MASCG: Multi-Atlas Segmentation Constrained Graph method for accurate segmentation of hip CT images

This paper addresses the issue of fully automatic segmentation of a hip CT image with the goal to preserve the joint structure for clinical applications in hip disease diagnosis and treatment. For this purpose, we propose a Multi-Atlas Segmentation Constrained Graph (MASCG) method. The MASCG method uses multi-atlas based mesh fusion results to initialize a bone sheetness based multi-label graph cut for an accurate hip CT segmentation which has the inherent advantage of automatic separation of the pelvic region from the bilateral proximal femoral regions. We then introduce a graph cut constrained graph search algorithm to further improve the segmentation accuracy around the bilateral hip joint regions. Taking manual segmentation as the ground truth, we evaluated the present approach on 30 hip CT images (60 hips) with a 15-fold cross validation. When the present approach was compared to manual segmentation, an average surface distance error of 0.30 mm, 0.29 mm, and 0.30 mm was found for the pelvis, the left proximal femur, and the right proximal femur, respectively. A further look at the bilateral hip joint regions demonstrated an average surface distance error of 0.16 mm, 0.21 mm and 0.20 mm for the acetabulum, the left femoral head, and the right femoral head, respectively.

Protrusio acetabuli: joint loading with severe pincer impingement and its theoretical implications for surgical therapy

Severe pincer impingement (acetabular protrusio) is an established cause of hip pain and osteoarthritis. The proposed underlying pathomechanism is a dynamic pathological contact of the prominent acetabular rim with the femoral head-neck junction. However, this cannot explain the classically described medial osteoarthritis in these hips. We therefore asked: (1) Does an overload exist in the medial aspect of the protrusio joint? and (2) What is the influence of three contemporary joint-preserving procedures on load distribution in protrusio hips? In vivo force and motion data for walking and standing to sitting were applied to six 3D finite element models (normal, dysplasia, protrusio, acetabular rim trimming, acetabular reorientation, and combined reorientation/rim trimming). Compared with dysplasia, the protrusio joint resulted in opposite patterns of von Mises stress and contact pressure during walking. In protrusio hips, we found an overload at the medial margin of the lunate surface (54% higher than normal). Isolated rim trimming further increased the medial overload (up to 28% higher than protrusio), whereas acetabular reorientation with/without rim trimming reduced stresses by up to 25%. Our results can be used as an adjunct for surgical decision making in the treatment of acetabular protrusio.

Validation of a standardized mapping system of the hip joint for radial MRA sequencing

OBJECTIVE Intraarticular gadolinium-enhanced magnetic resonance arthrography (MRA) is commonly applied to characterize morphological disorders of the hip. However, the reproducibility of retrieving anatomic landmarks on MRA scans and their correlation with intraarticular pathologies is unknown. A precise mapping system for the exact localization of hip pathomorphologies with radial MRA sequences is lacking. Therefore, the purpose of the study was the establishment and validation of a reproducible mapping system for radial sequences of hip MRA. MATERIALS AND METHODS Sixty-nine consecutive intraarticular gadolinium-enhanced hip MRAs were evaluated. Radial sequencing consisted of 14 cuts orientated along the axis of the femoral neck. Three orthopedic surgeons read the radial sequences independently. Each MRI was read twice with a minimum interval of 7 days from the first reading. The intra- and inter-observer reliability of the mapping procedure was determined. RESULTS A clockwise system for hip MRA was established. The teardrop figure served to determine the 6 o'clock position of the acetabulum; the center of the greater trochanter served to determine the 12 o'clock position of the femoral head-neck junction. The intra- and inter-observer ICCs to retrieve the correct 6/12 o'clock positions were 0.906-0.996 and 0.978-0.988, respectively. CONCLUSIONS The established mapping system for radial sequences of hip joint MRA is reproducible and easy to perform.

Computer Assisted Diagnosis and Treatment Planning of Femoroacetabular Impingement (FAI)

Femoroacetabular impingement (FAI) is a dynamic conflict of the hip defined by a pathological, early abutment of the proximal femur onto the acetabulum or pelvis. In the past two decades, FAI has received increasing focus in both research and clinical practice as a cause of hip pain and prearthrotic deformity. Anatomical abnormalities such as an aspherical femoral head (cam-type FAI), a focal or general overgrowth of the acetabulum (pincer-type FAI), a high riding greater or lesser trochanter (extra-articular FAI), or abnormal torsion of the femur have been identified as underlying pathomorphologies. Open and arthroscopic treatment options are available to correct the deformity and to allow impingement-free range of motion. In routine practice, diagnosis and treatment planning of FAI is based on clinical examination and conventional imaging modalities such as standard radiography, magnetic resonance arthrography (MRA), and computed tomography (CT). Modern software tools allow three-dimensional analysis of the hip joint by extracting pelvic landmarks from two-dimensional antero-posterior pelvic radiographs. An object-oriented cross-platform program (Hip2Norm) has been developed and validated to standardize pelvic rotation and tilt on conventional AP pelvis radiographs. It has been shown that Hip2Norm is an accurate, consistent, reliable and reproducible tool for the correction of selected hip parameters on conventional radiographs. In contrast to conventional imaging modalities, which provide only static visualization, novel computer assisted tools have been developed to allow the dynamic analysis of FAI pathomechanics. In this context, a validated, CT-based software package (HipMotion) has been introduced. HipMotion is based on polygonal three-dimensional models of the patient’s pelvis and femur. The software includes simulation methods for range of motion, collision detection and accurate mapping of impingement areas. A preoperative treatment plan can be created by performing a virtual resection of any mapped impingement zones both on the femoral head-neck junction, as well as the acetabular rim using the same three-dimensional models. The following book chapter provides a summarized description of current computer-assisted tools for the diagnosis and treatment planning of FAI highlighting the possibility for both static and dynamic evaluation, reliability and reproducibility, and its applicability to routine clinical use.

Surgical Technique: Reverse Periacetabular Osteotomy

Acetabular retroversion is the result of an externally rotated hemipelvis rather than a focal overgrowth of the anterior wall and/or hypoplasia of the posterior wall. Acetabular retroversion is a cause of pincer impingement which, if left untreated, can lead to hip pain and osteoarthritis. The causal surgical treatment in hips with acetabular retroversion is acetabular reorientation with a reverse periacetabular osteotomy (PAO). Indication is based on a positive correlation among symptoms (typically groin pain), physical findings on examination (positive anterior impingement test and decreased flexion and internal rotation), and radiographic signs for acetabular retroversion. These include a positive crossover, posterior wall, and ischial spine sign. A reverse PAO is performed with four osteotomies and a controlled fracture. Unlike reorientation of the acetabular fragment in dysplastic hips, correction for acetabular retroversion is achieved by a combined extension and internal rotation of the acetabular fragment. Typically, a small supra-acetabular wedge resection is required to allow sufficient extension of the fragment. The quality of acetabular reorientation is evaluated by intraoperative AP pelvic radiographs. In addition, intraoperative testing of range of motion following acetabular reorientation is mandatory. An arthrotomy and offset correction of the femoral head-neck area is indicated in hips with decreased internal rotation following acetabular reorientation. In a 10-year follow-up study of reverse PAO, a favorable outcome with preservation of all native joints was found. Correct acetabular orientation and, if necessary, a concomitant offset correction were the keys of successful outcome.

Prevention of cement leakage into the hip joint by a standard cement plug during PFN-A cement augmentation: a technical note.

Medial penetration of the helical blade into the hip joint after fixation of trochanteric fractures using the proximal femur nail antirotation (PFN-A) is a potential failure mode. In low demand patients a blade exchange with cement augmentation may be an option if conversion to total hip arthroplasty is unfeasible to salvage the cut-through. This article describes a technique to avoid intraarticular cement leakage using a cement plug to close the defect in the femoral head caused by the cut-through.

Characterization of tissue transglutaminase in human osteoblast-like cells

Tissue transglutaminase (tTG) is a calcium-dependent and guanosine 5'-triphosphate (GTP) binding enzyme, which catalyzes the post-translational modification of proteins by forming intermolecular ε(ϒ-glutamyl)lysine cross-links. In this study, human osteoblasts (HOBs) isolated from femoral head trabecular bone and two osteosarcoma cell lines (HOS and MG-63) were studied for their expression and localization of tTG. Quantitative evaluation of transglutaminase (TG) activity determined using the [1,414C]-putrescine incorporation assay showed that the enzyme was active in all cell types. However, there was a significantly higher activity in the cell homogenates of MG-63 cells as compared with HOB and HOS cells (p <0.001). There was no significant difference between the activity of the enzyme in HOB and HOS cells. All three cell types also have a small amount of active TG on their surface as determined by the incorporation of biotinylated cadaverine into fibronectin. Cell surface-related tTG was further shown by preincubation of cells with tTG antibody, which led to inhibition of cell attachment. Western blot analysis clearly indicated that the active TG was tTG and immunocytochemistry showed it be situated in the cytosol of the cells. In situ extracellular enzyme activity also was shown by the cell-mediated incorporation of fluorescein cadaverine into extracellular matrix (ECM) proteins. These results clearly showed that MG-63 cells have high extracellular activity, which colocalized with the ECM protein fibronectin and could be inhibited by the competitive primary amine substrate putrescine. The contribution of tTG to cell surface/matrix interactions and to the stabilization of the ECM of osteoblast cells therefore could by an important factor in the cascade of events leading to bone differentiation and mineralization.

In-Vivo Corrosion and Fretting of Modular TI-6AL-4V/CO-CR-MO Hip Prostheses: The Influence of Microstructure and Design Parameters

The purpose of this study was to evaluate the incidence of corrosion and fretting in 48 retrieved titanium-6aluminum-4vanadium and/or cobalt-chromium-molybdenum modular total hip prosthesis with respect to alloy material microstructure and design parameters. The results revealed vastly different performance results for the wide array of microstructures examined. Severe corrosion/fretting was seen in 100% of as-cast, 24% of low carbon wrought, 9% of high carbon wrought and 5% of solution heat treated cobalt-chrome. Severe corrosion/fretting was observed in 60% of Ti-6Al-4V components. Design features which allow for fluid entry and stagnation, amplification of contact pressure and/or increased micromotion were also shown to play a role. 75% of prosthesis with high femoral head-trunnion offset exhibited poor performance compared to 15% with a low offset. Large femoral heads (>32mm) did not exhibit poor corrosion or fretting. Implantation time was not sufficient to cause poor performance; 54% of prosthesis with greater than 10 years in-vivo demonstrated none or mild corrosion/fretting.

EVIDENCE FOR MECHANICAL STRAINS INFLUENCE IN OSTEOPHYTE DEVELOPMENT

Purpose: Osteophytes are osteo-cartilaginous metaplastic tissue outgrowths of bone capped by cartilage usually found in degenerative and inflammatory joint disease. The presence and degree of maturity of osteophytes, along with joint space narrowing, are the main radiographic criteria for diagnosis and grading osteoarthritis (OA). Although osteophytes are known for being anatomic signs of advanced OA, they can occur in non-symptomatic joints, in joints with no other observable alterations, and in early stage OA. It remains unclear if they develop from molecular, physiological and/or mechanical stimuli. We hypothesized that mechanical strains play a role in osteophyte development. The overall objective of this thesis was to find evidence that osteophytes are influenced by mechanical strains. Methods: The first project was to develop a mechanically-induced osteophyte animal model. One single impact load that was reported to induce moderate joint damage was applied to the periosteum of the rat knee. Animals were sacrificed at four time points to characterize the evolution of damaged tissue and the joint by histology. A second study using human mature hip osteophytes was conducted to evaluate if mature osteophyte presented histological signs of proliferating and developmental processes. The histological characterization of mature osteophyte was used to compare findings of the mechanically-induced osteophyte in the animal model to validate the use of this rodent model in studying some aspect of osteophyte development of human. Lastly, a detailed three-dimensional (3D) radiological morphometric analysis was performed on microscopic computed tomography (µCT) scanned femoral heads collected from total hip arthroplasty patients presenting mature hip osteophytes. Quantitative morphometric measures of osteophytes internal structure was compared to three regions of the femoral head of known quality of organisation and mechanical constraint. Results and Conclusion: Osteophyte can be mechanically induced by a single load impact to the joint periosteum, indicating that a moderate trauma to the periosteal layer of the joint may play a role in osteophyte development. Mature osteophytes have proliferation, developing and remodelling zones and have trabecular structures. Mechanically-induced osteophytes and mature osteophytes presented similar histological composition. Mature osteophytes have organized internal structure. These results provide evidence that mechanical strain can influence osteophyte development.

The role of chondrocyte senescence in the pathogenesis of canine osteoarthritis

The aims of this study were to (1) evaluate cellular senescence in chondrocytes from osteoarthritic articular cartilage, (2) investigate the hypothesis that oxidative stress is a feature of canine OA chondrocytes and that oxidative stress contributes to cellular senescence in canine chondrocytes, (3) investigate the hypothesis that osteoarthritic chondrocytes alter the gene expression of adjacent normal chondrocytes in OA joints leading to modulation of genes known to play a role in the pathogenesis of OA and (4) evaluate the presentation of dogs undergoing femoral head excision in veterinary referral practice in the UK as a treatment for osteoarthritis of the coxofemoral joint, and to categorise the distribution and severity of associated pathological lesions. Chondrocytes from osteoarthritic and normal cartilage were examined for levels of senescence. Initially chondrocytes were cultured using an alginate bead culture system, thought to mimic the extracellular matrix of articular cartilage. However, these chondrocytes showed almost no growth as compared to monolayer culture where they grew rapidly. OA chondrocytes entered the senescent state after 1.5 to 4.9 population doublings in monolayer culture, while normal chondrocytes underwent 4.8 to 14.6 population doublings before entering the senescent state. Osteoarthritic chondrocytes had increased levels of markers of cellular senescence (senescence associated beta-galactosidase accumulation and p16 protein accumulation) as compared to normal chondrocytes, suggesting that chondrocyte senescence is a feature of canine osteoarthritis. An experimental model for the induction of oxidative stress in chondrocyte cell culture was developed using tert-Butyl hydroperoxide and total cellular glutathione was measured as an indicator of cellular oxidative stress levels. Experimental induction of oxidative stress in both normal and osteoarthritic chondrocytes in cell culture resulted in increased amounts of cellular senescence, shown by an increase in levels of senescence associated beta-galactosidase accumulation and decreased replicative capacity. Experimental induction of oxidative stress also resulted in altered gene expression of three genes important to the degradation of the extracellular matrix; MMP-13, MMP-3 and Col-3A1, measured by RT-PCR, in normal canine chondrocytes in monolayer cell culture. MMP-3 showed the greatest relative expression change, with a fold-change of between 1.43 and 4.78. MMP-13 had a fold change of 1.16 to 1.38. Col-3A1 was down regulated, with a fold-change of between 0.21 and 0.31. These data demonstrate that experimentally induced oxidative stress in chondrocytes in monolayer culture increases levels of cellular senescence and alters the expression of genes relevant to the pathogenesis of canine OA. Coculture of osteoarthritic chondrocytes with normal canine chondrocytes resulted in gene modulation in the normal chondrocytes. Altered gene expression of ten genes known to play a role in the pathogenesis of osteoarthritis was detected in the normal chondrocytes (fold change shown in brackets); TNF-alpha (11.95), MMP-13 (5.93), MMP-3 (5.48), IL-4 (7.03), IL-6 (5.3), IL-8 (4.92), IL-F3 (4.22), COL-3A1 (4.12), ADAMTS-4 (3.78) and ADAMTS-5 (4.27). In total, 594 genes were significantly modulated suggesting that osteoarthritic chondrocytes contribute to the disease propagation by altering the gene expression of adjacent normal chondrocytes, thus recruiting them into the disease process. Gene expression changes were measured by microarray analysis and validated by RT-PCR and Western blot analysis. An epidemiological study of femoral heads collected from dogs undergoing total hip replacement surgery as a treatment for osteoarthritis of the coxofemoral joint secondary to canine hip dysplasia revealed that there was no characteristic pattern of cartilage lesion for canine hip dysplasia. Severe pathology of the femoral head with cartilage erosion occurred in 63.9% of cases and exposure of subchondral bone in 31.3% of cases. The work presented in this thesis has demonstrated that cellular senescence is a feature of chondrocytes from canine osteoarthritic cartilage and suggests that cellular senescence and oxidative stress play an important role in the pathogenesis of osteoarthritis in dogs.

Avaliação radiográfica da cobertura acetabular à cabeça femoral, após osteotomia tripla e aplicação de cunha sacroilíaca, em pelve de cadáveres de cães

Na tentativa de evitar algumas das dificuldades associadas à osteotomia pélvica tripla (OPT), foi desenvolvido experimentalmente o método de aplicação de cunha na junção sacroilíaca para aumentar a ventroversão acetabular. O objetivo deste estudo foi aplicar as técnicas de cunha sacroilíaca e OPT em cadáveres e avaliar radiograficamente a eficácia da ventroversão acetabular. Para tal, foram utilizados 10 cadáveres de cães, adultos, com 15-25 kg. Em cada hemipelve direita foi realizada OPT com placas de 20° e 40°. Na hemipelve esquerda foram aplicadas cunhas nas articulações sacroilíacas de 20° e 40°. Avaliações radiográficas em projeções ventrodorsais foram realizadas para mensuração da cobertura acetabular à cabeça femoral nas duas técnicas. De acordo com os dados obtidos pode-se observar que não houve diferença entre a técnica de OPT e o uso de cunha sacroilíaca utilizando implantes de 20° e 40°, mas ocorreu diferença significativa (p<0,05) entre os cães antes e após a aplicação dos implantes de 20° e 40°, e também entre os que receberam implantes de 20° e os de 40°. A aplicação de cunha sacroilíaca produziu resultados semelhantes à OPT, e também se mostrou como de mais fácil aplicação.

Lytic lesions of the femoral neck in rheumatoid arthritis simulating pigmented villonodular synovitis or malignancy

Large lytic lesions, relatively asymptomatic, involving the femoral neck and the base of the head are described in two patients suffering from a classical seropositive rheumatoid arthritis. Histological examination failed to reveal signs of malignancy, infection or pigmented villonodular synovitis. There were no rheumatoid nodules but a chronic hypertrophic villous synovitis was found. Rheumatoid synovium may invade the superior extremity of the femur; this fact is important in the differential diagnosis of destructive lesions of the femoral neck in RA.

Avaliação radiográfica da cobertura acetabular à cabeça femoral, após osteotomia tripla e aplicação de cunha sacroilíaca, em pelve de cadáveres de cães

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Diversity of the proximal femur in humans: morphological variations of the head-neck junction

The proximal femur is a high-diversity region of the human skeleton, especially at the anterior junction between head and neck, where various bony morphologies have been recognized since mid nineteenth century. Classical literature on this topic is chaotic and contradictory, making almost impossible the comparison of data from different researches. Starting from an extensive bibliographic review, the first standardized method to score these traits has been created. This method allows representing both the anatomical diversity of the region already described in literature and a part of variability not considered before, giving few and univocal definitions and allowing to collect comparable data. The method has been applied to three identified and five archaeological European skeletal collections, with the aim of investigating the distribution of these features by sex, age and side, in different places and time periods. It has also been applied to 3D digital reconstructions of femurs from CT scan files of coxo-femoral joints from fresh cadavers. In addition to the osseous traits described in the standardized method, the presence and frequency of some features known as herniation pits have been scored both on bones and on CT scans. The various osseous traits of the proximal femur are present at similar frequencies in skeletal samples from different countries and different historical periods, even if with clear local differentiation. Some of the features examined show significant trends related to their distribution by gender and age. Some hypotheses are proposed about the etiology of these morphologies and their possible implication with the acquisition of bipedalism in Humans. It is therefore highlighted the possible relation of some of these traits with the development of disorders of the hip joint. Moreover, it is not recommended the use of any of these features as a specific activity-related marker.