A new, accurate predictive model for incident hypertension

Application of artificial intelligence in hypertension

Hypertension is an important modifiable risk factor for morbidity and mortality associated with cardiovascular disease. The incidence of hypertension is increasing not only in Korea but also in many Western countries due to the aging of the population and the increase in unhealthy lifestyles. However, hypertension control rates remain low due to poor adherence to antihypertensive medications, low awareness of hypertension, and numerous factors that contribute to hypertension, including diet, environment, lifestyle, obesity, and genetics. Because artificial intelligence (AI) involves data-driven algorithms, AI is an asset to understanding chronic diseases that are influenced by multiple factors, such as hypertension. Although several hypertension studies using AI have been published recently, most are exploratory descriptive studies that are often difficult for clinicians to understand and have little clinical relevance. This review aims to provide a clinician-centered perspective on AI by showing recent studies on the relevance of AI for patients with hypertension. The review is organized into sections on blood pressure measurement and hypertension diagnosis, prognosis, and management.

Prognostic risk models for incident hypertension: A PRISMA systematic review and meta-analysis

OBJECTIVE

Our goal was to review the available literature on prognostic risk prediction for incident hypertension, synthesize performance, and provide suggestions for future work on the topic.

METHODS

A systematic search on PUBMED and Web of Science databases was conducted for studies on prognostic risk prediction models for incident hypertension in generally healthy individuals. Study-quality was assessed using the Prediction model Risk of Bias Assessment Tool (PROBAST) checklist. Three-level meta-analyses were used to obtain pooled AUC/C-statistic estimates. Heterogeneity was explored using study and cohort characteristics in meta-regressions.

RESULTS

From 5090 hits, we found 53 eligible studies, and included 47 in meta-analyses. Only four studies were assessed to have results with low risk of bias. Few models had been externally validated, with only the Framingham risk model validated more than thrice. The pooled AUC/C-statistics were 0.82 (0.77-0.86) for machine learning models and 0.78 (0.76-0.80) for traditional models, with high heterogeneity in both groups (I2 > 99%). Intra-class correlations within studies were 60% and 90%, respectively. Follow-up time (P = 0.0405) was significant for ML models and age (P = 0.0271) for traditional models in explaining heterogeneity. Validations of the Framingham risk model had high heterogeneity (I2 > 99%).

CONCLUSION

Overall, the quality of included studies was assessed as poor. AUC/C-statistic were mostly acceptable or good, and higher for ML models than traditional models. High heterogeneity implies large variability in the performance of new risk models. Further, large heterogeneity in validations of the Framingham risk model indicate variability in model performance on new populations. To enable researchers to assess hypertension risk models, we encourage adherence to existing guidelines for reporting and developing risk models, specifically reporting appropriate performance measures. Further, we recommend a stronger focus on validation of models by considering reasonable baseline models and performing external validations of existing models. Hence, developed risk models must be made available for external researchers.

Development of risk models of incident hypertension using machine learning on the HUNT study data

In this study, we aimed to create an 11-year hypertension risk prediction model using data from the Trøndelag Health (HUNT) Study in Norway, involving 17 852 individuals (20-85 years; 38% male; 24% incidence rate) with blood pressure (BP) below the hypertension threshold at baseline (1995-1997). We assessed 18 clinical, behavioral, and socioeconomic features, employing machine learning models such as eXtreme Gradient Boosting (XGBoost), Elastic regression, K-Nearest Neighbor, Support Vector Machines (SVM) and Random Forest. For comparison, we used logistic regression and a decision rule as reference models and validated six external models, with focus on the Framingham risk model. The top-performing models consistently included XGBoost, Elastic regression and SVM. These models efficiently identified hypertension risk, even among individuals with optimal baseline BP (< 120/80 mmHg), although improvement over reference models was modest. The recalibrated Framingham risk model outperformed the reference models, approaching the best-performing ML models. Important features included age, systolic and diastolic BP, body mass index, height, and family history of hypertension. In conclusion, our study demonstrated that linear effects sufficed for a well-performing model. The best models efficiently predicted hypertension risk, even among those with optimal or normal baseline BP, using few features. The recalibrated Framingham risk model proved effective in our cohort.

Early Diagnosis of Cardiovascular Diseases in the Era of Artificial Intelligence: An In-Depth Review

Cardiovascular diseases (CVDs) are significant health issues that result in high death rates globally. Early detection of cardiovascular events may lower the occurrence of acute myocardial infarction and reduce death rates in people with CVDs. Traditional data analysis is inadequate for managing multidimensional data related to the risk prediction of CVDs, heart attacks, medical image interpretations, therapeutic decision-making, and disease prognosis due to the complex pathological mechanisms and multiple factors involved. Artificial intelligence (AI) is a technology that utilizes advanced computer algorithms to extract information from large databases, and it has been integrated into the medical industry. AI methods have shown the ability to speed up the advancement of diagnosing and treating CVDs such as heart failure, atrial fibrillation, valvular heart disease, hypertrophic cardiomyopathy, congenital heart disease, and more. In clinical settings, AI has shown usefulness in diagnosing cardiovascular illness, improving the efficiency of supporting tools, stratifying and categorizing diseases, and predicting outcomes. Advanced AI algorithms have been intricately designed to analyze intricate relationships within extensive healthcare data, enabling them to tackle more intricate jobs compared to conventional approaches.

Chronic disease outcome metadata from German observational studies - public availability and FAIR principles

Metadata from epidemiological studies, including chronic disease outcome metadata (CDOM), are important to be findable to allow interpretability and reusability. We propose a comprehensive metadata schema and used it to assess public availability and findability of CDOM from German population-based observational studies participating in the consortium National Research Data Infrastructure for Personal Health Data (NFDI4Health). Additionally, principal investigators from the included studies completed a checklist evaluating consistency with FAIR principles (Findability, Accessibility, Interoperability, Reusability) within their studies. Overall, six of sixteen studies had complete publicly available CDOM. The most frequent CDOM source was scientific publications and the most frequently missing metadata were availability of codes of the International Classification of Diseases, Tenth Revision (ICD-10). Principal investigators' main perceived barriers for consistency with FAIR principles were limited human and financial resources. Our results reveal that CDOM from German population-based studies have incomplete availability and limited findability. There is a need to make CDOM publicly available in searchable platforms or metadata catalogues to improve their FAIRness, which requires human and financial resources.

Risk-prediction model for incident hypertension in patients with obstructive sleep apnea based on SpO2 signals

This work proposes a method utilizing oxygen saturation (SpO2) for predicting incident hypertension in patients with obstructive sleep apnea (OSA). We extracted time domain features and frequency domain features from the SpO2 signal. For prediction, we employed several machine learning algorithms to establish the 3-year risk prediction model in the Chinese Sleep Health Study, including 250 subjects without baseline hypertension who underwent sleep monitoring. The proposed random forest model achieved an accuracy of 84.4%, a sensitivity of 77.0%, a specificity of 91.5% and an area under the receiver operator characteristic of 84.3% using 10-fold crossvalidation. We show that the model outperformed two sleep medicine specialists using clinical experience to predict hypertension. Furthermore, we applied the prediction results in the public Sleep Heart Health Study database and showed the subjects who were predicted to have hypertension would be at a higher risk in 4-6 years. This work shows the potential of SpO2 signal during sleep for the prediction of hypertension and could be beneficial to the early detection and timely treatment of hypertension in OSA patients.Clinical Relevance-There is no prediction model for incident hypertension in OSA patients in clinical practice. Most patients are unaware of health complexity, symptoms and risk factors before hypertension. Establishing an accurate prediction model can effectively provide early intervention for OSA patients and reduce the prevalence of hypertension.

Estimation of Physiologic Pressures: Invasive and Non-Invasive Techniques, AI Models, and Future Perspectives

The measurement of physiologic pressure helps diagnose and prevent associated health complications. From typical conventional methods to more complicated modalities, such as the estimation of intracranial pressures, numerous invasive and noninvasive tools that provide us with insight into daily physiology and aid in understanding pathology are within our grasp. Currently, our standards for estimating vital pressures, including continuous BP measurements, pulmonary capillary wedge pressures, and hepatic portal gradients, involve the use of invasive modalities. As an emerging field in medical technology, artificial intelligence (AI) has been incorporated into analyzing and predicting patterns of physiologic pressures. AI has been used to construct models that have clinical applicability both in hospital settings and at-home settings for ease of use for patients. Studies applying AI to each of these compartmental pressures were searched and shortlisted for thorough assessment and review. There are several AI-based innovations in noninvasive blood pressure estimation based on imaging, auscultation, oscillometry and wearable technology employing biosignals. The purpose of this review is to provide an in-depth assessment of the involved physiologies, prevailing methodologies and emerging technologies incorporating AI in clinical practice for each type of compartmental pressure measurement. We also bring to the forefront AI-based noninvasive estimation techniques for physiologic pressure based on microwave systems that have promising potential for clinical practice.

Development and Validation of a Hypertension Risk Calculator for Living Kidney Donors

BACKGROUND

Ideally, when deciding whether to donate, kidney donor candidates would understand their long-term risks. Using single-center data (N = 4055; median [quartiles] follow-up: 18 [9-28] y), we developed a calculator for postdonation hypertension and validated it using long-term data from an external single-center cohort (N = 1189, median [quartiles] follow-up: 9 [5-17] y).

METHODS

Risk factors considered were routinely obtained at evaluation from donor candidates. Two modeling approaches were evaluated: Cox proportional hazards and random survival forest models. Cross-validation prediction error and Harrell's concordance-index were used to compare accuracy for model development. Top-performing models were assessed in the validation cohort using the concordance-index and net reclassification improvement.

RESULTS

In the development cohort, 34% reported hypertension at a median (quartiles) of 16 (8-24) y postdonation; and in the validation cohort, 29% reported hypertension after 17 (10-22) y postdonation. The most accurate model was a Cox proportional hazards model with age, sex, race, estimated glomerular filtration rate, systolic and diastolic blood pressure, body mass index, glucose, smoking history, family history of hypertension, relationship with recipient, and hyperlipidemia (concordance-index, 0.72 in the development cohort and 0.82 in the validation cohort).

CONCLUSIONS

A postdonation hypertension calculator was developed and validated; it provides kidney donor candidates, their family, and care team a long-term projection of hypertension risk that can be incorporated into the informed consent process.

A comparison of machine learning algorithms and traditional regression-based statistical modeling for predicting hypertension incidence in a Canadian population

Risk prediction models are frequently used to identify individuals at risk of developing hypertension. This study evaluates different machine learning algorithms and compares their predictive performance with the conventional Cox proportional hazards (PH) model to predict hypertension incidence using survival data. This study analyzed 18,322 participants on 24 candidate features from the large Alberta's Tomorrow Project (ATP) to develop different prediction models. To select the top features, we applied five feature selection methods, including two filter-based: a univariate Cox p-value and C-index; two embedded-based: random survival forest and least absolute shrinkage and selection operator (Lasso); and one constraint-based: the statistically equivalent signature (SES). Five machine learning algorithms were developed to predict hypertension incidence: penalized regression Ridge, Lasso, Elastic Net (EN), random survival forest (RSF), and gradient boosting (GB), along with the conventional Cox PH model. The predictive performance of the models was assessed using C-index. The performance of machine learning algorithms was observed, similar to the conventional Cox PH model. Average C-indexes were 0.78, 0.78, 0.78, 0.76, 0.76, and 0.77 for Ridge, Lasso, EN, RSF, GB and Cox PH, respectively. Important features associated with each model were also presented. Our study findings demonstrate little predictive performance difference between machine learning algorithms and the conventional Cox PH regression model in predicting hypertension incidence. In a moderate dataset with a reasonable number of features, conventional regression-based models perform similar to machine learning algorithms with good predictive accuracy.

Population Based Average Parotid Gland Volume and Prevalence of Incidental Tumors in T1-MRI

Representative epidemiologic data on the average volume of the parotid gland in a large population-based MRI survey is non-existent. Within the Study of Health in Pomerania (SHIP), we examined the parotid gland in 1725 non-contrast MRI-scans in T1 weighted sequence of axial layers. Thus, a reliable standard operating procedure (Intraclass Correlation Coefficient > 0.8) could be established. In this study, we found an average, single sided parotid gland volume of 27.82 cm3 (95% confidence interval (CI) 27.15 to 28.50) in male and 21.60 cm3 (95% CI 21.16 to 22.05) in female subjects. We observed positive associations for age, body mass index (BMI), as well as male sex with parotid gland size in a multivariate model. The prevalence of incidental tumors within the parotid gland regardless of dignity was 3.94% in the Northeast German population, slightly higher than assumed. Further epidemiologic investigations regarding primary salivary gland diseases are necessary.

Prediction of hypertension using traditional regression and machine learning models: A systematic review and meta-analysis

OBJECTIVE

We aimed to identify existing hypertension risk prediction models developed using traditional regression-based or machine learning approaches and compare their predictive performance.

METHODS

We systematically searched MEDLINE, EMBASE, Web of Science, Scopus, and the grey literature for studies predicting the risk of hypertension among the general adult population. Summary statistics from the individual studies were the C-statistic, and a random-effects meta-analysis was used to obtain pooled estimates. The predictive performance of pooled estimates was compared between traditional regression-based models and machine learning-based models. The potential sources of heterogeneity were assessed using meta-regression, and study quality was assessed using the PROBAST (Prediction model Risk Of Bias ASsessment Tool) checklist.

RESULTS

Of 14,778 articles, 52 articles were selected for systematic review and 32 for meta-analysis. The overall pooled C-statistics was 0.75 [0.73-0.77] for the traditional regression-based models and 0.76 [0.72-0.79] for the machine learning-based models. High heterogeneity in C-statistic was observed. The age (p = 0.011), and sex (p = 0.044) of the participants and the number of risk factors considered in the model (p = 0.001) were identified as a source of heterogeneity in traditional regression-based models.

CONCLUSION

We attempted to provide a comprehensive evaluation of hypertension risk prediction models. Many models with acceptable-to-good predictive performance were identified. Only a few models were externally validated, and the risk of bias and applicability was a concern in many studies. Overall discrimination was similar between models derived from traditional regression analysis and machine learning methods. More external validation and impact studies to implement the hypertension risk prediction model in clinical practice are required.

Prediction Models for the 5- and 10-Year Incidence of Home Morning Hypertension: The Ohasama Study

BACKGROUND

We aimed to develop risk prediction models for new-onset home morning hypertension.

METHODS

We followed up 978 participants without home hypertension in the general population of Ohasama, Japan (men: 30.1%, age: 53.3 years). The participants were divided into derivation (n = 489) and validation (n = 489) cohorts by their residential area. The C-statistics and calibration plots were assessed after the 5- or 10-year follow-up.

RESULTS

In the derivation cohort, sex, age, body mass index, smoking, office systolic blood pressure (SBP), and home SBP at baseline were selected as significant risk factors for new-onset home hypertension (≥135/85 mm Hg or the initiation of antihypertensive treatment) using the Cox model. In the validation cohort, Harrell's C-statistic for the 5-/10-year home hypertension was 0.7637 (0.7195-0.8100)/0.7308 (0.6932-0.7677), when we used the full model, which included the significant risk factors in the derivation cohort. The calibration test revealed good concordance between the observed and predicted 5-/10-year home hypertension probabilities (P ≥ 0.19); the regression slope of the observed probability on the predicted probability was 1.10/1.02, and the intercept was -0.04/0.06, respectively. A model without home SBP was also developed; for the 10-year home hypertension risk, the calibration test revealed a good concordance (P = 0.19) but Harrell's C-statistic was 0.6689 (0.6266-0.7067).

CONCLUSIONS

The full model revealed good ability to predict the 5- and 10-year home morning hypertension risk. Although the model without home SBP is acceptable, the low C-statistic implies that home BP should be measured to predict home morning hypertension precisely.

The Use of Machine Learning for the Care of Hypertension and Heart Failure

Machine learning (ML) is a branch of artificial intelligence that combines computer science, statistics, and decision theory to learn complex patterns from voluminous data. In the last decade, accumulating evidence has shown the utility of ML for prediction, diagnosis, and classification of hypertension and heart failure (HF). In addition, ML-enabled image analysis has potential value in assessing cardiac structure and function in an accurate, scalable, and efficient way. Considering the high burden of hypertension and HF in China and worldwide, ML may help address these challenges from different aspects. Indeed, prior studies have shown that ML can enhance each stage of patient care, from research and development, to daily clinical practice and population health. Through reviewing the published literature, the aims of the current systemic review are to summarize the utilities of ML for the care of those with hypertension and HF.

A risk scoring system to predict the risk of new-onset hypertension among patients with type 2 diabetes

Hypertension (HTN), which frequently co‐exists with diabetes mellitus, is the leading major cause of cardiovascular disease and death globally. This study aimed to develop and validate a risk scoring system considering the effects of glycemic and blood pressure (BP) variabilities to predict HTN incidence in patients with type 2 diabetes. This research is a retrospective cohort study that included 3416 patients with type 2 diabetes without HTN and who were enrolled in a managed care program in 2001–2015. The patients were followed up until April 2016, new‐onset HTN event, or death. HTN was defined as diastolic BP (DBP) ≥ 90 mm Hg, systolic BP (SBP) ≥ 140 mm Hg, or the initiation of antihypertensive medication. Cox proportional hazard regression model was used to develop the risk scoring system for HTN. Of the patients, 1738 experienced new‐onset HTN during an average follow‐up period of 3.40 years. Age, sex, physical activity, body mass index, type of DM treatment, family history of HTN, baseline SBP and DBP, variabilities of fasting plasma glucose, SBP, and DBP and macroalbuminuria were significant variables for the prediction of new‐onset HTN. Using these predictors, the prediction models for 1‐, 3‐, and 5‐year periods demonstrated good discrimination, with AUC values of 0.70–0.76. Our HTN scoring system for patients with type 2 DM, which involves innovative predictors of glycemic and BP variabilities, has good classification accuracy and identifies risk factors available in clinical settings for prevention of the progression to new‐onset HTN.

Genome-Wide and Candidate Gene Association Analyses Identify a 14-SNP Combination for Hypertension in Patients With Type 2 Diabetes

BACKGROUND

High blood pressure is common and comorbid with type 2 diabetes (T2D). Almost 50% of patients with T2D have high blood pressure. Patients with both conditions of hypertension (HTN) and T2D are at risk for cardiovascular diseases and mortality. The study aim was to investigate genetic risk factors for HTN in T2D patients.

METHODS

This study included 999 T2D (cohort 1) patients for the first genome scan stage and 922 T2D (cohort 2) patients for the replication stage. Here, we investigated the genetic susceptibility and cumulative weighted genetic risk score for HTN in T2D patients of Han Chinese descent in Taiwan.

RESULTS

Thirty novel genetic single nucleotide polymorphisms (SNPs) were associated with HTN in T2D after adjusting for age and body mass index (P value <1 × 10-4). Eight blood pressure-related and/or HTN-related genetic SNPs were associated with HTN in T2D after adjusting for age and body mass index (P value <0.05). Linkage disequilibrium and cumulative weighted genetic risk score analyses showed that 14 of the 38 SNPs were associated with risk of HTN in a dose-dependent manner in T2D (Cochran-Armitage trend test: P value <0.0001). The 14-SNP cumulative weighted genetic risk score was also associated with increased regression tendency of systolic blood pressure in T2D (SBP = 122.05 + 0.8 × weighted genetic risk score; P value = 0.0001).

CONCLUSIONS

A cumulative weighted genetic risk score composed of 14 SNPs is important for HTN, increased tendency of systolic blood pressure, and may contribute to HTN risk in T2D in Taiwan.

Development of the prediction model for hypertension in patients with idiopathic inflammatory myopathies

Cardiac involvement is an important cause of morbidity and mortality in patients with idiopathic inflammatory myopathies (IIMs). Hypertension, an important cardiovascular risk factor for the general population, has a crucial role in heart involvement. However, few studies have focused on the hypertension associated with IIMs. This study aimed to develop and assess the prediction model for incident hypertension in patients with IIMs. A retrospective cohort study was performed on 362 patients with IIMs, of whom 54 (14.9%) were given a diagnosis of new-onset hypertension from January 2008 to December 2018. The predictors of hypertension in IIMs were selected by least absolute shrinkage and selection operator (LASSO) regression, multivariable logistic regression, and clinically relevance, and then these predictors were used to draw the nomogram. Discrimination, calibration and clinical usefulness of the model were evaluated using the C-index, calibration plot, and decision curve analysis, respectively. The predicting model was validated by the bootstrapping validation. The nomogram mainly included predictors such as age, diabetes mellitus, triglyceride, low-density lipoprotein-cholesterol (LDL-C), antinuclear antibodies (ANA), and smoking. This prediction model demonstrated good discrimination with a C-index of 0.754 (95%CI, 0.684 to 0.824) and good calibration. The C-index of internal validation was 0.728, and decision curve analysis demonstrated that this nomogram was clinically useful. Clinicians can use this prediction model to assess the risk of hypertension in IIMs patients, and early preventive measures should be taken to reduce the incidence of hypertension in high-risk patients.

Applications of artificial intelligence for hypertension management

The prevalence of hypertension is increasing along with an aging population, causing millions of premature deaths annually worldwide. Low awareness of blood pressure (BP) elevation and suboptimal hypertension diagnosis serve as the major hurdles in effective hypertension management. The advent of artificial intelligence (AI), however, sheds the light of new strategies for hypertension management, such as remote supports from telemedicine and big data-derived prediction. There is considerable evidence demonstrating the feasibility of AI applications in hypertension management. A foreseeable trend was observed in integrating BP measurements with various wearable sensors and smartphones, so as to permit continuous and convenient monitoring. In the meantime, further investigations are advised to validate the novel prediction and prognostic tools. These revolutionary developments have made a stride toward the future model for digital management of chronic diseases.

From heterogeneous healthcare data to disease-specific biomarker networks: A hierarchical Bayesian network approach

In this work, we introduce an entirely data-driven and automated approach to reveal disease-associated biomarker and risk factor networks from heterogeneous and high-dimensional healthcare data. Our workflow is based on Bayesian networks, which are a popular tool for analyzing the interplay of biomarkers. Usually, data require extensive manual preprocessing and dimension reduction to allow for effective learning of Bayesian networks. For heterogeneous data, this preprocessing is hard to automatize and typically requires domain-specific prior knowledge. We here combine Bayesian network learning with hierarchical variable clustering in order to detect groups of similar features and learn interactions between them entirely automated. We present an optimization algorithm for the adaptive refinement of such group Bayesian networks to account for a specific target variable, like a disease. The combination of Bayesian networks, clustering, and refinement yields low-dimensional but disease-specific interaction networks. These networks provide easily interpretable, yet accurate models of biomarker interdependencies. We test our method extensively on simulated data, as well as on data from the Study of Health in Pomerania (SHIP-TREND), and demonstrate its effectiveness using non-alcoholic fatty liver disease and hypertension as examples. We show that the group network models outperform available biomarker scores, while at the same time, they provide an easily interpretable interaction network.

High-dimensional and heterogeneous healthcare data, such as electronic health records or epidemiological study data, contain much information on yet unknown risk factors that are associated with disease development. The identification of these risk factors may help to improve prevention, diagnosis, and therapy. Bayesian networks are powerful statistical models that can decipher these complex relationships. However, high dimensionality and heterogeneity of data, together with missing values and high feature correlation, make it difficult to automatically learn a good model from data. To facilitate the use of network models, we present a novel, fully automated workflow that combines network learning with hierarchical clustering. The algorithm reveals groups of strongly related features and models the interactions among those groups. It results in simpler network models that are easier to analyze. We introduce a method of adaptive refinement of such models to ensure that disease-relevant parts of the network are modeled in great detail. Our approach makes it easy to learn compact, accurate, and easily interpretable biomarker interaction networks. We test our method extensively on simulated data as well as data from the Study of Health in Pomerania (SHIP-Trend) by learning models of hypertension and non-alcoholic fatty liver disease.

Identifying Interactions between Dietary Sodium, Potassium, Sodium-Potassium Ratios, and FGF5 rs16998073 Variants and Their Associated Risk for Hypertension in Korean Adults

Hypertension is affected by both genetic and dietary factors. This study aimed to examine the interaction between dietary sodium/potassium intake, sodium–potassium ratios, and FGF5 rs16998073 and link these with increased risk for developing hypertension. Using data from the Health Examinee (HEXA) Study of the Korean Genome and Epidemiologic Study (KoGES), we were able to identify a total of 17,736 middle-aged Korean adults who could be included in our genome-wide association study (GWAS) to confirm any associations between hypertension and the FGF5 rs16998073 variant. GWAS analysis revealed that the FGF5 rs16698073 variant demonstrated the strongest association with hypertension in this population. Multivariable logistic regression was used to examine the relationship between dietary intake of sodium, potassium, and sodium–potassium ratios and the FGF5 rs16998073 genotypes (AA, AT, TT) and any increased risk of hypertension. Carriers with at least one minor T allele for FGF5 rs16998073 were shown to be at significantly higher risk for developing hypertension. Male TT carriers with a daily sodium intake ≥2000 mg also demonstrated an increased risk for developing hypertension compared to the male AA carriers with daily sodium intake <2000 mg (adjusted odds ratio (AOR) = 2.41, 95% confidence intervals (CIs) = 1.84–3.15, p-interaction < 0.0001). Female AA carriers with a daily potassium intake ≥3500 mg showed a reduced risk for hypertension when compared to female AA carriers with a daily potassium intake <3500 mg (AOR = 0.75. 95% CIs = 0.58–0.95, p-interaction < 0.0001). Male TT carriers in the mid-tertile for sodium–potassium ratio values showed the highest odds ratio for hypertension when compared to male AA carriers in the lowest-tertile for sodium–potassium ratio values (AOR = 3.03, 95% CIs = 2.14–4.29, p-interaction < 0.0001). This study confirmed that FGF5 rs16998073 variants do place their carriers (men and women) at increased risk for developing hypertension. In addition, we showed that high daily intake of sodium exerted a synergistic effect for hypertension when combined with FGF5 rs16998073 variants in both genders and that dietary sodium, potassium, and sodium–potassium ratios all interact with FGF5 rs16998073 and alter the risk of developing hypertension in carriers of either gender among Koreans.

Future possibilities for artificial intelligence in the practical management of hypertension

The use of artificial intelligence in numerous prediction and classification tasks, including clinical research and healthcare management, is becoming increasingly more common. This review describes the current status and a future possibility for artificial intelligence in blood pressure management, that is, the possibility of accurately predicting and estimating blood pressure using large-scale data, such as personal health records and electronic medical records. Individual blood pressure continuously changes because of lifestyle habits and the environment. This review focuses on two topics regarding controlling changing blood pressure: a novel blood pressure measurement system and blood pressure analysis using artificial intelligence. Regarding the novel blood pressure measurement system, we compare the conventional cuff-less method with the analysis of pulse waves using artificial intelligence for blood pressure estimation. Then, we describe the prediction of future blood pressure values using machine learning and deep learning. In addition, we summarize factor analysis using "explainable AI" to solve a black-box problem of artificial intelligence. Overall, we show that artificial intelligence is advantageous for hypertension management and can be used to establish clinical evidence for the practical management of hypertension.

A novel predicted model for hypertension based on a large cross-sectional study

Hypertension is a global public health issue and leading risk for death and disability. It is urgent to search novel methods predicting hypertension. Herein, we chose 73158 samples of physical examiners in central China from June 2008 to June 2018. After strict exclusion processes, 33570 participants with hypertension and 35410 healthy controls were included. We randomly chose 70% samples as the train set and the remaining 30% as the test set. Clinical parameters including age, gender, height, weight, body mass index, triglyceride, total cholesterol, low-density lipoprotein, blood urea nitrogen, uric acid, and creatinine were significantly increased, while high-density lipoprotein was decreased in the hypertension group versus controls. Nine optimal markers were identified by a logistic regression model, and achieved AUC value of 76.52% in the train set and 75.81% in the test set for hypertension. In conclusions, this study is the first to establish predicted models for hypertension using the logistic regression model in Central China, which provide risk factors and novel prediction method to predict and prevent hypertension.

Recent development of risk-prediction models for incident hypertension: An updated systematic review

BACKGROUND

Hypertension is a leading global health threat and a major cardiovascular disease. Since clinical interventions are effective in delaying the disease progression from prehypertension to hypertension, diagnostic prediction models to identify patient populations at high risk for hypertension are imperative.

METHODS

Both PubMed and Embase databases were searched for eligible reports of either prediction models or risk scores of hypertension. The study data were collected, including risk factors, statistic methods, characteristics of study design and participants, performance measurement, etc.

RESULTS

From the searched literature, 26 studies reporting 48 prediction models were selected. Among them, 20 reports studied the established models using traditional risk factors, such as body mass index (BMI), age, smoking, blood pressure (BP) level, parental history of hypertension, and biochemical factors, whereas 6 reports used genetic risk score (GRS) as the prediction factor. AUC ranged from 0.64 to 0.97, and C-statistic ranged from 60% to 90%.

CONCLUSIONS

The traditional models are still the predominant risk prediction models for hypertension, but recently, more models have begun to incorporate genetic factors as part of their model predictors. However, these genetic predictors need to be well selected. The current reported models have acceptable to good discrimination and calibration ability, but whether the models can be applied in clinical practice still needs more validation and adjustment.

Prediction of Postoperative Clinical Recovery of Drop Foot Attributable to Lumbar Degenerative Diseases, via a Bayesian Network

BACKGROUND

Drop foot resulting from degenerative lumbar diseases can impair activities of daily living. Therefore, predictors of recovery of this symptom have been investigated using univariate or/and multivariate analyses. However, the conclusions have been somewhat controversial. Bayesian network models, which are graphic and intuitive to the clinician, may facilitate understanding of the prognosis of drop foot resulting from degenerative lumbar diseases.

QUESTIONS/PURPOSES

(1) To show a layered correlation among predictors of recovery from drop foot resulting from degenerative lumbar diseases; and (2) to develop support tools for clinical decisions to treat drop foot resulting from lumbar degenerative diseases.

METHODS

Between 1993 and 2013, we treated 141 patients with decompressive lumbar spine surgery who presented with drop foot attributable to degenerative diseases. Of those, 102 (72%) were included in this retrospective study because they had drop foot of recent development and had no diseases develop that affect evaluation of drop foot after surgery. Specifically, 28 (20%) patients could not be analyzed because their records were not available at a minimum of 2 years followup after surgery and 11 (8%) were lost owing to postoperative conditions that affect the muscle strength evaluation. Eight candidate variables were sex, age, herniated soft disc, duration of the neurologic injury (duration), preoperative tibialis anterior muscle strength (pretibialis anterior), leg pain, cauda equina syndrome, and number of involved levels. Manual muscle testing was used to assess the tibialis anterior muscle strength. Drop foot was defined as a tibialis anterior muscle strength score of less than 3 of 5 (5 = movement against gravity and full resistance, 4 = movement against gravity and moderate resistance, 3 = movement against gravity through full ROM, 3- = movement against gravity through partial ROM, 2 = movement with gravity eliminated through full ROM, 1 = slight contraction but no movement, and 0 = no contraction). The two outcomes of interest were postoperative tibialis anterior muscle strength (posttibialis anterior) of 3 or greater and posttibialis anterior strength of 4 or greater at 2 years after surgery. We developed two separate Bayesian network models with outcomes of interest for posttibialis anterior strength of 3 or greater and posttibialis anterior strength of 4 or greater. The two outcomes correspond to "good" and "excellent" results based on previous reports, respectively. Direct predictors are defined as variables that have the tail of the arrow connecting the outcome of interest, whereas indirect predictors are defined as variables that have the tail of the arrow connecting either direct predictors or other indirect predictors that have the tail of the arrow connecting direct predictors. Sevenfold cross validation and receiver-operating characteristic (ROC) curve analyses were performed to evaluate the accuracy and robustness of the Bayesian network models.

RESULTS

Both of our Bayesian network models showed that weaker muscle power before surgery (pretibialis anterior ≤ 1) and longer duration of neurologic injury before treatment (> 30 days) were associated with a decreased likelihood of return of function by 2 years. The models for posttibialis anterior muscle strength of 3 or greater and posttibialis anterior muscle strength of 4 or greater were the same in terms of the graphs, showing that the two direct predictors were pretibialis anterior muscle strength (score ≤ 1 or ≥ 2) and duration (≤ 30 days or > 30 days). Age, herniated soft disc, and leg pain were identified as indirect predictors. We developed a decision-support tool in which the clinician can enter pretibialis anterior muscle strength and duration, and from this obtain the probability estimates of posttibialis anterior muscle strength. The probability estimates of posttibialis anterior muscle strength of 3 or greater and posttibialis anterior muscle strength of 4 or greater were 94% and 85%, respectively, in the most-favorable conditions (pretibialis anterior ≥ 2; duration ≤ 30 days) and 18% and 14%, respectively, in the least-favorable conditions (pretibialis anterior ≤ 1; duration > 30 days). On the sevenfold cross validation, the area under the ROC curve yielded means of 0.78 (95% CI, 0.68-0.87) and 0.74 (95% CI, 0.64-0.84) for posttibialis anterior muscle strength of 3 or greater and posttibialis anterior muscle strength of 4 or greater, respectively.

CONCLUSIONS

The results of this study suggest that the clinician can understand intuitively the layered correlation among predictors by Bayesian network models. Based on the models, the decision-support tool successfully provided the probability estimates of posttibialis anterior muscle strength to treat drop foot attributable to lumbar degenerative diseases. These models were shown to be robust on the internal validation but should be externally validated in other populations.

LEVEL OF EVIDENCE

Level III, therapeutic study.

Diagnostic and predictive accuracy of blood pressure screening methods with consideration of rescreening intervals: a systematic review for the U.S. Preventive Services Task Force

BACKGROUND

Elevated blood pressure (BP) is the largest contributing risk factor to all-cause and cardiovascular mortality.

PURPOSE

To update a systematic review on the benefits and harms of screening for high BP in adults and to summarize evidence on rescreening intervals and diagnostic and predictive accuracy of different BP methods for cardiovascular events.

DATA SOURCES

Selected databases searched through 24 February 2014.

STUDY SELECTION

Fair- and good-quality trials and diagnostic accuracy and cohort studies conducted in adults and published in English.

DATA EXTRACTION

One investigator abstracted data, and a second checked for accuracy. Study quality was dual-reviewed.

DATA SYNTHESIS

Ambulatory BP monitoring (ABPM) predicted long-term cardiovascular outcomes independently of office BP (hazard ratio range, 1.28 to 1.40, in 11 studies). Across 27 studies, 35% to 95% of persons with an elevated BP at screening remained hypertensive after nonoffice confirmatory testing. Cardiovascular outcomes in persons who were normotensive after confirmatory testing (isolated clinic hypertension) were similar to outcomes in those who were normotensive at screening. In 40 studies, hypertension incidence after rescreening varied considerably at each yearly interval up to 6 years. Intrastudy comparisons showed at least 2-fold higher incidence in older adults, those with high-normal BP, overweight and obese persons, and African Americans.

LIMITATION

Few diagnostic accuracy studies of office BP methods and protocols in untreated adults.

CONCLUSION

Evidence supports ABPM as the reference standard for confirming elevated office BP screening results to avoid misdiagnosis and overtreatment of persons with isolated clinic hypertension. Persons with BP in the high-normal range, older persons, those with an above-normal body mass index, and African Americans are at higher risk for hypertension on rescreening within 6 years than are persons without these risk factors.

PRIMARY FUNDING SOURCE

Agency for Healthcare Research and Quality.

Development of a risk prediction model for incident hypertension in a working-age Japanese male population

The aim of this study was to develop a risk prediction model for incident hypertension in a Japanese male population. Study participants included 15,025 nonhypertensive Japanese male workers (mean age, 38.8±8.9 years) who underwent an annual medical checkup at a company. The participants were followed-up for a median of 4.0 years to determine new-onset hypertension, defined as a systolic blood pressure (BP) ⩾140 mm Hg, a diastolic BP ⩾90 mm Hg, or the initiation of antihypertensive medication. Participants were divided into the following two cohorts for subsequent analyses: the derivation cohort (n=12,020, 80% of the study population) and the validation cohort (n=3005, the remaining 20% of the study population). In the derivation cohort, a multivariate Cox proportional hazards model demonstrated that age, body mass index, systolic and diastolic BP, current smoking status, excessive alcohol intake and parental history of hypertension were independent predictors of incident hypertension. Using these variables, a risk prediction model was constructed to estimate the 4-year risk of incident hypertension. In the validation cohort, the risk prediction model demonstrated high discrimination ability and acceptable calibration, with a C-statistic of 0.861 (95% confidence interval 0.844, 0.877) and a modified Hosmer-Lemeshow χ2 statistic of 15.2 (P=0.085). A risk score sheet was constructed to enable the simple calculation of the approximate 4-year probability of incident hypertension. In conclusion, a practical risk prediction model for incident hypertension was successfully developed in a working-age Japanese male population.