Independent Registry Research Articles

Machine learning prediction of in-hospital mortality and external validation in patients with cardiogenic shock: the RESCUE score.

Despite advances in mechanical circulatory support, mortality rates in cardiogenic shock (CS) remain high. A reliable risk stratification system could serve as a valuable guide in the clinical management of patients with CS. This study aimed to develop and externally validate a risk prediction model for in-hospital mortality in CS patients using machine learning (ML) algorithms. Data from 1247 patients with all-cause CS in the RESCUE registry (January 2014-December 2018) were analyzed. Key predictive variables were identified using 4 ML algorithms. A risk prediction model, the RESCUE score, was developed using logistic regression based on the selected variables. Internal validation was conducted within the RESCUE registry, and external validation was performed using an independent CS registry of 750 patients. The 4 ML models identified 7 predictors: age, vasoactive inotropic score, left ventricular ejection fraction, lactic acid level, in-hospital cardiac arrest at presentation, need for continuous renal replacement therapy, and mechanical ventilation. The RESCUE score demonstrated strong predictive performance, with an AUC of 0.86 (95%CI, 0.83-0.88) for in-hospital mortality. Ten-fold internal cross-validation yielded an AUC of 0.86 (95%CI, 0.77-0.95). External validation showed an AUC of 0.80 (95%CI, 0.76-0.84). Our ML-based risk-scoring system, the RESCUE score, demonstrated excellent predictive performance for in-hospital mortality in all patients with CS patients, regardless of cause. The system could be a useful and reliable tool to estimate risk stratification of CS in everyday clinical practice. NCT02985008.

Population-based cross-sectional survey of cervical cancer screening prevalence and socio-demographic correlates in Bangladeshi women.

Cervical cancer, albeit preventable, is the second-most deadly gynecological cancer in developing nations. Little is known about cervical cancer among Bangladeshi women. This study aims to estimate the prevalence of cervical cancer screening and demographic correlates to identify potential variabilities in screening rates among different demographic groups and regions. This study used secondary data from the WHO STEPS 2018 Survey. We used Bayesian regression to perform the bivariate analyses between the outcome and each explanatory factor, as it generates more acceptable results and improves parameter estimates. The top-ranked socio-demographic factors were identified using a two-step cluster analysis. This method determines the relevance of predictor variables and automatically establishes the number of clusters. The prevalence of Bangladeshi women who had ever been screened for cervical cancer was 6.2%. In the adjusted model, women with the following socio-demographic factors had a higher likelihood of developing cervical cancer: being 18-29 years old (AOR = 3.3, 95% CI: 0.24, 15.27) or 45-59 years old (AOR = 2.8, 95% CI: 1.22, 6.0), currently married (AOR = 2.3, 95% CI: 1.36, 3.70), and employed (AOR = 2.4, 95% CI: 1.40, 4.06). Women in the Barisal division were found to have higher odds of being screened for cervical cancer (AOR = 21, 95% CI: 0.66, 121.97). Cluster analysis found residence status predisposes women to cervical cancer screening. There is a significant potential for substantial reductions in the burden of cervical cancer in Bangladesh by strengthening the application of cervical cancer screening. Future studies should examine how socioeconomic status, culture, and healthcare access affect cervical cancer screening trends for different divisions in Bangladesh. An independent national cancer registry is urgently needed to evaluate screening trends and outcomes.

Establishing Treatment Effectiveness in Fabry Disease: Observation-Based Recommendations for Improvement.

Fabry disease (FD, OMIM #301500) is caused by pathogenic GLA gene (OMIM #300644) variants, resulting in a deficiency of the α-galactosidase A enzyme with accumulation of its substrate globotriaosylceramide and its derivatives. The phenotype of FD is highly variable, with distinctive disease features and course in classical male patients but more diverse and often nonspecific features in non-classical and female patients. FD-specific therapies have been available for approximately two decades, yet establishing robust evidence for long-term effectiveness remains challenging. This review aims to identify the factors contributing to this lack of robust evidence for the treatment of FD with enzyme replacement therapy (ERT) (agalsidase-alfa and -beta and pegunigalsidase alfa) and chaperone therapy (migalastat). Major factors that have been identified are study population heterogeneity (concerning sex, age, phenotype, disease stage) and differences in study design (control groups, outcomes assessed), as well as the short duration of studies. To address these challenges, we advocate for patient matching to improve control group compatibility in future FD therapy studies. We recommend international collaboration and harmonization, facilitated by an independent FD registry. We propose a stepwise approach for evaluating the effectiveness of novel treatments, including recommendations for surrogate outcomes and required study duration.

Abstract PO1-02-02: Update of RSClin with extended TAILORx follow-up and development and validation of a new tool for risk of late distant recurrence

Abstract Background: RSClin® (Sparano et al J Clin Oncol 2020, PMID 33306425) is an educational tool that estimates the 10-year risk of distant recurrence (DR) and the absolute benefit of chemotherapy in HR-positive, HER2-negative, node-negative early breast cancer. The RSClin tool integrates the Recurrence Score® (RS) result with the clinical-pathological features tumor grade, tumor size and patient age using patient-specific meta-analysis (PSMA) applied to the NSABP B-14, TAILORx and NSABP B-20 studies. Approximately 3 years of additional follow-up data have accumulated in the TAILORx study since the RSClin tool was developed, including 50% more DR events in the RS 11-25 group (375 vs. 250), and 46% more DR events in the overall population (561 vs. 384). Methods: The RSClin tool estimates were updated to include the extended follow-up data from TAILORx. As before, the PSMA included 10,004 women with hormone-receptor positive, HER2-negative, node-negative breast cancer from the NSABP B-14 (n=577) trial receiving endocrine therapy (ET) alone and the TAILORx trial receiving ET alone (n=4854) or chemotherapy (CT) plus ET (n=4573). The baseline risk estimate used TAILORx event rates with ET alone. Since there was evidence of differing hazard ratios in years 0-5 versus 5+, models using time-varying effects were used. A patient-specific estimator of absolute CT benefit was computed by combining a PSMA estimate of the individualized relative CT effect from the TAILORx and NSABP B-20 (n=550 HER2-negative) trials with risk estimates for ET alone. Similar methods were used to develop and validate a new tool for assessing the risk of late DR (from year 5 to 10) in patients who survive DR-free for 5 years on ET alone after surgery. External validation of DR risk estimation was performed in an independent cohort of 1098 women enrolled in the Clalit Health Service registry. Results: With the updated data, the model integrating RS result with clinical-pathological features was more informative for DR risk than either RS result alone or the clinical-pathological features alone, both for 10-year DR risk and for late DR risk (likelihood ratio tests, each p≤.001). The updated 10-year RSClin risk estimates are similar to the existing RSClin tool risk estimates with slightly lower risk estimates for very high RS results and more precise estimates for mid-range RS results. The updated chemotherapy effect estimates are slightly lower than for the existing RSClin. The updated RSClin tool 10-year risk estimates were strongly associated with DR risk in the independent Clalit registry (Cox regression p< .001) and closely approximated the observed DR risk (Lin concordance 0.87). The new tool’s late DR risk estimates beyond 5 years were strongly associated with late DR risk in the Clalit registry (Cox regression p=.002) and closely approximated the observed late DR risk (Lin concordance 0.92). Conclusions: The RSClin tool 10-year DR risk estimates have been updated to incorporate extended follow-up data from the TAILORx study. A new tool for assessing late DR risk has been developed and validated that provides more prognostic information than either clinical-pathological features or genomic risk used individually. Citation Format: Joseph Sparano, Michael Crager, Robert Gray, Gong Tang, Jess Hoag, Frederick Baehner, Charles Geyer, Steven Shak, Norman Wolmark, Salomon Stemmer. Update of RSClin with extended TAILORx follow-up and development and validation of a new tool for risk of late distant recurrence [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-02-02.

Penile cancer care in the Netherlands: increased incidence, centralisation, and improved survival.

To evaluate penile squamous cell carcinoma (PSCC) incidence and centralisation trends in the Netherlands over the past three decades, as well as the effect of centralisation of PSCC care on survival. In the Netherlands PSCC care is largely centralised in one national centre of expertise (Netherlands Cancer Institute [NCI], Amsterdam). For this study, the Netherlands Cancer Registry, an independent nationwide cancer registry, provided per-patient data on age, clinical and pathological tumour staging, follow-up, and vital status. Patients with treatment at the NCI were identified and compared to patients who were treated at all other centres. The age-standardised incidence rate was calculated with the European Standard Population. The probability of death due to PSCC was estimated using the relative survival. Multivariable Cox regression analysis was performed to evaluate predictors of survival. A total of 3160 patients were diagnosed with PSCC between 1990 and 2020, showing a rising incidence (P < 0.001). Annual caseload increased at the NCI (1% in 1990, 65% in 2020) and decreased at other (regional) centres (99% to 35%). Despite a relatively high percentage of patients with T2-4 (64%) and N+ (33%) at the NCI, the 5-year relative survival was higher (86%, 95% confidence interval [CI] 82-91%) compared to regional centres (76%, 95% CI 73-80%, P < 0.001). Patients with a pathological T2 tumour were treated with glans-sparing treatment more often at the reference centre than at the regional centres (16% vs 5.0%, P < 0.001). After adjusting for age, histological grading, T-stage, presence of lymph node involvement and year of diagnosis, treatment at regional centres remained a predictor for worse survival (hazard ratio 1.22, 95% CI 1.05-1.39; P = 0.006). The incidence of PSCC in the Netherlands has been gradually increasing over the past three decades, with a noticeable trend towards centralisation of PSCC care and improved relative survival rate.

The contribution of pain catastrophizing, depression and anxiety symptoms among patients with persistent pain and opioid misuse behaviours.

Anxiety, depression and pain catastrophizing are independently associated with risk of opioid misuse in patients with persistent pain but their relationship to current opioid misuse, when considered together, is poorly understood. This study will assess the relative contribution of these modifiable, and distinct psychological constructs to current opioid misuse in patients with persistent pain. One hundred and twenty-seven patients referred to a specialized opioid management clinic for prescription opioid misuse within a tertiary pain service were recruited for this study. The Pain Catastrophizing Scale, Depression, Anxiety and Stress Scales and the Current Opioid Misuse Measure were administered pre-treatment. Pain severity and morphine equivalent dose based on independent registry data were also recorded. Higher levels of pain catastrophizing, depression, and anxiety were significantly associated with higher current opioid misuse (r = .475, 0.599, and 0.516 respectively, p < .01). Pain severity was significantly associated with pain catastrophizing (r = .301, p < .01). Catastrophizing, depression, and anxiety explained an additional 11.56% of the variance (R2 change = 0.34, p < .01) over and above age, gender, pain severity and morphine equivalent dose. Depression was the only significant variable at Step 2 (β = 0.62, p < .01). Findings show that in a sample of people with persistent pain referred for treatment for opioid misuse, depression contributes over and above that of anxiety and pain catastrophizing. Theoretical and clinical practice implications are presented.

Endothelial activation and stress index as a prognostic factor of diffuse large B-cell lymphoma: the report from the nationwide multi-center Thai Lymphoma Study Group.

Several prognostic models have been introduced to predict outcomes of patients with diffuse large B-cell lymphoma (DLBCL). Endothelial activation and stress index (EASIX) is a surrogate of endothelial dysfunction which has been shown to predict outcomes of patients with various hematologic malignancies. However, the prognostic implication of EASIX for DLBCL is limited and warrants exploration. We conducted a retrospective study enrolling adult DLBCL patients including a discovery cohort from the single-centered university hospital database and a validation cohort from the independent nationwide multi-center registry. EASIX scores were calculated using creatinine, lactate dehydrogenase, and platelet levels. The receiver operating characteristic curve analysis was used to determine optimal cutoff. Statistical analysis explored the impact of EASIX on survival outcomes. A total of 323 patients were included in the discovery cohort. The optimal EASIX cutoff was 1.07 stratifying patients into low (53.9%) and high EASIX (46.1%) groups. Patients with high EASIX had worse 2-year progression-free survival (PFS) (53.4% vs. 81.5%, p<0.001) and overall survival (OS) (64.4% vs. 88.7%, p<0.001) than patients with low EASIX. Multivariate analysis revealed that older age, bulky disease, impaired performance status, and high EASIX were associated with an unfavorable OS. In the validation cohort of 499 patients, the optimal EASIX cutoff was 1.04. Similar to the discovery cohort, high EASIX score was associated with high-risk diseases, worse PFS, and inferior OS. In conclusion, EASIX score was significantly associated with survival outcomes and may be used as a simple prognostic tool to better risk-classify DLBCL.

Investigating significant health trends in growth hormone treatments registry: rationale, aims and design of a nationwide prospective registry (study protocol)

BackgroundSomatropin treatment is indicated in a variety of disorders including growth hormone (GH) deficiency, Prader–Willi and Turner syndrome, chronic renal insufficiency and others. To date, almost all studies have been limited to single GH products, and no independent registry across indications and somatropin products was ever established.AimThe present investigator-initiated registry named INSIGHTS-GHT aims to provide comprehensive information on various aspects of somatropin treatment in Germany in approved indications within routine clinical practice: drug utilization, effectiveness (including real final height, body composition), tolerability, quality of life, other patient related outcomes (PRO), and health economic variables.MethodsRegistry (prospective observational study) in specialised pediatric and adult endocrinology centres in Germany. Patients of any age are eligible for documentation, if they are on ongoing or newly initiated treatment with any approved somatropin or somatropin-related product within the labelling, available for long term follow-up documentation, and if they provided informed consent. Subjects may switch, discontinue/interrupt or initiate somatropin products at any time. They are followed up for at least 3 years (minimal study duration). Documentation is planned once or twice per year to record somatropin utilisation (product, dosing), other medications, laboratory status (glucose, lipids, GH function including stimulation tests, IGF-I, IGFBP3), if applicable, pubertal development, auxological parameters, body composition and bone age. Patient reported outcome (PRO) measures include, but are not limited to, Short Form 12 in adults and adolescents aged 14 years and over. Safety reporting includes adverse events.ConclusionsThe registry documents children and adults in one joint registry, includes, at present, patients in Germany and allows documentation of patients on all approved somatropin and other growth hormone preparations. It will allow to describe the transition of subjects from adolescence to adulthood (treatment and height), to describe switches between somatotropin preparations, to perform responder analyses, and to analyse differences and similarities of somatropin utilization (by age group, sex, setting, and PRO instrument). INSIGHTS-GHT offers a broad, comprehensive research platform to assess multiple relevant aspects of somatropin treatment and outcomes (including the transition of subjects from adolescence to adulthood), allows the documentation of all GH products including long-acting GH preparations after their introduction, and will evaluate the data independently of funders.Trial registration BfArM Nr. NIS7492, DRKS registry DRKS00027394.

Reduced methylation correlates with diabetic nephropathy risk in type 1 diabetes.

Diabetic nephropathy (DN) is a polygenic disorder with few risk variants showing robust replication in large-scale genome-wide association studies. To understand the role of DNA methylation, it is important to have the prevailing genomic view to distinguish key sequence elements that influence gene expression. This is particularly challenging for DN because genome-wide methylation patterns are poorly defined. While methylation is known to alter gene expression, the importance of this causal relationship is obscured by array-based technologies since coverage outside promoter regions is low. To overcome these challenges, we performed methylation sequencing using leukocytes derived from participants of the Finnish Diabetic Nephropathy (FinnDiane) type 1 diabetes (T1D) study (n = 39) that was subsequently replicated in a larger validation cohort (n = 296). Gene body-related regions made up more than 60% of the methylation differences and emphasized the importance of methylation sequencing. We observed differentially methylated genes associated with DN in 3 independent T1D registries originating from Denmark (n = 445), Hong Kong (n = 107), and Thailand (n = 130). Reduced DNA methylation at CTCF and Pol2B sites was tightly connected with DN pathways that include insulin signaling, lipid metabolism, and fibrosis. To define the pathophysiological significance of these population findings, methylation indices were assessed in human renal cells such as podocytes and proximal convoluted tubule cells. The expression of core genes was associated with reduced methylation, elevated CTCF and Pol2B binding, and the activation of insulin-signaling phosphoproteins in hyperglycemic cells. These experimental observations also closely parallel methylation-mediated regulation in human macrophages and vascular endothelial cells.

External Validation of a Machine Learning Algorithm for Predicting Clinically Meaningful Functional Improvement After Arthroscopic Hip Preservation Surgery

Background: Individualized risk prediction has become possible with machine learning (ML), which may have important implications in enhancing clinical decision making. We previously developed an ML algorithm to predict propensity for clinically meaningful outcome improvement after hip arthroscopy for femoroacetabular impingement syndrome. External validity of prognostic models is critical to determine generalizability, although it is rarely performed. Purpose: To assess the external validity of an ML algorithm for predicting clinically meaningful improvement after hip arthroscopy. Study Design: Cohort study; Level of evidence, 3. Methods: An independent hip preservation registry at a tertiary academic medical center was queried for consecutive patients/athletes who underwent hip arthroscopy for femoroacetabular impingement syndrome between 2015 and 2017. By assuming a minimal clinically important difference (MCID) outcome/event proportion of 75% based on the original study, a minimum sample of 132 patients was required. In total, 154 patients were included. Age, body mass index, alpha angle on anteroposterior pelvic radiographs, Tönnis grade and angle, and preoperative Hip Outcome Score–Sports Subscale were used as model inputs to predict the MCID for the Hip Outcome Score–Sports Subscale 2 years postoperatively. Performance was assessed using identical metrics to the internal validation study and included discrimination, calibration, Brier score, and decision curve analysis. Results: The concordance statistic in the validation cohort was 0.80 (95% CI, 0.71 to 0.87), suggesting good to excellent discrimination. The calibration slope was 1.16 (95% CI, 0.74 to 1.61) and the calibration intercept 0.13 (95% CI, –0.26 to 0.53). The Brier score was 0.15 (95% CI, 0.12 to 0.18). The null model Brier score was 0.20. Decision curve analysis revealed favorable net treatment benefit for patients with use of the algorithm as compared with interventional changes made for all and no patients. Conclusion: The performance of this algorithm in an independent patient population in the northeast region of the United States demonstrated superior discrimination and comparable calibration to that of the derivation cohort. The external validation of this algorithm suggests that it is a reliable method to predict propensity for clinically meaningful improvement after hip arthroscopy and is an essential step forward toward introducing initial use in clinical practice. Potential uses include integration into electronic medical records for automated prediction, enhanced shared decision making, and more informed allocation of resources to optimize patient outcomes.

Development of a Prediction Model for COVID-19 Acute Respiratory Distress Syndrome in Patients With Rheumatic Diseases: Results From the Global Rheumatology Alliance Registry.

ObjectiveSome patients with rheumatic diseases might be at higher risk for coronavirus disease 2019 (COVID‐19) acute respiratory distress syndrome (ARDS). We aimed to develop a prediction model for COVID‐19 ARDS in this population and to create a simple risk score calculator for use in clinical settings.MethodsData were derived from the COVID‐19 Global Rheumatology Alliance Registry from March 24, 2020, to May 12, 2021. Seven machine learning classifiers were trained on ARDS outcomes using 83 variables obtained at COVID‐19 diagnosis. Predictive performance was assessed in a US test set and was validated in patients from four countries with independent registries using area under the curve (AUC), accuracy, sensitivity, and specificity. A simple risk score calculator was developed using a regression model incorporating the most influential predictors from the best performing classifier.ResultsThe study included 8633 patients from 74 countries, of whom 523 (6%) had ARDS. Gradient boosting had the highest mean AUC (0.78; 95% confidence interval [CI]: 0.67‐0.88) and was considered the top performing classifier. Ten predictors were identified as key risk factors and were included in a regression model. The regression model that predicted ARDS with 71% (95% CI: 61%‐83%) sensitivity in the test set, and with sensitivities ranging from 61% to 80% in countries with independent registries, was used to develop the risk score calculator.ConclusionWe were able to predict ARDS with good sensitivity using information readily available at COVID‐19 diagnosis. The proposed risk score calculator has the potential to guide risk stratification for treatments, such as monoclonal antibodies, that have potential to reduce COVID‐19 disease progression.

Machine learning-derived major adverse event prediction of patients undergoing transvenous lead extraction: Using the ESC EHRA EORP European lead extraction ConTRolled ELECTRa registry.

Transvenous lead extraction (TLE) remains a high-risk procedure. The purpose of this study was to develop a machine learning (ML)-based risk stratification system to predict the risk of major adverse events (MAEs) after TLE. A MAE was defined as procedure-related major complication and procedure-related death. We designed and evaluated an ML-based risk stratification system trained using the European Lead Extraction ConTRolled (ELECTRa) registry to predict the risk of MAEs in 3555 patients undergoing TLE and tested this on an independent registry of 1171 patients. ML models were developed, including a self-normalizing neural network (SNN), stepwise logistic regression model ("stepwise model"), support vector machines, and random forest model. These were compared with the ELECTRa Registry Outcome Score (EROS) for MAEs. There were 53 MAEs (1.7%) in the training cohort and 24 (2.4%) in the test cohort. Thirty-two clinically important features were used to train the models. ML techniques were similar to EROS by balanced accuracy (stepwise model: 0.74 vs EROS: 0.70) and superior by area under the curve (support vector machines: 0.764 vs EROS: 0.677). The SNN provided a finite risk for MAE and accurately identified MAE in 14 of 169 "high (>80%) risk" patients (8.3%) and no MAEs in all 198 "low (<20%) risk" patients (100%). ML models incrementally improved risk prediction for identifying those at risk of MAEs. The SNN has the additional advantage of providing a personalized finite risk assessment for patients. This may aid patient decision making and allow better preoperative risk assessment and resource allocation.

How long does an elbow replacement last? A systematic review and meta-analysis of case-series and national registry reports with more than 10 years of follow-up.

This study aims to determine, for the first time, generalizable data on the longevity and long-term function of elbow replacements. In this systematic review and meta-analysis, we searched MEDLINE and Embase for articles reporting 10-year or greater survival of total elbow replacements (TERs) and distal humeral hemiarthroplasty. Implant survival and patient reported outcome measures (PROMs) data were extracted. National joint replacement registries were also analyzed. We weighted each series and calculated a pooled survival estimate at 10, 15, and 20 years. For PROMs we pooled the standardized mean difference (SMD) at 10 years. Despite its widespread use, we identified only 9 series reporting all-cause survival of 628 linked TERs and 610 unlinked TERs and no series for distal humeral hemiarthroplasty. The studied population was treated for rheumatoid arthritis in over 90% of cases. The estimated 10-year survival for linked TERs was 92% (95% CI 90-95) and unlinked TERs 84% (CI 81-88). 2 independent registries contributed 32 linked TERs and 530 unlinked TERs. The pooled registry 10-year survival for unlinked TERs was 86% (CI 83-89). Pooled 10-year PROMs from 164 TERs (33 linked and 131 unlinked), revealed a substantial improvement from baseline scores (SMD 2.7 [CI 1.6-3.8]). Over 80% of all elbow replacements and over 90% of linked elbow replacements can last more than 10 years with sustained patient-reported benefits. This information is long overdue and will be particularly useful to patients as well as healthcare providers.

The subcutaneous defibrillator in patients with low BMI - insights from a large European multicenter registry

Abstract Funding Acknowledgements Type of funding sources: None. Background The subcutaneous implantable cardioverter defibrillator (S-ICD) has become an alternative to transvenous ICDs (tv-ICD), especially in young patients without a need for pacing. One of the current limitations of the S-ICD is the relatively large size of the generator compared to tv-ICDs. There is little evidence whether the size of the current S-ICD generator is associated with an elevated risk of device-related complications in patients with a low body mass index (BMI). Purpose To compare the device-related complications and long-term outcomes in a large real world cohort of S-ICD recipients in patients with a BMI &amp;lt;18 kg/m2 compared to patients with a BMI &amp;gt;18 kg/m2. Methods The iSuSI registry is a European, multi-center, open-label, independent, and physician-initiated observational registry. A total of twenty-two Public and Private Healthcare Institutions from 4 different countries in Europe were involved in the registry. All consecutive patients meeting current guideline indications for ICD implantation and undergoing implantation of a S-ICD device (Boston Scientific, Marlborough, Massachusetts, USA) at 21 European institutions enrolled in the registry were used for the current analysis. Patients were classified into two cohorts, depending on the BMI at the time of device implantations: BMI &amp;lt; 18 kg/m2 versus &amp;gt; 18 kg/m2. Results Out of a total of 1497 pts, 58 pts (3.9%) had a BMI &amp;lt; 18 kg/m2. Patients with BMI &amp;lt;18 kg/m2 were younger (44.6±2.4 vs 50.8±0.4; p=0.004) and more frequently female (58.6% vs 22.3%, p&amp;lt;0.001). No differences in any of the other baseline characteristic were observed. Implantation techniques resulted comparable between the groups (Rates of 2-incision technique: 87.8% vs 91.9%; p=0.256; inter-muscular placement: 89.7% vs 83.3%; p=0.198). Of note, the mean PRAETORIAN score at implantation of patients with BMI &amp;lt;18 kg/m2 was significantly lower (33.8±9.1 vs 54.1±47.3; p=0.035), although the vast majority of pts in both cohorts qualify as at low risk of conversion failure (100% vs 91.4%; p=0.436). Over a median follow up time of 22.4 [11.6–36.8] months, both overall device-related complications (5.2% vs 7.4%) and rates of inappropriate shocks (12.0% vs 8.8%) resulted comparable between the two groups (p =0.517 and p=0.385, respectively). Figure1 reports Kaplan-Meier curves reporting the combined incidence of device-related complications and inappropriate shocks in the two groups (log-rank p = 0.576). Conclusion No difference in device-related complications and long-term outcomes after S-ICD implantation were observed in patients with BMI &amp;lt;18 kg/m2 compared to the remaining recipients from a large, multi-centered S-ICD registry. Figure 1: Kaplan-Meier-survival curve for the combined endpoint of inappropriate shocks (IAS) and device-related complications (DRC)

PO-619-02 MACHINE LEARNING-DERIVED MAJOR ADVERSE EVENT PREDICTION OF PATIENTS UNDERGOING TRANSVENOUS LEAD EXTRACTION: UTILISING THE ESC EHRA EORP EUROPEAN LEAD EXTRACTION CONTROLLED ELECTRA REGISTRY

Transvenous lead extraction (TLE) remains a high risk procedure. The purpose of this study was to develop a machine learning (ML)-based risk stratification system to predict risk of major adverse events (MAEs) following TLE. MAE was defined as procedure-related major complication and procedure-related deaths. We designed and evaluated a ML-based risk stratification trained using the ELECTRa registry to predict risk of MAEs in 3555 patients undergoing TLE and tested this on an independent registry of 1171 patients. ML models were developed, including a self-normalising neural network ( SNN ), stepwise logistic regression (“stepwise model”), support vector machines (SVM) and random forest (RF) model. These were compared to the Electra Registry Outcome Score (EROS) for MAEs. There were 53 MAEs (1.7%) in the training cohort and 24 (2.4%) in the test cohort. 32 clinically important features were used to train the models. ML techniques were similar to EROS by balanced accuracy (stepwise model: 0.74 vs EROS: 0.70) and superior by area under the curve (SVM: 0.764 vs EROS: 0.677). The SNN provided a finite risk for MAE, and accurately identified MAE in 14 out of 169 (8.3%) “high (>80%) risk” patients, and no MAEs in all 198 (100%) “low (<20%) risk” patients. ML models incrementally improved risk prediction for identifying those at risk for MAEs. The SNN has the additional advantage of providing a personalised, finite risk assessment for patients. This may aid patient decision making and allow better pre-operative risk assessment and resource allocation.

Semantic modelling of common data elements for rare disease registries, and a prototype workflow for their deployment over registry data

BackgroundThe European Platform on Rare Disease Registration (EU RD Platform) aims to address the fragmentation of European rare disease (RD) patient data, scattered among hundreds of independent and non-coordinating registries, by establishing standards for integration and interoperability. The first practical output of this effort was a set of 16 Common Data Elements (CDEs) that should be implemented by all RD registries. Interoperability, however, requires decisions beyond data elements - including data models, formats, and semantics. Within the European Joint Programme on Rare Diseases (EJP RD), we aim to further the goals of the EU RD Platform by generating reusable RD semantic model templates that follow the FAIR Data Principles.ResultsThrough a team-based iterative approach, we created semantically grounded models to represent each of the CDEs, using the SemanticScience Integrated Ontology as the core framework for representing the entities and their relationships. Within that framework, we mapped the concepts represented in the CDEs, and their possible values, into domain ontologies such as the Orphanet Rare Disease Ontology, Human Phenotype Ontology and National Cancer Institute Thesaurus. Finally, we created an exemplar, reusable ETL pipeline that we will be deploying over these non-coordinating data repositories to assist them in creating model-compliant FAIR data without requiring site-specific coding nor expertise in Linked Data or FAIR.ConclusionsWithin the EJP RD project, we determined that creating reusable, expert-designed templates reduced or eliminated the requirement for our participating biomedical domain experts and rare disease data hosts to understand OWL semantics. This enabled them to publish highly expressive FAIR data using tools and approaches that were already familiar to them.

Development and External Validation of a Model to Predict Complex Treatment After Radiofrequency Ablation for Barrett’s Esophagus With Early Neoplasia

Endoscopic eradication therapy for Barrett's esophagus (BE)-related neoplasia is safe and leads to complete eradication in the majority of patients. However, a subgroup will experience a more complex treatment course with a risk for failure or disease progression. Early identification of these patients may improve patient counseling and treatment outcomes. We aimed to develop a prognostic model for a complex treatment course. We collected data from a nationwide registry that captures outcomes for all patients undergoing endoscopic eradication therapy for early BE neoplasia. A complex treatment course was defined as neoplastic progression, treatment failure, or the need for endoscopic resection during the radiofrequency ablation treatment phase. We developed a prognostic model using logistic regression. We externally validated our model in an independent registry. A total of 1386 patients were included, of whom 78 (6%) had a complex treatment course. Our model identified patients with a BE length of 9 cm or longer with a visible lesion containing high-grade dysplasia/cancer, and patients with less than 50% squamous conversion after radiofrequency ablation were identified as high risk for a complex treatment. This applied to 8% of the study population and included 93% of all treatment failures and 76% of all patients with advanced neoplastic progression. The model appeared robust in multiple sensitivity analyses and performed well in external validation (area under the curve, 0.84). We developed a prognostic model that identified patients with a BE length of 9 cm or longer and high-grade dysplasia/esophageal adenocarcinoma and those with poor squamous regeneration as high risk for a complex treatment course. The good performance in external validation suggests that it may be used in clinical management (Netherlands Trial Register: NL7039).

Clinical disease characteristics of patients with Niemann-Pick Disease Type C: findings from the International Niemann-Pick Disease Registry (INPDR)

BackgroundNiemann-Pick Disease Type C (NPC) is an autosomal recessive rare disease characterised by progressive neurovisceral manifestations. The collection of on-going large-scale NPC clinical data may generate better understandings of the natural history of the disease. Here we report NPC patient data from the International Niemann-Pick Disease Registry (INPDR).MethodThe INPDR is a web-based, patient-led independent registry for the collection of prospective and retrospective clinical data from Niemann-Pick Disease patients. Baseline data from NPC patients enrolled into the INPDR from September 2014 to December 2019 was extracted to analyse the demographic, genetic and clinical features of the disease.ResultsA total of 203 NPC patients from six European countries were included in this study. The mean age (SD) at diagnosis was 11.2 years (14.2). Among enrolled patients, 168 had known neurological manifestations: 43 (24.2%) had early-infantile onset, 47 (26.4%) had late-infantile onset, 41 (23.0%) had juvenile onset, and 37 (20.8%) had adult onset. 10 (5.6%) patients had the neonatal rapidly fatal systemic form. Among the 97 patients with identified NPC1 variants, the most common variant was the c. 3182T > C variant responsible for the p.lle1061Thr protein change, reported in 35.1% (N = 34) of patients. The frequencies of hepatomegaly and neonatal jaundice were greatest in patients with early-infantile and late-infantile neurological onset. Splenomegaly was the most commonly reported observation, including 80% of adult-onset patients. The most commonly reported neurological manifestations were cognitive impairment (78.5%), dysarthria (75.9%), ataxia (75.9%), vertical supranuclear gaze palsy (70.9%) and dysphagia (69.6%). A 6-domain composite disability scale was used to calculate the overall disability score for each neurological form. Across all with neurological onset, the majority of patients showed moderate to severe impairments in all domains, except for ‘swallowing’ and ‘seizure’. The age at diagnosis and death increased with increased age of neurological symptom onset. Miglustat use was recorded in 62.4% of patients and the most common symptomatic therapies used by patients were antiepileptics (32.9%), antidepressants (11.8%) and antacids (9.4%).ConclusionThe proportion of participants at each age of neurological onset was relatively equal across the cohort. Neurological manifestations, such as ataxia, dysphagia, and dysarthria, were frequently observed across all age categories.

Risk factors for total ankle arthroplasty failure: A Dutch Arthroplasty Register study

BackgroundStudies concerning total ankle arthroplasty could be influenced by several forms of bias. Independent national arthroplasty registries represent objective data on survival and patient reported outcomes. The aim of this study was to determine survival and identify risk factors for early failure in a nationwide series of total ankle arthroplasties from the Dutch Arthroplasty Register (LROI). Patients and methodsData of 810 patients, who received 836 total ankle arthroplasties between 2014 and 2020 were obtained from the Dutch Arthroplasty Register (LROI) with a median follow-up of 38 months (range 1–84 months). Survival was expressed in Kaplan-Meier analysis and associated hazard ratios for implant failure were determined. Implant failure was defined as the need for revision surgery for any reason or (pan)arthrodesis. ResultsDuring follow-up, we recorded 39 failures (4.7%) resulting in a implant survival of 95.3% with a median follow-up of 38 months (range 1–84 months). Medial malleolus osteotomy (HR = 2.27), previous surgery (HR = 1.83), previous osteotomy (HR = 2.82) and previous ligament reconstruction (HR = 2.83) all showed potentially clinically meaningful associations with a higher incidence of implant failure, yet only previous OCD treatment (HR = 6.21), BMI (HR = 1.09) and age (HR = 0.71) were statistically significant. InterpretationExcellent short-term survival (95.3%) with a median follow-up of 38 months was reported for TAA patients from the Dutch Arthroplasty Register. Patients with a lower age, a higher BMI or who had a prior surgical OCD treatment before TAA surgery appear to have a higher risk for revision after short-term clinical follow-up. Thorough patient selection with emphasis on risk factors associated with early implant failure might be essential to improve TAA survivorship.

Spatially Enabled Web Application for Urban Cultural Heritage Monitoring and Metrics Reporting for the SDGs

UNESCO and the United Nations have recently identified cultural heritage (CH) as a key enabler of sustainability by incorporating it into several Sustainable Development Goals (SDGs). Accurate and efficient reporting on CH is considered fundamental despite known limitations due to the lack of sufficient and harmonized data. This paper presents a spatially enabled web application for urban CH monitoring for the city of Thessaloniki in Northern Greece. The objective was to integrate the information provided by several independent public registries on CH into a common 2D mapping and reporting platform and to enrich it with additional data provided by other built environment agencies. An estimation of the expected cost for the structural evaluation by experts of the city’s CH assets was also implemented for SDG’s Indicator 11.4.1. The methodology involved stakeholder identification, data collection and pre-processing, field verification and documentation, calculation of Indicator 11.4.1, and the actual coding process. The application can be found online, providing useful insights and statistical information on the city’s heritage in a dashboard format. The key challenges included the lack of updated data, the existence of several individual registries, and the need for regular field inspection due to the rapidly changing urban fabric.