Midshaft is the most commonly used site for behavioral and robusticity inferences in cross-sectional studies of the femur. This work tries to revive an alternative location because the midshaft, as much as it is easy to locate, does not necessarily reflect the same adaptive remodeling in every individual. Femoral waist which is defined as the shaft’s mechanically weakest point is reintroduced as an alternative. The aim of this work is to describe waist’s general patterns of distribution along the shaft, to test which morphological characteristics influence its position, to compare its inferential potential with the midshaft, and to evaluate applicability of the concept under the mechanical predictions about stress and strain distribution along the femur. A total of 251 individuals representing four temporal samples spanning the Eneolithics to 19th century were analyzed using CT-derived cross-sectional properties. The results showed that the femoral waist tends to center around the midshaft but with a rather large amount of variation and that the samples do not seem to differ in this respect. General levels of mechanical loading and robusticity may influence its distribution as evidenced from the Early Middle Ages males who were the most robust group (in body size adjusted parameters) and had their waists positioned more proximally. Variables that influence waist’s location are primarily related to strain distribution but not its magnitude. Thus, neck-shaft angle, anteversion angle, femoral inclination, crural index, and femoral curvature proved to be significant predictors (their importance varies between sexes), while body size measures were insignificant. Behavioral and robusticity inferences made from the femoral waist and midshaft are rather incompatible. In closely related populations, results from the two locations would probably lead to different interpretations while in rather distant units (species or genera), this would not be as problematic. Whenever possible, true cross-sectional properties should be used to locate the femoral waist. External methods using subperiosteal contour are more acceptable than methods estimating cross-sectional properties only from external dimensions. Lastly, the femoral waist position can potentially be used for taxonomic purposes in earlier hominins as well as for the reconstruction of other femoral characteristics (e.g., femoral length). / acase@tulane.edu

We introduce a new way to accurately segment the 3D femur from pelvic CT scans. The femur is a difficult target for segmentation due to its proximity to the acetabulum, irregular shape and the varying thickness of its hardened outer shell. Atypical bone morphologies, such as the ones present in hips suffering from Femoral Acetabular Impingements (FAIs) can also provide additional challenges to segmentation. We overcome these difficulties by (a) dividing the femur into the femur head and body regions (b) analysis of the femur-head and neighbouring acetabulum’s composition (c) segmentations with two levels of detail – rough and fine contours. Segmentations of the CT volume are performed iteratively, on a slice-by-slice basis and contours are extracted using the morphological snake algorithm. Our methodology was designed to require little initialization from the user and to deftly handle the large variation in femur shapes, most notably from deformations attributed to cam-type FAIs. Our efforts are to provide physicians with a new tool that creates patient-specific and high-quality 3D femur models while requiring much less time and effort. We tested our methodology on a database of 20 CT volumes acquired at the Ottawa General Hospital during a study into FAIs. We selected 6 CT scans from the database, for a total of 12 femurs, considering wide inter-patient variations. Of the 6 patients, 4 had unilateral cam-type FAIs, 1 had a bilateral cam-type FAI and the last was from a control group. The femurs segmented with our method achieved an average volume overlap error of 2.71 ± 0.44% and an average symmetric surface distance of 0.28 ± 0.04 mm compared against the same, manually segmented femurs. These results are better than all comparable literature and accurate enough to be used to in the creation of patient-specific 3D models.

We introduce a new way to accurately segment the 3D femur from pelvic CT scans. The femur is a difficult target for segmentation due to its proximity to the acetabulum, irregular shape and the varying thickness of its hardened outer shell. Atypical bone morphologies, such as the ones present in hips suffering from Femoral Acetabular Impingements (FAIs) can also provide additional challenges to segmentation. We overcome these difficulties by (a) dividing the femur into the femur head and body regions (b) analysis of the femur-head and neighbouring acetabulum’s composition (c) segmentations with two levels of detail – rough and fine contours. Segmentations of the CT volume are performed iteratively, on a slice-by-slice basis and contours are extracted using the morphological snake algorithm. Our methodology was designed to require little initialization from the user and to deftly handle the large variation in femur shapes, most notably from deformations attributed to cam-type FAIs. Our efforts are to provide physicians with a new tool that creates patient-specific and high-quality 3D femur models while requiring much less time and effort. We tested our methodology on a database of 20 CT volumes acquired at the Ottawa General Hospital during a study into FAIs. We selected 6 CT scans from the database, for a total of 12 femurs, considering wide inter-patient variations. Of the 6 patients, 4 had unilateral cam-type FAIs, 1 had a bilateral cam-type FAI and the last was from a control group. The femurs segmented with our method achieved an average volume overlap error of 2.71 ± 0.44% and an average symmetric surface distance of 0.28 ± 0.04 mm compared against the same, manually segmented femurs. These results are better than all comparable literature and accurate enough to be used to in the creation of patient-specific 3D models.

No description available.

Femoral shaft fractures are commonly encountered in orthopaedics and are typically treated using intramedullary (IM) nailing under fluoroscopic guidance. Inaccuracies in the location of the entry point of the nail and the alignment of the reduced fracture are not uncommon during this procedure. This can greatly increase the risk of iatrogenic fractures, malunions and, potentially, secondary degenerative joint disease. Fluoroscopy-based computer-assisted navigation workflows have been developed but are, as yet, not widely used. As such, there is a need to investigate the performance of these systems as well as the possibility of using newer imaging methods to enhance the reduction accuracy. This body of work investigated the impact of off-angle fluoroscopic images on the accuracy and precision of the selection of the entry points used in IM nailing and found that, while images were considered to be clinically acceptable, they resulted in large deviations in the selection of the entry point. Although higher precision was achieved with navigation, it did not improve the accuracy. This work extended the investigation of off-angle images by investigating the impact of the variation in the landmarks used by current navigation methods on quantification of femoral anteversion. The observed landmark variations were propagated through the calculation of femoral anteversion and yielded errors exceeding current clinical tolerances. As an alternative to fluoroscopy, this work developed two, semi-automated algorithms to quantify femoral shaft fracture alignment in six degrees of freedom (6DOF) based on a single, intraoperative cone-beam CT scan. Both algorithms were able to accurately quantify malalignment in all 6DOF with high repeatability and limited user interaction over a range of complex femoral shaft fractures, even in cases with severe comminution. The time requirements for the utilization of these algorithms were reasonable with respect to the time required for current, fluoroscopy-based navigation. Therefore, both of these algorithms would provide an efficient, robust and accurate alternative for the quantification of malalignment in 6DOF. Such an accurate and robust quantification of malalignment, when paired with the high precision tracking in current navigation systems, may enable the improvement of reduction accuracy in the treatment of complex femoral shaft fractures.

We introduce a new way to accurately segment the 3D femur from pelvic CT scans. The femur is a difficult target for segmentation due to its proximity to the acetabulum, irregular shape and the varying thickness of its hardened outer shell. Atypical bone morphologies, such as the ones present in hips suffering from Femoral Acetabular Impingements (FAIs) can also provide additional challenges to segmentation. We overcome these difficulties by (a) dividing the femur into the femur head and body regions (b) analysis of the femur-head and neighbouring acetabulum’s composition (c) segmentations with two levels of detail – rough and fine contours. Segmentations of the CT volume are performed iteratively, on a slice-by-slice basis and contours are extracted using the morphological snake algorithm. Our methodology was designed to require little initialization from the user and to deftly handle the large variation in femur shapes, most notably from deformations attributed to cam-type FAIs. Our efforts are to provide physicians with a new tool that creates patient-specific and high-quality 3D femur models while requiring much less time and effort. We tested our methodology on a database of 20 CT volumes acquired at the Ottawa General Hospital during a study into FAIs. We selected 6 CT scans from the database, for a total of 12 femurs, considering wide inter-patient variations. Of the 6 patients, 4 had unilateral cam-type FAIs, 1 had a bilateral cam-type FAI and the last was from a control group. The femurs segmented with our method achieved an average volume overlap error of 2.71 ± 0.44% and an average symmetric surface distance of 0.28 ± 0.04 mm compared against the same, manually segmented femurs. These results are better than all comparable literature and accurate enough to be used to in the creation of patient-specific 3D models.

Femoral shaft fractures are commonly encountered in orthopaedics and are typically treated using intramedullary (IM) nailing under fluoroscopic guidance. Inaccuracies in the location of the entry point of the nail and the alignment of the reduced fracture are not uncommon during this procedure. This can greatly increase the risk of iatrogenic fractures, malunions and, potentially, secondary degenerative joint disease. Fluoroscopy-based computer-assisted navigation workflows have been developed but are, as yet, not widely used. As such, there is a need to investigate the performance of these systems as well as the possibility of using newer imaging methods to enhance the reduction accuracy. This body of work investigated the impact of off-angle fluoroscopic images on the accuracy and precision of the selection of the entry points used in IM nailing and found that, while images were considered to be clinically acceptable, they resulted in large deviations in the selection of the entry point. Although higher precision was achieved with navigation, it did not improve the accuracy. This work extended the investigation of off-angle images by investigating the impact of the variation in the landmarks used by current navigation methods on quantification of femoral anteversion. The observed landmark variations were propagated through the calculation of femoral anteversion and yielded errors exceeding current clinical tolerances. As an alternative to fluoroscopy, this work developed two, semi-automated algorithms to quantify femoral shaft fracture alignment in six degrees of freedom (6DOF) based on a single, intraoperative cone-beam CT scan. Both algorithms were able to accurately quantify malalignment in all 6DOF with high repeatability and limited user interaction over a range of complex femoral shaft fractures, even in cases with severe comminution. The time requirements for the utilization of these algorithms were reasonable with respect to the time required for current, fluoroscopy-based navigation. Therefore, both of these algorithms would provide an efficient, robust and accurate alternative for the quantification of malalignment in 6DOF. Such an accurate and robust quantification of malalignment, when paired with the high precision tracking in current navigation systems, may enable the improvement of reduction accuracy in the treatment of complex femoral shaft fractures.

Total hip arthroplasty (THA) is a successful surgical technique that can be used for the effective treatment of fractured neck of femur, osteoarthritis, tumours, avascular necrosis, failed internal fixation, developmental dysplasia and rheumatoid arthritis. Revision surgery is necessary if loosening allows relative motion between the femoral stem and femur, causing pain and mechanical instability of the THA. The large number of revision operations undertaken each year as a result of implant failure emphasises the need for better biomechanical understanding of the femoral implant system. During 2001-02 in Australia 26,689 hip replacement operations were performed, with 3,710 of these being revision operations. The Exeter stem is the most commonly used cemented stem for primary and revision hip replacement in Australia. It is therefore very important to understand the mechanics of this clinically successful implant. Few studies have presented a through investigation into the mechanics of the Exeter stem from a fundamental perspective. To address these issues, mechanical and finite element (FE) methods were used to conduct experiments and numerical investigations into the mechanics of the Exeter stem. The femur geometry, for both the experimental and FE studies, was based upon the Sawbones model 3303 medium left third generation femur. The stem orientation for all specimens of the study was replicated from the orientation achieved by the senior surgeon implanting into the Sawbones femur. Test rigs were designed specifically to constrain the femur for the purposes of loading and stability measurements. The experimental investigation was used to investigate the torsional mechanical stability of the stem and to monitor this stability following periods of cyclic loading, using a resultant hip contact force, while monitoring the distal migration of the stem. The experimental investigation was also able to provide data for the validation of the finite element model. The resultant hip contact force was represented experimentally by a cyclic load of 1Hz applied to the head of the implant. The specimen was tested for four days. The loading regime for the initially implanted specimen involved the application of load for 6 hours a day, allowing the specimen to relax under no load for 18 hours a day. The mechanical stability of the initially implanted specimen was tested prior to the application of the cyclic load and immediately after the loading periods, prior to relaxation. Further tests were undertaken to assess the mechanical stability of the stem following the removal and reimplantation of the same stem without the use of additional bone cement (a procedure used surgically when only the acetabular component requires replacement). The reimplanted specimens were tested for a further two days following reimplantation. The six hours of loading for the reimplanted specimen was achieved using three, two hour loading periods. The stability of the reimplaned stem was assessed following each loading period. Initial studies found that the material properties of the Sawbones femurs were highly temperature dependent. If the temperature of the short glass fibre reinforced (SGFR) epoxy used for the cortical bone analogue was increased from room temperature to body temperature there was a reduction in the Young's modulus of up to 37 percent. This finding led to further investigation into the strain state of the femur for varus and neutral stem orientations to reduce femur failure during cyclic loading. The strains of the varus stem orientation were found to be higher than the strains of the neutral stem. The experiments investigating the mechanical stability under cyclic loading continued using the neutral stem orientation. For the neutral stem orientation it was found that there was no perceivable variation in the torsional stiffness of the initially implanted system during the cyclic loading period even though distal migration was observed. Torsional stiffness was observed to be compromised immediately after reimplantation. However, the torsional stiffness of the reimplanted specimen was recovered within the first two hour loading period. No perceivable variation in the torsional stiffness was observed between the initially implanted specimens and the reimplanted specimens following the first two hours of loading. The finite element model (FEM) found good agreement with the experimental investigation in terms of measured strain at two of three rosette positions and failure of the cortical bone. Trends for the stress-strain state of the stem showed good agreement with the clinical findings of failure and wear of the stem. The stress-strain state of the cement predicted the expected compressive and hoop stresses once debonding of the stem-cement interface had progressed. Strain on the surface of the femur was well predicted for pure torsional loading. The FEM has provided a valuable tool for future investigation of the effect of factors such as implant positioning on femoral component mechanics. The experimental and finite element models developed within the scope of this project have provided a powerful analysis tool for the investigation of the femoral component mechanics in THA. Application of the model to clinically relevant problems has given valuable insight into the mechanisms behind the success of this particular implant type. Models such as this will provide information on implant failure modes that will further lead to an increased implant life expectancy and a reduction in the number of revision operations performed.

Thesis (MEng)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: Bacterial colonisation and biofilm formation onto total hip replacement femoral stems remain a serious complication detrimental to the success of total hip arthroplasty. Current treatment procedures are accompanied by a heavy financial burden and morbidity for the patient while failing to guarantee a successful outcome with no reinfection. In fact, infection rates after revision surgeries are typically higher than those for primary hip arthroplasties. This study investigates conceptual drug delivery channels to be incorporated within cementless femoral stems by applying additive manufacturing as enabler technology. Drug delivery from these features is aimed at both prophylaxis and treatment of infection, with the latter emphasising the concept of creating a reinforceable antimicrobial depot inside the implant. The novelty lies in facilitating the administration of multiple drug dosages from within the implant instead of the once-off implant-based release strategies currently employed. Samples containing internal channels were designed based on analogies to drug delivery studies reporting on the commercial antibiotic loaded bone cement, Palacos R+G loaded with gentamicin. These samples were manufactured by LaserCUSING® from Ti-6Al-4V ELI powder. For prophylactic proof of concept, testing the channels were filled with Palacos R+G and challenged with two clinical isolates of Staphylococcus aureus in a bacterial growth inhibition study. Gentamicin-susceptible S.aureus Xen 36 was prevented from colonising for a minimum of 72 hours, whereas gentamicin resistant S.aureus Xen 31 reached the material within 24 h, signifying the importance of drug selection according to pathogen. Hence, a solution of vancomycin in phosphate buffered saline pH 7.4 was used during in vitro reservoir release testing. Three dosage injections were made into each of six samples during a cumulative incubation period of 100 h. A biocompatible 5,000 Da molecular weight cut off polyethersulfone nanoporous membrane was employed as release rate-controlling device. Released vancomycin was quantified with reversed phase high performance liquid chromatography. The resulting release profile was characterised by means of the Korsmeyer-and-Peppas model for diffusion based drug delivery. Constraint diffusion was identified as the mechanism controlling release, implying interplay between Fickian diffusion and polymer relaxation for effecting vancomycin release from within the reservoir. The concept created in this study provides a basis towards the development of full scale intelligent implants with multiple dose in situ drug delivery capabilities. Implants incorporating this concept could aid in the perpetual struggle against infection by providing a new strategy for delivery of high level antibiotics directly to the site of infection. / AFRIKAANSE OPSOMMING: Kolonisering van bakterieë en die vorming van biofilms op totale heupvervanging femorale stamme bly ʼn ernstige komplikasie tot die sukses van ʼn totale heupvervanging. Huidige behandelingsprosedures gaan gepaard met ʼn swaar finansiële las en infeksies vir die pasiënt sonder versekering van ʼn suksesvolle uitkoms. Die voorkoms van infeksie is tipies hoër vir revisie prosedures as vir primêre heupvervangings. Hierdie studie ondersoek konseptuele dwelmvoorsieningskanale wat in sementvrye femorale stamme geïnkorporeer kan word deur middel van toevoegingsvervaardiging as bemagtigingstegnologie. Dwelmtoediening deur hierdie strukture is gemik op beide voorkoming en behandeling van infeksie met ʼn klem op die konsep van ʼn herlaaibare antimikrobiese depot in die implantaat. Die nuwigheid lê in die fasilitasie van toediening van veelvuldige dwelm dosisse vanaf binne die implantaat in plaas van die huidige eenmalige implantaatgebaseerde toedieningstrategieë. Die ontwerp van monsters wat interne kanale bevat is baseer op vergelykings met literatuur op kommersiële beensement, Palacos R+G, wat met gentamisien belaai is. Hierdie monsters is vervaardig met LaserCUSING® van Ti-6Al-4V ELI poeier. Kanale is gevul met Palacos R+G en uitgedaag met twee kliniese isolate van Staphylococcus aureus in ʼn bakteriële groei inhibisie studie. Gentamisien-sensitiewe S.aureus Xen 36 is verhoed tot kolonisasie vir ʼn minimum van 72 uur, terwyl gentamisien-weerstandige S.aureus Xen 31 die materiaal binne 24 uur bereik het. Vervolgens was ‘n oplossing van vankomisien in fosfaatbuffer soutoplossing pH 7.4 gebruik tydens in vitro vrystellingstoetse. Drie dosisse inspuitings is in elk van ses monsters oor ʼn inkubasietydperk van 100 uur toegedien. ʼn Biokompatibiele 5,000 Da molekulêre massa afsnypunt polietersulfoon nanoporeuse membraan is as vrystellingskoers regulerende apparaat gebruik. Vrygestelde vankomisien is gekwantifiseer met omgekeerde fase hoë verrigting vloeistof-chromatografie. Die vrystellingsprofiel is gekarakteriseer met die Korsmeyer-en-Peppas model vir diffusie gebaseerde dwelmtoediening. Beperkte diffusie is geïdentifiseer as die meganisme wat vrystelling beheer. Dit impliseer ʼn tussenspel in Fickiaanse diffusie en polimeer verslapping vir die tot stand bring van vankomisien vrystelling vanaf binne die reservoir. Hierdie konsep van die studie verskaf ʼn basis vir die ontwikkeling van volskaalse intelligente implantate met ʼn veelvuldige in situ dwelmtoedienings vermoë. Implantate met hierdie konseptuele strukture kan steun gee aan die voortdurende stryd teen infeksie deur ʼn nuwe strategie daar te stel vir die toediening van hoë vlak antibiotika direk by die infeksie area.

Hydroxyproline (Hyp), a non-proteinogenic amino acid is a component of the organic material in bone. It has been used for 14C-dating of bone and the measurement of Hyp could be used as a biomarker in bone metabolism. Hydroxyproline is a component of collagen, the main structural protein in bone. The analyses of 14C in collagen and Hyp in human bones may provide timing information about bone processes and diseases, such as osteoarthritis and osteoporosis. The analysis of Hyp in bones (e.g., the determination of Hyp content) primarily relies on a spectrometric technique, liquid chromatography-mass spectrometry (LC-MS), and the determination of 14C content requires accelerator mass spectrometry (AMS). Moreover, to obtain these materials from bone requires the successful extraction of collagen and thr separation of Hyp from the collagen. This study aims at comparing methods for extracting collagen from bone, which do not destroy the Hyp. These methods include the use of either NaOH, KOH or HCl in one stage of the extraction process and separating sufficient Hyp for 14C analysis. This will provide information to determine whether Hyp can be used as a biomarker for bone diseases like osteoarthritis and osteoporosis. A preliminary 14C AMS analysis on collagen extracted by the NaOH method was carried out on human bones previously analyzed for forensic purposes. This demonstrated the ability of this technique to provide recent (post 1950) timing information. The collagen extractions by three different methods were first conducted on modern chicken bone, and the results showed that KOH method is the best bone collagen extraction method, yielding a largest quantity of Hyp. The KOH method was then employed to extract collagen from cow bone as a test of a more human-like (mammalian) material. As this was successful, collagen was extracted from diseased human bone fragments, obtained from the Ottawa Hospital. The data revealed that Hyp was successfully obtained from these bones. The study demonstrates that the extraction as well as the separation methods (preparative HPLC) can provide sufficient Hyp from bones for 14C AMS analysis. This will lead to future studies of Hyp in bone turnover, which may lead to its use as a novel biomarker for bone diseases such as osteoarthritis and osteoporosis.