Transforming Hypertension Diagnosis and Management in The Era of Artificial Intelligence: A 2023 National Heart, Lung, and Blood Institute (NHLBI) Workshop Report.

Hypertension is among the most important risk factors for cardiovascular disease, chronic kidney disease, and dementia. The artificial intelligence (AI) field is advancing quickly, and there has been little discussion on how AI could be leveraged for improving the diagnosis and management of hypertension. AI technologies, including machine learning tools, could alter the way we diagnose and manage hypertension, with potential impacts for improving individual and population health. The development of successful AI tools in public health and health care systems requires diverse types of expertise with collaborative relationships between clinicians, engineers, and data scientists. Unbiased data sources, management, and analyses remain a foundational challenge. From a diagnostic standpoint, machine learning tools may improve the measurement of blood pressure and be useful in the prediction of incident hypertension. To advance the management of hypertension, machine learning tools may be useful to find personalized treatments for patients using analytics to predict response to antihypertension medications and the risk for hypertension-related complications. However, there are real-world implementation challenges to using AI tools in hypertension. Herein, we summarize key findings from a diverse group of stakeholders who participated in a workshop held by the National Heart, Lung, and Blood Institute in March 2023. Workshop participants presented information on communication gaps between clinical medicine, data science, and engineering in health care; novel approaches to estimating BP, hypertension risk, and BP control; and real-world implementation challenges and issues.

Branched chain α-ketoacids aerobically activate HIF1α signaling in vascular cells.

Hypoxia-inducible factor 1α (HIF1α) is a master regulator of numerous biological processes under low oxygen tensions. Yet, the mechanisms and biological consequences of aerobic HIF1α activation by intrinsic factors, particularly in primary cells remain elusive. Here, we show that HIF1α signaling is activated in several human primary vascular cells under ambient oxygen tensions, and in vascular smooth muscle cells (VSMCs) of normal human lung tissue, which contributed to a relative resistance to further enhancement of glycolytic activity in hypoxia. Mechanistically, aerobic HIFα activation is mediated by paracrine secretion of three branched chain α-ketoacids (BCKAs), which suppress prolyl hydroxylase domain-containing protein 2 (PHD2) activity via direct inhibition and via lactate dehydrogenase A (LDHA)-mediated generation of L-2-hydroxyglutarate (L2HG). Metabolic dysfunction induced by BCKAs was observed in the lungs of rats with pulmonary arterial hypertension (PAH) and in pulmonary artery smooth muscle cells (PASMCs) from idiopathic PAH patients. BCKA supplementation stimulated glycolytic activity and promoted a phenotypic switch to the synthetic phenotype in PASMCs of normal and PAH subjects. In summary, we identify BCKAs as novel signaling metabolites that activate HIF1α signaling in normoxia and that the BCKA-HIF1α pathway modulates VSMC function and may be relevant to pulmonary vascular pathobiology.

Pulmonary hypertension across the spectrum of left heart and lung disease.

AIMS: Patients with pulmonary hypertension (PH) are grouped based upon clinical and haemodynamic characteristics. Groups 2 (G2, left heart disease [LHD]) and 3 (G3, lung disease or hypoxaemia) are most common. Many patients display overlapping characteristics of heart and lung disease (G2-3), but this group is not well-characterized. METHODS AND RESULTS: Patients with PH enrolled in the prospective, NHLBI-sponsored PVDOMICS network underwent intensive clinical, biomarker, imaging, gas exchange and exercise phenotyping. Patients with pure G2, pure G3, or overlapping G2-3 PH were compared across multiple phenotypic domains. Of all patients with predominant G2 (n = 136), 66 (49%) were deemed to have secondary lung disease/hypoxaemia contributors (G2/3), and of all patients categorized as predominant G3 (n = 172), 41 (24%) were judged to have a component of secondary LHD (G3/2), such that 107 had G2-3 (combined G2/3 and G3/2). As compared with G3, patients with G2 and G2-3 were more obese and had greater prevalence of hypertension, atrial fibrillation, and coronary disease. Patients with G2 and G2-3 were more anaemic, with poorer kidney function, more cardiac dysfunction, and higher N-terminal pro-B-type natriuretic peptide than G3. Lung diffusion was more impaired in G3 and G2-3, but commonly abnormal even in G2. Exercise capacity was severely and similarly impaired across all groups, with no differences in 6-min walk distance or peak oxygen consumption, and pulmonary vasoreactivity to nitric oxide did not differ. In a multivariable Cox regression model, patients with G2 had lower risk of death or transplant compared with G3 (hazard ratio [HR] 0.51, 95% confidence interval [CI] 0.30-0.86), and patients with G2-3 also displayed lower risk compared with G3 (HR 0.57, 95% CI 0.38-0.86). CONCLUSIONS: Overlap is common in patients with a pulmonary or cardiac basis for PH. While lung structure/function is clearly more impaired in G3 and G2-3 than G2, pulmonary abnormalities are common in G2, even when clinically judged as isolated LHD. Further study is required to identify optimal systematic evaluations to guide therapeutic innovation for PH associated with combined heart and lung disease. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov NCT02980887.

Diagnostic Performance of Coronary Angiography Derived Computational Fractional Flow Reserve.

BACKGROUND: Computational fluid dynamics can compute fractional flow reserve (FFR) accurately. However, existing models are limited by either the intravascular hemodynamic phenomarkers that can be captured or the fidelity of geometries that can be modeled. METHODS AND RESULTS: This study aimed to validate a new coronary angiography-based FFR framework, FFRHARVEY, and examine intravascular hemodynamics to identify new biomarkers that could augment FFR in discerning unrevascularized patients requiring intervention. A 2-center cohort was used to examine diagnostic performance of FFRHARVEY compared with reference wire-based FFR (FFRINVASIVE). Additional biomarkers, longitudinal vorticity, velocity, and wall shear stress, were evaluated for their ability to augment FFR and indicate major adverse cardiac events. A total of 160 patients with 166 lesions were investigated. FFRHARVEY was compared with FFRINVASIVE by investigators blinded to the invasive FFR results with a per-stenosis area under the curve of 0.91, positive predictive value of 90.2%, negative predictive value of 89.6%, sensitivity of 79.3%, and specificity of 95.4%. The percentage ofdiscrepancy for continuous values of FFR was 6.63%. We identified a hemodynamic phenomarker, longitudinal vorticity, as a metric indicative of major adverse cardiac events in unrevascularized gray-zone cases. CONCLUSIONS: FFRHARVEY had high performance (area under the curve: 0.91, positive predictive value: 90.2%, negative predictive value: 89.6%) compared with FFRINVASIVE. The proposed framework provides a robust and accurate way to compute a complete set of intravascular phenomarkers, in which longitudinal vorticity was specifically shown to differentiate vessels predisposed to major adverse cardiac events.

Preacinar Arterial Dilation Mediates Outcomes of Quantitative Interstitial Abnormalities in COPDGene.

RATIONALE: Quantitative interstitial abnormalities (QIA) are a computed tomography (CT) measure of early parenchymal lung disease associated with worse clinical outcomes including exercise capacity and symptoms. The presence of pulmonary vasculopathy in QIA and its role in the QIA-outcome relationship is unknown. OBJECTIVES: To quantify radiographic pulmonary vasculopathy in quantitative interstitial abnormalities (QIA) and determine if this vasculopathy mediates the QIA-outcome relationship. METHODS: Ever-smokers with QIA, outcome, and pulmonary vascular mediator data were identified from the COPDGene cohort. CT-based vascular mediators were: right ventricle-to-left ventricle ratio (RV/LV), pulmonary artery-to-aorta ratio (PA/Ao), and pre-acinar intraparenchymal arterial dilation (PA volume 5-20mm2 in cross-sectional area, normalized to total arterial volume). Outcomes were: six-minute walk distance (6MWD) and modified Medical Council Research Council (mMRC) Dyspnea score ≥2. Adjusted causal mediation analyses were used to determine if the pulmonary vasculature mediated the QIA effect on outcomes. Associations of pre-acinar arterial dilation with select plasma biomarkers of pulmonary vascular dysfunction were examined. MAIN RESULTS: Among 8,200 participants, QIA burden correlated positively with vascular damage measures including pre-acinar arterial dilation. Pre-acinar arterial dilation mediated 79.6% of the detrimental impact of QIA on 6MWD (56.2-100%, p<0.001). PA/Ao was a weak mediator and RV/LV was a suppressor. Similar results were observed in the QIA-mMRC relationship. Pre-acinar arterial dilation correlated with increased pulmonary vascular dysfunction biomarker levels including angiopoietin-2 and NT-proBNP. CONCLUSIONS: Parenchymal quantitative interstitial abnormalities (QIA) deleteriously impact outcomes primarily through pulmonary vasculopathy. Pre-acinar arterial dilation may be a novel marker of pulmonary vasculopathy in QIA.

Health-Related Quality of Life Across the Spectrum of Pulmonary Hypertension.

BACKGROUND: Health-related quality of life (HRQOL) is frequently impaired in pulmonary arterial hypertension. However, little is known about HRQOL in other forms of pulmonary hypertension (PH). RESEARCH QUESTION: Does HRQOL vary across groups of the World Symposium on Pulmonary Hypertension (WSPH) classification system? STUDY DESIGN AND METHODS: This cross-sectional study included patients with PH from the Pulmonary Vascular Disease Phenomics (PVDOMICS) cohort study. HRQOL was assessed by using emPHasis-10 (e-10), the 36-item Medical Outcomes Study Short Form survey (physical component score [PCS] and mental component score), and the Minnesota Living with Heart Failure Questionnaire. Pearson correlations between HRQOL and demographic, physiologic, and imaging characteristics within each WSPH group were tested. Multivariable linear regressions compared HRQOL across WSPH groups, adjusting for demographic characteristics, disease prevalence, functional class, and hemodynamics. Cox proportional hazards models were used to assess associations between HRQOL and survival across WSPH groups. RESULTS: Among 691 patients with PH, HRQOL correlated with functional class and 6-min walk distance but not hemodynamics. HRQOL was severely depressed across WSPH groups for all measures except the 36-item Medical Outcomes Study Short Form survey mental component score. Compared with Group 1 participants, Group 2 participants had significantly worse HRQOL (e-10 score, 29 vs 24 [P = .001]; PCS, 32.9 ± 8 vs 38.4 ± 10 [P < .0001]; and Minnesota Living with Heart Failure Questionnaire score, 50 vs 38 [P = .003]). Group 3 participants similarly had a worse e-10 score (31 vs 24; P < .0001) and PCS (33.3 ± 9 vs 38.4 ± 10; P < .0001) compared with Group 1 participants, which persisted in multivariable models (P < .05). HRQOL was associated in adjusted models with survival across Groups 1, 2, and 3. INTERPRETATION: HRQOL was depressed in PH and particularly in Groups 2 and 3 despite less severe hemodynamics. HRQOL is associated with functional capacity, but the severity of hemodynamic disease poorly estimates the impact of PH on patients' lives. Further studies are needed to better identify predictors and treatments to improve HRQOL across the spectrum of PH.

Implications of Mean Pulmonary Arterial Wedge Pressure Trajectories in Pulmonary Arterial Hypertension.

Rationale: The mean pulmonary arterial wedge pressure (mPAWP) is the critical hemodynamic factor differentiating group 1 pulmonary arterial hypertension (PAH) from group 2 pulmonary hypertension associated with left heart disease. Despite the discrepancy between the mPAWP upper physiologic normal and current PAH definitions, the implications of the initial mPAWP for PAH clinical trajectory are poorly understood. Objectives: To model longitudinal mPAWP trajectories in PAH over 10 years and examine the clinical and hemodynamic factors associated with trajectory membership. Methods: Adult patients with PAH with two or more right heart catheterizations were identified from a multiinstitution healthcare system in eastern Massachusetts. mPAWP trajectories were constructed via group-based trajectory modeling. Feature selection was performed in least absolute shrinkage and selection operator regression. Logistic regression was used to assess associations between trajectory membership, baseline characteristics, and transplant-free survival. Measurements and Main Results: Among 301 patients with PAH, there were two distinct mPAWP trajectories, termed "mPAWP-high" (n = 71; 23.6%) and "mPAWP-low" (n = 230; 76.4%), based on the ultimate mPAWP value. Initial mPAWP clustered around median 12 mm Hg (interquartile range [IQR], 8-14 mm Hg) in the mPAWP-high and 9 mm Hg (IQR, 6-11 mm Hg) in the mPAWP-low trajectories (P < 0.001). After feature selection, initial mPAWP ⩾12 mm Hg predicted an mPAWP-high trajectory (odds ratio, 3.2; 95% confidence interval, 1.4-6.1; P = 0.0006). An mPAWP-high trajectory was associated with shorter transplant-free survival (vs. mPAWP-low, median, 7.8 vs. 11.3 yr; log-rank P = 0.017; age-adjusted P = 0.217). Conclusions: Over 10 years, the mPAWP followed two distinct trajectories, with 25% evolving into group 2 pulmonary hypertension physiology. Using routine baseline data, longitudinal mPAWP trajectory could be predicted accurately, with initial mPAWP ⩾12 mm Hg as one of the strongest predictors.

Sleep-Related Hypoxia, Right Ventricular Dysfunction, and Survival in Patients With Group 1 Pulmonary Arterial Hypertension.

BACKGROUND: Group 1 pulmonary arterial hypertension (PAH) is a progressive fatal condition characterized by right ventricular (RV) failure with worse outcomes in connective tissue disease (CTD). Obstructive sleep apnea and sleep-related hypoxia may contribute to RV dysfunction, though the relationship remains unclear. OBJECTIVES: The aim of this study was to prospectively evaluate the association of the apnea-hypopnea index (AHI) and sleep-related hypoxia with RV function and survival. METHODS: Pulmonary Vascular Disease Phenomics (National Heart, Lung, and Blood Institute) cohort participants (patients with group 1 PAH, comparators, and healthy control participants) with sleep studies were included. Multimodal RV functional measures were examined in association with AHI and percentage of recording time with oxygen saturation <90% (T90) per 10-unit increment. Linear models, adjusted for demographics, oxygen, diffusing capacity of the lungs for carbon monoxide, pulmonary hypertension medications, assessed AHI and T90, and RV measures. Log-rank test/Cox proportional hazards models adjusted for demographics, oxygen, and positive airway pressure were constructed for transplantation-free survival analyses. RESULTS: Analysis included 186 participants with group 1 PAH with a mean age of 52.6 ± 14.1 years; 71.5% were women, 80.8% were Caucasian, and there were 43 events (transplantation or death). AHI and T90 were associated with decreased RV ejection fraction (on magnetic resonance imaging), by 2.18% (-2.18; 95% CI: -4.00 to -0.36; P = 0.019) and 0.93% (-0.93; 95% CI: -1.47 to -0.40; P < 0.001), respectively. T90 was associated with increased RV systolic pressure (on echocardiography), by 2.52 mm Hg (2.52; 95% CI: 1.61 to 3.43; P < 0.001); increased mean pulmonary artery pressure (on right heart catheterization), by 0.27 mm Hg (0.27; 95% CI: 0.05 to 0.49; P = 0.019); and RV hypertrophy (on electrocardiography), 1.24 mm (1.24; 95% CI: 1.10 to 1.40; P < 0.001). T90, but not AHI, was associated with a 17% increased 5-year risk for transplantation or death (HR: 1.17; 95% CI: 1.07 to 1.28). In non-CTD-associated PAH, T90 was associated with a 21% increased risk for transplantation or death (HR: 1.21; 95% CI: 1.08 to 1.34). In CTD-associated PAH, T90 was associated with RV dysfunction, but not death or transplantation. CONCLUSIONS: Sleep-related hypoxia was more strongly associated than AHI with measures of RV dysfunction, death, or transplantation overall and in group 1 non-CTD-associated PAH but only with RV dysfunction in CTD-associated PAH. (Pulmonary Vascular Disease Phenomics Program [PVDOMICS]; NCT02980887).

Kynurenine pathway metabolism evolves with development of preclinical and scleroderma-associated pulmonary arterial hypertension.

Understanding metabolic evolution underlying pulmonary arterial hypertension (PAH) development may clarify pathobiology and reveal disease-specific biomarkers. Patients with systemic sclerosis (SSc) are regularly surveilled for PAH, presenting an opportunity to examine metabolic change as disease develops in an at-risk cohort. We performed mass spectrometry-based metabolomics on longitudinal serum samples collected before and near SSc-PAH diagnosis, compared with time-matched SSc subjects without PAH, in a SSc surveillance cohort. We validated metabolic differences in a second cohort and determined metabolite-phenotype relationships. In parallel, we performed serial metabolomic and hemodynamic assessments as the disease developed in a preclinical model. For differentially expressed metabolites, we investigated corresponding gene expression in human and rodent PAH lungs. Kynurenine and its ratio to tryptophan (kyn/trp) increased over the surveillance period in patients with SSc who developed PAH. Higher kyn/trp measured two years before diagnostic right heart catheterization increased the odds of SSc-PAH diagnosis (OR 1.57, 95% CI 1.05-2.36, P = 0.028). The slope of kyn/trp rise during SSc surveillance predicted PAH development and mortality. In both clinical and experimental PAH, higher kynurenine pathway metabolites correlated with adverse pulmonary vascular and RV measurements. In human and rodent PAH lungs, expression of TDO2, which encodes tryptophan 2,3 dioxygenase (TDO), a protein that catalyzes tryptophan conversion to kynurenine, was significantly upregulated and tightly correlated with pulmonary hypertensive features. Upregulated kynurenine pathway metabolism occurs early in PAH, localizes to the lung, and may be modulated by TDO2. Kynurenine pathway metabolites may be candidate PAH biomarkers and TDO warrants exploration as a potential novel therapeutic target.NEW & NOTEWORTHY Our study shows an early increase in kynurenine pathway metabolism in at-risk subjects with systemic sclerosis who develop pulmonary arterial hypertension (PAH). We show that kynurenine pathway upregulation precedes clinical diagnosis and that this metabolic shift is associated with increased disease severity and shorter survival times. We also show that gene expression of TDO2, an enzyme that generates kynurenine from tryptophan, rises with PAH development.

Classification and Predictors of Right Ventricular Functional Recovery in Pulmonary Arterial Hypertension.

BACKGROUND: Normative changes in right ventricular (RV) structure and function have not been characterized in the context of treatment-associated functional recovery (RV functional recovery [RVFnRec]). The aim of this study is to assess the clinical relevance of a proposed RVFnRec definition. METHODS: We evaluated 63 incident patients with pulmonary arterial hypertension by right heart catheterization and cardiac magnetic resonance imaging at diagnosis and cardiac magnetic resonance imaging and invasive cardiopulmonary exercise testing following treatment (≈11 months). Sex, age, ethnicity matched healthy control subjects (n=62) with 1-time cardiac magnetic resonance imaging and noninvasive cardiopulmonary exercise testing were recruited from the PVDOMICS (Redefining Pulmonary Hypertension through Pulmonary Vascular Disease Phenomics) project. We examined therapeutic cardiac magnetic resonance imaging changes relative to the evidence-based peak oxygen consumption (VO2peak)>15 mL/(kg·min) to define RVFnRec by receiver operating curve analysis. Afterload was measured as mean pulmonary artery pressure, resistance, compliance, and elastance. RESULTS: A drop in RV end-diastolic volume of -15 mL best defined RVFnRec (area under the curve, 0.87; P=0.0001) and neared upper 95% CI RV end-diastolic volume of controls. This cutoff was met by 22 out of 63 (35%) patients which was reinforced by freedom from clinical worsening, RVFnRec 1 out of 21 (5%) versus no RVFnRec 17 out of 42, 40% (log-rank P=0.006). A therapy-associated increase of 0.8 mL/mm Hg in compliance had the best predictive value of RVFnRec (area under the curve, 0.76; [95% CI, 0.64-0.88]; P=0.001). RVFnRec patients had greater increases in stroke volume, and cardiac output at exercise. CONCLUSIONS: RVFnRec defined by RV end-diastolic volume therapeutic decrease of -15 mL predicts exercise capacity, freedom from clinical worsening, and nears normalization. A therapeutic improvement of compliance is superior to other measures of afterload in predicting RVFnRec. RVFnRec is also associated with increased RV output reserve at exercise.

Frequency of acute vasodilator response (AVR) in incident and prevalent patients with pulmonary arterial hypertension: Results from the pulmonary vascular disease phenomics study.

The prevalence of acute vasodilator response (AVR) to inhaled nitric oxide (iNO) during right heart catheterization (RHC) is 12% in idiopathic pulmonary arterial hypertension (IPAH). AVR, however, is reportedly lower in other disease-associated pulmonary arterial hypertension (PAH), such as connective tissue disease (CTD). The prevalence of AVR in patients on PAH therapy (prevalent cases) is unknown. We sought to determine AVR prevalence in Group 1 PH in the PVDOMICS cohort of incident and prevalent patients undergoing RHC. AVR was measured in response to 100% O2 and O2 plus iNO, with positivity defined as (1) decrease in mean pulmonary artery pressure (mPAP) by ≥10 mmHg to a value ≤40 mmHg, with no change or an increase in cardiac output (definition 1); or (2) decrease in mPAP by ≥12% and pulmonary vascular resistance by ≥30% (definition 2). AVR rates and cumulative survival were compared between incident and prevalent patients. In 338 mainly prevalent (86%) patients, positive AVR to O2-only was <2%, and 5.1% to 16.9%, based on definition 1 and 2 criteria, respectively; following O2 + iNO. IPAH AVR prevalence (4.1%-18.7%) was similar to prior reports. AVR positivity was 7.7% to 15.4% in mostly CTD-PAH prevalent cases, and 2.6% to 11.8% in other PAH groups. Survival was 89% in AVR responders versus 77% in nonresponders from PAH diagnosis, and 91% versus 86% from PVDOMICS enrollment (log-rank test p = 0.04 and p = 0.05, respectively). In conclusion, AVR in IPAH patients is similar to prior studies. AVR in non-IPAH patients was higher than previously reported. The relationship between PAH therapy, AVR response, and survival warrants further investigation.

Elevated Pulmonary Arterial Compliance Is Associated with Survival in Pulmonary Hypertension: Results from a Novel Network Medicine Analysis.

Rationale: Predictors of adverse outcome in pulmonary hypertension (PH) are well established; however, data that inform survival are lacking. Objectives: We aim to identify clinical markers and therapeutic targets that inform the survival in PH. Methods: We included data from patients with elevated mean pulmonary artery pressure (mPAP) diagnosed by right heart catheterization in the U.S. Veterans Affairs system (October 1, 2006-September 30, 2018). Network medicine framework was used to subgroup patients when considering an N of 79 variables per patient. The results informed outcome analyses in the discovery cohort and a sex-balanced validation right heart catheterization cohort from Vanderbilt University (September 24, 1998-December 20, 2013). Measurements and Main Results: From an N of 4,737 complete case patients with mPAP of 19-24 mm Hg, there were 21 distinct subgroups (network modules) (all-cause mortality range = 15.9-61.2% per module). Pulmonary arterial compliance (PAC) drove patient assignment to modules characterized by increased survival. When modeled continuously in patients with mPAP ⩾19 mm Hg (N = 37,744; age, 67.2 yr [range = 61.7-73.8 yr]; 96.7% male; median follow-up time, 1,236 d [range = 570-1,971 d]), the adjusted all-cause mortality hazard ratio was <1.0 beginning at PAC ⩾3.0 ml/mm Hg and decreased progressively to ∼7 ml/mm Hg. A protective association between PAC ⩾3.0 ml/mm Hg and mortality was also observed in the validation cohort (N = 1,514; age, 60.2 yr [range = 49.2-69.1 yr]; 48.0% male; median follow-up time, 2,485 d [range = 671-3,580 d]). The association was strongest in patients with precapillary PH at the time of catheterization, in whom 41% (95% confidence interval, 0.55-0.62; P < 0.001) and 49% (95% confidence interval, 0.38-0.69; P < 0.001) improvements in survival were observed for PAC ⩾3.0 versus <3.0 ml/mm Hg in the discovery and validation cohorts, respectively. Conclusions: These data identify elevated PAC as an important parameter associated with survival in PH. Prospective studies are warranted that consider PAC ⩾3.0 ml/mm Hg as a therapeutic target to achieve through proven interventions.