Abstract Introduction Machine learning (ML) models are increasingly being developed to predict outcomes in patients undergoing kidney transplantation. Little attention has been paid to assessing their performance across diverse patient subgroups, with risk of bias in predictive performance in subgroups of patients with protected characteristics. This study aimed to evaluate and address potential underperformance of a graft and patient survival prediction model in clinically relevant patient subgroups. Methods We assessed the predictive accuracy of a DeepHit survival analysis model, trained on over 25 years of transplant outcome data. The model's performance was evaluated across subgroups defined by patient ethnicity, sex, age, and pre-transplant sensitization status. To mitigate identified performance disparities, we applied several strategies to re-balance the model's training process to ensure all patient populations were adequately considered. Results Analysis revealed that the model's performance for graft survival was consistent across all investigated subgroups. However, a significant disparity was observed in patient survival predictions. The model demonstrated notably worse predictive accuracy for Black patients and patients over 70 years old, indicating a clear underperformance in these specific populations. Using bias mitigation strategies, we were able to partially close this performance gap. Conclusions This study highlights the critical importance of evaluating ML models for subgroup-specific underperformance in a clinical context. Our findings underscore that while overall model performance may appear strong, it can mask significant disparities. The application of mitigation strategies is essential for developing equitable and reliable decision support tools to ensure fair outcomes for all patients.

Abstract Introduction Best medical therapy (BMT), including antithrombotic and lipid-lowering treatment, is the standard of care for patients undergoing major lower limb amputation (MLLA), aiming to reduce major adverse cardiac events (MACE) and mortality. We assessed whether patients undergoing MLLA are receiving appropriate BMT. Methods This single-centre study included patients undergoing MLLA with clinical or radiological evidence of atherosclerosis. Medication use at admission and discharge was recorded. Triglyceride and LDL-C levels were measured at baseline, 3 months, and 6 months postoperatively. Results A total of 116 patients (81.9% male) were included. Comorbidities: diabetes (67.2%), hypertension (60.3%), heart failure (17.2%), and chronic kidney disease (28.5%). High-intensity statin use (75.0 to 82.5%, P = 0.0209) and ezetimibe increased significantly (6.9 to 11.3%, P = 0.0253). No patients were prescribed inclisiran, evolocumab, or icosapent ethyl, despite 10 meeting potential eligibility criteria. Antithrombotic use improved: full dose DOACs increased from 29.0 to 42.0% and warfarin from 0.9 to 2.7% (P < 0.001), COMPASS regimen increased from 12.9 to 21.4% (P = 0.0067). No statistical difference in single dose antiplatelet use was observed; patients receiving no antithrombotic therapy declined significantly (32 to 14, P < 0.001). Median LDL-C levels were 1.4 mmol/L baseline and at 6 months (P = 0.5443). Triglycerides levels increased 1.4 baseline to 2.0 mmol/L at 6 months (P = 0.1876). Conclusions While BMT prescribing improved following MLLA, adjunct lipid-lowering therapies remain underutilised reflected in no overall change in biochemical parameters. Broader implementation of full BMT is necessary to optimise cardiovascular risk reduction.

Abstract Background Hospital-acquired venous thromboembolism (VTE) remains a leading yet preventable cause of mortality. Based on National Institute for Health and Care Excellence (NICE) recommendations, all hospitalised patients should have a VTE risk assessment. This study examined factors impacting UK clinician engagement with the NICE VTE Clinical Practice Guidelines (CPGs) and VTE risk assessment tools (VTE-RATs). Methods The adapted-validated Clinician Guideline Determinants Questionnaire (CGDQ) was disseminated through clinical and trainee networks. Results Among the 100 respondents, most were doctors (78%), followed by nurses (15%) and other healthcare professionals (7%). 89% reported routinely completing VTE risk assessments, predominantly using bespoke Trust RATs (62%). Clinicians valued VTE-RATs for patient safety; two-thirds of respondents rated them as highly important. However, only 57% found them highly influential for thromboprophylaxis decisions. Enablers to using RATs were electronic integration, ease of use, and compulsory completion. Time pressures, lack of integration with patient records, and lack of clarity for complex cases were common barriers to RAT usage. Regarding perceptions of guidelines, most respondents (84%) felt guidelines optimise healthcare delivery and outcomes. Engagement with the NICE VTE CPGs was varied: 47% had read it frequently, 36% had once, and 14% were aware of it but hadn’t read it. Conclusions Despite widespread VTE-RAT usage, engagement with NICE VTE CPGs remained limited. The perception of VTE-RATs as important for patient safety but less so for guiding thromboprophylaxis decisions highlights a gap between clinical practice and guideline awareness. Better integration and standardisation of VTE-RATs may help align clinician practice with national guidelines.

Abstract Objectives Pelvic floor dysfunction is prevalent yet under-discussed, disproportionately affecting women and reducing quality of life. Many fail to seek help, particularly those from ethnic communities. The aim was to understand the patients being referred and if demographics are representative of the local population, to identify possible access issues across different communities. Methods This is a retrospective service evaluation across Cardiff and Vale University and Aneurin Bevan University Health Boards. We retrospectively compared Service Users (SU) (n = 549) over a one-month period attending pelvic health clinics (physiotherapy, nurses and consultants) to population data. Data points included age, sex, deprivation index (WIMD). For Cardiff, an additional questionnaire for patients collected data on language, ethnicity, religion, sexual orientation, employment and educational background. Results CAVUHB: SU deprivation index was representative of the local population. 91.7% of survey respondents identified as White-British, compared to 80% of local population. ABUHB: There was no significant difference in the overall WIMD of SU but older patients (>60 years) attending were less deprived than local population (WIMD median 1028 versus 799; P = 0.003). Younger patients were more likely to be deprived (t = −2.33, P = 0.02) and on logistic regression were more likely to be referred to physiotherapy. Conclusions In Cardiff, there is under-representation of minority groups amongst users, with higher education linked to better access. In ABUHB, there is under-representation of older patients from deprived areas, with younger patients being referred through a postnatal physiotherapy referral pathway. More research into what the barriers are, and how to overcome them is needed.

Abstract Inflammatory bowel disease (IBD), primarily presenting as ulcerative colitis or Crohn’s disease, is a chronic inflammatory condition of the gut leading to increased risk of developing colorectal cancer (CRC). IBD-induced CRC, termed colitis-associated cancer (CAC), with earlier onset and higher mortality rate than sporadic CRC, follows a flat inflammation-to-dysplasia-to-carcinoma path of tumorigenesis with a mixed immune and stromal cell-rich tumour microenvironment. Long non-coding RNAs (lncRNAs), a class of non-protein-coding RNA molecules more than 200 nucleotides long are expressed in a high tissue- and disease-specific manner and may represent promising targets for therapy. We aimed to identify a lncRNA that is dysregulated in CAC patients and can be targeted as a therapy to block the development of CAC from IBD. We have identified an endothelial cell-enriched lncRNA, provisionally named LINC-ECER, that is upregulated in both colon cancer and IBD. Knockdown of LINC-ECER and functional assays in Human Umbilical Vein Endothelial Cells (HUVECs) and lymphatic endothelial cells (HLECs) in vitro revealed that LINC-ECER is an important regulator of endothelial cell survival, growth and angiogenesis. Overall, our data proposes that LINC-ECER is a critical regulator of endothelial cell properties and can be targeted in IBD patients to specifically block angiogenesis and prevent CAC development.

Abstract Background Idiopathic granulomatous mastitis (IGM) is a rare, benign breast disease characterised by granulomatous inflammation. Current literature mainly comprises of small-sized retrospective studies with no well-designed clinical trials and limited representation of UK population. With no consensus to guide clinical practice there is a significant risk of delayed diagnosis and variation in management strategies that may adversely affect patient outcomes Methods ENIGMA is a multi-stage programme of work aiming to develop guidelines for the diagnosis and management of IGM in the UK. A scoping review will assess the range and quality of existing evidence, focusing on its relevance to UK practice. Findings from this review will inform a national practice questionnaire (NPQ) to capture current approaches to IGM care across the UK. These data, combined with a modified Delphi process involving experts and stakeholders, will support guideline development and help define priority research questions to advance best practice. Progress The scoping review identified 225 papers, involving 13 062 patients, mostly from Turkey, the USA, and China. Only three studies, covering 25 patients, originated from the UK. Commonly reported ethnic groups were Hispanic and Turkish. Risk factors include smoking and hormonal influences such as oestrogen and prolactin. Reported treatment strategies range from conservative approaches to medical therapies (antibiotics, steroids, immunosuppressants) and surgical interventions (drainage or excision). Preliminary results of the NPQ highlight substantial heterogeneity in UK diagnostic and management practices. Once the NPQ analysis is complete, the working group will initiate the Delphi process to establish consensus and inform guideline development.

Abstract Introduction Positive surgical margins (PSM) after robotic-assisted radical prostatectomy (RARP) can lead to negative oncological outcomes. This study aimed to identify sources of avoidable PSMs to improve outcomes. Methods Data from 100 consecutive cases with PSMs in 2022 were extracted for biopsy and MRI cancer location identification, RARP planning recommendations, surgeon operation notes, and final pathology results. Exclusion criteria were Neurosafe procedures, absence of pre-operative MRI, or undocumented RARP planning. PSMs were identified at staging, planning, and/or execution and classified as avoidable or unavoidable based on available information and decision-making. Results Out of 100 patients, 45 had extracapsular extension (ECE). 112 distinct PSMs were identified and classified as apical (32%, n = 36), basal (8%, n = 9), anterior (3.5%, n = 4) or postero-lateral (56%, n = 63). 49% of PSMs were intra-prostatic (n = 55). Sources of PSM included staging (28.5%), planning (∼20%) and execution (∼52%). However, ∼90% of execution errors were unavoidable, with PSMs occurring most commonly despite no function-sparing (∼70%) or adherence to the planning recommendation. Conversely, all planning errors were avoidable and occurred due to inappropriate functional sparing (78%), or unrecognised ECE or overlooked patient factors. Meanwhile, within staging errors, ∼28% were avoidable due to biopsy errors while the remaining were unavoidable (MRI-invisible disease). The most common avoidable PSM was due to inappropriate intra-fascial nerve-sparing recommended in the planning stage. Conclusions Overall, ∼30% of PSMs were avoidable. Reductions are achievable through improving biopsy targeting accuracy and MRI interpretation. Adherence to surgical planning and ongoing technical refinement can further improve oncological outcomes.

Abstract Background Long-term antiplatelet therapy reduces recurrent stroke after symptomatic carotid endarterectomy (sCEA). While both aspirin and clopidogrel are recommended, regional guidelines differ—and real-world practice may not always align. This study surveyed vascular surgeons to assess global prescribing patterns and compare them to local guideline recommendations. Methods Consultant vascular surgeons in four regions (UK, Europe, Australasia, Brazil) received an online questionnaire regarding their preferred long-term antiplatelet regimen following sCEA. We collected 357 responses (UK:192; Europe:107; Brazil:23; Australasia:35) and compared drug choices to regional guidelines. Mann–Whitney U tests (P < 0.05) and binary logistic regression were used to identify predictors of guideline adherence. Results Prescribing patterns varied significantly by region (P < 0.001). Clopidogrel was most common in the UK (80%), aligning with NICE guidance. Aspirin predominated in Australasia (89%) and Brazil (44%), consistent with local protocols. Guideline adherence was 80% in both the UK and Australasia (P > 0.05), but only 40% in Brazil. In Europe, aspirin and clopidogrel were similarly prescribed (43% 40%), despite ESVS guidance favouring clopidogrel, yielding 55% adherence. Dual-antiplatelet therapy was used by 9.5% overall, most frequently in Brazil (P < 0.01), despite guideline discouragement. Surgeon annual CEA volume and teaching-hospital status did not influence prescribing. Binary logistic regression confirmed region as the sole independent predictor of guideline adherence (P < 0.01). Conclusions Antiplatelet regimens post-sCEA vary globally. While most regions follow local guidelines, adherence is poorest in Europe and Brazil. Off-guideline dual therapy persists in 10% of cases. Improved dissemination and implementation strategies are needed to standardise prescribing and optimise stroke prevention.

Abstract Background Ischaemia–reperfusion injury (IRI) is a key factor in early morbidity and graft loss after pancreas transplantation. Oxygenated hypothermic machine perfusion (HMPO₂) has been shown to mitigate the consequences of IRI in liver and kidney preservation and is promising preclinically in pancreas preservation. The metabolic mechanisms underlying the benefit of HMPO₂ are poorly defined. This study aimed to highlight the key metabolites and pathways involved using metabolomic profiling. Methods 18 porcine pancreases, with 19 min of warm ischaemia, were preserved in three groups (n = 6): 8h static cold storage (SCS), 6 h SCS followed by 2 h HMPO2 and 8 h HMPO2, and then assessed by 60-minute normothermic reperfusion (NR) with whole blood. Human validation cohort: seven pancreases declined for transplantation arrived in SCS, underwent 2h HMPO₂ followed by 60-minute NR. Tissue samples were collected at start and end of cold preservation, and end of NR for untargeted metabolomics analysis using liquid chromatography-mass spectrometry (LC-MS). Results Key porcine tissue metabolic findings included:Succinate, a mitochondrial marker of ischaemia, increased during SCS and decreased during HMPO₂.Malate, a TCA metabolite, was consumed during HMPO₂ relative to SCS.S-adenosylmethionine was lower in continuous HMPO₂ than SCS, indicating methylation demand.L-cystine, a glutathione precursor, was lower in HMPO₂ than SCS.Human graft metabolite trends mirrored the porcine HMPO₂ groups, supporting translational relevance. Conclusions Compared with SCS, HMPO₂ was associated with changes in key metabolites involved in energy metabolism and respiration. This supports that oxygen delivery during HMPO₂ sustains residual metabolism under hypothermia and may mitigate IRI.

Abstract Background The British Rhinological Society Multidisciplinary Consensus on the Hospital Management of Epistaxis published in 2016 recommends the continuation of anticoagulants and antiplatelets in patients with uncomplicated epistaxis. The aim of this audit is to improve adherence to the current consensus guideline. Methods Patients who were admitted under the ear, nose and throat (ENT) department at the Norfolk and Norwich University Hospital (NNUH) with uncomplicated epistaxis from May to September 2024 (cycle one) and were on anticoagulants or antiplatelets were included in our audit. We calculated how many had their antiplatelets or anticoagulants suspended during their admission. We implemented teaching on the national consensus guideline at the December 2024 ENT departmental induction for new ENT doctors. We then recollected the data from December 2024 to February 2025 (second cycle). Results In the first cycle, 22 patients were identified. Anticoagulants or antiplatelets were suspended in 68% of these patients. In the second cycle, 25 patients were identified, and 44% had their anticoagulants or antiplatelets suspended. There was a decrease in the percentage of patients who had their anticoagulants or antiplatelets suspended between the first and second cycles (P < 0.1), leading to an improvement in the adherence to the national consensus guideline. Conclusions Teaching at the ENT department induction improved adherence to the current guideline recommendations to continue anticoagulation and antiplatelets in patients admitted with epistaxis. We aim to share the guideline with emergency department and medical colleagues, and to continue teaching on these guidelines at departmental induction for new ENT doctors.