Patient Data Research Articles

Risk of death in patients with early cardiovascular complications following ischaemic stroke: an individual patient data pooled analysis from the VISTA database

Abstract Background Early cardiovascular complications may occur in patients following ischaemic stroke (stroke-heart-syndrome, SHS). We investigated the risk of death in patients with SHS, and across the SHS manifestations. Methods Data were sought from the Virtual International Stroke Trials Archive (VISTA), an international repository of clinical trials data. We reviewed relevant adverse events and classified patients into two cohorts based on the incidence of cardiovascular complications within 30 days post-stroke (acute coronary syndrome [ACS] including acute myocardial injury, heart failure [HF]/left ventricular [LV] dysfunction, atrial fibrillation [AF]/atrial flutter [AFL], other arrhythmia/ ECG abnormalities, and cardio-respiratory arrest [CRA]), i.e. those with SHS and those with stroke-only (without SHS). Using Cox proportional hazards model, we evaluated the 90-day risk of death trajectories associated with SHS. We also compared the 90-day risk of death across SHS manifestations excluding CRA, using multivariate analysis. Results Among 15,054 patients with ischaemic stroke (mean age 72±11 years; 55% male), 1,787 (11.8%; 95% confidence interval [CI]: 11.3-12.3, Table 1) developed SHS, with incidence peaked at day 3 post-stroke (interquartile range: 2-5). Other arrhythmia/ECG abnormalities were the most prevalent (6.5%; 95%CI: 6.1-6.9), followed by AF/AFL, HF/LV dysfunction, ACS/acute myocardial injury, and CRA. Patients who developed SHS between 10-30 days post-stroke had significantly higher risks of death compared to those with SHS within the first 0-3 days (adjusted hazard ratio [HR]: 2.13 [1.57-2.89]). In the multivariate-adjusted analysis which includes 13 background factors such as age and sex, SHS manifested as ACS/acute myocardial injury and HF/LV dysfunction were associated with highest risk of death within 90-day post-stroke (adjusted hazard ratio [HR] [95% CI], HR 2.13 (1.08-4.18) and HR 2.28 (1.42-3.66), respectively, Figure 1). Conclusions Stroke-heart syndrome is associated with high risk of death, with greater risk observed with the delayed SHS onset. Among the SHS manifestations, ACS/acute myocardial injury and HF/LV dysfunction were associated with the highest risk of death post-stroke.Figure 1

Lower early mortality after acute pulmonary embolism in patients with WHO-defined obesity: results from a post-hoc cohort analysis with external validation

Abstract Background The relationship between obesity, defined as a Body Mass Index (BMI) ≥ 30 kg/m2, and mortality in venous thrombo-embolism remains controversial. Objectives To compare early outcomes after pulmonary embolism (PE) between obese patients, and non-obese, non-underweight patients. Methods We performed a post-hoc analysis based on prospectively recorded individual patient data from the multicenter BFC-FRANCE registry. We compared rates of outcomes using multivariable logistic regression or a Cox model for 30-day and 6-month outcomes respectively. We assessed the incremental value of adding BMI information on top of the 30-day European Society of Cardiology (ESC) prognostic algorithm. Results A total of 2,390 patients with a BMI ≥ 18.5 kg/m2 (mean age, 66.9±16.8 years; 1,188 men [49.7%]) were included, 686 patients [28.7%] in the obese group, and 1,704 patients [71.3%] in the non-obese group. Mortality rates were significantly lower in obese patients versus non-obese patients at 30 days (OR, 0.59; 95% CI, 0.35-0.98), and 6 months. Cox-model-derived adjusted curves for all-cause death at 6 months, with hazard ratios (HRs) between obese and non-obese patients, after acute PE, are shown in the Figure. Rates of secondary non-fatal outcomes (including recurrent VTE) did not differ between groups. The addition of the obesity information on top of the ESC prognostic model improved global model fit, and discriminatory and calibration capacities, yielding significant reclassification based on the observed mortality rates with the ESC model as reference. Findings were confirmed in an external validation using 35,796 PE patients from the RIETE registry. Conclusion We present evidence indicating lower early and mid-term mortality after PE in patients classified as obese based on BMI, compared with non-obese, non-underweight patients. BMI should likely be incorporated into algorithms or scoring systems for predicting early mortality following PE.

Digital health and some thematic shifts in bioethics in academic publications after the pandemic

Abstract Background Digital health (DH) presents vast opportunities, but also dynamic shift in viewpoints, which is reflected in academic works. Since DH concerns patient data, a major issue is complying with the bioethical principles, and ensuring patients’ privacy and security. The aim of this research is to explore avenues, in which ethics and privacy are reflected in DH, especially in terms of research, legislation, artificial intelligence (AI) and blockchain applications in medicine and public health. Methods Literature search for full text publications in PubMed, Scopus and ScienceDirect focused on ethics and privacy in DH in the period 2019-2023 found 325 articles in English authored by academicians. Six thematic categories were identified: ethics (fundamental principles; ethics in using novel DH tools); privacy (in use of apps/wearables; as a challenge/barrier to DH interventions/technologies); ethics/privacy in: DH research (data gathering, use, sharing); policy and legislation (for DH applications); AI and blockchains (in medicine and public health). Results Articles were from 32 countries: USA-24.3%; UK-13.5%; Germany-7.7%; Canada-7.4%; Australia-6.7%; Netherlands, Italy, Switzerland-4.7% each; China, India-3.0% each and 20.3% from 22 other countries; 86.5% in medical/DH journals. Only 10.2% were published in 2019; 17.8% - 2020; 19.4% - 2021; 27.1% - 2022 and 25.5% - 2023. Privacy was discussed in 26.5%; ethics-23.7%; research-19.4%; AI-15.1%; policy and legislation-9.5% and blockchains-5.5% of the articles. The publication activity of all six categories increased after the unfold of the pandemic in 2020. In 2023 publications for ethics (13%), privacy (5.8%) and research (6.3%) decreased; for legislation (12.9%), AI (14.3%) and blockchains in healthcare (16.8%) increased. Conclusions The pandemic and its aftermath present a change in academic interest. Traditional ethical fundaments in DH slightly lose position in favor of top notch technologies and apt legislation. Key messages • The pandemic and modern technologies enforced new digital health applications. Further research will reveal the needs of AI and blockchain applications respecting privacy and the ethical principles. • Besides existing legislation and regulations, in order to catch up with the dynamic developments in DH, the legal base needs to become more flexible. Technologies change, privacy importance does not.

The impact of ethnicity on clinical outcomes in patients with heart failure and atrial fibrillation

Abstract Introduction Heart failure (HF) and atrial fibrillation (AF) are increasingly prevalent and associated with substantial morbidity (1,2). When HF and AF co-exist, the outcome profile is significantly worse than for either condition in isolation (3). Ethnic background is known to influence disease outcomes and treatment efficacy (4), but there is limited knowledge about how HF, AF and ethnicity interact on mortality and hospital admission. Purpose This study aims to address the null hypothesis of no difference in death and cardiovascular events comparing non-white and white patients with HF and concomitant AF, and assess the differential impact of ethnicity in patients with these conditions. Methods Individual patient data were obtained from 12 randomized controlled trials (RCTs) in patients with HF. RCT data were harmonised and meta-analysed to investigate the interaction of baseline heart rhythm (sinus rhythm versus atrial fibrillation) and ethnicity (non-white versus white). The primary outcome of all-cause mortality and secondary outcome of cardiovascular (CV)-related mortality were analysed using adjusted Cox regression models and propensity-score matching. Results 16,713 HF patients were included of which 1,899 (11%) were non-white, median age 66 (interquartile range [IQR] 57-73) and 4,703 (28%) women. During median follow-up of 1.4 years (IQR 0.8-2.3), death occurred in 394 (21%) non-white patients and 2,142 (15%) white patients; adjusted hazard ratio (HR) 1.36, 95% CI 1.20-1.54; p<0.001. The impact of ethnicity was greater in patients with coexisting HF and AF (non-white versus white HR 2.05; 95% CI 1.55-2.70; p<0.001), than in those with HF in sinus rhythm (non-white versus white HR 1.24; 95% CI 1.08-1.41; p=0.002). A significant interaction was evident between ethnicity and rhythm status for all-cause mortality (p=0.003), with similar findings using propensity-score matching (Figure 1). An exploratory analysis of beta-blockers versus placebo demonstrated divergent efficacy with respect to mortality for non-white (HR 1.00, 95% CI 0.79-1.27; p=0.99) and white patients (HR 0.70, 95% CI 0.63-0.78; p<0.001); p-interaction=0.008 (Figure 2). For cardiovascular hospitalisation, the non-white cohort also had higher rates, with greatest disparity in patients with combined HF and AF (non-white versus white HR 1.80, 95% CI 1.63-2.37; p<0.001) compared to HF in sinus rhythm (non-white versus white HR 1.23, 95% CI 1.11-1.36; p<0.001); p-interaction=0.012 in propensity-matching. Conclusion Non-white patients with HF suffer from substantially higher rates of death and hospitalisation than corresponding white patients, with the presence of comorbid AF significantly worsening this ethnicity-related disparity. In those with HF in sinus rhythm, beta blockers were found to be significantly less effective at reducing mortality in non-white patients.Figure 1 Figure 2

Cardiovascular outcomes in acute coronary syndrome patients with cancer: a statewide population-linkage study

Abstract Background Advances in oncology treatment have led to an increase in the number of cancer survivors. As a result, the number of acute coronary syndrome (ACS) patients with cancer is increasing and the outcomes of these patients are not well described. Purpose This study aimed to examine revascularisation strategies and all cause and cardiovascular death in ACS patients with and without cancer. Methods Hospitalised patients in New South Wales, Australia, from July 2006 to December 2020 with a primary diagnosis of ACS were identified from the statewide Admitted Patient Data Collection database. Patients were stratified into those with or without a history of cancer (non-melanoma skin cancers excluded). Propensity scoring matching (PSM) for the variables of age, gender and cardiovascular risk factors was performed with each cancer patient matched to two non-cancer patients. Cox proportional hazards regression model was used to evaluate the hazard ratios of all-cause mortality between individuals with a history of cancer in the 5 years preceding their hospital admission, and those without. Additionally, Fine-Gray competing risk analysis was performed for cardiovascular death with non-cardiovascular mortality as the competing event. Results During the study period, 193,848 patients were admitted with ACS. 12,543 patients had a history of cancer and were PSM to 25,085 patients without cancer. The median age was 75 (Q1-Q3 67-82) years with 69% male and hypertension the most prevalent risk factor (55%) (Table 1). ACS patients with cancer had clinically comparable, rates of percutaneous revascularisation (36 vs 37%, p=0.07) and coronary artery bypass grafting (11 vs 13%, p<0.001) with ACS patients without cancer. During a median follow up of 8.6 (95% CI 8.4-8.8) years, ACS patients with a history of cancer had a significant higher risk of all cause death (HR 1.42 95% CI [1.38-1.46]) and cardiovascular death (SHR 1.14 [1.11-1.18]). The cumulative incidence function demonstrated cardiovascular death in those with a cancer history occurred in 20% by approximately 1.4 years follow up compared to 2.5 years in those without cancer (Figure 1). Conclusion ACS patients with cancer have a higher risk of all cause death and cardiovascular death compared to those without a history of cancer with clinically comparably revascularisation rates. Further studies to determine the cause of increased cardiovascular death in ACS patients with cancer are required.Figure 1

Evidence for AI in healthcare

Abstract In healthcare, novel AI solutions are being created to address some of the biggest challenges in the prognosis, diagnosis, and treatment of disease, as well as clinician workflows and service improvement. The emergence of generative AI presents a compelling opportunity to revolutionize diagnostics, treatment planning, medical research, and patient engagement. The hypothesized uses of generative AI are broad, ranging from medical education, providing information to patients, generating synthetic patient data for the validation of AI tools, to the analysis of continuous data from wearables to detect early signs of disease. Adoption of digital solutions and AI in healthcare is slower than in other industries. The majority of clinicians don’t have direct experience with AI technologies, only a quarter have recommended a digital therapeutic, and less than a fifth have prescribed one. Safety, quality and confidence can be built through appropriate governance, testing, careful implementation and appropriate clinical use. There is a responsibility for AI developers, health systems and providers, as well as regulators, to create clear expectations for evidence and solutions. AI solutions require unique model evidence (evidence for the underlying algorithm) and solution evidence (evidence for the product in which the algorithm is embedded). Models will need validation first on internal and then on external datasets (internal and external validation). This will indicate how accurate and reliable the model is, but will be limited to the dataset that a client has access to. By examining ways to evaluate evidence for AI in healthcare, we can develop a better understanding of how best to use AI to its most appropriate advantages.

Prognostic value of echocardiographic parameters in predicting major adverse cardiac events one year following st-segment elevation myocardial infarction

Abstract Approximately 50% of patients who survive an ST-segment elevation myocardial infarction (STEMI) experience irreversible damage to the heart muscle, which increases long-term risk of developing heart failure. Moreover, cardiovascular mortality constitutes half of all deaths. This research aimed to determine the best combination of patient data collected during a one-year follow-up period to enhance the accuracy of outcome predictions for patients with STEMI. Methods In this study, 246 patients who experienced STEMI and older than 18 were prospectively monitored. All participants received primary percutaneous coronary intervention treatment. The study's primary outcomes, observed over the year following hospital discharge, included cardiovascular death, recurrent myocardial infarction, newly diagnosed heart failure, and arrhythmias or conduction issues. Of the patients, 18% of patients reached the primary outcomes. Two-dimensional transthoracic echocardiography was conducted 3-5 days after the initial event to assess left ventricular global longitudinal strain (GLS) using 2D speckle-tracking echocardiography on the apical four-chamber, two-chamber, and long-axis views. Left ventricular mechanical dispersion was calculated as the standard deviation of the time from the Q/R wave onset to peak GLS across 17 segments of the left ventricle. Results The findings revealed that the overall GLS was -10.70 (-12.61 to -8.40), but in patients who met the study's primary outcomes, GLS was significantly lower at -7.83 (-8.70 to -6.90), compared to -12.02 (-13.11 to -10.17) in those who did not, with a significant difference (p=0.00001). Mechanical dispersion was 40.24±26.56 in the entire cohort but significantly higher at 65.62±22.57 in the outcome group compared to 25.79±14.81 in the non-outcome group (p=0.00001). The ejection fraction averaged 48.97±7.69% across all patients, 47.22±7.16% in those with outcomes, and 49.92±7.84% in those without, with a statistically significant difference (p=0.039). Cox regression analysis identified key predictors of major adverse cardiac events (MACE) within one year after STEMI, including global longitudinal strain (p=0.0006), left atrium volume index (p=0.0448), left ventricular end-diastolic diameter (p=0.045), and mechanical dispersion (p<0.0001). A multivariate logistic regression analysis demonstrated that a combination of GLS, left ventricular end-systolic volume (LVESV), mechanical dispersion, and ejection fraction could predict 1-year MACE with high accuracy (χ2 = 85.18; p<0.0001, AUC=0.961 [0.903-0.989]). In conclusion, this study identifies crucial metrics, such as global longitudinal strain, left ventricular end-systolic volume, mechanical dispersion, and ejection fraction, as predictive factors for MACE one year following STEMI. This predictive model is straightforward, reproducible, and can be easily implemented in cardiology practice.

VA-ECMO-related complications seem not causally related to mortality in patients with acute myocardial infarction complicated by cardiogenic shock

Abstract Background Recently published randomized controlled trials and an individual patient data meta-analysis showed that the routine use of veno-arterial extracorporeal membrane oxygenation (VA-ECMO) did not reduce mortality compared to medical therapy alone in patients with acute myocardial infarction complicated by cardiogenic shock (AMI-CS). Purpose The aim of this study was to investigate as to whether possible VA-ECMO-related medical complications might be associated with an increased mortality risk, which might explain the observed equipoise among mortality outcomes between VA-ECMO and usual care in patients with AMI-CS. Methods Using data from the randomized ECLS-SHOCK trial (Extracorporeal Life Support in Infarct-Related Cardiogenic Shock), which compared routine VA-ECMO use versus medical therapy alone. Causal mediation analyses (primary analysis; VA-ECMO-related complications used as a mediator), logistic regression and causes of death analyses (secondary analyses) were utilised to test potential effects of possible VA-ECMO-related complications on 30-day mortality. Analyses were adjusted for risk factors of bleeding and vascular complications. Results In total, 417 patients were included in this analysis. Moderate and severe bleeding as well as peripheral vascular complications requiring intervention were identified as possible VA-ECMO-related complications. A total of 88 patients (21.1%) developed possible VA-ECMO-related complications, which mostly occurred within five days after randomisation and more frequently in the VA-ECMO group than in the control group: 49 (23.4%) vs. 20 (9.6%) patients, p<0.001 for bleeding and 23 (11.0%) vs. 8 (3.8%) patients, p=0.01 for peripheral vascular complications, respectively. Causal mediation analyses revealed no significant mediation effect of possible VA-ECMO-related complications on 30-day mortality (p=n.s.; Figure). Logistic regression and causes of death analyses of possible VA-ECMO-related complications confirmed no significant associations between the incidence of VA-ECMO-related complications and mortality (p=n.s.). Conclusion VA-ECMO use in patients with AMI-CS was associated with increased incidences of moderate to severe bleeding as well as vascular complications warranting intervention. In this causal mediation analysis, VA-ECMO-related complications do not appear to mediate an increased mortality risk.

Chronic kidney disease in patients with heart failure in German primary care practices: the InspeCKD study

Abstract Background Chronic kidney disease (CKD) is, alongside type 2 diabetes mellitus, the most common comorbidity in patients with heart failure (HF), with a prevalence of 40-50%. HF patients should therefore be screened, monitored, and treated according to international Kidney Disease Improving Global Outcomes (KDIGO) conclusions for early identification and intervention of CKD, with annual determination of eGFR to monitor kidney function and annual urine albumin-to-creatinine ratio (UACR) testing as appropriate. Early diagnosis is important as therapeutic options can be offered to slow the progression of CKD and reduce the risk of cardiovascular complications and mortality. Purpose Currently, there are only limited data available regarding screening, diagnosis and treatment of CKD for HF patients in German primary care physician (PCP) practices. Therefore, this analysis of the InspeCKD study was designed to examine the frequency of use of CKD-specific laboratory diagnostics in HF patients in routine PCP care. Methods In the current data analysis, fully anonymized patient data from German PCP practices were evaluated. In accordance with the screening recommendation of the KDIGO guideline, adult patients with diabetes, hypertension and/or cardiovascular disease, including HF, with at least one year of observation period were included in the analysis. Results The overall patient cohort comprised 448.837 patients from 1.244 German PCP practices. Of the 7.9% HF patients, 6.1% participated in a disease management program (DMP) for coronary heart disease. The average age of HF patients was 74 years. The prevalence of diagnosed CKD in HF patients during the observational period (mean: 1.7 years) was 26.8%. Within the subgroup of patients with HF, serum creatinine to estimate GFR was measured at least once in 48.5% of patients. Urine dipstick testing for albuminuria was performed in 7.4% of patients and UACR was measured at least once in 0.3% of patients. In patients with at least one documented measurement, eGFR and UACR were checked 2.2 and 0.9 times, respectively, per patient and year. Of the HF patients with laboratory-confirmed CKD according to KDIGO criteria, 81.8% were not diagnosed with CKD. Conclusion The analysis shows that, given the high prevalence of CKD in HF patients and the current KDIGO recommendations for early identification and intervention of CKD, a substantial proportion of HF patients in German PCP practices did not receive adequate laboratory diagnostics for early diagnosis of CKD. In addition, the majority of patients with laboratory-confirmed CKD were not diagnosed. Early detection of CKD and the use of evidence-based therapeutic options can significantly reduce the risk of CKD progression. In conclusion, there is an urgent need to raise awareness among PCPs about secondary and tertiary prevention of CKD in patients with HF and to include coordinated CKD screening recommendations in HF guidelines.

The importance of heart rate on left ventricular global longitudinal strain in echocardiography: novel insights facilitated by automated deep learning methods

Abstract Background Heart rate (HR) is assumed to influence echocardiographic measurements. Consequently, HR needs to be considered for accurate assessment of cardiac function. Despite left ventricular global longitudinal strain (LV GLS) being recognised as a valuable measure of cardiac function, its normal reference ranges remain confined to patients within a limited range of resting HR. So far, time-consuming manual analyses and high measurement variability have precluded comprehensive assessment of the interplay between HR and LV GLS. Using fully automated GLS measurements provided by novel deep learning (DL) methods, it is now possible to evaluate changes in LV GLS across a wide range of HR. Purpose To improve echocardiographic assessments across a wide range of HR intervals, we aimed to study the influence of controlled HR variations on LV GLS using programmable pacemakers and fully automated DL measurements. Methods Participants with sinus node dysfunction and implanted programmable pacemakers were enrolled. Echocardiographic acquisitions were obtained at baseline HR and during stepwise increments by 10 beats/min, targeting a range from 50 to 140 beats/min. Recordings of cine-loops including 10 cardiac cycles were obtained for the four-chamber, two-chamber, and apical long axis view. LV GLS was measured for all cycles using a fully automated DL method developed by our group, and presented as mean of the three standard apical views for all patients at each HR level. A linear mixed model was used for construction of an overall trendline based on the estimated fixed effects across all 50 patients, stating patient number as the random effect. The correlation between HR and LV GLS was evaluated by a likelihood ratio test. Results Fifty patients were included, 52% females, mean age 68 years (27 to 88 years). All patients had preserved atrioventricular conduction. Resting HR ranged from 50 to 87 beats/min (mean 64 beats/min). At baseline echocardiographic acquisition, most participants were within normal ranges for LV volumes, systolic and diastolic function. A total of about 13,000 cardiac cycles were automatically analysed by the DL method. The overall trend demonstrated a linear relation between increasing HR and decreasing absolute LV GLS (Figure 1). Across HR increments from 50 to 140 beats/min, there was a 32% (22%–42%) reduction in absolute LV GLS, declining from -17.1% to -11.7%. The strong correlation between changes in HR and LV GLS was further supported by a highly significant likelihood ratio test (p-value <0.001). Conclusions There was a reduction of absolute LV GLS with increasing HR. These findings emphasize the importance of integrating HR considerations into clinical echocardiography practice. Moreover, the results highlight that DL methods can unlock substantially richer insights from individual patient data, offering a new pathway towards more personalized and precise cardiac diagnostics.Figure 1

Transcatheter versus surgical aortic valve replacement in lower-risk patients: an updated meta-analysis of randomised controlled trials

Abstract Background Long-term clinical outcomes after transcatheter [TAVI] and surgical [SAVR] aortic valve replacement are of utmost importance in lower-risk patients. Event rates are diminished in lower-risk patients, therefore individual trials may be separately underpowered to assess low-frequency outcomes. New randomised data has emerged, including 4, 5 and 10-year follow-up from the Evolut Low Risk, PARTNER-3 and NOTION trials, respectively. Purpose To perform a systematic review and meta-analysis of randomised trials studying long-term outcomes in patients with severe aortic stenosis at low-surgical risk undergoing TAVI versus SAVR. Methods Trials randomising patients at lower surgical risk in trials comparing TAVI with SAVR were systematically identified. The prespecified primary endpoint was all-cause death. Key secondary endpoints included stroke, death or disabling stroke, cardiac death, new permanent pacemaker, aortic valve reintervention, and new-onset atrial fibrillation. Reconstructed individual patient data [IPD] was obtained via digitisation of Kaplan-Meier survival curves. Using these data, Cox proportional hazards regression was performed using frailty models on the primary outcome of all-cause death. To assess total lifetime lost, the restricted mean survival time [RMST] was calculated. Conventional pairwise random effects meta-analyses were also performed. This analysis was prospectively registered [CRD42023484976]. Results 4 eligible trials contributed data from 3557 patients with a weighted mean follow-up duration of approximately 4 years (47.6 months). In the reconstructed IPD pooled analyses, there was no significant difference in the hazard of death [hazard ratio 1.15, 95% CI 0.93 to 1.41, p=0.20] or RMST [Δ-0.8 months, p=0.09] (Figure 1). In pairwise meta-analyses, there was no difference between TAVI and SAVR in all-cause death [relative risk (RR) 0.94, 95% confidence interval (CI) 0.78 to 1.13, p=0.51] (Figure 2). There was also no significant difference for stroke [RR 0.94, 95% CI 0.59 to 1.50, p=0.80], or a composite of death or disabling stroke [RR 0.92, 95% CI 0.70 to 1.22, p=0.57]. There was an increase in new pacemaker implantation with TAVI [RR 2.15, 95% CI 1.50 to 3.07, p<0.001], and a reduction in new AF with TAVI [RR 0.43, 95% CI 0.26 to 0.70, p<0.001]; there was no significant difference in any other endpoint (Figure 2). Conclusions The present analysis demonstrates no difference in major clinical outcomes with TAVI or SAVR in lower-risk patients at 4-years follow-up. These results were consistent when analysed using pooled reconstructed IPD or conventional meta-analyses and there were no differences in RMST. These data are informative for lower-risk patients and clinicians, but longer-term follow-up is required to confirm whether equivalence is maintained.Kaplan-Meier plot comparing TAVI vs SAVRForest plot for key clinical endpoints

Association Between Delay to First Shock and Successful First-Shock Ventricular Fibrillation Termination in Patients With Witnessed Out-of-Hospital Cardiac Arrest.

In patients with out-of-hospital cardiac arrest who present with an initial shockable rhythm, a longer delay to the first shock decreases the probability of survival, often attributed to cerebral damage. The mechanisms of this decreased survival have not yet been elucidated. Estimating the probability of successful defibrillation and other factors in relation to the time to first shock may guide prehospital care systems to implement policies that improve patient survival by decreasing time to first shock. Patients with a witnessed out-of-hospital cardiac arrest and ventricular fibrillation (VF) as an initial rhythm were included using the prospective ARREST registry (Amsterdam Resuscitation Studies). Patient and resuscitation data, including time-synchronized automated external defibrillator and manual defibrillator data, were analyzed to determine VF termination at 5 seconds after the first shock. Delay to first shock was defined as the time from initial emergency call until the first shock by any defibrillator. Outcomes were the proportion of VF termination, return of organized rhythm, transportation with return of spontaneous circulation, and survival to discharge, all in relation to the delay to first shock. A Poisson regression model with robust standard errors was used to estimate the association between delay to first shock and outcomes. Among 3723 patients, the proportion of VF termination declined from 93% when the delay to first shock was <6 minutes to 75% when that delay was >16 minutes (Ptrend<0.001). Every additional minute in VF from emergency call was associated with 6% higher probability of failure to terminate VF (adjusted relative risk, 1.06 [95% CI, 1.04-1.07]), 4% lower probability of return of organized rhythm (adjusted relative risk, 0.96 [95% CI, 0.95-0.98]), and 6% lower probability of surviving to discharge (adjusted relative risk, 0.94 [95% CI, 0.93-0.95]). Every minute of delay to first shock was associated with a significantly lower proportion of VF termination and return of organized rhythm. This may explain the worse outcomes in patients with a long delay to defibrillation. Reducing the time interval from emergency call to first shock to ≤6 minutes could be considered a key performance indicator of the chain of survival.

Missing Risk Factor Prediction in Cardiovascular Disease Using a Blended Dataset and Optimizing Classification With a Stacking Algorithm

ABSTRACTMachine learning is important in the treatment of heart disease because it is capable of analyzing large amounts of patient data, such as medical records, imaging tests, and genetic information, in order to identify patterns and predict the risk of developing heart disease. However, most ML algorithms require more accurate data in order to build an accurate prediction model and do not tolerate missing values. Handling missing risk factors is critical during dataset preprocessing and becomes more difficult when the risk factors are completely missing. Removing this completely missing feature may result in the loss of critical information, but there are no readily available imputation methods, which presents a significant challenge. To overcome this difficulty, in this study, we take an attempt to impute using statistical multiple linear regression and Huber regression (HR) methods using four blended datasets (Statlog, Cleveland, Hungarian, and Switzerland) sourced from the UCI ML repository. The entire dataset comprises 14 attributes, including one target variable; however, in the Switzerland dataset, one feature value (“serum cholesterol”) is entirely missing. Missing “serum cholesterol” is recognized as a predisposing factor including “chest pain,” “supreme heartbeat rate,” “type of defect,” “exercise induced ST stress related to rest,” and “exercise generated angina” in the proposed imputation methods. We also proposed applying the majority voting ensemble technique in an individual's and integrated dataset using ML algorithms as part of the risk factor identification strategy. The results show that our proposed stacked algorithm for the combined dataset with the ensemble features significantly improved accuracy by 93.47%, and an AUC score of 94.50% demonstrated more accurate and early prediction than the previous research and also provided the model's diversity, resilience, generalization, and adaptability to varied datasets.

Machine learning discrimination of Gleason scores below GG3 and above GG4 for HSPC patients diagnosis

This study aims to develop machine learning (ML)-assisted models for analyzing datasets related to Gleason scores in prostate cancer, conducting statistical analyses on the datasets, and identifying meaningful features. We retrospectively collected data from 717 hormone-sensitive prostate cancer (HSPC) patients at Yunnan Cancer Hospital. Of these, data from 526 patients were used for modeling. Seven auxiliary models were established using Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), Extreme gradient boosting tree (XGBoost), Adaptive Boosting (Adaboost), and artificial neural network (ANN) based on 21 clinical biochemical indicators and features. Evaluation metrics included accuracy (ACC), precision (PRE), specificity (SPE), sensitivity (SEN) or regression rate(Recall), and f1 score. Evaluation metrics for the models primarily included ACC, PRE, SPE, SEN or Recall, f1 score, and area under the curve(AUC). Evaluation metrics were visualized using confusion matrices and ROC curves. Among the ensemble learning methods, RF, XGBoost, and Adaboost performed the best. RF achieved a training dataset score of 0.769 (95% CI: 0.759—0.835) and a testing dataset score of 0.755 (95% CI: 0.660—0.760) (AUC: 0.786, 95%CI: 0.722—0.803), while XGBoost achieved a training dataset score of 0.755 (95% CI: 95%CI: 0.711—0.809) and a testing dataset score of 0.745 (95% CI: 0.660—0.764) (AUC: 0.777, 95% CI: 0.726—0.798). Adaboost scored 0.789 on the training dataset (95% CI: 0.782—0.857) and 0.774 on the testing dataset (95% CI: 0.651—0.774) (AUC: 0.799, 95% CI: 0.703—0.802). In terms of feature importance (FI) in ensemble learning, Bone metastases at first visit, prostatic volume, age, and T1-T2 have significant proportions in RF’s FI. fPSA, TPSA, and tumor burden have significant proportions in Adaboost’s FI, while f/TPSA, LDH, and testosterone have the highest proportions in XGBoost. Our findings indicate that ensemble learning methods demonstrate good performance in classifying HSPC patient data, with TNM staging and fPSA being important classification indicators. These discoveries provide valuable references for distinguishing different Gleason scores, facilitating more accurate patient assessments and personalized treatment plans.

Knee joint dislocations-Current epidemiology and treatment in Germany.

Knee joint dislocations, though rare, present significant challenges due to potential complications like vascular and nerve damage, and are known to be often linked to sports injuries, accidents and obesity. This study aims to analyse the epidemiology, incidence and treatment approaches for knee dislocations in Germany from 2019 to 2022. This retrospective cohort study utilized data from the German Institute for the Hospital Remuneration System to examine knee dislocation cases across German medical institutions. Patient data coded under International Statistical Classification of Diseases and Related Health Problems 10 for 'knee dislocation' enabled detailed analysis by age, sex and surgical procedures categorized by operation and procedure codes. The Patient Clinical Complexity Level (PCCL) assessed complication severity. Analysis of 1643 knee dislocation cases revealed an incidence rate of 0.44-0.54 per 100,000 inhabitants annually. During the years 2020 and 2021, there were fewer cases of knee dislocations. Male patients comprised 50%-56% of cases, with an average hospital stay of 11 days. Most cases were PCCL 0 (62%-72%) and predominantly affected patients aged 18-29 years. Anterior tibial dislocations were common among classified cases. Injuries included ligament ruptures, meniscus lesions and grade I soft-tissue injuries. Patients with pre-existing knee prostheses constituted 0%-16% annually. Treatment involved closed reduction, external fixation and surgeries like capsuloligamentous reconstructions and arthroscopic procedures. Revision knee arthroplasty was required in 2%-9% of cases, with obesity rates up to 7%. This study provides valuable insights into the epidemiology, incidence and treatment of knee dislocations in Germany, with a focus on demographic risk factors, treatment complexities and the impact of obesity and knee prostheses. The findings emphasize the importance of specialized care in larger hospitals, comprehensive management of concomitant injuries and the need for improved coding accuracy. Future research should aim to refine treatment protocols. Level III.

Matching-adjusted indirect comparison of Acalabrutinib versus Ibrutinib in relapsed/refractory mantle cell lymphoma

ABSTRACT Objective In the absence of head-to-head clinical trials, matching-adjusted indirect comparison (MAIC) was used to compare 2 Bruton tyrosine kinase inhibitors (BTKis) approved for the treatment of relapsed/refractory (R/R) mantle cell lymphoma (MCL). This analysis compares the efficacy and safety of acalabrutinib versus ibrutinib using a more mature dataset than a previously published MAIC. Methods Individual patient data from 122 patients treated with acalabrutinib in a phase 2 study were weighted to match aggregate baseline characteristics of patients pooled from 3 separate trials of ibrutinib. Patients were matched on Eastern Cooperative Oncology Group performance status, simplified Mantle Cell Lymphoma International Prognostic Index, lactate dehydrogenase, prior lines of therapy, tumor burden, and blastoid histology. Outcomes assessed included progression-free survival (PFS), overall survival (OS), and adverse events. Results After matching, differences in PFS between acalabrutinib (median 17.8 months) and ibrutinib (median 12.8 months) were not statistically significant (hazard ratio [HR], 0.92; 95% confidence interval [CI], 0.74–1.15; P = 0.48). Similarly, after matching, OS differences between acalabrutinib (median 36.5 months) and ibrutinib (median 27.9 months) did not reach statistical significance (HR, 0.87; 95% CI, 0.64–1.17; P = 0.35). Acalabrutinib was associated with an improved safety profile compared with ibrutinib, with statistically significantly lower rates of grade ≥3 atrial fibrillation and thrombocytopenia. Conclusions This comparison of 2 BTKis used in the treatment of R/R MCL showed that PFS and OS risk was not statistically different between the treatments; however, acalabrutinib had an improved safety profile compared with ibrutinib.

Older women with non-ST-elevation acute coronary syndrome undergoing invasive or conservative management: an individual patient data meta-analysis

Abstract Aims Women and older patients are underrepresented in randomised controlled trials (RCTs) investigating treatment strategies following acute coronary syndrome. To evaluate the benefit of invasive versus conservative strategy of older women with non-ST-elevation acute coronary syndrome (NSTEACS). Methods This analysis from an individual patient data meta-analysis included six RCTs comparing an invasive management with a conservative management in older NSTEACS patients. The primary endpoint was the composite of all-cause mortality or myocardial infarction (MI). Secondary endpoints included all-cause mortality, cardiovascular death, MI, urgent revascularisation and stroke. Follow-up time was censored at 1 year. Results In total, 717 women (median age 84.0 [interquartile range 81.0-87.0] years) were included. The primary endpoint occurred in 21.0% in the invasive strategy versus 27.8% in the conservative strategy (hazard ratio [HR] 0.77, 95% confidence interval [CI] 0.52-1.13, P=0.160 using random effect) at 1-year follow-up. The invasive management was associated with reduced risk of MI (HR 0.49, 95% CI 0.32-0.73, P&amp;lt;0.001) and urgent revascularisation (HR 0.44, 95% CI 0.20-0.98, P=0.045). No significant differences were identified in the risk of all-cause mortality, cardiovascular death and stroke. Among males, there was no significant association between the treatment strategy and primary or secondary endpoints. Conclusions An invasive strategy compared to a conservative strategy did not reduce the composite outcome of all-cause mortality or MI in older NSTEACS women at 1-year follow-up. An invasive strategy reduced the individual risk of MI and urgent revascularisation. Our results support the beneficial role of the invasive strategy in older NSTEACS women. Registration This meta-analysis is registered with PROSPERO (CRD42023379819).

The Epidemiology of Pediatric Shoulder Dislocations in the United States.

Shoulder dislocations are the most common of all major joint dislocations and are a frequent cause of emergency department (ED) presentations. While the epidemiology of shoulder dislocations has been characterized in adults, it has not yet been done for children in the United States. The primary goal of this study was to calculate the incidence rate of pediatric shoulder dislocations in the United States. Secondary goals were to characterize the demographic information of children with shoulder dislocations as well as trends in management. The US Collaborative Network in TriNetX, a network of clinical data repositories containing patient data from over 100,000,000 unique individuals within the United States, was queried for patients younger than 18 years old diagnosed with shoulder dislocation from 2014 to 2024 using ICD codes. The demographic details were then extracted from the data set, and treatment approaches were determined by CPT coding. Over the past 10 years, there were 16,460 pediatric and adolescent patients diagnosed with a first-time shoulder dislocation in either an ED or ambulatory clinic. The overall incidence rate was 60.31 per 100,000 patients, while the incidence rate was 116.61 per 100,000 patients presenting in the ED and 33.95 per 100,000 patients presenting in ambulatory clinics that were under the age of 18 years. Most patients were male (73%), White (59%), and came from the southeastern region of the United States (36%). The most common treatment was a closed reduction (25%), followed by arthroscopic surgery (17%). The number of annual shoulder dislocations has increased over the past decade without an increase in the number of patients undergoing surgical treatment. There remains a high incidence rate of shoulder dislocations in the pediatric population of the United States. Despite evidence that early surgical treatment of shoulder dislocations offers improved outcomes, nonoperative treatment continues to be the most common modality without any increase in the percentage of patients undergoing surgical treatment. Level IV-descriptive epidemiological analysis.

Percutaneous endoscopic gastrostomy helped to normalise feeding problems and gastrointestinal symptoms in Silver-Russell syndrome.

This study evaluated feeding problems and gastrointestinal symptoms in children with Silver-Russell syndrome (SRS), which is a rare epigenetic disorder. It also compared the symptoms experienced during different feeding methods, including percutaneous endoscopic gastrostomy (PEG). The national expert team for children with SRS at Queen Silvia Children's Hospital, Gothenburg, studied 46 referrals (63% male) who were born with SRS in Sweden from 1984 to 2018. Patient data were extracted from the Paediatric National Growth Hormone Registry. The medical records covered a median of 68% of the time of the patients' childhood, with a median follow-up of 9 years. Their symptoms were most prevalent during infancy and decreased when they were toddlers. Feeding problems and gastrointestinal symptoms were reported in 91% of the 46 patients, with vomiting in 57% and constipation in 46%. There were 19 children who relied on enteral feeding for their nutrition and 13 of those received PEG. Their body mass index (BMI) increased significantly 2 years after PEG started (p = 0.005). Feeding problems and gastrointestinal symptoms were very common in children with SRS, but partly disappeared during childhood. Providing treatment, such as PEG, normalised the BMIs of children with SRS and reduced their symptoms.

Development and Modeling of the Internet of Things Network for Patients IT Diagnostics

The work of a network that implements algorithms for IT diagnostics of neurological diseases based on the Internet of Things technology has been developed and modeled. The network includes a smartphone, a platform, a neural network, and applications. First, the voices of sick patients are entered from the smartphone to train the neural network, and then the examined patients for IT diagnostics. Data is transferred between the smartphone and the platform (ThingSpeak) via the MQTT protocol. The smartphone’s mobile application extracts the voice functions of the examined patients and records them on the Internet of Things network platform. Recognition is performed using the trained neural network. The structure and algorithm of the ThingSpeak platform are presented. IT diagnostics data are displayed in the application on the smartphone. The patient data used in the study are taken from the ADReSS 2020 Challenge program, which contains speech data of patients with Alzheimer’s disease and healthy people.