data of the studied population. Gender Female 13 (14.3) Male 78 (85.7) Age (years), mean (range) 21 (12–33) <18 years 31 (27.7) ≥18 years 81 (72.3) Laterality Right limb 58 (51.8) Left limb 54 (48.2) Dominant limb examined 66 (58.9) Sport Association football 83 (91.2) Futsal 8 (8.8) FIGURE 5 Body part examined Oblique coronal 3D proton density image. Measurement of the Hip 7 (6.2) direct tendon slope (direct tendon lateral margin/craniocaudal axis). Thigh 105 (93.8) Reason for exploration (injured body part) Hamstrings 45 (40.2) Patients with a clinical indication, some volunteers who willingly Rectus femoris 26 (23.2) participated in the study and could understand the reason behind it, and Adductors 18 (16.1) those who had signed informed consent were included for a hip or Femoroacetabular impingement 4 (3.6) proximal thigh MR. We excluded seven cases where artifacts or Psoas iliacus 4 (3.6) movement caused an inhomogeneity of the field under study, one case Contract signing exploration 4 (3.6) with a poor definition of tendon structures due to acute tendon injury, and Volunteer 4 (3.6) one case who did not consent to participate in the study (Figure 1). Quadriceps 2 (1.8) Each patient’s demographic data were recorded (age, gender, Other 5 (4.5) football discipline, lateral side, dominance, examination results, Figures are absolute numbers (and %) unless otherwise stated reason for examination, and history or acute injury of the proximal MTJ injury of the RF) in the database. allowed new oblique planes oriented to the anatomy of both Image acquisition tendons. 3D sequences were used for multiplanar reconstructions to avoid obtaining oblique sequences with post- The explorations were performed using a Canon Medical Vantage processing artifacts. Titan 3 T MR scanner, with a maximum gradient strength of 45 mT/ m, 203 T/m/s slew rate, and 32 receiver channels. An axial sequence was added to the standard thigh or hip protocol consisting of a 3D DT-IT angle plane proton density (PD) contrast; TR 5200 ms; TE 44–60 ms; echo train spacing 7.5; slice thickness 2.5–3.5 mm; in-plane resolution 1.4 × The DT-IT angle was measured in an oblique sagittal plane 0.88 mm2; and FOV 256 × 256 mm. The examination duration for this obtained by multiplanar reconstruction based on the orientation of sequence was 8 min. This sequence provides datasets that can be the IT axis in the axial plane and the DT axis in the coronal plane retrospectively post-processed for viewing in freely selectable (Figure 2). orientations, thus being advantageous for evaluating the complex The section where the IT was thickest in a greater length in the anatomy associated with the PTC (Mugler, 2014). axial plane was chosen. The inclination was oriented, taking the most The subjects were placed in a supine position, with the limb lateral margin of the anterior portion of the tendon as a reference covered by a 16-element phased array coil (Body). Twenty elements of (Figure 2A). In the coronal plane, the section was selected where the the table-integrated coil were also used, combined in a total of transverse diameter of the DT was greatest. The inclination was 32 independent receiver channels. The feet were fixed in a neutral oriented with respect to the most lateral margin of the DT (Figure 2B). position using a strap to standardize and fix the position of the extremities, and the symmetrical horizontal position of the femoral heads was verified using an MR pelvis locator scan. Totally, 172 axial IT slope plane images of the proximal portion of the thigh were obtained from the lower margin of the anterior superior iliac spine to the origin of the On the sagittal images, the plane in which the junction of the IT proximal anterior myoaponeurotic junction of the RF. with the CT was the clearest was chosen, and an oblique axial plane The images were post-processed with the validated software was obtained after the IT orientation in the portion closest to the CT Vitrea, and a multiplanar reconstruction was obtained, which (Figure 3). Frontiers in Physiology 04 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 TABLE 2 Descriptions of radiological measurements of DT-IT angle, IT slope, and DT slope for observer 1 and observer 2. Radiological measurements (grades) Observer 1 Observer 2 First measure Second measure First measure Second measure Direct tendon–indirect tendon angle 56.74 ± 9.37 (54.98–58.49) 57.05 ± 9.49 (55.28–58.83) 57.13 ± 9.75 (55.30–58.95) 57.37 ± 9.83 (55.53–59.21) Indirect tendon slope −7.90 ± 7.49 (−9.30–(−6.50)) −7.33 ± 7.44 (−8.72–(−5.93)) −7.98 ± 7.69 (−9.42–(−6.54)) −7.60 (11.85) −26.93–11.51 Direct tendon slope 22.16 ± 5.88 (21.06–23.26) 22.36 ± 5.66 (21.30–23.42) 22.22 ± 5.88 (21.12–23.33) 22.27 ± 6.00 (21.15–23.39) The variables normally distributed are shown as mean ± SD, 95% confidence interval for mean (lower bound and upper bound), whereas non-normally distributed data are shown as statistic median (interquartile range, IQR) and minimum–maximum. DT slope plane second randomized reading, with an interval between 4 and 12 months (a mean of 8 months). The DT slope was obtained using a coronal plane, selecting the plane with the greatest transverse diameter of the tendon. The inclination was oriented with respect to the most lateral margin of the DT. Data analysis The categorical variables are shown as absolute frequencies and Measurements percentages. The normality of distribution in the continuous variables was The obtained reconstructions were exported in batch format to PACS determined by the Kolmogorov–Smirnov test (n = 112). The normally to perform measurements. The reconstructions and the 3D PD axial distributed variables are shown as mean ± SD, 95% confidence interval sequence with Alma 2D Dicom Viewer were analyzed using a layout of for mean (lower and upper bound), whereas the non-normally different views (usually four). The three desired angles (DT-IT angle, IT distributed data are shown as median (interquartile range, IQR) slope, and DT slope) were obtained with the “Angle with Baseline” tool. and minimum–maximum. Double obliquity sagittal reconstruction was used to measure the DT- The Bland–Altman plot (Bland JM and Altman DG, 1999; IT angle (Figure 4). The first lateral section was chosen to obtain the IT Giavarina D, 2015) was used to evaluate mean biases and 95% component, where its upper margin was clearest (hypointense fine line), limit of inter- and intra-observer agreements. In the case of the and the line delimiting this margin was drawn (Figure 4A). Only the inter-observer analysis, observer 1 was taken as a reference, and portion of the free tendon was considered (tendinous portion without the analysis was made between the first measure of each observer. insertional expansions to the acetabular sulcus and joint capsule). The The B&A plot difference percentage was also performed. This is useful section corresponding to the middle third of the tendon was chosen to when there is an increase in variability of the differences as the delimit the DT line, considering the coronal image as a reference. The line magnitude of measurements increases (Bland JM and Altman DG, delimiting the anterior margin of the DT was drawn (Figure 4B). 1999). The sample size for the Bland–Altman plot was estimated for Oblique axial reconstruction was used to measure the IT slope. In Type I error (α, significance) of 0.01 and Type II error (β, 1-Power) of the middle section of the IT, the inclination of this tendon was 0.20 using MedCalc statistical software version 19.0.3 (MedCalc measured with respect to the anteroposterior axis, tracing the line Software bvba, Ostend, Belgium). of the IT on the most lateral margin of the free tendinous portion. A The intraclass correlation coefficient (ICC) was used to study the negative value was considered when, in the axial plane, the junction of intra- and inter-observer reliability for the measurement of the the IT with the CT was located in a more lateral location than the radiological parameters (DT-IT angle, IT slope, and DT slope). The acetabular insertion of the IT (Figure 3). data were analyzed using a mean of 112 measurements, absolute Coronal reconstruction was used to measure the DT slope. In the agreement, and a two-way mixed-effects model (Koo and Li, 2016). middle section of the DT, the inclination of this tendon was measured One-way ANOVA test and Student’s t-test were used to determine with respect to the craniocaudal axis, considering the lateral margin of statistical differences in the radiological parameters (DT-IT angle, IT the tendon as a reference point (Figure 5). slope, and DT slope), considering gender (male and female), football Post-processing of the obtained images was conducted by two discipline (futsal and association football), lateral side (right and left), radiologists specialized in musculoskeletal imaging with 13 (SM) and 2 dominance (right and left), and age (<18 y and ≥18 y). Linear (RL) years of experience. Both radiologists conducted a previous 3- regression analysis was also used to determine statistical differences month work to agree on the most appropriate measurement methods. in the radiological parameters considering age. Each radiologist assessed the images, performed the post-processing, The statistical software IBM® SPSS® version 26.0 (Armonk, NY: and measured the DT-IT angle and the slopes independently (they IBM Corp, United States) was used for data analysis. The level of were blind to the affiliation and clinical data of patients and the results statistical significance was set at p < 0.05. of each other). To assess the intra-observer agreement of the measurement method, each radiologist undertook this measurement process twice, using the same images. In cases of Results disagreement, the causes were analyzed (most were due to poor definition of the tendon margins), and the methodology was Post-processing and measurement of 112 examinations of revised. The intra-observer agreement was evaluated by means of a 91 players were performed (Figure 1). Frontiers in Physiology 05 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 FIGURE 6 Bland and Altman: inter-observer analysis (A: Bland and Altman plot; (B) plot of differences between observer 1 and observer 2 as percentages of the values on the axis vs. the mean of the two measurements) for (A) direct tendon–indirect tendon angle, (B) indirect tendon slope, and (C) direct tendon slope. The clinical and demographic characteristics of the participants a history of proximal MTJ injury of the RF, or an acute injury was are described in Table 1. The mean age was 21 (range 12–33) years. identified in this location on the examination. Dividing the age into ranges resulted in two categories: The descriptions of radiological measurements for both observers 81 participants older than 18 and 31 participants younger than are shown in Table 2. All the measurements are normally distributed 18 . Most participants were football players (83, 91.2%), and the variables, except for the second measurement of the IT slope of dominant limb was examined in 66 (58.9%) cases. Most observer 2. The mean of the first measure of the DT-IT angle of examinations were performed on the proximal thigh (105, both observers was similar (~57°). The mean of the first measure of the 93.8%) for injuries in the hamstrings (45, 40.2%), the RF (26, IT slope of both observers was similar (~−7.5°). The mean of the first 23.2%), or the adductor (18, 16.1%). In 35 cases (31.25%), there was measure of the DT slope of both observers was similar (~22°). In all Frontiers in Physiology 06 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 The inter-observer agreement B&A analysis for the different radiological measurements is represented in Figure 6. The mean difference of the first measure between observer 1 and observer 2 for angle DT-IT (Figure 6A), IT slope (Figure 6B), and DT slope (Figure 6C) was ~0 grade. The intra-observer agreement B&A analysis for the different radiological measurements is represented in Figure 7. In the case of the angle DT-IT and DT slope measurement (Figures 7A, C), the intra-observer agreement was similar for both observers, showing values of ~ −0.2 and ~ −0.1, and 95% of the data points lie within ± 2 s of the mean difference. In the case of the IT slope measure of observer 1, the intra-observer agreement B&A analysis showed that 0 is not included in the confidence interval limits for mean difference (Figure 7B). However, the mean difference is close to 0 (−0.6). The ICC results are described in Table 3, showing reproducibility (with a tendency to absolute concordance) of the measurement method. The ICC value for all cases was greater than 0.9. The case with less correlation was the intra-observer 2 DT slope measurement, in which the ICC value was 0.92 with a 95% confidence interval of 0.888–0.946. Likewise, the results of Student’s t-test and one-way ANOVA test showed no statistically significant differences in the measured DT-IT angle and tendon inclinations for the means of the distributions corresponding to gender, dominant side, examined limb, and football discipline (Tables 4, 5). However, the Student’s t-test (Table 6) showed a correlation between the DT slope and the age range, and the older group showed a steeper slope. Conversely, linear regression analysis in Figure 8 showed that the participant’s age and the measured DT-IT angle or tendon inclinations (i.e., IT and DT slopes) were not significantly correlated. The population available in the study is small to be able to assess the correlation between the radiological parameters and the risk of proximal MTJ injury of the RF. However, a small sample of three players was taken as an example, considering the history or presence of RF injury [two players with a history or acute injury (case B-C) and one player with no history of injury (case A)]. It could be seen that high values of the DT-IT angle (77.84°, 63.27°) and rectification of the inclinations of the IT (−1.41°) and the DT (16.14°) were characteristics of the injured cases. Therefore, they can be considered risk characteristics (Table 7). Discussion This study presented unprecedented radiological parameters to measure the angle between the DT and IT of the PTC of the RF. In FIGURE 7 Bland and Altman: intra-observer analysis (A: Bland–Altman plot; addition, the inclination of these tendons in their respective (B) plot of differences between measure 1 and measure 2 for observer craniocaudal and anteroposterior courses in the coronal and axial 1 and observer 2 as percentages of the values on the axis vs. the mean of planes was also reported. The measurements were carried out in the two measurements) for (A) direct tendon–indirect tendon angle, (B) indirect tendon slope, and (C) direct tendon slope. 112 MR scans of football players of different gender, sport discipline, and age. The inter- and intra-observer agreement analyses in the proposed measurements of the radiological parameters (DT-IT angle, IT slope, and three cases, the mean of the second measure of both observers was DT slope) have been evaluated through the B&A plot (Bland JM and similar to that of the first measure. Altman DG, 1999; Giavarina D, 2015). The B&A analysis recommended Considering the sample calculation for Bland–Altman (B&A) that the criteria for good agreement are that the mean difference of the analysis, the maximum required number of pairs for the angle DT- measure between observers should be ~0 and that 95% of the data points IT measurement was 18 cases; for the IT slope, 11 cases; and for the DT should lie within ± 2 s of the mean difference (Bland JM and Altman DG, slope, 15 cases. 1999; Giavarina D, 2015). In all three radiological parameters, the B&A Frontiers in Physiology 07 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 TABLE 3 ICC intra- and inter-observer analysis of the DT-IT angle, IT slope, and DT slope measurements, including intraclass correlation coefficient (ICC) and 95% confidence interval (CI). Observer 1 95% CI F test with true value 0 ICC Lower bound Upper bound F-value df1 df2 p-valuea DT-IT angle 0.987 0.981 0.991 152.059 111 111 0.00 IT slope 0.98 0.971 0.986 97.878 111 111 0.00 DT slope 0.934 0.906 0.954 29.477 111 111 0.00 Observer 2 DT-IT angle 0.988 0.982 0.992 163.440 111 111 0.00 IT slope 0.978 0.968 0.985 90.203 111 111 0.00 DT slope 0.922 0.888 0.946 24.586 111 111 0.00 Observer 1 vs. Observer 2 95% CI F test with true value 0 ICC Lower bound Upper bound F-value df1 df2 p-valuea DT-IT angle 0.957 0.938 0.97 45.42 111 111 0.00 IT slope 0.951 0.93 0.966 39.951 111 111 0.00 DT slope 0.946 0.923 0.963 36.326 111 111 0.00 DT, direct tendon; IT, indirect tendon. a Statistical significance of ICC analysis. TABLE 4 Statistical Student’s t-test correlation of the DT-IT angle, IT slope, and DT slope measured by observer 1 with demographic factors, including 95% confidence interval (CI). DT-IT angle 95% CI Mean differences Standard error differences Lower bound Upper bound p-valuea Gender 0.003 2.69 −5.3 5.3 0.84 Examined limb 2.4 1.77 −1.09 5.9 0.9 Dominant limb −0.15 2.16 −4.4 4.7 0.94 Sport −3.5 3.1 −9.7 2.6 0.69 IT slope Gender −1.08 2.15 −5.34 3.17 0.62 Examined limb 1.9 1.4 −0.86 4.73 0.43 Dominant limb −0.98 1.13 −1.37 3.34 0.39 Sport 1.36 2.5 −3.57 6.3 0.55 DT slope Gender 2.45 1.67 −0.86 5.76 0.44 Examined limb −1.5 1.12.35 −3.7 0.68 0.38 Dominant limb −0.47 1.22 −2.08 3.02 0.7 Sport −2.5 1.94 −6.3 1.4 0.78 DT, direct tendon; IT, indirect tendon. a Statistical significance of Student’s t-test. Frontiers in Physiology 08 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 TABLE 5 One-way ANOVA test of the DT-IT angle, IT slope, and DT slope measured by observer 1 depending on the demographic factors, including 95% confidence interval (CI) and F-Fisher. DT-IT angle 95% CI Mean ± SD Lower bound Upper bound F p-valuea Gender Male 56.74 ± 9.4 54.85 58.62 0 0.99 Female 56.73 ± 9.4 51.74 61.72 Examined limb Right 57.89 ± 9.2 55.46 60.32 1.85 0.18 Left 55.49 ± 9.5 52.97 58 Dominant limb Dominant 55.06 ± 8.98 51.21 58.91 0.003 0.95 Non-dominant 55.21 ± 8.47 51.36 59.06 Sport Football 56.42 ± 9.3 54.58 58.25 1.31 0.25 Futsal 59.97 ± 9.9 54.1 65.84 IT slope Gender Male −8.04 ± 7.5 −9.54 −6.53 0.25 0.6 Female −6.95 ± 7.62 −10.93 −2.97 Examined limb Right −6.97 ± 8.01 −8.91 −5.03 1.88 0.17 Left −8.9 ± 6.8 −10.9 −6.89 Dominant limb Dominant −7.18 ± 7.44 −10.14 −4.21 0.22 0.64 Non-dominant −6.19 ± 5.89 −9.15 −3.23 Sport Football −7.78 ± 7.34 −9.25 −6.3 0.3 0.58 Futsal −9.14 ± 9.23 −13.85 −4.44 DT slope Gender Male 22.46 ± 5.98 21.29 23.63 2.15 0.14 Female 20.01 ± 4.74 16.91 23.11 Examined limb Right 21.43 ± 6.02 19.9 22.95 1.88 0.17 Left 22.94 ± 5.67 21.36 24.52 Dominant limb Dominant 22.11 ± 6.41 19.68 24.54 0.08 0.78 Non-dominant 22.58 ± 4.45 20.15 25.02 Sport Football 21.94 ± 5.82 20.79 23.09 1.6 0.21 Futsal 24.4 ± 6.34 20.72 28.07 DT, direct tendon; IT, indirect tendon. a Statistical significance of one-way ANOVA test. plot demonstrated inter-observer agreement (Figure 6). In the IT slope, The intra-observer agreement B&A analysis for the different the inter-observer agreement is where more outliers were found. It should radiological measurements is presented in Figure 7. In the case of be due to the wavy morphology and lateral convex margin of the IT. These the angle DT-IT and the DT slope, there was intra-observer characteristics make it difficult to decide which trajectory should follow agreement. In the case of the IT slope measurement of the IT line to measure the inclination. observer 1, the intra-observer agreement B&A analysis showed Frontiers in Physiology 09 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 TABLE 6 Student’s t-test analysis of the DT-IT angle, IT slope, and DT slope measured by observer 1 depending on the age ranges (≥18 y and <18 y). Student’s t-test 95% CI Mean ± SD Mean differences SE differences Lower Bound Upper Bound p-valuea DT-IT angle ≥18 56.46 ± 9.19 −1.01 1.99 −4.94 2.93 0.61 <18 57.46 ± 9.95 IT slope ≥18 −7.4 ± 7.60 1.81 1.58 −1.32 4.94 0.64 <18 −9.2 ± 7.10 DT slope ≥18 22.61 ± 6.29 1.62 1.24 −0.83 4,08 0.04 <18 20.98 ± 4.52 DT, direct tendon; IT, indirect tendon. a Statistical significance of Student’s t-test. that 0 is excluded from the confidence interval limits for here and that measured by Atzmon et al. (2020) between the RF mean difference (Figure 7B). When the plot of differences is shown as muscle belly and the free IT lies in the origin of the measurement; the B&A plot difference percentage, 0 is included in the confidence interval line along the DT path was drawn, which follows an axis different from limits for the mean difference. Therefore, the intra-observer agreement that of the muscular belly. B&A analysis for the IT slope measure of observer 1 meets criteria B&A. To draw the IT line, the free tendon portion was considered as Differences in the intra-observer agreement between observers could be Atzmon et al. (2020) did, because the insertional portion is a complex due to experience. The more experienced observer (observer 1) could, at network difficult to define by MR. The IT in its free tendon portion can some point, draw a line between some parameters following the instinct present cranial or caudal expansions. However, they can be clearly that develops with their own experience. Conversely, in the case of the less defined and do not hinder the correct definition of the superior margin experienced observer (observer 2), they probably follow the rules of the of the tendon when measuring the angle. Other studies have also been measurement method with accuracy. conducted to evaluate the measurements of the insertional portions of The radiological measurements of the PTC proposed by us (DT-IT the DT and IT, demonstrating that these attachments were located in a angle, IT slope, and DT slope) showed inter- and intra-observer broad area of the anterolateral pelvis and close to critical agreement, although the IT slope measurement showed a slightly neurovascular structures (Hapa O et al., 2013; Ryan JM et al., lower degree of inter- and intra-observer agreement according to B&A 2014). These neurovascular structures can be identified by MRI analysis, which could be due to the IT wavy morphology and its and are differentiated from the typical expansions of IT by the convex lateral margin. Conversely, the DT slope measurement showed characteristic irregular and serpiginous path of vascular structures. the highest inter- and intra-observer agreement according to B&A Previous studies have described in detail the anatomy of the PTC in analysis, probably due to its well-defined margins in most cases. cadavers (Testut L, 1884; Paturet G, 1951; Hasselman CT et al., 1995), Furthermore, according to the ICC results, the inter- and intra-observer and some have used MRI techniques (Bordalo-Rodrigues M and reliability for measuring the radiological parameters was excellent. Rosenberg ZS, 2005; Ouellette H et al., 2006; Gyftopoulos S et al., In addition, population variability was found in all our measurements, 2008; Kassarjian A et al., 2012). However, to the best of our knowledge, and neither the DT-IT angle nor the tendon inclinations was correlated this work is the first study to assess the degree of the amplitude of the with gender, dominant limb, examined limb, or football discipline. In the angle between the IT and the DT when they merge (in the CT), case of the age range, there was no correlation between the DT-IT angle without being influenced by the complexity of the insertion of the IT and the IT inclination, but the DT slope showed a correlation with age, or the DT. In addition, the inclination presented by the IT and DT has probably because the samples between both ranges are different. Another already been reported in their free tendon path up to their respective limitation of the study is the small sample size of bilateral studies to insertions in the axial and coronal planes. compare dominant and non-dominant limbs. In an ultrasound study of the IT anatomy, Moraux et al. (2015) assessed The PTC of the RF has been extensively studied. Atzmon et al. the thickness, length, and inclination of the IT in the axial plane according to (2020) analyzed the IT in an anatomical study with human cadavers to the inclination of the transducer (29.7°). In agreement with our results, they assess its clinical applications in labral surgical reconstruction. The found no statistical differences in tendon measurements and angulation of authors described different PTC variables, among which the angle the IT between the left and right sides nor between men and women between the longitudinal axis of the muscle belly of the RF and the free (Moraux A et al., 2015). Although Atzmon et al. (2020) suggested that IT (mean 135.8°) and the angle between the free IT and its footprint are musculoskeletal structures could evolve with age (Atzmon R et al., 2020), included (Atzmon R et al., 2020). The latter cannot be measured by their results did not show any statistical differences according to that MRI, given the poor definition of the craniocaudal margins of the parameter. In addition, in this study, no statistically significant differences footprint portion. The difference between the DT-IT angle described were found between the two studied different football disciplines Frontiers in Physiology 10 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 FIGURE 8 Linear regression plots of DT-IT angle (A), IT slope (B), and DT slope (C) with respect to age (y). (association football and futsal) or between lateral dominance, suggesting no et al. (Hughes C et al., 1995; Balius R et al., 2009; Mendiguchia J adaptive variation of the PTC morphology to the sporting activity. et al., 2013; Mariluis and CupitoMamone, 2015). The most frequent injuries in athletes are muscle injuries, and, The results demonstrated a population variability in the degree of in the case of football and its variants, 19% are myoconnective opening of the DT-IT angle and the inclination of these tendons and injuries of the RF (Ekstrand K et al., 2011; Isern-Kebschull J et al., that this variability did not depend on demographic factors, except for 2020). Several studies have monitored these injuries in Australian the DT slope and the older age group. This study is a prelude to the and Spanish football players and have shown that they usually correlation between radiological parameters and the risk of injury. represent a long time of rehabilitation (Mendiguchia J et al., 2013), After we have demonstrated the reproducibility of the radiological particularly if they present three out of the following four parameters and their utility in assessing a PTC, the next step will be to radiological signs: proximal location, injury seen in T1 weighted validate a screening program. Following the process of development images, cross-sectional area >50%, and intermuscular fluid and validation of a screening program, where Bahr recommends three (Rodrigo RM, et al., presented at the 2009 annual meeting of steps (Bahr R, 2016; Lee Dow C et al., 2021) to analyze the predictive the American Roentgen Ray Society) (Kassarjian A et al., 2012). capacity of the radiological parameters of the CTP as risk factors for The most frequent RF lesion involves the central septum, which injury, the cohort of men and soccer players should be selected. In the belongs to the proximal MTJ and was first described by Hughes future, once we achieve an adequate sample size to consolidate the Frontiers in Physiology 11 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 TABLE 7 DT-IT angle, IT slope, and DT slope in one male football player with no rectus femoris injury and two male football players with different rectus femoris injuries. Bold values are outliers. DT, direct tendon; IT, indirect tendon. Type of injury Angle DT-IT IT slope DT slope Injury figures Case A (no rectus femoris injury) 57.46° 26.25° −9.21° Case B Paraseptal acute and previous muscle injuries Scar of previous paraseptal muscle injury (arrowhead) and interstitial edema of paraseptal acute injury (arrow) 77.74° 19.82° −13.73° Case C common tendon tear (arrow) 63.26° −1.41° 16.14° Angle DT-IT: double obliquity sagittal reconstruction, last slice where the DT line is drawn and the IT line is projected. IT slope: oblique axial reconstruction and inclination of IT with respect to the anteroposterior axis. DT slope: coronal reconstruction and inclination of DT with respect to the craniocaudal axis. results, we will be able to demonstrate the correlation between the Data availability statement parameters and the risk of injury. We can only suggest the possibility of an anatomical pattern that may constitute a risk factor for injury in The raw data supporting the conclusion of this article will be made the proximal MTJ, markedly open DT-IT angle, and the rectification available by the authors, without undue reservation. of the inclinations of the IT and DT, independently (Table 7, Case B) or in combination (Table 7, Case C). Detecting early anatomical patterns that can exert biomechanical Ethics statement effects that play an important role in the pathogenesis of The studies involving human participants were reviewed and myoconnective lesions of the proximal MTJ of the RF should be approved by the Comité d’Etica’d Investigacions Clíniques de one of the main challenges for sports medicine departments. This, in l’Administració Esportiva de Catalunya. Consell Català de l’Esport turn, will foster the establishment of more individualized preventive Països Catalans Ave, 40–48. 08950-Esplugues de Llobregat Barcelona- programs. Spain. Written informed consent to participate in this study was The main limitation of this study was the relatively small sample provided by the participants’ legal guardian/next of kin. size. This precluded the evaluation of possible anatomical predisposition to RF injury based on the degree of opening of the DT-IT angle and the inclination of these tendons. Author contributions In conclusion, this study presented and analyzed unprecedented radiological parameters to measure the DT-IT angle and the IT and All authors listed have made a substantial, direct, and intellectual DT slope to further characterize the elements forming the RF. All contribution to the work and approved it for publication. measurements showed inter- and intra-observer agreement and anatomical variability of the PTC unrelated to the demographic factors of the population. These results allow going a step further Conflict of interest in the detection of possible morphological patterns that represent a risk factor for lesions in the proximal MTJ of the RF in professional The authors declare that the research was conducted in the football and futsal players, extending the analysis to other sports absence of any commercial or financial relationships that could be disciplines. construed as a potential conflict of interest. Frontiers in Physiology 12 frontiersin.org Mechó et al. 10.3389/fphys.2023.986872 Publisher’s note organizations or those of the publisher, the editors, and the reviewers. Any product that may be evaluated in this article, or All claims expressed in this article are solely those of the claim that may be made by its manufacturer, is not guaranteed or authors and do not necessarily represent those of their affiliated endorsed by the publisher. References Atzmon, R., Sharfman, Z. T., Atoun, E., Sampson, T. G., Amar, E., and Rath, E. (2020). Hughes, C., Hasselman, C. T., Best, T. M., Martínez, S., and Garrett, W. E. (1995). The anatomical properties of the indirect head of the rectus femoris tendon: A cadaveric Incomplete, intrasubstance strain injuries of the rectus femoris muscle. Am. J. Sports Med. study with clinical significance for labral reconstruction surgery. Archives Orthop. Trauma 23 (4), 500–506. doi:10.1177/036354659502300422 Surg. 140, 85–92. doi:10.1007/s00402-019-03293-6 Isern-Kebschull, J., Mechó, S., Pruna, R., Kassarjian, A., Valle, X., Yanguas, X., et al. Bahr, R. (2016). Why screening tests to predict injury do not work-and probably never (2020). Sports-related lower limb muscle injuries: Pattern recognition approach and MRI will. . .: A critical review:a critical review. Br. J. Sports Med. 50, 776–780. doi:10.1136/ review. Insights into imaging 11, 108–125. doi:10.1186/s13244-020-00912-4 bjsports-2016-096256 Kassarjian, A., Rodrigo, R. M., and Santisteban, J. M. (2012). Current concepts in MRI of Balius, R., Maestro, A., Pedret, C., Estruch, A., Mota, J., and Rodriguez, L. (2009). Central rectus femoris musculotendinous (myotendinous) and myofascial injuries in elite athletes. aponeurosis tears of the rectus femoris: Practical sonographic prognosis. Br. J. Sports Med. Eur. J. radiology 81 (12), 3763–3771. doi:10.1016/j.ejrad.2011.04.002 43, 818–824. doi:10.1136/bjsm.2008.052332 Koo, T. K., and Li, M. Y. (2016). A guideline of selecting and reporting intraclass Bianchi, S., Martinoli, C., Peiris Waser, N., Bianchi-Zamorani, M. P., Federici, E., and correlation coefficients for reliability research. J. Chiropr. Med. 15, 155–163. doi:10.1016/j. Fasel, J. (2002). Central aponeurosis tears of the rectus femoris: Sonographic findings. jcm.2016.02.012 Skelet. Radiol. 31, 581–586. doi:10.1007/s00256-002-0559-z Larruskain, J., Lekue, J. A., Diaz, N., Odriozola, A., and Gil, S. M. (2018). A comparison Bland, J. M., and Altman, D. G. (1999). Measuring agreement in method of injuries in elite male and female football players: A five-season prospective study. Scand. comparison studies. Stat. Methods Med. Res. 8, 135–160. doi:10.1177/ J. Med. Sci. Sports 28, 237–245. doi:10.1111/sms.12860 096228029900800204 Lee Dow, C., Timmins, R. G., Ruddy, J. D., Williams, M. D., Maniar, N., Hickey, J. T., Bordalo-Rodrigues, M., and Rosenberg, Z. S. (2005). MR imaging of the proximal rectus et al. (2021). Prediction of hamstring injuries in Australian football using biceps femoris femoris musculotendinous unit. Magnetic Reson. imaging Clin. N. Am. 13 (4), 717–725. architectural risk factors derived from soccer. Am. J. Sports Med. 49 (13), 3687–3695. doi:10.1016/j.mric.2005.08.005 doi:10.1177/03635465211041686 Brukner, P., and Connell, D. (2015). Serious thigh muscle strains”: Beware the Mariluis, C. A., Cupito, J., and Mamone, F. (2015). Desgarros del músculo recto femoral. intramuscular tendon which plays an important role in difficult hamstring and Actual. RM. Rev. Argent. Radiol. 79 (4), 182–191. doi:10.1016/j.rard.2015.08.002 quadriceps muscle strains. Br. J. Sports Med. 0, 205–208. doi:10.1136/bjsports-2015- Mechó, S., Iriarte, I., Pruna, R., Pérez-Andrés, R., and Rodríguez-Baeza, A. (2022). A 095136 newly discovered membrane at the origin of the proximal tendinous complex of the rectus Cross, T. M., Gibbs, N., Houang, M. T., and Cameron, M. (2004). Acute quadriceps femoris. Surg. Radiol. Anat. 44, 835–843. doi:10.1007/s00276-022-02954-3 muscle StrainsMagnetic resonance imaging features and prognosis. Am. J. Sports Med. 32 Mendiguchia, J., Alentorn-Geli, E., Idoate, F., and Myer, G. D. (2013). Rectus femoris (3), 710–719. doi:10.1177/0363546503261734 muscle injuries in football: A clinically relevant review of mechanisms of injury, risk Ekstrand, J., Hägglund, M., and Waldén, M. (2011). Epidemiology of muscle injuries in factors and preventive strategies. Br. J. Sports Med. 47, 359–366. doi:10.1136/bjsports- professional football (soccer). Am. J. Sports Med. 39 (6), 1226–1232. doi:10.1177/ 2012-091250 0363546510395879 Moraux, A., Wawer, R., Lefevbre, G., Demondion, X., and Cotten, A. (2015). An Entwisle, T., Carey, B., Marovic, P., Connell, D. A., and Schneider, M. (2019). anatomical study of the indirect tendon of the rectus femoris using ultrasonography. Eur. Rectus femoris injuries in professional athletes: MRI analysis correlated with time Radiol. 25 (12), 3614–3619. doi:10.1007/s00330-015-3769-1 interval to return to play. Semin. Musculoskelet. Radiol. 23 (S 02), 1–18. doi:10.1055/ Morelli, V., and Weaver, V. (2005). Groin injuries and groin pain in athletes: Part 1. s-0039-1692554 Prim. Care Clin. Off. Pract. 32, 163–183. doi:10.1016/j.pop.2004.11.011 Gamradt, S. C., Brophy, R. H., Barnes, R., Warren, R. F., Thomas Byrd, J. W., and Kelly, Mugler, J. P., III (2014). Optimized three-dimensional fast-spin-echo MRI. J. Magnetic B. T. (2009). Nonoperative treatment for proximal avulsion of the rectus femoris in Reson. Imaging 39, 745–767. doi:10.1002/jmri.24542 professional American football. Am. J. Sports Med. 37 (7), 1370–1374. doi:10.1177/ 0363546509333477 Ochard, J., Thomas, M. B., and George, M. V. (2005). Return to Play following muscle strains. Clin. J. Sport Med. 15, 436–441. doi:10.1097/01.jsm.0000188206.54984.65 Giavarina, D. (2015). Understanding Bland altman analysis. Biochem. Medica 25 (2), 141–151. doi:10.11613/BM.2015.015 Ouellette, H., Thomas, B. J., Nelson, E., and Torriani, M. (2006). MR imaging of rectus femoris origin injuries. Skelet. Radiol. 35 (9), 665–672. doi:10.1007/s00256-006-0162-9 Guermazi, A., Roemer, F. W., Robinson, P., Tol, J. L., Regatte, R. R., and Crema, M. D. (2017). Imaging of muscle injuries in sports medicine: Sports imaging series. Radiology 282 Paturet, G. (1951). Traité D’Anatomie Humaine. Membres supérieur e inférieur. Tome II. (3), 646–663. doi:10.1148/radiol.2017160267 Éditeurs, Paris: Masson and Cie, 794–796. Gyftopoulos, S., Rosenberg, Z. S., Schweitzer, M. E., and Bordalo-Rodrigues, M. (2008). Renstrom, P. A. (1992). Tendon and muscle injuries in the groin area. Clin. sports Med. Normal anatomy and strains of the deep musculotendinous junction of the proximal 11 (4), 815–831. doi:10.1016/s0278-5919(20)30487-7 rectus femoris: MRI features. AJR 190, W182–W186. doi:10.2214/AJR.07.2947 Ryan, J. M., Harris, J. D., Graham, W. C., Virk, S. S., and Ellis, T. J. (2014). Origin of the Hapa, O., Bedi, A., Gursan, O., Akar, M. S., Guvencer, M., Havitcioğlu, H., et al. (2013). direct and reflected head of the rectus femoris: An anatomic study. J. Arthrosc. Relat. Surg. Anatomic footprint of the direct head of the rectus femoris origin: Cadaveric study and 30 (7), 796–802. doi:10.1016/j.arthro.2014.03.003 clinical series of hips after arthroscopic anterior inferior iliac spine/subspine Testut, L. (1884). Les anomalies musculaires chez l’homme. Paris: G. Masson, 698. decompression. Arthroscopy 29 (12), 1932–1940. doi:10.1016/j.arthro.2013.08.023 Tubbs, R. S., Stetler, W., Savage, A. J., Shoja, M. M., Shakeri, A. B., Loukas, M., et al. Hasselman, C. T., Best, T. M., Hughes, C., 4th, Martinez, S., and Garrett, W. E. J. (1995). (2006). Does a third head of the rectus femoris muscle exist? Folia Morphol. 65 (4), An explanation for various rectus femoris strain injuries using previously undescribed 377–380. muscle architecture. Am. J. Sports Med. 23 (4), 493–499. doi:10.1177/036354659502300421 Frontiers in Physiology 13 frontiersin.org