Isolated Diastolic Hypertension and Cardiovascular Disease Risk in Young Adults

Abstract

Background:

Whether isolated diastolic hypertension (IDH) in young adults represents a benign, age-specific phenotype or confers increased cardiovascular disease (CVD) risk remains uncertain. We assessed the associations between hypertension subtypes - IDH, isolated systolic hypertension (ISH), and systolic-diastolic hypertension (SDH) - with atherosclerotic CVD (ASCVD) and heart failure (HF), and characterized longitudinal transitions in hypertension subtypes from young adulthood into later life.

Methods:

Adults aged ≥18 years without CVD were included from four US cohorts. Hypertension subtypes were defined using BP thresholds of ≥130/80 mm Hg. Cox proportional hazard models estimated the associations between hypertension subtypes with ASCVD and HF.

Results:

Among 23,957 participants (mean age 44.5 years, 42.5% young adults aged 18–39 years, 45.1% male), 2,320 ASCVD and 1,243 HF events occurred over a median follow-up of 17.2 years. In young adults, the adjusted hazard ratios (HR [95% CI]) for ASCVD were 1.36 (1.11–1.68) for IDH, 1.54 (1.02–2.32) for ISH, and 1.84 (1.46–2.32) for SDH, compared to normotension. Corresponding HRs for HF were 1.69 (1.24–2.31) for IDH, 1.92 (1.09–3.40) for ISH, and 1.69 (1.17–2.43) for SDH. Associations were consistent among adults aged ≥40 years. In young adults with IDH, 29.8% remained with IDH while 36.5% progressed to SDH later in life.

Conclusions:

IDH in young adulthood represents an early marker of elevated CVD risk and a precursor to more adverse hypertension subtypes. These findings underscore the importance of early recognition, ongoing BP monitoring, and guideline-concordant pharmacologic treatment when indicated among young adults with IDH to reduce long-term CVD risk.

Graphical Abstract

Introduction

Isolated diastolic hypertension (IDH), defined as diastolic blood pressure (DBP) in the hypertensive range with systolic blood pressure (SBP) below the hypertensive threshold, is more common in younger adults and less prevalent in middle-aged or older adults.^1–3 However, it remains uncertain whether IDH represents a benign, age-related phenomenon in young adults or confers an increased risk of cardiovascular disease (CVD). Previous studies have reported inconsistent associations between IDH and CVD outcomes. Analyses from US cohorts and the UK Biobank found no association between IDH and CVD risk,^4,5 whereas analyses of the Japan Medical Data Center database reported a higher risk of CVD among individuals with IDH.⁶ These discrepancies may be partly attributable to differences in the DBP threshold used to define IDH (DBP ≥80 vs. ≥90 mm Hg), the choice of reference group (normotension on both SBP and DBP vs. non-IDH), and differences in study populations (younger vs. older adults). Furthermore, analyses of the Korean National Health Insurance Service database suggested that IDH may be more strongly associated with CVD risk among young adults aged 20–39 years than older age groups, with associations attenuated with advancing age.^7,8 Whether IDH confers consistent CVD risk among US young adults remains unclear.

Young adults with IDH may also experience transitions over time to systolic-diastolic hypertension (SDH: both SBP and DBP in the hypertensive range) or other hypertension subtypes later in life, but the prevalence and longitudinal patterns of these transitions remain poorly characterized in diverse US cohorts. Moreover, the optimal DBP level for minimizing CVD risk among young adults with IDH is unknown.^9–12 The 2025 AHA/ACC BP guideline has identified this evidence gap as a critical area for future research.¹

The current study sought to (1) assess the associations of IDH with incident atherosclerotic CVD (ASCVD) and heart failure (HF) among US young adults; (2) characterize the within-person longitudinal transitions in hypertension subtypes from young adulthood into later life; and (3) identify DBP levels associated with the lowest risk of ASCVD and HF among individuals with SBP <130 mm Hg.

Method

Study Design and Cohorts

Because of the sensitive nature of the data collected for this study and restrictions from participating cohort studies, qualified researchers trained in human subject confidentiality protocols may request access through the respective cohort study coordinating centers, in accordance with each cohort’s established data access procedures.

This study analyzed data from four large, population-based prospective epidemiologic cohorts in the US: (1) Coronary Artery Risk Development in Young Adults (CARDIA);¹³ (2) Framingham Heart Study (FHS; including data from the Offspring, Third Generation, Omni 1, and Omni 2 cohorts);^14–16 (3) Jackson Heart Study (JHS);¹⁷ and (4) Multi-Ethnic Study of Atherosclerosis (MESA).¹⁸ The designs of each cohort have been described previously and are detailed in the Supplemental Methods. All study protocols were approved by institutional review boards at participating institutions, and all participants provided written informed consent.

In the current analysis, we included participants aged 18 years or older (Figure S1). We excluded individuals with a history of CVD at baseline (N=756), defined as a history of acute coronary syndrome, myocardial Infarction (MI), stroke, transient ischemic attack, HF, cardiac arrest, stable or unstable angina, coronary revascularization procedures, or peripheral arterial disease. We also excluded individuals with missing hypertension subtype data (N=1,294), missing traditional CVD risk factors required for multivariable adjustment (N=733), or missing follow-up for incident ASCVD and HF events (N=619). The final analytic sample included 23,957 participants. The first visit at which participants met all inclusion and exclusion criteria was designated as the baseline for this analysis. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.

Clinical Data Collection

Demographic, traditional ASCVD and HF risk factors, and BPs were measured using standardized protocols in each study.^13–18 Information on age, sex, race, ethnicity, smoking status, history of CVD, and medication use was self-reported. Laboratory assays included fasting glucose, low-density lipoprotein (LDL-C), and high-density lipoprotein cholesterol (HDL-C). Diabetes was defined as a fasting glucose level ≥126 mg/dL, the use of anti-diabetic medication, or self-reported diabetes status.

BP Measurement and Subtype Classification

BP was measured by trained staff using standardized methods in each cohort. In CARDIA, BP was measured three times at each study visit, and the average of the second and third readings was used for the analysis. At year 0 to 15, BP was measured using a random-zero sphygmomanometer (Hawksley & Sons, Ltd); beginning at year 20, BP was measured with an automated oscillometric device (Omron HEM907XL).¹⁹ The oscillometric values were calibrated to the random-zero scale using study-specific equations to ensure comparability across exams.²⁰ In FHS, BP was measured twice at each visit using a mercury sphygmomanometer, and the average of the two readings was used.²¹ In JHS, BP was measured twice and the two readings were averaged. At visit 1, BP was measured using a random-zero sphygmomanometer (Hawksley & Sons, Ltd); at visit 2 and 3, BP was measured using an oscillometric device (Omron HEM-907XL). Random-zero values were calibrated to the oscillometric device using robust regression.²² In MESA, BP was measured three times using an automated oscillometric device (Dinamap PRO 100). The average of the second and third measurements was used for analysis.²³

In all four cohorts, we classified individuals into four hypertension subtypes using their baseline BP: (1) normotension, defined as SBP <120 mm Hg and DBP <80 mm Hg without antihypertensive medication; (2) IDH, defined as SBP <130 mm Hg and DBP ≥80 mm Hg; (3) isolated systolic hypertension (ISH), defined as SBP ≥130 mm Hg and DBP <80 mm Hg; and (4) SDH, defined as SBP ≥130 mm Hg and DBP ≥80 mm Hg.¹ Individuals with SBP <120 mm Hg and DBP <80 mm Hg who were taking antihypertensive medication were not classified as normotensive because their untreated BP phenotype could not be reliably determined. These individuals, along with those with SBP 120–129 mm Hg and DBP <80 mm Hg, were not assigned to one of the four prespecified BP phenotype categories and were excluded from the analyses.

Follow-Up and outcome Events

The outcomes were time to the first incident ASCVD event and time to the first incident HF event. ASCVD was defined as nonfatal MI, death from coronary heart disease, and fatal or nonfatal stroke. Events were ascertained and adjudicated using the specific protocol of each cohort.^13–18 Additional details are provided in the Supplemental Methods.

Statistical Analysis

Participant characteristics at the baseline visit were summarized for the overall pooled cohort, by individual study, and by age group (18–39 years and ≥40 years). Incidence rates of ASCVD and HF events were calculated for overall, by hypertension subtype, and within age group.

To examine the associations between hypertension subtypes at baseline and the risk of incident ASCVD and HF events, we used Cox proportional hazards regression with age as the time scale after confirming that the assumption of proportional hazards was met. In primary analysis, we performed stratified analyses by age (18–39 vs. ≥40 years), given our a priori hypothesis that the association between hypertension subtypes and CVD may differ by age. In secondary analyses, we also examined heterogeneity by sex (male vs. female), race and ethnicity (White vs. non-White), body mass index (BMI) (<30 vs. ≥30 kg/m²), diabetes status, and smoking status (non-current vs. current), as these are key demographic and clinical factors relevant to cardiovascular risk. Statistical interactions were tested using likelihood ratio tests. Models were adjusted for established cardiovascular risk factors. Model 1 was adjusted for sex and race/ethnicity. Model 2 was further adjusted for BMI, smoking status, LDL-C, HDL-C, antihypertensive medication use, lipid-lowering therapy, and diabetes status. To account for potential cohort effects, all models were stratified by study cohort, allowing the baseline hazard function to vary across cohorts.

To characterize longitudinal transitions in hypertension subtypes during follow-up, we assessed in a subgroup of young adult participants who had follow-up BP measures during middle-to-older adulthood period (50–69 years). For each participant, the hypertension subtype was defined based on BP measures from the earliest available visit within each period. Pairwise transitions between subtypes were summarized and visualized with a Sankey diagram.

To determine age-specific (18–39 vs. ≥40 years) DBP levels associated with the lowest ASCVD and HF risk among adults with SBP <130 mm Hg, we modeled DBP as a continuous variable using restricted cubic splines in Cox proportional hazards models. Nonlinearity was assessed by testing the significance of the nonlinear spline terms.

For sensitivity analyses, we restricted the analyses to participants not taking antihypertensive medication therapy at baseline and censored follow-up at the initiation of therapy. Additionally, we re-defined hypertension subtypes using alternative BP thresholds of 140/90 mm Hg.²⁴ All analyses were conducted using R version 4.5.0 (R Foundation for Statistical Computing). Two-sided P values <0.05 were considered statistically significant.

Results

A total of 23,957 participants were included in the analyses, with mean (SD) age 44.5 (16.9) years, 10,191 (42.5%) were young adults aged 18–39 years, 10,805 (45.1%) were male, and 13,535 (56.5%) self-identified as non-Hispanic White (Table 1). At baseline, 12,371 (51.6%) had normotension, 3,669 (15.3%) had IDH, 3,534 (14.8%) had ISH, and 4,383 (18.3%) had SDH. Detailed characteristics of the study participants by individual study are shown in Table S1.

Table 1.

Characteristics of participants by age^a

Characteristic	Overall N = 23,957	Age 18–39 N = 10,191	Age ≥40 N = 13,766
Age	44.5 ± 16.9	28.3 ± 5.7	56.6 ± 11.4
Study
CARDIA	5,070 (21.2%)	5,060 (49.7%)	10 (0.1%)
FHS	9,444 (39.4%)	4,728 (46.4%)	4,716 (34.3%)
JHS	3,108 (13.0%)	403 (4.0%)	2,705 (19.6%)
MESA	6,335 (26.4%)	0	6,335 (46.0%)
Sex
Male	10,805 (45.1%)	4,635 (45.5%)	6,170 (44.8%)
Female	13,152 (54.9%)	5,556 (54.5%)	7,596 (55.2%)
Race/Ethnicity
Non-Hispanic White	13,535 (56.5%)	7,010 (68.8%)	6,525 (47.4%)
Non-Hispanic Black	7,685 (32.1%)	3,037 (29.8%)	4,648 (33.8%)
Other	2,737 (11.4%)	144 (1.4%)	2,593 (18.8%)
BMI, kg/m2	27.0 ± 5.9	25.1 ± 5.4	28.5 ± 5.9
Smoking status
Not smoking	12,775 (53.3%)	5,450 (53.5%)	7,325 (53.2%)
Previously smoked	5,753 (24.0%)	1,577 (15.5%)	4,176 (30.3%)
Currently Smoking	5,429 (22.7%)	3,164 (31.0%)	2,265 (16.5%)
SBP, mm Hg	121.4 ± 19.0	112.5 ± 12.5	128.0 ± 20.3
DBP, mm Hg	74.3 ± 10.8	71.9 ± 10.5	76.1 ± 10.7
Hypertension subtypes
Normotension	12,371 (51.6%)	7,341 (72.0%)	5,030 (36.5%)
IDH	3,669 (15.3%)	1,685 (16.5%)	1,984 (14.4%)
ISH	3,534 (14.8%)	271 (2.7%)	3,263 (23.7%)
SDH	4,383 (18.3%)	894 (8.8%)	3,489 (25.3%)
Total cholesterol, mg/dL	191.3 ± 37.2	181.3 ± 34.7	198.7 ± 37.4
LDL cholesterol, mg/dL	118.1 ± 34.4	112.1 ± 32.8	122.6 ± 34.9
HDL cholesterol, mg/dL	52.7 ± 15.1	52.7 ± 14.0	52.7 ± 15.8
Diabetes	1,577 (6.6%)	100 (1.0%)	1,477 (10.7%)
Antihypertensive medication	3,973 (16.6%)	141 (1.4%)	3,832 (27.8%)
Lipid-lowering medication	1,840 (7.7%)	50 (0.5%)	1,790 (13.0%)

^aNumbers are mean (SD) for normally distributed data, and n (%) for categorical variables.

BMI, body mass index; BP, blood pressure; DBP, diastolic blood pressure; HDL cholesterol: high-density lipoprotein cholesterol; IDH: isolated diastolic hypertension; ISH: isolated systolic hypertension; LDL cholesterol: low-density lipoprotein cholesterol; SDH: systolic and diastolic hypertension; SBP: systolic blood pressure.

During a median follow-up of 17.2 years, a total of 2,320 incident ASCVD and 1,243 incident HF events occurred (Table S2). Among young adults aged 18–39 years, the hazard ratio (HR) for ASCVD was 1.36 (95% CI: 1.11 to 1.68) for IDH, 1.54 (95% CI: 1.02 to 2.32) for ISH, and 1.84 (95% CI: 1.46 to 2.32) for SDH in the multivariable-adjusted model, compared with normotension (Figure 1). Similarly, HR for HF was 1.69 (95% CI: 1.24 to 2.31) for IDH, 1.92 (95% CI: 1.09 to 3.40) for ISH, and 1.69 (95% CI: 1.17 to 2.43) for SDH, compared to young adults with normotension (Figure 2). No significant interaction between hypertension subtypes and age was observed for ASCVD or HF (P for interaction=0.87 and 0.38, respectively). In secondary analyses, no significant interactions were observed for ASCVD across any subgroup factor (all P for interaction >0.05) (Figure S2). For HF, significant interactions were observed for race and ethnicity and diabetes status (both P for interaction =0.04).

Figure 1.

Associations between hypertension subtypes and ASCVD.

Caption: Model 1: adjusted for sex, race/ethnicity. Model 2: Model 1 + body mass index, smoking status, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, diabetes, antihypertensive medication, lipid lowering medication. All models were stratified by study cohort. ASCVD: atherosclerotic cardiovascular disease; HR: hazard ratio; IDH: isolated diastolic hypertension; ISH: isolated systolic hypertension; SDH: systolic and diastolic hypertension.

Figure 2.

Associations between hypertension subtypes and HF.

Caption: Model 1: adjusted for sex, race/ethnicity. Model 2: Model 1 + body mass index, smoking status, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, diabetes, antihypertensive medication, lipid lowering medication. All models were stratified by study cohort. HF: heart failure; HR: hazard ratio; IDH: isolated diastolic hypertension; ISH: isolated systolic hypertension; SDH: systolic and diastolic hypertension.

In sensitivity analysis excluding participants taking antihypertensive medication at baseline and censoring follow-up at medication initiation (N=19,984), associations between hypertension subtypes with ASCVD and HF remained similar to those observed in the main analyses (Figure S3 and Figure S4). Additionally, re-defining hypertension subtypes using alternative BP threshold of 140/90 mm Hg found consistent associations (Table S3, Figure S5 and Figure S6).

Among 5,459 young adults with follow-up BP measures during middle-to-older adulthood (50–69 years), the prevalence of normotension, IDH, ISH, and SDH was 70.8%, 17.0%, 2.4%, and 9.7%, respectively, in young adulthood; the corresponding prevalence in middle-to-older adulthood was 46.7%, 18.0%, 8.0%, and 27.3% (Figure 3). Among young adults with IDH, 29.8% remained with IDH, 24.3% transitioned to normotension, 9.4% to ISH, and 36.5% to SDH in middle-to-older adulthood. Compared to young adults with IDH who remained with IDH in later life, those who transitioned to SDH were more likely to be female and non-Hispanic Black, and had higher levels of SBP and HDL-C (Table S4). Sensitivity analysis excluding participants who were taking antihypertensive medications at baseline and censoring follow-up at medication initiation (N=5,409) showed similar transition patterns. Among young adults with IDH, 29.7% remained with IDH, 25.0% transitioned to normotension, 9.2% to ISH, and 36.1% to SDH in middle-to-older adulthood (Figure S7).

Figure 3.

Sankey plot of hypertension subtype transitions from young adulthood (18–39 years) to middle-to-older adulthood (50–69 years).

Caption: IDH: isolated diastolic hypertension; ISH: isolated systolic hypertension; SDH: systolic and diastolic hypertension.

Among young adults with SBP <130 mm Hg, restricted cubic spline models suggested a nonlinear association between DBP and ASCVD (P for nonlinearity =0.006), with lower DBP associated with a lower risk of ASCVD at least down to a DBP of 66 mm Hg after adjustment for CVD risk factors and SBP (Figure 4). A similar nonlinear association was observed for HF among adults aged ≥40 years (P for nonlinearity =0.002), whereas the association between DBP and HF appeared to be linear among young adults (P for nonlinearity =0.18; Figure 5).

Figure 4.

Restricted cubic spline modeling the relationship between DBP with ASCVD among individuals with SBP <130 mm Hg, stratified by age.

Caption: Model 1: adjusted for sex, race/ethnicity, body mass index, smoking status, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, diabetes, antihypertensive medication, lipid lowering medication. Model 2: Model 1 + SBP. All models were stratified by study cohort. Reference point (black dot) was set at the median DBP. ASCVD: atherosclerotic cardiovascular disease; DBP: diastolic blood pressure; SBP: systolic blood pressure.

Figure 5.

Restricted cubic spline modeling the relationship between DBP with HF among individuals with SBP <130 mm Hg, stratified by age.

Caption: Model 1: adjusted for sex, race/ethnicity, body mass index, smoking status, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, diabetes, antihypertensive medication, lipid lowering medication. Model 2: Model 1 + SBP. All models were stratified by study cohort. Reference point (black dot) was set at the median DBP. HF: heart failure; DBP: diastolic blood pressure; SBP: systolic blood pressure.

Discussion

In this analysis of 23,957 adults from four large US prospective cohort studies, the prevalence of IDH was 16.5% in young adults aged 18–39 years and 14.4% in those ≥40 years. Young adults with IDH, ISH, or SDH all had significantly higher risks of ASCVD and HF compared to those with normotension, and these associations were consistent across age groups (18–39 vs. ≥40 years). Longitudinally, progression to SDH was the most common trajectory among young adults with IDH at baseline, of whom 36.5% transitioned to SDH later in life. Among young adults with SBP <130 mm Hg, DBP showed a nonlinear association with ASCVD, with lower DBP associated with a lower risk of ASCVD at least down to DBP of 66 mm Hg, whereas the relationship with HF appeared to be linear. These findings suggest that IDH in young adulthood represents an early marker of elevated CVD risk and a precursor to more adverse hypertension subtypes.

Prior studies of IDH have yielded conflicting results, with some suggesting it is a benign phenotype and others associating it with increased CVD risk. In analyses of middle-aged and older adults, IDH generally showed no significant association with incident CVD. For example, in the Atherosclerosis Risk in Communities (ARIC) study (N=8,703; mean age 56 years), IDH was not associated with incident ASCVD (HR: 1.06; 95% CI, 0.89–1.26) or HF (HR: 0.91; 95% CI, 0.76–1.09) compared with no IDH (SBP <130 and DBP <80 mm Hg).⁴ Similarly, in MESA (N=5,104; mean age 60 years), IDH was not associated with incident CVD (HR: 1.19; 95% CI, 0.77–1.84) compared with no IDH.²⁵ In the UK Biobank (N=151,831; mean age 54 years), IDH was not associated with CVD compared with no IDH when defined by BP thresholds of 130/80 mm Hg, but showed a modest association (HR: 1.15; 95% CI, 1.04–1.29) when defined using the higher threshold (SBP <140 and DBP ≥90 mm Hg vs. SBP <140 and DBP <90 mm Hg).⁵ In contrast, analyses of young adults aged 20–39 years from the Korean National Health Insurance Service (N=6,424,090) found higher CVD risk with IDH compared to normotension;⁷ and in an expanded cohort (N=8,109,484; ages 20–89 years), risk associations were modified by age and stronger at younger ages.⁸ Additionally, the International Database on Ambulatory Blood Pressure in Relation to Cardiovascular Outcomes (IDACO) study reported that that IDH based on 24-hour Ambulatory Blood Pressure Monitoring (ABPM) was associated with higher cardiovascular risk among adults aged <50 years (HR, 2.87 [95% CI, 1.72–4.80]) but not among older adults.²⁶ The current analysis extends prior evidence by focusing specifically on a diverse cohort of US young adults, and found that IDH was associated with higher risks of ASCVD and HF compared with normotension. Importantly, the magnitude of these associations was similar across age groups (18–39 vs. ≥40 years), suggesting that the adverse cardiovascular implications of IDH may begin early in adulthood and persist over time. These findings were robust across multiple sensitivity analyses, including re-defining hypertension subtypes using alternative BP thresholds, suggesting prior inconsistent results in the literature may reflect differences in study design and population characteristics rather than the absence of cardiovascular risk associated with IDH.

These findings are clinically relevant in the context of evolving hypertension guidelines. The 2025 AHA/ACC BP guideline classifies DBP 80–89 mm Hg as stage 1 hypertension and recommends consideration of antihypertensive therapy when BP remains ≥130/80 mm Hg after an initial 3–6 months of lifestyle modification to prevent target organ damage and mitigate further BP progression.¹ In contrast, the 2024 European Society of Cardiology (ESC) hypertension guideline defines IDH using a higher threshold of SBP <140 and DBP ≥90 mm Hg and recognizes IDH as a phenotype associated with modestly increased relative cardiovascular risk, particularly among younger adults.²⁷ However, the ESC guideline notes that the absolute cardiovascular risk in individuals with IDH is generally low and that it remains unclear whether IDH should routinely prompt initiation of BP-lowering medication, while emphasizing the need for follow-up because individuals with IDH are at increased risk for developing systolic hypertension.²⁷ Findings from the current study support the view that IDH in young adulthood should not be dismissed as benign, while underscoring the need for future studies to determine whether early pharmacologic treatment, beyond lifestyle intervention and monitoring, reduces long-term cardiovascular risk.

Hypertension subtypes are dynamic and evolve over the life course. In the current study, 29.8% of young adults with IDH remained with IDH in later life, while 36.5% progressed to SDH. Few prior studies have characterized such long-term trajectories of hypertension subtypes. In FHS, IDH was more likely to progress to SDH than to ISH, and ISH or SDH rarely reverted to IDH.²⁸ In MESA, baseline IDH strongly predicted the development of ISH or SDH in later life.²⁵ These transitions are biologically plausible, as arterial stiffening with age raises SBP while DBP plateaus or falls, resulting in widened pulse pressure.^29,30 Taken together, these findings suggest that IDH in young adulthood is neither benign nor stable, but rather represents an early marker of elevated CVD risk and subsequent progression to more harmful hypertension subtypes such as SDH. Clinically, this underscores the importance of early recognition and ongoing monitoring of young adults with IDH, as timely intervention may help prevent progression to SDH and mitigate long-term CVD risk.^31,32 While most BP-lowering trials have focused on SBP targets, the optimal DBP target, particularly in young adults, remains unclear.^9–12 A post-hoc analysis pooling data from Systolic Blood Pressure Intervention Trial and Action to Control Cardiovascular Risk in Diabetes BP trial found that DBP levels of 70–80 mm Hg were associated with the lowest CVD risk among individuals with treated SBP <130 mm Hg, with higher CVD risk at both lower and higher DBP values.³³ Similarly, in the Treating to New Targets trial of patients with coronary artery disease, DBP showed a J-shaped relationship with future cardiovascular events, with the lowest risk observed at DBP 81 mm Hg.³⁴ The current analysis extends these observations from high-risk older adults to lower-risk young adults and found a nonlinear association between DBP and ASCVD when SBP was <130 mm Hg, with lower DBP associated with a lower risk of ASCVD at least down to DBP of 66–68 mm Hg in both young and older adults. In contrast, the pattern for HF differed by age: among middle-aged and older adults, the spline analyses suggested the lower DBP was associated with a lower risk of HF at least down to a DBP of around 70 mm Hg, whereas among young adults, the association appeared approximately linear without clear evidence of a nonlinear nadir. This difference may reflect age-related vascular changes, such as increased arterial stiffness and impaired diastolic function, which make older adults more vulnerable to inadequate coronary perfusion at lower DBP levels, whereas younger adults may tolerate lower DBP without the same hemodynamic compromise.^29,35,36 Alternatively, the relatively small number of HF events in young adults may have limited our ability to detect non-linear associations in this group.

Strengths and Limitations

This study has several strengths, including pooling data from four large prospective cohorts with high-quality exposure and outcome assessments, large sample sizes of both young and older adults, and long follow-up durations. This study also has several limitations. First, heterogeneity in BP measurement protocols across cohorts and reliance on single-visit BP measurements for hypertension subtype assignment may have led to misclassification. Second, most young adults in this analysis were from CARDIA and FHS, whereas participants ≥40 years were primarily from FHS, JHS, and MESA. This raises the possibility of cohort effects in addition to age effects. Although we stratified Cox models by cohort to allow baseline hazards to vary, residual cohort-related influences cannot be fully excluded. Lastly, the current study population was comprised primarily of non-Hispanic White and non-Hispanic Black adults, with relatively fewer participants from Hispanic, Asian, or other racial and ethnic groups. This may limit the generalizability of our findings to those underrepresented populations.

Perspective

In this analysis of 23,957 adults from four large US cohorts, IDH, ISH, and SDH were each associated with higher risks of ASCVD and HF compared with normotension, with similar associations across age groups. More than one-third of young adults with IDH progressed to SDH in later life. These findings suggest that IDH is not a benign phenotype in young adults and underscore the importance of early recognition, lifestyle-based risk reduction, ongoing BP monitoring, and guideline-concordant pharmacologic treatment when indicated among young adults with IDH to reduce long-term CVD risk.

Supplementary Material

Supplemental_Material

Supplemental Methods

Supplemental References

Table S1. Characteristics of participants by study

Table S2. Incidence rates of ASCVD and HF outcomes

Table S3. Incidence rates of ASCVD and HF outcomes in sensitivity analyses using alternative BP thresholds (140/90 mm Hg)

Table S4. Characteristics of participants with IDH during young adulthood (18–39 years) stratified by hypertension subtypes in middle-to-older adulthood (50–69 years)

Figure S1. Flow chart

Figure S2. Associations between hypertension subtypes and ASCVD and HF outcomes, stratified by baseline characteristics.

Figure S3. Associations between hypertension subtypes and ASCVD, excluding antihypertensive medication use at baseline and censoring follow-up at medication initiation

Figure S4. Associations between hypertension subtypes and HF, excluding antihypertensive medication use at baseline and censoring follow-up at medication initiation

Figure S5. Associations between hypertension subtypes and ASCVD, using alternative BP thresholds (140/90 mm Hg)

Figure S6. Associations between hypertension subtypes and HF, using alternative BP thresholds (140/90 mm Hg)

Figure S7. Sankey plot of hypertension subtype transitions from young adulthood (18–39 years) to middle-to-older adulthood (50–69 years), excluding antihypertensive medication use at baseline and censoring follow-up at medication initiation

Novelty and Relevance.

What Is New?

In this pooled analysis of four US cohorts, IDH in young adults aged 18–39 years was associated with higher risks of ASCVD and HF compared with normotension. More than one-third of young adults with IDH progressed to SDH later in life.

What Is Relevant?

IDH is young adulthood is not a benign phenotype, but rather an early marker of elevated CVD risk and a precursor to more adverse hypertension subtypes.

Clinical/Pathophysiological Implications?

Early recognition of IDH in young adults may help identify individuals at elevated long-term cardiovascular risk. Lifestyle-based risk reduction, longitudinal BP monitoring, and guideline-concordant pharmacologic treatment when indicated may help reduce long-term cardiovascular risk in this population.

Abbreviations

ACC

American College of Cardiology

AHA

American Heart Association

ASCVD

atherosclerotic cardiovascular disease

BP

blood pressure

CARDIA

Coronary Artery Risk Development in Young Adults

CI

confidence interval

CVD

cardiovascular disease

DBP

diastolic blood pressure

FHS

Framingham Heart Study

HF

heart failure

HR

hazard ratio

IDH

isolated diastolic hypertension

ISH

isolated systolic hypertension

JHS

Jackson Heart Study

MESA

Multi-Ethnic Study of Atherosclerosis

SBP

systolic blood pressure

SDH

systolic-diastolic hypertension

Data sharing statement

Data used in this study may be requested from the Biologic Specimen and Data Repository Information Coordinating Center (BioLINCCa) repository and the database of Genotypes and Phenotypes (dbGaP). Additionally, investigators with a manuscript proposal or ancillary study proposal that has been approved by study committee may request data directly from the study coordinating centers.