Evaluation of patient specific spinopelvic mobility requires the detection of bony landmarks in lateral functional radiographs. Current manual landmarking methods are inefficient, and subjective. This study proposes a deep learning model to automate landmark detection and derivation of spinopelvic measurements (SPM).

A deep learning model was developed using an international multicenter imaging database of 26,109 landmarked preoperative, and postoperative, lateral functional radiographs (HREC: Bellberry: 2020-08-764-A-2). Three functional positions were analysed: 1) standing, 2) contralateral step-up and 3) flexed seated. Landmarks were manually captured and independently verified by qualified engineers during pre-operative planning with additional assistance of 3D computed tomography derived landmarks. Pelvic tilt (PT), sacral slope (SS), and lumbar lordotic angle (LLA) were derived from the predicted landmark coordinates. Interobserver variability was explored in a pilot study, consisting of 9 qualified engineers, annotating three functional images, while blinded to additional 3D information. The dataset was subdivided into 70:20:10 for training, validation, and testing.

The model produced a mean absolute error (MAE), for PT, SS, and LLA of 1.7°±3.1°, 3.4°±3.8°, 4.9°±4.5°, respectively. PT MAE values were dependent on functional position: standing 1.2°±1.3°, step 1.7°±4.0°, and seated 2.4°±3.3°, p< 0.001. The mean model prediction time was 0.7 seconds per image. The interobserver 95% confidence interval (CI) for engineer measured PT, SS and LLA (1.9°, 1.9°, 3.1°, respectively) was comparable to the MAE values generated by the model.

The model MAE reported comparable performance to the gold standard when blinded to additional 3D information. LLA prediction produced the lowest SPM accuracy potentially due to error propagation from the SS and L1 landmarks. Reduced PT accuracy in step and seated functional positions may be attributed to an increased occlusion of the pubic-symphysis landmark. Our model shows excellent performance when compared against the current gold standard manual annotation process.

The Coronal Plane Alignment of the Knee (CPAK) is a recent method for classifying knees using the hip-knee-ankle angle and joint line obliquity to assist surgeons in selection of an optimal alignment philosophy in total knee arthroplasty (TKA)1. It is unclear, however, how CPAK classification impacts pre-operative joint balance. Our objective was to characterise joint balance differences between CPAK categories.

A retrospective review of TKA's using the OMNIBotics platform and BalanceBot (Corin, UK) using a tibia first workflow was performed. Lateral distal femoral angle (LDFA) and medial proximal tibial angle (MPTA) were landmarked intra-operatively and corrected for wear. Joint gaps were measured under a load of 70–90N after the tibial resection. Resection thicknesses were validated to recreate the pre-tibial resection joint balance.

Knees were subdivided into 9 categories as described by MacDessi et al.1 Differences in balance at 10°, 40° and 90° were determined using a one-way 2-tailed ANOVA test with a critical p-value of 0.05.

1124 knees satisfied inclusion criteria. The highest proportion of knees (60.7%) are CPAK I with a varus aHKA and Distal Apex JLO, 79.8% report a Distal Apex JLO and 69.3% report a varus aHKA. Greater medial gaps are observed in varus (I, IV, VII) compared to neutral (II, V, VIII) and valgus knees (III, VI, IX) (p<0.05 in all cases) as well as in the Distal Apex (I, II, III) compared to Neutral groups (IV, V, VI) (p<0.05 in all cases). Comparisons could not be made with the Proximal Apex groups due to low frequency (≤2.5%).

Significant differences in joint balance were observed between and within CPAK groups. Although both hip-knee-ankle angle and joint line orientation are associated with joint balance, boney anatomy alone is not sufficient to fully characterize the knee.

Aim

Patellofemoral Arthroplasty (PFA) prosthesis with asymmetric trochlear component was introduced as an improvement from existing designs for surgical treatment of symptomatic isolated patellofemoral arthritis. The purpose of this study was to evaluate midterm results in patients who underwent PFA procedure using such prosthesis.

This aim of this study was to identify common factors in patients with the shortest length of hospital stay following total hip arthroplasty (THA). This would then allow a means of targeting suitable patients to reduce their length of stay.

This was a retrospective cohort study of all patients undergoing primary THA at our institution between September 2013 and August 2014. Demographic data were collected from the patient record. The cohort was divided into those discharged to home within two days of operation and the rest of the THA population. The demographics (age, gender, ASA grade, body mass index (BMI), primary diagnosis, socioeconomic status (Scottish Index of Multiple Deprivation, SIMD and SIMD health domain) were compared between groups. In addition for the early discharge group information on comorbidities, family support at home and independent transport were collected.

The study cohort was 1292 patients. 119 patients were discharged home on the first post-operative day. Those discharged earlier were on average younger (p<0.0001), more likely to be male (p<0.0001) and had a lower ASA grade (p<0.00001). Other demographics did not differ between groups. Patients who were discharged early also appeared to have few comorbidities (Diabetes 5.9%, Cardiac disease 7.6%, Respiratory disease 9%), high levels of family support at home (95%) and high levels of independent transport arrangements (97%).

Factors associated with those patients with the shortest lengths of stay were identified. Such factors could be used to target patients who are suitable for streamlined recovery programmes aimed at early discharge after THA and assist with service planning.

Biological reconstruction techniques after diaphyseal tumour resection have increased in popularity in recent years. High complication and failure rates have been reported with intercalary allografts, with recent studies questioning their role in limb-salvage surgery. We developed a technique in which large segment allografts are augmented with intramedullary cement and fixed using compression plating. The goal of this study was to evaluate the survivorship, complications and functional outcomes of these intercalary reconstructions.

Forty-two patients who had reconstruction with an intercalary allograft following tumour resection between 1989 and 2010 were identified from our prospectively collected database. Allograft survival, local recurrence-free, disease-free and overall survival were assessed using the Kaplan-Meier method. Patient function was assessed using the Musculoskeletal Tumour Society (MSTS) scoring system and the Toronto Extremity Salvage Score (TESS).

The 23 women and 19 men had a mean age of 33 years (14–77). The most common diagnoses were osteosarcoma (n=16) and chondrosarcoma (n=9). There were 9 humerus, 18 femur and 15 tibia reconstructions. At a mean follow-up of 95 months (5–288), 31 patients were alive without disease, 10 were dead of disease and 1 was deceased of other causes. There were 4 local recurrences and 11 patients developed metastatic disease. 5-year local recurrence free survival was 92%, 5-year disease-free survival was 70% and overall survival was 75%. Fourteen of 42 patients (33%) experienced complications: 5 wound healing complications, 4 infections, 2 non-unions, 2 fractures and 1 nerve palsy. Four allografts (9.5%) were revised for complications and 2 (5%) for local recurrence. Mean allograft survival was 85 months (4–288). Mean time to union was 8.2 (3–36) months for the proximal osteotomy site and 8.1 (3–23) months for the distal osteotomy site. The mean score for MSTS 87 was 29.4 (+/− 4.4), MSTS 93 was 83.7 (+/−14.8) and TESS was 81.6 (+/−16.9).

An intercalary allograft augmented with intramedullary cement and compression plate fixation provides a reliable and durable method of reconstruction after tumour resection. Complication rates are comparable to the literature and are associated with high levels of patient function and satisfaction.

Long-term biological fixation and stability of uncemented acetabular implant are influenced by peri-prosthetic bone ingrowth which is known to follow the principle of mechanoregulatory tissue differentiation algorithm. A tissue differentiation is a complex set of cellular events which are largely influenced by various mechanical stimuli. Over the last decade, a number of cell-phenotype specific algorithms have been developed in order to simulate these complex cellular events during bone ingrowth. Higher bone ingrowth results in better implant fixation. It is hypothesized that these cellular events might influence the peri-prosthetic bone ingrowth and thereby implant fixation. Using a three-dimensional (3D) microscale FE model representing an implant-bone interface and a cell-phenotype specific algorithm, the objective of the study is to evaluate the influences of various cellular activities on peri-prosthetic tissue differentiation. Consequently the study aims at identifying those cellular activities that may enhance implant fixation.

The 3D microscale implant-bone interface model, comprising of Porocast Bead of BHR implant, granulation tissue and bone, was developed and meshed in ANSYS (Fig. 1b). Frictional contact (µ=0.5) was simulated at all interfaces. The displacement fields were transferred and prescribed at the top and bottom boundaries of the microscale model from a previously investigated macroscale implanted pelvis model (Fig. 1a) [4]. Periodic boundary conditions were imposed on the lateral surfaces. Linear elastic, isotropic material properties were assumed for all materials. Young's modulus and Poisson's ratios of bone and implant were mapped from the macroscale implanted pelvis [4]. A cell-phenotype specific mechanoregulatory algorithm was developed where various cellular activities and tissue formation were modeled with seven coupled differential equations [1, 2]. In order to evaluate the influence of various cellular activities, a Plackett-Burman DOE scheme was adopted. In the present study each of the cellular activity was assumed to be an independent factor. A total of 20 independent two-level factors were considered in this study which resulted in altogether 24 different combinations to be investigated. All these cellular activities were in turn assumed to be regulated by local mechanical stimulus [3]. The mechano-biological simulation was run until a convergence in tissue formation was attained.

The cell-phenotype specific algorithm predicted a progressive transformation of granulation tissue into bone, cartilage and fibrous tissue (Fig. 1c). Various cellular activities were found to influence the time to reach equilibrium in tissue differentiation and, thereby, attainment of sufficient implant fixation (Fig. 2, Table 1). Negative regression coefficients were predicted for the significant factors, differentiation rate of MSCs and bone matrix formation rate, indicating that these cellular activities favor peri-prosthetic bone ingrowth by facilitating rapid peri-prosthetic bone ingrowth. Osteoblast differentiation rate, on the contrary, was found to have the highest positive regression coefficient among the other cellular activities, indicating that an increase in this cellular activity delays the attainment of equilibrium in bone ingrowth prohibiting rapid implant fixation.

To view tables/figures, please contact authors directly.

The success of a cementless Total Hip Arthroplasty (THA) depends not only on initial micromotion, but also on long-term failure mechanisms, e.g., implant-bone interface stresses and stress shielding. Any preclinical investigation aimed at designing femoral implant needs to account for temporal evolution of interfacial condition, while dealing with these failure mechanisms. The goal of the present multi-criteria optimization study was to search for optimum implant geometry by implementing a novel machine learning framework comprised of a neural network (NN), genetic algorithm (GA) and finite element (FE) analysis. The optimum implant model was subsequently evaluated based on evolutionary interface conditions.

The optimization scheme of our earlier study [1] has been used here with an additional inclusion of an NN to predict the initial fixation of an implant model. The entire CAD based parameterization technique for the implant was described previously [1]. Three objective functions, the first two based on proximal resorbed Bone Mass Fraction (BMF) [1] and implant-bone interface failure index [1], respectively, and the other based on initial micromotion, were formulated to model the multi-criteria optimization problem. The first two objective functions, e.g., objectives f1 and f2, were calculated from the FE analysis (Ansys), whereas the third objective (f3) involved an NN developed for the purpose of predicting the post-operative micromotion based on the stem design parameters. Bonded interfacial condition was used to account for the effects of stress shielding and interface stresses, whereas a set of contact models were used to develop the NN for faster prediction of post-operative micromotion. A multi-criteria GA was executed up to a desired number of generations for optimization (Fig. 1). The final trade-off model was further evaluated using a combined remodelling and bone ingrowth simulation based on an evolutionary interface condition [2], and subsequently compared with a generic TriLock implant.

The non-dominated solutions obtained from the GA execution were interpolated to determine the 3D nature of the Pareto-optimal surface (Fig. 2). The effects of all failure mechanisms were found to be minimized in these optimized solutions (Fig. 2). However, the most compromised solution, i.e., the trade-off stem geometry (TSG), was chosen for further assessment based on evolutionary interfacial condition. The simulation-based combined remodelling and bone ingrowth study predicted a faster ingrowth for TSG as compared to the generic design. The surface area with post-operative (i.e., iteration 1) ingrowth was found to be ∼50% for the TSG, while that for the TriLock model was ∼38% (Fig. 3). However, both designs predicted similar long-term ingrowth (∼89% surface area). The long-term proximal bone resorption (upto lesser trochanter) was found to be ∼30% for the TSG, as compared to ∼37% for the TriLock model. The TSG was found to be bone-preserving with prominent frontal wedge and rectangular proximal section for better rotational stability; features present in some recent designs. The optimization scheme, therefore, appears to be a quick and robust preclinical assessment tool for cementless femoral implant design.

To view tables/figures, please contact authors directly.

Introduction

There is a growing recognition that evaluation should use patient-reported outcome tools and assessments of satisfaction in procedures like total knee replacement. These ensure that the patient's perception of outcome is included in the evaluation. Considering the increasing demands on physical function from the aging population, it is important to evaluate demanding physical activities for the population with end stage arthritis assigned for TKR.

Purpose

To validate accuracy of transepicondylar axis as a reference for femoral component rotation in primary total knee arthroplasty.

Purpose:

To compare accuracy of transepicondylar axis as a reference for femoral component rotation in primary navigated versus non navigated total knee arthroplasty in severely deformed knees.

Purpose

To assesment of geometric center of knee as a reference for femoral component rotation in primary total knee arthroplasty.

Despite the generally inferior clinical performance of acetabular prostheses as compared to the femoral implants, the causes of acetabular component loosening and the extent to which mechanical factors play a role in the failure mechanism are not clearly understood yet. The study was aimed at investigating the load transfer and bone remodelling around the uncemented acetabular prosthesis.

The 3-D FE model of a natural right hemi-pelvis was developed using CT-scan data. The same bone was implanted with two uncemented hemispherical acetabular components, one metallic (CoCrMo alloy) and the other ceramic (Biolox delta), with 54 mm outer diameter and 48 mm bearing diameter. The FE models of the implanted pelvis (containing ∼116000 quadratic tetrahedrals) were generated using a submodelling approach, which were based on an overall full model of implanted pelvis (containing ∼217600 quadratic tetrahedrals) acted upon by hip joint force and twenty one muscle forces. The apparent density (ρ in g cm⁻³) of each cancellous bone element was calculated using linear calibration of CT numbers of bone, from which the Young's modulus (E in MPa) was determined using the relationship, E = 2017.3 ρ^2.46 [1]. Implant-bone interface conditions, fully bonded and debonded with friction coefficient μ = 0.5, were simulated using contact elements. Applied loading conditions consist of two load cases during a gait cycle, corresponding to 13% and 52% of the walking cycle. Fixed constraints were prescribed at the pubis and at the sacroiliac joint. The bone remodelling algorithm was based on strain energy based site-specific formulation [2]. The FE analysis, in combination with the bone remodelling simulation, was performed using ANSYS FE software.

The predicted changes in peri-prosthetic bone density were similar for the metallic and the ceramic implant. For debonded implant-bone interface, stress shielding led to ∼20% reductions in bone density at supero-anterior, infero-anterior and posterior part of the acetabulum (Fig. 1). However, bone apposition was observed at the supero-posterior part of the acetabulum, where implantation led to ∼60% increase in bone density (Fig. 1). The effect of bone resorption was higher for the fully bonded implant-bone interface, wherein bone density reductions of 20–50% were observed in the cancellous bone underlying the implant (Fig. 1), which is indicative of implant loosening over time. However, implantation led to an increase in bone density around the acetabular rim for both the interface conditions (Fig. 1). These results are well corroborated by the earlier studies [3, 4]. Implantation with a ceramic component resulted in 2–7% increase in bone density at supero-posterior part of the acetabulum as compared to the metallic component, for the debonded interface condition. Considering better wear resistant properties and absence of metal ion release, results of this study suggest that the ceramic component might be a viable alternative to the metallic prosthesis.

The effects of metal ion release and wear particle debris in metal-on-metal articulation warrants an investigation of alternative material, like ceramics, as a low-wear bearing couple [1]. Short-stem resurfacing femoral implant, with a stem-tip located at the centre of the femoral head, appears to provide a better physiological load transfer within the femoral head and therefore seems to be a promising alternative to the long-stem design [2]. The objective of this study was to investigate the effect of evolutionary bone adaptation on load transfer and interfacial failure in cemented metallic and ceramic resurfacing implant.

Bone geometry and material properties of 3D finite element (FE) models (intact, short-stem metallic and ceramic resurfaced femurs of 44 mm head diameter) were derived from the CT scan data. The FE models consisted of 170352 quadratic tetrahedral elements and 238111 nodes with frictional contact at the implant-cement (μ = 0.3) and stem-bone interfaces (μ = 0.4) and fully bonded cement-bone interface. Normal walking and stair climbing were considered as two different loading conditions. A time-dependant “site specific” bone remodelling simulation was based on the strain energy density and internal free surface area of bone [3]. The variable time-step was determined after each remodelling iteration. The Hoffman failure criterion was used to assess cement-bone interfacial failure.

Predicted change in bone density due to bone remodelling was very much similar in both the metallic and ceramic resurfaced femurs (Fig. 1). Both the metallic and ceramic implant resulted in strain reduction in the proximal regions (Region of interest, ROI 2 and 4) and subsequent bone resorption, average bone density reduction by 72% (Fig. 1). Higher strains were generated in ROI 5 and 7, which caused bone apposition, an average increase in bone density of 145% (Fig. 1). The tensile stresses in the resurfacing implants increased with change in bone density; a maximum stress of 83 MPa and 63 MPa were observed in the ceramic and the metallic implants, respectively. The tensile stress in the cement mantle also increased with bone remodelling. Although the cement-bone interface was secure against interface debonding in the post-operative situation, calculations of Hoffman number indicated that risk of cement-bone interfacial failure was increased with peri-prosthetic bone adaptation.

During the remodelling simulation, maximum tensile stress in the implant and the cement was far below its strength. However, with bone adaptation greater volume of cement mantle was exposed to higher stresses which, in-turn, resulted in greater risk of interfacial failure around the periphery of the cement mantle. Both the short-stem ceramic and metallic resurfacing component, under debonded stem-bone interface, resulted in more physiological stress distribution across the femoral head. Based on these results, short-stem ceramic resurfacing component appears to be a viable alternative to the metallic design.

Introduction

Infection rates following arthroplasty surgery are reported between 1–4%, with considerably higher rates in revision surgery. The associated costs of treating infected arthroplasty cases are over 4 times the cost of primary arthroplasties, with significantly worse functional and satisfaction outcomes. In addition, multiple antibiotic resistant bacteria are developing, so to reduce the infection rates and costs associated with arthroplasty surgery, new preventative methods are required. HINS-light is a novel blue light inactivation technology which kills bacteria through a photodynamic process, and is proven to have bactericidal activity against a wide range of species. The aim of this study was to investigate the efficacy of HINS-light for the inactivation of bacteria isolated from infected arthoplasty cases.