BASK 2026 Free Paper Presentations

Podium Presentations

Day 1 – Tuesday 14th April

Charlotte McMain1, Petra Bonacic-Bartolin1, David Haslhofer2, Simon Ball2, Andy Williams2, Andrew Amis1

1Imperial College, London, United Kingdom. 2Fortius Clinic, London, United Kingdom

Abstract

Introduction: The objective was to explain the mechanism of non-contact ACL injuries of elite footballers by analysing MR images and reconstructing the bone displacements when bone bruises (BB) occurred. Calculating ligament strains would predict damaged secondary restraints.

Methods: 90 MRIs were obtained post ACL injury and 20 were analysed. 3D bone models with BBs were manipulated to oppose the femoral and tibial BBs. Ligament attachments defined each major ligament in the anatomical posture, then the bones were moved in 20 steps towards the BB posture and ligament strains calculated. Ruptures were predicted from published tensile data.

Results: None of 90 MRIs had medial compartment impact BBs, so 100% of the injuries were in valgus. What appeared to be medial “BB”s were avulsions of femoral deep MCL or tibial ramp lesions. BBs were in the sulcus of the lateral femoral condyle and posterolateral tibial plateau. With the BBs opposed the tibia was flexed 60deg and externally rotated 15deg. The ACL rupture occurred at step 8/20 (20mm anterior drawer) and the deep MCL at step 11/20, thus both in 100% of the knees, with the POL rupturing in 65%, the superficial MCL 35%, the ALL in 30% and the LCL in 0%.

Conclusions: There was a valgus plus external rotation mechanism in 100%. Medial bone lesions were avulsions. The deep MCL failed in 100% of knees analysed. The ACL ruptured at less than half the subluxation towards the BB impact, releasing the tibia to move further and damage the secondary restraints.

Helen Sankey, Nicholas Bowman

Maidstone & Tunbridge Wells NHS Trust, Tunbridge Wells, United Kingdom

Abstract

Purpose Anterior Cruciate Ligament (ACL) rupture is increasingly identified in children, but few prospective research studies reporting outcomes following specific surgical interventions in this population exist.  We report two-year outcomes of skeletally immature patients in the UK undergoing endoscopic ACL reconstruction using living donor hamstring tendon allograft (LDHA) from a parent.

Methods Twenty skeletally immature patients were recruited, from those routinely referred to the Knee injury clinics of Maidstone & Tunbridge Wells NHS Trust, with ACL rupture confirmed on MRI, and a parent willing to donate hamstring tendon.  For all cases a trans-physeal technique was used and Lateral extra-articular tenodesis performed.  Patient reported IKDC and Lysholm ratings were recorded and Long leg radiographs performed preoperatively and at 1 and 2 years post operatively.

Results The children’s mean age was 14 years & 6 months (range 10-17).  90% were male. In 60%, the cause of primary injury was football.  55% had associated meniscal tears (4 lateral, 4 medial, 3 medial & lateral) all were repaired at the time of surgery. Average graft size achieved from the parent donor was 8.4mm (± 0.6mm).

At 2-year post operation there were no incidence graft rupture, contralateral ACL rupture or growth disturbance. Mean IKDC scores and Lysholm scores were 96.1 and 96.8 respectively.

Conclusions In a prospective cohort of 20 children, 2 years post ACL reconstruction with LDHA, there were no graft ruptures and no incidence of growth arrest.  Excellent patient scores were reported. Good hamstring graft size was achieved from maternal and paternal donors.

Florence Shekleton, Ben Lankester

Yeovil District Hospital, Yeovil, United Kingdom

Abstract

Introduction

Monoblock all-polyethylene tibial (APT) components were used in the earliest total knee replacements (TKR) but were replaced by modular metal-backed designs after early failures.  With improvements in polyethylene processing, implant geometry, and surgical technique, contemporary APT designs have demonstrated excellent survivorship. However, their clinical use remains limited, accounting for only 1.3% of TKRs in the UK National Joint Registry (NJR).

Methodology

This study aimed to evaluate the utilisation, revision rate, and cost implications of the Triathlon® APT TKR (Stryker, Limerick, Ireland) performed in a single centre, comparing against metal-backed tibial component within the same period using NJR outcome data.

A retrospective review was conducted of all primary TKR procedures using APT components. Demographic data, revision rates, and Kaplan–Meier survival analyses were performed and compared against metal-backed tibial data. A cost analysis of APT versus modular metal-backed TKRs was undertaken.

Results

A total of 2155 TKRs performed between 2014-2025 were included. 35% utilised the APT. APT implants demonstrated improved survival rates compared to modular counterparts, with statistically significant reduction in total revisions (p: 0.001), aseptic loosening (p: 0.01) and infection (p: 0.041). Cost analysis revealed an estimated saving of over £50,000 across the primary TKR cohort over 10 years.

Conclusion

The findings, corroborated by multiple clinical studies and registry data, confirm the non-inferiority of APT components in the TKR. Their lower cost, reduced wear, and simplified revision process offer potential advantages. Given these benefits and equivalent clinical outcomes, the underutilisation of APT components in modern arthroplasty warrants reconsideration.

James Murray1, Andrew Porteous1, Max Ettinger2, Peter Savov2, Matteo Innocenti33, Gijs van Hellemondt4, Koen DeFoort4, Juan Carlos Martinez-Pastor5, Peter Bollars6, Francesco Zambianchi7, Alberto Belluati8, GIanluca Piovan9, Etienne Cavaignac10, Cat Whitall11

1Southmead Hospital, Bristol, United Kingdom. 22University Clinic For Orthopaedic surgery, Oldenburg, Germany. 33Centro Traumatologico Ortopedico di Firenze, Turin, Italy. 4Sint Marteenskliniek, Nijmegen, Netherlands. 5Hospital Clinic Barcelona, Barcelona, Spain. 66Sint-Trudo Ziekenhuis, Sint-Truiden, Belgium. 7Azienda Ospedaliero Universitaria di Modena, Modena, Italy. 8Ospedale S. Maria delle Croci, Ravenna, Italy. 9IRCCS Ospedale Sacro Cuore Don Calabria, Calabria, Italy. 10Hôpital Pierre Paul Riquet, Bordeaux, France. 11Smith and Nephew, Watford, United Kingdom

Abstract

Introduction: This prospective multicentre project evaluated the impact of handheld robotic‑assisted (RA) technology on operating room staff wellbeing during revision total knee arthroplasty (rTKA). The aim was to determine whether RA using the CORI™ Surgical System (Smith & Nephew, USA) reduced perceived workload compared with conventional manual instrumentation, measured by the NASA Task Load Index (NASA‑TLX) instrument.

Methodology: NASA‑TLX surveys were completed by surgeons, trainees, scrub nurses, and circulating nurses after rTKA procedures performed with either RA or manual instrumentation across six European centres. The tool assesses six domains—Mental, Physical, Temporal demand, Performance, Effort, and Frustration—to produce an Overall Workload Score (OWS; −50 best to +50 worst). In total, 212 complete responses were analysed (manual n=94; RA n=118), with a subgroup analysis of the lead surgeons only (manual n=39; RA n=61).

Results: The results demonstrated a significant reduction in the overall workload with RA compared with manual instrumentation (median overall workload score 0.0 vs 8.5 respectively, p=0.0003). Domain-specific analyses revealed significantly lower mental demand, physical demand, and effort in the RA group without compromising perceived surgical performance, temporal demand, or frustration. A subgroup analysis of the lead surgeons showed consistent workload reductions consistent with the overall findings

Conclusion: This evaluation provides compelling real-world evidence that the handheld RA surgical system significantly reduces intraoperative workload during rTKA, as measured by the NASA-TLX. Overall, these findings support the integration of RA as a strategy to enhance surgical comfort, reduce physical strain, and enhance safety in revision knee arthroplasty

Ben Tyas1,2, Catherine Hewitt1, David Deehan3,4, Paul Baker2,1

1University of York, York, United Kingdom. 2University Hospitals Tees, Middlesbrough, United Kingdom. 3Newcastle Hospitals NHS Foundation Trust, Newcastle, United Kingdom. 4Newcastle University, Newcastle Upon Tyne, United Kingdom

Abstract

Introduction:

Using data from the National Joint Registry for England, Wales, Northern Ireland and the Isle of Man, this study aimed to characterise temporal trends and factors associated with implant selection at first-time elective aseptic revision knee arthroplasty (rTKA).

Methodology:

Patients undergoing first-time elective aseptic rTKA between 2009-2023 were included. The outcome of interest was implant selection; this included constraint level, stem design, use of metaphyseal fixation devices and the use of fixed versus mobile bearings. Descriptive analyses and logistic regression were used to identify associations between patient and clinical characteristics, mean annual surgeon volume, and implant selection.

Results:

12,950 cases were analysed. The most common level of constraint used was condylar-constrained (53.9%). There was an increasing use of condylar-constrained and hinged prostheses over time. Revisions were most frequently undertaken using cementless stems (57.3%). The use of metaphyseal fixation increased from 8.5% of revisions in 2009 to 38.0% in 2021. This was accompanied by a decreasing use of long cementless stems. Low-volume surgeons were more likely to use unconstrained/posterior-stabilised implants (HR=1.445, p<0.001) and perform rTKA without intramedullary stems (HR=1.559, p<0.001). Revision for instability was associated with an increased use of hinged prostheses (HR=3.973, p<0.001).

Conclusion:

This study highlights temporal changes in implant selection in aseptic rTKA over the last 15 years. Surgeons have increasingly adopted higher-constraint designs, whilst fixation strategies are evolving to include the use of metaphyseal fixation devices. Further studies are required to determine whether these changes are associated with improved outcomes for patients.

Day 2 – Wednesday 15th April

Omar Javed1, Sabri Bleibleh1, Hammad Sadique1, Francisco Barbosa1, Tom Kurien1,2, Jimmy Ng1

1Trauma & Orthopaedics Department, Nottingham University Hospitals NHS Trust, Nottingham, United Kingdom. 2Academic Orthopaedics, Trauma and Sports Medicine, University of Nottingham, Nottingham, United Kingdom

Abstract

Introduction

Multiligament knee injury (MLKI) commonly affects young, active patients and is associated with significant morbidity. Debate exists regarding whether a single-stage or two-stage approach balances early rehabilitation and stabilisation against complications such as arthrofibrosis. This study assessed unplanned re-operation and outcomes following single-stage or two-stage MLKI surgery.

Methodology

This was a single-centre observational cohort study of patients undergoing MLKI surgery between January 2023 and January 2026. Staging strategy was at the treating surgeons’ discretion. The primary outcome was unplanned re-operation, excluding any planned second-stage surgery. Secondary outcomes included International Knee Documentation Committee (IKDC) score and range of motion (ROM).

Results

Fifty-seven patients were included (50 single-stage, 7 two-stage), with 41 males and 16 females (median 21 years, range 15-56). Primary outcome data were complete. Fifteen patients required ≥1 unplanned re-operation, with similar rates in the single-stage (13/50, 26%) and two-stage groups (2/7, 29%). Re-operations included surgery for arthrofibrosis (manipulation under anaesthesia [MUA] or arthroscopic release), secondary meniscal or graft-related pathology, infection, and implant-related symptoms. The most common indication for re-operation was arthrofibrosis. At 12 months, IKDC scores were available for 21 patients (median 58, IQR 50-74). At 6 months, ROM data were available for 25 patients, with 88% achieving full extension and 84%achieving flexion ≥110°. Patients with persistent ROM deficit underwent subsequent intervention.

Conclusion

Unplanned re-operation was common following both single-stage and two-stage MLKI surgery. Arthrofibrosis following persistent ROM deficit was the main indication, highlighting the importance of early identification and management through coordinated surgical and rehabilitation strategies.

Muaaz Tahir, Samuel Harris, Danielle Johnson, Adrian Davidescu, Nicholas Smith, Feisal Shah, Peter Thompson, Nicola MacKay, Tim Spalding

University Hospitals Coventry & Warwickshire, Coventry, United Kingdom

Abstract

Introduction
Meniscal allograft transplantation (MAT) is infrequently performed in the paediatric population, resulting in limited evidence regarding graft survivorship and clinical outcomes. While MAT is well established in adults for reducing pain and improving function, paediatric data remain scarce. This study aims to evaluate clinical and functional outcomes following MAT in children and adolescents.

Methodology
A prospectively maintained database comprising 480 patients was analysed retrospectively. Patients aged ≤18 years at the time of surgery with minimum 1-year follow-up were included. Demographic data, concomitant procedures, complications, reoperations, and patient-reported outcome measures (PROMs) were analysed at latest follow-up.

Results
A total of 53 patients were included (58% female), with median age of 16 years (range 8-18). Majority of procedures involved the lateral meniscus (81%), several patients underwent concomitant procedures including osteotomy (9%), ligament reconstruction (12%), or cartilage procedures (13%). Mean follow-up was 7.8±3.8 years. Significant improvements were observed across all PROMs, including Lysholm (57±19 to 78±18, p<0.001), Tegner (3±1 to 5±2, p<0.001), IKDC (44±16 to 68±19, p<0.001) and all KOOS subscales. Thirteen patients (23%) required further arthroscopic intervention. There were 3 graft failures (5.7%), and mean time to failure was 5.3±3 years. Other complications included superficial wound infection (4%), and deep-vein-thrombosis (2%).

Conclusion
Paediatric MAT demonstrates favourable mid- to long-term outcomes, with improved function and excellent graft survivorship. These findings support MAT as an effective treatment for the symptomatic meniscal-deficient paediatric knee. Early referral should be considered for children and adolescents with post-meniscectomy symptoms.

Neel Badhe1, Robert Kerslake2, Thomas Kurien1

1University of Nottingham, Nottingham, United Kingdom. 2Versus Arthritis Pain Centre, Nottingham, United Kingdom

Abstract

Introduction: Radiographic findings often fail to reflect knee osteoarthritis pain. This study aimed to evaluate associations between MRI-detected features of knee osteoarthritis (articular cartilage loss, bone marrow lesions (BMLs), and synovitis) and pain thresholds at the knee and distant sites, measured objectively by quantitative sensory testing (QST).

Methods: Forty-two knee osteoarthritis patients and 21 healthy controls completed Oxford Knee Score (OKS), Visual Analogue Scale (VAS) pain, Total Intermittent and Constant Osteoarthritis Pain (ICOAP), and PainDETECT questionnaires. Pressure pain thresholds (PPTs) at the affected knee, distal (tibialis anterior), and remote (extensor carpi radialis longus) sites were assessed using QST. MOAKS quantified MRI-detected articular cartilage loss, BMLs, and synovitis. Correlations between PPTs and MRI findings were evaluated.

Results: Knee OA patients exhibited significantly worse Patient-Reported Outcome Measures (PROMs) scores compared to healthy controls. Knee PPTs correlated with overall cartilage loss (r = -0.4210) and more strongly with medial cartilage loss (r = -0.5775). Tibialis anterior PPTs correlated with medial cartilage loss (r = -0.3728), while ECRL PPTs correlated with both overall (r = -0.3786) and medial cartilage loss (r = -0.4765). BMLs were associated with PPTs at the knee (r = -0.5075), tibialis anterior (r = -0.4299) and ECRL (r = -0.3716). No associations between PPTs and synovitis were observed.

Conclusion: Cartilage loss and BMLs in knee OA patients are associated with reduced PPTs locally and at distant sites, suggesting peripheral and central sensitisation. These associations may help identify patients at risk of heightened pain sensitivity and suboptimal outcomes following knee arthroplasty.

Shuntaro Nejima1,2, Jonathan Holt2, Richard van Arkel2, Andy Metcalfe3, Nick Smith3, Andrew Amis2

1Yokohama City University, Yokohama, Japan. 2Imperial College, London, United Kingdom. 3Warwick University, Coventry, United Kingdom

Abstract

Introduction: Posterior tibial slope (PTS) influences tibiofemoral kinematics and anterior cruciate ligament (ACL) function. However, the isolated roles the medial or lateral PTS remain unclear. This study aimed to evaluate the influence of medial and lateral PTS changes on anterior translation and internal rotation under axial loading.

Methods: Eight fresh-frozen knees were mounted in a compression loading machine. Tibiofemoral kinematics were measured at 500 N load at 0° and 20° flexion using optical tracking. Unicondylar slope-changing osteotomies were performed with a custom cutting guide, preserving key ligament and meniscal root attachments. Medial (M) and lateral (L) PTS were adjusted independently across -5° to +10° using 3D-printed wedges. Repeated-measures ANOVA, post-hoc t-tests and Pearson correlations were used.

Results: No significant difference was found between the intact and L0°M0° conditions, indicating no procedural effect. Increasing lateral PTS significantly increased anterior translation (r=0.966, P=0.0004), while increasing medial PTS did not (r=0.091, P=0.847 at 20°). Internal rotation was most influenced by delta PTS at both 0° and 20° flexion (r=0.908, P=0.0047 at 20°), altering lateral or medial PTS alone did not significantly change rotation. At 20° flexion the largest rotational differences were observed between wedge combinations with opposing delta PTS values: L5M−5 vs L0M10 (13.5°, P=0.003), L10M0 vs L0M10 (12.5°, P=0.006).

Conclusions:

Lateral PTS was the primary determinant of anterior translation, whereas delta PTS was the principal determinant of internal rotation under axial loading. These findings suggest the potential utility of unicondylar slope-changing osteotomy in ACL reconstruction with rotational instability.

Short Presentations (2 Minutes 2 Slides) – Day 2 – Wednesday 15th April

Chris Madden-McKee, Kerry Bowsie, Roslyn Cassidy, Richard Napier, Paul Karayiannis, David Beverland

Musgrave Park Hospital, Belfast, United Kingdom

Abstract

Introduction

It is widely perceived that morbid obesity is associated with unacceptable rates of complication after total knee arthroplasty (TKA), with many surgeons using a body mass index (BMI) of 40 as a cut-off. The aim of this study was to assess the outcomes of two patient groups undergoing cementless TKA in a high-volume unit: one with morbid obesity (BMI 40 to 44.9) and the other non-obese (BMI 20 to 29.9). 

Methodology

This was a retrospective single-centre comparative case series of 2,674 fully cementless TKAs. Of these, 568 were morbidly obese and 2,106 were non-obese. Median follow-up was 7.3 years (range, 1 to 13). Data was collected via electronic databases and radiology systems.

Results

All-cause revision rates were 0.7% (4 of 568) in the morbidly obese, and 0.9% (19 of 2,106) in the non-obese (P=0.801). There were no revisions for non-infective causes in the morbidly obese group. Other key outcomes, such as mortality, venous thromboembolism (VTE), and patient-reported outcomes scores (PROMs), were also comparable.

The reoperation rate for periprosthetic fracture (PPF) was significantly lower in the morbidly obese (1.1%, [6 of 568]) compared to the non-obese (3.6% [76 of 2,106], P=0.002). Reoperation rate, excluding PPF, was also significantly lower in the morbidly obese (1.1% [6 of 568]) compared to the non-obese (3.0% [63 of 2,106], P=0.010).

Conclusions

Cementless TKA can be performed safely, and with a significantly lower reoperation rate, in the morbidly obese with BMI up to at least 45, as compared to the non-obese, in a high-volume environment.

Rudi Carrino, Sophie Stewart, Audrey Kemayou, Luke Farrow

Aberdeen Royal Infirmary, Aberdeen, United Kingdom

Abstract

Despite growing utilisation of Robotic arm-assisted total knee arthroplasty (RA-TKA), there remains a limited understanding of how individual patient anatomy and the resulting intraoperative decisions influence patient-reported outcome measures (PROMs).

A retrospective cohort study was undertaken with relevant perioperative data collected from patients undergoing RA-TKA at a large University teaching hospital between 2022-2024. Outcomes were 6 week and 1-year Oxford Knee Scores (OKS) Minimum Clinically Important Difference (MCID) attainment. Multivariable logistic regression (adjusting for age, sex, baseline OKS, and pre-operative Coronal Plane Alignment of the Knee (CPAK) classification) was performed to assess the relationship between OKS MCID attainment and: (1) Change in CPAK classification (2) Implant positioning outside industry defined radiological parameters, and (3) Degree of coronal deformity correction. Statistical significance was set at p≤0.05.

A total of 374 patients (mean age 72, 56% female) were included, of which 229 and 142 had 6 week and 1-year OKS available respectively. 82% patients changed CPAK classification post-op, with no significant association with 6 week / 1-year MCID attainment. Femoral or tibial coronal alignment outside the recommended parameters was however associated with significantly greater 1-year MCID attainment (OR 4.57 (1.09-24.35, p – 0.05). Every 1 degree of alignment correction towards neutral was also associated with 3% greater chance of 1-year PROMs MCID attainment (Coefficient 0.03 0.01 to 0.06, p = 0.012).

Patient anatomy and implant positioning do appear to influence PROMs following RA-TKA. Further work is required to understand optimal implant placement for specific anatomical knee phenotypes at an individual patient level.

Babar Kayani, Andreas Fontalis, Jenni Tahmassebi, Sujith Konan, Ricci Plastow, Usman Wazir, Munuzza Shah, Sam Oussedik, Fares S. Haddad

Department of Trauma and Orthopaedic Surgery, University College Hospital, 235 Euston Road, Bloomsbury, London, NW1 2BU, London, United Kingdom

Abstract

Introduction:
 This study reports interim outcomes from a randomised controlled trial comparing component positioning accuracy, early recovery, patient-reported outcomes, and complications between conventional jig-based unicompartmental knee arthroplasty (CO UKA) and robotic-arm assisted unicompartmental knee arthroplasty (RO UKA) in patients with medial compartment knee osteoarthritis.

Methodology:
 A total of 107 patients with symptomatic medial compartment knee osteoarthritis were prospectively randomised to undergo CO UKA (n=52) or RO UKA (n=55). All patients underwent CT-based preoperative planning, procedures were performed via a medial parapatellar approach and a standardised postoperative rehabilitation protocol was applied. Predefined outcomes were collected at regular intervals up to two years postoperatively.

Results:
 RO UKA demonstrated significantly greater accuracy in achieving planned femoral (p<0.001) and tibial (p<0.001) component positions compared with CO UKA. RO UKA was associated with lower inpatient pain scores (p<0.001), reduced opioid consumption (p=0.008) and shorter time to hospital discharge (p=0.004). There were no significant differences between groups in the Oxford Knee Score (p=0.299), Knee Injury and Osteoarthritis Outcome Score (p=0.261) or Western Ontario and McMaster Universities Arthritis Index (p=0.281) at two-year follow-up. RO UKA was associated with higher Forgotten Joint Scores at six months (p=0.002) and two years (p=0.021). Recruitment declined during the trial due to patient preference for RO UKA and loss of surgeon equipoise.

Conclusion:
 Interim results demonstrate that RO UKA improves component positioning accuracy, enhances early recovery and reduces length of hospital stay compared with CO UKA, with superior joint awareness outcomes at short-term follow-up.

Florence Judge-Clayden1, Ben Tyas1,2, Jonathan Murray3,4, Paul Baker1,2,4

  1. 1 Trauma and Orthopaedics, University Hospitals Tees, Middlesborough, United Kingdom. 2University of York, York, United Kingdom. 33. Renal Unit, University Hospitals Tees, Middlesborough, United Kingdom. 4School of Health & Life Sciences, Teesside University, Middlesborough, United Kingdom

Abstract

Introduction:

This study compared patient and implant survival, and the incidence of 90-day medical complications, between patients with prior solid-organ transplant (SOT) and those without solid-organ transplant (Non-SOT) following primary total knee arthroplasty (TKA).

Methods:

Data were extracted from TriNetX, a global healthcare database covering 21 countries. Patients undergoing TKA between 2005-2025 were first classified as SOT or non-SOT patients, and SOT patients further categorised by transplant type. Descriptive analyses reported incidence of prosthetic joint infection (PJI), patient and implant survival at 1, 2 and 5 years, and 90-day medical complications, stratified by SOT status and transplant type. Propensity score-matched analysis compared outcomes between SOT and non-SOT cohorts. Cox regression identified covariate-adjusted relative risk of PJI at two years between SOT and non-SOT patient groups.

Results:

In total 671,661 patients were included with a median follow-up of 4.7 years. Patients in the SOT group (n=2802) had significantly inferior 5-year patient survivorship (87.1% vs 96.1%, p<0.001). Liver transplant patients had the highest incidence (4.8%) and hazard of PJI at 2-years compared to non-SOT group (HR 2.09,   p<0.001). After matching, there was no difference in 5-year implant survival (HR= 1.25, p= 0.146). Solid-organ transplant patients had an increased risk of nearly all 90-day medical complications, most notably acute kidney injury (RR=6.68, p<0.001).

Conclusion:

While peri-operative risks following TKA were higher for SOT patients, 5-year implant survival was equivalent. Solid-organ transplant status alone should not be a contraindication to TKA. Patients should be appropriately optimised pre-operatively through a multidisciplinary approach.

Conor Hennessy1, Simon Abram2

1Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford,, Oxford, United Kingdom. 2Musculoskeletal Research Unit, Translational Health Sciences, University of Bristol,, Bristol, United Kingdom

Abstract

Objectives:

Knee osteotomy is an established joint-preserving surgical procedure. Despite this, there has been little data to inform our understanding of subsequent rates of conversion to knee arthroplasty. We sought to use the National Health Service (NHS), England UK, procedure registry database to examine the arthroplasty conversion rate to a maximum of 25-years following knee osteotomy.

Methods:

Using NHS Hospital Episode Statistics (HES) data between 01/04/98 to 31/03/2023 we identified isolated knee osteotomies in patients between 16 to 80 years of age. Only the first osteotomy per patient, per side was included; bilateral cases within <6 months were excluded. Conversion to arthroplasty was determined using OPCS-4 codes linked by recorded laterality; the primary outcome was arthroplasty free survival.

Results:

24,947 knee osteotomies (18,871 HTO, 4,490 DFO, 1,586 doubles; 23,371 patients) were included (mean age: 40.83 years; 62.12% male; Charlson index 0: 80.42%). Overall arthroplasty free survival was 93.45% (95% CI 93.1-93.8) at 5 years, 84.9% (95% CI 84.3-85.4) at 10 years, and 67.95% (95% CI 66.1-69.8) at 25 years. Female sex and older age were predictive of higher conversion rates,  with female patients aged 40-59 showing highest rate of conversion to arthroplasty at 5 years (88.7%; 95% CI 87.5-89.7), 10 years (74.2%; 95% CI 72.2-76.1%) and 25 years (52.5%; 95% CI 48.2-56.5%).

Conclusions:

Including the whole cohort, the observed rate of conversion to arthroplasty in this study was roughly 1.5% per year. Higher rates of conversion were seen in the 40-59 age group, particularly in female patients.