Perceptions of a Secure Cloud-Based Solution for Data Sharing During Acute Stroke Care: Qualitative Interview Study

BACKGROUND

Acute stroke care demands fast procedures performed through the collaboration of multiple professionals across multiple organizations. Cloud computing and the wide adoption of electronic medical records (EMRs) enable health care systems to improve data availability and facilitate sharing among professionals. However, designing a secure and privacy-preserving EMR cloud-based application is challenging because it must dynamically control the access to the patient's EMR according to the needs for data during treatment.

OBJECTIVE

We developed a prototype of a secure EMR cloud-based application. The application explores the security features offered by the eHealth cloud-based framework created by the Advanced Secure Cloud Encrypted Platform for Internationally Orchestrated Solutions in Health Care Horizon 2020 project. This study aimed to collect impressions, challenges, and improvements for the prototype when applied to the use case of secure data sharing among acute care teams during emergency treatment in the Netherlands.

METHODS

We conducted 14 semistructured interviews with medical professionals with 4 prominent roles in acute care: emergency call centers, ambulance services, emergency hospitals, and general practitioner clinics. We used in-depth interviews to capture their perspectives about the application's design and functions and its use in a simulated acute care event. We used thematic analysis of interview transcripts. Participants were recruited until the collected data reached thematic saturation.

RESULTS

The participants' perceptions and feedback are presented as 5 themes identified from the interviews: current challenges (theme 1), quality of the shared EMR data (theme 2), integrity and auditability of the EMR data (theme 3), usefulness and functionality of the application (theme 4), and trust and acceptance of the technology (theme 5). The results reinforced the current challenges in patient data sharing during acute stroke care. Moreover, from the user point of view, we expressed the challenges of adopting the Advanced Secure Cloud Encrypted Platform for Internationally Orchestrated Solutions in Health Care Acute Stroke Care application in a real scenario and provided suggestions for improving the proposed technology's acceptability.

CONCLUSIONS

This study has endorsed a system that supports data sharing among acute care professionals with efficiency, but without compromising the security and privacy of the patient. This explorative study identified several significant barriers to and improvement opportunities for the future acceptance and adoption of the proposed system. Moreover, the study results highlight that the desired digital transformation should consider integrating the already existing systems instead of requesting migration to a new centralized system.

The gut microbiota and depressive symptoms across ethnic groups

The gut microbiome is thought to play a role in depressive disorders, which makes it an attractive target for interventions. Both the microbiome and depressive symptom levels vary substantially across ethnic groups. Thus, any intervention for depression targeting the microbiome requires understanding of microbiome-depression associations across ethnicities. Analysing data from the HELIUS cohort, we characterize the gut microbiota and its associations with depressive symptoms in 6 ethnic groups (Dutch, South-Asian Surinamese, African Surinamese, Ghanaian, Turkish, Moroccan; N = 3211), living in the same urban area. Diversity of the gut microbiota, both within (α-diversity) and between individuals (β-diversity), predicts depressive symptom levels, taking into account demographic, behavioural, and medical differences. These associations do not differ between ethnic groups. Further, β-diversity explains 29%-18% of the ethnic differences in depressive symptoms. Bacterial genera associated with depressive symptoms belong to mulitple families, prominently including the families Christensenellaceae, Lachnospiraceae, and Ruminococcaceae. In summary, the results show that the gut microbiota are linked to depressive symptom levels and that this association generalizes across ethnic groups. Moreover, the results suggest that ethnic differences in the gut microbiota may partly explain parallel disparities in depression.

Lipoprotein(a) levels in children with suspected familial hypercholesterolaemia: a cross-sectional study

AIMS

Familial hypercholesterolaemia (FH) predisposes children to the early initiation of atherosclerosis and is preferably diagnosed by DNA analysis. Yet, in many children with a clinical presentation of FH, no mutation is found. Adult data show that high levels of lipoprotein(a) [Lp(a)] may underlie a clinical presentation of FH, as the cholesterol content of Lp(a) is included in conventional LDL cholesterol measurements. As this is limited to adult data, Lp(a) levels in children with and without (clinical) FH were evaluated.

METHODS AND RESULTS

Children were eligible if they visited the paediatric lipid clinic (1989-2020) and if Lp(a) measurement and DNA analysis were performed. In total, 2721 children (mean age: 10.3 years) were included and divided into four groups: 1931 children with definite FH (mutation detected), 290 unaffected siblings/normolipidaemic controls (mutation excluded), 108 children with probable FH (clinical presentation, mutation not detected), and 392 children with probable non-FH (no clinical presentation, mutation not excluded). In children with probable FH, 32% were found to have high Lp(a) [geometric mean (95% confidence interval) of 15.9 (12.3-20.6) mg/dL] compared with 10 and 10% [geometric means (95% confidence interval) of 11.5 (10.9-12.1) mg/dL and 9.8 (8.4-11.3) mg/dL] in children with definite FH (P = 0.017) and unaffected siblings (P = 0.002), respectively.

CONCLUSION

Lp(a) was significantly higher and more frequently elevated in children with probable FH compared with children with definite FH and unaffected siblings, suggesting that high Lp(a) may underlie the clinical presentation of FH when no FH-causing mutation is found. Performing both DNA analysis and measuring Lp(a) in all children suspected of FH is recommended to assess possible LDL cholesterol overestimation related to increased Lp(a).

A cross-sectional study on fatigue, anxiety, and symptoms of depression and their relation with medical status in adult patients with Marfan syndrome. Psychological consequences in Marfan syndrome

Marfan syndrome (MFS) is a connective tissue disorder affecting the cardiovascular, ocular, and skeletal system, which may be accompanied by psychological features. This study aimed to determine the prevalence of fatigue, anxiety, and symptoms of depression in MFS patients, and to assess the degree to which sociodemographic and clinical variables are associated with fatigue and psychological aspects. The prevalence of fatigue, anxiety, and symptoms of depression were assessed in two cohorts of MFS patients and compared with healthy controls. The checklist individual strength (CIS), and hospital anxiety and depression scale (HADS) questionnaires were utilized. Medical status was assessed (family history of MFS, aortic root dilatation >40 mm, previous aortic surgery, aortic dissection, chronic pain, skeletal involvement, and scoliosis). Severe fatigue was experienced by 37% of the total MFS cohort (n = 155). MFS patients scored significantly higher on the CIS questionnaire, concerning severe fatigue, as compared with the general Dutch population (p < 0.0001). There were no differences in HADS anxiety or depression scores. In older MFS patients, with a more severe cardiovascular phenotype, chronic pain, and a higher unemployment rate, significantly more symptoms of depression were observed, when compared with the general population (p = 0.027) or compared with younger MFS patients (p = 0.026). Multivariate analysis, showed that anxiety was associated with chronic pain (p = 0.022) and symptoms of depression with unemployment (p = 0.024). MFS patients report significantly more severe fatigue as compared with the general population. Since the cause of fatigue is unclear, more research may be needed. Psychological intervention, for example, cognitive behavioral therapy, may contribute to a reduction in psychological symptoms.

Between-hospital variation in time to glioblastoma surgery: a report from the Quality Registry Neuro Surgery in the Netherlands

OBJECTIVE

Patients with glioblastoma are often scheduled for urgent elective surgery. Currently, the impact of the waiting period until glioblastoma surgery is undetermined. In this national quality registry study, the authors determined the wait times until surgery for patients with glioblastoma, the risk factors associated with wait times, and the risk-standardized variation in time to surgery between Dutch hospitals. The associations between time to surgery and patient outcomes were also explored.

METHODS

Data from all 4589 patients who underwent first-time glioblastoma surgery between 2014 and 2019 in the Netherlands were collected by 13 hospitals in the Quality Registry Neuro Surgery. Time to surgery comprised 1) the time from first MR scan to surgery (MTS), and 2) the time from first neurosurgical consultation to surgery (CTS). Long MTS was defined as more than 21 days and long CTS as more than 14 days. Potential risk factors were analyzed in multivariable logistic regression models. The standardized rate of long time to surgery was analyzed using funnel plots. Patient outcomes including Karnofsky Performance Scale (KPS) score change, complications, and survival were analyzed by multivariable logistic regression and proportional hazards models.

RESULTS

The median overall MTS and CTS were 18 and 9 days, respectively. Overall, 2576 patients (56%) had an MTS within 3 weeks and 3069 (67%) had a CTS within 2 weeks. Long MTS was significantly associated with older age, higher preoperative KPS score, higher American Society of Anesthesiologists comorbidity class, season, lower hospital case volume, university affiliation, and resection. Long CTS was significantly associated with higher baseline KPS score, university affiliation, resection, more recent year of treatment, and season. In funnel plots, considerable practice variation was observed between hospitals in patients with long times to surgery. Fewer patients with KPS score improvement were observed after a long time until resection. Long CTS was associated with longer survival. Complications and KPS score decline were not associated with time to surgery.

CONCLUSIONS

Considerable between-hospital variation among Dutch hospitals was observed in the time to glioblastoma surgery. A long time to resection impeded KPS score improvement, and therefore, patients who may improve should be identified for more urgent resection. Longer survival was observed in patients selected for longer time until surgery after neurosurgical consultation (CTS).

Six-lead device superior to single-lead smartwatch ECG in atrial fibrillation detection

This was a head-to-head comparative study on different electrocardiogram (ECG)-based smartwatches and devices for atrial fibrillation detection. We prospectively included 220 patients scheduled for electrical cardioversion and recorded ECGs with 3 different devices (Withings Move ECG, Apple Watch 5, Kardia Mobile 6-leads) as well as the standard 12-lead ECG (gold standard), both before and after cardioversion. All atrial fibrillation detection algorithms had high accuracy (sensitivity and specificity: 91-99%) but were hampered by uninterpretable recordings (20-24%). In cardiologists' interpretation, the 6-lead device was superior (sensitivity 99%, specificity 97%) to both single-lead smartwatches (P < .05) for atrial fibrillation detection.

bnormal aortic hemodynamics are associated with risk factors for aortic complications in patients with marfan syndrome

Background

It is difficult to assess the risk for aortic dissection beyond the aortic root in patients with Marfan syndrome (MFS). To aid risk assessment in these patients, we investigated aortic flow and wall shear stress (WSS) by 4D flow magnetic resonance imaging (MRI) in patients with MFS and compared the results with healthy volunteers. We hypothesized that MFS patients with a high-risk profile for aortic dissection would show abnormal hemodynamics in aortic regions associated with aortic dissection.

Methods

MFS patients (n = 55) and healthy subjects (n = 25), matched for age and sex, prospectively underwent 4D flow MRI. 4D flow maps were constructed to detect elevated (defined as higher than the three-dimensional 95 % confidence interval) and deviant directed (defined as vector angle differences higher than 120°) WSS in MFS patients as compared to the controls. Univariate and multivariate associations with risk factors for aortic dissection in MFS patients were assessed.

Results

The maximum incidence for elevated WSS was 20 % (CI 9 %-31 %) and found in the ascending aorta. The maximum for deviant directed WSS was 39 % (CI 26 %-52 %) and found in the inner descending aorta. Significantly more male patients had deviant directed WSS in the inner proximal descending aorta (63 % vs 24 %, p = 0.014). Multivariate analysis showed that deviant directed WSS was associated with male sex (p = 0.019), and a haplo-insufficient FBN1 mutation type (p = 0.040). In 60 % of MFS patients with a previous aortic root replacement surgery, abnormal hemodynamics were found in the ascending aorta. No significant differences between hemodynamics were found in the descending aorta between operated and non-operated patients.

Conclusion

Deviant directed WSS in the proximal descending aorta is associated with known risk factors for aortic dissection in MFS patients, namely male sex and a haploinsufficient FBN1 mutation type.

P11.37.B When to resect or biopsy for patients with supratentorial glioblastoma: a multivariable prediction model

Background

The prospects of a patient with suspected glioblastoma may rely heavily on the indication for surgical resection versus biopsy only. Biopsy percentages vary considerably across hospitals and guidelines for treatment of glioblastoma lack criteria for surgical decision-making. To identify patient and tumor characteristics associated with the decision to resect or biopsy a glioblastoma and to develop and validate a prediction model for decision support.

Material and Methods

Clinical data and pre-operative MRI scans were collected for adults who underwent first-time surgery for supratentorial glioblastoma from a registry-based cohort study of 12 hospitals from the Netherlands, Germany, France, Italy, and the United States between 1st of January 2007 and 31st of December 2011. The main outcome was the type of surgical procedure: surgical resection or biopsy only. Predictors were patient- and tumor-related characteristics. Radiological factors were extracted from MRI using an automated tumor segmentation method. A prediction model was constructed using multivariable logistic regression analysis. The model was cross-validated and externally validated with a leave-one-hospital-out approach.

Results

Out of 1053 patients treated for glioblastoma, 28% underwent biopsy only. Biopsy rates varied from 15-40% across hospitals. The prediction model showed excellent discrimination with an average area under the curve of 0.86. Of the patient-related characteristics, younger age was associated more with resection and Karnofsky Performance Score of 60 or less with biopsy. Of the tumor-related characteristics, a location in the right hemisphere, unifocality, no tumor midline crossing, and no involvement of the cortical spinal tract, were associated with resection, as well as a high expected resectability index, a location in the right occipital lobe, and a higher percentage of tumor in Schaefer’s dorsal or ventral attention, limbic, and default networks. External validation proved acceptable to outstanding discrimination with areas under the curve ranging between 0.79 and 0.92 for hospitals.

Conclusion

A prediction model is presented and validated to support the decision to resect or to biopsy a patient with a suspected supratentorial glioblastoma. In this prediction model, tumor-related characteristics were more informative than patient-related factors. This may support surgical decision-making for individual patients, or facilitate comparisons of patient cohorts between surgeons or institutions.

Gut microbiome associations with host genotype vary across ethnicities and potentially influence cardiometabolic traits

Previous studies in mainly European populations have reported that the gut microbiome composition is associated with the human genome. However, the genotype-microbiome interaction in different ethnicities is largely unknown. We performed a large fecal microbiome genome-wide association study of a single multiethnic cohort, the Healthy Life in an Urban Setting (HELIUS) cohort (N = 4,117). Mendelian randomization was performed using the multiethnic Pan-UK Biobank (N = 460,000) to dissect potential causality. We identified ethnicity-specific associations between host genomes and gut microbiota. Certain microbes were associated with genotype in multiple ethnicities. Several of the microbe-associated loci were found to be related to immune functions, interact with glutamate and the mucus layer, or be expressed in the gut or brain. Additionally, we found that gut microbes potentially influence cardiometabolic health factors such as BMI, cholesterol, and blood pressure. This provides insight into the relationship of ethnicity and gut microbiota and into the possible causal effects of gut microbes on cardiometabolic traits.

BAHAMA: A Bayesian Hierarchical Model for the Detection of MedDRA®-Coded Adverse Events in Randomized Controlled Trials

INTRODUCTION

Patients participating in randomized controlled trials (RCTs) are susceptible to a wide range of different adverse events (AE) during the RCT. MedDRA® is a hierarchical standardization terminology to structure the AEs reported in an RCT. The lowest level in the MedDRA hierarchy is a single medical event, and every higher level is the aggregation of the lower levels.

METHOD

We propose a multi-stage Bayesian hierarchical Poisson model for estimating MedDRA-coded AE rate ratios (RRs). To deal with rare AEs, we introduce data aggregation at a higher level within the MedDRA structure and based on thresholds on incidence and MedDRA structure.

RESULTS

With simulations, we showed the effects of this data aggregation process and the method's performance. Furthermore, an application to a real example is provided and compared with other methods.

CONCLUSION

We showed the benefit of using the full MedDRA structure and using aggregated data. The proposed model, as well as the pre-processing, is implemented in an R-package: BAHAMA.

Preoperative Brain Tumor Imaging: Models and Software for Segmentation and Standardized Reporting

For patients suffering from brain tumor, prognosis estimation and treatment decisions are made by a multidisciplinary team based on a set of preoperative MR scans. Currently, the lack of standardized and automatic methods for tumor detection and generation of clinical reports, incorporating a wide range of tumor characteristics, represents a major hurdle. In this study, we investigate the most occurring brain tumor types: glioblastomas, lower grade gliomas, meningiomas, and metastases, through four cohorts of up to 4,000 patients. Tumor segmentation models were trained using the AGU-Net architecture with different preprocessing steps and protocols. Segmentation performances were assessed in-depth using a wide-range of voxel and patient-wise metrics covering volume, distance, and probabilistic aspects. Finally, two software solutions have been developed, enabling an easy use of the trained models and standardized generation of clinical reports: Raidionics and Raidionics-Slicer. Segmentation performances were quite homogeneous across the four different brain tumor types, with an average true positive Dice ranging between 80 and 90%, patient-wise recall between 88 and 98%, and patient-wise precision around 95%. In conjunction to Dice, the identified most relevant other metrics were the relative absolute volume difference, the variation of information, and the Hausdorff, Mahalanobis, and object average symmetric surface distances. With our Raidionics software, running on a desktop computer with CPU support, tumor segmentation can be performed in 16-54 s depending on the dimensions of the MRI volume. For the generation of a standardized clinical report, including the tumor segmentation and features computation, 5-15 min are necessary. All trained models have been made open-access together with the source code for both software solutions and validation metrics computation. In the future, a method to convert results from a set of metrics into a final single score would be highly desirable for easier ranking across trained models. In addition, an automatic classification of the brain tumor type would be necessary to replace manual user input. Finally, the inclusion of post-operative segmentation in both software solutions will be key for generating complete post-operative standardized clinical reports.

SOURCE beyond first-line: A survival prediction model for patients with metastatic esophagogastric adenocarcinoma after failure of first-line palliative systemic therapy

4037

Background: Prediction models for survival may aid shared decision making between physicians and patients. Prior models have been developed that predict survival for patients with potentially curable esophagogastric cancer and patients with metastatic esophageal cancer who start first-line therapy (the SOURCE models). The aim of this study was to develop and internally validate a registry-based clinical prediction model, called SOURCE beyond first-line, for survival of patients with metastatic esophagogastric cancer after failure of first-line palliative systemic therapy. Methods: Patients with unresectable or metastatic (synchronous or metachronous) esophageal or gastric cancer who received first-line systemic therapy (N = 1067) between 2015-2017 were selected from the Netherlands Cancer Registry. Follow-up data were retrieved in 2019. Patient, tumor and treatment characteristics at primary diagnosis and at progression of disease, were used to develop the prediction model. A Cox proportional hazards regression prediction model was developed through forward and backward selection using Akaike’s Information Criterion. The model was internally validated through 10-fold cross-validations to assess performance on unseen data. Model discrimination (C-index) and calibration (slope and intercept) were used to evaluate performance of the complete and cross-validated models. Results: The final model consisted of 10 patient, tumor and treatment characteristics. The C-index was 0.75 (0.74-0.77), the calibration slope 0.99 (0.98-0.99) and the calibration intercept 0.02 (0.01-0.02). Internal cross-validation of the model showed that the model performed adequately on unseen data: C-index 0.79 (0.76-0.80), calibration slope 1.02 (1.00-1.04) and calibration intercept -0.01 (-0.01-0.02). Conclusions: The SOURCE beyond first-line model predicted survival with fair discriminatory ability and good calibration, and is a valuable addition to the existing SOURCE prediction models. In the future this model will be integrated in an online decision support tool that can be used in clinical practice to aid personalized treatment.

A deep learning approach identifies new ECG features in congenital long QT syndrome

BACKGROUND

Congenital long QT syndrome (LQTS) is a rare heart disease caused by various underlying mutations. Most general cardiologists do not routinely see patients with congenital LQTS and may not always recognize the accompanying ECG features. In addition, a proportion of disease carriers do not display obvious abnormalities on their ECG. Combined, this can cause underdiagnosing of this potentially life-threatening disease.

METHODS

This study presents 1D convolutional neural network models trained to identify genotype positive LQTS patients from electrocardiogram as input. The deep learning (DL) models were trained with a large 10-s 12-lead ECGs dataset provided by Amsterdam UMC and externally validated with a dataset provided by University Hospital Leuven. The Amsterdam dataset included ECGs from 10000 controls, 172 LQTS1, 214 LQTS2, and 72 LQTS3 patients. The Leuven dataset included ECGs from 2200 controls, 32 LQTS1, and 80 LQTS2 patients. The performance of the DL models was compared with conventional QTc measurement and with that of an international expert in congenital LQTS (A.A.M.W). Lastly, an explainable artificial intelligence (AI) technique was used to better understand the prediction models.

RESULTS

Overall, the best performing DL models, across 5-fold cross-validation, achieved on average a sensitivity of 84 ± 2%, 90 ± 2% and 87 ± 6%, specificity of 96 ± 2%, 95 ± 1%, and 92 ± 4%, and AUC of 0.90 ± 0.01, 0.92 ± 0.02, and 0.89 ± 0.03, for LQTS 1, 2, and 3 respectively. The DL models were also shown to perform better than conventional QTc measurements in detecting LQTS patients. Furthermore, the performances held up when the DL models were validated on a novel external cohort and outperformed the expert cardiologist in terms of specificity, while in terms of sensitivity, the DL models and the expert cardiologist in LQTS performed the same. Finally, the explainable AI technique identified the onset of the QRS complex as the most informative region to classify LQTS from non-LQTS patients, a feature previously not associated with this disease.

CONCLUSIONS

This study suggests that DL models can potentially be used to aid cardiologists in diagnosing LQTS. Furthermore, explainable DL models can be used to possibly identify new features for LQTS on the ECG, thus increasing our understanding of this syndrome.

Neoadjuvant Chemoradiotherapy Versus Upfront Surgery for Resectable and Borderline Resectable Pancreatic Cancer: Long-Term Results of the Dutch Randomized PREOPANC Trial

PURPOSE

The benefit of neoadjuvant chemoradiotherapy in resectable and borderline resectable pancreatic cancer remains controversial. Initial results of the PREOPANC trial failed to demonstrate a statistically significant overall survival (OS) benefit. The long-term results are reported.

METHODS

In this multicenter, phase III trial, patients with resectable and borderline resectable pancreatic cancer were randomly assigned (1:1) to neoadjuvant chemoradiotherapy or upfront surgery in 16 Dutch centers. Neoadjuvant chemoradiotherapy consisted of three cycles of gemcitabine combined with 36 Gy radiotherapy in 15 fractions during the second cycle. After restaging, patients underwent surgery followed by four cycles of adjuvant gemcitabine. Patients in the upfront surgery group underwent surgery followed by six cycles of adjuvant gemcitabine. The primary outcome was OS by intention-to-treat. No safety data were collected beyond the initial report of the trial.

RESULTS

Between April 24, 2013, and July 25, 2017, 246 eligible patients were randomly assigned to neoadjuvant chemoradiotherapy (n = 119) and upfront surgery (n = 127). At a median follow-up of 59 months, the OS was better in the neoadjuvant chemoradiotherapy group than in the upfront surgery group (hazard ratio, 0.73; 95% CI, 0.56 to 0.96; P = .025). Although the difference in median survival was only 1.4 months (15.7 months v 14.3 months), the 5-year OS rate was 20.5% (95% CI, 14.2 to 29.8) with neoadjuvant chemoradiotherapy and 6.5% (95% CI, 3.1 to 13.7) with upfront surgery. The effect of neoadjuvant chemoradiotherapy was consistent across the prespecified subgroups, including resectable and borderline resectable pancreatic cancer.

CONCLUSION

Neoadjuvant gemcitabine-based chemoradiotherapy followed by surgery and adjuvant gemcitabine improves OS compared with upfront surgery and adjuvant gemcitabine in resectable and borderline resectable pancreatic cancer.

Estimating the Risk of Severe Peanut Allergy Using Clinical Background and IgE Sensitization Profiles

Background: It is not well-understood why symptom severity varies between patients with peanut allergy (PA). Objective: To gain insight into the clinical profile of subjects with mild-to-moderate and severe PA, and investigate individual and collective predictive accuracy of clinical background and IgE to peanut extract and components for PA severity. Methods: Data on demographics, patient history and sensitization at extract and component level of 393 patients with probable PA (symptoms ≤ 2 h + IgE sensitization) from 12 EuroPrevall centers were analyzed. Univariable and penalized multivariable regression analyses were used to evaluate risk factors and biomarkers for severity. Results: Female sex, age at onset of PA, symptoms elicited by skin contact with peanut, family atopy, atopic dermatitis, house dust mite and latex allergy were independently associated with severe PA; birch pollen allergy with mild-to-moderate PA. The cross-validated AUC of all clinical background determinants combined (0.74) was significantly larger than the AUC of tests for sensitization to extract (0.63) or peanut components (0.54-0.64). Although larger skin prick test wheal size, and higher IgE to peanut extract, Ara h 1 and Ara h 2/6, were associated with severe PA, and higher IgE to Ara h 8 with mild-to-moderate PA, addition of these measurements of sensitization to the clinical background model did not significantly improve the AUC. Conclusions: Models combining clinical characteristics and IgE sensitization patterns can help establish the risk of severe reactions for peanut allergic patients, but clinical background determinants are most valuable for predicting severity of probable PA in an individual patient.

Electrocardiogram-based mortality prediction in patients with COVID-19 using machine learning

Background and purpose

The electrocardiogram (ECG) is frequently obtained in the work-up of COVID-19 patients. So far, no study has evaluated whether ECG-based machine learning models have added value to predict in-hospital mortality specifically in COVID-19 patients.

Methods

Using data from the CAPACITY-COVID registry, we studied 882 patients admitted with COVID-19 across seven hospitals in the Netherlands. Raw format 12-lead ECGs recorded within 72 h of admission were studied. With data from five hospitals (n = 634), three models were developed: (a) a logistic regression baseline model using age and sex, (b) a least absolute shrinkage and selection operator (LASSO) model using age, sex and human annotated ECG features, and (c) a pre-trained deep neural network (DNN) using age, sex and the raw ECG waveforms. Data from two hospitals (n = 248) was used for external validation.

Results

Performances for models a, b and c were comparable with an area under the receiver operating curve of 0.73 (95% confidence interval [CI] 0.65–0.79), 0.76 (95% CI 0.68–0.82) and 0.77 (95% CI 0.70–0.83) respectively. Predictors of mortality in the LASSO model were age, low QRS voltage, ST depression, premature atrial complexes, sex, increased ventricular rate, and right bundle branch block.

Conclusion

This study shows that the ECG-based prediction models could be helpful for the initial risk stratification of patients diagnosed with COVID-19, and that several ECG abnormalities are associated with in-hospital all-cause mortality of COVID-19 patients. Moreover, this proof-of-principle study shows that the use of pre-trained DNNs for ECG analysis does not underperform compared with time-consuming manual annotation of ECG features.

Supplementary Information

The online version of this article (10.1007/s12471-022-01670-2) contains supplementary material, which is available to authorized users.

Abstract 51: Age Stratified Risk Of Cerebral Venous Sinus Thrombosis After Sars-Cov-2 Vaccination

Introduction: Cerebral Venous Sinus Thrombosis (CVST) as a part of the thrombosis and thrombocytopenia syndrome is a rare adverse drug reaction of SARS-CoV-2 vaccination. The estimated background rate of CVST in adults is around 1 case per million per month, and CVST with thrombocytopenia accounts for 8% of all CVST. We assessed the age-stratified risk of CVST with and without thrombocytopenia after SARS-CoV-2 vaccination.

Methods: We estimated the absolute risk of any CVST, CVST with thrombocytopenia, and CVST without thrombocytopenia, within 28 days of first dose SARS-CoV-2 vaccination, using data from the European Medicines Agency’s EudraVigilance database (until 13 June 2021). As a denominator, we used data on vaccine delivery from 31 European countries. For 22.8 million adults from 25 countries we estimated the absolute risk of CVST after the first dose of ChAdOx1 nCov-19 per age category.

Results: The absolute risk of CVST within 28 days of first dose vaccination was 7.5 (95%CI 6.9-8.3), 0.7 (95%CI 0.2-2.4), 0.6 (95%CI 0.5-0.7) and 0.6 (95%CI 0.3-1.1) per million of first doses of ChAdOx1 nCov-19, Ad26.COV2.S, BNT162b2 and mRNA-1273, respectively. The absolute risk of CVST with thrombocytopenia within 28 days of first dose vaccination was 4.4 (95%CI 3.9-4.9), 0.7 (95%CI 0.2-2.4), 0.0 (95%CI 0.0-0.1) and 0.0 (95%CI 0.0-0.2) per million of first doses of ChAdOx1 nCov-19, Ad26.COV2.S, BNT162b2 and mRNA-1273, respectively. In recipients of ChAdOx1 nCov-19, the risk of CVST, both with and without thrombocytopenia, was the highest in the 18-24 years age group (7.3 per million, 95%CI 2.8-18.8 and 3.7, 95%CI 1.0-13.3, respectively). The risk of CVST with thrombocytopenia was the lowest in ChAdOx1 nCov-19 recipients ≥70 years (0.2, 95%CI 0.0-1.3). Age <60 compared to ≥60 was a predictor for CVST with thrombocytopenia (incidence rate ratio 5.79; 95%CI 2.98-11.24, p<0.001).

Discussion: The risk of CVST with thrombocytopenia within 28 days of first dose vaccination with ChAdOx1 nCov-19 was higher in younger age groups. The risk of CVST with thrombocytopenia was slightly increased in patients receiving Ad26.COV2.S, comparing with the estimated background risk. The risk of CVST with thrombocytopenia was not increased in recipients of mRNA vaccines for SARS-CoV-2.

Differences in SARS-CoV-2 infections during the first and second wave of SARS-CoV-2 between six ethnic groups in Amsterdam, the Netherlands: A population-based longitudinal serological study

BACKGROUND

Surveillance data in high-income countries have reported more frequent SARS-CoV-2 diagnoses in ethnic minority groups. We examined the cumulative incidence of SARS-CoV-2 and its determinants in six ethnic groups in Amsterdam, the Netherlands.

METHODS

We analysed participants enrolled in the population-based HELIUS cohort, who were tested for SARS-CoV-2-specific antibodies and answered COVID-19-related questions between June 24-October 9, 2020 (after the first wave) and November 23, 2020-March 31, 2021 (during the second wave). We modelled SARS-CoV-2 incidence from January 1, 2020-March 31, 2021 using Markov models adjusted for age and sex. We compared incidence between ethnic groups over time and identified determinants of incident infection within ethnic groups.

FINDINGS

2,497 participants were tested after the first wave; 2,083 (83·4%) were tested during the second wave. Median age at first visit was 54 years (interquartile range=44-61); 56·6% were female. Compared to Dutch-origin participants (15·9%), cumulative SARS-CoV-2 incidence was higher in participants of South-Asian Surinamese (25·0%; adjusted hazard ratio [aHR]=1·66; 95%CI=1·16-2·40), African Surinamese (28·9%, aHR=1·97; 95%CI=1·37-2·83), Turkish (37·0%; aHR=2·67; 95%CI=1·89-3·78), Moroccan (41·9%; aHR=3·13; 95%CI=2·22-4·42), and Ghanaian (64·6%; aHR=6·00; 95%CI=4·33-8·30) origin. Compared to those of Dutch origin, differences in incidence became wider during the second versus first wave for all ethnic minority groups (all p-values for interaction<0·05), except Ghanaians. Having household members with suspected SARS-CoV-2 infection, larger household size, and low health literacy were common determinants of SARS-CoV-2 incidence across groups.

INTERPRETATION

SARS-CoV-2 incidence was higher in the largest ethnic minority groups of Amsterdam, particularly during the second wave. Prevention measures, including vaccination, should be encouraged in these groups.

FUNDING

ZonMw, Public Health Service of Amsterdam, Dutch Heart Foundation, European Union, European Fund for the Integration of non-EU immigrants.

SARS-CoV-2 antibody prevalence and correlates of six ethnic groups living in Amsterdam, the Netherlands: a population-based cross-sectional study, June-October 2020

OBJECTIVES

It has been suggested that ethnic minorities have been disproportionally affected by the COVID-19. We aimed to determine whether prevalence and correlates of past SARS-CoV-2 exposure varied between six ethnic groups in Amsterdam, the Netherlands.

DESIGN, SETTING, PARTICIPANTS

Participants aged 25-79 years enrolled in the Healthy Life in an Urban Setting population-based prospective cohort (n=16 889) were randomly selected within ethnic groups and invited to participate in a cross-sectional COVID-19 seroprevalence substudy.

OUTCOME MEASURES

We tested participants for SARS-CoV-2-specific antibodies and collected information on SARS-CoV-2 exposures. We estimated prevalence and correlates of SARS-CoV-2 exposure within ethnic groups using survey-weighted logistic regression adjusting for age, sex and calendar time.

RESULTS

Between 24 June and 9 October 2020, we included 2497 participants. Adjusted SARS-CoV-2 seroprevalence was comparable between ethnic Dutch (24/498; 5.1%, 95% CI 2.8% to 7.4%), South-Asian Surinamese (22/451; 4.9%, 95% CI 2.2% to 7.7%), African Surinamese (22/400; 8.3%, 95% CI 3.1% to 13.6%), Turkish (30/408; 7.9%, 95% CI 4.4% to 11.4%) and Moroccan (32/391; 7.2%, 95% CI 4.2% to 10.1%) participants, but higher among Ghanaians (95/327; 26.3%, 95% CI 18.5% to 34.0%). 57.1% of SARS-CoV-2-positive participants did not suspect or were unsure of being infected, which was lowest in African Surinamese (18.2%) and highest in Ghanaians (90.5%). Correlates of SARS-CoV-2 exposure varied across ethnic groups, while the most common correlate was having a household member suspected of infection. In Ghanaians, seropositivity was associated with older age, larger household sizes, living with small children, leaving home to work and attending religious services.

CONCLUSIONS

No remarkable differences in SARS-CoV-2 seroprevalence were observed between the largest ethnic groups in Amsterdam after the first wave of infections. The higher infection seroprevalence observed among Ghanaians, which passed mostly unnoticed, warrants wider prevention efforts and opportunities for non-symptom-based testing.

Age-Stratified Risk of Cerebral Venous Sinus Thrombosis After SARS-CoV-2 Vaccination

BACKGROUND AND OBJECTIVES

Cerebral Venous Sinus Thrombosis (CVST) as a part of the thrombosis and thrombocytopenia syndrome is a rare adverse drug reaction of SARS-CoV-2 vaccination. Estimated background rate of CVST with thrombocytopenia is 0.1 per million per month. We assessed the age-stratified risk of CVST with and without thrombocytopenia after SARS-CoV-2 vaccination.

METHODS

We estimated the absolute risk of CVST with and without thrombocytopenia within 28 days of first dose of four SARS-CoV-2 vaccinations, using data from the European Medicines Agency's EudraVigilance database (until 13 June 2021). As a denominator, we used data on vaccine delivery from 31 European countries. For 22.8 million adults from 25 countries, we estimated the absolute risk of CVST after the first dose of ChAdOx1 nCov-19 per age category.

RESULTS

The absolute risk of CVST within 28 days of first dose vaccination was 7.5 (95%CI 6.9-8.3), 0.7 (95%CI 0.2-2.4), 0.6 (95%CI 0.5-0.7) and 0.6 (95%CI 0.3-1.1) per million of first doses of ChAdOx1 nCov-19, Ad26.COV2.S, BNT162b2 and mRNA-1273, respectively. The absolute risk of CVST with thrombocytopenia within 28 days of first dose vaccination was 4.4 (95%CI 3.9-4.9), 0.7 (95%CI 0.2-2.4), 0.0 (95%CI 0.0-0.1) and 0.0 (95%CI 0.0-0.2) per million of first doses of ChAdOx1 nCov-19, Ad26.COV2.S, BNT162b2 and mRNA-1273, respectively. In recipients of ChAdOx1 nCov-19, the absolute risk of CVST, both with and without thrombocytopenia, was the highest in the 18-24 years age group (7.3 per million, 95%CI 2.8-18.8 and 3.7 per million, 95%CI 1.0-13.3, respectively). The risk of CVST with thrombocytopenia in ChAdOx1 nCov-19 recipients was the lowest in the age group≥70 years (0.2, 95%CI 0.0-1.3). Age <60 compared to ≥60 was a predictor for CVST with thrombocytopenia (incidence rate ratio 5.79; 95%CI 2.98-11.24, p<0.001).

DISCUSSION

The risk of CVST with thrombocytopenia within 28 days of first dose vaccination with ChAdOx1 nCov-19 was higher in younger age groups. The risk of CVST with thrombocytopenia was slightly increased in patients receiving Ad26.COV2.S, compared with the estimated background risk. The risk of CVST with thrombocytopenia was not increased in recipients of SARS-CoV-2 mRNA vaccines.

Genome-wide methylation patterns in Marfan syndrome

Background

Marfan syndrome (MFS) is a connective tissue disorder caused by mutations in the Fibrillin-1 gene (FBN1). Here, we undertook the first epigenome-wide association study (EWAS) in patients with MFS aiming at identifying DNA methylation loci associated with MFS phenotypes that may shed light on the disease process.

Methods

The Illumina 450 k DNA-methylation array was used on stored peripheral whole-blood samples of 190 patients with MFS originally included in the COMPARE trial. An unbiased genome-wide approach was used, and methylation of CpG-sites across the entire genome was evaluated. Additionally, we investigated CpG-sites across the FBN1-locus (15q21.1) more closely, since this is the gene defective in MFS. Differentially Methylated Positions (DMPs) and Differentially Methylated Regions (DMRs) were identified through regression analysis. Associations between methylation levels and aortic diameters and presence or absence of 21 clinical features of MFS at baseline were analyzed. Moreover, associations between aortic diameter change, and the occurrence of clinical events (death any cause, type-A or -B dissection/rupture, or aortic surgery) and methylation levels were analyzed.

Results

We identified 28 DMPs that are significantly associated with aortic diameters in patients with MFS. Seven of these DMPs (25%) could be allocated to a gene that was previously associated with cardiovascular diseases (HDAC4, IGF2BP3, CASZ1, SDK1, PCDHGA1, DIO3, PTPRN2). Moreover, we identified seven DMPs that were significantly associated with aortic diameter change and five DMP’s that associated with clinical events. No significant associations at p < 10^–8 or p < 10^–6 were found with any of the non-cardiovascular phenotypic MFS features. Investigating DMRs, clusters were seen mostly on X- and Y, and chromosome 18–22. The remaining DMRs indicated involvement of a large family of protocadherins on chromosome 5, which were not reported in MFS before.

Conclusion

This EWAS in patients with MFS has identified a number of methylation loci significantly associated with aortic diameters, aortic dilatation rate and aortic events. Our findings add to the slowly growing literature on the regulation of gene expression in MFS patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13148-021-01204-4.

F. L. de Vos F, ten Tije AJ, Beerepoot LV, Tjan‐Heijnen VCG, van Laarhoven HWM, Hamberg P, Vulink AJE, Los M, Zwinderman AH, Ferwerda B, Lolkema MPJK, Steeghs N, Büller HR, Kamphuisen PW, van Es N. Evaluation of the Khorana, PROTECHT, and 5-SNP scores for prediction of venous thromboembolism in patients with cancer

BACKGROUND

The Khorana score is a validated tool to identify cancer patients at higher risk of venous thromboembolism (VTE).

OBJECTIVE

We compared its predictive performance to that of the clinical PROTECHT and the polygenic 5-SNP scores in patients who participated in the Dutch CPCT-02 study.

PATIENTS/METHODS

Data on VTE and its risk factors were retrospectively collected for 2729 patients with advanced stage solid tumors planned for systemic cancer treatment. Patients were followed for 6 months. Overall discriminatory performance of the scores was evaluated by time-dependent c-indices. The scores were additionally evaluated dichotomously in competing risk models.

RESULTS

A total of 160 (5.9%) patients developed VTE during follow-up. Time-dependent c-indices at 6 months for the Khorana, PROTECHT, and 5-SNP scores were 0.57 (95% confidence interval [CI]: 0.55-0.60), 0.60 (95% CI: 0.57-0.62), and 0.54 (95% CI: 0.51-0.57), respectively. The dichotomous scores classified 9.6%, 16.8%, and 9.5% as high-risk, respectively. VTE risk was about 2-fold higher among high-risk patients than low-risk patients for the Khorana (subdistribution hazard ratio [SHR] 1.9, 95% CI: 1.3-3.0), PROTECHT (SHR 2.1, 95% CI: 1.5-3.0), and 5-SNP scores (SHR 1.7, 95% CI: 1.03-2.8). The sensitivity at 6 months was 16.6% (95% CI: 10.5-22.7), 28.9% (95% CI: 21.5-36.3), and 14.9% (95% CI: 8.5-21.2), respectively.

CONCLUSIONS

Performance of the PROTECHT or 5-SNP score was not superior to that of the Khorana score. The majority of cancer patients who developed VTE during 6-month follow-up were not identified by these scores. Future directions for studies on cancer-associated VTE prediction may include combined clinical-genetic scores.

Statin therapy and lipoprotein(a) levels: a systematic review and meta-analysis

AIMS 

Lipoprotein(a) [Lp(a)] is a causal and independent risk factor for cardiovascular disease (CVD). People with elevated Lp(a) are often prescribed statins as they also often show elevated low-density lipoprotein cholesterol (LDL-C) levels. While statins are well-established in lowering LDL-C, their effect on Lp(a) remains unclear. We evaluated the effect of statins compared to placebo on Lp(a) and the effects of different types and intensities of statin therapy on Lp(a).

METHODS AND RESULTS 

We conducted a systematic review and meta-analysis of randomized trials with a statin and placebo arm. Medline and EMBASE were searched until August 2019. Quality assessment of studies was done using Cochrane risk-of-bias tool (RoB 2). Mean difference of absolute and percentage changes of Lp(a) in the statin vs. the placebo arms were pooled using a random-effects meta-analysis. We compared effects of different types and intensities of statin therapy using subgroup- and network meta-analyses. Certainty of the evidence was determined using GRADE (Grading of Recommendations, Assessment, Development, and Evaluation). Overall, 39 studies (24 448 participants) were included. Mean differences (95% confidence interval) of absolute and percentage changes in the statin vs. the placebo arms were 1.1 mg/dL (0.5-1.6, P < 0.0001) and 0.1% (-3.6% to 4.0%, P = 0.95), respectively (moderate-certainty evidence). None of the types of statins changed Lp(a) significantly compared to placebo (very low- to high-certainty evidence), as well as intensities of statin therapy (low- to moderate-certainty evidence).

CONCLUSION 

Statin therapy does not lead to clinically important differences in Lp(a) compared to placebo in patients at risk for CVD. Our findings suggest that in these patients, statin therapy will not change Lp(a)-associated CVD risk.