Pin-site fracture is a catastrophic postoperative complication of navigated and robotic total knee arthroplasty (TKA). In these techniques, fixation pins are installed and removed again intraoperatively, leaving an unfilled pin track in the femoral shaft1,2. The incidence of pin-site fractures varies greatly in the literature3,4. Given the large volume and continued rapid growth of navigated and robotic TKA5,6, the assessment of the risk of pin-site fractures and their prevention are of great clinical importance.
It has been suggested that pin-site fractures are associated with obesity, osteoporosis, and poor drilling locations7-9. Our surgical team also encountered a case of femoral pin-site fracture in a femur that had been eccentrically drilled (i.e., resulting in a pin track that did not pass through the center of the bone). Operative factors that affect the risk of pin-site fracture may include the location of the pin track (height and eccentricity), diameter of the pin track, number of cortical layers involved, and iatrogenic bone destruction. Current studies have focused on the treatment of pin-site fracture10. However, there is a lack of relevant biomechanical evidence regarding their causation. For example, to our knowledge the severity of eccentricity as a risk factor for pin-site fractures remains undiscussed.
Rabbit femora have been used in mechanical simulations of human femora for decades11,12 because of the similarity in their shape (including anterior and lateral bowing of the femora) and biomechanical properties. Finite-element analysis (FEA) is also widely used as an effective method for studying femoral biomechanics13. In this study, biomechanical tests combined with FEA were used to investigate the biomechanical effect of pin track location on pin-site fracture occurrence.
Materials and Methods
Specimens and Preparation
Fifty adult New Zealand rabbits with a weight range of 2.5 to 3.5 kg were purchased from the Animal Experimental Center of Soochow University and raised in a clean animal house. After euthanasia, both femora were harvested in their entirety without soft tissue and wrapped with gauze soaked in normal saline solution. The femora were stored at −20 °C for the subsequent experiments.
Drilling and Embedding
Seventy-five of the femoral specimens were used for destructive biomechanical tests following drilling: 25 right femora were used for the torsional test, 25 left femora were used for the 3-point bending test, and 15 right and 10 left femora were used for the compression test. The remaining 25 femora underwent micro-computed tomography (CT) scanning. The specimens for each test were randomly assigned to 5 groups: the intact control group (no drilling), standard drilling group, slightly eccentric drilling group, severely eccentric drilling group, and high drilling group (n = 5 each).
The diameter and direction of the drill hole were set in reference to the OrthoPilot (Aesculap) navigated TKA system (Figs. 1-A and 1-B). Details regarding the size correspondence between rabbit and human femora are provided in Figure Sup1 and Table Sup1 in the Appendix. Drill holes were created using a 1.2-mm-diameter drill operated at a constant speed. Drilling was performed perpendicular to the bone, from anteromedial to posterolateral” (at 45° to the anteroposterior axis of the femur); the drilling height and eccentricity of each group are shown in Figure 1-C. Photographs and radiographs of each group were recorded (Figs. 1-D and 1-E).
After drilling, femora for torsional and compression tests were embedded in the same manner to allow for the subsequent biomechanical tests. Both ends were aligned along the same central axis of the femur (see Figure Sup2 in the Appendix).
Radiographic and Micro-CT Analysis
As radiography and CT provide important diagnostic evidence for fractures, we also performed imaging examinations. Anteroposterior radiography (Siemens) of the femoral specimens was performed to document the gross appearance of the femora before and after the biomechanical testing of all groups. In addition, the femora were scanned using high-resolution micro-CT (Bruker SkyScan 1176; 18 μm per layer, 0.7° rotation step). Analyzed parameters including the cortical defect area and distance between the centers of the cortical drill holes divided by the femoral diameter (center distance/transverse diameter). In addition, 3D models were created for better views of the drilling channels.
Femora were thawed at room temperature and stored in normal saline solution until tested. Biomechanical parameters of the femora were measured with a universal material testing system (Instron ElectroPuls E10000). The failure load, failure displacement, and stiffness were recorded and analyzed using Bluehill 2.0 and WaveMatrix software (Instron). The system was programmed to stop immediately after the occurrence of fracture.
For the torsional tests, the cylinders in which the ends of the femur had been potted were clamped and twisted until femoral fracture occurred. Data including maximum torque, peak twist (twist at which fracture occurred), and torsional stiffness were recorded and analyzed. For the 3-point bending tests, the femoral samples were supported at both ends, with a span of 60 mm and the anterior side of the bone on the bottom. Displacement-controlled loading was applied to the posterior end of the femur until fracture occurred. The measured biomechanical properties included failure load and failure displacement. For the axial compression tests, femoral specimens were axially compressed at a fixed speed. The failure load, failure displacement, and compressive stiffness were recorded. The loading rates in the tests were held constant at values determined on the basis of related research in the literature14-17: 5°/min in the torsional tests and 3 mm/min in the 3-point bending and axial compression tests. The femora were kept moist with normal saline solution throughout the tests. The failure load and failure displacement were derived from the load-displacement curves, which typically exhibited a pronounced peak near the end of the curve.
CT images of a femur from a normal adult (a 30-year-old man 1.73 m tall and weighing 65 kg) were imported into Mimics 21.0 (Materialise) for 3D modeling of the femur. The femoral model consisted of cancellous and cortical bone. The geometric model of the femur and pin track was created by Geomagic 12.0 (Raindrop) and Creo Parametric 5.0 (PTC) engineering software. A 4.5-mm Kirschner wire model was created and superimposed on the femoral model; subtraction of the overlap yielded a 4.5-mm-diameter pin track in the femur. The drilling location varied among the groups: in the intact control group and standard, slightly eccentric, and severely eccentric drilling groups, the drilling was performed 100 mm from the distal end the of the femur. In the high drilling group, the drilling was at a height of 133 mm. The direction and eccentricity for each group were consistent with those used in the biomechanical tests (see Fig. Sup3 in the Appendix).
The abovementioned femoral models with pin tracks were then imported into Hypermesh (Altair) for mesh creation and definition of material properties, as in previous studies18. The elastic modulus was set to 16,800 MPa for cortical bone and 840 MPa for cancellous bone, and the Poisson ratio was set to 0.3 for both cortical and cancellous bone. This finite-element model of the femur has been validated in previous studies19. The total number of elements and total nodal points in the model are shown in Table I. ANSYS 19.1 (Ansys) was used to set loads and constraints and to obtain stress and deformation results.
TABLE I -
Number of Elements and Nodal Points in FEA Modeling
||Total Nodal Points
Boundary and Loading Conditions
All nodes of the distal femoral condyle were fully constrained13. For torsional load, 12 Nm of torque was applied to the femoral head. For 3-point bending, the displacement of both the proximal and distal nodes of the femur was constrained, and 500 N of force was applied perpendicular to the middle of the femur. For compression, 1,500 N of axial compression was applied to the femoral head to simulate standing on 1 leg. We analyzed the von Mises stress distribution, the maximum von Mises stress, and deformation in each simulation. The finite-element modeling and the procedure have been validated in our published study20.
Definition of the Safe Range
FEA was used to identify the safe range of drilling eccentricity. Nineteen equally spaced points were used to divide the transverse diameter of the femur into 20 equal 1.68-mm portions at the standard drilling height. A finite-element model corresponding to a drilling axis passing through each of these points was created (resulting in a total of 19 models). The models were grouped according to eccentricity (drilling through the center of the bone, L1 through L9 for increasing anterolateral eccentricity, and M1 through M9 for increasing posteromedial eccentricity). We analyzed the von Mises stress distribution, the maximum von Mises stress, and the deformation in torsion, compression, and bending simulations. For each simulation, the cumulative sum (CUSUM) method, a reliable way to define the turning point in a series21, was applied to find the inflection point in each of the 2 eccentricity groups.
Each mechanical test was performed on at least 5 specimens, and all were successful and were included in the analysis. The results were expressed as the mean and standard deviation (SD). GraphPad Prism 7.0 was used for statistical analysis. One-way analysis of variance (ANOVA) was used to determine significance; differences were considered significant if p < 0.05.
Source of Funding
This work was supported by grants from the National Natural Science Foundation of China (82072425, 82072498, 81873991, 81873990), Young Medical Talents of Jiangsu Province (QNRC2016751), Natural Science Foundation of Jiangsu Province (BK20180001, BE2021650), Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD), and Special Project of Diagnosis and Treatment Technology for Key Clinical Diseases in Suzhou (LCZX202003).
Micro-CT images of transverse and longitudinal sections of the pin track showed that the drilling results in each group met the plan (Fig. 2-A). The drilling edges were straight and neat, which could be observed in 3D models viewed from the perspective of the drilling direction (Fig. 2-B). It is worth noting that in the severely eccentric group, the drilling path notched the inner cortex of the femur. No fractures or additional bone destruction occurred while drilling. The mean area of the cortical defect was greater for eccentric drilling and high drilling than for the standard position, especially in the severely eccentric group (Fig. 2-C). The results of the center distance/transverse diameter showed that cortical drill holes were significantly closer in the severely eccentric group (Fig. 2-D).
Biomechanical tests confirmed that severely eccentric drilling significantly weakened the biomechanical strength of the femora. The torsional test (Fig. 3-A) showed that the maximum torque and peak twist in the severely eccentric drilling group were significantly smaller than those in the control group (Figs. 3-B and 3-C), indicating that the ability of a femur to resist torsion is impaired by severely eccentric drilling. Slightly eccentric or high drilling did not lead to this phenomenon. There was no significant difference in torsional stiffness among all groups (Fig. 3-D). Furthermore, spiral fractures were the predominant fracture type in each group in the torsional test (Figs. 3-E and 3-F). Both drill holes were included in the fracture lines in all 5 specimens in the severely eccentric group, revealing the severe stress concentration at the drill holes (Table II). In contrast, no drill holes were involved in the fractures in any of the other groups.
TABLE II -
Summary of Fracture Types and Fracture Lines in Biomechanical Experiments
||3-Point Bending Test
||Axial Compression Test
||Fractured Drill Holes*
||Fractured Drill Holes*
||Fractured Drill Holes*
||1 incomplete, 4 transverse
||2 incomplete, 2 oblique, 1 transverse
||4 spiral, 1 incomplete
||3 incomplete, 2 transverse
||3 incomplete, 1 oblique, 1 comminuted
||4 spiral, 1 incomplete
||3 incomplete, 2 transverse
||1 incomplete, 2 splitting, 1 transverse, 1 comminuted
||4 oblique, 1 comminuted
||3 incomplete, 1 transverse, 1 oblique
||1 incomplete, 4 oblique
||2 incomplete, 2 comminuted, 1 splitting
*Number of drill holes involved in fractures.
The failure load of the femur in the 3-point bending test (Fig. 4-A) was significantly lower in the severely eccentric drilling group than in the control group. There were no significant differences in failure displacement and stiffness among the groups (Figs. 4-B, 4-C, and 4-D). Transverse and incomplete fractures made up all of the fractures caused by 3-point bending in the intact control group, the standard drilling group, and the slightly eccentric group, and no drill holes were included in the fracture lines in those groups (Figs. 4-E and 4-F, Table II).
The compression test (Fig. 5-A) showed no significant differences in maximum compression load, maximum deformation, or stiffness among the drilling groups (Figs. 5-B, 5-C, and 5-D). Compression resulted in comminuted fractures in the standard, slightly eccentric, and high drilling groups. The number of drill holes included in the fractures varied from 2 to 4 per group (Figs. 5-E and 5-F, Table II). Interestingly, the only drill breakage occurred in the severely eccentric group.
The FEA results provided comparisons of the stress distribution and deformation of the human femur under simulated stress. The von Mises stress distribution is shown in the maps in Figures 6-A, 6-B, and 6-C; the maximum values are shown in Tables III and IV. In the torsional simulation, the local stress near the drill hole was highest in the severely eccentric group: 31.02 MPa, which was 88.1% higher than that in the standard group. Differences in deformation across the groups were very small, suggesting that torsional deformation did not depend on drilling eccentricity. In the compression simulation, the maximum von Mises stress was highest in the severely eccentric group, as was the deformation. In the 3-point bending simulation, the maximum von Mises stress and deformation results showed little difference with the amount of eccentricity.
TABLE III -
Maximum von Mises Stress in Finite-Element Analysis
||Maximum von Mises Stress (MPa)
TABLE IV -
Deformation in Finite-Element Analysis
The overall safe range of eccentricity was found to lie between 50% of the radius in the anterolateral direction and 70% in the posteromedial direction. The stress distribution in additional torsion and compression analyses is shown in Figure Sup4 in the Appendix. The inflection points of von Mises stress were found at the same eccentricities in all of these torsion and compression analyses as well: 50% anterolateral and 70% posteromedial. In the 3-point bending analysis, there was no obvious inflection point (Figs. 7-A through 7-F). Taking these 2 inflection points as boundaries, the range of possible eccentricities can be divided into a safe zone and a risk zone (Fig. 7-G). Within the safe zone, we further defined a subzone with maximum torsional von Mises stress of <20 MPa and maximum compressive von Mises stress of <60 MPa (ranging from 10% anterolateral eccentricity to 20% posteromedial eccentricity) as strongly recommended for drilling, and the remainder of the safe zone as moderately recommended (Fig. 7-H).
Navigational and robotic tools have been increasingly utilized in TKA because of their advantages in accuracy of alignment restoration and soft-tissue balancing22-25. However, surgeons should also be aware of the potential complications related to navigation26. During drilling in navigated or robotic TKA, local periosteal tissue remains unstripped and direct vision and intraoperative fluoroscopy are not used, increasing the risk of eccentric drilling. Also, drill holes remain unfilled after the procedure, increasing the risk of a pin-site fracture after a transient low-energy event.
Pin-site fracture has been described as an uncommon complication of navigated TKA28. Wysocki et al. reported a pin-site fracture incidence of approximately 1%, close to the incidence of deep infection29. Surgical factors predisposing to pin-site fractures may include a poor prosthesis position, soft-tissue trauma, an oversized component, transcortical pin fixation, and repeated drilling8,30.
Our study systematically explored the role of pin track location in the pin-site fracture risk from a biomechanical point of view. We found that the eccentricity is an important pin-site parameter. Severely eccentric drilling should be considered an iatrogenic “unforced error.” The effect of eccentric drilling on fracture risk varied with the type of biomechanical mechanism. Drilling that was slightly eccentric (by 1/4 of the diameter) or high acted comparably to the control drilling parameters in terms of biomechanical safety. However, severe eccentricity significantly affected the biomechanical strength of the femur, including in torsion, compression, and 3-point bending, with torsional resistance being weakened the most obviously. Furthermore, drill holes in the severely eccentric drilling group were more often included in the fracture lines in the torsional group, confirming the predicted stress concentration at severely eccentric drill holes. In addition to elevating the pin-site fracture risk, eccentric drilling can also elevate the risks of other complications. A study based on TKA using the Mako robotic arm (Stryker) suggested that transcortical drilling may lead to thermal osteonecrosis and even pin-site infection26. Interestingly, the only occurrence of drill breakage in the present study was in the severely eccentric group, suggesting that the drill experienced greater shear force when notching the cortex.
The assessment of the safe range indicated that medial eccentricity was better tolerated than lateral eccentricity, and severe eccentricity should be avoided in order to reduce the risk of pin-site fracture. The importance of drilling within the safe zone must be emphasized. It should also be noted that our findings are not limited to femoral pin-site fractures following TKA; they should be valuable in other procedures that involve femoral drilling, such as traction pin insertion and internal fixation. Spiral fractures were the most common type in the biomechanical and clinical results. Therefore, patients with drilling eccentricity in the risk zone should be warned of the risk that femoral torsion can result in a fracture, and should be avoided while the bone is healing.
Additional factors may be related to the risk of pin-site fractures. When designing the workflow, the diameter and number of drill holes should be minimized, as should the volume of anchors. Patient factors predisposing to femoral fracture may include obesity31, primary osteoporosis32, diabetes33, and malignancies34. Obesity results in greater weight-bearing, and osteoporosis implies reduced biomechanical strength of the femur. We conjecture that the safe zone for drilling in obese or osteoporotic patients would therefore be narrower. Perioperative treatments for such underlying diseases, as well as increased vigilance, are advised in these patients. Extra attention should also be given to patients with a high body mass index because of the reduced drilling accuracy due to thickened soft tissue in addition to the greater load on the femur.
We propose the following clinical recommendations based on our analysis of pin-site fracture risk. First, all arthroplasty systems in which pinning is performed should provide an optimal drilling location or at least a safe zone. Second, eccentricity correction devices need to be developed to improve drilling accuracy. Orthopaedic surgeons should appreciate the learning curve when utilizing new navigational or robotic systems. Evaluation of the pin-site fracture risk should be included throughout management of TKA. Surgeons should be alert to postoperative thigh pain in high-risk patients (obese, osteoporotic, and/or those with eccentric drilling) and limit their weight-bearing during recovery. Most pin-site fractures occurred within 3 months after surgery35. Perioperative drill hole management should be noted, including bone grafting, material filling, bone-targeted medication, etc. Surgical treatment of pin-site fractures depends on the fracture type and site26,35. We found that the stress release at the pin-site actually protected the periprosthetic area. Hence, most pin-site fractures can be treated as simple femoral shaft fractures that do not require revision of the prosthesis28,36,37.
There are limitations to this study that need to be acknowledged. Investigation of the loading rate variation would improve the understanding of risk factors for pin-site fractures. In addition, underlying diseases including obesity and osteoporosis should be discussed. Two-pin navigation and robotic TKA are also commonly used, which calls for comparative studies. These provide directions for future research regarding femoral pin-site fractures.
Supporting material provided by the authors is posted with the online version of this article as a data supplement at jbjs.org (https://links.lww.com/JBJS/H144).
1. Picard F, Deep K, Jenny JY. Current state of the art in total knee arthroplasty computer navigation. Knee Surg Sports Traumatol Arthrosc. 2016 Nov;24(11):3565-74.
2. van der List JP, Chawla H, Joskowicz L, Pearle AD. Current state of computer navigation and robotics in unicompartmental and total knee arthroplasty: a systematic review with meta-analysis. Knee Surg Sports Traumatol Arthrosc. 2016 Nov;24(11):3482-95.
3. Brooks DB, Burstein AH, Frankel VH. The biomechanics of torsional fractures. The stress concentration effect of a drill hole. J Bone Joint Surg Am. 1970 Apr;52(3):507-14.
4. Burstein AH, Currey J, Frankel VH, Heiple KG, Lunseth P, Vessely JC. Bone strength. The effect of screw holes. J Bone Joint Surg Am. 1972 Sep;54(6):1143-56.
5. Australian Orthopaedic Association National Joint Replacement Registry. Hip, Knee & Shoulder Arthroplasty 2018 Annual Report. Accessed 2022 June 8. https://aoanjrr.sahmri.com/annual-reports-2018
6. Jones CW, Jerabek SA. Current Role of Computer Navigation in Total Knee Arthroplasty. J Arthroplasty. 2018 Jul;33(7):1989-93.
7. Brown MJ, Matthews JR, Bayers-Thering MT, Phillips MJ, Krackow KA. Low Incidence of Postoperative Complications With Navigated Total Knee Arthroplasty. J Arthroplasty. 2017 Jul;32(7):2120-6.
8. Jung HJ, Jung YB, Song KS, Park SJ, Lee JS. Fractures associated with computer-navigated total knee arthroplasty. A report of two cases. J Bone Joint Surg Am. 2007 Oct;89(10):2280-4.
9. Skibicki HE, Ponzio DY, Brustein JA, Post ZD, Ong AC, Orozco FR. A cautionary case: osteoporotic femur fracture after robotic-assisted total knee arthroplasty. Osteoporos Int. 2021 Oct;32(10):2125-9.
10. Bonutti P, Dethmers D, Stiehl JB. Case report: femoral shaft fracture resulting from femoral tracker placement in navigated TKA. Clin Orthop Relat Res. 2008 Jun;466(6):1499-502.
11. Omeroğlu H, Ateş Y, Akkuş O, Korkusuz F, Biçimoğlu A, Akkaş N. Biomechanical analysis of the effects of single high-dose vitamin D3 on fracture healing in a healthy rabbit model. Arch Orthop Trauma Surg. 1997;116(5):271-4.
12. Ren Y, Zhao H, Yang K, Zhang Y. Biomechanical compatibility of high strength nickel free stainless steel bone plate under lightweight design. Mater Sci Eng C Mater Biol Appl. 2019 Aug;101:415-22.
13. Papini M, Zdero R, Schemitsch EH, Zalzal P. The biomechanics of human femurs in axial and torsional loading: comparison of finite element analysis, human cadaveric femurs, and synthetic femurs. J Biomech Eng. 2007 Feb;129(1):12-9.
14. Zhang J, He F, Zhang W, Zhang M, Yang H, Luo ZP. Mechanical force enhanced bony formation in defect implanted with calcium sulphate cement. Bone Res. 2015 Jan 20;3:14048.
15. Maitirouzi J, Yanna L, Abulizi A, Aihemaitiniyazi A, Kuerban S, Shaojun H. Effects of Uygur sand therapy on the mechanical properties of femurs in osteoarthritic rabbits. Biomed Mater Eng. 2017;28(6):633-42.
16. Sevil F, Kara ME. The effects of ovariectomy on bone mineral density, geometrical, and biomechanical characteristics in the rabbit femur. Vet Comp Orthop Traumatol. 2010;23(1):31-6.
17. Massie AM, Kapatkin AS, Garcia TC, Guzman DS, Chou PY, Stover SM. Effects of Hole Diameter on Torsional Mechanical Properties of the Rabbit Femur. Vet Comp Orthop Traumatol. 2019 Jan;32(1):51-8.
18. Sowmianarayanan S, Chandrasekaran A, Kumar RK. Finite element analysis of a subtrochanteric fractured femur with dynamic hip screw, dynamic condylar screw, and proximal femur nail implants—a comparative study. Proc Inst Mech Eng H. 2008 Jan;222(1):117-27.
19. Aikebaier T, Ajimu K, Xie Z, Zhang W. A finite element analysis of two internal fixation methods for unstable femoral neck fractures in young adults. Chinese J Orthop Trauma. 2020;22:793-8.
20. Ren W, Zhang W, Jiang S, Peng J, She C, Li L, Mao Y, Zhou H, Xu W. The Study of Biomechanics and Clinical Anatomy on a Novel Plate Designed for Posterolateral Tibial Plateau Fractures via
Anterolateral Approach. Front Bioeng Biotechnol. 2022 Mar 8;10:818610.
21. Kayani B, Konan S, Huq SS, Tahmassebi J, Haddad FS. Robotic-arm assisted total knee arthroplasty has a learning curve of seven cases for integration into the surgical workflow but no learning curve effect for accuracy of implant positioning. Knee Surg Sports Traumatol Arthrosc. 2019 Apr;27(4):1132-41.
22. de Steiger RN, Liu YL, Graves SE. Computer navigation for total knee arthroplasty reduces revision rate for patients less than sixty-five years of age. J Bone Joint Surg Am. 2015 Apr 15;97(8):635-42.
23. Kim YH, Park JW, Kim JS. The Clinical Outcome of Computer-Navigated Compared with Conventional Knee Arthroplasty in the Same Patients: A Prospective, Randomized, Double-Blind, Long-Term Study. J Bone Joint Surg Am. 2017 Jun 21;99(12):989-96.
24. Saragaglia D, Rubens-Duval B, Gaillot J, Lateur G, Pailhé R. Total knee arthroplasties from the origin to navigation: history, rationale, indications. Int Orthop. 2019 Mar;43(3):597-604.
25. Lee YS, Howell SM, Won YY, Lee OS, Lee SH, Vahedi H, Teo SH. Kinematic alignment is a possible alternative to mechanical alignment in total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc. 2017 Nov;25(11):3467-79.
26. Kamara E, Berliner ZP, Hepinstall MS, Cooper HJ. Pin Site Complications Associated With Computer-Assisted Navigation in Hip and Knee Arthroplasty. J Arthroplasty. 2017 Sep;32(9):2842-6.
27. Johnson BA, Fallat LM. The effect of screw holes on bone strength. J Foot Ankle Surg. 1997 Nov-Dec;36(6):446-51.
28. Beldame J, Boisrenoult P, Beaufils P. Pin track induced fractures around computer-assisted TKA. Orthop Traumatol Surg Res. 2010 May;96(3):249-55.
29. Wysocki RW, Sheinkop MB, Virkus WW, Della Valle CJ. Femoral fracture through a previous pin site after computer-assisted total knee arthroplasty. J Arthroplasty. 2008 Apr;23(3):462-5.
30. Owens RF Jr, Swank ML. Low incidence of postoperative complications due to pin placement in computer-navigated total knee arthroplasty. J Arthroplasty. 2010 Oct;25(7):1096-8.
31. Soliman SS, Jordan GB, Bilaniuk JW, Benfante A, Kong K, Rolandelli RH, Curran T, Nemeth ZH. The impact of BMI on morbidity and mortality after femoral fractures. Eur J Trauma Emerg Surg. 2021 Oct 12:1-7.
32. Lane NE. Epidemiology, etiology, and diagnosis of osteoporosis. Am J Obstet Gynecol. 2006 Feb;194(2)(Suppl):S3-11.
33. Vestergaard P, Rejnmark L, Mosekilde L. Relative fracture risk in patients with diabetes mellitus, and the impact of insulin and oral antidiabetic medication on relative fracture risk. Diabetologia. 2005 Jul;48(7):1292-9.
34. Body JJ. Increased fracture rate in women with breast cancer: a review of the hidden risk. BMC Cancer. 2011 Aug 29;11:384.
35. Jung KA, Lee SC, Ahn NK, Song MB, Nam CH, Shon OJ. Delayed femoral fracture through a tracker pin site after navigated total knee arthroplasty. J Arthroplasty. 2011 Apr;26(3):505.e9-11.
36. Blue M, Douthit C, Dennison J, Caroom C, Jenkins M. Periprosthetic Fracture through a Unicortical Tracking Pin Site after Computer Navigated Total Knee Replacement. Case Rep Orthop. 2018 Sep 16;2018:2381406.
37. Li CH, Chen TH, Su YP, Shao PC, Lee KS, Chen WM. Periprosthetic femoral supracondylar fracture after total knee arthroplasty with navigation system. J Arthroplasty. 2008 Feb;23(2):304-7.