Aims. It is increasingly appreciated that coordinated regulation of angiogenesis and osteogenesis is needed for bone formation. How this regulation is achieved during peri-implant bone healing, such as osseointegration, is largely unclear. This study examined the relationship between angiogenesis and osteogenesis in a unique model of osseointegration of a mouse tibial implant by pharmacologically blocking the vascular endothelial growth factor (VEGF) pathway. Materials and Methods. An implant was inserted into the right tibia of 16-week-old female C57BL/6 mice (n = 38). Mice received anti-VEGF receptor-1 (VEGFR-1) antibody (25 mg/kg) and VEGF receptor-2 (VEGFR-2) antibody (25 mg/kg; n = 19) or an isotype control antibody (n = 19). Flow cytometric (n = 4/group) and immunofluorescent (n = 3/group) analyses were performed at two weeks post-implantation to detect the distribution and density of CD31. hi. EMCN. hi. endothelium. RNA sequencing analysis was performed using sorted CD31. hi. EMCN. hi. endothelial cells (n = 2/group). Osteoblast lineage cells expressing osterix (OSX) and osteopontin (OPN) were also detected with immunofluorescence. Mechanical pull-out testing (n = 12/group) was used at four weeks post-implantation to determine the strength of the bone-implant interface. After pull-out testing, the tissue attached to the implant surface was harvested. Whole mount immunofluorescent staining of OSX and OPN was performed to determine the amount of osteoblast lineage cells. Results. Flow cytometry revealed that anti-VEGFR treatment decreased CD31. hi. EMCN. hi. vascular endothelium in the peri-implant bone versus controls at two weeks post-implantation. This was confirmed by the decrease of CD31 and endomucin (EMCN) double-positive cells detected with immunofluorescence. In addition, treated mice had more OPN-positive cells in both peri-implant bone and tissue on the implant surface at two weeks and four weeks, respectively. More OSX-positive cells were present in peri-implant bone at two weeks. More importantly, anti-VEGFR treatment decreased the maximum load of pull-out testing compared with the control. Conclusion. VEGF pathway controls the coupling of angiogenesis and osteogenesis in orthopaedic implant osseointegration by affecting the formation of CD31. hi. EMCN. hi. endothelium. Cite this article: Bone Joint J 2019;101-B(7 Supple C):108–114

1. The routes by which adult human articular cartilage can receive its nutrition is still a subject of controversy. 2. Microscopic examination of normal adult human femoral heads has revealed vascular channels which penetrate the subchondral plate and calcified cartilage. These channels bring the medullary soft tissue into contact with the articular cartilage. 3. A fluorescent dye migration technique was used to show that the observed vascular channels are pathways for dye from the medullary cavity to the articular cartilage. It is suggested that these pathways could also be routes by which articular cartilage receives part of its nutrition. 4. The nutritional mechanism in the mature rabbit and adult human femoral heads cannot be compared because histological studies revealed differences in their subchondral structures

Aims . To investigate whether pre-operative functional mobility is a determinant of delayed inpatient recovery of activities (IRoA) after total knee arthroplasty (TKA) in three periods that coincided with changes in the clinical pathway. Patients and Methods. All patients (n = 682, 73% women, mean age 70 years, standard deviation 9) scheduled for TKA between 2009 and 2015 were pre-operatively screened for functional mobility by the Timed-up-and-Go test (TUG) and De Morton mobility index (DEMMI). The cut-off point for delayed IRoA was set on the day that 70% of the patients were recovered, according to the Modified Iowa Levels of Assistance Scale (mILAS) (a 5-item activity scale). In a multivariable logistic regression analysis, we added either the TUG or the DEMMI to a reference model including established determinants. Results. Both the TUG (Odds Ratio (OR) 1.10 per second, 95% confidence intervals (CI) 1.06 to 1.15) and the DEMMI (OR 0.96 per point on the 100-point scale, 95% CI 0.95 to 0.98) were statistically significant determinants of delayed IRoA in a model that also included age, BMI, ASA score and ISAR score. These associations did not depend on the time period during which the TKA took place, as assessed by tests for interaction. . Conclusion. Functional mobility, as assessed pre-operatively by the TUG and DEMMI, is an independent and stable determinant of delayed inpatient recovery of activities after TKA. Future research, focusing on improvement of pre-operative functional mobility through tailored physiotherapy intervention, should indicate whether such intervention enhances post-operative recovery among high-risk patients. Cite this article: Bone Joint J 2017;99-B:211–17

We obtained samples of spinal accessory nerve from patients undergoing radical surgery for tumours or nerve grafting in the neck. These were analysed by light and electron microscopy for the type of fibre. All contained fibres consistent with non-proprioceptive sensory function including pain.

We performed a prospective study to examine the influence of the patient’s position on the location of the abdominal organs, to investigate the possibility of a true lateral approach for transforaminal endoscopic lumbar discectomy. Pre-operative abdominal CT scans were taken in 20 patients who underwent endoscopic lumbar discectomy. Axial images in parallel planes of each intervertebral disc from L1 to L5 were achieved in both supine and prone positions. The most horizontal approach angles possible to avoid injury to the abdominal organs were measured. The results demonstrated that the safe approach angles were significantly less (i.e., more horizontal) in the prone than in the supine position. Obstacles to a more lateral approach were mainly the liver, the spleen and the kidneys at L1/2 (39 of 40, 97.5%) and L2/3 (28 of 40, 70.0%), and the intestines at L3/4 (33 of 40, 82.5%) and L4/5 (30 of 30, 100%). A true lateral approach from each side was possible for 30 of the 40 discs at L3/4 (75%) and 23 of the 30 discs at L4/5 (76.7%). We concluded that a more horizontal approach for transforaminal endoscopic lumbar discectomy is possible in the prone position but not in the supine. Prone abdominal CT is more helpful in determining the trajectory of the endoscope. While a true lateral approach is feasible in many patients, our study shows it is not universally applicable.

Aims. Our aim was to estimate the total costs of all hospitalizations for treating periprosthetic joint infection (PJI) by main management strategy within 24 months post-diagnosis using activity-based costing. Additionally, we investigated the influence of individual PJI treatment pathways on hospital costs within the first 24 months. Methods. Using admission and procedure data from a prospective observational cohort in Australia and New Zealand, Australian Refined Diagnosis Related Groups were assigned to each admitted patient episode of care for activity-based costing estimates of 273 hip PJI patients and 377 knee PJI patients. Costs were aggregated at 24 months post-diagnosis, and are presented in Australian dollars. Results. The mean cost per hip and knee PJI patient was $64,585 (SD $53,550). Single-stage revision mean costs were $67,029 (SD $47,116) and $80,063 (SD $42,438) for hip and knee, respectively. Two-stage revision costs were $113,226 (SD $66,724) and $122,425 (SD $60,874) for hip and knee, respectively. Debridement, antibiotics, and implant retention in hips and knees mean costs were $53,537 (SD$ 39,342) and $48,463 (SD $33,179), respectively. Suppressive antibiotic therapy without surgical management mean costs were $20,296 (SD $8,875) for hip patients and $16,481 (SD $6,712) for knee patients. Hip patients had 16 different treatment pathways and knee patients had 18 treatment pathways. Additional treatment, episodes of care, and length of stay contributed to substantially increased costs up to a maximum of $369,948. Conclusion. Treating PJI incurs a substantial cost burden, which is substantially influenced by management strategy. With an annual PJI incidence of 3,900, the cost burden would be in excess of $250 million to the Australian healthcare system. Treatment pathways with additional surgery, more episodes of care, and a longer length of stay substantially increase the associated hospital costs. Prospectively monitoring individual patient treatment pathways beyond initial management is important when quantifying PJI treatment cost. Our study highlights the importance of optimizing initial surgical treatment, and informs treating hospitals of the resources required to provide care for PJI patients. Cite this article: Bone Joint J 2024;106-B(10):1084–1092

Aims. The aim of this study was to describe the introduction of a virtual pathway for the management of patients with a suspected fracture of the scaphoid, and to report patient-reported outcome measures (PROMs) and satisfaction following treatment using this service. Methods. All adult patients who presented with a clinically suspected scaphoid fracture that was not visible on radiographs at the time of presentation during a one-year period were eligible for inclusion in the pathway. Demographic details, findings on examination, and routine four-view radiographs at the time of presentation were collected. All radiographs were reviewed virtually by a single consultant hand surgeon, with patient-initiated follow-up on request. PROMs were assessed at a minimum of one year after presentation and included the abbreviated version of the Disabilities of the Arm, Shoulder and Hand Score (QuickDASH), the EuroQol five-dimension five-level health questionnaire (EQ-5D-5L), the Net Promoter Score (NPS), and return to work. Results. A total of 221 patients were referred to the virtual pathway. Their mean age was 41 years (range 16 to 87) and there were 99 male patients (45%). A total of 189 patients (86%) were discharged with advice and 19 (9%) were recalled for clinical review: seven with an undisplaced scaphoid fracture, six with another fracture of the hand or wrist, two with a scapholunate ligament injury, and four in whom no abnormality was detected. A total of 13 patients (6%) initiated follow-up with the hand service: no fracture or ligament injury was identified in this group. PROMs were available for 179 patients (81%) at a mean follow-up of 19 months (range 13 to 33). The median QuickDASH score was 2.3 (interquartile range (IQR) 0 to 15.9), the median EQ-5D-5L was 0.85 (IQR 0.73 to 1.00), the NPS was 76, and 173 patients (97%) were satisfied with their treatment. There were no documented cases of symptomatic nonunion one year following injury. Conclusion. We describe the introduction of a virtual pathway for the management of patients with a suspected scaphoid fracture. We found high levels of patient satisfaction, excellent PROMs, and no detrimental effects in the vast majority of cases. Cite this article: Bone Joint J 2022;104-B(6):709–714

Aims. Weightbearing instructions after musculoskeletal injury or orthopaedic surgery are a key aspect of the rehabilitation pathway and prescription. The terminology used to describe the weightbearing status of the patient is variable; many different terms are used, and there is recognition and evidence that the lack of standardized terminology contributes to confusion in practice. Methods. A consensus exercise was conducted involving all the major stakeholders in the patient journey for those with musculoskeletal injury. The consensus exercise primary aim was to seek agreement on a standardized set of terminology for weightbearing instructions. Results. A pre-meeting questionnaire was conducted. The one-day consensus meeting, including patient representatives, identified three agreed terms only to be used in defining the weightbearing status of the patient: 1) non-weightbearing; 2) limited weightbearing; and 3) unrestricted weightbearing. Conclusion. This study represents the first and only exercise in standardizing rehabilitation terminology in orthopaedics, as agreed by all major stakeholders in the patient pathway and the patients themselves. The standardization of language allows for higher-quality and more accurate research to be conducted, and is one small part of the bigger picture in increasing the mobility of patients after orthopaedic injury or surgery. Cite this article: Bone Joint J 2024;106-B(9):1016–1020

In the UK, multidisciplinary teamwork for patients with hip fracture has been shown to reduce mortality and improves health-related quality of life for patients, while also reducing hospital bed days and associated healthcare costs. However, despite rapidly increasing numbers of fragility fractures, multidisciplinary shared care is rare in low- and middle-income countries around the world. The HIPCARE trial will test the introduction of multidisciplinary care pathways in five low- and middle-income countries in South and Southeast Asia, with the aim to improve patients’ quality of life and reduce healthcare costs. Cite this article: Bone Joint J 2024;106-B(12):1369–1371

Aims. A national screening programme has existed in the UK for the diagnosis of developmental dysplasia of the hip (DDH) since 1969. However, every aspect of screening and treatment remains controversial. Screening programmes throughout the world vary enormously, and in the UK there is significant variation in screening practice and treatment pathways. We report the results of an attempt by the British Society for Children’s Orthopaedic Surgery (BSCOS) to identify a nationwide consensus for the management of DDH in order to unify treatment and suggest an approach for screening. Methods. A Delphi consensus study was performed among the membership of BSCOS. Statements were generated by a steering group regarding aspects of the management of DDH in children aged under three months, namely screening and surveillance (15 questions), the technique of ultrasound scanning (eight questions), the initiation of treatment (19 questions), care during treatment with a splint (ten questions), and on quality, governance, and research (eight questions). A two-round Delphi process was used and a consensus document was produced at the final meeting of the steering group. Results. A total of 60 statements were graded by 128 clinicians in the first round and 132 in the second round. Consensus was reached on 30 out of 60 statements in the first round and an additional 12 in the seond. This was summarized in a consensus statement and distilled into a flowchart to guide clinical practice. Conclusion. We identified agreement in an area of medicine that has a long history of controversy and varied practice. None of the areas of consensus are based on high-quality evidence. This document is thus a framework to guide clinical practice and on which high-quality clinical trials can be developed. Cite this article: Bone Joint J 2023;105-B(2):209–214

Aims. The aim of this study was to determine the consensus best practice approach for the investigation and management of children (aged 0 to 15 years) in the UK with musculoskeletal infection (including septic arthritis, osteomyelitis, pyomyositis, tenosynovitis, fasciitis, and discitis). This consensus can then be used to ensure consistent, safe care for children in UK hospitals and those elsewhere with similar healthcare systems. Methods. A Delphi approach was used to determine consensus in three core aspects of care: 1) assessment, investigation, and diagnosis; 2) treatment; and 3) service, pathways, and networks. A steering group of paediatric orthopaedic surgeons created statements which were then evaluated through a two-round Delphi survey sent to all members of the British Society for Children’s Orthopaedic Surgery (BSCOS). Statements were only included (‘consensus in’) in the final agreed consensus if at least 75% of respondents scored the statement as critical for inclusion. Statements were discarded (‘consensus out’) if at least 75% of respondents scored them as not important for inclusion. Reporting these results followed the Appraisal Guidelines for Research and Evaluation. Results. A total of 133 children’s orthopaedic surgeons completed the first survey, and 109 the second. Out of 43 proposed statements included in the initial Delphi, 32 reached ‘consensus in’, 0 ‘consensus out’, and 11 ‘no consensus’. These 11 statements were then reworded, amalgamated, or deleted before the second Delphi round of eight statements. All eight were accepted as ‘consensus in’, resulting in a total of 40 approved statements. Conclusion. In the many aspects of medicine where relevant evidence is not available for clinicians to base their practice, a Delphi consensus can provide a strong body of opinion that acts as a benchmark for good quality clinical care. We would recommend clinicians managing children with musculoskeletal infection follow the guidance in the consensus statements in this article, to ensure care in all medical settings is consistent and safe. Cite this article: Bone Joint J 2023;105-B(7):815–820

Aims. Historically, patients undergoing surgery for adolescent idiopathic scoliosis (AIS) have been nursed postoperatively in a critical care (CC) setting because of the challenges posed by prone positioning, extensive exposures, prolonged operating times, significant blood loss, major intraoperative fluid shifts, cardiopulmonary complications, and difficulty in postoperative pain management. The primary aim of this paper was to determine whether a scoring system, which uses Cobb angle, forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and number of levels to be fused, is a valid method of predicting the need for postoperative critical care in AIS patients who are to undergo scoliosis correction with posterior spinal fusion (PSF). Methods. We retrospectively reviewed all AIS patients who had undergone PSF between January 2018 and January 2020 in a specialist tertiary spinal referral centre. All patients were assessed preoperatively in an anaesthetic clinic. Postoperative care was defined as ward-based (WB) or critical care (CC), based on the preoperative FEV1, FVC, major curve Cobb angle, and the planned number of instrumented levels. Results. Overall, 105 patients were enrolled. Their mean age was 15.5 years (11 to 25) with a mean weight of 55 kg (35 to 103). The mean Cobb angle was 68° (38° to 122°). Of these, 38 patients were preoperatively scored to receive postoperative CC. However, only 19% of the cohort (20/105) actually needed CC-level support. Based on these figures, and an average paediatric intensive care unit stay of one day before stepdown to ward-based care, the potential cost-saving on the first postoperative night for this cohort was over £20,000. There was no statistically significant difference between the Total Pathway Score (TPS), the numerical representation of the four factors being assessed, and the actual level of care received (p = 0.052) or the American Society of Anesthesiologists grade (p = 0.187). Binary logistic regression analysis of the TPS variables showed that the preoperative Cobb angle was the only variable which significantly predicted the need for critical care. Conclusion. Most patients undergoing posterior fusion surgery for AIS do not need critical care. Of the readily available preoperative measures, the Cobb angle is the only predictor of the need for higher levels of care, and has a threshold value of 74.5°. Cite this article: Bone Joint J 2024;106-B(7):713–719

Aims. There is a lack of high-quality research investigating outcomes of Ponseti-treated idiopathic clubfeet and correlation with relapse. This study assessed clinical and quality of life (QoL) outcomes using a standardized core outcome set (COS), comparing children with and without relapse. Methods. A total of 11 international centres participated in this institutional review board-approved observational study. Data including demographics, information regarding presentation, treatment, and details of subsequent relapse and management were collected between 1 June 2022 and 30 June 2023 from consecutive clinic patients who had a minimum five-year follow-up. The clubfoot COS incorporating 31 parameters was used. A regression model assessed relationships between baseline variables and outcomes (clinical/QoL). Results. Overall, 293 patients (432 feet) with a median age of 89 months (interquartile range 72 to 113) were included. The relapse rate was 37%, with repeated relapse in 14%. Treatment considered a standard part of the Ponseti journey (recasting, repeat tenotomy, and tibialis anterior tendon transfer) was performed in 35% of cases, with soft-tissue release and osteotomies in 5% and 2% of cases, respectively. Predictors of relapse included duration of follow-up, higher initial Pirani score, and poor Evertor muscle activity. Relapse was associated with poorer outcomes. Conclusion. This is the first multicentre study using a standardized COS following clubfoot treatment. It distinguishes patients with and without relapse in terms of clinical outcomes and QoL, with poorer outcomes in the relapse group. This tool allows comparison of treatment methods and outcomes, facilitates information sharing, and sets family expectations. Predictors of relapse encourage us to create appropriate treatment pathways to reduce relapse and improve outcome. Cite this article: Bone Joint J 2024;106-B(7):735–743

Aims. Postoperative length of stay (LOS) and discharge dispositions following arthroplasty can be used as surrogate measurements for improvements in patients’ pathways and costs. With the increasing use of robotic technology in arthroplasty, it is important to assess its impact on LOS. The aim of this study was to identify factors associated with decreased LOS following robotic arm-assisted total hip arthroplasty (RO THA) compared with the conventional technique (CO THA). Methods. This large-scale, single-institution study included 1,607 patients of any age who underwent 1,732 primary THAs for any indication between May 2019 and January 2023. The data which were collected included the demographics of the patients, LOS, type of anaesthetic, the need for treatment in a post-anaesthesia care unit (PACU), readmission within 30 days, and discharge disposition. Univariate and multivariate logistic regression models were used to identify factors and the characteristics of patients which were associated with delayed discharge. Results. The multivariate model identified that age, female sex, admission into a PACU, American Society of Anesthesiologists grade > II, and CO THA were associated with a significantly higher risk of a LOS of > two days. The median LOS was 54 hours (interquartile range (IQR) 34 to 78) in the RO THA group compared with 60 hours (IQR 51 to 100) in the CO THA group (p < 0.001). The discharge dispositions were comparable between the two groups. A higher proportion of patients undergoing CO THA required PACU admission postoperatively, although without reaching statistical significance (7.2% vs 5.2%, p = 0.238). Conclusion. We found that among other baseline characteristics and comorbidities, RO THA was associated with a significantly shorter LOS, with no difference in discharge destination. With the increasing demand for THA, these findings suggest that robotic assistance in THA could reduce costs. However, randomized controlled trials are required to investigate the cost-effectiveness of this technology. Cite this article: Bone Joint J 2024;106-B(3 Supple A):24–30

We aim to explore the potential technologies for monitoring and assessment of patients undergoing arthroplasty by examining selected literature focusing on the technology currently available and reflecting on possible future development and application. The reviewed literature indicates a large variety of different hardware and software, widely available and used in a limited manner, to assess patients’ performance. There are extensive opportunities to enhance and integrate the systems which are already in existence to develop patient-specific pathways for rehabilitation. Cite this article: Bone Joint J 2022;104-B(10):1104–1109

Aims. Approximately 10% to 20% of knee arthroplasty patients are not satisfied with the result, while a clear indication for revision surgery might not be present. Therapeutic options for these patients, who often lack adequate quadriceps strength, are limited. Therefore, the primary aim of this study was to evaluate the clinical effect of a novel rehabilitation protocol that combines low-load resistance training (LL-RT) with blood flow restriction (BFR). Methods. Between May 2022 and March 2024, we enrolled 45 dissatisfied knee arthroplasty patients who lacked any clear indication for revision to this prospective cohort study. All patients were at least six months post-surgery and had undergone conventional physiotherapy previously. The patients participated in a supervised LL-RT combined with BFR in 18 sessions. Primary assessments included the following patient-reported outcome measures (PROMs): Knee injury and Osteoarthritis Outcome Score (KOOS); Knee Society Score: satisfaction (KSSs); the EuroQol five-dimension five-level questionnaire (EQ-5D-5L); and the pain catastrophizing scale (PCS). Functionality was assessed using the six-minute walk Test (6MWT) and the 30-second chair stand test (30CST). Follow-up timepoints were at baseline, six weeks, three months, and six months after the start. Results. Six weeks of BFR with LL-RT improved all the PROMs except the sports subscale of the KOOS compared to baseline. Highest improvements after six weeks were found for quality of life (QoL) (mean 28.2 (SD 17.2) vs 19 (SD 14.7); p = 0.002), activities of daily living (mean 54.7 (SD 18.7) vs 42.9 (SD 17.3); p < 0.001), and KSSs (mean 17.1 (SD 8.8) vs 12.8 (SD 6.7); p < 0.001). PROMs improvements continued to be present at three-month and six-month follow-up compared to baseline. However, no significant differences were observed in the paired comparisons of the six-week, three-month, and six-month follow-up. The same trends are observed for the 6MWT and 30CST. Conclusion. The reported regime demonstrates improved QoL and function of dissatisfied knee arthroplasty patients. In light of this, the pathway described may provide a valuable and safe treatment option for dissatisfied knee arthroplasty patients for whom therapeutic options are limited. Cite this article: Bone Joint J 2024;106-B(12):1416–1425

Aims. To examine whether natural language processing (NLP) using a clinically based large language model (LLM) could be used to predict patient selection for total hip or total knee arthroplasty (THA/TKA) from routinely available free-text radiology reports. Methods. Data pre-processing and analyses were conducted according to the Artificial intelligence to Revolutionize the patient Care pathway in Hip and knEe aRthroplastY (ARCHERY) project protocol. This included use of de-identified Scottish regional clinical data of patients referred for consideration of THA/TKA, held in a secure data environment designed for artificial intelligence (AI) inference. Only preoperative radiology reports were included. NLP algorithms were based on the freely available GatorTron model, a LLM trained on over 82 billion words of de-identified clinical text. Two inference tasks were performed: assessment after model-fine tuning (50 Epochs and three cycles of k-fold cross validation), and external validation. Results. For THA, there were 5,558 patient radiology reports included, of which 4,137 were used for model training and testing, and 1,421 for external validation. Following training, model performance demonstrated average (mean across three folds) accuracy, F1 score, and area under the receiver operating curve (AUROC) values of 0.850 (95% confidence interval (CI) 0.833 to 0.867), 0.813 (95% CI 0.785 to 0.841), and 0.847 (95% CI 0.822 to 0.872), respectively. For TKA, 7,457 patient radiology reports were included, with 3,478 used for model training and testing, and 3,152 for external validation. Performance metrics included accuracy, F1 score, and AUROC values of 0.757 (95% CI 0.702 to 0.811), 0.543 (95% CI 0.479 to 0.607), and 0.717 (95% CI 0.657 to 0.778) respectively. There was a notable deterioration in performance on external validation in both cohorts. Conclusion. The use of routinely available preoperative radiology reports provides promising potential to help screen suitable candidates for THA, but not for TKA. The external validation results demonstrate the importance of further model testing and training when confronted with new clinical cohorts. Cite this article: Bone Joint J 2024;106-B(7):688–695

Aims. Machine learning (ML), a branch of artificial intelligence that uses algorithms to learn from data and make predictions, offers a pathway towards more personalized and tailored surgical treatments. This approach is particularly relevant to prevalent joint diseases such as osteoarthritis (OA). In contrast to end-stage disease, where joint arthroplasty provides excellent results, early stages of OA currently lack effective therapies to halt or reverse progression. Accurate prediction of OA progression is crucial if timely interventions are to be developed, to enhance patient care and optimize the design of clinical trials. Methods. A systematic review was conducted in accordance with PRISMA guidelines. We searched MEDLINE and Embase on 5 May 2024 for studies utilizing ML to predict OA progression. Titles and abstracts were independently screened, followed by full-text reviews for studies that met the eligibility criteria. Key information was extracted and synthesized for analysis, including types of data (such as clinical, radiological, or biochemical), definitions of OA progression, ML algorithms, validation methods, and outcome measures. Results. Out of 1,160 studies initially identified, 39 were included. Most studies (85%) were published between 2020 and 2024, with 82% using publicly available datasets, primarily the Osteoarthritis Initiative. ML methods were predominantly supervised, with significant variability in the definitions of OA progression: most studies focused on structural changes (59%), while fewer addressed pain progression or both. Deep learning was used in 44% of studies, while automated ML was used in 5%. There was a lack of standardization in evaluation metrics and limited external validation. Interpretability was explored in 54% of studies, primarily using SHapley Additive exPlanations. Conclusion. Our systematic review demonstrates the feasibility of ML models in predicting OA progression, but also uncovers critical limitations that currently restrict their clinical applicability. Future priorities should include diversifying data sources, standardizing outcome measures, enforcing rigorous validation, and integrating more sophisticated algorithms. This paradigm shift from predictive modelling to actionable clinical tools has the potential to transform patient care and disease management in orthopaedic practice. Cite this article: Bone Joint J 2024;106-B(11):1216–1222

Aims. Knee osteoarthritis (OA) is characterized by a chronic inflammatory process involving multiple cytokine pathways, leading to articular cartilage degeneration. Intra-articular therapies using pharmaceutical or autologous anti-inflammatory factors offer potential non-surgical treatment options. Autologous protein solution (APS) is one such product that uses the patient’s blood to produce a concentrate of cells and anti-inflammatory cytokines. This study evaluated the effect of a specific APS intra-articular injection (nSTRIDE) on patient-reported outcome measures compared to saline in moderate knee OA. Methods. A parallel, double-blinded, placebo-controlled randomized controlled trial was conducted, where patients with unilateral moderate knee OA (Kellgren-Lawrence grade 2 or 3) received either nSTRIDE or saline (placebo) injection to their symptomatic knee. The primary outcome was the difference in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) total score at 12 months post-intervention. Secondary outcomes included WOMAC component scores, Knee injury and Osteoarthritis Outcome Score (KOOS), and visual analogue scale (VAS) scores at all follow-up timepoints (three, six, and 12 months). Results. A total of 40 patients were analyzed (21 nSTRIDE; 19 saline) in the study. No significant difference was found between nSTRIDE and saline groups for WOMAC total score at 12 months (mean difference -10.4 (95% CI -24.4 to 3.6; p = 0.141). There were no significant differences in WOMAC or KOOS scores across all timepoints. VAS scores favoured the saline group for both rest and worst pain scales at 12 months post-injection (mean difference (worst) 12 months 21.5 (95% CI 6.2 to 36.8; p = 0.008); mean difference (rest) 12 months 17.8 (95% CI 2.2 to 33.4; p = 0.026)). There were no adverse events recorded in either study group. Conclusion. Our study demonstrates no significant differences between nSTRIDE and saline groups in KOOS and WOMAC scores over time. Notably, APS injection resulted in significantly worse pain symptoms at 12 months compared to saline injection. Cite this article: Bone Joint J 2024;106-B(9):907–915