Simulated Cam Morphology of the Hip Changes Sacroiliac Motion During Hip Motion and Loading in a Cadaveric Model

Abstract

Purpose

To determine the relationship between cam morphology of the hip and ipsilateral sacroiliac motion compared to the native hip in a cadaveric model.

Methods

A simulated cam state was created using a 3-dimensional printed cam secured to the head-neck junction of 5 cadaveric hips. Hips were studied using a computed tomography–based optic metrology system and a 6 degree-of-freedom robot to exert an internal rotation torque at 3 different torque levels (6 N-m, 12 N-m, 18 N-m). Outcomes included translational and rotational movement about 3 axes and composite (total) translational motion at the ipsilateral sacroiliac (SI) joint. Statistical analysis included a linear mixed model regression with repeated measures.

Results

The presence of a simulated cam was associated with medial motion in the coronal plane (P = .03) and posterior motion in the sagittal plane (P < .01) but not composite motion (P = .37). Motion in the axial plane was in an inferior direction (P = .08). Cam morphology significantly changed rotation in the sagittal plane (P < .01) but not in the coronal (P = .63) or axial plane (P = .18). Composite motion was related to the amount of torque applied to the hip (P < .01). The amount of torque applied to the hip was related to rotation in the coronal plane (P < .01), axial plane (P < .01), and sagittal plane (P < .01) with increased effects as torque increased. Torque was not associated with translation movement in any of the anatomic planes.

Conclusions

The presence of simulated cam morphology is associated with motion in a more medial, inferior, and posterior direction at the ipsilateral SI joint relative to a native state. Increasing torque affects the magnitude of translation, but not its direction, which in this study is primarily influenced by cam morphology.

Clinical Relevance

This biomechanical connection between cam-type femoroacetabular impingement syndrome and the ipsilateral SI joint provides insight into SI joint dysfunction in patients with femoroacetabular impingement syndrome.

Femoroacetabular impingement syndrome (FAIS) is a common condition affecting young, active adults that may predispose patients to early hip osteoarthritis.¹ Recognition and treatment of FAIS have increased over the last decade and have resulted in improvement in functional outcomes for patients.^2,3

Cam morphology of the proximal femur changes the biomechanics of the hip joint.^4,5 This change in the loading pattern has been shown to affect movement at the pubic symphysis, and correction of cam morphology can reduce force at the chondrolabral junction.^6,7 The prevalence of other periarticular conditions such as osteitis pubis, ischiofemoral impingement, and sacroiliac disease in the setting of FAIS suggests that the altered biomechanics experienced in FAIS may predispose to adjacent pathology at other points of pelvic articulation.

The effect of cam morphology on areas elsewhere in the pelvis is supported by studies showing improvement in the T2 magnetic resonance imaging signal in many patients at the pubis after hip arthroscopy and cam resection.⁸ Although there is evidence to support a relationship between pubic symphyseal pathology and FAIS, literature investigating the relationship between FAIS and ipsilateral sacroiliac (SI) joint dysfunction is sparse, particularly as they relate to quantitative biomechanical data. There are observational data linking poor hip range of motion with low-back pain among various age groups, and the prevalence of radiographic SI joint abnormalities in patients with FAIS, and vice versa, ranges from 1 in 4 to 3 in 4.9, 10, 11, 12, 13, 14 This is of importance, as SI dysfunction may affect outcomes after surgical treatment for FAIS.¹⁵

Therefore, the purpose of the study was to determine the relationship between cam morphology of the hip and ipsilateral SI motion compared with a native hip in a cadaveric model. Our hypothesis was that the presence of cam-type FAIS would result in significant alterations in motion through the ipsilateral SI joint compared with a native, non-FAIS state.

Methods

After institutional review board and institutional human cadaveric use committee approval, human cadavers were screened for baseline arthritis and dysplasia using a standard fluoroscopy unit. Eighteen cadavers were available and thus screened. The primary criteria for exclusion were any cadaver with clear joint space narrowing of the hip, frank dysplasia, previous hip surgery, or previous SI joint surgery. Three cadavers were excluded for hip osteoarthritis, 3 were excluded because of having had previous hip surgeries, and 1 was excluded for having had frank hip dysplasia. After the initial round of exclusions, we then identified specimens with adequate femoral bone stock for instrumentation (at least 8 cm of bone distal to the lesser trochanter). Of these specimens with adequate femoral bone stock, we selected relatively younger specimens when possible. For instance, if 2 cadavers with similar bone stock were identified, we decided to test the 88-year-old cadaver rather than the 101-year-old cadaver. Eight subsequent patients were excluded on the basis of relative age and/or available bone stock for testing. Therefore, testing included 3 eviscerated human pelves with a total of 5 tested hips. There were 2 female cadavers and 1 male, with a mean age of 85 years (range, 75-93 years).

Before testing, each specimen was prepared for an optical metrology–based analysis. An array of 5 to 8 fiducial tracking landmarks were rigidly fixed to the right and left ilia, and sacrum. After placing of the fiducial landmarks, the specimens underwent a computed tomography scan. The right and left ilia and sacrum were then segmented into separate 3-dimensional (3D) mesh files with the fiducial landmarks included in each mesh (Fig 1). Tracking stickers were placed on the exposed tip of each landmark, allowing the landmarks to be tracked. Two 12-megapixel cameras equipped with Titanar 24-mm lenses (Aramis SRX; Trilion Quality Systems, King of Prussia, PA) were positioned such that each camera could see all tracking markers. With this arrangement, the optical metrology system could capture motion of each tracking marker in 3D space with an accuracy of 0.01 mm. The tracking marker arrays of the right and left ilia and sacrum enabled each bone to be tracked as a rigid body in 3D space using the Aramis software (Aramis Pro 2018; Trilion Quality Systems). After initial alignment, the position and orientation of the right and left ilia and sacrum can be tracked in 3D space with an accuracy of 0.04 mm. These tracked rigid bodies were used to quantify the relative motion between the sacrum and ilia.

Fig 1.

Three-dimensional computed tomography reconstruction of a pelvis demonstrating fiducial markers. The ovals denote the groups of fiducial markers used to identify the ilia and sacrum.

Anatomic coordinate systems, joint kinematics, and kinetics of the pelvis and femur were defined according to the recommendations of the International Society of Biomechanics.¹⁶ The anatomical landmarks used for defining the coordinate systems were digitized using a digitizing stylus and camera-based motion capture system (Optitrak; Corvallis, OR). The coordinate systems were defined by anatomical landmark digitization and were optimized by moving the hip through a flexion-extension movement cycle from 10° to 90° of flexion with all other forces and torques minimized. Six components of motion of the ilia relative to the sacrum were tracked: sagittal plane translation (x), axial plane translation (y), coronal plane translation (z), ilia coronal rotation (rotation about x), ilia axial rotation (rotation about y), and ilia sagittal rotation (rotation about z) (Fig 2).

Fig 2.

Coordinate system used to track relative sacroiliac translation and rotation. X-axis = red, Y-axis = green, and Z-axis = blue. Direction of arrows indacte positive direction for motion of a right ilia relative to the sacrum. Translation and rotational terms were defined. The X-axis translation is referred to as translation in the sagittal plane, whereas rotation around this is rotation in the coronal plane. The Y-axis translation along axial plane and rotation along this is in the axial plane. The Z-axis translation is along the coronal plane whereas rotation about this axis is in the sagittal plane.

Specimens were thawed according to anatomy laboratory protocol, and testing was performed at a similar time after thawing for each specimen. The skin and subcutaneous tissues were removed from the distal end of the exposed femoral shaft such that approximately 6 cm of bone was exposed. Each distal femur was then potted in 2.5-inch diameter tubes using a polyurethane resin (Smooth-On 300D, Macungie, PA). Hips were initially tested following a direct anterior approach with a longitudinal capsulotomy and subsequent repair. This was followed by a simulated cam state via the same approach and capsular repair. The longitudinal capsulotomy was repaired with numerous interrupted 5-0 ETHIBOND sutures in both experimental states. The simulated cam state was achieved using a 3D-printed simulated cam that was designed to contour along the anterior femoral head and neck and be placed at the 10-o’clock position and 2-o’clock position of right and left hips respectively, at the femoral head-neck junction (Fig 3). The simulated cam state was taken through a range of motion to ensure impingement. Cam size measured 3 mm tall by 24 mm wide with a beveled edge. The cam was secured with a countersunk 4.5-mm Depuy-Synthes cortical screw. The sacrum was augmented to withstand repeated testing with polymethylmethacrylate avoiding the SI articulation and ligamentous attachments. The pelvis was then secured via the sacrum to a rigid plate using a custom fixture such that the sacrum was fixed. Potted specimens were mounted to a 6 degree-of-freedom robotic system (KUKA, Augsburg, Germany) with one distal femur mounted to the robot’s manipulator equipped with a 6-axis load cell (Omega160 IP65; ATI Industrial Automation, Apex, NC) and the sacrum to a floor-mounted platform (Fig 4). The robot was controlled using simVITRO software (Cleveland Clinic, Cleveland, OH), which enables real-time, force-feedback control in anatomically relevant reference frames.

Fig 3.

(A) Photograph demonstrating the simulated cam positioned and fixed on a right femoral head-neck junction. The cam was 3-dimensionally printed and right/left specific to contour the head-neck junction at the 2:00-o’clock on a right hip and the 10:00-o’clock position on a left hip (cam positioned anteriorly at common location of cam). Dimensions of the cam included a height of 3 mm and diameter of 24 mm, the distal flange facilitated cam placement. (B) Photograph demonstrating the simulated cam positioned and fixed on a right femoral head-neck junction on a bone model.

Fig 4.

Experimental setup. This photograph demonstrates the experimental setup with the sacrum fixed to a pedestal and the right femur to the robotic arm. The optical metrology camera system is viewing from the right.

Hips were tested at 90° of flexion, neutral adduction, and at varying degrees of an internal rotation torque. The torques used included 6 N-m, 12 N-m, and 18 N-m, as these are documented approximations of hip internal rotation torque during activities of daily living, and these have previously been used to study motion of the symphysis in an FAIS model.^6,17 In addition, these torque levels were decided upon to represent a level of reasonable activity and to allow a specimen fixed via the sacrum to undergo repetitive testing without compromising bony integrity or fixation. The hip was internally rotated at a rate of 1°/sec until the target internal rotation torque was reached, at which point the joint was returned to its original position. For both the no-cam state and simulated cam surgical conditions, the hip was internally rotated to target torques of 6 N-m, 12 N-m, and 18 N-m with 1 trial completed for each target torque. The small number of trials performed on each specimen was due to concerns of disrupting the sacrum’s rigid fixation with repeated testing.

The primary outcome included translational and rotational motion in 3 planes in addition to composite translational motion through the ipsilateral SI joint in a cam and no-cam state. We secondarily aimed to compare changes in torque and any influence these changes have with respect to motion at the ipsilateral SI joint. Lastly, we analyzed possible interactions between the cam state and torque level as they relate to the primary outcome.

Statistical Analysis

Hips were analyzed individually as the resultant SI motion varied between hips. The effect of cam and torque on the study outcomes was analyzed with a linear mixed model regression with repeated measures on both factors. Throughout the analysis, no interaction between cam and torque level was identified. Thus, analysis was undertaken using an appropriate multiple comparisons procedure to control the overall type-I error rate while evaluating pairwise differences between the 3 torque conditions and the cam and no-cam conditions. All analyses were performed with BlueSky, version 10.4.1 (Chicago, IL). All statistical tests were 2-sided, and P values less than .05 were be considered statistically significant.

Results

SI Translation

The presence of a cam was not associated with composite motion (P = .37) but was associated with medial motion in the coronal plane (P = .03) and posterior motion in the sagittal plane (P < .01). Motion in the axial plane was not statistically significant (P = .08) (Table 1). Composite motion was related to the amount of torque applied to the hip (P < .01). The effect of torque increased from 12 N-m to 18 N-m. Torque was not associated with translation movement in the coronal plane (P = .88), axial plane (P = .93), or the sagittal plane (P = .59) (Table 2). Normalized net movement in each individual plane in a cam state relative to the specimen-matched native state is demonstrated in Figure 5.

Table 1.

The Effect of Cam Morphology of the Proximal Femur on Translational and Rotational Movement at the SI Joint During Rotation of the Hip

Composite Translation, mm		Translation in the Coronal Plane, mm		Translation in the Axial Plane, mm		Translation in the Sagittal Plane, mm
Effect Size	P Value	Effect Size	P Value	Effect Size	P Value	Effect Size	P Value
–0.12	.35	–0.31	.028	–0.34	.076	–0.33	.005
Rotation in the Coronal Plane, °			Rotation in the Axial Plane, °		Rotation in the Sagittal Plane, °
Effect Size	P Value		Effect Size	P Value	Effect Size		P Value
0.03	.63		0.05	.18	–0.07		.006

NOTE. The effect size refers to the estimated direction of movement as a result of a cam state. The sign of the effect size refers to the direction of movement. For example, translation in the sagittal plane is 0.33 mm posterior (along the sagittal plane with a positive number being anterior and a negative number being posterior).

SI, sacroiliac.

Table 2.

The Effect of Internal Rotation Torque of the Hip on Translational and Rotational Movement at the SI Joint

Composite Translation, mm				Translation in the Coronal Plane, mm				Translation in the Axial Plane, mm				Translation in the Sagittal Plane, mm
12 N-m IR Torque		18 N-m IR Torque		12 N-m IR Torque		18 N-m IR Torque		12 N-m IR Torque		18 N-m IR Torque		12 N-m IR Torque		18 N-m IR Torque
Effect Size	P Value	Effect Size	P Value	Effect Size	P Value	Effect Size	P Value	Effect Size	P Value	Effect Size	P Value	Effect Size	P Value	Effect Size	P Value
0.61	<.001	1.04	<.001	–0.17	.27	–0.21	.22	–0.15	.48	–0.18	.44	–0.03	.84	0.01	.94

Rotation in the Coronal Plane, °				Rotation in the Axial Plane, °				Rotation in the Sagittal Plane, °
12 N-m IR Torque		18 N-m IR Torque		12 N-m IR Torque		18 N-m IR Torque		12 N-m IR Torque		18 N-m IR Torque
Effect Size	P Value	Effect Size	P Value	Effect Size	P Value	Effect Size	P Value	Effect Size	P Value	Effect Size	P Value
–0.32	<.001	–0.59	<.001	0.23	<.001	0.38	<.001	0.10	.002	0.03	<.001

NOTE. Effect size refers to the effect of 12 N-m compared with 6 N-m and 18 N-m with 6 N-m, i.e., the effect of increasing torque to a given level up from 6 N-m.

IR, internal rotation; SI, sacroiliac.

Fig 5.

Motion at the SI joint in the cam state relative to a specimen matched normal state. Movement away from 0 on the Y-axis represents mean movement in the cam-state relative to native state across specimens. (A) Graphical representation of composite SI joint motion with internal rotation torque (N-m) on the X-axis and translation (mm) on the Y-axis. The black line represents the difference in motion in the cam state relative to the normal state, which is denoted by the straight line at 0 on the Y-axis. Vertical bars represent the 95% CI. (B) Graphical representation of SI joint motion in the sagittal plane (blue), axial plane (red), and coronal plane (green) with internal rotation torque (N-m) on the X-axis and translation (mm) on the Y-axis. Vertical bars represent the 95% CI. (C) Graphical representation of SI joint rotation in the coronal plane (blue), axial plane (red), and sagittal plane (green) with internal rotation torque (N-m) on the X-axis and rotation (degrees) on the Y-axis. Vertical bars represent the 95% CI. (CI, confidence interval; SI, sacroiliac.)

SI Rotation

Rotation in the coronal plane was not statistically related to the presence of a cam (P = .65) nor in the axial plane (P = .18) but was related to rotation in the sagittal plane (P = .007) (Table 1). Coronal plane rotation was related to the amount of torque applied (P < .001), this effect increased from 12 N-m to 18 N-m. Axial plane rotation was related to the amount of torque applied to the hip (P < .001), this effect increased from 12 N-m to 18 N-m. Sagittal plane rotation was related to the amount of torque applied to the hip (<.001) and this effect increased with increasing torque (Table 2). Normalized net rotation in each individual plane in a cam state relative to the specimen matched native state is demonstrated in Figure 5.

Discussion

The major findings of this study demonstrate that the cam morphology of the proximal femur changes the direction of motion of the ipsilateral SI joint. When the cam state data are taken together, the presence of a cam pushes the ilium more medial, inferior, and posterior while not changing total composite movement. Of note, torque plays a role primarily in total composite motion in addition to changes in the way the ilia rotate about the sacrum during loading. These findings are clinically relevant in that they suggest that normalization (resection) of cam morphology may lead to improvements in postoperative SI biomechanics. As such, patients with clinically symptomatic SI dysfunction should be evaluated for the presence of cam morphology and, conversely, patients undergoing FAIS surgery may, in turn, be counseled on potential improvements in SI mechanics and symptoms after cam resection.

Other studies have evaluated the effect of cam type FAIS on the biomechanical loading and motion of the hip and pelvis. Ng et al.,⁷ in a study of skeletonized cadaver hips, were able to determine experimentally that approximately 25% of the resistance to flexion and internal rotation could be removed with cam resection. Birmingham et al.,⁶ in a simulated cam model, demonstrated that a simulated cam state affected pubic symphyseal motion. Dynamic fluoroscopy evaluating pelvic and hip kinematics has also shown that there are subtle differences in pelvic and hip motion in FAIS patients compared with normal controls.⁴ Notably, patients with FAIS typically stand with their hips in a slightly extended position and have less anterior pelvic tilt during heel strike compared with control patients.⁴ However, a rigorous evaluation of a biomechanical link between femoral cam morphology in FAIS and the SI joint is lacking in the literature. Our results suggest that the presence of cam morphology of the proximal femur pushes the ilium away from a typical axis of load. That is, the ilium moves in, down, and back from the sacrum. This change in motion may result in abnormal SI joint load distribution and stress on the posterior and inferior ligaments, resulting in clinical symptoms.

This biomechanical link has been suggested clinically, as SI joint abnormalities or pain are common in patients presenting with FAIS.¹² In a study of more than 1,000 consecutive patients undergoing hip arthroscopy for FAIS, approximately 25% of patients demonstrated radiographic SI joint abnormalities.¹² More importantly, these patients had worse functional scores and were less likely to meet the minimum clinical important difference status postarthroscopy.¹² Separately, in a matched-cohort study comparing patients with SI joint pain in the setting of FAIS with those without, those with SI joint pain had lower patient-reported outcomes both preoperatively and postoperatively.¹⁵ Despite this, the improvement in patient-reported outcome scores was proportionally the same between the SI pain group and the control group, suggesting that patients with SI joint dysfunction in the setting of FAIS syndrome benefit from hip arthroscopy similarly to those without SI joint dysfunction.¹⁵ In our group’s experience, patients who have SI pain in the setting of concomitant hip pathology often experience improvement of SI pain after treatment of hip pathology, especially in the presence of large cam lesions with broad loss of femoral head-neck offset. However, this remains anecdotal, highlighting the potential utility of future clinical studies. This clinical improvement may relate to correcting abnormal stresses at the SI joint with removal of an aggravating stimulus (whether it be FAIS or hip arthritis).

Inherent to studying SI joint motion includes the conceptualization of lumbopelvic motion and load transfer in the setting of FAIS syndrome. Outside of the growing literature on the “hip-spine syndrome” as it relates to arthroplasty and spine surgery, there is generally a paucity of clinical or experimental evidence linking prearthritic hips and FAIS with spine problems or motion. In the hip arthroplasty setting, there are data to suggest low back pain can significantly improve after total hip arthroplasty in those presenting with pain in both areas,^18,19 whereas in the prearthritic population, 3 clinical studies show a correlation between young athletes with decreased hip range of motion and low back pain.9, 10, 11 One experimental study does suggest that reduced lumbar spine mobility places an FAIS hip into a position where impingement is more likely.²⁰ However, further research will be required to better evaluate whether improvements in range of motion of the hip and/or altered transfer of lumbopelvic stresses after treatment of FAIS can improve SI joint and low-back pain.

Apart from the effect of cam morphology, the amount of torque played a role in SI joint motion. This finding is intuitive. Although no interaction was identified between torque and cam (i.e., the effect of torque on SI joint motion was not altered by the presence of a cam and vice versa), this may be the result of inadequate power to detect the interaction. Figure 5 demonstrates the normalized motion relative to a “zeroed” control line. Observing these trends, there is a suggestion that with increasing torque, the effect of the cam does get larger, though this was not borne out statistically. More specimens may be needed to further delineate this relationship.

Limitations

The main limitations of this study include the inherent limitations of translating biomechanical evaluations to clinical consequences, the simulated cam state, and the age of the cadavers. The precision of our experimental instrumentation allowed us to detect small movements about the SI joint, and the clinical conclusions of such small movements may be limited. The simulated cam does not perfectly replicate normal joint mechanics and thus the findings may not be generalized to an organic cam versus no-cam or postresection state. In addition, the physiologic characteristics of the hips studied may not be representative of a younger, active population. Another limitation is that we did not evaluate the native hip head-neck offset, the version of the proximal femur, acetabular version, or anterior inferior iliac spine differences; these may reasonably alter the magnitude of difference between the native and cam states. However, because specimens were analyzed individually with repeated measures, these parameters are effectively controlled for within each hip. Lastly, our study may have been underpowered to detect a difference in some of the outcome variables.

Conclusions

The presence of simulated cam morphology is associated with motion in a more medial, inferior, and posterior direction at the SI joint relative to a native state. Increasing torque affects the magnitude of translation, but not its direction, which in this study is primarily influenced by cam morphology.

Disclosures

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Funding was provided by the Mayo Clinic Orthopaedic Research Review Committee and support from the Foderaro-Quattrone Musculoskeletal-Orthopaedic Surgery Research Innovation Fund. W.W.C. reports IP royalties and paid consultant for OsteoCentric. A.J.K. reports research support from Aesculap/B. Braun; editorial or governing board of American Journal of Sports Medicine; IP royalties and paid consultant for Arthrex; and board or committee member for the Arthroscopy Association of North America and International Cartilage Repair Society. M.H. reports paid consultant for DJO-Enovis; publishing royalties and financial or material support from Elsevier; editorial or governing board of the Journal of Cartilage and Joint Preservation; and paid consultant for Moximed and Vericel. All other authors (M.E.U., A.W.H, M.A., Z.V.B., M.J.N., E.M.D.) declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.