Original Article
Does cam osteochondroplasty compromise proximal femur strength?
Proc IMechE Part H: J Engineering in Medicine 2014, Vol. 228(12) 1235–1240 Ó IMechE 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0954411914561051 pih.sagepub.com
Chandni Nigam1, Milad Masjedi2, James Houston2, Charles Marquardt2, Adeel Aqil2 and Justin Cobb2
Abstract Little is known about the effect on load bearing ability of cam-type femurs following osteochondroplasty. The aim of this study was to compare the change in deformation undergone by cam-type femoral acetabular impingement femur models after resection of different volumes. Dry-bone replicas (N = 10) of two cam-type femurs (cam A and B) underwent resections of increasing volume (Surgery I, II and III) representing conservative, adequate and radical resections. Deformation under cyclic loading of 700 N for five cycles after each surgery was compared. The 360° alpha angle and the change in head to neck ratio at four equidistant points along the femoral neck were used as measures of surgical efficacy and volume resected. Intact cam A and B replicas had a maximum alpha angle of 88° and 90°, respectively, which were reduced to 55° and 54° post Surgery I. Cam A replicas showed a significant reduction (p \ 0.01) in mean axial displacement after Surgery I (up to 10% reduction in neck volume) and an increase after Surgery III (~20%–40% reduction in neck volume) compared to unresected controls (p \ 0.01). Surgery II (~10%–15% reduction in neck volume) produced no significant change in mean displacement (p . 0.05). Cam B models exhibited lower mean displacement after Surgery I, II and III (p \ 0.01) compared to unresected controls. Conservative surgery appears to improve the axial load bearing ability of dry-bone models. Radical resections may significantly decrease the fracture-resistant properties of bone following osteochondroplasty which should be noted when planning such a procedure.
Keywords Cam femoral acetabular impingement, osteochondroplasty, femoral strength, fracture risk, resection volume, head–neck ratio, alpha angle
Date received: 11 June 2014; accepted: 3 November 2014
Background Femoral acetabular impingement (FAI) is a degenerative hip pathology that presents with bony deformity of the femoral head–neck junction or acetabulum. It is linked to high levels of sporting activity during childhood1 and is regarded as a cause of premature onset osteoarthritis of the hip.2 It is a painful condition associated with reduced range of motion in the hip and damage to the labrum from abutment of the bony deformity affecting the femoral head and neck.2 Osteochondroplasty surgery aims to remove the impinging bone, thereby restoring the patient’s range of motion and relieving pain caused by the impingement.2 Little is known about the effect of the bone resection on proximal femur strength. When performing osteochondroplasty, complications can arise from resecting either too much or too little bone. On one hand, if too much bone is removed, there is the implied decrease in the strength of the femoral neck. Mardones et al.3
found that the upper boundary of safe resection was 30% of the femoral neck diameter (measured twodimensionally). On the other hand, if too little bone is removed, a revision may be required to fully correct the FAI and improve range of motion.4 Revision brings further risks to the patient that include nerve injuries, fluid extravasation, avascular necrosis and inconvenience to the patient.5 Due to patients having high functional requirements, in that those most commonly affected by cam-type pathology are highly active in
1
Imperial College School of Medicine, Imperial College London, London, UK 2 MSK Lab, Department of Orthopaedics, Imperial College London, Charing Cross Hospital, London, UK Corresponding author: Chandni Nigam, Imperial College School of Medicine, Imperial College London, London SW7 2AZ, UK. Email: [email protected]
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
1236
Proc IMechE Part H: J Engineering in Medicine 228(12)
sports, it is important that neither femoral neck strength nor range of motion is compromised. The force required to break the femoral neck depends on various factors including bone mineral density and geometry of the femoral neck. In cases of osteoporosis, occurring most commonly in the elderly, the force required to fracture the femur is lower as bone mineral density decreases.6 In addition, age-related microfractures can propagate and accelerate deterioration in bone strength where their formation exceeds repair processes7 increasing skeletal fragility.8 The effect of the geometry of the femoral neck, however, has not been consistently reported in the literature. Some studies have shown that neck width and fracture risk are positively correlated – generally acknowledged to be owing to a compensatory mechanism in the bone that increases in size but compromises on mineral density.9–11 While other studies have shown that thinner femoral necks are more prone to fracture.3,12 The instance of fracture directly attributed to osteochondroplasty is essentially nil; therefore, we do not pose that cam resection will create an immediate and significant fracture risk. There is, however, a logical conclusion that neck over-resection in a middle aged patient may manifest itself later in life and increase the patient’s risk of fracture, as bone mineral density decreases and the occurrence of microfractures increases.7 In this study, we evaluated femoral neck strength following cam surgery.
Figure 1. Images of the unresected dry-bone models bones cam A (left) and cam B (right).
Methods Computed tomography (CT) scans of two subjects with cam-type deformity were used to make proximal drybone femur models, cam A and cam B (Medical Models Ltd, Bristol, UK). Both cam A and B represented severe cam deformities present on the antero-superior aspect of the femur; however, both femurs differed in their overall morphology of the head and neck (Figure 1). Mechanical load testing was carried out on five drybone replicas each of cam A and B. The external shape of the femoral shaft was used as a template to produce custom clamping apparatus for each cam-type model using a three-dimensional (3D) printer (Objet, Rehovot, Israel). The clamp angled the femurs to 12° flexion from the vertical to recreate the natural angle of highest loading on the femoral head at walking speeds13 and to restrict femoral shaft displacement over neck displacement. An artificial acetabular surface, based on the radius of the femoral head, was also custom made for the cam-type femur models using the 3D printing. A cam model in the custom clamp was fixed into an axial loading apparatus (MultiTest 10-i; Mecmesin, Slinfold, England) (Figure 2). The MultiTest 10-i is a twin-column tensile and compression test system with a loading capacity up to 10 kN. It has a load resolution of 1:6500 and a load cell accuracy of 60.1% of full scale from 2 to 2500 N and a speed range of 1–1000 mm/min.
Figure 2. Setup of bone model in Mecmesin loading machine.
A load control method was used whereby a cyclic axial force between 10 and 700 N was applied to the femoral head at a rate of 5 mm/min for five cycles, while the displacement was recorded to a resolution of 60.01 mm. This specific loading pattern was based on preliminary results that showed 700 N to be the maximum force that the specimens could withstand without undergoing plastic deformation or failure. In addition, the relationship between fracture strength of the drybone models and 700 N was comparable to the relationship between the fracture strength of cadaveric bone and the natural loads experienced during gait.
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
Nigam et al.
1237
Figure 3. Diagrammatic representation of the alpha angle (a) (left) and four equally spaced points along the femoral head radius (R) where head–neck ratio (HNR) was evaluated (R/4, R/2. 3R/4, R) (right); 0 denotes centre point of the femoral head.
Subsequently, each bone model underwent three separate levels of resection: Surgery I, II and III equating to conservative, adequate and radical resections, respectively. First, a surgery judged to be standard but conservative for osteochondroplasty was performed by an orthopaedic surgeon (Surgery I). Open template guides were used for each subsequent resection (Surgery II and III), moving the margins of resection 5 mm inferior and 5 mm posterior each time. The open templates allowed the surgeon a good visual on the volume to be resected for each surgery. Following each resection, the models were remounted in the Mecmesin apparatus and the displacement was recorded under cyclical loading, as per the method outlined above. The displacement was compared between surgeries for each dry-bone model. Following each resection, the specimens were also scanned using NextEngine Desktop 3D Scanner (NextEngine, Santa Monica, CA, USA) and virtual models their 3D surface were created. These were input into custom-written software14 to calculate alpha angles, 360° around the head–neck junction after each surgical resection.15 The alpha angle is the angle between the longitudinal axis of the neck and the imaginary straight line joining the femoral head centre to the point on the head–neck junction where the surface contour exceeds the femoral head radius (Figure 3). Its value indicates the extent of deformity; less than 55° is considered normal range for the alpha angle.16 The software also facilitated the measurement of the 3D head– neck ratio (HNR) at four points along the femoral head radius (R) moving distally from the femoral head centre (Figure 3), the values of which were summated.17 The HNR is the ratio of the cross-sectional area of the femoral head to that of the femoral neck.
Statistical tests Values for displacement of the femur models acquired under the loading cycles did not display normal
Table 1. Mean 6 standard deviation of alpha angles (°) for cam A and cam B at unresected and after each surgery measured in the area of impingement (superior–anterior aspect) of the femoral head.
Unresected Surgery I Surgery II Surgery III
Cam A (°)
Cam B (°)
88 55 6 3 46 6 4 40 6 4
90 54 6 4 45 6 2 41 6 4
distribution. Therefore, a two-tailed Wilcoxon signedrank test for paired samples was used to determine the significance of changes in axial deformation, comparing unresected models to each subsequent surgery.
Results Alpha angle measurements extracted from the customwritten software showed a maximum alpha angle of 88° and 90° for the unresected cam A and B models, respectively (Table 1). Mean alpha angles decreased with each subsequent surgery in both cam A and B, as would be expected when the impingement is progressively corrected by the surgery (Table 1). After conservative surgery (Surgery I), the mean alpha angle is 55° 6 3° for cam A replicas and 54° 6 4° for cam B. Placing unresected cam femurs under cyclical loading produced closed hysteresis loops with variation in the value of peak axial displacement. Cycle 1 was excluded from all analyses as there was an initial linear portion in this cycle separate from the hysteresis loop under axial loading representing slippage between the bone model and loading components. Cycle 5 was also excluded as it was incomplete. Unresected cam A models had a mean displacement value across cycles 2–4 of 1.5 60.2 mm, while the mean
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
1238
Proc IMechE Part H: J Engineering in Medicine 228(12)
Figure 4. Bar graph representing mean displacement (cycles 2– 4) in unresected models and with three levels of resection (Surgery I, II and III) for (a) cam A and (b) cam B specimens. *p \ 0.01 statistical significance of displacement value compared with unresected model.
for unresected cam B models was greater at 2.7 6 0.2 mm. Bar graphs representing comparison of mean displacement between the unresected models and the three surgeries are shown in Figure 4(a) and (b). Both cam A and B models exhibited a decrease in mean displacement after Surgery I compared to that of unresected models (9.5% and 4.8% decrease in displacement for cam A and B, respectively). In cam A models, Surgery II caused a non-significant change in axial displacement compared to unresected controls. Surgery III represented the greatest displacement observed in the cam A models, with the mean displacement being 49% greater than that of unresected models, whereas in cam B, there was a 3% reduction in mean displacement compared to the intact cam B controls. Finally, mean displacement for each surgery was plotted against percentage change in the sum of the HNRs to determine the relationship between the two variables shown in Figure 5. Figure 5 shows a general decrease in the mean displacement for cam A after Surgery I, where up to 10% of the femoral neck has been removed (p \ 0.01, based on the change in the sum of HNR). Mean displacement rises to above that experienced by unresected cam A models after Surgery III (p \ 0.01), representing the greatest level of resection in this study and corresponds to more than 20% reduction in neck volume. However, there appears to be no correlation between mean
Figure 5. Mean displacement against percentage change in head–neck ratio (HNR) values compared to unresected model for (a) cam A and (b) cam B. HNR was calculated at four points along the femoral head radius and summated for each model. Missing data points are the result of inadequate laser scanning of some models.
displacement and change in the HNR for cam B (Figure 5(b)). Displacement remains similar after each surgery despite statistical significance of the mean displacement values compared to unresected (Figure 4(b)). The difference between the minimum (Surgery I) and maximum (Surgery III) mean displacement for cam B is 0.06 mm compared to 0.72 mm for cam A.
Discussion In this study, two sets of femur models with cam lesions were subjected to cyclical loading up to 700 N after three levels of resection (Surgery I, II and III) representing conservative, adequate and radical resection, to evaluate the change in femoral neck strength. Plotting axial displacement against load produced hysteresis loops with non-overlapping loading and unloading curves representing dissipated energy and viscoelastic behaviour of the material, also exhibited by natural bone.18 Bone deformity seems to be strongly related to external bony morphology. For cam A, bone resection affected bone strength with statistically significant differences in deformation when comparing the
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
Nigam et al.
1239
unresected controls with Surgery I and III. Surprisingly, Surgery I (5%–10% reduction in the neck cross-sectional area) was associated with a lower axial displacement than that for unresected models (p \ 0.01) (Figure 4). The reduction in the alpha angle to a normal range ( \ 55°) represents adequacy of conservative resection. This potentially leads to lower microcrack development which, once it has occurred, can reduce the load intensity and number of cycles required for crack propagation and bone failure,19 and in this way, microdamage reduces fracture-resistant properties of bone.20 However, any loss in strength in this instance of decreased deformation (Surgery I) may not be apparent with the downward axial force that was applied. Further studies with multidirectional deformation measurement capabilities could analyse such hypotheses in more detail. In this study, radical resection in cam A models, which removed up to 40% of the femoral neck volume, may have been a direct cause of the increased deformation observed, compared to intact cam A models, due to the drastic change in femoral neck geometry.3,12 Whether the increased displacement observed with this surgery is clinically sufficient to increase the chances of fracture is unknown. Similar studies have provided evidence of a significant reduction in load to failure with increasing depth or area of resection.21,22 Wijdicks et al.22 used a larger sample size of 30 specimens to evaluate the biomechanical parameters of composite femurs following cam resection with additional iatrogenic notching. In the study by Wijdicks, cam resection which reduced the alpha angle from 61° to 45° did not alter the mean ultimate load, stiffness or energy to failure compared to controls. Additional resection of the femoral neck of depths equal to or greater than 4 mm and width of 5.5 mm caused a statistically significant decrease ( . 17%) in the load to failure. However, these studies were limited by their ability to quantify the resections made to the bone models in only two dimensions and did not give consideration to the severity of cam impingement, as done in this study. In contrast to cam A femurs, the results from the cam B femurs suggest that there is little change in displacement when comparing resected to unresected bone even when approximately 30% of the femoral neck had been removed (Figure 5). An appreciable change may be detectable with a greater level of resection; however, 30% greatly exceeds what would normally be removed in a real-world surgical setting. Unlike previous studies, which stated a safe 30% removal zone,3 we think the margin of safe resection is subject specific depending on the morphology of the femoral bone. The discrepancy between cam A and B results suggests that the morphology of the cam deformity present influences the load bearing ability of the bone. Our study was non-cadaveric and used dry-bone models customised to replicate actual bone structure (cortical and trabecular) and shape. Previous studies have shown that plastic models can simulate the
mechanical properties of natural bone under elastic deformation.23 Peak force observed in this study was 700 N, whereas strenuous activities, such as running, can create a maximum force of 11.4 times body weight;24 therefore, a range of stress values should be investigated. This study repeatedly loaded the same bone following each bony resection. Although the loading pattern chosen of 700 N is unlikely to have altered the mechanical properties of the models via plastic deformation, the cyclic nature of the loading alongside surgical resection may have led to increased deformation. However, the decrease in mean displacement following conservative resections (Surgery I) suggests that the neck volume is the determining factor for neck strength. There was great variability in the percentage change in the HNR with each surgery, and therefore, consistency in bone resection would have resulted in a more controlled setup. Future studies can perhaps use robotic technology to control this variable.25 Femoral osteochondroplasty could lead to reduced bone strength, which is potentially serious if compounded with age-associated osteoporosis. This study demonstrates that radical neck resections may compromise the integrity of femoral neck strength due to drastic changes in femoral neck geometry. Conservative surgery results in the most favourable results in terms of axial force bearing capacity of bone models. Further studies are warranted to investigate the clinical significance of the change in displacement in multidirectional settings with an aim of changing the probability of fracture. Declaration of conflicting interests The authors declare that they have no conflict of interest. All research was performed in accordance with the Declaration of Helsinki and all local, regional and national law. Funding This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) (grant no. WSSU/PS1203). References 1. Siebenrock K, Ferner F, Noble P, et al. The cam-type deformity of the proximal femur arises in childhood in response to vigorous sporting activity. Clin Orthop Relat Res 2011; 469(11): 3229–3240. 2. Chakraverty J and Snelling N.Anterior hip pain – have you considered femoroacetabular impingement? Int J Osteopath Med 2012; 15(1): 22–27. 3. Mardones RM, Gonzalez C, Chen Q, et al. Surgical treatment of femoroacetabular impingement: evaluation of the effect of the size of the resection. J Bone Joint Surg Am 2005; 87(2): 273–279. 4. Ilizaliturri V Jr.Complications of arthroscopic femoroacetabular impingement treatment: a review. Clin Orthop 2009; 467(3): 760–768.
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
1240
Proc IMechE Part H: J Engineering in Medicine 228(12)
5. Byrd JT.Complications associated with hip arthroscopy. In: JT Byrd (ed.) Operative hip arthroscopy. New York: Springer, 2005, pp.229–235. 6. Muldoon MP, Padgett DE, Sweet DE, et al. Femoral neck stress fractures and metabolic bone disease. J Orthop Trauma 2001; 15(3): 181–185. 7. Fazzalari NL.Trabecular microfracture. Calcif Tissue Int 1993; 53(1): S143–S147. 8. Burr DB, Forwood MR, Fyhrie DP, et al. Bone microdamage and skeletal fragility in osteoporotic and stress fractures. J Bone Miner Res 1997; 12(1): 6–15. 9. Mikhail M, Vaswani A and Aloia J.Racial differences in femoral dimensions and their relation to hip fracture. Osteoporosis Int 1996; 6(1): 22–24. 10. Gnudi S, Ripamonti C, Gualtieri G, et al. Geometry of proximal femur in the prediction of hip fracture in osteoporotic women. Br J Radiol 1999; 72(860): 729–733. 11. Dincxel VE, S xengelen M, Sepici V, et al. The association of proximal femur geometry with hip fracture risk. Clin Anat 2008; 21(6): 575–580. 12. Cheng X, Lowet G, Boonen S, et al. Assessment of the strength of proximal femur in vitro: relationship to femoral bone mineral density and femoral geometry. Bone 1997; 20(3): 213–218. 13. Bergmann G, Deuretzbacher G, Heller M, et al. Hip contact forces and gait patterns from routine activities. J Biomech 2001; 34(7): 859–871. 14. Masjedi M, Nightingale CL, Azimi DY, et al. The threedimensional relationship between acetabular rim morphology and the severity of femoral cam lesions. Bone Joint J 2013; 95-B(3): 314–319. 15. Masjedi M, Azimi DY, Nightingale CL, et al. A method of assessing the severity of cam type femoro-acetabular impingement in three dimensions. Hip Int 2012; 22(6): 677–682.
16. Neumann M, Cui Q, Siebenrock KA, et al. Impingementfree hip motion: the ‘normal’ angle alpha after osteochondroplasty. Clin Orthop 2009; 467(3): 699–703. 17. Masjedi M, Marquardt CS, Drummond IM, et al. Cam type femoro-acetabular impingement: quantifying the diagnosis using three dimensional head-neck ratios. Skeletal Radiol 2013; 42(3): 329–333. 18. Lakes RS, Katz JL and Sternstein SS. Viscoelastic properties of wet cortical bone – I. Torsional and biaxial studies. J Biomech 1979; 12(9): 657–678. 19. An YH and Draughn RA.Mechanical testing of bone and the bone-implant interface. Boca Raton, FL: CRC Press, 1999. 20. Zioupos P.Accumulation of in-vivo fatigue microdamage and its relation to biomechanical properties in ageing human cortical bone. J Microsc 2001; 201(2): 270–278. 21. Davis ET, Olsen M, Zdero R, et al. A biomechanical and finite element analysis of femoral neck notching during hip resurfacing. J Biomech Eng 2009; 131(4): 041002. 22. Wijdicks CA, Balldin BC, Jansson KS, et al. Cam lesion femoral osteoplasty: in vitro biomechanical evaluation of iatrogenic femoral cortical notching and risk of neck fracture. Arthroscopy 2013; 29(10): 1608–1614. 23. Landsman AS and Chang TJ.Can synthetic bone models approximate the mechanical properties of cadaveric first metatarsal bone?J Foot Ankle Surg 1998; 37(2): 122–127; discussion 172. 24. Edwards WB, Gillette JC, Thomas JM, et al. Internal femoral forces and moments during running: implications for stress fracture development. Clin Biomech 2008; 23(10): 1269–1278. 25. Masjedi M, Tan WL, Jaskaranjit S, et al. Use of robotic technology in cam femoroacetabular impingement corrective surgery. Int J Med Robot 2013; 9(1): 23–28.
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
Does cam osteochondroplasty compromise proximal femur strength?
Proc IMechE Part H: J Engineering in Medicine 2014, Vol. 228(12) 1235–1240 Ó IMechE 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0954411914561051 pih.sagepub.com
Chandni Nigam1, Milad Masjedi2, James Houston2, Charles Marquardt2, Adeel Aqil2 and Justin Cobb2
Abstract Little is known about the effect on load bearing ability of cam-type femurs following osteochondroplasty. The aim of this study was to compare the change in deformation undergone by cam-type femoral acetabular impingement femur models after resection of different volumes. Dry-bone replicas (N = 10) of two cam-type femurs (cam A and B) underwent resections of increasing volume (Surgery I, II and III) representing conservative, adequate and radical resections. Deformation under cyclic loading of 700 N for five cycles after each surgery was compared. The 360° alpha angle and the change in head to neck ratio at four equidistant points along the femoral neck were used as measures of surgical efficacy and volume resected. Intact cam A and B replicas had a maximum alpha angle of 88° and 90°, respectively, which were reduced to 55° and 54° post Surgery I. Cam A replicas showed a significant reduction (p \ 0.01) in mean axial displacement after Surgery I (up to 10% reduction in neck volume) and an increase after Surgery III (~20%–40% reduction in neck volume) compared to unresected controls (p \ 0.01). Surgery II (~10%–15% reduction in neck volume) produced no significant change in mean displacement (p . 0.05). Cam B models exhibited lower mean displacement after Surgery I, II and III (p \ 0.01) compared to unresected controls. Conservative surgery appears to improve the axial load bearing ability of dry-bone models. Radical resections may significantly decrease the fracture-resistant properties of bone following osteochondroplasty which should be noted when planning such a procedure.
Keywords Cam femoral acetabular impingement, osteochondroplasty, femoral strength, fracture risk, resection volume, head–neck ratio, alpha angle
Date received: 11 June 2014; accepted: 3 November 2014
Background Femoral acetabular impingement (FAI) is a degenerative hip pathology that presents with bony deformity of the femoral head–neck junction or acetabulum. It is linked to high levels of sporting activity during childhood1 and is regarded as a cause of premature onset osteoarthritis of the hip.2 It is a painful condition associated with reduced range of motion in the hip and damage to the labrum from abutment of the bony deformity affecting the femoral head and neck.2 Osteochondroplasty surgery aims to remove the impinging bone, thereby restoring the patient’s range of motion and relieving pain caused by the impingement.2 Little is known about the effect of the bone resection on proximal femur strength. When performing osteochondroplasty, complications can arise from resecting either too much or too little bone. On one hand, if too much bone is removed, there is the implied decrease in the strength of the femoral neck. Mardones et al.3
found that the upper boundary of safe resection was 30% of the femoral neck diameter (measured twodimensionally). On the other hand, if too little bone is removed, a revision may be required to fully correct the FAI and improve range of motion.4 Revision brings further risks to the patient that include nerve injuries, fluid extravasation, avascular necrosis and inconvenience to the patient.5 Due to patients having high functional requirements, in that those most commonly affected by cam-type pathology are highly active in
1
Imperial College School of Medicine, Imperial College London, London, UK 2 MSK Lab, Department of Orthopaedics, Imperial College London, Charing Cross Hospital, London, UK Corresponding author: Chandni Nigam, Imperial College School of Medicine, Imperial College London, London SW7 2AZ, UK. Email: [email protected]
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
1236
Proc IMechE Part H: J Engineering in Medicine 228(12)
sports, it is important that neither femoral neck strength nor range of motion is compromised. The force required to break the femoral neck depends on various factors including bone mineral density and geometry of the femoral neck. In cases of osteoporosis, occurring most commonly in the elderly, the force required to fracture the femur is lower as bone mineral density decreases.6 In addition, age-related microfractures can propagate and accelerate deterioration in bone strength where their formation exceeds repair processes7 increasing skeletal fragility.8 The effect of the geometry of the femoral neck, however, has not been consistently reported in the literature. Some studies have shown that neck width and fracture risk are positively correlated – generally acknowledged to be owing to a compensatory mechanism in the bone that increases in size but compromises on mineral density.9–11 While other studies have shown that thinner femoral necks are more prone to fracture.3,12 The instance of fracture directly attributed to osteochondroplasty is essentially nil; therefore, we do not pose that cam resection will create an immediate and significant fracture risk. There is, however, a logical conclusion that neck over-resection in a middle aged patient may manifest itself later in life and increase the patient’s risk of fracture, as bone mineral density decreases and the occurrence of microfractures increases.7 In this study, we evaluated femoral neck strength following cam surgery.
Figure 1. Images of the unresected dry-bone models bones cam A (left) and cam B (right).
Methods Computed tomography (CT) scans of two subjects with cam-type deformity were used to make proximal drybone femur models, cam A and cam B (Medical Models Ltd, Bristol, UK). Both cam A and B represented severe cam deformities present on the antero-superior aspect of the femur; however, both femurs differed in their overall morphology of the head and neck (Figure 1). Mechanical load testing was carried out on five drybone replicas each of cam A and B. The external shape of the femoral shaft was used as a template to produce custom clamping apparatus for each cam-type model using a three-dimensional (3D) printer (Objet, Rehovot, Israel). The clamp angled the femurs to 12° flexion from the vertical to recreate the natural angle of highest loading on the femoral head at walking speeds13 and to restrict femoral shaft displacement over neck displacement. An artificial acetabular surface, based on the radius of the femoral head, was also custom made for the cam-type femur models using the 3D printing. A cam model in the custom clamp was fixed into an axial loading apparatus (MultiTest 10-i; Mecmesin, Slinfold, England) (Figure 2). The MultiTest 10-i is a twin-column tensile and compression test system with a loading capacity up to 10 kN. It has a load resolution of 1:6500 and a load cell accuracy of 60.1% of full scale from 2 to 2500 N and a speed range of 1–1000 mm/min.
Figure 2. Setup of bone model in Mecmesin loading machine.
A load control method was used whereby a cyclic axial force between 10 and 700 N was applied to the femoral head at a rate of 5 mm/min for five cycles, while the displacement was recorded to a resolution of 60.01 mm. This specific loading pattern was based on preliminary results that showed 700 N to be the maximum force that the specimens could withstand without undergoing plastic deformation or failure. In addition, the relationship between fracture strength of the drybone models and 700 N was comparable to the relationship between the fracture strength of cadaveric bone and the natural loads experienced during gait.
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
Nigam et al.
1237
Figure 3. Diagrammatic representation of the alpha angle (a) (left) and four equally spaced points along the femoral head radius (R) where head–neck ratio (HNR) was evaluated (R/4, R/2. 3R/4, R) (right); 0 denotes centre point of the femoral head.
Subsequently, each bone model underwent three separate levels of resection: Surgery I, II and III equating to conservative, adequate and radical resections, respectively. First, a surgery judged to be standard but conservative for osteochondroplasty was performed by an orthopaedic surgeon (Surgery I). Open template guides were used for each subsequent resection (Surgery II and III), moving the margins of resection 5 mm inferior and 5 mm posterior each time. The open templates allowed the surgeon a good visual on the volume to be resected for each surgery. Following each resection, the models were remounted in the Mecmesin apparatus and the displacement was recorded under cyclical loading, as per the method outlined above. The displacement was compared between surgeries for each dry-bone model. Following each resection, the specimens were also scanned using NextEngine Desktop 3D Scanner (NextEngine, Santa Monica, CA, USA) and virtual models their 3D surface were created. These were input into custom-written software14 to calculate alpha angles, 360° around the head–neck junction after each surgical resection.15 The alpha angle is the angle between the longitudinal axis of the neck and the imaginary straight line joining the femoral head centre to the point on the head–neck junction where the surface contour exceeds the femoral head radius (Figure 3). Its value indicates the extent of deformity; less than 55° is considered normal range for the alpha angle.16 The software also facilitated the measurement of the 3D head– neck ratio (HNR) at four points along the femoral head radius (R) moving distally from the femoral head centre (Figure 3), the values of which were summated.17 The HNR is the ratio of the cross-sectional area of the femoral head to that of the femoral neck.
Statistical tests Values for displacement of the femur models acquired under the loading cycles did not display normal
Table 1. Mean 6 standard deviation of alpha angles (°) for cam A and cam B at unresected and after each surgery measured in the area of impingement (superior–anterior aspect) of the femoral head.
Unresected Surgery I Surgery II Surgery III
Cam A (°)
Cam B (°)
88 55 6 3 46 6 4 40 6 4
90 54 6 4 45 6 2 41 6 4
distribution. Therefore, a two-tailed Wilcoxon signedrank test for paired samples was used to determine the significance of changes in axial deformation, comparing unresected models to each subsequent surgery.
Results Alpha angle measurements extracted from the customwritten software showed a maximum alpha angle of 88° and 90° for the unresected cam A and B models, respectively (Table 1). Mean alpha angles decreased with each subsequent surgery in both cam A and B, as would be expected when the impingement is progressively corrected by the surgery (Table 1). After conservative surgery (Surgery I), the mean alpha angle is 55° 6 3° for cam A replicas and 54° 6 4° for cam B. Placing unresected cam femurs under cyclical loading produced closed hysteresis loops with variation in the value of peak axial displacement. Cycle 1 was excluded from all analyses as there was an initial linear portion in this cycle separate from the hysteresis loop under axial loading representing slippage between the bone model and loading components. Cycle 5 was also excluded as it was incomplete. Unresected cam A models had a mean displacement value across cycles 2–4 of 1.5 60.2 mm, while the mean
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
1238
Proc IMechE Part H: J Engineering in Medicine 228(12)
Figure 4. Bar graph representing mean displacement (cycles 2– 4) in unresected models and with three levels of resection (Surgery I, II and III) for (a) cam A and (b) cam B specimens. *p \ 0.01 statistical significance of displacement value compared with unresected model.
for unresected cam B models was greater at 2.7 6 0.2 mm. Bar graphs representing comparison of mean displacement between the unresected models and the three surgeries are shown in Figure 4(a) and (b). Both cam A and B models exhibited a decrease in mean displacement after Surgery I compared to that of unresected models (9.5% and 4.8% decrease in displacement for cam A and B, respectively). In cam A models, Surgery II caused a non-significant change in axial displacement compared to unresected controls. Surgery III represented the greatest displacement observed in the cam A models, with the mean displacement being 49% greater than that of unresected models, whereas in cam B, there was a 3% reduction in mean displacement compared to the intact cam B controls. Finally, mean displacement for each surgery was plotted against percentage change in the sum of the HNRs to determine the relationship between the two variables shown in Figure 5. Figure 5 shows a general decrease in the mean displacement for cam A after Surgery I, where up to 10% of the femoral neck has been removed (p \ 0.01, based on the change in the sum of HNR). Mean displacement rises to above that experienced by unresected cam A models after Surgery III (p \ 0.01), representing the greatest level of resection in this study and corresponds to more than 20% reduction in neck volume. However, there appears to be no correlation between mean
Figure 5. Mean displacement against percentage change in head–neck ratio (HNR) values compared to unresected model for (a) cam A and (b) cam B. HNR was calculated at four points along the femoral head radius and summated for each model. Missing data points are the result of inadequate laser scanning of some models.
displacement and change in the HNR for cam B (Figure 5(b)). Displacement remains similar after each surgery despite statistical significance of the mean displacement values compared to unresected (Figure 4(b)). The difference between the minimum (Surgery I) and maximum (Surgery III) mean displacement for cam B is 0.06 mm compared to 0.72 mm for cam A.
Discussion In this study, two sets of femur models with cam lesions were subjected to cyclical loading up to 700 N after three levels of resection (Surgery I, II and III) representing conservative, adequate and radical resection, to evaluate the change in femoral neck strength. Plotting axial displacement against load produced hysteresis loops with non-overlapping loading and unloading curves representing dissipated energy and viscoelastic behaviour of the material, also exhibited by natural bone.18 Bone deformity seems to be strongly related to external bony morphology. For cam A, bone resection affected bone strength with statistically significant differences in deformation when comparing the
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
Nigam et al.
1239
unresected controls with Surgery I and III. Surprisingly, Surgery I (5%–10% reduction in the neck cross-sectional area) was associated with a lower axial displacement than that for unresected models (p \ 0.01) (Figure 4). The reduction in the alpha angle to a normal range ( \ 55°) represents adequacy of conservative resection. This potentially leads to lower microcrack development which, once it has occurred, can reduce the load intensity and number of cycles required for crack propagation and bone failure,19 and in this way, microdamage reduces fracture-resistant properties of bone.20 However, any loss in strength in this instance of decreased deformation (Surgery I) may not be apparent with the downward axial force that was applied. Further studies with multidirectional deformation measurement capabilities could analyse such hypotheses in more detail. In this study, radical resection in cam A models, which removed up to 40% of the femoral neck volume, may have been a direct cause of the increased deformation observed, compared to intact cam A models, due to the drastic change in femoral neck geometry.3,12 Whether the increased displacement observed with this surgery is clinically sufficient to increase the chances of fracture is unknown. Similar studies have provided evidence of a significant reduction in load to failure with increasing depth or area of resection.21,22 Wijdicks et al.22 used a larger sample size of 30 specimens to evaluate the biomechanical parameters of composite femurs following cam resection with additional iatrogenic notching. In the study by Wijdicks, cam resection which reduced the alpha angle from 61° to 45° did not alter the mean ultimate load, stiffness or energy to failure compared to controls. Additional resection of the femoral neck of depths equal to or greater than 4 mm and width of 5.5 mm caused a statistically significant decrease ( . 17%) in the load to failure. However, these studies were limited by their ability to quantify the resections made to the bone models in only two dimensions and did not give consideration to the severity of cam impingement, as done in this study. In contrast to cam A femurs, the results from the cam B femurs suggest that there is little change in displacement when comparing resected to unresected bone even when approximately 30% of the femoral neck had been removed (Figure 5). An appreciable change may be detectable with a greater level of resection; however, 30% greatly exceeds what would normally be removed in a real-world surgical setting. Unlike previous studies, which stated a safe 30% removal zone,3 we think the margin of safe resection is subject specific depending on the morphology of the femoral bone. The discrepancy between cam A and B results suggests that the morphology of the cam deformity present influences the load bearing ability of the bone. Our study was non-cadaveric and used dry-bone models customised to replicate actual bone structure (cortical and trabecular) and shape. Previous studies have shown that plastic models can simulate the
mechanical properties of natural bone under elastic deformation.23 Peak force observed in this study was 700 N, whereas strenuous activities, such as running, can create a maximum force of 11.4 times body weight;24 therefore, a range of stress values should be investigated. This study repeatedly loaded the same bone following each bony resection. Although the loading pattern chosen of 700 N is unlikely to have altered the mechanical properties of the models via plastic deformation, the cyclic nature of the loading alongside surgical resection may have led to increased deformation. However, the decrease in mean displacement following conservative resections (Surgery I) suggests that the neck volume is the determining factor for neck strength. There was great variability in the percentage change in the HNR with each surgery, and therefore, consistency in bone resection would have resulted in a more controlled setup. Future studies can perhaps use robotic technology to control this variable.25 Femoral osteochondroplasty could lead to reduced bone strength, which is potentially serious if compounded with age-associated osteoporosis. This study demonstrates that radical neck resections may compromise the integrity of femoral neck strength due to drastic changes in femoral neck geometry. Conservative surgery results in the most favourable results in terms of axial force bearing capacity of bone models. Further studies are warranted to investigate the clinical significance of the change in displacement in multidirectional settings with an aim of changing the probability of fracture. Declaration of conflicting interests The authors declare that they have no conflict of interest. All research was performed in accordance with the Declaration of Helsinki and all local, regional and national law. Funding This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) (grant no. WSSU/PS1203). References 1. Siebenrock K, Ferner F, Noble P, et al. The cam-type deformity of the proximal femur arises in childhood in response to vigorous sporting activity. Clin Orthop Relat Res 2011; 469(11): 3229–3240. 2. Chakraverty J and Snelling N.Anterior hip pain – have you considered femoroacetabular impingement? Int J Osteopath Med 2012; 15(1): 22–27. 3. Mardones RM, Gonzalez C, Chen Q, et al. Surgical treatment of femoroacetabular impingement: evaluation of the effect of the size of the resection. J Bone Joint Surg Am 2005; 87(2): 273–279. 4. Ilizaliturri V Jr.Complications of arthroscopic femoroacetabular impingement treatment: a review. Clin Orthop 2009; 467(3): 760–768.
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
1240
Proc IMechE Part H: J Engineering in Medicine 228(12)
5. Byrd JT.Complications associated with hip arthroscopy. In: JT Byrd (ed.) Operative hip arthroscopy. New York: Springer, 2005, pp.229–235. 6. Muldoon MP, Padgett DE, Sweet DE, et al. Femoral neck stress fractures and metabolic bone disease. J Orthop Trauma 2001; 15(3): 181–185. 7. Fazzalari NL.Trabecular microfracture. Calcif Tissue Int 1993; 53(1): S143–S147. 8. Burr DB, Forwood MR, Fyhrie DP, et al. Bone microdamage and skeletal fragility in osteoporotic and stress fractures. J Bone Miner Res 1997; 12(1): 6–15. 9. Mikhail M, Vaswani A and Aloia J.Racial differences in femoral dimensions and their relation to hip fracture. Osteoporosis Int 1996; 6(1): 22–24. 10. Gnudi S, Ripamonti C, Gualtieri G, et al. Geometry of proximal femur in the prediction of hip fracture in osteoporotic women. Br J Radiol 1999; 72(860): 729–733. 11. Dincxel VE, S xengelen M, Sepici V, et al. The association of proximal femur geometry with hip fracture risk. Clin Anat 2008; 21(6): 575–580. 12. Cheng X, Lowet G, Boonen S, et al. Assessment of the strength of proximal femur in vitro: relationship to femoral bone mineral density and femoral geometry. Bone 1997; 20(3): 213–218. 13. Bergmann G, Deuretzbacher G, Heller M, et al. Hip contact forces and gait patterns from routine activities. J Biomech 2001; 34(7): 859–871. 14. Masjedi M, Nightingale CL, Azimi DY, et al. The threedimensional relationship between acetabular rim morphology and the severity of femoral cam lesions. Bone Joint J 2013; 95-B(3): 314–319. 15. Masjedi M, Azimi DY, Nightingale CL, et al. A method of assessing the severity of cam type femoro-acetabular impingement in three dimensions. Hip Int 2012; 22(6): 677–682.
16. Neumann M, Cui Q, Siebenrock KA, et al. Impingementfree hip motion: the ‘normal’ angle alpha after osteochondroplasty. Clin Orthop 2009; 467(3): 699–703. 17. Masjedi M, Marquardt CS, Drummond IM, et al. Cam type femoro-acetabular impingement: quantifying the diagnosis using three dimensional head-neck ratios. Skeletal Radiol 2013; 42(3): 329–333. 18. Lakes RS, Katz JL and Sternstein SS. Viscoelastic properties of wet cortical bone – I. Torsional and biaxial studies. J Biomech 1979; 12(9): 657–678. 19. An YH and Draughn RA.Mechanical testing of bone and the bone-implant interface. Boca Raton, FL: CRC Press, 1999. 20. Zioupos P.Accumulation of in-vivo fatigue microdamage and its relation to biomechanical properties in ageing human cortical bone. J Microsc 2001; 201(2): 270–278. 21. Davis ET, Olsen M, Zdero R, et al. A biomechanical and finite element analysis of femoral neck notching during hip resurfacing. J Biomech Eng 2009; 131(4): 041002. 22. Wijdicks CA, Balldin BC, Jansson KS, et al. Cam lesion femoral osteoplasty: in vitro biomechanical evaluation of iatrogenic femoral cortical notching and risk of neck fracture. Arthroscopy 2013; 29(10): 1608–1614. 23. Landsman AS and Chang TJ.Can synthetic bone models approximate the mechanical properties of cadaveric first metatarsal bone?J Foot Ankle Surg 1998; 37(2): 122–127; discussion 172. 24. Edwards WB, Gillette JC, Thomas JM, et al. Internal femoral forces and moments during running: implications for stress fracture development. Clin Biomech 2008; 23(10): 1269–1278. 25. Masjedi M, Tan WL, Jaskaranjit S, et al. Use of robotic technology in cam femoroacetabular impingement corrective surgery. Int J Med Robot 2013; 9(1): 23–28.
Downloaded from pih.sagepub.com at The University of Iowa Libraries on June 24, 2015
