Zacharias Mandalenakis
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zacharias Mandalenakis has a total of 172 co-authors. Here are the top co-authors by number of shared publications:
Mikael Dellborg
Institute of Medicine Department of Molecular and Clinical Medicine Sahlgrenska AcademyUniversity of Gothenburg Sweden
22 shared publicationsAnnika Rosengren
Institute of Medicine, Department of Molecular and Clinical Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden
18 shared publicationsPeter Eriksson
Department of Molecular and Clinical Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg , Diagnosvägen 11, SE-416 50 Gothenburg
15 shared publications- Mikael Dellborg
Department of Molecular and Clinical Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg , Gothenburg
14 shared publications - Kok Wai Giang
University of Gothenburg
13 shared publications - Peter Eriksson
Sahlgrenska University Hospital
12 shared publications
Latest Publications
- Open Access
- 27 September 2026
- Published by Wiley in Acta Anaesthesiologica Scandinavica
- Vol. 70 (10) , e70346
A National Pediatric Cohort Study on In‐Hospital Cardiac Arrest in Sweden
Background: National data on pediatric in‐hospital cardiac arrest (pIHCA) are limited, and cohorts from highly specialized pediatric centers may not reflect the broader hospital population. We aimed to describe pIHCA reported across Swedish hospitals, with particular focus on hospitals outside the two national centers for highly specialized pediatric cardiac care. Methods: This retrospective observational registry study included patients aged 0–18 years with pIHCA reported to the Swedish Registry for Cardiopulmonary Resuscitation between January 1, 2005 and September 25, 2025. Cases from the two highly specialized centers were analyzed separately and as part of the complete national registry‐reported cohort. The primary outcome was 30‐day survival; secondary outcomes included return of spontaneous circulation (ROSC) and 1‐year survival. Results: Among 525 pIHCA events, 344 occurred outside the two highly specialized centers and 181 at these centers. In the broader Swedish hospital population, 31.7% of arrests occurred on hospital wards, 26.2% in emergency departments, and 22.7% in intensive care units. CPR was initiated immediately in 81.4% of events. Forty‐nine percent survived to 30 days, and 44% to 1 year. At the two highly specialized centers, infants, congenital heart disease, and arrests in intensive care units were substantially more common, and 30‐day survival was 70%. Conclusion: Patient characteristics, arrest location, and survival differed substantially between the broader Swedish hospital population and the two highly specialized centers. These findings highlight the heterogeneity of pIHCA across hospital settings and the importance of considering hospital type and case mix when interpreting pIHCA data. Editorial Comment: Pediatric patient in‐hospital cardiac arrest is uncommon, though they registered in a national database in Sweden. This analysis presents 20 years of experience and follow‐up for these cases describing associated factors.
- 27 September 2026
- Published by Wiley in Acta Anaesthesiologica Scandinavica
- Vol. 70 (10) , e70346
Open AccessA National Pediatric Cohort Study on In‐Hospital Cardiac Arrest in Sweden
Background: National data on pediatric in‐hospital cardiac arrest (pIHCA) are limited, and cohorts from highly specialized pediatric centers may not reflect the broader hospital population. We aimed to describe pIHCA reported across Swedish hospitals, with particular focus on hospitals outside the two national centers for highly specialized pediatric cardiac care. Methods: This retrospective observational registry study included patients aged 0–18 years with pIHCA reported to the Swedish Registry for Cardiopulmonary Resuscitation between January 1, 2005 and September 25, 2025. Cases from the two highly specialized centers were analyzed separately and as part of the complete national registry‐reported cohort. The primary outcome was 30‐day survival; secondary outcomes included return of spontaneous circulation (ROSC) and 1‐year survival. Results: Among 525 pIHCA events, 344 occurred outside the two highly specialized centers and 181 at these centers. In the broader Swedish hospital population, 31.7% of arrests occurred on hospital wards, 26.2% in emergency departments, and 22.7% in intensive care units. CPR was initiated immediately in 81.4% of events. Forty‐nine percent survived to 30 days, and 44% to 1 year. At the two highly specialized centers, infants, congenital heart disease, and arrests in intensive care units were substantially more common, and 30‐day survival was 70%. Conclusion: Patient characteristics, arrest location, and survival differed substantially between the broader Swedish hospital population and the two highly specialized centers. These findings highlight the heterogeneity of pIHCA across hospital settings and the importance of considering hospital type and case mix when interpreting pIHCA data. Editorial Comment: Pediatric patient in‐hospital cardiac arrest is uncommon, though they registered in a national database in Sweden. This analysis presents 20 years of experience and follow‐up for these cases describing associated factors.
- Open Access
- 21 April 2026
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 15 (8) , e045788
Long‐Term Outcomes of Antithrombotic Treatment in a Cohort of 864 Patients With Fontan Circulation
Background: An effective antithrombotic strategy for patients with univentricular heart is required. We determined the prevalence of antithrombotic treatment and survival in patients with Fontan circulation, with secondary outcomes mortality, cardiovascular disease, acute myocardial infarction, and ischemic stroke. Methods: We retrospectively included patients with univentricular heart born in Sweden between January 1, 1970, and December 31, 2017. Data from the National Patient Register, the Cause of Death Register, and the Prescribed Drug Register were used. Patients who underwent total cavopulmonary connection surgery, resulting in Fontan circulation, were included and divided into groups based on treatment: anticoagulation, antiplatelets, and no antithrombotic therapy. Results: In total, 864 patients (median [interquartile range] 14.4 [3.9–24.9] years follow‐up) were included, 511 of whom (59%) were men. Thirty‐six percent received no antithrombotic treatment, and the mortality rate during the follow‐up was 58%. Among the patients treated with antithrombotic treatment, 75% received antiplatelets, with a 4% mortality rate. The mortality incidence rate was 2.40 (95% CI, 1.49–3.67) for antithrombotic treatment and 49.42 (95% CI, 42.50–57.15) without. The hazard ratio for mortality was 22.32 (95% CI, 14.04–35.48) without antithrombotic treatment. The incidence rate of ischemic stroke was 2.45 (95% CI, 1.52–3.75) with antithrombotic treatment and 2.47 (95% CI, 1.13–4.68) without. The incidence rate of acute myocardial infarction was 0.69 (95% CI, 0.25–1.50) with antithrombotic treatment and 2.76 (95% CI, 1.32–5.07) without treatment. Conclusions: Treatment with antiplatelets was associated with lower mortality compared with no antithrombotic treatment.
- 21 April 2026
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 15 (8) , e045788
Open AccessLong‐Term Outcomes of Antithrombotic Treatment in a Cohort of 864 Patients With Fontan Circulation
Background: An effective antithrombotic strategy for patients with univentricular heart is required. We determined the prevalence of antithrombotic treatment and survival in patients with Fontan circulation, with secondary outcomes mortality, cardiovascular disease, acute myocardial infarction, and ischemic stroke. Methods: We retrospectively included patients with univentricular heart born in Sweden between January 1, 1970, and December 31, 2017. Data from the National Patient Register, the Cause of Death Register, and the Prescribed Drug Register were used. Patients who underwent total cavopulmonary connection surgery, resulting in Fontan circulation, were included and divided into groups based on treatment: anticoagulation, antiplatelets, and no antithrombotic therapy. Results: In total, 864 patients (median [interquartile range] 14.4 [3.9–24.9] years follow‐up) were included, 511 of whom (59%) were men. Thirty‐six percent received no antithrombotic treatment, and the mortality rate during the follow‐up was 58%. Among the patients treated with antithrombotic treatment, 75% received antiplatelets, with a 4% mortality rate. The mortality incidence rate was 2.40 (95% CI, 1.49–3.67) for antithrombotic treatment and 49.42 (95% CI, 42.50–57.15) without. The hazard ratio for mortality was 22.32 (95% CI, 14.04–35.48) without antithrombotic treatment. The incidence rate of ischemic stroke was 2.45 (95% CI, 1.52–3.75) with antithrombotic treatment and 2.47 (95% CI, 1.13–4.68) without. The incidence rate of acute myocardial infarction was 0.69 (95% CI, 0.25–1.50) with antithrombotic treatment and 2.76 (95% CI, 1.32–5.07) without treatment. Conclusions: Treatment with antiplatelets was associated with lower mortality compared with no antithrombotic treatment.
- Open Access
- 2 April 2026
- Published by Oxford University Press (OUP) in European Heart Journal - Valvular and Structural Heart Disease
- Vol. 2 (2)
The contemporary and historical burden of arrhythmias in patients with congenital heart disease
Background: Arrhythmias constitute a significant concern in patients with congenital heart disease (CHD) but temporal trends remain unknown. Aims: To describe the burden of arrhythmias and dissect the impact of era of birth, lesion complexity, and age in both children and adults with CHD born in Sweden. Methods and results: In a population-based cohort study, patients diagnosed with CHD (n = 63 431) born in 1970–2017 were identified in the Swedish National Patient Register. Each patient was matched by age and sex with 10 control subjects (n = 583 100) from the general population. Fine–Gray competing risk and Cox proportional-hazards regression were used; risk estimates were compared with matched controls. During median follow-up of 19 years (total range: 0–53 years), 1895 new events of supraventricular arrhythmias were registered, yielding a hazard ratio (HR) of 8.0 [95% confidence interval (CI) 7.6–8.5], along with 371 new ventricular arrhythmias (HR 13.0; 95% CI 11.2–15.2). Complex CHD was associated with higher risk, irrespective of type, than non-complex CHD, and patients born in 1990 or later displayed lower risk for both supraventricular and ventricular arrhythmias, irrespective of age and CHD complexity, than those born before 1990. Patients aged 18–39 years with non-complex and complex CHD born before 1990 displayed HRs of 6.9 (95% CI 6.1–7.8) and 23.9 (95% CI 19.4–29.5) for supraventricular arrhythmias, respectively, versus 3.2 (95% CI 2.7–3.9) and 13.9 (95% CI 9.8–19.7) in those born in 1990 or later. Conclusions: The risk of arrhythmias was generally high across the CHD population and increased with increasing age and lesion complexity. Importantly, patients born in 1990 or later demonstrated lower risk for both supraventricular and ventricular arrhythmias than those born before 1990.
- 2 April 2026
- Published by Oxford University Press (OUP) in European Heart Journal - Valvular and Structural Heart Disease
- Vol. 2 (2)
Open AccessThe contemporary and historical burden of arrhythmias in patients with congenital heart disease
Background: Arrhythmias constitute a significant concern in patients with congenital heart disease (CHD) but temporal trends remain unknown. Aims: To describe the burden of arrhythmias and dissect the impact of era of birth, lesion complexity, and age in both children and adults with CHD born in Sweden. Methods and results: In a population-based cohort study, patients diagnosed with CHD (n = 63 431) born in 1970–2017 were identified in the Swedish National Patient Register. Each patient was matched by age and sex with 10 control subjects (n = 583 100) from the general population. Fine–Gray competing risk and Cox proportional-hazards regression were used; risk estimates were compared with matched controls. During median follow-up of 19 years (total range: 0–53 years), 1895 new events of supraventricular arrhythmias were registered, yielding a hazard ratio (HR) of 8.0 [95% confidence interval (CI) 7.6–8.5], along with 371 new ventricular arrhythmias (HR 13.0; 95% CI 11.2–15.2). Complex CHD was associated with higher risk, irrespective of type, than non-complex CHD, and patients born in 1990 or later displayed lower risk for both supraventricular and ventricular arrhythmias, irrespective of age and CHD complexity, than those born before 1990. Patients aged 18–39 years with non-complex and complex CHD born before 1990 displayed HRs of 6.9 (95% CI 6.1–7.8) and 23.9 (95% CI 19.4–29.5) for supraventricular arrhythmias, respectively, versus 3.2 (95% CI 2.7–3.9) and 13.9 (95% CI 9.8–19.7) in those born in 1990 or later. Conclusions: The risk of arrhythmias was generally high across the CHD population and increased with increasing age and lesion complexity. Importantly, patients born in 1990 or later demonstrated lower risk for both supraventricular and ventricular arrhythmias than those born before 1990.
- Open Access
- 6 January 2026
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 15 (1) , e043652
Trends in Congenital Heart Disease Hospitalizations: What Nearly 50 Years of Longitudinal Data and Segmented Regression Reveal
Background: Understanding long‐term health care utilization patterns in congenital heart disease (CHD) is essential for optimizing care delivery. While previous studies often assume linear trends, such assumptions could provide flawed analysis if abrupt changes in treatment or care occur, causing nonlinearity in trends. This study aims to determine if and where significant nonlinearity exists in hospitalization trends within the CHD pediatric population. Methods: Using national Swedish registers, we conducted a retrospective cohort study using segmented regression analysis of pediatric CHD health care utilization from infancy through adolescence, stratified by disease complexity. Breakpoints were identified to assess changes in trend over time, and subgroup analyses compared segmented and linear regression models. Results: Segmented regression revealed distinct utilization patterns across age groups, with highest health care use in infancy followed by a progressive decline through childhood and adolescence. Contrary to prior research, utilization declined over time regardless of CHD complexity. Identified breakpoints within the general CHD pediatric population in 1978, 1985, 1992, and 2003 aligned with key developments in CHD care. Conclusions: These findings challenge prevailing assumptions of increasing health care use over time among the pediatric CHD population. Given the identified breakpoints, caution is warranted when conducting comparative analyses that span these periods, as doing so may obscure meaningful shifts in underlying trends.
- 6 January 2026
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 15 (1) , e043652
Open AccessTrends in Congenital Heart Disease Hospitalizations: What Nearly 50 Years of Longitudinal Data and Segmented Regression Reveal
Background: Understanding long‐term health care utilization patterns in congenital heart disease (CHD) is essential for optimizing care delivery. While previous studies often assume linear trends, such assumptions could provide flawed analysis if abrupt changes in treatment or care occur, causing nonlinearity in trends. This study aims to determine if and where significant nonlinearity exists in hospitalization trends within the CHD pediatric population. Methods: Using national Swedish registers, we conducted a retrospective cohort study using segmented regression analysis of pediatric CHD health care utilization from infancy through adolescence, stratified by disease complexity. Breakpoints were identified to assess changes in trend over time, and subgroup analyses compared segmented and linear regression models. Results: Segmented regression revealed distinct utilization patterns across age groups, with highest health care use in infancy followed by a progressive decline through childhood and adolescence. Contrary to prior research, utilization declined over time regardless of CHD complexity. Identified breakpoints within the general CHD pediatric population in 1978, 1985, 1992, and 2003 aligned with key developments in CHD care. Conclusions: These findings challenge prevailing assumptions of increasing health care use over time among the pediatric CHD population. Given the identified breakpoints, caution is warranted when conducting comparative analyses that span these periods, as doing so may obscure meaningful shifts in underlying trends.
- Open Access
- 25 October 2025
- Published by Elsevier in International Journal of Cardiology Congenital Heart Disease
- Vol. 22
Physical fitness in adults with isolated secundum atrial septal defects
- 25 October 2025
- Published by Elsevier in International Journal of Cardiology Congenital Heart Disease
- Vol. 22
Open AccessPhysical fitness in adults with isolated secundum atrial septal defects
- Open Access
- 16 June 2025
- Published by Elsevier in International Journal of Cardiology Congenital Heart Disease
- Vol. 21, 100601
Fatigue in adults with congenital heart disease aged over 40 years
- 16 June 2025
- Published by Elsevier in International Journal of Cardiology Congenital Heart Disease
- Vol. 21, 100601
Open AccessFatigue in adults with congenital heart disease aged over 40 years
- Open Access
Ischemic stroke in patients with atrial fibrillation and low cardiovascular risk profile: A nationwide, registry-based cohort study
Background Although there is substantial evidence supporting the prevention of ischemic stroke (IS) in high-risk patients with atrial fibrillation (AF), knowledge regarding AF patients with a low cardiovascular risk profile remains limited. Furthermore, the necessity of anticoagulant therapy in this population has been widely debated. Methods Data from Swedish health registers were utilized to identify all patients diagnosed with AF but without prior cardiovascular conditions between 1987 and 2018. The risk of IS was assessed using Cox regression models, with AF patients compared to age- and sex-matched controls without AF. Results The study included a total of 229,075 patients with AF and 455,541 matched controls without AF. The overall risk of IS was 2.3 times higher (95% confidence interval [CI] 2.2–2.3) in patients with AF compared to controls over a mean follow-up period of 9 years (range 3–32 years). Women with AF had a 4.4-fold increased risk of developing IS within the first year following diagnosis (hazard ratio [HR] 4.4, CI 4.2–4.7) compared to matched women without AF. Additionally, younger patients with AF (aged 35–49 years) exhibited the highest risk of IS within the first year after diagnosis (HR 8.3, 95% CI 4.0– 17.1). Conclusions This large, nationwide, register-based cohort study found that even in the absence of cardiovascular conditions, patients with atrial fibrillation had more than double the risk of ischemic stroke compared to matched controls. The risk was especially elevated in women and younger individuals, particularly within the first year after AF diagnosis. These findings underscore the urgent need to refine risk stratification and explore preventive strategies beyond traditional clinical risk factors.
Open AccessIschemic stroke in patients with atrial fibrillation and low cardiovascular risk profile: A nationwide, registry-based cohort study
Background Although there is substantial evidence supporting the prevention of ischemic stroke (IS) in high-risk patients with atrial fibrillation (AF), knowledge regarding AF patients with a low cardiovascular risk profile remains limited. Furthermore, the necessity of anticoagulant therapy in this population has been widely debated. Methods Data from Swedish health registers were utilized to identify all patients diagnosed with AF but without prior cardiovascular conditions between 1987 and 2018. The risk of IS was assessed using Cox regression models, with AF patients compared to age- and sex-matched controls without AF. Results The study included a total of 229,075 patients with AF and 455,541 matched controls without AF. The overall risk of IS was 2.3 times higher (95% confidence interval [CI] 2.2–2.3) in patients with AF compared to controls over a mean follow-up period of 9 years (range 3–32 years). Women with AF had a 4.4-fold increased risk of developing IS within the first year following diagnosis (hazard ratio [HR] 4.4, CI 4.2–4.7) compared to matched women without AF. Additionally, younger patients with AF (aged 35–49 years) exhibited the highest risk of IS within the first year after diagnosis (HR 8.3, 95% CI 4.0– 17.1). Conclusions This large, nationwide, register-based cohort study found that even in the absence of cardiovascular conditions, patients with atrial fibrillation had more than double the risk of ischemic stroke compared to matched controls. The risk was especially elevated in women and younger individuals, particularly within the first year after AF diagnosis. These findings underscore the urgent need to refine risk stratification and explore preventive strategies beyond traditional clinical risk factors.
- Open Access
- 4 November 2024
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 13 (21) , e037127
Morbidity Burden in Patients With Ebstein Anomaly: The Natural History
Background: The lifetime morbidity burden of patients with Ebstein anomaly (EA) has not been well described. Methods and Results: Through an extensive 2‐country register‐based collaboration, patients diagnosed with EA who were born between 1930 and 2017 were identified in Danish and Swedish nationwide medical registries. Each patient was matched by age and sex with 10 control subjects from the general population. Cox proportional‐hazards regression, Fine–Gray competing risk regression, and Kaplan–Meier failure function were used to estimate the morbidity burden. The study included 794 patients diagnosed with EA and 7940 controls, with a median follow‐up period of 33 years. Among patients with EA, approximately half (n=442) had isolated EA, and 28% (n=218) had concomitant atrial septal defect. Patients with complex anatomy demonstrated the highest cardiovascular morbidity burden, followed by those with concomitant atrial septal defect and isolated EA. The lifetime cumulative incidence of supraventricular arrhythmia and ventricular preexcitation in patients with EA, with or without atrial septal defect, was approximately 70% and 19%, respectively. Supraventricular arrhythmia substantially increased the risk of ischemic stroke (hazard ratio [HR] 22.6 [95% CI, 11.1–45.9]). Presence of atrial septal defect significantly affected arrhythmia and heart failure burden compared with isolated EA. In the total cohort of patients with EA, supraventricular arrhythmia onset led to an immediate high incidence of heart failure, with a 10‐year cumulative incidence of 18%. Conclusions: The natural history of EA, whether isolated or not, involves a substantial burden of cardiovascular morbidity and thus a highly vulnerable long‐term prognosis.
- 4 November 2024
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 13 (21) , e037127
Open AccessMorbidity Burden in Patients With Ebstein Anomaly: The Natural History
Background: The lifetime morbidity burden of patients with Ebstein anomaly (EA) has not been well described. Methods and Results: Through an extensive 2‐country register‐based collaboration, patients diagnosed with EA who were born between 1930 and 2017 were identified in Danish and Swedish nationwide medical registries. Each patient was matched by age and sex with 10 control subjects from the general population. Cox proportional‐hazards regression, Fine–Gray competing risk regression, and Kaplan–Meier failure function were used to estimate the morbidity burden. The study included 794 patients diagnosed with EA and 7940 controls, with a median follow‐up period of 33 years. Among patients with EA, approximately half (n=442) had isolated EA, and 28% (n=218) had concomitant atrial septal defect. Patients with complex anatomy demonstrated the highest cardiovascular morbidity burden, followed by those with concomitant atrial septal defect and isolated EA. The lifetime cumulative incidence of supraventricular arrhythmia and ventricular preexcitation in patients with EA, with or without atrial septal defect, was approximately 70% and 19%, respectively. Supraventricular arrhythmia substantially increased the risk of ischemic stroke (hazard ratio [HR] 22.6 [95% CI, 11.1–45.9]). Presence of atrial septal defect significantly affected arrhythmia and heart failure burden compared with isolated EA. In the total cohort of patients with EA, supraventricular arrhythmia onset led to an immediate high incidence of heart failure, with a 10‐year cumulative incidence of 18%. Conclusions: The natural history of EA, whether isolated or not, involves a substantial burden of cardiovascular morbidity and thus a highly vulnerable long‐term prognosis.
- Open Access
- 30 October 2024
- Published by BMJ in Open Heart
- Vol. 11 (2) , e002937
Predicting troponin biomarker elevation from electrocardiograms using a deep neural network
Background: Elevated troponin levels are a sensitive biomarker for cardiac injury. The quick and reliable prediction of troponin elevation for patients with chest pain from readily available ECGs may pose a valuable time-saving diagnostic tool during decision-making concerning this patient population. Methods and results: The data used included 15 856 ECGs from patients presenting to the emergency rooms with chest pain or dyspnoea at two centres in Sweden from 2015 to June 2023. All patients had high-sensitivity troponin test results within 6 hours after 12-lead ECG. Both troponin I (TnI) and TnT were used, with biomarker-specific cut-offs and sex-specific cut-offs for TnI. On this dataset, a residual convolutional neural network (ResNet) was trained 10 times, each on a unique split of the data. The final model achieved an average area under the curve for the receiver operating characteristic curve of 0.7717 (95% CI±0.0052), calibration curve analysis revealed a mean slope of 1.243 (95% CI±0.075) and intercept of −0.073 (95% CI±0.034), indicating a good correlation between prediction and ground truth. Post-classification, tuned for F1 score, accuracy was 71.43% (95% CI±1.28), with an F1 score of 0.5642 (95% CI±0.0052) and a negative predictive value of 0.8660 (95% CI±0.0048), respectively. The ResNet displayed comparable or surpassing metrics to prior presented models. Conclusion: The model exhibited clinically meaningful performance, notably its high negative predictive accuracy. Therefore, clinical use of comparable neural networks in first-line, quick-response triage of patients with chest pain or dyspnoea appears as a valuable option in future medical practice.
- 30 October 2024
- Published by BMJ in Open Heart
- Vol. 11 (2) , e002937
Open AccessPredicting troponin biomarker elevation from electrocardiograms using a deep neural network
Background: Elevated troponin levels are a sensitive biomarker for cardiac injury. The quick and reliable prediction of troponin elevation for patients with chest pain from readily available ECGs may pose a valuable time-saving diagnostic tool during decision-making concerning this patient population. Methods and results: The data used included 15 856 ECGs from patients presenting to the emergency rooms with chest pain or dyspnoea at two centres in Sweden from 2015 to June 2023. All patients had high-sensitivity troponin test results within 6 hours after 12-lead ECG. Both troponin I (TnI) and TnT were used, with biomarker-specific cut-offs and sex-specific cut-offs for TnI. On this dataset, a residual convolutional neural network (ResNet) was trained 10 times, each on a unique split of the data. The final model achieved an average area under the curve for the receiver operating characteristic curve of 0.7717 (95% CI±0.0052), calibration curve analysis revealed a mean slope of 1.243 (95% CI±0.075) and intercept of −0.073 (95% CI±0.034), indicating a good correlation between prediction and ground truth. Post-classification, tuned for F1 score, accuracy was 71.43% (95% CI±1.28), with an F1 score of 0.5642 (95% CI±0.0052) and a negative predictive value of 0.8660 (95% CI±0.0048), respectively. The ResNet displayed comparable or surpassing metrics to prior presented models. Conclusion: The model exhibited clinically meaningful performance, notably its high negative predictive accuracy. Therefore, clinical use of comparable neural networks in first-line, quick-response triage of patients with chest pain or dyspnoea appears as a valuable option in future medical practice.
- Open Access
- 10 October 2024
- Published by Oxford University Press (OUP) in European Heart Journal Open
- Vol. 4 (6) , oeae089
Risk of venous thromboembolism in patients with congenital heart disease: a nationwide, register-based, case–control study
Aims: Patients with congenital heart disease (CHD) have an increased risk of developing acquired cardiovascular diseases. However, the risk of venous thromboembolism (VTE) in patients with CHD is unknown. We aimed to investigate the incidence and risk of VTE in patients with CHD compared with matched controls without CHD. Methods and results: Data from Swedish health registers were used to identify all patients with CHD between 1970 and 2017 in Sweden. Each patient with CHD was matched with 10 controls from the Swedish Total Population Register. The primary outcome of the study was an event of VTE. Follow-up was from birth until VTE, death, or the end of the study (2017). Cox proportional hazard models were used to investigate the risk of VTE in patients with CHD and controls. A total of 67 814 patients with CHD and 583 709 matched controls were identified and included in the study. During a mean follow-up of 15.9 (SD ± 12.5) years, 554 (0.8%) patients with CHD and 1571 (0.3%) controls developed VTE. The risk of VTE was 3.3 [95% confidence interval [CI] 2.6–3.4) times higher in patients with CHD than in controls. Patients with conotruncal defects had the highest risk of VTE (hazard ratio 7.06, 95% CI 5.52–9.03). Conclusion: In our nationwide study, we found that the risk of VTE in patients with CHD was more than three times higher than in matched controls. The highest risk of VTE was in patients with the most complex lesions. Further research is crucial to clarify the underlying risk factors and prevent VTE in patients with CHD.
- 10 October 2024
- Published by Oxford University Press (OUP) in European Heart Journal Open
- Vol. 4 (6) , oeae089
Open AccessRisk of venous thromboembolism in patients with congenital heart disease: a nationwide, register-based, case–control study
Aims: Patients with congenital heart disease (CHD) have an increased risk of developing acquired cardiovascular diseases. However, the risk of venous thromboembolism (VTE) in patients with CHD is unknown. We aimed to investigate the incidence and risk of VTE in patients with CHD compared with matched controls without CHD. Methods and results: Data from Swedish health registers were used to identify all patients with CHD between 1970 and 2017 in Sweden. Each patient with CHD was matched with 10 controls from the Swedish Total Population Register. The primary outcome of the study was an event of VTE. Follow-up was from birth until VTE, death, or the end of the study (2017). Cox proportional hazard models were used to investigate the risk of VTE in patients with CHD and controls. A total of 67 814 patients with CHD and 583 709 matched controls were identified and included in the study. During a mean follow-up of 15.9 (SD ± 12.5) years, 554 (0.8%) patients with CHD and 1571 (0.3%) controls developed VTE. The risk of VTE was 3.3 [95% confidence interval [CI] 2.6–3.4) times higher in patients with CHD than in controls. Patients with conotruncal defects had the highest risk of VTE (hazard ratio 7.06, 95% CI 5.52–9.03). Conclusion: In our nationwide study, we found that the risk of VTE in patients with CHD was more than three times higher than in matched controls. The highest risk of VTE was in patients with the most complex lesions. Further research is crucial to clarify the underlying risk factors and prevent VTE in patients with CHD.