Skip main navigation

Cumulative Diastolic Blood Pressure Burden in Normal Systolic Blood Pressure and Cardiovascular Disease

Originally published 2024;81:273–281



The clinical significance of isolated diastolic hypertension defined by the 2017 American College of Cardiology/American Heart Association blood pressure (BP) guidelines remains inconsistent. We examined whether long-term diastolic burden predicts the first major adverse cardiovascular event in participants with sustained and untreated normal systolic BP.


The Mass General Brigham Biobank is a New England health care–based cohort recruited between 2010 and 2021. A total of 15 979 participants aged 18 to 64 years and without prior cardiovascular disease, antihypertensives, or high systolic BP were studied. The cumulative diastolic burden was determined as the area under the curve for diastolic BP (DBP) ≥80 mm Hg over 5 years before enrollment. Major adverse cardiovascular event was defined as a composite of first incident ischemic heart disease, stroke, heart failure, or all-cause death.


Of the 15 979 participants, mean (SD) age at enrollment was 47.6 (14.3) years, 11 950 (74.8%) were women, and the mean (SD) systolic BP and DBP were 118.0 (12.9) and 72.2 (9.3) mm Hg, respectively. Over a median (interquartile range) follow-up of 3.5 (1.8–5.4) years, 2467 (15.4%) major adverse cardiovascular events occurred. Using Cox proportional hazards regression, each SD increase in cumulative DBP was independently associated with a hazard ratio (95% CI) of 1.06 (1.02–1.10) without effect modification by sex (P=0.65), age (P=0.46), or race/ethnicity (P=0.24). In addition to traditional risk factors, cumulative DBP modestly improved the discrimination C index (95% CI) from 0.74 (0.72–0.75) to 0.75 (0.74–0.76; likelihood ratio test, P=0.037).


Among individuals with normal systolic BP, cumulative DBP may augment cardiovascular disease risk stratification beyond a single DBP measure and traditional risk factors.


What Is New?

In a health care cohort study of New England adults, each SD increase in cumulative diastolic blood pressure (BP) burden was independently associated with excess cardiovascular disease risk by 1.06-fold and nominally improved cardiovascular event discrimination beyond traditional risk factors.

What Is Relevant?

Persistence of high diastolic BP may inform the determination of lifestyle and pharmacological interventions among individuals without cardiovascular disease and with normal systolic BP.

Clinical/Pathophysiological Implications?

Event-based trials tailored to isolated diastolic hypertension are yet to exist, and the modern cardiovascular disease risk stratification paradigms do not account for diastolic BP. Whether earlier diastolic BP-augmented treatment initiation outweighs the resultant harm in low-risk isolated diastolic hypertension individuals should be undertaken.

High blood pressure (BP) remains the most prevalent and attributable risk factor for cardiovascular disease (CVD)–related mortality and disability worldwide and carries the largest population-attributable risk.1 With a continuous, log-linear association between BP and CVD established,2,3 the 2017 American College of Cardiology (ACC)/American Heart Association (AHA) BP guideline has redefined hypertension from systolic BP (SBP)/diastolic BP (DBP) of 140/90 mm Hg to more stringent 130/80 mm Hg cutoffs.4 Subsequently, population-based cohort studies have confirmed higher CVD risks associated with stage 1 hypertension even in its isolated subtypes.5–7

Nevertheless, the clinical relevance of isolated diastolic hypertension (IDH; SBP <130 mm Hg and DBP ≥80 mm Hg) is unclear and contemporary BP management guidelines4 do not provide specific therapeutic guidance.8 The adoption of the 2017 ACC/AHA BP guideline has substantially increased the prevalence of IDH notably in young adults.9 Furthermore, preventive interventions are commonly guided by estimated 10-year risk for atherosclerotic CVD,10 which do not require the input of DBP. Observational studies report inconsistent associations between IDH and CVD event risk potentially due to BP measurement variability, confounding by SBP, effect modification by age, and varying reference groups.11

Elucidating a putative association between IDH and CVD may enable unaddressed opportunities in CVD prevention. As long-term SBP load, trajectories, and variability have demonstrated incremental prognostic improvements for CVD prediction beyond cross-sectional risk factors,12,13 longitudinal DBP measures may, likewise, augment risk stratification. Here, we examined whether cumulative elevated diastolic burden above the ≥80-mm Hg threshold independently predicts the first major adverse cardiovascular events (MACE) in antihypertensive-naive adults without elevated SBP.


Data Availability

The present study followed the Strengthening the Reporting of Observational Studies in Epidemiology guidelines. Because of the sensitive nature of the data collected for this study, requests to access the data set from qualified researchers trained in human subject confidentiality protocols may be sent to Mass General Brigham Biobank at .

Study Population

The Mass General Brigham Biobank14 is a volunteer health care cohort recruited across 8 tertiary care and affiliated community hospitals within the Mass General Brigham health care network in New England. Its main objective is to study independent and combined effects of clinical, genetic, lifestyle, and socioenvironmental risk factors on health outcomes and disease pathophysiology. The Mass General Brigham Biobank collects and links genotype and electronic health records for its participants. The present analysis used data requested on April 6, 2022, comprising 131 155 individuals enrolled between March 10, 2010, and December 11, 2021.

To reliably assess longitudinal DBP burden, we designated a 5-year baseline period preceding the enrollment date to minimize spurious association from BP variability by leveraging an adequate number of BP measurements and to minimize the possibility of immortal time bias before inclusion in the analyses. Based on the electronic health records, we identified 15 979 participants (1) aged 18 to 64 years; (2) without prior MACE; (3) with normal (<130 mm Hg) outpatient SBP without a prescription for antihypertensives; and (4) with sustained normal untreated time-averaged SBP throughout the baseline period (Figure 1; Figure S1). Outcome ascertainment began on the date of enrollment and thereafter (observation period).

Figure 1.

Figure 1. Study design. We used a 5-year baseline period to assess longitudinal blood pressure (BP) burden and observed whether patients had major adverse cardiovascular events during the observation period. To avoid misclassification, participants with at least 3 BP measures during the baseline period were included. Participants who had a period of ≥2 years between the latest BP measurement in the observation period and the end of the observation period had their observations truncated before the gap. Baseline period includes 5 years preceding the Mass General Brigham Biobank (MGBB) enrollment date. IDH indicates isolated diastolic hypertension.

Determination of Cumulative DBP Burden

To avoid misclassification, participants with at least 3 outpatient BP measurements during the baseline period were included. BPs were measured using automated oscillometric devices provided by each health care site; BPs collected during invasive procedure, ambulatory BP monitoring, or special examinations (eg, disability assessment) were excluded. A total of 37 (0.20%) participants who had a period of 2 years or longer between the latest BP measurement in the observation period (follow-up) and the end of the observation period had their observations truncated before the gap. To avoid introducing potentially spurious readings, we excluded measurements with SBP ≥240 or <60 mm Hg or DBP ≥160 or <30 mm Hg.15

The diastolic hypertension burden was calculated by rezeroing the interpolated DBP values above ≥80 mm Hg such that only values above the threshold contribute to the absolute cumulative diastolic burden (Figure 2).15 For each participant, we performed linear interpolation on all DBP measurements to connect the measured values throughout the baseline period. Graphically, the absolute cumulative DBP burden represents the area under the curve between DBP readings ≥80 and 80 mm Hg threshold aligned with the 2017 ACC/AHA BP guideline.4 Participants who sustain DBP below 80 mm Hg during the baseline period would have zero absolute cumulative diastolic burden. As a secondary outcome, we also calculated proportional cumulative diastolic burden, which represents the percentage of the cumulative burden from DBP ≥80 mm Hg relative to the area under the curve of the entire DBP range.

Figure 2.

Figure 2. Determination of cumulative diastolic blood pressure (DBP) burden. A, The diastolic hypertension burden above a DBP ≥80 mm Hg threshold is determined by rezeroing the interpolated pressure values at the threshold, such that only values above the threshold contribute to the absolute cumulative diastolic burden (red). Linear interpolation is performed on all DBP measurements of a participant to connect the measured values across time. B, Absolute cumulative diastolic burden represents long-term elevated diastolic burden spent with DBP ≥80 mm Hg (area under the curve [AUC] between DBP measures ≥80 and 80 mm Hg threshold; red) indexed to 1 year.

Demographics, Lifestyle, and Clinical Risk Factors

Covariates data were pooled from the electronic health records within 1 year and the most proximate to the enrollment date. Self-reported sex was categorized into female or male. Self-identified race and ethnicity were classified into Hispanic or Latino, non-Hispanic Asian, non-Hispanic Black or African American, non-Hispanic White, and others including those reporting multiple categories. Current smoker was defined as ever responding yes to tobacco smoking and without records of cessation. Health insurance coverage and claims were verified by the billing department. Blood biochemistry and anthropometry measurements were collected in adherence to the manufacturers’ guidelines through routine clinical procedures; negative or zero values were excluded. The estimated glomerular filtration rate was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation. Diabetes was identified as glycated hemoglobin ≥6.5% or diagnosis records containing the term diabetes but excluding non-type 1 and·2 diabetes indications, such as prediabetes, gestational, or neonatal diabetes. BP- and lipid-lowering medications were extracted in adherence to the United States Adopted Names nomenclature classifications.16 For example, prescription records containing the suffix sartan were considered angiotensin receptor blocker antihypertensives (Supplemental Methods S1).


The primary outcome was a first composite MACE, defined as ischemic heart disease (pertaining to International Classification of Diseases Tenth Revision: I20–I25), stroke (I60–I66), heart failure (I50), or all-cause death ascertained by diagnostic codes, inpatient records, and All Patient Refined Diagnostic-Related Groups classifications. Death that occurred within the Mass General Brigham network was verified by the US Social Security Death Index; out-of-network death was collected by the participant’s next of kin, whenever possible. If a participant had >1 event during follow-up, the first event was counted as satisfying the outcome. In secondary analyses, we separately assessed the first event of each component of the primary outcome. Outcomes were available through February 27, 2022. Outcomes curation was independently validated by a medical doctor (R.S.Z.) and clinical research coordinator (S.G.) based on manual chart review among a random nonoverlapping 30 cases and 100 controls. The positive predictive value was 0.93 and the negative predictive value was 1.00. Cohen κ statistic for interrater reliability was 0.96 indicating a near-perfect agreement.

Statistical Analysis

Baseline characteristics were reported as mean (SD), median (interquartile range), or count (percentage). The cumulative incidence of MACE was estimated using the Kaplan-Meier method. The hazard ratio (HR) and 95% CIs for MACE associated with each SD increase in enrollment DBP and cumulative DBP burden were separately calculated using Cox proportional hazards models. HRs were adjusted for age, sex, race and ethnicity, body mass index, SBP, total and high-density lipoprotein cholesterols, estimated glomerular filtration rate, diabetes, cigarette smoking, and statin prescription assessed at enrollment. Additionally, antihypertensive initiation was included as a time-varying covariate to reflect the individual variability in intake duration during the follow-up. Proportional hazards assumptions were confirmed by log-minus-log plot and Schoenfeld residuals.

The prognostic value of enrollment (cross-sectional) and cumulative DBP was measured by the Harrell C index, which estimates the probability of a model assigning a higher risk to participants who sustain a CVD event within a shorter time frame. From the base model (including age, sex, race and ethnicity, body mass index, SBP, lipids, diabetes, estimated glomerular filtration rate, smoking, and medications), we compared improvements in model performance by separately adding enrollment DBP and cumulative DBP burden. The likelihood ratio test determined whether the inclusion of the respective DBP index significantly improves the model goodness-of-fit compared with that of the base model. Akaike information criterion was calculated from the maximum log-likelihood estimation to check for overfitting. Continuous net reclassification index quantified whether the addition of each DBP measure provides clinically relevant improvements in prediction.17

Seven sensitivity analyses were conducted. First, we examined whether higher enrollment and cumulative DBP significantly predict escalation to systolic and diastolic hypertension (SBP ≥130 mm Hg and DBP ≥80 mm Hg) during follow-up. Second, to eliminate the possibility of confounding and misclassification by postenrollment SBP, we excluded participants who had at least 1 SBP reading ≥130 mm Hg during follow-up. Alternatively, we censored participants when their SBP reached ≥130 mm Hg during follow-up. Third, we checked for effect modification by enrollment DBP level on the association between cumulative diastolic burden and MACE. Fourth, to account for outlying health care utilization pattern, we excluded participants with several DBP readings ≥95th percentile (more frequent than once per 29 days) and, conversely, the mean interval between DBP readings ≤5th percentile (<13 days between each reading). Fifth, age-, sex-, and race and ethnicity-stratified analyses were performed to assess subgroup-specific HRs for incident MACE as BP phenotypes are known to be transient over the life course and heterogenous across demographics.18 Sixth, we quantified whether cumulative SBP burden above ≥130 mm Hg threshold improves CVD prediction beyond traditional risk factors. Lastly, we calculated participant-specific DBP trajectory during the baseline period, defined as the slope (mean derivative) of the curve (Supplemental Methods S2), and tested whether such longitudinal DBP associates with MACE both independently and jointly with the cumulative diastolic burden.

All statistical tests were 2 sided, and statistical significance was set at P<0.05. All analyses were performed using R version 4.1.0 (R Foundation for Statistical Computing, Vienna, Austria). Study protocols complied with the tenets of the Declaration of Helsinki and were approved by the Massachusetts General Hospital Institutional Review Board (2018P001236). All participants provided signed consent.


Observation Period BP Measures and Baseline Characteristics

The study included 15 979 participants (mean [SD] age, 47.6 [14.3] years; 11 950 [74.8%] women) without prior CVD at enrollment (Table 1). Throughout the baseline period, participants had a median (interquartile range) of 16 (8–29) BP readings with a median (interquartile range) of 66 (37–111) days between each reading (Figure S2 and S3). While 7845 (49.1%) participants had sustained normotension, 8134 (50.9%) had high DBP (≥80 mm Hg) at least on 1 occasion, accumulating to a mean (SD) annualized cumulative DBP burden of 1.20 (1.41) DBParea under curve mm Hg year while maintaining a normal time-averaged SBP (Figures S4 and S5). Compared with those who sustained normal DBP for the entirety of the baseline period, participants with IDH (DBP ≥80 mm Hg on at least 1 occasion) were more likely to be older, male sex, current smokers, and to have lipid-lowering drug prescription, diabetes, and higher predicted 10-year CVD risk (P<0.001; Table S1). At enrollment, the mean (SD) SBP and DBP were 118.0 (12.9) and 72.2 (9.3) mm Hg. Both enrollment (r=0.59) and cumulative (r=0.44) DBPs were moderately correlated with their systolic counterparts (Figure S6).

Table 1. Mass General Brigham Biobank Participants With Normal Systolic BP and Without Antihypertensive Prescriptions at Enrollment

CharacteristicsCount (%)
Total, N15 979
Age at enrollment, y; mean (SD)47.62 (14.34)
Female sex11 950 (74.79)
Self-reported race
 Hispanic or Latino454 (2.84)
 Non-Hispanic Asian or Pacific Islander691 (4.32)
 Non-Hispanic Black or African American712 (4.46)
 Non-Hispanic White13 427 (84.03)
 Other including multiracial695 (4.56)
Current smoker379 (2.37)
Health insurance coverage
 Government programs3952 (24.73%)
 Private/single service plan11 960 (74.85%)
 None67 (0.42%)
Body mass index, kg/m2; mean (SD)26.30 (5.89)
Systolic BP, mm Hg; mean (SD)117.97 (12.93)
Diastolic BP, mm Hg; mean (SD)72.18 (9.25)
Antihypertensive initiation during follow-up3540 (22.15)
Total cholesterol, mg/dL; mean (SD)191.50 (30.83)
HDL cholesterol, mg/dL; mean (SD)62.22 (15.58)
Lipid-lowering medication1393 (8.72)
Estimated glomerular filtration rate, mL/min per 1.73 m286.40 (18.24)
Diabetes938 (5.87)
10-y ASCVD risk, %; median (IQR)2.28 (0.90–5.84)
Follow-up duration, y; median (IQR)3.54 (1.80–5.37)

Data are presented as mean (SD), median (IQR), or count (percentage). Baseline period refers to the 5-y period in which longitudinal BP burden was assessed before outcome ascertainment. 10-y ASCVD risk was calculated based on the pooled cohort equation. ASCVD indicates atherosclerotic cardiovascular disease; BP, blood pressure; HDL, high-density lipoprotein; and IQR, interquartile range.

Cumulative DBP Burden and MACE

During a median (interquartile range ) follow-up of 3.5 (1.8–5.4) years, 2467 (15.4%) new MACE occurred. Each SD increase in enrollment DBP was not significantly associated with MACE (HR, 1.01 [95% CI, 0.96–1.06]); instead, it delineated a nonlinear association with low and high DBP levels embodying increased cardiovascular risk (Figure 3; Table 2). However, compared with participants with consistently normal DBP, each SD increase in cumulative DBP burden was independently associated with HR of 1.06 (95% CI, 1.02–1.10), adjusting for demographics, lifestyle, clinical risk factors, and medications. Consistent effect estimates were shown irrespective of enrollment DBP threshold of <80 or ≥80 mm Hg (Figure S7) and BP measurement frequency and interval (Figure S8). In alignment, each SD increase in cumulative DBP burden also associated with greater progression risk to systolic and diastolic hypertension (1.11 [1.09–1.13]; Table S2). When restricting to participants who maintained normal SBP levels until the end of follow-up or censoring when reaching high SBP, cumulative DBP burden was more pronouncedly associated with MACE by 1.17 (1.11–1.24) and 1.13 (1.07–1.20)-fold, respectively (Tables S3 and S4). In end point–specific analyses, we consistently observed continuous associations of cumulative DBP burden with ischemic heart disease, stroke, and all-cause death but not heart failure (Table S5; Figure S9). Meanwhile, enrollment DBP is significantly associated with incident stroke (per SD, 1.10 [1.00–1.23]).

Table 2. Association Between Cumulative Diastolic Burden and Risk for Major Adverse Cardiovascular Events

DBP indicesEvent, %HR (95% CI)
Major adverse cardiovascular events2467 (15.44)
 Enrollment DBP, per SD1.01 (0.96–1.06)
 Absolute cumulative DBP burden, per SD*1.06 (1.02–1.10)
 Proportional cumulative DBP burden, per SD1.06 (1.02–1.10)

Models are adjusted for enrollment age, sex, race and ethnicity, body mass index, systolic blood pressure, total cholesterol, high-density lipoprotein cholesterol, estimated glomerular filtration rate, diabetes, cigarette smoking, statin prescription, and antihypertensive initiation as a time-varying covariate. AUC indicates area under the curve; DBP, diastolic blood pressure; and HR, hazard ratio.

* Absolute cumulative diastolic burden represents long-term diastolic burden spent with DBP ≥80 mm Hg (AUC between DBP measures ≥80 and 80 mm Hg threshold).

† Proportional cumulative diastolic burden illustrates the percentage of cumulative burden from DBP ≥80 mm Hg over total DBP AUC.

Figure 3.

Figure 3. Restricted cubic splines of hazard ratios (HRs) of diastolic blood pressure (DBP) with cardiovascular outcomes. Associations of (A) enrollment DBP and (B) absolute cumulative diastolic burden with incident major adverse cardiovascular events. Absolute cumulative diastolic burden represents long-term diastolic burden spent with DBP ≥80 mm Hg (area under the curve [AUC] between DBP measures ≥80 and 80 mm Hg threshold) indexed to 1 year. HRs are adjusted for enrollment age, sex, race, body mass index, systolic blood pressure, total cholesterol, high-density lipoprotein cholesterol, diabetes, cigarette smoking, statin prescription, and antihypertensive initiation as a time-varying covariate.

In age-stratified analyses, absolute cumulative DBP burden consistently associated with MACE among individuals <50 (1.07 [1.01–1.13]) or >50 (1.08 [1.03–1.13]) years without significant effect modification by age (P=0.46; Table S6). Simultaneously, higher enrollment DBP was significantly associated with pronounced cardiovascular risk among participants aged <50 (1.09 [1.00–1.19]) years but not in older individuals (P=0.04). Likewise, cumulative diastolic burden was associated with increased cardiovascular risk without significant heterogeneity by sex or race (Tables S7 and S8; Figure S10).

Prognosis of DBP Measures

Table 3 illustrates the discrimination ability and net reclassification index of enrollment DBP and cumulative diastolic burden beyond recognized risk factors. In addition to demographic, clinical (including SBP), and lifestyle risk factors, the inclusion of enrollment DBP did not enhance the discrimination ability for MACE prediction or improve the model goodness of fit (P=0.618) and net reclassification index (0.00 [95% CI, −0.00 to 0.01]). Absolute cumulative diastolic burden modestly improved C index from 0.735 (95% CI, 0.724–0.747) to 0.746 (95% CI, 0.735–0.760) with nominal continuous net reclassification improvement by 0.04 (0.02–0.06). The contribution of cumulative DBP measures, but not SBP, consistently augmented discrimination and reclassification abilities for all CVD subtypes and death (Tables S9 and S10).

Table 3. Prognostic Value of Enrollment vs Cumulative DBP Measures for Predicting Major Adverse Cardiovascular Events

ModelAkaike information criterionC index (95% CI)*Likelihood ratio test P valueContinuous net reclassification improvement
Base model41049.320.735 (0.724 to 0.747)
 +Enrollment DBP41051.050.735 (0.724 to 0.748)0.6180.00 (−0.00 to 0.01)
 +Absolute cumulative DBP40050.380.746 (0.735 to 0.760)0.0370.04 (0.02 to 0.06)
 +Proportional cumulative DBP40050.270.746 (0.735 to 0.758)0.0320.04 (0.02 to 0.06)

Cumulative diastolic burden represents long-term diastolic burden spent with DBP ≥80 mm Hg. DBP indicates diastolic blood pressure.

* The discrimination C index estimates the probability of a model assigning a higher risk to participants who undergo cardiovascular events compared with those without incidence.

P value <0.05 indicates that the inclusion of the corresponding DBP index significantly improves the model goodness of fit compared with the base model.

‡ Base model is adjusted for age, sex, race and ethnicity, body mass index, systolic blood pressure, total cholesterol, high-density lipoprotein cholesterol, estimated glomerular filtration rate, diabetes, cigarette smoking, statin prescription, and antihypertensives initiation as a time-varying covariate.

Secondary Analyses

As an alternative approach to characterize longitudinal DBPs, we reassessed the association between participant-specific DBP trajectory and MACE (Table S11). Compared with participants exhibiting unchanged or decreasing DBP throughout the baseline period, those with rapidly increasing DBP had greater cumulative incidence and hazard (1.19 [1.05–1.35]) for MACE (P trend, 0.029). However, the incidence between individuals with stable and modestly increasing DBP was less discernible.

Nevertheless, when simultaneously considering both cumulative DBP burden and trajectory, high (≥75th percentile) diastolic load and increasing trajectory associated with HR of 1.41 (95% CI, 1.23–1.63) compared with the persistently normal DBP group (P trend, <0.001; Table S12). Furthermore, high cumulative DBP with stable trajectory also exhibited 1.26-fold greater risk of MACE. Cumulative diastolic load and DBP trajectory conjointly predicted MACE incidence (P interaction, 0.03).


In a large prospective cohort of young-to-middle-aged US adults with untreated normal SBP, persistence of DBP above the 80-mm Hg threshold was associated with increased CVD and death compared with true low-risk normotensive individuals across all age, sex, and race strata. Longitudinal diastolic burden nominally improved MACE prediction beyond traditional risk factors. Furthermore, cumulative DBP load and trajectory additively predicted CVD incidence. These observations permit several conclusions regarding DBP-associated cardiovascular risk and management considerations for patients with IDH.

Our study contributes to our understanding of the clinical significance of IDH defined by the 2017 ACC/AHA BP guideline by leveraging longitudinal DBP measures. IDH, established using a single-occasion BP, has been shown to be associated with cardiovascular and kidney disease, including myocardial infarction, angina pectoris, atrial fibrillation, and chronic kidney disease, even after adjusting for comorbidity and competing risk of death.19–22 Findings from the IDACO (International Database on Ambulatory Blood Pressure in Relation to Cardiovascular Outcomes) showed that CVD, but not all-cause death, is linked to 24-hour out-of-office DBP only for individuals <50 years (HR, 2.87 [95% CI, 1.72–4.80]).23 Nevertheless, analysis of the UK Biobank participants with normal SBP (mean age, 54 years; 40% male sex) demonstrated nonsignificant association between IDH defined by the 2017 ACC/AHA guideline (HR, 1.08 [95% CI, 0.98–1.18]), whereas IDH by the European Society of Cardiology/National Institute for Health and Care Excellence definition (SBP <140 mm Hg and DBP ≥90 mm Hg) increased cardiovascular risk (HR, 1.15 [95% CI, 1.04–1.29]) irrespective of sex and antihypertensives intake.24 The absence of age effect modification in the present study may be attributable to longitudinal study design and rigorous identification of younger participants (mean age, 47.6 years) sustaining normal, untreated SBP. In contrast, prior literature25 reporting nonsignificant association between IDH by the 2017 ACC/AHA guideline and CVD has focused on middle-to-older-aged participants whose life course BP trajectory is largely described by elevated SBP.26 Therefore, a simple adjustment for single-occasion SBP may have incompletely captured risk conferred by cumulative SBP burden before baseline or unaccounted for residual confounding from increasing SBP during follow-up. In align, findings from general outpatient setting have illustrated a J-curve association between DBP threshold ≥80 mm Hg and coronary artery disease among 1.3 million adults with a wide SBP range and denoted relatively greater contribution of systolic hypertension to cardiovascular complications in the lower DBP range.15 Furthermore, given practical challenges with BP measurement variability, misclassification of hypertension status (eg, white coat hypertension) is not uncommon. Based on a median of 16 health care visits, our study ameliorates the possibility of spurious BP readings and illustrates a linear association between persistent IDH and MACE in all age, sex, race, and ethnicity subgroups otherwise overlooked based on cross-sectional DBP.

In parallel, cumulative diastolic burden improves discrimination ability and reclassification in addition to recognized risk factors to a modest extent. The results are akin to additive prognostic values of time averaged, cumulative load, and time at target SBP in predicting subclinical (eg, cardiac structure, function) and MACE.27,28 These longitudinal indices have underscored the criticality of primordial prevention as increased CVD risk from premature elevation to suboptimal BP levels cannot be entirely reversed albeit subsequent restoration to normal BP.19,29 However, whether cumulative DBP can be used in conjunction to its systolic counterpart and traditional risk factors (eg, lipids, diabetes) to assist clinical decision-making in individuals without apparent atherosclerotic indications or across full SBP range is unknown. Given the (1) different hemodynamic patterns and vascular pathophysiology across life course18, (2) limited applicability of guideline-supported CVD risk estimator to younger or older adults, and (3) heterogeneous attributions of BPs to CVD burden across ethnicities,30 DBP may differentially inform risk stratification across populations. Event-based trials tailored to stage 1 IDH are yet to exist and the modern CVD risk stratification paradigm does not underscore DBP. Nevertheless, our findings demonstrate that greater cumulative diastolic burden associates with higher risk of systolic and diastolic hypertension compared with the consistently normotensive category. Considering the high progression rate and established association between systolic and diastolic hypertension and CVD, whether earlier DBP-augmented treatment initiation outweighs the resultant harm, cost, and stigma in low-risk IDH individuals should be investigated in future clinical trials.

Strengths and Limitations

We relied on real-world data to rigorously characterize diastolic burden using outpatient BP measurements over a 5-year baseline period, minimizing the concern for misclassification bias. By including young-to-middle-aged multiethnic adults with sustained normal SBP levels, we were able to quantify the relative risk-associated IDH using a true low-risk normotensive comparator.8

Nevertheless, our results should be interpreted in the context of potential limitations. First, we relied on oscillometric BP measures from routine clinical practice with varied devices, environments, and adherence to the measurement protocol. Therefore, direct comparison to findings from observational cohort studies using auscultatory or ambulatory methods may be limited.15 Second, although our cumulative diastolic burden was quantified based on individuals untreated with antihypertensives during the exposure period, residual confounding is possible as 22.2% (17.4% with consistently normal DBP versus 26.7% with IDH) initiate treatment during follow-up. Whether the medication initiation, up/down titration, and adherence modify the association merits further studies with linkage to prescription dispense data. Third, the obligate 5-year baseline period may contribute to the potential for immortal time bias. However, the period was before enrollment and is a consideration for the use of all potential longitudinal risk factors. Furthermore, the IDH-associated hazards are likely to be biased toward the null as the excluded individuals would have conferred high cardiovascular risk and underwent MACE. Fourth, due to limited data on the cause of death, we were unable to account for competing risks of death from non-CVD-related origins; subdistribution proportional hazards model may better accurately describe the cardiovascular effects of IDH. Fifth, external validation is required to confirm generalizability as the present study comprised a subset of patients seeking health care primarily in tertiary or specialty hospitals with higher event rate compared to that of the general population. Conversely, patients may have migrated to out-of-network practices. Nevertheless, the Mass General Brigham is the largest multi-institutional health system in New England spanning tertiary care and community hospitals allowing a granular estimation of IDH burden and exhaustive outcome ascertainment. Expanding analyses to more geographically and racially diverse populations may identify subgroups who merit earlier and more aggressive interventions attributed to DBP-associated CVD risk.


Persistence of IDH was significantly associated with incident MACE risk among young-to-middle-aged adults with sustained normal SBP. With the current lack of atherosclerotic CVD prevention-oriented randomized controlled trials for patients with 2017 ACC/AHA BP guideline-defined stage 1 hypertension who confer substantial lifetime benefit from earlier interventions, cumulative DBP measures may more rigorously stratify CVD risk in this group and highlight new opportunities for CVD prevention.


In New England-based healthcare cohort participants with normal, untreated systolic BP, cumulative DBP independently associated with greater risk of MACE irrespective of age, sex, race and ethnicity. Persistent monitoring of DBP may augment future CVD risk identification.


Nonstandard Abbreviations and Acronyms


American College of Cardiology


American Heart Association


blood pressure


cardiovascular disease


diastolic blood pressure


hazard ratio


isolated diastolic hypertension


major adverse cardiovascular events


systolic blood pressure

Disclosures M.C. Honigberg reports consulting fees from CRISPR Therapeutics, advisory board service for Miga Health, and research support from Genentech; P. Natarajan reports personal consulting fees from Allelica, Amgen, Apple, AstraZeneca, Blackstone Life Sciences, Foresite Labs, Genentech/Roche, Novartis, and TenSixteen Bio, investigator-initiated grants from Apple, AstraZeneca, Amgen, Genentech/Roche, Novartis, and Boston Scientific is a cofounder of TenSixteen Bio, is a scientific advisory board member of Esperion Therapeutics, TenSixteen Bio, and geneXwell, and spousal employment at Vertex, all unrelated to the present work. The other authors report no conflicts.


For Sources of Funding and Disclosures, see page 280.

Supplemental Material is available at

Correspondence to: Pradeep Natarajan, Cardiovascular Research Center, Massachusetts General Hospital, 185 Cambridge St, CPZN 3.184, Boston, MA 02114. Email


  • 1. Tsao CW, Aday AW, Almarzooq ZI, Alonso A, Beaton AZ, Bittencourt MS, Boehme AK, Buxton AE, Carson AP, Commodore-Mensah Y, et al. Heart disease and stroke statistics-2022 update: a report from the American Heart Association.Circulation. 2022; 145:e153–e639. doi: 10.1161/CIR.0000000000001052LinkGoogle Scholar
  • 2. Lewington S, Clarke R, Qizilbash N, Peto R, Collins R; Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies.Lancet. 2002; 360:1903–1913. doi: 10.1016/s0140-6736(02)11911-8CrossrefMedlineGoogle Scholar
  • 3. Wright JT, Williamson JD, Whelton PK, Snyder JK, Sink KM, Rocco MV, Reboussin DM, Rahman M, Oparil S, Lewis CE, et al; SPRINT Research Group. A randomized trial of intensive versus standard blood-pressure control.N Engl J Med. 2015; 373:2103–2116. doi: 10.1056/NEJMoa1511939CrossrefMedlineGoogle Scholar
  • 4. Whelton PK, Carey RM, Aronow WS, Casey DE, Collins KJ, Dennison Himmelfarb C, DePalma SM, Gidding S, Jamerson KA, Jones DW, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: executive summary: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines.Hypertension. 2018; 71:e13–e115. doi: 10.1161/HYP.0000000000000065LinkGoogle Scholar
  • 5. Lee H, Yano Y, Cho SMJ, Park S, Lloyd-Jones DM, Kim HC. Cardiovascular risk of isolated diastolic hypertension defined by the 2017 American College of Cardiology/American Heart Association blood pressure guideline: a nationwide age-stratified cohort study.Hypertension. 2020; 76:e44–e46. doi: 10.1161/HYPERTENSIONAHA.120.16018LinkGoogle Scholar
  • 6. Li FR, He Y, Yang HL, Liu HM, Zhou R, Chen GC, Wu XX, Zou MC, Wang JY, Wu XB. Isolated systolic and diastolic hypertension by the 2017 American College of Cardiology/American Heart Association guidelines and risk of cardiovascular disease: a large prospective cohort study.J Hypertens. 2021; 39:1594–1601. doi: 10.1097/HJH.0000000000002805CrossrefMedlineGoogle Scholar
  • 7. Wu S, Ji C, Shi J, Chen S, Huang Z, Jonas JB. Isolated diastolic hypertension as defined by the 2017 American College of Cardiology/American Heart Association blood pressure guideline and incident cardiovascular events in Chinese.J Hypertens. 2021; 39:519–525. doi: 10.1097/HJH.0000000000002659CrossrefMedlineGoogle Scholar
  • 8. Yano Y, Kim HC, Lee H, Azahar N, Ahmed S, Kitaoka K, Kaneko H, Kawai F, Mizuno A, Viera AJ. Isolated diastolic hypertension and risk of cardiovascular disease: controversies in hypertension-pro side of the argument.Hypertension. 2022; 79:1563–1570. doi: 10.1161/HYPERTENSIONAHA.122.18459LinkGoogle Scholar
  • 9. Cho SMJ, Lee H, Kim HC. Differences in prevalence of hypertension subtypes according to the 2018 Korean Society of Hypertension and 2017 American College of Cardiology/American Heart Association guidelines: the Korean National Health and Nutrition Examination Survey, 2007-2017 (KNHANES IV-VII).Clin Hypertens. 2019; 25:26. doi: 10.1186/s40885-019-0129-5CrossrefMedlineGoogle Scholar
  • 10. Goff DC, Lloyd-Jones DM, Bennett G, Coady S, D’Agostino RB, Gibbons R, Greenland P, Lackland DT, Levy D, O’Donnell CJ, et al; American College of Cardiology/American Heart Association Task Force on Practice Guidelines. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association task force on practice guidelines.Circulation. 2014; 129:S49–S73. doi: 10.1161/01.cir.0000437741.48606.98LinkGoogle Scholar
  • 11. Jacobsen AP, McKittrick M, Daya N, Rifai MA, McEvoy JW. Isolated diastolic hypertension and risk of cardiovascular disease: controversies in hypertension-con side of the argument.Hypertension. 2022; 79:1571–1579. doi: 10.1161/HYPERTENSIONAHA.122.18458LinkGoogle Scholar
  • 12. Allen NB, Krefman AE, Labarthe D, Greenland P, Juonala M, Kähönen M, Lehtimäki T, Day RS, Bazzano LA, Van Horn LV, et al. Cardiovascular health trajectories from childhood through middle age and their association with subclinical atherosclerosis.JAMA Cardiol. 2020; 5:557–566. doi: 10.1001/jamacardio.2020.0140CrossrefMedlineGoogle Scholar
  • 13. Stevens SL, Wood S, Koshiaris C, Law K, Glasziou P, Stevens RJ, McManus RJ. Blood pressure variability and cardiovascular disease: systematic review and meta-analysis.BMJ. 2016; 354:i4098. doi: 10.1136/bmj.i4098CrossrefMedlineGoogle Scholar
  • 14. Castro VM, Gainer V, Wattanasin N, Benoit B, Cagan A, Ghosh B, Goryachev S, Metta R, Park H, Wang D, et al. The mass general Brigham Biobank portal: an i2b2-based data repository linking disparate and high-dimensional patient data to support multimodal analytics.JAMIA. 2022; 29:643–651. doi: 10.1093/jamia/ocab264CrossrefGoogle Scholar
  • 15. Flint AC, Conell C, Ren X, Banki NM, Chan SL, Rao VA, Melles RB, Bhatt DL. Effect of systolic and diastolic blood pressure on cardiovascular outcomes.N Engl J Med. 2019; 381:243–251. doi: 10.1056/NEJMoa1803180CrossrefMedlineGoogle Scholar
  • 16. United States Adopted Names naming guidelines. American Medical Association. 2021. Accessed March 18, 2023. Scholar
  • 17. Cook NR. Quantifying the added value of new biomarkers: how and how not.Diagn Progn Res. 2018; 2:14. doi: 10.1186/s41512-018-0037-2CrossrefMedlineGoogle Scholar
  • 18. Mitchell GF, Wang N, Palmisano JN, Larson MG, Hamburg NM, Vita JA, Levy D, Benjamin EJ, Vasan RS. Hemodynamic correlates of blood pressure across the adult age spectrum: noninvasive evaluation in the Framingham Heart Study.Circulation. 2010; 122:1379–1386. doi: 10.1161/CIRCULATIONAHA.109.914507LinkGoogle Scholar
  • 19. Lee H, Yano Y, Cho SMJ, Park JH, Park S, Lloyd-Jones DM, Kim HC. Cardiovascular risk of isolated systolic or diastolic hypertension in young adults.Circulation. 2020; 141:1778–1786. doi: 10.1161/CIRCULATIONAHA.119.044838LinkGoogle Scholar
  • 20. Kaneko H, Itoh H, Yotsumoto H, Kiriyama H, Kamon T, Fujiu K, Morita K, Michihata N, Jo T, Takeda N, et al. Association of isolated diastolic hypertension based on the cutoff value in the 2017 American College of Cardiology/American Heart Association blood pressure guidelines with subsequent cardiovascular events in the general population.J Am Heart Assoc. 2020; 9:e017963. doi: 10.1161/JAHA.120.017963LinkGoogle Scholar
  • 21. Lee SR, Han KD, Choi EK, Ahn HJ, Oh S, Lip GYH. Risk of atrial fibrillation in young adults with isolated diastolic, isolated systolic, and systolic-diastolic hypertension.Hypertension. 2021; 78:1248–1258. doi: 10.1161/HYPERTENSIONAHA.121.17399LinkGoogle Scholar
  • 22. Bae EH, Lim SY, Jung JH, Oh TR, Choi HS, Kim CS, Ma SK, Han K-D, Kim SW. Chronic kidney disease risk of isolated systolic or diastolic hypertension in young adults: a nationwide sample based-cohort study.J Am Heart Assoc. 2021; 10:e019764. doi: 10.1161/JAHA.120.019764LinkGoogle Scholar
  • 23. McEvoy JW, Yang WY, Thijs L, Zhang ZY, Melgarejo JD, Boggia J, Hansen TW, Asayama K, Ohkubo T, Dolan E, et al; International Database on Ambulatory Blood Pressure in Relation to Cardiovascular Outcomes (IDACO) Investigators. Isolated diastolic hypertension in the IDACO study: an age-stratified analysis using 24-hour ambulatory blood pressure measurements.Hypertension. 2021; 78:1222–1231. doi: 10.1161/HYPERTENSIONAHA.121.17766LinkGoogle Scholar
  • 24. McGrath BP, Kundu P, Daya N, Coresh J, Selvin E, McEvoy JW, Chatterjee N. Isolated diastolic hypertension in the UK Biobank: comparison of ACC/AHA and ESC/NICE guideline definitions.Hypertension. 2020; 76:699–706. doi: 10.1161/HYPERTENSIONAHA.120.15286LinkGoogle Scholar
  • 25. McEvoy JW, Daya N, Rahman F, Hoogeveen RC, Blumenthal RS, Shah AM, Ballantyne CM, Coresh J, Selvin E. Association of isolated diastolic hypertension as defined by the 2017 ACC/AHA blood pressure guideline with incident cardiovascular outcomes.JAMA. 2020; 323:329–338. doi: 10.1001/jama.2019.21402CrossrefMedlineGoogle Scholar
  • 26. Allen NB, Khan SS. Blood pressure trajectories across the life course.Am J Hypertens. 2021; 34:234–241. doi: 10.1093/ajh/hpab009CrossrefMedlineGoogle Scholar
  • 27. Teramoto K, Nadruz W, Matsushita K, Claggett B, John JE, Skali H, Solomon S, Cheng S, Shah AM. Mid- to late-life time-averaged cumulative blood pressure and late-life cardiac structure, function, and heart failure.Hypertension. 2020; 76:808–818. doi: 10.1161/HYPERTENSIONAHA.120.14833LinkGoogle Scholar
  • 28. Wang N, Harris K, Hamet P, Harrap S, Mancia G, Poulter N, Williams B, Zoungas S, Woodward M, Chalmers J, et al. Cumulative systolic blood pressure load and cardiovascular risk in patients with diabetes.J Am Coll Cardiol. 2022; 80:1147–1155. doi: 10.1016/j.jacc.2022.06.039CrossrefMedlineGoogle Scholar
  • 29. Cho SMJ, Koyama S, Ruan Y, Lannery K, Wong M, Ajufo E, Lee H, Khera AV, Honigberg MC, Natarajan P. Measured blood pressure, genetically predicted blood pressure, and cardiovascular disease risk in the UK Biobank.JAMA Cardiol. 2022; 7:1129–1137. doi: 10.1001/jamacardio.2022.3191CrossrefMedlineGoogle Scholar
  • 30. Peters SAE, Wang X, Lam TH, Kim HC, Ho S, Ninomiya T, Knuiman M, Vaartjes I, Bots ML, Woodward M; Asia Pacific Cohort Studies Collaboration. Clustering of risk factors and the risk of incident cardiovascular disease in Asian and Caucasian populations: results from the Asia Pacific Cohort Studies Collaboration.BMJ Open. 2018; 8:e019335. doi: 10.1136/bmjopen-2017-019335CrossrefMedlineGoogle Scholar


eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.