Assessing Machine Learning for Diagnostic Classification of Hypertension Types Identified by Ambulatory Blood Pressure Monitoring.

Background: Inaccurate blood pressure (BP) classification results in inappropriate treatment. We tested whether machine learning (ML), using routine clinical data, can serve as a reliable alternative to ambulatory BP monitoring (ABPM) in classifying BP status. Methods: This study employed a multicentre approach involving 3 derivation cohorts from Glasgow, Gdansk, and Birmingham, and a fourth independent evaluation cohort. ML models were trained using office BP, ABPM, and clinical, laboratory, and demographic data, collected from patients referred for hypertension assessment. Seven ML algorithms were trained to classify patients into 5 groups, named as follows: Normal/Target; Hypertension-Masked; Normal/Target-White-Coat (WC); Hypertension-WC; and Hypertension. The 10-year cardiovascular outcomes and 27-year all-cause mortality risks were calculated for the ML-derived groups using the Cox proportional hazards model. Results: Overall, extreme gradient boosting (using XGBoost open source software) showed the highest area under the receiver operating characteristic curve of 0.85-0.88 across derivation cohorts, Glasgow (n = 923; 43% female; age 50.7 ± 16.3 years), Gdansk (n = 709; 46% female; age 54.4 ± 13 years), and Birmingham (n = 1222; 56% female; age 55.7 ± 14 years). But accuracy (0.57-0.72) and F1 (harmonic mean of precision and recall) scores (0.57-0.69) were low across the 3 patient cohorts. The evaluation cohort (n = 6213; 51% female; age 51.2 ± 10.8 years) indicated elevated 10-year risks of composite cardiovascular events in the Normal/Target-WC and the Hypertension-WC groups, with heightened 27-year all-cause mortality observed in all groups, except the Hypertension-Masked group, compared to the Normal/Target group. Conclusions: ML has limited potential in accurate BP classification when ABPM is unavailable. Larger studies including diverse patient groups and different resource settings are warranted.

Contexte: Les erreurs dans la classification des valeurs de la pression artérielle (PA) entraînent une inadéquation du traitement. Nous avons tâché de déterminer si l'apprentissage machine, à l'aide de données cliniques routinières, constituait une solution de rechange fiable à la surveillance ambulatoire de la PA pour définir le statut de la PA. Méthodologie: Cette étude a utilisé une approche multicentrique incluant trois cohortes de dérivation de Glasgow, Gdansk et Birmingham, et une quatrième cohorte d'évaluation indépendante. Les modèles d'apprentissage machine ont été développés en analysant les données démographiques, les valeurs de la PA mesurée au cabinet, les données relatives à la surveillance ambulatoire de la PA et aux épreuves de laboratoire recueillies auprès de patients adressés pour une évaluation de l'hypertension. Sept algorithmes d'apprentissage machine ont été appliqués pour classer les patients en cinq groupes : Normale/Cible; Hypertension-Masquée; Normal/Cible-Blouse blanche; Hypertension-Blouse blanche; Hypertension. Les événements cardiovasculaires sur 10 ans et le risque de mortalité toutes causes confondues sur 27 ans ont été calculés dans les groupes dérivés de l'apprentissage machine à l'aide d'un modèle de risques proportionnels de Cox. Résultats: D'une manière générale, l'amplification de gradient extrême (à l'aide du logiciel ouvert XGBoost) a mis en évidence l'aire sous la courbe de la fonction d'efficacité du récepteur (courbe ROC pour Receiver Operating Characteristic) la plus haute, soit 0,85 à 0,88, pour toutes les cohortes de dérivation : Glasgow (n = 923; 43 % de femmes; âge : 50,7 ± 16,3 ans); Gdansk (n = 709; 46 % de femmes; âge : 54,4 ± 13 ans); Birmingham (n = 1 222; 56 % de femmes; âge : 55,7 ± 14 ans). La précision (0,57 ­ 0,72) et le score F1 (moyenne harmonique de la précision et du rappel) (0,57 ­ 0,69) ont été faibles dans les trois cohortes de patients. La cohorte d'évaluation (n = 6 213; 51 % de femmes; âge : 51,2 ± 10,8 ans) a indiqué un risque d'événements cardiovasculaires composites sur 10 ans élevé dans les groupes Normale/Cible-Blouse blanche et Hypertension-Blouse blanche, tandis qu'une hausse de la mortalité toutes causes confondues sur 27 ans a été observée dans tous les groupes, sauf dans le groupe Hypertension-Masquée, comparativement au groupe Normale/Cible. Conclusions: Le potentiel d'exactitude de la classification de la PA à l'aide de l'apprentissage machine lorsque la surveillance ambulatoire de la PA n'est pas possible est limité. Des études de plus grande envergure portant sur des groupes de patients et des niveaux de ressources diversifiés s'imposent.

Establishing plausibility of cardiovascular adverse effects of immunotherapies using Mendelian randomisation.

Immune checkpoint inhibitors (ICIs) and Janus kinase inhibitors (JAKis) have raised concerns over serious unexpected cardiovascular adverse events. The widespread pleiotropy in genome-wide association studies offers an opportunity to identify cardiovascular risks from in-development drugs to help inform appropriate trial design and pharmacovigilance strategies. This study uses the Mendelian randomization (MR) approach to study the causal effects of 9 cardiovascular risk factors on ischemic stroke risk both independently and by mediation, followed by an interrogation of the implicated expression quantitative trait loci (eQTLs) to determine if the enriched pathways can explain the adverse stroke events observed with ICI or JAKi treatment. Genetic predisposition to higher systolic blood pressure (SBP), diastolic blood pressure (DBP), body mass index (BMI), waist-to-hip ratio (WHR), low-density lipoprotein cholesterol (LDL), triglycerides (TG), type 2 diabetes (T2DM), and smoking index were associated with higher ischemic stroke risk. The associations of genetically predicted BMI, WHR, and TG on the outcome were attenuated after adjusting for genetically predicted T2DM [BMI: 53.15% mediated, 95% CI 17.21%-89.10%; WHR: 42.92% (4.17%-81.67%); TG: 72.05% (10.63%-133.46%)]. JAKis, programmed cell death protein 1 and programmed death ligand 1 inhibitors were implicated in the pathways enriched by the genes related to the instruments for each of SBP, DBP, WHR, T2DM, and LDL. Overall, MR mediation analyses support the role of T2DM in mediating the effects of BMI, WHR, and TG on ischemic stroke risk and follow-up pathway enrichment analysis highlights the utility of this approach in the early identification of potential harm from drugs.

Survey and Evaluation of Hypertension Machine Learning Research.

Background Machine learning (ML) is pervasive in all fields of research, from automating tasks to complex decision-making. However, applications in different specialities are variable and generally limited. Like other conditions, the number of studies employing ML in hypertension research is growing rapidly. In this study, we aimed to survey hypertension research using ML, evaluate the reporting quality, and identify barriers to ML's potential to transform hypertension care. Methods and Results The Harmonious Understanding of Machine Learning Analytics Network survey questionnaire was applied to 63 hypertension-related ML research articles published between January 2019 and September 2021. The most common research topics were blood pressure prediction (38%), hypertension (22%), cardiovascular outcomes (6%), blood pressure variability (5%), treatment response (5%), and real-time blood pressure estimation (5%). The reporting quality of the articles was variable. Only 46% of articles described the study population or derivation cohort. Most articles (81%) reported at least 1 performance measure, but only 40% presented any measures of calibration. Compliance with ethics, patient privacy, and data security regulations were mentioned in 30 (48%) of the articles. Only 14% used geographically or temporally distinct validation data sets. Algorithmic bias was not addressed in any of the articles, with only 6 of them acknowledging risk of bias. Conclusions Recent ML research on hypertension is limited to exploratory research and has significant shortcomings in reporting quality, model validation, and algorithmic bias. Our analysis identifies areas for improvement that will help pave the way for the realization of the potential of ML in hypertension and facilitate its adoption.

A genomic deep field view of hypertension.

Blood pressure is regulated by a complex neurohumoral system including the renin-angiotensin-aldosterone system, natriuretic peptides, endothelial pathways, the sympathetic nervous system, and the immune system. This review charts the evolution of our understanding of the genomic basis of hypertension at increasing resolution over the last 5 decades from monogenic causes to polygenic associations, spanning ∼30 monogenic rare variants and >1500 single nucleotide variants. Unexpected early wins from blood pressure genomics include deepening of our understanding of the complex causation of hypertension; refinement of causal estimates bidirectionally between blood pressure, risk factors, and outcomes through Mendelian randomization; risk stratification using polygenic risk scores; and opportunities for precision medicine and drug repurposing.

Cardiovascular precision medicine - A pharmacogenomic perspective.

Precision medicine envisages the integration of an individual's clinical and biological features obtained from laboratory tests, imaging, high-throughput omics and health records, to drive a personalised approach to diagnosis and treatment with a higher chance of success. As only up to half of patients respond to medication prescribed following the current one-size-fits-all treatment strategy, the need for a more personalised approach is evident. One of the routes to transforming healthcare through precision medicine is pharmacogenomics (PGx). Around 95% of the population is estimated to carry one or more actionable pharmacogenetic variants and over 75% of adults over 50 years old are on a prescription with a known PGx association. Whilst there are compelling examples of pharmacogenomic implementation in clinical practice, the case for cardiovascular PGx is still evolving. In this review, we shall summarise the current status of PGx in cardiovascular diseases and look at the key enablers and barriers to PGx implementation in clinical practice.