Incorporation of Biomarkers into Risk Assessment for Allocation of Antihypertensive Medication According to the 2017 ACC/AHA High Blood Pressure Guideline: A Pooled Cohort Analysis

Abstract

Background.

Risk for atherosclerotic cardiovascular disease (ASCVD) was a novel consideration for antihypertensive medication initiation in the 2017 ACC/AHA Blood Pressure (BP) guideline. Whether biomarkers of chronic myocardial injury [high-sensitivity cardiac troponin T (hs-cTnT) ≥ 6 ng/L] and stress [N-terminal pro-B-type natriuretic peptide (NT-proBNP) ≥ 100 pg/mL)] can inform cardiovascular (CV) risk stratification and treatment decisions among adults with elevated BP and hypertension is unclear.

Methods.

Participant-level data from 3 cohort studies (Atherosclerosis Risk in Communities Study, Dallas Heart Study, and Multiethnic Study of Atherosclerosis) were pooled, excluding individuals with prevalent CV disease and those taking antihypertensive medication at baseline. Participants were analyzed according to BP treatment group from the 2017 ACC/AHA BP guideline and those with high BP (BP:120–159/<100 mm Hg) were further stratified by biomarker status. Cumulative incidence rates for CV event (ASCVD or heart failure), and the corresponding 10-year number needed to treat to prevent one event (NNT¹⁰) with intensive BP lowering (to target systolic BP <120 mm Hg), were estimated for BP and biomarker-based subgroups.

Results.

The study included 12,987 participants (mean age:55 years, 55% women, 21.5% with elevated hs-cTnT, 17.7% with elevated NT-proBNP) with 825 incident CV events over 10-year follow-up. Participants with elevated BP or hypertension not recommended for antihypertensive medication with vs. without either elevated hs-cTnT or NT-proBNP had a 10-year CV incidence rate of 11.0% and 4.6%, with an NNT¹⁰ for intensive BP lowering of 36 and 85, respectively. Among participants with stage 1 or stage 2 hypertension recommended for antihypertensive medication with BP <160/100 mm Hg, those with vs. without an elevated biomarker had a 10-year CV incidence rate of 15.1% and 7.9%, with an NNT¹⁰ of 26 and 49, respectively.

Conclusions.

Elevations in hs-cTnT or NT-proBNP identify individuals with elevated BP or hypertension not currently recommended for antihypertensive medication who are at high risk for CV events. The presence of non-elevated biomarkers, even in the setting of stage 1 or stage 2 hypertension, was associated with lower risk. Incorporation of biomarkers into risk assessment algorithms may lead to more appropriate matching of intensive BP control with patient risk.

INTRODUCTION

Hypertension is common and associated with increased risk of atherosclerotic cardiovascular disease (ASCVD) and heart failure (HF).^1–3 Recent evidence suggests that most incident cardiovascular (CV) events occur in adults with blood pressure (BP) <140/90 mm Hg, a level at which there are conflicting recommendations for initiation of antihypertensive medication across clinical practice guidelines.^4–7 In the Systolic Blood Pressure Intervention Trial (SPRINT), intensive blood pressure (BP) control targeting a systolic BP <120 mm Hg reduced the risk of adverse CV events, particularly incident HF and CV mortality.⁸ Accordingly, the 2017 American College of Cardiology/American Heart Association (ACC/AHA) Guideline for the Prevention, Detection, Evaluation and Management of High Blood Pressure in Adults recommended a lower BP threshold for diagnosing hypertension and initiating antihypertensive medication.⁷ Implementation of the 2017 ACC/AHA BP guideline has led to a substantial increase in the prevalence of hypertension in the United States (US).⁹ A large proportion of individuals who are newly identified with elevated BP or hypertension are low-risk, younger adults with intermediate range BP (BP 120–139/<90 mm Hg) who are not currently recommended for antihypertensive medication based on the 2017 ACC/AHA BP guideline.⁹

The 2017 ACC/AHA BP guideline incorporates 10-year predicted ASCVD risk to determine which individuals with stage 1 hypertension (BP 130–139/80–89 mmHg) are likely to derive the most benefit from antihypertensive medication initiation. However, the ACC/AHA Pooled Cohort Equations (PCE), which are the recommended strategy for assessing ASCVD risk, only provide modest discrimination for ASCVD events, and have overestimated risk in contemporary cohorts.^10–13 Importantly, these equations do not assess risk for HF, an important limitation of a tool used to guide initiation of BP lowering medications, given the benefits of intensive BP control on incident HF.⁸ Thus, there is a need for improved risk stratification strategies to identify individuals with intermediate range BP levels that are at the highest risk for ASCVD and HF events and may benefit from intensive BP control.

Cardiac troponin T (cTnT) and N-terminal pro-B-type natriuretic peptide (NT-proBNP) are biomarkers of chronic myocardial injury and stress, are detectable in the general population and are associated with adverse CV events.^14–17 Mildly elevated levels of these biomarkers identify intermediate phenotypes along the progression from hypertension to clinical HF.^{18, 19} The utility of these biomarkers to identify individuals with elevated BP or hypertension who may benefit from intensive BP control is not well established. The purpose of this study was to evaluate whether cTnT measured by a high-sensitivity assay (hs-cTnT) and NT-proBNP can improve stratification of CV risk among adults with elevated BP or hypertension according to the 2017 ACC/AHA BP guideline. We hypothesized that elevated biomarker levels would identify individuals with elevated BP or hypertension not currently recommended for treatment who may benefit from antihypertensive medication.

METHODS

The data from the present study will not be made available by the authors for use by other investigators.

Study population

We pooled individual-level data from the following three prospective cohort studies: Atherosclerosis Risk in Communities (ARIC), Dallas Heart Study (DHS), and Multiethnic Study of Atherosclerosis (MESA). Study design, participant characteristics, and event adjudication have been reported previously for each of these cohort studies.^20–22 Briefly, ARIC is a prospective, community-based cohort that enrolled 15,792 participants between 45 and 64 years of age from 1987 to 1989 in four communities across the US.²⁰ For the present analysis, ARIC visit 2 (from 1990–1992) was considered the baseline visit as hs-cTnT and NT-proBNP were measured using samples obtained from this visit (n=14,348) ²³. DHS is a multiethnic, probability-based cohort study of adults from Dallas County with deliberate oversampling of African Americans.^{14, 21} Data were collected at 3 separate visits from 2000 to 2002. At the second in-home visit, fasting blood was collected from 3,557 adults, 30 to 65 years of age. MESA is a multiethnic, community-based cohort study that enrolled 6,814 participants, 45 to 84 years of age, with no known CV disease (CVD) from 2000 to 2002 across six US communities.^{22, 24} The initial phases of DHS and MESA each occurred from 2000 to 2002 and are considered the baseline visits.

The present study included participants from all three cohorts who, at baseline for this study, were free of known CVD (coronary heart disease[CHD], stroke, and HF), not taking antihypertensive medication and had available biomarker (hs-cTnT and NT-proBNP) data and parameters needed to calculate estimated glomerular filtration rate (eGFR) as well as 10-year ASCVD risk according to the ACC/AHA PCE.¹⁰ The protocol for each study was approved by the Institutional Review Boards of each participating center. All study participants provided written informed consent. The present analysis was approved by the Institutional Review Board at UT Southwestern and the coordinating centers for each of the cohorts.

Assessment of baseline covariates in the study cohorts

Study participants from all three cohorts underwent comprehensive examinations that included self-reported and measured clinical characteristics using standardized protocols described previously.^20–22 Age, sex, race, smoking status, history of hypertension, and medication use were self-reported. Body mass index (BMI), total cholesterol, and high-density lipoprotein cholesterol were measured. Diabetes mellitus and eGFR were defined as previously described.^20–22 Left ventricular (LV) hypertrophy (LVH) by 12-lead electrocardiogram was determined according to the Sokolow-Lyon criteria across all three cohorts. In the cohorts with available cardiac MRI data (MESA and DHS), LVH was determined using sex-specific cut-offs for indexed LV mass as defined previously in the DHS (women: LV mass/BSA≥89 g/m²; men: LV mass/BSA≥112 g/m²)¹⁸ and MESA (women: LV mass/BSA ≥84.6 g/m²; men: LV mass/BSA ≥106.2 g/m²).²⁵ Subclinical LV dysfunction was also determined in participants with available cardiac MRI data based on LV ejection fraction (EF) <50%. Antihypertensive medication use patterns during follow-up was assessed using data from subsequent visits after the baseline biomarker assessment across the study cohorts (visit 2 for DHS, visit 3 for MESA, and visit 4 for ARIC).

Blood pressure measurement at baseline

Study participants from all three cohorts had their BP measured in the seated position after five minutes of rest.^26–28 In ARIC, a technician obtained three BP measurements using a random-zero sphygmomanometer and BP was defined as the mean of the last two BP readings ²⁶. In DHS, a trained professional obtained five seated BP measurements with an automated oscillometric device (Series #52,000, Welch Allyn, Arden, North Carolina) and the mean of the last three BP recordings was used in these analyses.²⁷ In MESA, an automated oscillometric device (Dinamap Monitor Pro100, GE Healthcare, Milwaukee, WI) was used at two-minute intervals to obtain three BP readings and BP was defined as the mean of the last two BP measurements.²⁸

Assessment of biomarkers of interest (hs-cTnT & NT-proBNP)

Both hs-cTnT and NT-proBNP were measured using commercially available immunoassays, with Roche Elecsys 2010 analyzer used in ARIC and DHS, and the Roche Cobas e601 used in MESA (Roche Diagnostics, Indianapolis, IN).^{14, 15, 29} Based on clinical practice limits of quantification in the US, hs-cTnT ≥6 ng/L was defined as elevated. For NT-proBNP, a level ≥100 pg/mL was considered elevated based on prior studies.^{30, 31} Data-derived optimal cut points for these biomarker assays to predict incident CV events were also determined using the receiver operator characteristic analysis and the maximum value of Youden’s index and validated using bootstrapping analysis with 1,000 bootstraps.³²

Study groups based on 2017 ACC/AHA BP guideline recommendations for treatment & biomarker status

The 2017 ACC/AHA BP guideline identifies three groups based on BP level and indication for antihypertensive medication initiation: 1) normal BP (BP <120/80 mm Hg); 2) elevated BP (BP 120–129/<80 mm Hg) or low-risk stage 1 hypertension (BP 130–139/80–89 mm Hg) not recommended for antihypertensive medication; and 3) high-risk stage 1 hypertension (BP 130–139/80–89 mm Hg) or stage 2 hypertension (BP ≥140/90 mm Hg) recommended for antihypertensive medication. In accordance with the 2017 ACC/AHA BP guideline, high-risk stage 1 hypertension was defined by the presence of any of the following: PCE-estimated 10-year ASCVD risk ≥10%, diabetes mellitus, estimated glomerular filtration rate <60 mL/min per 1.73 m², or age ≥65 years with systolic BP ≥130 mm Hg.⁷ In the absence of all of these risk factors, individuals with stage 1 hypertension were classified as low-risk. In the current study, the PCE for non-Hispanic whites were used to estimate 10-year ASCVD risk for participants who identified as non-African American (white, Hispanic, and Asian).

Study participants were categorized into six groups based on a combination of the 2017 ACC/AHA BP guideline recommendation for antihypertensive medication initiation and categories of hs-cTnT or NT-proBNP: Group A: normal BP (<120/80 mm Hg); Group B: elevated BP (120–129/<80 mm Hg) or low-risk stage 1 hypertension (130–139/80–89 mm Hg) with non-elevated biomarkers; Group C: elevated BP (120–129/<80 mm Hg) or low-risk stage 1 hypertension (130–139/80–89 mm Hg) with elevated hs-cTnT and/or NT-proBNP; Group D: high-risk stage 1 hypertension (130–139/80–89 mm Hg) or stage 2 hypertension (140–159/90–99 mm Hg) with non-elevated biomarkers; Group E: high-risk stage 1 hypertension (130–139/80–89 mm Hg) or stage 2 hypertension (140–159/90–99 mm Hg) with elevated hs-cTnT and/or NT-proBNP; and Group F: stage 2 hypertension (≥160/100 mm Hg). As there is general consensus across various guideline recommendations regarding management of individuals with normal BP and those with BP ≥160/100 mm Hg, groups A and F were not stratified by biomarker status.^5–7

Primary and secondary outcomes of interest

The primary outcome of interest was an incident CV event defined as a composite of incident ASCVD (non-fatal myocardial infarction [MI], non-fatal stroke, or CV death) or HF. The secondary outcome of interest was incident HF identified by the first HF hospitalization event. In each of the three cohort studies, an expert committee adjudicated all non-fatal CV events as described previously and detailed in the Supplemental Methods I.^{18, 30, 33, 34}

Statistical analysis

Individual-level participant data were merged, and participants were categorized into Groups A through F based on the BP groups and biomarker status (elevated vs. non-elevated). Baseline characteristics were compared across the study groups using Chi-square test for categorical variables and Jonckheere-Terpstra test for continuous variables.

Unadjusted risks of the incident CV event (ASCVD or HF) and HF were assessed across the BP/ biomarker-based study groups using cumulative incidence plots and log-rank tests. The follow-up time was censored at 10 years in the pooled cohort to allow for comparable follow-up time across cohorts. The 10-year number needed to treat (NNT¹⁰) to prevent an incident CV event or HF was estimated assuming a 25% and 38% relative risk reduction (RRR) in these outcomes, respectively, based on the treatment effect of intensive BP lowering to target systolic BP <120 mm Hg in SPRINT.⁸ Given that one-third of the participants in SPRINT had systolic BP ≤132 mm Hg and intensive BP control led to similar reductions in CV event rates regardless of baseline BP level, we used the same RRR across all groups in the present analysis. Based on the 2017 ACC/AHA BP guideline systematic review, we also performed sensitivity analyses using RRRs of 19% and 25% for CV and HF, respectively, to calculate the NNT¹⁰ to prevent each outcome.³⁵

Multivariable adjusted analysis was also performed to evaluate the association between the BP/biomarker-based study groups and risk of incident CV event and incident HF independent of other risk factors. Separate Cox models were constructed for each outcome with adjustment for the following potential confounders: demographics (age, sex, race), CV risk factors (BMI, diabetes mellitus status, smoking status), laboratory values (total cholesterol, HDL cholesterol, eGFR), medications (statin use), and study cohort. The contribution of the biomarker-based approach to risk stratification toward the prediction of incident CV and HF event above and beyond the traditional CV risk factors and BP treatment categories was assessed by calculating the Harrell C statistic for adjusted Cox models with the 2017 ACC/AHA BP guideline recommended treatment groups and the BP/biomarker-based study groups. The C statistic derived from the model with 2017 ACC/AHA BP guideline recommended treatment groups was compared with that calculated for the BP/biomarker-based study groups and were compared using bootstrapping method.

Several sensitivity analyses were also performed to assess the robustness of the observed associations between BP/biomarker-based study groups and risk of adverse cardiovascular events. First, to account for heterogeneity in study cohorts, a cohort-stratified pooled analysis was performed by pooling cohort-specific 10-year KM estimates for incident CV events associated with different BP/biomarker-based groups using random effect modeling technique as described previously ³⁶. Second, sensitivity analyses were performed excluding individuals with LVH based on ECG at baseline. Furthermore, in the subset of participants with available cardiac MRI data, sensitivity analyses were also performed excluding individuals with LVH or subclinical LV dysfunction (LV EF <50%). Third, sensitivity analysis was also performed to evaluate the 10-year risk of incident CV event across different BP/biomarker-based groups among participants without any of the following high-risk features: prevalent diabetes mellitus, current smoker, eGFR <60 mL/min per 1.73 m², and statin use. Finally, race- (black vs. non-black) and sex- (male vs. female) based subgroup analyses were performed examining risk of CV events across the BP/biomarker-based study groups. Statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, North Carolina, USA). Two-sided p-values < 0.05 were considered statistically significant without adjustment for multiple comparisons.

Alternative biomarkers of chronic myocardial injury and stress

A similar analytic approach was taken in the Jackson Heart Study (JHS) using the same BP level categories but different biomarkers for chronic myocardial injury and stress. Specifically, cardiac troponin I (cTnI) was assessed using a highly-sensitive assay (hs-cTnI, Roche Diagnostics) and B-type natriuretic peptide (BNP) was measured at the baseline visit ¹⁹. JHS design and study participants as well as details regarding assessment of baseline covariates (including BP, hs-cTnI, and BNP) and event adjudication are described in detail in Supplemental Methods II.

RESULTS

The present study included 12,987 participants (mean age 55 years, 55% women). ARIC contributed 50.2% of participants to the final study cohort while DHS and MESA contributed 17.3% and 32.5%, respectively (Supplemental Figure 1). In the pooled cohort, 54.2% of participants had normal BP and 22.8% had elevated BP or low-risk stage 1 hypertension (antihypertensive medication not recommended). The prevalence of high-risk stage 1 hypertension or stage 2 hypertension with BP<160/100 mm Hg (antihypertensive medication recommended) was 19.9% while 3.2% of study participants had stage 2 hypertension with BP ≥160/100 mm Hg in the present study.

Among all study participants, elevated hs-cTnT and NT-proBNP was noted in 21.5% and 17.7%, respectively. The prevalence of elevated hs-cTnT or NT-proBNP was 32.3% among participants with elevated BP or low-risk stage 1 HTN (not recommended for antihypertensive medication) and 46.5% among those with high-risk stage 1 hypertension or stage 2 hypertension with BP <160/100 mm Hg (recommended for antihypertensive medication) (Figure 1).

Figure 1. Prevalence of elevated hs-cTnT or NT-proBNP across 2017 ACC/AHA BP guideline recommended treatment groups.

Elevated BP: 120–129/<80 mm Hg; stage 1 HTN: 130–139/80–89 mm Hg; stage 2 HTN: ≥140/90 mm Hg; high-risk stage 1 HTN was defined by the presence of any of the following: PCE-estimated 10-year ASCVD risk ≥10%, diabetes mellitus, estimated GFR <60 mL/min per 1.73 m², or age ≥65 years with systolic BP ≥130 mmHg; in the absence of all of these risk factors, individuals with stage 1 HTN were classified as low-risk; hs-cTnT (+): hs-cTnT ≥6 ng/L; NT-proBNP (+): NT-proBNP ≥100 pg/mL; any biomarker (+): hs-cTnT ≥6 ng/L and/or NT-proBNP ≥100 pg/mL.

Abbreviations: anti-HTN= antihypertensive; ASCVD = atherosclerotic cardiovascular disease; BP = blood pressure; GFR = glomerular filtration rate; hs-cTnT = high-sensitivity cardiac troponin T; HTN = hypertension; NT-proBNP = N-terminal pro-B-type natriuretic peptide; PCE = pooled cohort equation.

Baseline characteristics of the study population categorized by the 2017 ACC/AHA BP guideline recommended treatment groups and according to biomarker status are shown in Table 1. As expected, the burden of traditional CV risk factors was higher in participants with higher BP and those with indications for antihypertensive medication initiation. Participants with vs. without elevated biomarkers were older, more commonly men, and had higher burden of diabetes mellitus, smoking, and LVH. Among participants with any elevated biomarkers across both BP treatment groups, the median biomarker levels were only slightly above the pre-defined cut-offs and well below the clinical thresholds used for diagnosis of myocardial infarction (for hs-cTnT) or heart failure (for NT-proBNP) (Table 1).

Table 1.

Baseline characteristics across 2017 ACC/AHA BP guideline recommended treatment groups stratified by biomarker status

	Not recommended to receive anti-HTN medication			Recommended to receive anti-HTN medication			P value
	Normal BP (n = 7,038)	Elevated BP or low-risk stage 1 HTN, both biomarker (−) (n = 2,005)	Elevated BP or low-risk stage 1 HTN, any biomarker (+) (n = 955)	High-risk stage 1 HTN or stage 2 HTN (BP <160/100 mm Hg), both biomarker (−) (n = 1,380)	High-risk stage 1 HTN or stage 2 HTN (BP <160/100 mm Hg), any biomarker (+) (n = 1,199)	Stage 2 HTN (BP ≥ 160/100 mm Hg) (n = 410)
Age, years	53 (49 – 59)	52 (47 – 59)	58 (52 – 64)	57 (51 – 63)	63 (57 – 69)	61 (54 – 68)	<0.001
Men, %	2,885 (41.0)	936 (46.7)	498 (52.2)	709 (51.4)	640 (53.4)	177 (43.2)	<0.001
Black, %	1,288 (18.3)	690 (34.4)	230 (24.1)	367 (26.6)	275 (22.9)	168 (41.0)	<0.001
BMI, kg/m2	25.9 (23.3 – 29.1)	28.0 (25.1 – 31.9)	27.6 (24.5 – 31.5)	28.0 (25.1 – 31.4)	27.4 (24.3 – 31.2)	28.9 (25.3 – 33)	<0.001
Systolic BP, mm Hg	108 (102 – 114)	125 (122 – 129)	125 (122 – 128)	138 (132 – 145)	140 (133 – 147)	168 (162 – 176)	<0.001
Diastolic BP, mm Hg	67 (62 – 72)	77 (72 – 81)	74 (69 – 78)	82 (76 – 87)	80 (73 – 85)	89 (80 – 98)	<0.001
Total cholesterol, mg/dL	195 (173 – 222)	198 (174 – 220)	194 (174 – 223)	204 (179 – 229)	203 (180 – 228)	196 (176 – 228)	<0.001
HDL cholesterol, mg/dL	49 (40 – 60)	47 (40 – 57)	47 (39 – 61)	46 (38 – 57)	47 (40 – 60)	49 (40 – 58)	<0.001
Estimated GFR, mL/min per 1.73 m2	68.9 (58.0 – 85.8)	79.0 (64.2 – 95.4)	68.8 (58.9 – 83.3)	66.9 (57.0 – 85.9)	67.0 (56.7 – 81.5)	73.9 (61.4 – 90.4)	0.25
Diabetes, %	309 (4.4)	95 (4.7)	66 (6.9)	151 (10.9)	160 (13.3)	52 (12.7)	<0.001
Current smoker, %	1,450 (20.6)	414 (20.7)	160 (16.8)	284 (20.6)	185 (15.4)	88 (21.5)	<0.001
Biomarkers
hs-cTnT (+), %	1,170 (16.6)	0 (0)	619 (64.8)	0 (0)	840 (70.1)	167 (40.7)	<0.001
NT-proBNP (+), %	1,061 (15.1)	0 (0)	454 (47.5)	0 (0)	619 (51.6)	162 (39.5)	<0.001
both (+), %	217 (3.1)	0 (0)	118 (12.4)	0 (0)	260 (21.7)	82 (20.0)	<0.001
hs-cTnT, ng/L	3.0 (1.5 – 4.4)	1.5 (1.5 – 3.6)	6.8 (3.0 – 9.0)	1.5 (1.5 – 4.0)	7.0 (4.7 – 10.0)	5.0 (1.5 – 8.1)	<0.001
NT-proBNP, pg/mL	40.9 (21.4 – 74.9)	31.9 (15.4 – 53.9)	90.9 (32.7 – 136.7)	38.2 (20.8 – 62.0)	102.0 (41.8 – 156.3)	76.9 (41.1 – 152.3)	<0.001
Left ventricular hypertrophy by ECG	256 (3.7)	153 (7.9)	87 (9.3)	120 (8.9)	134 (11.4)	96 (24.0)	<0.001
Study group
ARIC	3,780 (53.7)	800 (39.9)	488 (51.1)	764 (55.4)	558 (46.5)	131 (32.0)	<0.001
DHS	1,124 (16.0)	619 (30.9)	126 (13.2)	210 (15.2)	74 (6.2)	88 (21.5)	<0.001
MESA	2,134 (30.3)	586 (29.2)	341 (35.7)	406 (29.4)	567 (47.3)	191 (46.6)	<0.001

Data presented as median (interquartile range) or n (percentile). Comparison across groups performed using Chi-square test for categorical variables and Jonckheere-Terpstra test for continuous variables. Normal BP: <120/80 mm Hg; elevated BP: 120–129/<80 mm Hg; stage 1 HTN: 130–139/80–89 mm Hg; stage 2 HTN: ≥140/90 mm Hg; high-risk stage 1 HTN was defined by the presence of any of the following: PCE-estimated 10-year ASCVD risk ≥10%, diabetes mellitus, estimated GFR <60 mL/min per 1.73 m², or age ≥65 years with systolic BP ≥130 mmHg; in the absence of all of these risk factors, individuals with stage 1 HTN were classified as low-risk; both biomarker (−): hs-cTnT <6 ng/L and NT-proBNP <100 pg/mL; any biomarker (+): hs-cTnT ≥6 ng/L and/or NT-proBNP ≥100 pg/mL; hs-cTnT (+): hs-cTnT ≥6 ng/L; NT-proBNP (+): NT-proBNP ≥100 pg/mL; both biomarkers (+): hs-cTnT ≥6 ng/L and NT-proBNP ≥100 pg/mL. Number of participants with available ECG data used to estimate the proportion with left ventricular hypertrophy by ECG is slightly lower than the total number of participants in each group.

Abbreviations: ARIC = Atherosclerosis Risk in Communities Study; ASCVD = atherosclerotic cardiovascular disease; BMI = body mass index; BP = blood pressure; DHS = Dallas Heart Study; ECG: Electrocardiogram; GFR = glomerular filtration rate; HDL = high-density lipoprotein; hs-cTnT = high-sensitivity cardiac troponin T; HTN: hypertension; MESA = Multi-Ethnic Study of Atherosclerosis; NT-proBNP = N-terminal pro-B-type natriuretic peptide

Antihypertensive medication use pattern at the time of the follow-up visit (~5 years from baseline visit) across different BP/biomarker-based groups is shown in Supplemental Figure 2. Antihypertensive medication initiation during follow-up increased across higher baseline BP groups. However, antihypertensive medication use at the follow-up visit was comparable among individuals with vs. without biomarker elevation at baseline across both BP treatment groups.

Incident CV events among the 2017 ACC/AHA BP guideline recommended treatment groups

Overall, there were 825 incident CV events and 261 HF events over a maximum 10-year follow-up. The cumulative incidence of CV events and HF was lowest among participants with normal BP. Among those with elevated BP or hypertension, the incidence of CV events and HF were higher in the subgroup with vs. without guideline recommendations for antihypertensive medication initiation, as expected (Supplemental Figure 3).

Incident CV events across groups stratified by BP level and biomarker status

Among participants with elevated BP or hypertension not recommended for antihypertensive medication, the incidence of CV events in the subgroup with non-elevated biomarkers was low and comparable to the incidence observed in the group with normal BP (4.6% vs. 3.8%, respectively; p-value = 0.11) (Figure 2A, Table 2). In contrast, those with elevated (vs. non-elevated) hs-cTnT or NT-proBNP had more than a 2-fold higher incidence of CV events (11.0% vs. 4.6%, respectively). Among participants recommended for antihypertensive medication based on the 2017 ACC/AHA BP guideline with BP <160/100 mm Hg, the risk of a CV event was higher in those with (vs. without) an elevated biomarker (15.1% vs. 7.9%, respectively; p-value <0.01) and approached the risk for participants with BP ≥160/100 mm Hg (17.3%). A similar pattern of association was noted for incident HF, with consistently higher risk among participants with vs. without elevated biomarkers across both treatment recommendation groups (Figure 2B, Table 2). Associations with risk for incident CV events and HF were similar for hs-cTnT and NT-proBNP when each biomarker was considered individually (Supplemental Tables 1 & 2).

Figure 2. Cumulative incidence of composite CV event (ASCVD or HF, Panel A) and HF (Panel B) across 2017 ACC/AHA BP guideline recommended blood pressure treatment groups stratified by presence of elevated hs-cTnT or NT-proBNP.

Normal BP: <120/80 mm Hg; Elevated BP: 120–129/<80 mm Hg; stage 1 HTN: 130–139/80–89 mm Hg; stage 2 HTN: ≥140/90 mm Hg; high-risk stage 1 HTN was defined by the presence of any of the following: PCE-estimated 10-year ASCVD risk ≥10%, diabetes mellitus, estimated GFR <60 mL/min per 1.73 m², or age ≥65 years with systolic BP ≥130 mmHg; in the absence of all of these risk factors, individuals with stage 1 HTN were classified as low-risk; both biomarker (−): hs-cTnT <6 ng/L and NT-proBNP <100 pg/mL; any biomarker (+): hs-cTnT ≥6 ng/L and/or NT-proBNP ≥100 pg/mL; Composite CV event = non-fatal MI, non-fatal stroke, HF, or CV death.

Abbreviations: anti-HTN= antihypertensive; ASCVD = atherosclerotic cardiovascular disease; BP = blood pressure; CV = cardiovascular; GFR = glomerular filtration rate; HF = heart failure; hs-cTnT = high-sensitivity cardiac troponin T; HTN = hypertension; MI = myocardial infarction; NT-proBNP = N-terminal pro-B-type natriuretic peptide; PCE = pooled cohort equation.

Table 2.

Proportion of participants with incident composite CV event (ASCVD or HF) and HF across 2017 ACC/AHA BP guideline recommended treatment groups stratified by biomarker status

	Participant Groups	Number of participants	Incident composite CV events (%)	Incident HF (%)
Not recommended to receive anti-HTN medication	Normal BP	7,038	267 (3.8)	85 (1.2)
	Elevated BP or low-risk stage 1 HTN, both biomarker (−)	2,005	92 (4.6)	17 (0.9)
	Elevated BP or low-risk stage 1 HTN, any biomarker (+)	955	105 (11.0)	41 (4.3)
Recommended to receive anti-HTN medication	High-risk stage 1 HTN or stage 2 HTN (BP <160/100 mm Hg), both biomarker (−)	1,380	109 (7.9)	28 (2.0)
	High-risk stage 1 HTN or stage 2 HTN (BP <160/100 mm Hg), any biomarker (+)	1,199	181 (15.1)	67 (5.6)
	Stage 2 HTN (BP ≥160/100 mm Hg)	410	71 (17.3)	23 (5.7)

Normal BP: <120/80 mm Hg; elevated BP: 120–129/<80 mm Hg; stage 1 HTN: 130–139/80–89 mm Hg; stage 2 HTN: ≥140/90 mm Hg; high-risk stage 1 HTN was defined by the presence of any of the following: PCE-estimated 10-year ASCVD risk ≥10%, diabetes mellitus, estimated GFR <60 mL/min per 1.73 m², or age ≥65 years with systolic BP ≥130 mmHg; in the absence of all of these risk factors, individuals with stage 1 HTN were classified as low-risk; both biomarker (−): hs-cTnT <6 ng/L and NT-proBNP <100 pg/mL; any biomarker (+): hs-cTnT ≥6 ng/L and/or NT-proBNP ≥100 pg/mL. Number at risk used to estimate the proportion of participants with incident HF is similar to or slightly lower than the total number of participants in each group.

Composite CV event = non-fatal MI, non-fatal stroke, HF, or CV death.

Abbreviations: anti-HTN = anti-hypertensive; ASCVD = atherosclerotic cardiovascular disease; BP = blood pressure; CV = cardiovascular; GFR: glomerular filtration rate; HF = heart failure; hs-cTnT = high-sensitivity cardiac troponin T; HTN = hypertension; MI = myocardial infarction; NT-proBNP = N-terminal pro-B-type natriuretic peptide; PCE = pooled cohort equation

In sensitivity analysis limited to the subset of individuals without LVH by ECG criteria as well as those without LVH or LV dysfunction by cardiac MRI, the patterns of association between BP/biomarker-based groups and risk of incident CV events and HF were consistent with those observed in the overall cohort (Supplemental Tables 3 & 4). In sensitivity analysis excluding individuals with prevalent risk factors including diabetes mellitus, current smoking, eGFR < 60 mL/min per 1.73 m², and statin use, the pattern of association between BP/biomarker-based study groups and risk of CV events was similar to that observed in the overall cohort (Supplemental Table 5). In sex- and race-based subgroup analyses, elevated biomarkers were associated with higher risk of CV events and HF across both BP treatment groups for both men and women (Supplemental Table 6) as well as both blacks and non-blacks (Supplemental Table 7).

Number needed to treat to prevent a CV event

Among participants with elevated BP or hypertension not recommended for antihypertensive medication, the NNT¹⁰ for prevention of one CV event, using the SPRINT treatment effect estimates, in those with vs. without an elevated biomarker was 36 and 85, respectively (Figure 3A). Among participants recommended for antihypertensive medication initiation with BP <160/100 mm Hg, the NNT¹⁰ for prevention of a CV event in those with vs. without an elevated biomarker was 26 and 49, respectively. The NNT¹⁰ to prevent an incident CV event for individuals with BP ≥160/100 mm Hg was 22. The NNT¹⁰ to prevent an incident HF event was also lower in individuals with vs. without an elevated biomarker for both BP treatment groups (not recommended for antihypertensive medication: 58 vs. 300; recommended for antihypertensive medication: 44 vs. 123, respectively) (Figure 3B). A similar pattern of results was observed using the RRR from the 2017 ACC/AHA BP guideline systematic review for intensive BP control, with a lower NNT¹⁰ to prevent an incident CV event or HF among those with an elevated vs. non-elevated biomarker across both BP treatment groups (Figure 4).

Figure 3. 10-year number needed to treat for prevention of an incident composite CV event (ASCVD or HF, left panel) and HF (right panel) across 2017 ACC/AHA BP guideline recommended treatment groups stratified by biomarker status using the treatment effect of intensive BP control in SPRINT.

The red dotted line represents the 10-year NNT to prevent a composite CV event (left panel) and HF (right panel) for the group with stage 2 HTN (BP ≥160/100 mm Hg) (CV event = 22; HF = 43).

The 10-year number needed to treat (NNT¹⁰) to prevent an incident composite CV event and HF was estimated assuming a 25% and 38% relative risk reduction (RRR) in these outcomes, respectively, based on the treatment effect of intensive BP lowering to target systolic BP <120 mm Hg in SPRINT.

Elevated BP: 120–129/<80 mm Hg; stage 1 HTN: 130–139/80–89 mm Hg; stage 2 HTN: ≥140/90 mm Hg; high-risk stage 1 HTN was defined by the presence of any of the following: PCE-estimated 10-year ASCVD risk ≥10%, diabetes mellitus, estimated GFR <60 mL/min per 1.73 m², or age ≥65 years with systolic BP ≥130 mmHg; in the absence of all of these risk factors, individuals with stage 1 HTN were classified as low-risk; both biomarker (−): hs-cTnT <6 ng/L and NT-proBNP <100 pg/mL; hs-cTnT (+): hs-cTnT ≥6 ng/L; NT-proBNP (+): NT-proBNP ≥100 pg/mL; any biomarker (+): hs-cTnT ≥6 ng/L and/or NT-proBNP ≥100 pg/mL; composite CV event = non-fatal MI, non-fatal stroke, HF, or CV death.

Abbreviations: anti-HTN= antihypertensive; ASCVD = atherosclerotic cardiovascular disease; BP = blood pressure; CV = cardiovascular; GFR = glomerular filtration rate; HF = heart failure; hs-cTnT = high-sensitivity cardiac troponin T; HTN = hypertension; MI = myocardial infarction; NNT: number needed to treat; NT-proBNP = N-terminal pro-B-type natriuretic peptide; PCE = pooled cohort equation; SPRINT = Systolic Blood Pressure Intervention Trial.

Figure 4: 10-year number needed to treat for prevention of an incident composite CV event (ASCVD or HF, left panel) and HF (right panel) across 2017 ACC/AHA BP guideline recommended treatment groups stratified by biomarker status using the treatment effect of intensive BP control in the 2017 ACC/AHA BP guideline systematic review.

The red dotted line represents the 10-year NNT to prevent an incident composite CV event (left panel) and HF (right panel) for the group with stage 2 HTN (BP ≥160/100 mm Hg) (CV event = 29; HF = 65).

Based on the 2017 ACC/AHA BP guideline systematic review, a 19% and 25% relative risk reduction was used to calculate the NNT¹⁰ for prevention of an incident composite CV event and HF, respectively.

Elevated BP: 120–129/<80 mm Hg; stage 1 HTN: 130–139/80–89 mm Hg; stage 2 HTN: ≥140/90 mm Hg; high-risk stage 1 HTN was defined by the presence of any of the following: PCE-estimated 10-year ASCVD risk ≥10%, diabetes mellitus, estimated GFR <60 mL/min per 1.73 m², or age ≥65 years with systolic BP ≥130 mmHg; in the absence of all of these risk factors, individuals with stage 1 HTN were classified as low-risk; both biomarker (−): hs-cTnT <6 ng/L and NT-proBNP <100 pg/mL; hs-cTnT (+): hs-cTnT ≥6 ng/L; NT-proBNP (+): NT-proBNP ≥100 pg/mL; any biomarker (+): hs-cTnT ≥6 ng/L and/or NT-proBNP ≥100 pg/mL; composite CV event = non-fatal MI, non-fatal stroke, HF, or CV death.

Abbreviations: anti-HTN= antihypertensive; ASCVD = atherosclerotic cardiovascular disease; BP = blood pressure; CV = cardiovascular; GFR = glomerular filtration rate; HF = heart failure; hs-cTnT = high-sensitivity cardiac troponin T; HTN = hypertension; MI = myocardial infarction; NNT = number needed to treat; NT-proBNP = N-terminal pro-B-type natriuretic peptide; PCE = pooled cohort equation.

In sensitivity analysis pooling cohort-specific 10-year KM estimates, the NNT¹⁰ to prevent incident CV events across different BP/biomarker-based study groups were similar to that observed in the primary analysis (Supplemental Figure 4).

Optimal data-derived biomarker assay cut points for risk of CV events

Data-derived optimal hs-cTnT and NT-proBNP cut points for predicting incident CV events and HF as derived by Youden’s statistic and validated using 1000 bootstraps are shown in Supplemental Table 8. For predicting an incident CV event, the optimal cut points for hs-cTnT and NT-proBNP were 8.5 ng/L and 95.9 pg/mL respectively. The optimal cut points for predicting an incident HF event were 7.7 ng/L for hs-cTnT and 102.3 pg/mL for NT-proBNP.

Adjusted association between BP/biomarker groups and risk of CV event

In adjusted analysis, compared with the normotensive group, the risk of incident CV event increased in a graded fashion across increasing BP groups, with a higher risk noted among individuals with (vs. without) elevated biomarkers across both BP treatment groups. The gradient in adjusted risk among individuals with (vs. without) biomarker elevation was even more pronounced for incident HF with a significantly higher risk noted only among those with biomarker elevation across both BP treatment groups (Figure 5). Compared with the 2017 ACC/AHA BP guideline recommended treatment groups, the BP/biomarker-based groups demonstrated a modest but statistically significant improvement in discrimination for incident CV events (C-index=0.732 vs. 0.745; p-value <0.001) and a more substantial improvement in discrimination for incident HF event (C-index=0.727 vs. 0.763; p-value <0.001).

Figure 5: Multivariable adjusted association between the 2017 ACC/AHA BP guideline recommended treatment groups stratified by biomarker status and incident composite CV event (ASCVD or HF) and HF.

Multivariable adjusted Cox models were constructed to evaluate the association between the BP/biomarker-based study groups and risk of outcome [composite CV event (non-fatal MI, non-fatal stroke, HF, or CV death) or HF (referent group = normal BP)] with adjustment for the following potential confounders: demographics (age, sex, race), CV risk factors (BMI, diabetes mellitus status, smoking status), laboratory values (total cholesterol, HDL cholesterol, estimated glomerular filtration rate), medications (statin use), and study cohort. Elevated BP: 120–129/<80 mm Hg; stage 1 HTN: 130–139/80–89 mm Hg; stage 2 HTN: ≥140/90 mm Hg; high-risk stage 1 HTN was defined by the presence of any of the following: PCE-estimated 10-year ASCVD risk ≥10%, diabetes mellitus, estimated GFR <60 mL/min per 1.73 m², or age ≥65 years with systolic BP ≥130 mmHg; in the absence of all of these risk factors, individuals with stage 1 HTN were classified as low-risk; any biomarker (+): hs-cTnT ≥6 ng/L and/or NT-proBNP ≥100 pg/mL.

Abbreviations: ASCVD = atherosclerotic cardiovascular disease; BMI: body mass index; BP = blood pressure; CV = cardiovascular; GFR: glomerular filtration rate; HDL: high density lipoprotein; HF = heart failure; hs-cTnT = high-sensitivity cardiac troponin T; HTN = hypertension; MI = myocardial infarction; NT-proBNP = N-terminal pro-B-type natriuretic peptide; PCE: pooled cohort equation.

Alternative measures of chronic myocardial injury and stress

To further assess a biomarker-based approach to CV risk prediction in the context of the 2017 ACC/AHA BP guideline, we evaluated 1,626 JHS participants without baseline CVD who were not taking antihypertensive medication and had alternative markers of chronic myocardial injury and stress available (cTnI and BNP, respectively) (Supplemental Figure 5, Supplemental Table 9). There were 117 incident adverse CV events (non-fatal MI, non-fatal stroke, HF or all-cause death) over 10-year follow-up. Similar to the primary study population, among participants with high BP levels (BP 120–159/<100 mmHg), 10-year CV event rates were higher among individuals with elevated vs. non-elevated biomarkers, whether they were not recommended (4.6 vs. 7.5%) or recommended (10.9 vs. 20.6%) for antihypertensive medication based on the 2017 ACC/AHA BP guideline (Supplemental Table 10).

DISCUSSION

In this large, pooled analysis of 3 cohort studies, there was substantial heterogeneity in the burden of chronic myocardial injury or stress, as defined by elevated hs-cTnT or NT-proBNP, respectively, among participants with elevated BP or hypertension. For example, approximately one-third of individuals with elevated BP or stage 1 hypertension currently not recommended for antihypertensive medication had elevated levels of at least one biomarker, while more than half of individuals currently recommended for antihypertensive medication with BP <160/100 mm Hg had no elevated biomarkers. Elevated levels of these biomarkers identified individuals at notably higher risk for adverse CV events and who thus would be expected to benefit more from initiation of BP lowering medications or targeting of more intensive BP goals. This was observed both among participants with and those without current guideline recommended indications for initiating antihypertensive medication. The absence of elevated biomarker levels identified lower 10-year CV risk among participants with elevated BP or hypertension. Among participants with elevated BP or low-risk stage 1 hypertension, non-elevated biomarkers were seen in approximately two-thirds, and in this group CV risk was comparable with the group with normal BP. Taken together, the study findings highlight the potential usefulness of a biomarker-based strategy to assist with BP treatment decisions among patients with elevated BP or hypertension.

Prior studies have demonstrated that chronic myocardial injury is associated with a higher risk of CV events across all BP levels.³³ The current findings extend prior observations regarding the prognostic utility of biomarkers in CV risk stratification, by considering the important context of the 2017 ACC/AHA BP guideline.^{33, 37} Specifically, a biomarker-based approach can improve risk stratification for CV events among individuals with elevated BP and hypertension with intermediate range BP levels. Based on the treatment effects derived from the SPRINT trial, the estimated NNT for prevention of a CV event among individuals with BP 120–159/<99 mm Hg and an elevated biomarker was comparable to that observed among individuals with BP ≥160/100 mm Hg. In contrast, the absence of elevated biomarkers identifies a lower-risk subset with a high NNT to prevent a CV event.

The results from the current study have important clinical implications. The current BP guideline has identified an additional 31.1 million individuals with hypertension in the US, of whom 4.2 million are newly recommended for antihypertensive medication.⁹ Our study findings suggest that supplementing the current guideline recommendations for antihypertensive medication initiation with a biomarker-based approach may help direct BP lowering medication to the highest risk individuals and thus maximize absolute CVD risk reduction (and lower NNT). An approach targeting medication initiation to those with the lowest NNT would have favorable risk- and cost-benefit implications. Along these lines, we observed that approximately one-third of the study participants with elevated BP or low-risk stage 1 HTN that are not currently recommended for antihypertensive medication had an elevated biomarker and higher CV risk. These individuals would be expected to benefit from initiation of antihypertensive pharmacotherapies based on the CV risk-based BP treatment approach recommended in the 2017 ACC/AHA BP guideline. Hs-cTnT and NT-proBNP testing are available in most countries worldwide at a relatively low cost. Considering the well-established cost-effectiveness of intensive BP control in individuals with high CV risk, similar to that observed in individuals with elevated biomarkers, it is anticipated that a biomarker driven approach to guide antihypertensive therapy initiation would be cost-effective in this group without current indications for therapy.³⁸ We also found that among individuals with stage 1 or 2 hypertension with BP <160/100 mm Hg currently recommended for antihypertensive pharmacotherapies, the absence of elevated biomarkers identified a lower risk subgroup of participants. However, considering the well-established evidence in favor of BP treatment, particularly for individuals with systolic BP ≥140 mm Hg, ^{5–7, 39} an approach that continues to favor initiating antihypertensive therapies in the absence of any contraindications may be considered, with selective use of biomarker testing to better inform the shared-decision making process among individuals with any clinical or patient-reported concerns related to treatment.

Similar to the biomarker-based approach used in our study, recent studies have demonstrated that coronary artery calcium (CAC), a marker of subclinical atherosclerosis, may also be used to identify high-risk individuals with hypertension that may benefit from intensive BP control.^{40, 41} While the role of CAC in ASCVD risk prediction is well-established, there are limited studies that have examined the independent association between CAC and HF risk, and CAC testing is more difficult to apply broadly than blood tests.⁴² In contrast, our study findings suggest that simple, relatively inexpensive blood tests to identify chronic myocardial injury or stress may provide prognostic information regarding both ASCVD and HF risk. Future clinical trials are needed to confirm the clinical utility and cost-effectiveness of biomarker-based assessment strategies to guide BP management among individuals with hypertension and intermediate range BP. It is noteworthy that intensive BP control is but one possible intervention that may modify risk among individuals identified at higher risk based on elevated biomarker levels. Other strategies including, but not limited to, intensive lifestyle interventions, intensive lipid lowering, sodium glucose cotransporter-2 inhibitors for type 2 DM may also be considered to further modify the risk of adverse cardiovascular events in these high-risk individuals. The abundance of effective therapies for CV risk reduction, targeting distinct modifiable biological pathways, argues for a more personalized approach to primary prevention that focuses on appropriate matching of the available treatment strategies to the underlying burden and pathophysiological drivers of the CV risk.⁴³

The primary strengths of our study include the large sample size, objective measurement of hs-cTnT and NT-proBNP in the primary analysis of the pooled cohorts, long follow-up duration, and adjudicated CV events. Additionally, our study cohort included nearly equal proportions of men and women with a large proportion of black participants. Furthermore, we validated the approach, using alternative biomarkers of chronic myocardial injury and stress (hs-cTnI and BNP) in a separate cohort, highlighting the robustness of the biomarker-based risk-stratification approach.

Several limitations to our study are noteworthy. First, participants included in this analysis were not taking antihypertensive medication prior to the baseline visit but were initiated on pharmacotherapies for BP treatment on follow-up. Antihypertensive medication initiation would be expected to lower incident CV event rates and may result in underestimation of CV risk in the present analysis. In the present study, the anti-hypertensive initiation during follow-up increased with higher baseline BP but was comparable among those with vs. without elevated biomarkers at baseline across both BP treatment groups. This suggests that the marked differences in the risks of adverse CV events among those with vs. without elevated biomarkers are unlikely to be related to the use of antihypertensive medications on follow-up. Second, BP assessment techniques varied slightly across different cohorts. Each study measured BP with a unique device, but BP was measured in the seated position after a rest period and the mean of multiple measurements was used in analyses. Third, we do not have data on risk of renal outcomes and safety endpoints (syncope, falls) that may also be influenced by BP levels, biomarkers, and intensive BP control. Fourth, we did not include albuminuria in the definition of chronic kidney disease in the present study due to non-uniform urine albumin assessment across cohorts. Fourth, the study cohort was relatively young, and these findings may not be generalizable to other populations that have distinct clinical characteristics from our cohort. Finally, owing to the observational nature of the study, there is a potential for residual confounding due to measured or unmeasured risk factors like subclinical coronary artery disease, silent ischemia or cardiomyopathy at baseline. However, the higher risk of adverse cardiovascular events among those with vs. without biomarker elevation across both BP treatment groups persisted in adjusted analysis accounting for other potential confounders and sensitivity analyses excluding participants with LVH and subclinical LV dysfunction. Furthermore, consistent patterns of associations were also observed in sensitivity analysis excluding higher risk individuals with comorbidities highlighting the robustness of our study findings.

In conclusion, elevated hs-cTnT or NT-proBNP identifies individuals with elevated BP and stage 1 hypertension not currently recommended for BP lowering medication who are at higher risk for CV events and may benefit from initiation of antihypertensive medication. In contrast, among individuals with intermediate BP levels and non-elevated biomarkers, event rates are comparable to those with normal BP. Incorporation of biomarkers into risk assessment algorithms may lead to more appropriate matching of intensive BP control with patient risk. Future clinical trials are needed to evaluate biomarker-based strategies for CV risk assessment to guide BP treatment decisions.

CLINICAL PERSPECTIVE.

What is new?

• Approximately one-third of adults with elevated BP or stage 1 hypertension who are not recommended for antihypertensive medication initiation according to the 2017 ACC/AHA BP guideline had elevated hs-cTnT or NT-proBNP, and these individuals had more than 10% risk of ASCVD or HF over the 10 year follow-up period.

• Among adults recommended antihypertensive medication with BP <160/100 mm Hg, more than one-half of individuals had non-elevated hs-cTnT and NT-proBNP, which was associated with less than 10% risk of an adverse CV event.

What are the clinical implications?

• A biomarker-based approach to CV risk assessment may help identify high risk individuals with elevated BP or stage 1 HTN who are currently not recommended antihypertensive medication according to the 2017 ACC/AHA BP guideline, but who may benefit from BP lowering therapy.

ABBRIVIATIONS

ASCVD

Atherosclerotic cardiovascular disease (ASCVD)

PCE

Pooled Cohort Equations

cTnT

Cardiac troponin T

NT-proBNP

N-terminal pro-B-type natriuretic peptide

ARIC

Atherosclerosis Risk in Communities

DHS

Dallas Heart Study (DHS)

MESA

Multiethnic Study of Atherosclerosis (MESA)

CVD

Cardiovascular disease

eGFR

Estimated glomerular filtration rate

LV

Left ventricle

LVH

Left ventricular hypertrophy

NNT¹⁰

10-year number needed to treat

RRR

Relative risk reduction