Percutaneous iliosacral fixation in external rotational pelvic fractures. A biomechanical analysis

Accepted Manuscript Title: Percutaneous Iliosacral Fixation in External Rotational Pelvic Fractures. A Biomechanical Analysis Author: Miguel A. Gir´ aldez-S´ anchez ´ Angel L´ azaro-Gonz´ alvez Javier Mart´ ınez-Reina David Serrano-Toledano Alfredo Navarro-Robles Pedro Cano-Luis Evangelos M. Fragkakis Peter V. Giannoudis PII: S0020-1383(14)00543-9 DOI: http://dx.doi.org/doi:10.1016/j.injury.2014.10.058 Reference: JINJ 5956 To appear in: Injury, Int. J. Care Injured Received date: 12-6-2014 Revised date: 18-8-2014 Accepted date: 22-10-2014 Please cite this article as: Gir´ aldez-S´ anchez MA, L´ azaro-Gonz´ alvez ´ A, Mart´ inez-Reina J, Serrano-Toledano D, Navarro-Robles A, Cano-Luis P, Fragkakis EM, Giannoudis PV, Percutaneous Iliosacral Fixation in External Rotational Pelvic Fractures. A Biomechanical Analysis, Injury (2014), http://dx.doi.org/10.1016/j.injury.2014.10.058 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Page 1 of 18 Accepted Manuscript TITLE PAGE. PERCUTANEOUS ILIOSACRAL FIXATION IN EXTERNAL ROTATIONAL PELVIC FRACTURES. A BIOMECHANICAL ANALYSIS. Miguel A. Giráldez-Sánchez1, Ángel Lázaro-Gonzálvez1, Javier Martínez-Reina2, David Serrano-Toledano1, Alfredo Navarro-Robles2, Pedro Cano-Luis1, Evangelos M. Fragkakis3, Peter V. Giannoudis3. 1. Clinical Orthopaedics, Trauma Surgery and Rheumatology Management Unit, Hospital Universitario Virgen del Rocío, Seville, Spain. 2. Department of Mechanical Engineering, University of Seville. Escuela Superior de Ingenieros, Seville, Spain. 3. Academic Department of Trauma &Orthopaedics, Clarendon Wing, Level A, Leeds General Infirmary Great George Street, LS1 3EX, Leeds, UK. Conflict of interest statement. There are no conflicts of interest. Keywords: biomechanic, pelvis, open-book fracture, Young-Burgess fracture, screw, minimally invasive fixation. Clinical Orthopaedics, Trauma Surgery and Rheumatology Management Unit, HU Virgen del Rocío, Avda. Manuel Siurot s/n, E-41013 Seville, Spain. E-mail addresses: [email protected] (D. Serrano-Toledano). Page 2 of 18 Accepted Manuscript ABSTRACT. Introduction: Although the gold standard in open book pelvic fractures remains the pubic symphysis (PS) plate fixation, the clinical outcomes are not satisfactory, despite the excellent anatomical reduction assessed radiologically. Some authors suggest that residual instability of the posterior pelvic elements may be responsible for the chronic pain and the early osteoarthritic changes in the sacroiliac joint (SIJ). Objective: To evaluate whether the isolated posterior fixation with one or two iliosacral screws (ISSs) is sufficient to provide adequate stability for the treatment of Burgess Young APC-II (YB APCII) type of pelvic ring injuries. Methods: Biomechanical experimental study using 7 fresh human pelvises, where an YB APC-II pelvic injury was previously implemented. The isolated posterior fixation of the pelvic ring with 1 or 2 ISSs directed in the S1 vertebra body was analysed in each specimen following an axial load of 300N. The different displacement of the SIJ and of the PS were analysed in all three spatial axes, using the validated optical measurement system 3D PONTOS 5M. A multivariate version of Friedman test (non-parametric ANOVA for repeated measures) was performed. Results: The isolated fixation of the SIJ with 1 ISS did not show any differences with respect to the intact pelvis (p = 0,851). Regarding the PS, both type of fixations (with 1 or 2 ISSs) confirmed an acceptable correction and adequate control of the PS even though with some differences compared to the intact pelvis (p = 0,01). The presence of the second ISS found not to offer any significant additional benefit. The three-dimensional analysis of the behaviour of the pelvic elements, in these two different types of fixation, did not show any statistical significant differences (p = 0.645). Conclusion: The posterior fixation with ISS can represent an alternative option for treatment of pelvic injuries associated with rotational instability. Further prospective clinical studies are necessary to determine, the influence of the residual pubic symphysis mobility in the every day life, when the above-mentioned technique is applied. KEYWORDS. Biomechanic, pelvis, open-book fracture, Young-Burgess fracture, screw, minimally invasive fixation. INTRODUCTION. Pubic symphysis (PS) plating is considered nowadays the gold standard of treatment of open book pelvic lesions with partial preservation of the posterior anatomical structures (Young Burgess APC-II (YB APC-II)1 or AO/Tile B12,3) 2,4,5. In spite of a high rate of radiologically confirmed anatomic reduction of the symphysis joint, clinical outcomes reveal that 32.2% of patients report persistent pelvic back pain in the long-term6. These clinical results are analogous to those pelvic lesions characterised by complete disruption of the posterior pelvic ring elements7,8. Residual pain and secondary arthritis of the sacroiliac joint (SIJ)4,9 have been reported in Page 3 of 18 Accepted Manuscript patients who sustained antero-posterior compression pelvic injuries with partial lesion of the SIJ treated solely by plating of the PS. Not surprisingly, several authors questioned whether this strategy of stabilisation of these pelvic ring lesions is sufficient10,11,12,13,14,15to adequately control the SIJ. It was argued that the potential micro-instability component in SIJ10,15,16 following the injury may represent the cause of the contradictory results between optimal post-operative radiological findings of the anatomical reduction, with respect to the less satisfactory functional outcome during the long term follow up. Osterhoff17 et al. have recently concluded that “percutaneous sacroiliac screw fixation alone is a sufficient technique for the stabilization of rotationally unstable pelvic fractures”. The authors of the study underlined the importance of restoring the integrity of the posterior pelvic arch. Considering that pain is mainly located in the posterior region of the pelvic ring, precisely at the SIJ, is one incorrect to postulate whether anterior fixation of the PS is insufficient to guarantee stability of the posterior elements of the pelvic ring? Moreover, is it possible to restore the pelvic ring anatomy and prevent the micro-movements of the SIJ by solely posterior fixation? The literature regarding this clinical situation remains obscure. The aim of this study therefore is to determine the degree of pelvic stability after isolated posterior fixation using one or two iliosacral screws (ISSs) as definitive treatment in pelvic external rotation injuries (Young Burgess APC-II (YB APC-II) or AO/Tile B1). MATERIALS AND METHODS. Seven female human freshly-frozen anatomical specimens were used, (i.e. not embalmed cadavers) for this study. The mean age of the specimens was 73,9 (SD ±13,85) years. None of the donors had any previous medical history of skeletal disorders. The cadaveric specimens were dissected in such way as to obtain a configuration including the L4-L5 vertebrae, the whole pelvis and also the lower third of both femurs. In all specimens the articular capsule and ligaments of the PS and SIJ were spared, including the sacrospinous and the sacrotuberous ligaments. The soft tissues of the lumbar spine and of the hip joints were spared as well. After preparation, the anatomical specimens were frozen to below -20 °C10,12,18,19,. Before running the tests, the specimens were defrosted and hydrated by immersion in room temperature water for 16-20 hours. In this way the mechanical changes of the bone and the ligaments due to the freezing process and the dehydration were minimised. For the same reason the anatomic specimens were moistened with water before and during the experiments12,19. For the purposes of this study an electromechanical machine type Zwick / Roell Z100 (BT1- FB100TN), able to apply vertical loads, controlled by the test Xpert II software, was used. A patented and registered device12,20 was attached to the electromechanical machine in order to appropriately anchor the pelvis specimens and simulate the standing position and alignment in the same sagittal plane of the anterior superior iliac spines and of the pubic tubercles19. Due to the properties of this specific device, it was possible to analyze the biomechanical behavior of the fractured pelvis. The specimens were fixed proximally at an angle of 130° on a metallic plate attached on the vertebral bodies with 4 screws, 4 mm each, and 6 mm screw bolts for the sacrum. The remaining space between the metallic angled plate and the bone was filled with surgical Page 4 of 18 Accepted Manuscript cement polymethylmethacrylate (PMMA) Palacos LV®. Both femurs were distally fixed at 15° of anteversion and 10° of valgus, using a bi-component polyurethane resin type fast setting (Feropur®) (PR55-E55). In all cadaveric specimens an YB APC-II pelvic injury was simulated (Table 1). Each specimen underwent a cyclic axial load of 300N in 3 phases: (A) Intact pelvis, (B) injured pelvis with iliacsacral synthesis by one screw at the lower portion of the S1 level (F1S), and (C) injured pelvis with iliac-sacral synthesis, adding a second screw to the previous fixation scheme, also in the S1 level (2FS), positioned superiorly with respect to the first screw. Following the YB APC-II type of injury pelvis reduction was achieved by manual manipulation until complete alignment was externally visualized around the anterior edge of the sacroiliac joint and the anterior-posterior pubic symphysis margins. X-rays were then used to both confirm reduction and perform osteosynthesis using Kirschner guide wires and titanium cannulated screws (Synthes®, Oberdof, Switzerland) of 7.3 mm diameter with washer as recommended by Kraemer21. Over each specimen, radiographic control was performed in inlet, outlet and lateral sacral views to ensure the proper position of the implant and reduction of injury (figure 1). In order to study the three-dimensional directions of the different components of the pelvic ring, the PS and SIJ were marked with special adhesive markers. The movement and the characteristics of the pelvis under axial load in each phase of the study, were analyzed, using a validated system for micro mobility biomechanical studies, PONTOS 5M system22 (GOM System, Optical Measuring Techniques). On the other hand, minor axial load cycles of 80N were performed after every main axial load cycle of 300N, with the final objective of verifying that the axial load of 300N had not determined any bone fracture or that the repeated axial loads had not negatively influenced the overall mechanical behaviour of the specimens, which would have invalidated the study12. The displacements (measured in millimetres) were analyzed and compared between the following structures: The superior and inferior level of the PS, the superior and inferior level of both sacroiliac injured joints, and the rotational movements (measured in degrees) of the whole injured hemipelvis with reference to the sacrum, in the three spatial axes. In order to achieve a more thorough analysis of the data obtained, a complementary elaboration of the overall deformity of the pelvis was performed, analyzing the data acquired during the minor axial load cycles of 80N. This data was expressed in a regression line diagram, illustrating the deformation of the whole pelvis under the load applied. A multivariate version23 of the Friedman test (non-parametric ANOVA for repeated measures) was done to compare the overall effect of the independent variable injury status of the pelvis (with 3 levels: Physiological, injured and fixed with 1 ISS and injured and fixed with 2 ISS) on the 7 dependent variables (4 displacements and 3 rotations). Basically, this test consists in performing a MANOVA on the ranks. There was a special interest in comparing the dependent variables separately. This way, the multivariate test was performed three times: a) For the two displacements at the right SIJ, b) for the two displacements at the PS and c) for the three rotations of the right iliac bone relative to the sacrum. Page 5 of 18 Accepted Manuscript RESULTS. Tables 2 and 3 show the variations in marker position for the different specimens in each test relative to the initial position with no load in an intact pelvis, osteosynthesed injured pelvis with one and two ISSs respectively. Regarding the displacements at the SIJ, the Box´s M test for homogeneity of covariance matrices was not significant (p= 0.484). Then, the multivariate version of the Friedman test revealed a non-significant multivariate main effect for injury status, Wilks’ λ = 0.925, F (4,34) = 0.337, p = 0.851, partial eta squared = 0.038. Power to detect the effect was 0.117 (Table 2). With regard to the displacements at the PS (Table 3) the Box´s M test for homogeneity of covariance matrices was not significant (p= 0.002). The multivariate version of the Friedman test revealed a significant multivariate main effect for injury status, Wilks’ λ = 0.466, F (4,34) = 3.954, p = 0.010, partial eta squared = 0.317. Power to detect the effect was 0.860. Given the significance of the overall test, the univariate main effects were examined. Significant univariate main effects for injury status were obtained for the displacement at the upper part of the symphysis, F (2,18) = 5.917, p = 0.011, partial eta square = 0.397, power = 0.814; and for the displacement at the lower part of the symphysis, F (2,18 ) = 9.629 , p = 0.001 , partial eta square = 0.517, power = 0.959. Additionally, post-hoc pair wise comparisons for each displacement at the symphysis were made. The p-values of those comparisons are shown in Table 5 and the significant differences (applying Holm’s sequential Bonferroni correction) are indicated in boldface. In the comparison of the rotations of the right iliac bone (Table 4), the Box´s M test for homogeneity of covariance matrices was not significant (p= 0.645). Then, the multivariate version of the Friedman test revealed a non-significant multivariate main effect for injury status, Wilks’ λ = 0.669, F (6,32) = 1.186, p = 0.339, partial eta squared = 0.182. Power to detect the effect was 0.396. When comparing distance variations and their signs, no significant differences were found in the SIJ (p = 0.851); however, there was a significant difference in the lower symphysis (p = 0.001) and upper symphysis (p = 0.011). Comparing rotations of the right iliac bone no significant differences were found between intact pelvis and osteosynthesed injured pelvis. Overall, no significant differences were found between one or two ISSs. DISCUSSION. The management of pelvic ring injuries has evolved over the years and the work of Letournel has significantly contributed to the advances that have been made24. While it is widely accepted that lesions with complete ligamentous disruption both anteriorly and posteriorly (YB APC-III), necessitate anterior and posterior reconstruction, the management of lesions with partial posterior ligamentous disruption (YB APC-II) remains a point of vivid discussion10,11,25. The fact that the published radiological results on this type of injury do not reflect the clinical outcome which is less satisfactory, but is comparable to clinical results related to more severe injuries where all the soft tissue elements of the posterior pelvic ring are completely disrupted8 has led some authors to consider whether the long term pain reported is the sequelae of residual Page 6 of 18 Accepted Manuscript instability of the posterior pelvic elements. This argument has been supported by the observation of the development of early arthritic changes in the SIJ4,9. Consequently, there is an open debate whether the routine plating of the PS should be additionally enhanced with stabilisation of the SIJ10,11,12,13,25. It is of note that recently, a clinical study by Osterhoffet al17, supported the view that the ISS fixation alone could be sufficient for the treatment of an injury with rotational instability. The results of this study revealed that isolated posterior fixation with ISS managed to restore the physiological stability/mobility on the SIJ in YB APC-II injuries. With respect to the PS movements, the displacement was restored with a minimal residual mobility. Very few studies have examined the biomechanical behaviour of the PS following isolated fixation of the posterior pelvic ring. Simonian et al11performed fixation of the SIJ using ISS or sacroiliac plate in a YB APC-II injury model without PS fixation. They observed a significant decrease in displacement in the pelvic symphysis with respect to the intact pelvis, in both cases without anterior fixation. Nevertheless, authors concluded that the isolated posterior fixation of the pelvic ring can not affect the mobility of the PS. Regarding the two configurations of SIJ fixation tested in our study, the correction of the displacement achieved was 90% approximately (90.9% median with 16,4% Interquartile Range (IQR) in upper PS and 88.4% (IQR 22,4%)in lower PS with one ISS; two ISSs showed 85,9% (IQR 18,3%) and 84.8% (IQR 22,3%) median of reduction in upper and lower PS, respectively) allowing minimal displacements anteriorly (maximum gap measured in the superior and inferior PS was 1,47 mm and 1,86 mm respectively). No statistically significant differences were observed between one or two sacroiliac screws. The necessity of the posterior fixation of the SIJ additionally to the standard stabilisation of the PS, in case of YB APC-II injury, represents a topic under debate still to be confirmed. Van den Bosch et al25 performed a biomechanical study evaluating the posterior fixation additionally to the PS standard fixation in open book injuries. Its influence in the pelvic ring stability was analysed and it wascompared to the intact pelvis behaviour. The outcome of the study revealed that the SIJ fixation did not provide any greater stability, and that the behaviour of the injured pelvic ring was similar with respect to intact pelvis. The pelvic damage caused by these authors was potentially less severe than the one in the present work, in order not to injure the sacrotuberous, sacrospinous, and interosseous ligaments. Conversely, Simonian11 and Dujardin10determined that the complete injury of the interosseous ligament, is able to increase significantly the mobility of the SIJ, in case of YB APC-II type of injury. They concluded that the combination of anterior and posterior fixation is necessary to restore the physiological mobility of the pelvic ring. Regarding the biomechanical study carried out by our group, the isolated posterior fixation with ISS managed to restore the physiological mobility of the SIJ, which represents the main biomechanical fulcrum for the axial loads of the human body. With respect to the PS movements, the displacement was restored by 90%, with a minimal residual mobility restricted in a millimetric range. Given that the mobility of PS was reduced and restricted in a narrow range of motion following SIJ fixation and also that an up to 2.5 cm PS diastasis is considered relatively stable and can be conservatively treated 1,26,27, results herein presented suggest that the isolated posterior fixation with ISS may represent a therapeutic option in these type of injuries. Furthermore, the isolated fixation of the injured hemipelvis with a single ISS did not present significant differences regarding to the physiological mobility of the hemipelvis in all spatial axes. The addition of a second ISS was not Page 7 of 18 Accepted Manuscript associated with any substantial improvement. A registered and patented system was used to analyse experimentally the mechanical behaviour of injured pelvis20. The model was designed to allow the simulating standing upright position described by Simonian11 and Comstock28, so that the anterior superior iliac spine and PS tubers were aligned in the coronal plane and femurs could be placed in 15º anteversion and 10º valgus. Therefore, the axial load direction passes through the PS and provides a biomechanical model for studying the effects of sacroiliac fixation in the PS. The versatility of this system allows specimens to be positioned and set correctly despite having a distinct separation between femurs. These goals were achieved through the use of a proximal bearing, which corrects angles and allows the correct inclination of the pelvis and a rolling system of linear motion guide lower, which allows the distance between femurs to be adjusted. This study presents limitations related to the unfeasibility to reproduce the real conditions of the mechanism of injury of the patient in a laboratory environment. In order to minimize the limitations, fresh specimens were used to preserve the biomechanical characteristics of all the osteo-capsular and ligamentous structures. Additionally, the data analyzed were obtained at a single moment in time (cross-sectional study). Another potential limitation relates to the possible biomechanical stress modifications of the specimens due to the repeated axial load cycles. To minimize this bias, the overall stiffness was analysed between the different phases of the study and it was statistically verified that the mechanical capabilities were not affected by the load cycles12. Given the characteristics of the study and the use of human pelvises, it was not possible to select randomly the cadaveric specimens. In order to obtain a more homogeneous sample of data, all seven specimens used were females, but with heterogeneous ages. This condition can potentially influence the outcomes and lead to bias. Due to the nature of this study it was not possible either to evaluate in a later stage the biomechanical impact of the physiological healing process that would happen in real conditions after trauma. It is still unknown whether the reported pain after this kind of injuries and the development of SIJ arthritis is merely due to SIJ instability and whether mechanical stability is only to be achieved by rigid fixation of SIJ which in turn leads to arthritis despite excellent radiologic results. The discrepancy between radiological and functional outcomes in case of YB APC-II injuries, treated surgically by isolated anterior fixation, could be due to a residual SIJ microinstability10,12,28. The isolated fixation of the SIJ appears to show sufficient stability of the posterior elements of the pelvic ring, and could be able to re-establish the physiological biomechanical behaviour of the SIJ. Nevertheless, this type of fixation has limitations to control the anterior displacement, allowing residual movements in a region with less biomechanic solicitation as the PS. Despite this, prospective clinical studies are necessary to determine whether these minimum displacements of PS are relevant from the clinical perspective. REFERENCES. 4. Young JW, Burgess AR, Brumback RJ, et al. Pelvic fractures: value of plain radiography in early assessment and management. Radiology. 1986;160:445–451. 5. Rüedi T, Buckley R, Moran C. AO principles of fracture management, 2nd ed., vol 2. Page 8 of 18 Accepted Manuscript Stuttgart/New York: Thieme; 2007. 6. Tile M. Pelvic ring fractures: Should they be fixed? J Bone Joint Surg1998;70B:1-12. 7. Matta JM, Saucedo T. Internal fixation of pelvic ring fractures. ClinOrthop1989;242:83– 97. 8. Matta JM. Indications for anterior fixation of pelvic fractures. ClinOrthopRelat Res 1996; 329: 88-96. 9. Pohlemann T, Gänsslen A, Hartung S. Beckenverletzungen/Pelvic injuries. Heft zur der Unfallchirurg 266. Berlin, Heidelberg, New York: Springer-Verlag, 1998. 10. Rommens PM, Hessmann MH. Staged reconstruction of pelvic ring disruption: Differences in morbidity, mortality, radiologic results, and functional outcomes between B1, B2/B3, and C-type lesions. J Orthop Trauma 2002;16(2):92–108. 11. Rommens PM, Gercek E, Hansen M, Hessmann MH. Mortality, morbidity and functional outcome after open book and lateral compression lesions of the pelvic ring. A retrospective analysis of 100 type B pelvic ring lesions according to Tile’s classification. Unfallchirurg 2003;106(7):542–9. 12. Dujardin FH, Hossenbaccus M, Duparc F, et al. Long-term functional prognosis of posterior injuries in high-energy pelvic disruption. J Orthop Trauma. 1998;12:145–150. 13. Dujardin FH, Roussignol X, Hossenbaccus M, Thomine JM. Experimental study of the sacroiliac joint micromotion in pelvic disruption. J Orthop trauma.2002; 2:99-103. 14. Simonian PT, Routt LC Jr, Harrington RM, Mayo KA, Tencer AF. Biomechanical simulation of the anteroposterior compression injury of the pelvis. An understanding of instability and fixation. ClinOrthop. 1994; 309:245-256. 15. Cano-Luis P, Giráldez-Sánchez MA, Martínez-Reina J, Serrano-Escalante FJ, Galleguillos- Rioboo C, Lázaro-Gonzálvez A, Garcia-Rodríguez J, Navarro A. Biomechanical analysis of a new minimallyinvasive system for osteosynthesis of pubis symphysis disruption. Injury. 2012 Dec;43 Suppl 2:S20-7. 16. MacAvoy MC, McClellan RT, Goodman SB, Chien CR, Allen W, van der Muellen M. Stability of open-book pelvic fractures using a new biomechanical model of single-limb stance. J Orthop Trauma. 1997;11:590–593. 17. Papathanasopoulos A, Tzioupis C, Giannoudis VP, Roberts C, Giannoudis PV. Biomechanical aspects of pelvic ring reconstruction techniques: Evidence today. Injury. 2010;41:1220-1227. 18. Phieffer LS, Lundberg WP, Templeman DC. Instability of the posterior pelvic ring associated with disruption of the pubic symphysis. OrthopClin North Am. 2004 Oct;35(4):445-9, v. 19. Sagi HC, Papp S. Comparative Radiographic and Clinical Outcome of two-Hole and Multi-Hole Symphyseal Plating. J Orthop Trauma 2008;22:373–378. 20. Osterhoff G, Ossendorf C, Wanner GA, Simmen HP, Werner CM. Posterior screw fixation in rotationally unstable pelvic ring injuries. Injury2011;42(10):992–6. 21. Woo, S.L., Orlando, C.A., Camp, J.F., Akeson, W.H., 1986. Effects of postmortem storage by freezing on ligament tensile behavior. Journal of Biomechanics 19 (5), 399–404. 22. Varga E, Hearn T, Powell J, Tile M. Effects of method of internal fixation of symphyseal disruptions on stability of the pelvic ring. Injury.1995; 26 (2):75-80. 23. Serrano Escalante FJ, Giráldez Sánchez MA, Cano Luis P, García Rodríguez J, Navarro Robles A, Carranza Bencano A. Diseño y desarrollo de un modelo experimental para estudios biomecánicos de fracturas tipo B de Tile del anillo pélvico. Trauma Fund MAPFRE.2011:22(3):174-83. Page 15 of 18 Accepted Manuscript Table 4: Rotations (degrees) of the right iliac bone relative to the sacrum. Sagittal plane Transversal plane Coronal plane Subject A B C A B C A B C 1 -0.208 -0.757 -0.617 0.038 0.306 0.290 -0.004 -0.155 -0.133 2 -0.881 -0.491 -0.902 0.284 0.236 0.077 0.004 -0.013 0.092 3 -0.491 -0.289 -0.217 0.070 -0.230 -0.013 -0.016 -0.144 -0.101 4 -1.718 -1.114 -1.060 0.015 -0.172 0.074 0.180 -0.302 -0.270 5 -0.244 -0.709 -0.274 -0.446 0.007 -0.570 0.204 -0.210 -0.023 6 -1.097 -0.518 -0.552 0.067 0.071 0.197 0.090 -0.015 0.040 7 -0.844 -0.555 -0.766 0.182 0.253 0.213 -0.191 -0.110 -0.092 Median -0.844 -0.555 -0.617 0.067 0.071 0.077 0.004 -0.144 -0.092 Table Page 16 of 18 Accepted Manuscript Table 5: p-values of the pairwise comparisons for each displacement at the symphysis. Significant differences (applying Holm’s sequential Bonferroni correction) are indicated in boldface. Upper disp. Physiol.-1 Screw Upper disp. Physiol.-2 Screws Upper disp. 1Screw-2 Screws Lower disp. Physiol.-1 Screw Lower disp. Physiol.-2 Screws Lower disp. 1Screw-2 Screws .011 0.048 .776 .003 .005 .954 Table Page 17 of 18 Accepted Manuscript Table 6: Correction (%) between the markers placed at the symphysis. Correction at the upper symphysis Correction at the lower symphysis Subject B C B C 1 90,9% 92,7% 88,4% 85,5% 2 99,7% 98,5% 99,1% 98,6% 3 3,9% 47,3% 18,6% 51% 4 85,9% 85,7% 68,4% 66,1% 5 67,3% 68,5% 67,4% 69,9% 6 92,3% 98,1% 89,4% 95,1% 7 93,6% 85,9% 91,1% 84,7% Median, IQR 90,9% 85,9% 88,4% 84,8% IQR 16,4% 18,3% 22,4% 22,3% Table Page 18 of 18 Accepted Manuscript Figure