Biomechanical Evaluation for Bone Allograft in Treating the Femoral Head Necrosis
Thorough Debridement or not?
Guangquan Zhou1,2,4; Wei He1; Zhihui Pang2; Xiumin Chen3; Yujing Xu4; Liao Shaoyi Stephen4; QinQun Chen5 1 The First Affiliated Hospital, Guangzhou University of Chinese Medicine, Guangzhou, China
2 Laboratory of National Key Discipline Orthopaedics and Traumatology of Chinese Medicine, Guangzhou University of Chinese Medicine, Guangzhou, China
3 Department of Rheumatology, Guangdong Provincial Hospital of Chinese Medicine, Guangzhou, China, and Postdoctoral Mobile Research Station, Guangzhou University of Chinese Medicine, Guangzhou, China
4 Department of Information Systems, City University of Hong Kong, Hong Kong, China
5 School of Medical Information Engineering, Guangzhou University of Chinese Medicine, Guangzhou, China
Abstract
Fibular allograft with impaction bone grafting (FAIBG) is an effective hip-preserving method for avoiding total hip replacement in the early stage of femoral head necrosis (FHN). However, the question of whether thorough debridement should be used with FAIBG is controversial. This study first proposes employing computational biomechanical technology to explore the different mechanical performance of FAIBG with or without thorough debridement, which provides a biomechanical basis for choosing the proper treatment in clinic.
1 Introduction
There has been a rapid increase in the incidence of femoral head necrosis (FHN) all over the world which is caused by the widespread use of steroids (Chan and Mok, 2012; Weinstein, 2012) and alcohol (Matuso et al., 1988; Hirota et al., 1993; Wang et al., 2003; Shigemura et al., 2012). FHN is associated with high morbidity and disability. Patients with FHN often have a high risk of collapse of the femoral head, arthritis, or dearticulation, finally resulting in hip replacement (HR). The medium- and long-term effects of hip implants are obviously unsatisfactory; thus, the young patient will require several surgical treatments (Ditri et al., 2006). Hence, various head-preserving procedures have been developed to protect the femoral head of patients and avoid HR, particularly in the early stage of FHN.
Fibular allograft with impaction bone grafting (FAIBG) is an effective head-preserving method for avoiding total hip replacement (THR) in the early stage of FHN. FAIBG provides both repaired materials and biomechanical structural support during the healing of the necrosis region (Brannon, 2007; Katz and Urbaniak, 2001; Malizos et al., 1995; Urbaniak et al., 1995). However, the question of whether thorough debridement should be used with FAIBG is controversial. “With thorough debridement” means that the necrotic bone should be cleaned up completely, while “without thorough debridement” means that the necrotic bone should undergo partial debridement. Theoretically, thorough debridement can better protect the anterolateral column and reduces the necrosis area of the stress concentration phenomenon compared to partial debridement, but it will cause a larger trauma region and higher incidence of complications and require more recovery time after surgery. In most cases, the choice is based on the experience and preference of different surgeons. At the same time, studies about comparing the risk of collapse of postoperative femoral head accompanied with thorough debridement and without thorough debridement are relatively rare.
To provide a scientific biomechanical basis for FAIBG, this study presents two subject-specific FHN cases without collapse of the femoral head to compare the mechanical performance between FAIBG with and without thorough debridement.
2 Materials and methods
2.1 JIC classification
In 2001, the Japanese Investigation Committee (JIC) (Sugano et al., 2002) revised the diagnostic criteria to clarify the definition of osteonecrosis of the femoral head (ONFH). According to the JIC classification criteria, FHN is classified into subtypes A, B, C1, and C2 based on the location of the lesion in the weight-bearing area. Type A lesions occupy the medial one-third or less of the weight-bearing portion. Type B lesions occupy the medial two-thirds or less of the weight-bearing portion. Type C1 lesions occupy more than the medial two thirds of the weight-bearing portion but do not extend laterally to the acetabular edge. Type C2 lesions occupy more than the medial two-thirds of the weight-bearing portion and extend laterally to the acetabular edge.
Recent studies showed that patients who conform to the JIC C criteria are suitable for FAIBG. However, these conclusions are mainly based on clinical observation experience and must be proved in both theory and practice. We postulated that the FAIBG procedure with different debridement regions has different biomechanical performances, which could affect the choice of treatment procedure for FHN. Hence, we reconstructed two subject-specific models (JIC C1 and C2; Figure 8.1) to provide a biomechanical basis for FAIBG to explore the performance of different debridement regions in treating FHN.
2.2 Generation of intact finite element models
A JIC C1 FHN-diagnosed patient (P1) with a weight of 70 kg and a JIC C2 FHN-diagnosed patient (P2) with a weight of 60 kg were selected for the biomechanical evaluation of the proximal femur. Computed tomography data sets (0.5 mm thickness; Toshiba Aquilion 64, Toshiba, US) of each case were used to reconstruct solid models with gray-level processing of the software MIMICS 15.1 based on the function of Thresholding, Edit Masks, and Calculate 3D. The solid models in the STereo Lithography (STL) format were entered into the Rapidform preprocessor, and surface fitting was performed. Based on the Mesh and Autosurfacing functions, we found the fit hip to generate the Non-Uniform Rational B-Splines (NURBS) models. The interface between the ilium and femoral head was used to identify thecartilage geometry. All NURBS models in the format of igs were entered into ABAQUS V6.13 (Simulia, Dassault Systemes, France) to generate nonlinear elastic finite element models. Based on the initial hip geometry, we simulated the physiological and pathological models using different materials.
Then, all models were input into ABAQUS V6.13 to generate isotropic 10-node tetrahedral elements. The mesh size was 4 mm. The initial models consisted of elements (146,879 of P1; 156,471 of P2) and nodes (213,970 of P1; 230,541 of P2). In these models, the single-legged stance was considered as a representative body position, and a ground reaction force equivalent to the body weight was performed on a rigid plate, which was tied to the distal part of the femur in Figure 8.2. Constraints were applied to the pubic symphysis and sacroiliac joint. All six degrees of freedom were constrained to zero. Seven muscles were modeled as axial connectors, and the muscle forces were set according to the literature (Sverdlova and Witzel, 2010): adductor longus = 560 N; adductor magnus = 600 N; gluteal maximus = 550 N; gluteal medius = 700 N; gluteal minimus = 300 N; piriformis = 500 N; tensor fascia latae = 300 N. The models consist of cortical, trabeculae, cartilage, and lesion bone. The material properties used in the biomechanical experiment were obtained from the literature (Brown and Hild, 1983; Brown et al., 1981; Grecu et al., 2010): Ecortical = 15,100 MPa, Etrabeculae = 445 MPa, Ecartilage = 10.5 MPa, Elesion = 124.6 MPa, νcortical = 0.3, νtrabeculae = 0.22, νcartilage = 0.45, and νlesion = 0.152.
The parametric analysis was designed to explore the effects of the debridement extent of necrotic bone for cases that require surgery. The maximum debridement radius was defined as r, and the debridement extent variants are schematically shown in Figure 8.3. We assumed that the anterolateral cortical stress corresponding to the debridement extent of necrotic lesions had an increased radius R (R = 1/4 r, 3/8 r, 1/2 r, 5/8 r, 3/4 r, 7/8 r and r), where R = 1/4r refers to the least debridement, and R = r denotes thorough debridement. To simulate the allogeneic fibular implant, the dimensions (80 mm in length and 6 mm in radius) were obtained from the manufacturer. The axial direction of fibula was defined by the entry point and lesion centroid. The entry point was located in the trochanteric lateral cortex of the femur. The distance of the cortical bone from the apex of the fibula was 5 mm. The remaining voids were occupied by impaction cancellous bone after the debridement.
3 Results
3.1 Stress transfer path
The principal stress transfer characteristics are the most important biomechanical index in the process of evaluating the performance of FHN. In all femoral heads, the principal stress transfer patterns are computed when a midstance gait occurs. The principal stress transfer efficiency reduces markedly (see Figure 8.4).
3.2 Stress of the anterolateral column
FAIBG has a considerably small risk of structure collapse compared with the untreated situation. Figures 8.5a and 8.6a show the healthy stress distribution of anterolateral cortical bone and the maximum stress values are 23.95 MPa of P1 and 25.99 MPa of P2. Figure 8.5b shows that JIC C1 stress of 30.31 MPa increases about 26.56%, which is more than the healthy condition (P1). Figure 8.6b shows that the JIC C2 stress of 34.58 MPa increases about 33.05%, which is higher than the healthy condition (P2). Figure 8.5c and 6C show that the postoperative stress is 23.52 MPa in P1 and 25.31 MPa in P2, which is approximately 22.4% less than the JIC C1 condition (P1) and 26.81% lower than the JIC C2 condition (P2) after the FAIBG procedure. The peak stresses of the two postoperative cases return to near-healthy levels. It is obviously that the stress concentration regions in JIC C1 and C2 are the areas that the red arrows point to. After the FAIBG procedure, the stress concentration regions disappeared. Figure 8.5d–i and 8.6d–i show that stress has no significant changes as the debridement radius increasing.
3.3 Peak stress of the residual necrotic bone
Figure 8.7 displays that the debridement size will affect the stress gradient in the residual necrotic bone. Seven different necrotic debridement sizes, ranging from 1/4r to r, are chosen to study the effect of the debridement radius on the residual necrotic bone. The relation between debridement size and the stress of the residual necrotic bone is shown in Figure 8.7. When the debridement radius is 1/4r, there is a 3762% increase in the peak stress compared to JIC C1 condition and a 1217% increase in the peak stress compared to the JIC C2 condition. When the debridement is not less than 3/8r, the peak stress in the residual lesion is rapidly falling and returns to the physiological level.
3.4 Model validation
The principal compressive trabecula loads the principal compressive stress of the femoral head (Figures 8.8c–d), which correlates well with the bone density distribution (Figure 8.8b) (Jang and Kim, 2008). The shape and location of the biomechanical transfer path for both load cases are consistent with the trabecular features in the cross-sections of the cadaver bone (Figure 8.8a) (Boyle and Kim, 2011; Jang and Kim, 2009). It is clearly the case that trabeculae in the corresponding areas are thinner. In the same time, there are strong similarities of stress patterns between the simulation results of our study and the previous results in the literature (Sverdlova and Witzel, 2010). Hence, we think that the FE results could mirror the physical phenomenon of the hip and evaluate the results.
4 Discussion
FAIBG represents a proven technique for maintaining the shape of the femoral head and reducing the risk of collapse of FHN in its early stages. Rosenwasser et al., (1994) first described thorough debridement and bone grafting for treating FHN. This technique is an effective method for young patients with FHN in early stages, which will delay the progression of osteoarthrosis and subsequent Total Hip Arthroplasty (THA). Tao et al. (2014) reported an 80% clinical success rate with a mean follow-up time of 24 months among 15 patients who had surgical therapy by thorough debridement with bone grafting. However, these procedures may cause serious artificial damage and complication by capsulotomies or the destruction of the cortical bone of the femur neck fundus, and also require relatively high surgical devices and technique. Shi et al. (2008) reported a study of 67 hips, with the treatment of internal bracket implanting with partial debridement for FHN. They came up with a 64.2% (43/67) success rate, with an average follow-up of 23 months. In 2013, Shi et al. commented on their results by treating 25 of 40 patients using the allograft fibula with partial debridement for FHN. They reported satisfactory results in 18/25 (72%) patients with 24 months follow-up. These minimally invasive procedures could reduce the artificial damage and complication but got a poorer clinical outcome, since they couldn’t provide both repaired materials and biomechanical structural support during the healing of the necrosis region. FAIBG with proper debridement is an effective head-preserving method, and we achieved an average clinical success rate of 90.3%, with a mean follow up time of 37.5 months (He et al., 2009). All these views are based on the clinical observation experience and lack of biomechanical basis. Hence, both thorough debridement and partial debridement are not accepted universally because these is no compelling evidence of which method could be better at reducing the collapse risk of the femoral head. It encourages us to describe our experiences of the computational biomechanical analysis of debridement extent to reduce the collapse risk of FHN.
In our study, we adopted a subject-specific computational approach to consider changes in stress distribution of anterolateral cortical bone and the residual necrotic bone. Figure 8.4 shows that the stress transfer path in both JIC C1 and C2 are completely broken off, which indicate that surgical intervention should be involved. The effect of debridement size with FAIBG on collapse risk is clearly demonstrated in Figure 8.5–8.6. After FAIBG, the stress of anterolateral cortical bone in all conditions could return to the physiological level and the decrement/increment of stress is less than 0.1% as the debridement radius increases in two cases; hence, the collapse risk of femoral head can be reduced effectively using allo-fibula support to bear the load. When debridement size is not less than 3/8r, the von Mises stress of the residual bone also returns to the pathological level, which denotes that the progression of necrosis wouldn’t deteriorate after surgical intervention. Our results provide specific biomechanical evidence to support the viewpoint that FAIBG can resist the collapse of FHN, and FAIBG with thorough debridement has a lower risk of collapse risk than with partial debridement.
Thorough debridement has been reported by previous studies (Rosenwasser et al., 1994; Tao et al., 2014; Meyers et al., 1983; Ko et al., 1995; Meyers and Convery, 1991; Gardeniers et al., 1999). However, this procedure is difficult and time consuming, which is associated with serious artificial damage. FAIBG with partial debridement can not only reduce the anterolateral cortical stress, but also ensure that the stress of the residual bone will not increase. This technique has a distinct biomechanical basis, and it is time saving and requires relatively fewer surgical devices. It also brings a low risk of damage. Hence, FAIBG without thorough debridement seems to be superior to FAIBG with thorough debridement.
5 Conclusion
In this chapter, we propose employing computational biomechanical technology to explore the different mechanical performances of FAIBG with or without thorough debridement, which provides a biomechanical basis for choosing the proper treatment in clinic. A total of 18 computational models were constructed and used to simulate two subtypes of FHN with seven debridement radii of FAIBG. The simulation results provide specific biomechanical evidence that the use of FAIBG can resist the collapse of FHN. Furthermore, FAIBG without thorough debridement, which not only requires relatively low surgical devices but also reduces damage, seems to be a better method of resisting the collapse of JIC C1 and JIC C2 FHN. This chapter is a preliminary approach to investigate FAIBG with thorough debridement; more detailed analysis will be reported in the near future.
6 Disclaimer
The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.
6.1 Funding
This study was supported by the Natural Science Foundation of Guangdong Province (2014A030310214). There are no financial and personal relationships with other people or organizations that could inappropriately influence our work.