Anterior approach total hip arthroplasty (AA-THA) has a steep learning curve, with higher complication rates in initial cases. Proper surgical case selection during the learning curve can reduce early risk. This study aims to identify patient and radiographic factors associated with AA-THA difficulty using Machine Learning (ML).

Consecutive primary AA-THA patients from two centres, operated by two expert surgeons, were enrolled (excluding patients with prior hip surgery and first 100 cases per surgeon). K- means prototype clustering – an unsupervised ML algorithm – was used with two variables - operative duration and surgical complications within 6 weeks - to cluster operations into difficult or standard groups.

Radiographic measurements (neck shaft angle, offset, LCEA, inter-teardrop distance, Tonnis grade) were measured by two independent observers. These factors, alongside patient factors (BMI, age, sex, laterality) were employed in a multivariate logistic regression analysis and used for k-means clustering. Significant continuous variables were investigated for predictive accuracy using Receiver Operator Characteristics (ROC).

Out of 328 THAs analyzed, 130 (40%) were classified as difficult and 198 (60%) as standard. Difficult group had a mean operative time of 106mins (range 99–116) with 2 complications, while standard group had a mean operative time of 77mins (range 69–86) with 0 complications. Decreasing inter-teardrop distance (odds ratio [OR] 0.97, 95% confidence interval [CI] 0.95–0.99, p = 0.03) and right-sided operations (OR 1.73, 95% CI 1.10–2.72, p = 0.02) were associated with operative difficulty. However, ROC analysis showed poor predictive accuracy for these factors alone, with area under the curve of 0.56. Inter-observer reliability was reported as excellent (ICC >0.7).

Right-sided hips (for right-hand dominant surgeons) and decreasing inter-teardrop distance were associated with case difficulty in AA-THA. These data could guide case selection during the learning phase. A larger dataset with more complications may reveal further factors.

The Posterior and Lateral approaches are most commonly used for Total Hip Arthroplasty (THA) in the United Kingdom (UK). Fewer than 5% of UK surgeons routinely use the Direct Anterior Approach (DAA). DAA THA is increasing, particularly among surgeons who have learned the technique during overseas fellowships. Whether DAA offers long-term clinical benefit is unclear.

We undertook a retrospective analysis of prospectively collected 10-year, multi-surgeon, multi-centre implant surveillance study data for matched cohorts of patients whose operations were undertaken by either the DAA or posterior approach. All operations were undertaken using uncemented femoral and acetabular components. The implants were different for the two surgical approaches. We report the pre-operative, and post operative six-month, two-year, five-year and 10-year Oxford Hip Score (OHS) and 10-year revision rates.

125 patients underwent DAA THA; these patients were matched against those undergoing the posterior approach through propensity score matching for age, gender and body mass index. The 10-year revision rate for DAA THA was 3.2% (4/125) and 2.4% (3/125) for posterior THA. The difference in revision rate was not statistically significant. Both DAA and Posterior THA pre-operative OHS were comparable at 19.85 and 19.12 respectively. At the six-month time point, there was an OHS improvement of 20.89 points for DAA and 18.82 points for Posterior THA and this was statistically significant (P-Value <0.001). At the two, five and 10-year time-points the OHS and OHS improvement from the pre-operative review were comparable. At the 10-year time point post-op the OHS for DAA THA was 42.63, 42.10 for posterior THA and the mean improvement from pre-op to 10-years post op was 22.78 and 22.98 respectively. There was no statistical difference when comparing the OHS or the OHS mean improvements at the two, five and 10-year point.

Whilst there was greater improvement and statistical significance during the initial six month time period, as time went on there was no statistically significant difference between the outcome measures or revision rates for the two approaches.

Over 8000 total hip arthroplasties (THA) in the UK were revised in 2019, half for aseptic loosening. It is believed that Artificial Intelligence (AI) could identify or predict failing THA and result in early recognition of poorly performing implants and reduce patient suffering.

The aim of this study is to investigate whether Artificial Intelligence based machine learning (ML) / Deep Learning (DL) techniques can train an algorithm to identify and/or predict failing uncemented THA.

Consent was sought from patients followed up in a single design, uncemented THA implant surveillance study (2010–2021). Oxford hip scores and radiographs were collected at yearly intervals. Radiographs were analysed by 3 observers for presence of markers of implant loosening/failure: periprosthetic lucency, cortical hypertrophy, and pedestal formation.

DL using the RGB ResNet 18 model, with images entered chronologically, was trained according to revision status and radiographic features. Data augmentation and cross validation were used to increase the available training data, reduce bias, and improve verification of results.

184 patients consented to inclusion. 6 (3.2%) patients were revised for aseptic loosening. 2097 radiographs were analysed: 21 (11.4%) patients had three radiographic features of failure.

166 patients were used for ML algorithm testing of 3 scenarios to detect those who were revised. 1) The use of revision as an end point was associated with increased variability in accuracy. The area under the curve (AUC) was 23–97%. 2) Using 2/3 radiographic features associated with failure was associated with improved results, AUC: 75–100%. 3) Using 3/3 radiographic features, had less variability, reduced AUC of 73%, but 5/6 patients who had been revised were identified (total 66 identified).

The best algorithm identified the greatest number of revised hips (5/6), predicting failure 2–8 years before revision, before all radiographic features were visible and before a significant fall in the Oxford Hip score. True-Positive: 0.77, False Positive: 0.29.

ML algorithms can identify failing THA before visible features on radiographs or before PROM scores deteriorate. This is an important finding that could identify failing THA early.

Abstract

Abstract

Patient Reported Outcome Measures (PROMs) can be completed using paper and postal services (pPROMS) or via computer, tablet or smartphone (ePROMs). We have investigated whether there are differences in scores depending on the method of PROMs acquisition for the Oxford Knee (OKS) and the EQ-5D scores, at one and two years post operatively.

Patient demographics, mode of preferred data collection and pre-and post-operative PROMs for Total Knee Replacements (TKRs) performed between 1st January 2018 and 31st December 2018 were collected.

During the study period, 1573 patients underwent TKRs. The average OHS and EQ-5D pre-operatively scores was 19.47 and 0.40 respectively. 71.46% opted to undertake post-operative questionnaires using ePROMs. The remaining 28.54% opted for pPROMS. The one and two-year OHS for ePROMS patients increased to 37.64 and 39.76 while the OHS scores for pPROMS patients were 35.71 and 36.83. At the one and two-year post-operative time intervals, a Mann-Whitney test showed statistical significance between the modes of administration for OHS (P-Value = 0.044 and 0.01 respectively). The one and two-year EQ-5D for ePROMS patients increased to 0.76 and 0.78 while the EQ-5D scores for pPROMS patients were 0.73 and 0.76. The P-Value for Mann-Whitney tests comparing the modes of administration for EQ-5D were 0.04 and 0.07 respectively.

There is no agreed mode of PROMs data acquisition for the OKS and EQ-5D Scores. While we have demonstrated an apparent difference in scores depending on the mode of administration, further work is required to establish the influence of potentially confounding factors.

Disease specific or generic Patient Reported Outcome Measures (PROMs) can be completed by patients using paper and postal services (pPROMS) or via computer, tablet or smartphone (ePROMs) or by hybrid data collection, which uses both paper and electronic questionnaires. We have investigated whether there are differences in scores depending on the method of PROMs acquisition for the Oxford Hip Score (OHS) and the EQ-5D scores, at one and two years post operatively.

Patients for this study were identified retrospectively from a prospectively compiled arthroplasty database held at the study centre. Patient demographics, mode of preferred data collection and pre- and post-operative PROMs for Total Hip Replacements (THRs) performed at this centre between 1^st January 2018 and 31^st December 2018 were collected.

During the study period, 1494 patients underwent THRs and had complete one and two-year PROMs data available for analysis. All pre-operative scores were obtained by pPROMS. The average OHS and EQ-5D pre-operatively scores were 19.51 and 0.36 respectively. 72.02% of the patients consented to undertake post-operative questionnaires using ePROMs. The remaining 27.98% opted for pPROMS. The one and two-year OHS for ePROMS patients increased to 41.31 and 42.14 while the OHS scores for pPROMS patients were 39.80 and 39.83. At the one and two-year post-operative time intervals, a Mann-Whitney test showed statistical significance between the modes of administration for OHS (P-Value =0.044 and 0.01 respectively). The one and two-year EQ-5D for ePROMS patients increased to 0.83 and 0.84 while the EQ-5D scores for pPROMS patients were 0.79 and 0.81. The P-Value for Mann-Whitney tests comparing the modes of administration for EQ-5D were 0.13 and 0.07 respectively.

Within Orthopaedics, PROMs have become the most widely used instrument to assess patients’ subjective outcomes. However, there is no agreed mode of PROMs data acquisition. While we have demonstrated an apparent difference in scores depending on the mode of administration, further work is required to establish the influence of potentially confounding factors such as patient age, gender and familiarity with computer technology.

Clinical decision-making is often based on evidence of outcome after a specific treatment. Surgeons and patients may, have different perceptions and expectations of what to achieve following a Total Hip Replacement (THR). Several studies have shown that unfulfilled expectations are a principal source of patient dissatisfaction and patients are typically overly optimistic with regards to expected outcomes following surgery. Published data on clinical and functional outcomes show that persistence of symptoms, such as pain, and failure to return to preoperative levels of function are normal. To measure patient's expectations we undertook prospective study reviewing patients' expectations in 1800 THRs over a 21-year period (1997–2018).

Of the whole cohort, 48.98% patients reported they wanted a THR to overcome unbearable pain. 11.75 % wanted a THR to be able to walk without a limp. 9.69% wanted to a THR to increase walk endurance. 61.97% reported it was extremely important to decrease pain following a THR. In 2001, the most important reason for a THR reported by patients was to relieve unbearable pain and this remained the same most important reason in 2018. This result was also statistically significant with a p-value of 0.001. 80.36% reported they anticipated ‘no pain’ after recovery from a THR, 16.75% reported they anticipated ‘some pain’ and 2.89% reported they anticipated ‘extreme pain’ following a THR. 74.71% reported it was extremely important to increase their ability to undertake normal activities. 22.06% reported it was very important, 2.40% reported it was moderately important, 0.55% slightly important and 0.28% reported it was not all to important to increase their ability to undertake normal activity.

In conclusion patients' want to reduce their pain, walk normally and increase their level activities. Differences in expectation fulfilment may be due to unrealistic expectations. To achieve optimal outcome managing patient expectations is vital.

Background

For total hip arthroplasty (THA), cognitive training prior to performing real surgery may be an effective adjunct alongside simulation to shorten the learning curve. This study sought to create a cognitive training tool to perform direct anterior approach THA, validated by expert surgeons; and test its use as a training tool compared to conventional material.

Patients need to know the benefits, risks and alternatives to any proposed treatment. Surgeons discussing the risk of a revision procedure becoming necessary, after a hip replacement can draw upon the orthopaedic literature and arthroplasty registries for long-term implant survival. However, early revision is required in a minority of cases. We have investigated the probability for revision hip replacement patients in terms of time-point and indication for revision.

Of the 9,411 Primary Total Hip Replacements (THR), undertaken by 22 surgeons, over an eleven-year period, between January 2004 and March 2015, 1.70% (160) were subsequently reported to the National Joint Registry (NJR) as revised. Each revision case was reviewed under the supervision of senior hip specialist consultants. The modes of failure of were identified through clinical, laboratory and imaging (x-rays, CT, MRI and Isotope scans) studies.

The revision rate for THRs was 0.58% in the first year. This was statistically higher than all subsequent years, P-Value <0.001. There was no statistical difference between any pair of subsequent years. Thereafter, the average revision rate was 0.30% per annum. The odds ratio for revision during the first post–operative year against the subsequent year average was 1.67.

The indications for the early hip revisions in the first three years were infection, dislocation and peri-prosthetic fracture. The data from this study can help better inform patients of the revision rates after a primary THR and allow surgeons to develop implant surveillance strategies among high-risk patients.

INTRODUCTION

Historically, the clinical performance of novel implants was usually reported by designer surgeons who were the first to acquire clinical data. Regional and national registries now provide rapid access to survival data on new implants and drive ODEP ratings. To assess implant performance, clinical and radiological data is required in addition to implant survival. Prospective, multi-surgeon, multi-centre assessments have been advocated as the most meaningful. We report the preliminary results of such a study for the MiniHip™femoral component and Trinity™ acetabular component (Corin Ltd, UK).

We have investigated whether the pattern of subchondral acetabular cyst formation reflects hip pathology and may provide a prognostic indicator for treatment.

A single surgeon series of sequential hip arthroscopies was reviewed to identify the most recent 200 cases undertaken on a previously un-operated joint with pre-operative plain radiographs and computed tomography or magnetic resonance scan available for review. Also, serial “non-arthritic hip scores” (NAHS) recorded pre-operatively, at 6 weeks and 3 months post-surgery. The acetabular Lateral Centre Edge Angle, the Acetabular Index, the FEAR index and the Kallgren and Lawrence grade were determined. All images were reviewed by two independent assessors and divided into four groups according to acetabular subchondral cyst distribution. NC - No cysts - 74 cases (mean: 40.1 years), SPC - a Single Peripheral Cyst – 58 cases (mean: 43.1 years), SDC - a Solitary Dome Cyst – 32 cases (mean: 33 years) and MC - Multiple Cysts – 36 cases (mean: 42.7 years).

No association was identified between gender and patient reported outcomes.

SDC patients were significantly younger than the other three groups (p <0.001). At three months after surgery, the average increase of the NAHS in the four groups was 25.3, 23.5, 4.2 and 4.9 respectively. Acetabular dysplasia was identified in 72% of the SDC group compared to 18%, 16% and 33% in NC, SPC and MC groups. Degenerative change was identified in 86% of the MC group compared to 18%, 40% and 41% of the NC, SPC and SDC groups.

The early patient reported outcome following hip arthroscopy was significantly affected by the location and number of acetabular subchondral cysts. A Solitary Dome Cyst is indicative of underling hip dysplasia. Multiple Cysts are indicative of degenerative disease. The majority of patients in these groups have poor three-month outcomes after hip arthroscopy.

Total Hip Replacement (THR) is widely assumed to resolve sleep disturbance commonly experienced by individuals with hip osteoarthritis (OA). We report a study of 329 THRs with mean age of 71.9 years comparing pre-operative and one and two year post-operative patient reported outcomes for sleep disturbance to determine the veracity of this expectation. Data was collected from the validated Oxford Hip Patient Reported Questionnaire. Specifically, Question 12: “During the past four weeks, have you been troubled by pain from your hip in bed at night?” Answers to the question were multiple choice: No nights (4 points), Only 1 or 2 nights (3 points), Some nights (2 points), Most nights (1 point) and Every Night (0 points)

Pre-operatively, the mean score for patients with hip OA was 1.2/4. This increased to 3.5 at one year and was also maintained at two years. The pre- to post-operative improvement was significant at both one and two years for THR with p <0.00001. Pre-operatively, only 6% of patients with arthritic hips reported that they were never woken from sleep because of their painful hip. One year after THR 72% always enjoyed pain free sleeping and at two years this had risen to 75%. When patients who only experienced disturbance one or two nights per month were included, the three figures increased from 13% to 83% and 83% respectively.

The study confirms that sleep disturbance affects over 90% of patients with arthritic hip joints. Over 80% of THR patients will enjoy sleep that is seldom or never disturbed by their artificial hip. The improvement achieved by THR occurs within a year of surgery and is preserved at two years. In this regard, hip replacement is a highly effective intervention.

Introduction

Instability accounts for one third of revision total hip arthroplasty(rTHA) performed in the United Kingdom. Removal of well-fixed femoral stems in rTHA is challenging with a risk of blood loss and iatrogenic damage to the femur. The Bioball Universal Adaptor (BUA) (Merete, Germany), a modular head neck extension adaptor, provides a mechanism for optimisation of femoral offset, leg length and femoral anteversion. This can avoid the need for femoral stem revision in selected cases. There is a relative paucity of clinical data available with the use of this device.

Introduction

The Metasul articulation was introduced in the early 1990's. It comprises a 28mm forged, cobalt-chrome head and a forged metal inlay, set into the polyethylene element of the acetabular component. During the 1990's and early 2000's anatomic sized MoM articulations were re-introduced for hip resurfacing and stemmed implant designs. Adverse tissue reactions to metal debris and corrosion products subsequently led to a decline in MoM usage.

Objectives

This study examines variations in knee arthroplasty patient reported outcome measures according to patient age.

Using general practitioner records, hospital medical notes and through direct telephone conversation with patients, we investigated the accuracy of nine patient-reported complications after elective joint replacement surgery of the hip and knee. A total of 402 post-discharge complications were reported after 8546 elective operations that were undertaken within a three-year period. These were reported by 136 men and 240 women with an overall mean age of 71.8 years (34.3–93.2). A total of 319 (79.4%; 95% confidence interval, 75.4%–83.3%) reported complications were confirmed to be correct. Very high rates of correct reporting were demonstrated for infection (94.5%) and further surgery (100%), whereas the rates of reporting deep venous thrombosis (DVT), pulmonary embolism, myocardial infarct and stroke were lower (75%–84.2%). Dislocation, periprosthetic fractures and nerve palsy were associated with modest rates of correct reporting (36%–57.1%). More patients who had had knee surgery delivered incorrect reports of dislocation (p = 0.001) and DVT (p = 0.013). Despite these variations in accuracy, it appears that post-operative complications may form part of a larger patient-reported outcome programme for monitoring outcome after elective joint replacement surgery.

Introduction

Total hip arthroplasty (THA) is undertaken to relieve pain and to restore mobility. The orthopaedic community remains divided on the influence of surgical approach in achieving functional recovery most quickly and effectively. We report a study comparing THA performed through a posterior (Posterior) against anterior approach (Heuter).

Introduction

The outcomes programme of our institution has been developed from a system first used at Epsom and St Helier NHS Trust 15 years ago. The system was implemented at our institution when it opened in 2004, and has been used to collect data on over 17,000 joint replacement operations so far. A bespoke database is used to collect, analyse and report outcome data.

Introduction

The MITCH PCR is an anatomic, flexible, horse-shoe shaped acetabular component, with 2 polar fins. The rationale of the PCR cup design is to reproduce a near-physiological stress distribution in the bone adjacent to the prosthesis. The thin composite cup is designed to fuse and flex in harmony with the surrounding bony structure. Only the pathological acetabular cartilage and underlying subchondral bone of the horseshoe-shaped, load-bearing portion of the acetabular socket is replaced, thus preserving viable bone stock. The PCR is manufactured from injection moulded carbon fibre reinforced polyetheretherketone (PEEK), with a two layer outer surface comprising hydroxyapatite and plasma sprayed commercially pure titanium. It is implanted in conjunction with a large diameter low wear femoral head, producing a bearing that will generate minimal wear debris with relatively inert particles.

Pre-clinical mechanical testing, finite element analysis and biocompatibility studies have been undertaken. FEA evaluation predicts preservation of host bone density in the load bearing segments. A pilot clinical study was completed on a proto-type version of the PCR cup (the “Cambridge” cup), achieving excellent 5 and 10 year results.

Introduction

The need for the stringent surveillance of new devices was recognised by the NICE review of hip replacement surgery in 2000 and led to the Orthopaedic Data Evaluation Panel (ODEP) developing criteria for post-marketing surveillance (PMS) studies. This requirement has been reinforced by the recent recall of ASR devices.