ORIGINAL ARTICLE
Biomechanical Testing of Unstable Slipped Capital Femoral Epiphysis Screw Fixation: Worth the Risk of a Second Screw? Matthew R. Schmitz, MD,* Christine L. Farnsworth, MS,w Joshua D. Doan, MS, MEng,z Diana A. Glaser, PhD,z Brian P. Scannell, MD,y and Eric W. Edmonds, MDw8
Background: In a prior biomechanical study, 2-screw fixation of anatomically reduced slipped capital femoral epiphysis (SCFE) demonstrated marginally greater stability than single-screw fixation. However, the authors judged the benefits of a second screw to be minimal compared with the additional complication risk. A similar evaluation of fixation stability in unstable moderately displaced SCFE is performed. Methods: SCFE model: Transverse periosteal incision and epiphyseal separation from the metaphysis by leveraging in 25-monthold porcine femurs. Four groups were evaluated: pinned (3.5 mm cortex screws; Synthes, Monument, CO) with no displacement (1 screw = group N1; 2 screws = group N2) or with moderate posterior-inferior displacement of 50% of the epiphyseal diameter (1 screw = group D1; 2 screws = group D2). Biomechanical testing: Cyclical shear forces (40 to 200 N, 1 Hz) were applied along the physeal plane. Maximum load increased by 100 N every 500 cycles until failure (epiphyseal translation greater than one third the epiphyseal diameter). Force cycles (the sum of the maximum cycle loads) and number of cycles to failure were reported. Results: A sample from each D1 and D2 had fixation problems (D1, D2: n = 4; N1, N2: n = 5). One D1 failed through the femoral neck; all others failed through the epiphysis. The data showed nonsignificant trends of greater force cycles for nondisplaced over displaced (P = 0.13) and for 2 screws over 1 (P = 0.19). Number of cycles to failure showed similar trends, with no significant differences between nondisplaced and displaced (P = 0.10) and screw number (P = 0.13). Force cycles were significantly greater in the N2 group than in the D1 group. Conclusions: A trend toward higher force cycles to failure in nondisplaced and 2-screw groups was observed. Higher force cycles correspond to greater physeal stability and thus decreased risk for subsequent displacement. Within displacement groups, From the *San Antonio Military Medical Center, Fort Sam Houston, TX; wDivision of Orthopedics; zOrthopedic Biomechanics Research Center, Rady Children’s Hospital - San Diego; 8Department of Orthopaedic Surgery, University of California San Diego, San Diego, CA; and yDepartment of Orthopaedic Surgery, Carolinas Medical Center, Charlotte, NC. Rady Children’s Specialists Foundation Orthopedic Research and Education Fund provided research support. The authors declare no conflicts of interest. Reprints: Eric W. Edmonds, MD, Division of Orthopedics, Rady Children’s Hospital - San Diego, 3030 Children’s Way, Suite #410, San Diego, CA 92123. E-mail: [email protected]. Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.
496 | www.pedorthopaedics.com
adding a second screw did not significantly increase stability. Reduction of displaced SCFE also did not significantly increase stability. Only the D1 and N2 groups were significantly different. Clinical Relevance: Nondisplaced SCFE does not require 2 screws. In situ fixation of displaced SCFE might be optimized with 2 screws. Key Words: slipped capital femoral epiphysis fixation, SCFE, fixation biomechanics (J Pediatr Orthop 2015;35:496–500)
S
lipped capital femoral epiphysis (SCFE) is a common painful hip disease in adolescence, which causes loss of mobility and decrease in activity, potentially resulting in progressive femoral head deformity and/or avascular necrosis (AVN). The condition occurs when the proximal epiphysis of the femur becomes disconnected from the metaphysis and “slips” through the open physis. Treatment goals include stabilizing the epiphysis, preventing further displacement, and avoiding the devastating complications of AVN and chondrolysis. Debate continues about the best manner to stabilize the physis. Surgical treatment by screw (or screws) fixation to hold the epiphysis onto the metaphysis in situ, without a reduction of the deformity for fear of inducing AVN, is considered the gold-standard treatment. In situ screw fixation with a single screw has become the most commonly accepted form of treatment for a stable SCFE.1 The management of choice for unstable SCFE remains controversial, although the goals remain the same. Recent authors have proposed the use of a surgical dislocation technique of the hip with reorientation of the capital femoral epiphysis.2 This requires extensive training and familiarity with the technique, and the risks for catastrophic complications including AVN, remain elevated.3 In situ fixation still remains the most common treatment for unstable SCFE; however, there continues to be debate regarding the optimal number of screws (single screw vs. 2 screws) and the type of screw used (fully threaded vs. partially threaded). A bovine model previously demonstrated greater biomechanical stability with use of 2 screws compared with 1; however, single-screw fixation was still recommended, as the added stability of a second screw was J Pediatr Orthop
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Volume 35, Number 5, July/August 2015
J Pediatr Orthop
Volume 35, Number 5, July/August 2015
minimal compared with the anticipated risk for complications.4 However, this study used an anatomically reduced SCFE model, which does not replicate the more clinically applicable in situ situation with residual displacement of the epiphysis. Research is warranted as to whether there is a biomechanical advantage in using fully threaded screws to pin a reduced unstable SCFE versus pinning in situ with residual displacement. The purpose of this study was to evaluate fixation strength of in situ pinning of unstable moderate slips with residual displacement compared with anatomically reduced slips with no displacement using a porcine model. Fixation using 1 or 2 screws was compared.
METHODS SCFE Model Twenty-five-month-old (70 to 80 kg body mass) porcine femurs were acquired (10 right-sided and 10 left-sided) wrapped in saline-soaked gauze and stored frozen ( 201C). This model was chosen, as the physes were easily visible radiographically and upon gross examination. Also, it has been used in similar SCFE biomechanical studies.5 Femurs were screened for abnormalities (congenital or traumatic) using anteroposterior (AP) and lateral (Lat) radiographs. An SCFE was created in each femur by cutting the periosteum over the proximal femoral physis and levering the epiphysis off of the metaphysis as previously described by Dragoni et al.5 A surgical marking pen was used to mark adjacent locations on the epiphyseal and the metaphyseal fragments before they were separated, so that the original orientation could be restored. The diameter of the physis was measured using precision calipers from the epiphyseal piece; an AP and a mediolateral measurement were made (mm) and the 2 averaged to provide the physeal diameter for each individual femur. Half of specimens (5 right and 5 left femurs) had the epiphysis placed back onto the metaphysis in the original position, verified by matching the markings. These formed the nondisplaced SCFE groups, simulating a perfectly reduced SCFE clinical situation. SCFEs were pinned by experienced, fellowship trained pediatric orthopaedic surgeons, with either 1 (group N1, n = 5) or 2 (group N2, n = 5) screws (3.5 mm fully threaded pelvic cortex self-tapping screws; Synthes, Monument, CO), alternating between right and left in each
Unstable SCFE Screw Biomechanics
group. The remaining half of the right and half of the left femurs had the epiphysis placed with a moderate posteriorinferior displacement of 50% of the physeal diameter, simulating a moderately slipped SCFE fixed in situ. These formed the displaced SCFE groups, which were pinned with either 1 (group D1, n = 5) or 2 (group D2, n = 5) screws, alternating between right and left in each group. For the single-screw groups (N1 and D1), pinning technique followed the standard surgical procedure with a screw path predrilled starting along the lateral cortex of the proximal femur to project in line with the femoral neck and perpendicular to the physis. The screw was placed across the physis with at least 3 threads within the epiphysis. For the 2-screw groups (N2 and D2), both screws were placed in a similar parallel manner, taking into account the decreased area within the proximal femoral metaphysis for screw placement perpendicular across the physis and entering the epiphysis. Screw lengths varied but were determined by direct measurement. Appropriate lengths of 60, 65, 70, or 75 mm were utilized. Biplanar radiographs were repeated to confirm pin placement and slip displacement (Fig. 1).
Biomechanical Testing Biomechanical testing was performed using a biaxial servohydraulic table-mounted MTS 858 Bionix machine (MTS Systems, Eden Prairie, MN). Both ends of the specimens (femoral heads and shafts) were embedded in polyester resin, and then secured into custom aluminium fixtures. Femurs were aligned so that the test machine applied shear forces along the physeal planes (Fig. 2). Each specimen was tested to determine fixation strength by cyclic AP loading through the femoral head until failure. Failure was defined as epiphyseal translation greater than one third of the physeal diameter. Cyclical forces between 40 and 200 N were applied along each femoral epiphysis at 1 Hz. A maximum compression of 200 N was based on typical body weight of the pigs and the net peak joint reaction force measured by Thorup et al6 for porcine hindlimbs (4.27 N/kg body weight). If failure did not occur after 500 cycles, the maximum compressive load was increased by 100 N in a stepwise manner until failure.7 Cycle number at failure and failure mode were recorded. Biplanar radiographs were taken after biomechanical testing to document failure location. Specimen fixation strength was expressed as force cycles,
FIGURE 1. Representative posteroanterior radiograph from each group. Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.pedorthopaedics.com |
497
J Pediatr Orthop
Schmitz et al
Volume 35, Number 5, July/August 2015
FIGURE 3. Average force cycles (N) with SD error bars.
N2). Although the data (Table 1) show trends of greater mean force cycles with nondisplaced over displaced (P = 0.13) and with 2 screws over 1 (P = 0.19), the differences were not significant with no interaction between the 2 (P = 0.58; Fig. 3). Number of cycles to failure (Table 2) showed the same trends, also with no significant differences between nondisplaced and displaced (P = 0.10) and number of screws (P = 0.13) with no interaction (P = 0.64; Fig. 4). Comparing only the N2 and D1 groups, force cycles was significantly greater in N2 than in D1 (660.2 ± 109.2 vs. 273.4 ± 121.8 kN, P = 0.0015) (Fig. 3). One femur in the D1 group failed through the femoral neck, all others failed through the epiphysis.
DISCUSSION
Upon initial radiographic evaluation, 1 sample had a femoral head with an abnormal shape; it was replaced, so that the total number of samples remained 20. The mean physeal diameter was 36.0 ± 1.8 mm. One sample from each D1 and D2 did not complete testing because of fixation problems (n = 4 for D1 and D2, n = 5 for N1 and
The purpose of this study was to evaluate fixation strength of in situ pinning of unstable moderate slips with residual displacement compared with anatomically reduced slips with no displacement using a porcine model. Two studies have previously addressed some of the concerns of fixation.4,5 Karol et al4 found increased stability with 2 screws compared with 1 screw in a model with no displacement. There is no clear evidence as to which type of screw, or if a second screw is used, represents the optimum treatment for unstable SCFE. A recent study by Dragoni et al5 compared fully threaded with partially threaded (both 16 and 32 mm) screws in a porcine model. They showed higher failure rates with the 16 mm partially threaded screws and concluded that one should use either the 32 mm partially threaded or fully threaded screws for fixation. However, they only studied anatomically reduced SCFE, not slips that were pinned with residual displacement, which is the current standard of care. One previous study compared fixation of nonreduced SCFE with those fixed in an anatomic reduction.8 The authors concluded that adding a second screw increased the torsional stability and concluded that the “increased rotational stability of
TABLE 1. Force Cycles (kN)
TABLE 2. Number of Cycles
FIGURE 2. Schematic of testing rig showing application of shear force orthogonal to the physeal plane.
the product of each cyclic maximum load and the number of cycles achieved at that load.7 Force cycles (N) and number of cycles to failure were compared between groups using 2 2-way analyses of variance. Screw number and displacement were compared using 1 1-way analysis of variance (StatView, v5.0; SAS Institute Inc., Cary, NC). Statistical significance was set at a critical a value of P < 0.05.
RESULTS
Nondisplaced Displaced
1 Screw
2 Screws
P
556 ± 376 273 ± 122
660 ± 109 526 ± 360
0.57 0.23
Data are averages ± SD.
Nondisplaced Displaced
1 Screw
2 Screws
P
1626 ± 769 1038 ± 356
1923 ± 211 1591 ± 690
0.15 0.20
Data are averages ± SD.
498 | www.pedorthopaedics.com
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Pediatr Orthop
Volume 35, Number 5, July/August 2015
FIGURE 4. Average cycles with SD error bars.
double-screw fixation under torsional loading conditions may justify its use in in situ stabilization of acute or unstable SCFE.” However, the authors found that the double-screw technique did not significantly increase stiffness under shear stress. Shear forces have previously been postulated to be the largest deforming force in SCFE,9,10 so the increase in stability related to torsion might not justify the increased risk of adding a second screw, contrary to the authors’ suggestions. Our data show that there appears to be a trend toward requiring higher force cycles to achieve displacement in both the nondisplaced SCFE and those fixed with 2 screws; however, this is not statistically significant. Higher force cycles correspond to a greater strength of stability at the physis and thus decreased risk for subsequent displacement. We found no significant differences in testing single-screw versus double-screw techniques in both nondisplaced and moderately displaced SCFE as a group. Yet, there was a significant difference between the nondisplaced double-screw fixation group and the displaced single-screw fixation group. This may possibly be attributed to the fact that in a residual displacement model fixed with a single screw, the screw enters the epiphysis eccentrically, potentially posing a biomechanical disadvantage. Although rotational stability was not tested in this model, eccentric single-screw placement in a displaced model could lead to decreased rotational control. Reduction of a displaced SCFE is a controversial topic. Surgeons are faced with a decision between leaving residual deformity in the proximal femur, which is unlikely to remodel, versus attempting to reduce the epiphysis and the associated potential complications of damaging the tenuous blood supply to the proximal femur. Our data suggest that there is no biomechanical advantage in physeal stability with reducing the epiphysis and fixing with either a single-screw or double-screw technique. This study shows that there is potentially a biomechanical advantage in stability by adding a second screw, especially in displaced SCFE pinned with residual displacement. However, it must be cautioned that by adding a second screw in a displaced slipped epiphysis allows less room for error. As the epiphysis displaces from the metaphysis, the relative area in the femoral neck to safely place a second screw across the physis decreases and the risk for joint penetration or implant malplaceCopyright
r
Unstable SCFE Screw Biomechanics
ment increases. This must be taken into account when clinically deciding whether to use a single-screw or double-screw construct. Karol et al4 previously showed a 33% increase in stiffness in a double-screw construct compared with a single-screw construct in a bovine model but concluded that the small amount of gains in stiffness did not offset the potential risks of adding a second screw. Segal et al8 showed a 25% increase in stiffness with a double-screw technique in their displaced model, but this value did not reach significance. In addition, Karol et al4 showed a 33% increase in stiffness with a double-screw construct, which they found to be significant. Our study calculated force cycles, a measure of fatigue strength, instead of stiffness to quantify construct strength. Using force cycles, our study also found that adding a second screw provided an increase in construct strength: from 273 to 526 kN in the displaced group, from 556 to 660 kN in the nondisplaced group, and from 430 to 601 kN overall. However, none of these results were statistically significant (P = 0.23, 0.56, 0.19). This seems to agree with the modest stiffness differences noted in previous studies. There are a number of limitations to this study. First and foremost, this is an in vitro study and caution should be advised when applying these data to in vivo situations. These specimens were devoid of soft tissue attachments and the periosteum was cut intentionally to create the model, both of which enhance stability. Also, this biomechanical study utilized 3.5 mm screws, smaller than those typically used for SCFE fixation in humans. However, the purpose of this study was to study any differences between single-screw and double-screw fixation, regardless of size, and given our comparative porcine model, an increase in screw diameter was not feasible to reliably place 2 screws across the physis and safely into the epiphysis. Yet, the biomechanical data can be extrapolated when comparing single-screw versus doublescrew fixation. This study was also underpowered because of the variability between samples. Screw location and the tenuous epiphyseal fixation to the testing rig were the most likely contributors to this variability. In conclusion, stability with in situ fixation of displaced SCFE was not augmented with reduction. Moreover, within displacement groups, the addition of a second screw for fixation did not significantly increase the stability in our model. However, the difference between 2 screws in the nondisplaced model and 1 screw in the moderately displaced model suggests that there may potentially be a mechanical advantage to placing 2 screws in a displaced SCFE. In the appropriate clinical setting, this may possibly outweigh the risks associated with secondscrew placement, although it was not proven to be statistically significant in our current study. REFERENCES 1. Aronsson DD, Loder RT, Breur GJ, et al. Slipped capital femoral epiphysis: current concepts. J Am Acad Orthop Surg. 2006;14:666–679. 2. Slongo T, Kakaty D, Krause F, et al. Treatment of slipped capital femoral epiphysis with a modified Dunn procedure. J Bone Joint Surg Am. 2010;92:2898–2908.
2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.pedorthopaedics.com |
499
J Pediatr Orthop
Schmitz et al
3. Sankar WN, Vanderhave KL, Matheney T, et al. The modified Dunn procedure for unstable slipped capital femoral epiphysis: a multicenter perspective. J Bone Joint Surg Am. 2013;95: 585–591. 4. Karol LA, Doane RM, Cornicelli SF, et al. Single versus double screw fixation for treatment of slipped capital femoral epiphysis: a biomechanical analysis. J Pediatr Orthop. 1992;12:741–745. 5. Dragoni M, Heiner AD, Costa S, et al. Biomechanical study of 16mm threaded, 32-mm threaded, and fully threaded SCFE screw fixation. J Pediatr Orthop. 2012;32:70–74. 6. Thorup VM, Laursen B, Jensen BR. Net joint kinetics in the limbs of pigs walking on concrete floor in dry and contaminated conditions. J Anim Sci. 2008;86:992–998.
500 | www.pedorthopaedics.com
Volume 35, Number 5, July/August 2015
7. Gardner MJ, Brophy RH, Campbell D, et al. The mechanical behavior of locking compression plates compared with dynamic compression plates in a cadaver radius model. J Orthop Trauma. 2005;19:597–603. 8. Segal LS, Jacobson JA, Saunders MM. Biomechanical analysis of in situ single versus double screw fixation in a nonreduced slipped capital femoral epiphysis model. J Pediatr Orthop. 2006;26: 479–485. 9. Early SD, Hedman TP, Reynolds RA. Biomechanical analysis of compression screw fixation versus standard in situ pinning in slipped capital femoral epiphysis. J Pediatr Orthop. 2001;21:183–188. 10. Pritchett JW, Perdue KD. Mechanical factors in slipped capital femoral epiphysis. J Pediatr Orthop. 1988;8:385–388.
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Biomechanical Testing of Unstable Slipped Capital Femoral Epiphysis Screw Fixation: Worth the Risk of a Second Screw? Matthew R. Schmitz, MD,* Christine L. Farnsworth, MS,w Joshua D. Doan, MS, MEng,z Diana A. Glaser, PhD,z Brian P. Scannell, MD,y and Eric W. Edmonds, MDw8
Background: In a prior biomechanical study, 2-screw fixation of anatomically reduced slipped capital femoral epiphysis (SCFE) demonstrated marginally greater stability than single-screw fixation. However, the authors judged the benefits of a second screw to be minimal compared with the additional complication risk. A similar evaluation of fixation stability in unstable moderately displaced SCFE is performed. Methods: SCFE model: Transverse periosteal incision and epiphyseal separation from the metaphysis by leveraging in 25-monthold porcine femurs. Four groups were evaluated: pinned (3.5 mm cortex screws; Synthes, Monument, CO) with no displacement (1 screw = group N1; 2 screws = group N2) or with moderate posterior-inferior displacement of 50% of the epiphyseal diameter (1 screw = group D1; 2 screws = group D2). Biomechanical testing: Cyclical shear forces (40 to 200 N, 1 Hz) were applied along the physeal plane. Maximum load increased by 100 N every 500 cycles until failure (epiphyseal translation greater than one third the epiphyseal diameter). Force cycles (the sum of the maximum cycle loads) and number of cycles to failure were reported. Results: A sample from each D1 and D2 had fixation problems (D1, D2: n = 4; N1, N2: n = 5). One D1 failed through the femoral neck; all others failed through the epiphysis. The data showed nonsignificant trends of greater force cycles for nondisplaced over displaced (P = 0.13) and for 2 screws over 1 (P = 0.19). Number of cycles to failure showed similar trends, with no significant differences between nondisplaced and displaced (P = 0.10) and screw number (P = 0.13). Force cycles were significantly greater in the N2 group than in the D1 group. Conclusions: A trend toward higher force cycles to failure in nondisplaced and 2-screw groups was observed. Higher force cycles correspond to greater physeal stability and thus decreased risk for subsequent displacement. Within displacement groups, From the *San Antonio Military Medical Center, Fort Sam Houston, TX; wDivision of Orthopedics; zOrthopedic Biomechanics Research Center, Rady Children’s Hospital - San Diego; 8Department of Orthopaedic Surgery, University of California San Diego, San Diego, CA; and yDepartment of Orthopaedic Surgery, Carolinas Medical Center, Charlotte, NC. Rady Children’s Specialists Foundation Orthopedic Research and Education Fund provided research support. The authors declare no conflicts of interest. Reprints: Eric W. Edmonds, MD, Division of Orthopedics, Rady Children’s Hospital - San Diego, 3030 Children’s Way, Suite #410, San Diego, CA 92123. E-mail: [email protected]. Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.
496 | www.pedorthopaedics.com
adding a second screw did not significantly increase stability. Reduction of displaced SCFE also did not significantly increase stability. Only the D1 and N2 groups were significantly different. Clinical Relevance: Nondisplaced SCFE does not require 2 screws. In situ fixation of displaced SCFE might be optimized with 2 screws. Key Words: slipped capital femoral epiphysis fixation, SCFE, fixation biomechanics (J Pediatr Orthop 2015;35:496–500)
S
lipped capital femoral epiphysis (SCFE) is a common painful hip disease in adolescence, which causes loss of mobility and decrease in activity, potentially resulting in progressive femoral head deformity and/or avascular necrosis (AVN). The condition occurs when the proximal epiphysis of the femur becomes disconnected from the metaphysis and “slips” through the open physis. Treatment goals include stabilizing the epiphysis, preventing further displacement, and avoiding the devastating complications of AVN and chondrolysis. Debate continues about the best manner to stabilize the physis. Surgical treatment by screw (or screws) fixation to hold the epiphysis onto the metaphysis in situ, without a reduction of the deformity for fear of inducing AVN, is considered the gold-standard treatment. In situ screw fixation with a single screw has become the most commonly accepted form of treatment for a stable SCFE.1 The management of choice for unstable SCFE remains controversial, although the goals remain the same. Recent authors have proposed the use of a surgical dislocation technique of the hip with reorientation of the capital femoral epiphysis.2 This requires extensive training and familiarity with the technique, and the risks for catastrophic complications including AVN, remain elevated.3 In situ fixation still remains the most common treatment for unstable SCFE; however, there continues to be debate regarding the optimal number of screws (single screw vs. 2 screws) and the type of screw used (fully threaded vs. partially threaded). A bovine model previously demonstrated greater biomechanical stability with use of 2 screws compared with 1; however, single-screw fixation was still recommended, as the added stability of a second screw was J Pediatr Orthop
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Volume 35, Number 5, July/August 2015
J Pediatr Orthop
Volume 35, Number 5, July/August 2015
minimal compared with the anticipated risk for complications.4 However, this study used an anatomically reduced SCFE model, which does not replicate the more clinically applicable in situ situation with residual displacement of the epiphysis. Research is warranted as to whether there is a biomechanical advantage in using fully threaded screws to pin a reduced unstable SCFE versus pinning in situ with residual displacement. The purpose of this study was to evaluate fixation strength of in situ pinning of unstable moderate slips with residual displacement compared with anatomically reduced slips with no displacement using a porcine model. Fixation using 1 or 2 screws was compared.
METHODS SCFE Model Twenty-five-month-old (70 to 80 kg body mass) porcine femurs were acquired (10 right-sided and 10 left-sided) wrapped in saline-soaked gauze and stored frozen ( 201C). This model was chosen, as the physes were easily visible radiographically and upon gross examination. Also, it has been used in similar SCFE biomechanical studies.5 Femurs were screened for abnormalities (congenital or traumatic) using anteroposterior (AP) and lateral (Lat) radiographs. An SCFE was created in each femur by cutting the periosteum over the proximal femoral physis and levering the epiphysis off of the metaphysis as previously described by Dragoni et al.5 A surgical marking pen was used to mark adjacent locations on the epiphyseal and the metaphyseal fragments before they were separated, so that the original orientation could be restored. The diameter of the physis was measured using precision calipers from the epiphyseal piece; an AP and a mediolateral measurement were made (mm) and the 2 averaged to provide the physeal diameter for each individual femur. Half of specimens (5 right and 5 left femurs) had the epiphysis placed back onto the metaphysis in the original position, verified by matching the markings. These formed the nondisplaced SCFE groups, simulating a perfectly reduced SCFE clinical situation. SCFEs were pinned by experienced, fellowship trained pediatric orthopaedic surgeons, with either 1 (group N1, n = 5) or 2 (group N2, n = 5) screws (3.5 mm fully threaded pelvic cortex self-tapping screws; Synthes, Monument, CO), alternating between right and left in each
Unstable SCFE Screw Biomechanics
group. The remaining half of the right and half of the left femurs had the epiphysis placed with a moderate posteriorinferior displacement of 50% of the physeal diameter, simulating a moderately slipped SCFE fixed in situ. These formed the displaced SCFE groups, which were pinned with either 1 (group D1, n = 5) or 2 (group D2, n = 5) screws, alternating between right and left in each group. For the single-screw groups (N1 and D1), pinning technique followed the standard surgical procedure with a screw path predrilled starting along the lateral cortex of the proximal femur to project in line with the femoral neck and perpendicular to the physis. The screw was placed across the physis with at least 3 threads within the epiphysis. For the 2-screw groups (N2 and D2), both screws were placed in a similar parallel manner, taking into account the decreased area within the proximal femoral metaphysis for screw placement perpendicular across the physis and entering the epiphysis. Screw lengths varied but were determined by direct measurement. Appropriate lengths of 60, 65, 70, or 75 mm were utilized. Biplanar radiographs were repeated to confirm pin placement and slip displacement (Fig. 1).
Biomechanical Testing Biomechanical testing was performed using a biaxial servohydraulic table-mounted MTS 858 Bionix machine (MTS Systems, Eden Prairie, MN). Both ends of the specimens (femoral heads and shafts) were embedded in polyester resin, and then secured into custom aluminium fixtures. Femurs were aligned so that the test machine applied shear forces along the physeal planes (Fig. 2). Each specimen was tested to determine fixation strength by cyclic AP loading through the femoral head until failure. Failure was defined as epiphyseal translation greater than one third of the physeal diameter. Cyclical forces between 40 and 200 N were applied along each femoral epiphysis at 1 Hz. A maximum compression of 200 N was based on typical body weight of the pigs and the net peak joint reaction force measured by Thorup et al6 for porcine hindlimbs (4.27 N/kg body weight). If failure did not occur after 500 cycles, the maximum compressive load was increased by 100 N in a stepwise manner until failure.7 Cycle number at failure and failure mode were recorded. Biplanar radiographs were taken after biomechanical testing to document failure location. Specimen fixation strength was expressed as force cycles,
FIGURE 1. Representative posteroanterior radiograph from each group. Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.pedorthopaedics.com |
497
J Pediatr Orthop
Schmitz et al
Volume 35, Number 5, July/August 2015
FIGURE 3. Average force cycles (N) with SD error bars.
N2). Although the data (Table 1) show trends of greater mean force cycles with nondisplaced over displaced (P = 0.13) and with 2 screws over 1 (P = 0.19), the differences were not significant with no interaction between the 2 (P = 0.58; Fig. 3). Number of cycles to failure (Table 2) showed the same trends, also with no significant differences between nondisplaced and displaced (P = 0.10) and number of screws (P = 0.13) with no interaction (P = 0.64; Fig. 4). Comparing only the N2 and D1 groups, force cycles was significantly greater in N2 than in D1 (660.2 ± 109.2 vs. 273.4 ± 121.8 kN, P = 0.0015) (Fig. 3). One femur in the D1 group failed through the femoral neck, all others failed through the epiphysis.
DISCUSSION
Upon initial radiographic evaluation, 1 sample had a femoral head with an abnormal shape; it was replaced, so that the total number of samples remained 20. The mean physeal diameter was 36.0 ± 1.8 mm. One sample from each D1 and D2 did not complete testing because of fixation problems (n = 4 for D1 and D2, n = 5 for N1 and
The purpose of this study was to evaluate fixation strength of in situ pinning of unstable moderate slips with residual displacement compared with anatomically reduced slips with no displacement using a porcine model. Two studies have previously addressed some of the concerns of fixation.4,5 Karol et al4 found increased stability with 2 screws compared with 1 screw in a model with no displacement. There is no clear evidence as to which type of screw, or if a second screw is used, represents the optimum treatment for unstable SCFE. A recent study by Dragoni et al5 compared fully threaded with partially threaded (both 16 and 32 mm) screws in a porcine model. They showed higher failure rates with the 16 mm partially threaded screws and concluded that one should use either the 32 mm partially threaded or fully threaded screws for fixation. However, they only studied anatomically reduced SCFE, not slips that were pinned with residual displacement, which is the current standard of care. One previous study compared fixation of nonreduced SCFE with those fixed in an anatomic reduction.8 The authors concluded that adding a second screw increased the torsional stability and concluded that the “increased rotational stability of
TABLE 1. Force Cycles (kN)
TABLE 2. Number of Cycles
FIGURE 2. Schematic of testing rig showing application of shear force orthogonal to the physeal plane.
the product of each cyclic maximum load and the number of cycles achieved at that load.7 Force cycles (N) and number of cycles to failure were compared between groups using 2 2-way analyses of variance. Screw number and displacement were compared using 1 1-way analysis of variance (StatView, v5.0; SAS Institute Inc., Cary, NC). Statistical significance was set at a critical a value of P < 0.05.
RESULTS
Nondisplaced Displaced
1 Screw
2 Screws
P
556 ± 376 273 ± 122
660 ± 109 526 ± 360
0.57 0.23
Data are averages ± SD.
Nondisplaced Displaced
1 Screw
2 Screws
P
1626 ± 769 1038 ± 356
1923 ± 211 1591 ± 690
0.15 0.20
Data are averages ± SD.
498 | www.pedorthopaedics.com
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
J Pediatr Orthop
Volume 35, Number 5, July/August 2015
FIGURE 4. Average cycles with SD error bars.
double-screw fixation under torsional loading conditions may justify its use in in situ stabilization of acute or unstable SCFE.” However, the authors found that the double-screw technique did not significantly increase stiffness under shear stress. Shear forces have previously been postulated to be the largest deforming force in SCFE,9,10 so the increase in stability related to torsion might not justify the increased risk of adding a second screw, contrary to the authors’ suggestions. Our data show that there appears to be a trend toward requiring higher force cycles to achieve displacement in both the nondisplaced SCFE and those fixed with 2 screws; however, this is not statistically significant. Higher force cycles correspond to a greater strength of stability at the physis and thus decreased risk for subsequent displacement. We found no significant differences in testing single-screw versus double-screw techniques in both nondisplaced and moderately displaced SCFE as a group. Yet, there was a significant difference between the nondisplaced double-screw fixation group and the displaced single-screw fixation group. This may possibly be attributed to the fact that in a residual displacement model fixed with a single screw, the screw enters the epiphysis eccentrically, potentially posing a biomechanical disadvantage. Although rotational stability was not tested in this model, eccentric single-screw placement in a displaced model could lead to decreased rotational control. Reduction of a displaced SCFE is a controversial topic. Surgeons are faced with a decision between leaving residual deformity in the proximal femur, which is unlikely to remodel, versus attempting to reduce the epiphysis and the associated potential complications of damaging the tenuous blood supply to the proximal femur. Our data suggest that there is no biomechanical advantage in physeal stability with reducing the epiphysis and fixing with either a single-screw or double-screw technique. This study shows that there is potentially a biomechanical advantage in stability by adding a second screw, especially in displaced SCFE pinned with residual displacement. However, it must be cautioned that by adding a second screw in a displaced slipped epiphysis allows less room for error. As the epiphysis displaces from the metaphysis, the relative area in the femoral neck to safely place a second screw across the physis decreases and the risk for joint penetration or implant malplaceCopyright
r
Unstable SCFE Screw Biomechanics
ment increases. This must be taken into account when clinically deciding whether to use a single-screw or double-screw construct. Karol et al4 previously showed a 33% increase in stiffness in a double-screw construct compared with a single-screw construct in a bovine model but concluded that the small amount of gains in stiffness did not offset the potential risks of adding a second screw. Segal et al8 showed a 25% increase in stiffness with a double-screw technique in their displaced model, but this value did not reach significance. In addition, Karol et al4 showed a 33% increase in stiffness with a double-screw construct, which they found to be significant. Our study calculated force cycles, a measure of fatigue strength, instead of stiffness to quantify construct strength. Using force cycles, our study also found that adding a second screw provided an increase in construct strength: from 273 to 526 kN in the displaced group, from 556 to 660 kN in the nondisplaced group, and from 430 to 601 kN overall. However, none of these results were statistically significant (P = 0.23, 0.56, 0.19). This seems to agree with the modest stiffness differences noted in previous studies. There are a number of limitations to this study. First and foremost, this is an in vitro study and caution should be advised when applying these data to in vivo situations. These specimens were devoid of soft tissue attachments and the periosteum was cut intentionally to create the model, both of which enhance stability. Also, this biomechanical study utilized 3.5 mm screws, smaller than those typically used for SCFE fixation in humans. However, the purpose of this study was to study any differences between single-screw and double-screw fixation, regardless of size, and given our comparative porcine model, an increase in screw diameter was not feasible to reliably place 2 screws across the physis and safely into the epiphysis. Yet, the biomechanical data can be extrapolated when comparing single-screw versus doublescrew fixation. This study was also underpowered because of the variability between samples. Screw location and the tenuous epiphyseal fixation to the testing rig were the most likely contributors to this variability. In conclusion, stability with in situ fixation of displaced SCFE was not augmented with reduction. Moreover, within displacement groups, the addition of a second screw for fixation did not significantly increase the stability in our model. However, the difference between 2 screws in the nondisplaced model and 1 screw in the moderately displaced model suggests that there may potentially be a mechanical advantage to placing 2 screws in a displaced SCFE. In the appropriate clinical setting, this may possibly outweigh the risks associated with secondscrew placement, although it was not proven to be statistically significant in our current study. REFERENCES 1. Aronsson DD, Loder RT, Breur GJ, et al. Slipped capital femoral epiphysis: current concepts. J Am Acad Orthop Surg. 2006;14:666–679. 2. Slongo T, Kakaty D, Krause F, et al. Treatment of slipped capital femoral epiphysis with a modified Dunn procedure. J Bone Joint Surg Am. 2010;92:2898–2908.
2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
www.pedorthopaedics.com |
499
J Pediatr Orthop
Schmitz et al
3. Sankar WN, Vanderhave KL, Matheney T, et al. The modified Dunn procedure for unstable slipped capital femoral epiphysis: a multicenter perspective. J Bone Joint Surg Am. 2013;95: 585–591. 4. Karol LA, Doane RM, Cornicelli SF, et al. Single versus double screw fixation for treatment of slipped capital femoral epiphysis: a biomechanical analysis. J Pediatr Orthop. 1992;12:741–745. 5. Dragoni M, Heiner AD, Costa S, et al. Biomechanical study of 16mm threaded, 32-mm threaded, and fully threaded SCFE screw fixation. J Pediatr Orthop. 2012;32:70–74. 6. Thorup VM, Laursen B, Jensen BR. Net joint kinetics in the limbs of pigs walking on concrete floor in dry and contaminated conditions. J Anim Sci. 2008;86:992–998.
500 | www.pedorthopaedics.com
Volume 35, Number 5, July/August 2015
7. Gardner MJ, Brophy RH, Campbell D, et al. The mechanical behavior of locking compression plates compared with dynamic compression plates in a cadaver radius model. J Orthop Trauma. 2005;19:597–603. 8. Segal LS, Jacobson JA, Saunders MM. Biomechanical analysis of in situ single versus double screw fixation in a nonreduced slipped capital femoral epiphysis model. J Pediatr Orthop. 2006;26: 479–485. 9. Early SD, Hedman TP, Reynolds RA. Biomechanical analysis of compression screw fixation versus standard in situ pinning in slipped capital femoral epiphysis. J Pediatr Orthop. 2001;21:183–188. 10. Pritchett JW, Perdue KD. Mechanical factors in slipped capital femoral epiphysis. J Pediatr Orthop. 1988;8:385–388.
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
