Arterial Stiffness Mechanisms and Orthostatic Hypotension in SPRINT

Abstract

Background

Hypertension is a leading cardiovascular disease risk factor. Intensive blood pressure (BP) control reduces cardiovascular disease events but may be limited by treatment-related serious adverse events (SAEs).

Objectives

The purpose of this study was to determine if baseline carotid to femoral pulse wave velocity (PWV)—including the component of structural stiffening, due to remodeling of the vessel wall, and load-dependent stiffening, due to the BP load on the arterial wall—is independently associated with orthostatic hypotension (OH) and SAEs in the SPRINT (Systolic Blood Pressure Intervention Trial).

Methods

SPRINT compared intensive (<120 mm Hg) vs standard (<140 mm Hg) systolic BP goals. Carotid-femoral PWV was measured in 642 participants at baseline. Structural stiffening and load-dependent stiffening were calculated by adjusting PWV to a 120/80 mm Hg reference BP with participant-specific models. The association of PWV with OH and with other SAEs was assessed using negative binomial regression.

Results

Over a 3.0-year median follow-up, the cumulative SAE incidence was 38.7% in the intensive group and 35.1% in the standard group. OH was the most frequent event (27.0% standard; 24.4% intensive). Higher load-dependent PWV was associated with greater numbers of OH events (P = 0.005) regardless of treatment group. Higher total PWV was associated with greater rates of SAEs (P = 0.001) and this was largely driven by the association with load-dependent PWV (P < 0.001).

Conclusions

Load-dependent stiffness is associated with SAEs, including OH, regardless of BP treatment intensity. Load-dependent PWV may serve as a valuable clinical tool to identify patients who require enhanced surveillance before initiating intensive BP treatment. (Systolic Blood Pressure Intervention Trial SPRINT]; NCT01206062)

Central Illustration

Hypertension remains the most prevalent modifiable risk factor for many devastating cardiovascular disease (CVD) outcomes including myocardial infarction and stroke.¹ Despite convincing evidence that more intensive blood pressure (BP) control improves CVD and cognitive impairment outcomes,2, 3, 4, 5 optimal systolic BP (SBP) targets for adults with hypertension remain a subject of ongoing debate.6, 7, 8 The controversy in BP treatment targets stems primarily from concerns about the increased risk of serious adverse events (SAEs) including hypotension, injurious falls, and poor end-organ perfusion.^9,10 The lack of consensus from major medical societies underscore the need for improved strategies to personalize BP care strategies that optimize cardiovascular health and minimize adverse outcomes associated with hypertension treatment.

A key limitation to achieving this goal in modern hypertension management is the reliance solely on seated and resting brachial BP thresholds to assess CVD risk, without considering the physiologic and biomechanical properties of how the arterial system changes with aging and treatment of hypertension.11, 12, 13 Arterial stiffness is a key age-related factor influencing arterial health. It is tightly linked to hypertension and could provide a necessary link between BP treatment and arterial physiology.

Numerous large cohort studies and meta-analyses over the past 2 decades have demonstrated that elevated pulse wave velocity (PWV), particularly carotid-femoral PWV (cfPWV), independently predicts cardiovascular events and mortality, though PWV remains a weaker marker of CVD events compared to other noninvasive tools.14, 15, 16 However, the incremental predictive value of conventional PWV measurements beyond established risk factors remains limited, partly because PWV is strongly associated with age and BP, which account for much of its variance.^12,17,18

Recently, a novel strategy to separate the components of arterial stiffness into structural stiffness, which reflects irreversible remodeling of the arterial wall from chronic processes such as intima-media thickening, elastin fragmentation, and collagen accumulation, and load-dependent stiffness, which represents the reversible increase in arterial stiffness that occurs acutely when BP rises.¹⁹ Load-dependent stiffness results from increased mechanical loading and engagement of collagen fibers at higher distending pressures, without requiring any change in the underlying arterial wall composition.^20,21 These distinct components of arterial stiffness have been linked to incident hypertension, CVD events, mortality, chronic kidney disease, dementia, and structural changes in the brain.^19,21,22

The PWV ancillary study to the SPRINT (Systolic Blood Pressure Intervention Trial)-PWV collected information on baseline arterial stiffness measures and SAEs over the 3-year period following randomization to intensive or standard treatment.²³ No prior study has evaluated if the components of arterial stiffness can improve our ability to predict SAEs associated with the treatment of hypertension. By evaluating the relationship between arterial stiffness components and SAEs in the context of intensive vs standard SBP control, we aim to further refine risk stratification strategies for hypertensive adults.

Methods

The SPRINT design has been previously described.²³ Briefly, 9,361 participants age 50 and older with hypertension and high CVD risk without diabetes were randomized to either a standard SBP target of <140 mm Hg or a more intensive SBP target of <120 mm Hg.²³

Participants enrolled in the SPRINT-PWV ancillary study were recruited from 11 of its 102 clinical sites—10 from a single clinical center network (University of Chicago, George Washington University, University of Texas Southwestern, University of Colorado, University of Utah, University of Pittsburgh, Tufts Medical Center, Vanderbilt University, Stanford University, and University of California-San Diego) plus the Houston VA Medical Center. The study protocol was approved by Institutional Review Boards at each of the sites, and all participants provided written informed consent. The only additional exclusion criterion for entry into the ancillary study was the presence of atrial fibrillation, since this arrhythmia precludes obtaining a valid PWV result using the SphygmoCor device.

Pulse wave velocity measurement

AtCor Medical provided each study site with the SphygmoCor CPV system device with software version 9.0 dedicated solely for use in this study. The cfPWV study protocol coincided with participants’ SPRINT randomization study visit. In some instances when this schedule could not be kept, a separate visit was scheduled within 1 month of randomization. The study’s standardized manual of procedures adhered to the recommendations published by the Consensus Conference on the Clinical Applications of Arterial Stiffness.²⁴ PWV study protocol details have been published previously.^23,25

Arterial stiffness modeling

Arterial mechanics were modeled using an exponential pressure-area relationship.²⁶ As previously described, measured cfPWV was used to calculate total PWV over a person’s diastolic blood pressure-SBP range, and structural PWV was calculated at the same 120/80 reference BP range for all participants as.^19,21,22 Load-dependent stiffness was calculated from the difference between total arterial stiffness and structural arterial stiffness.^19,21,22 If the participant’s BP was over 120/80, load-dependent stiffness was positive and if the participant’s BP was <120/80, load-dependent stiffness was negative. The mathematical equations for these calculations have been published previously.²⁵

Orthostatic hypotension and serious adverse events

SAEs and monitored clinical conditions were assessed over the follow-up period as previously reported in SPRINT.²² Orthostatic hypotension (OH) was defined as a drop in SBP of at least 20 mm Hg or in diastolic BP of at least 10 mm Hg after the participant stood up, as compared with the value obtained when the participant was seated. Standing BPs were measured at screening, baseline, 1 month, 6 months, 12 months, and yearly thereafter. Participants were asked if they felt dizzy at the time the orthostatic measure was taken. A SAE was defined as an event that was fatal of life-threatening, that resulted in clinically significant or persistent disability that required or prolonged hospitalization, or that was judged by the investigator to represent a clinically significant hazard or harm to the participant that might require medical or surgical intervention to prevent one of the other events listed above. SAEs were classified as hypotension, syncope, bradycardia, electrolyte abnormality, injurious fall, and acute kidney injury or acute renal failure. An injurious fall was defined as a fall that resulted in evaluation in an emergency department or that resulted in hospitalization. Acute injury or acute renal failure was coded if the diagnosis was listed in the hospital discharge summary and was believed by the safety officer to be one of the top 3 reasons for admission or continued hospitalization.

Statistical analysis

Continuous variables were summarized using median (25th, 75th percentile), and categorical variables were summarized as number and percent among participants randomized to intensive and standard treatment, separately. The primary exposures were total, structural, and load-dependent PWV at the SPRINT baseline visit. The primary outcomes were the number of OH events during the primary SPRINT follow-up period. Secondary outcomes included all SAEs during the primary SPRINT follow-up period. We examined heterogeneity in the association of PWV and each outcome in subgroups based on randomized treatment and frailty index at baseline. Statistical analyses were performed in R version 4.3.3 with assistance from multiple R packages, including MASS, ggplot2, and Hmisc.27, 28, 29 The number of events per participant were modeled with negative binomial regression including an offset for the duration of follow-up for each participant for SAEs and the number of possible OH assessments for each participant for OH.³⁰ A negative binomial model was selected to analyze event count because recurrent adverse events are important for both a patient’s quality of life and the clinical management of patients experiencing adverse events. The negative binomial regression models included multivariable adjustment for treatment group, age, sex, race, smoking status, mean arterial pressure (MAP), number of BP medications, serum creatinine, and prior CVD. Continuous variables were modeled to allow for moderate nonlinear effects using restricted cubic splines.³¹ We used 3 knots per spline and to minimize the risk of overfitting outlier data points, knots were set at the 10th and 90th percentile.³² A log-rank test comparing models with and without each exposure was used to calculate P values. The analysis code used in the manuscript is publicly available at https://github.com/ryanpewowaruk/SPRINT_PWV_SAE.

Results

A total of 771 SPRINT participants agreed to participate and signed informed consent in the PWV ancillary study. Valid PWV data were obtained from 652 participants at baseline (n = 119 excluded due to arrhythmias or an inadequate study). Seven subjects were excluded due to missing SAE follow-up data, and an additional 3 subjects were excluded due to missing covariates, resulting in a final sample size of 642 participants. At baseline, the standard and intensive treatment groups were similar with respect to age, gender, comorbidities, and SBP (Table 1). Baseline participant characteristics by quartile of total PWV, structural PWV, and load-dependent PWV are presented in Supplemental Tables 1 to 3.

Table 1.

Baseline Participant Characteristics

	Standard Treatment (n = 326)	Intensive Treatment (n = 316)
Age (y)	74 (67-79)	73 (65-79)
Female n (%)	138 (42.3%)	119 (37.7%)
Race/ethnicity n (%)
Black	72 (22.1%)	70 (22.2%)
Hispanic	13 (4.0%)	22 (7.0%)
White	234 (71.8%)	210 (66.5%)
Other	7 (2.1%)	14 (4.4%)
Body mass index (kg/m2)	26.8 (24.1-30.0)	27.6 (25.0-31.3)
Smoking status n (%)
Current	21 (6.4%)	22 (7.0%)
Former	150 (46.0%))	148 (46.8%)
Never	155 (47.5%)	146 (46.2%)
Systolic BP (mm Hg)	139 (131-150)	140 (130-149)
Diastolic BP (mm Hg)	75 (67-83)	76 (67-84)
Number of antihypertensive medications n (%)
0	29 (8.9%)	26 (8.2%)
1	112 (34.4%)	110 (34.8%)
2	114 (35.0%)	105 (33.2%)
3	45 (13.8%)	55 (17.4%)
4	25 (7.7%)	20 (6.3%)
5	1 (0.3%)	0 (0.0%)
Prior cardiovascular disease n (%)	46 (14.1%)	39 (12.3%)
Chronic kidney disease n (%)	114 (35.0%)	112 (35.4%)
Serum creatinine (mg/dL)	1.02 (0.83, 1.24)	1.00 (0.86, 1.22)
Frailty index (continuous)	0.15 (0.11, 0.21)	0.15 (0.11, 0.21)
Frailty index (categorical)
Fit	72 (22.2%)	60 (19.0%)
Prefrail	181 (55.4%)	176 (55.9%)
Frail	73 (22.5%)	79 (25.1%)
Measured cfPWV (m/s)	10.4 (8.5, 12.4)	10.6 (8.7, 12.1)
Total PWV (m/s)	12.4 (10.2, 14.9)	12.5 (10.5, 14.7)
Structural PWV (m/s)	12.0 (9.8, 14.5)	12.0 (10.1, 14.4)
Load-dependent PWV (m/s)	0.4 (−0.2, 0.9)	0.4 (−0.3, 1.0)

Values are median (25th, 75th percentile) unless otherwise indicated. Participant characteristics at the baseline SPRINT visit.

BP = blood pressure; cfPWV = carotid-femoral pulse wave velocity; PWV = pulse wave velocity; SPRINT = Systolic Blood Pressure Intervention Trial.

Over a median follow-up period of 3.0 (2.7, 3.4) years, 88 participants had 130 OH events in the standard treatment group and 77 participants had 103 OH events in the intensive treatment group (P value for difference in number of events from negative binomial regression = 0.13). The number of participants who reported dizziness with OH was 39 in the standard treatment group and 44 in the intensive treatment group. One hundred twenty-six participants had 206 SAEs in the standard treatment group vs 111 participants had 174 SAEs in the intensive treatment group (P value for difference in number of events from negative binomial regression = 0.004). The most common defined SAEs (Table 2) in both groups were injurious falls (standard treatment 18 events in 13 participants vs intensive treatment 19 events in 16 participants). Histograms of the numbers of OH events and SAEs per event are presented in Supplemental Figure 1.

Table 2.

Number and Percent of Subjects With Orthostatic Hypotension and Serious Adverse Events

	Standard Treatment of Participants	Intensive Treatment (n = 316)
Orthostatic hypotensiona	88 (130)	77 (103)
With dizziness	39 (49)	44 (57)
Serious adverse eventb	126 (206)	111 (174)
Hypotension	7 (8)	2 (2)
Syncope	10 (11)	6 (6)
Bradycardia	6 (6)	5 (5)
Electrolyte abnormality	11 (12)	12 (14)
Injurious fallc	13 (18)	16 (19)
Acute kidney injury or acute renal failured	9 (11)	12 (14)
Undefined serious adverse events	84 (164)	67 (133)

Values are n (events).

Abbreviation as in Table 1.

^aOrthostatic hypotension was defined as a drop in systolic BP of at least 20 mm Hg or in diastolic BP of at least 10 mm Hg after the participant stood up, as compared with the value obtained when the participant was seated. Standing BPs were measured at screening, baseline, 1 mo, 6 mo, 12 mo, and yearly thereafter. Participants were asked if they felt dizzy at the time the orthostatic measure was taken.

^bA serious adverse event was defined as an event that was fatal of life-threatening, that resulted in clinically significant or persistent disability that required or prolonged hospitalization, or that was judged by the investigator to represent a clinically significant hazard or harm to the participant that might require medical or surgical intervention to prevent one of the other events listed above.

^cAn injurious fall was defined as a fall that resulted in evaluation in an emergency department or that resulted in hospitalization.

^dAcute injury or acute renal failure was coded if the diagnosis was listed in the hospital discharge summary and was believed by the safety officer to be 1 of the top 3 reasons for admission or continued hospitalization.

Higher total and structural PWV were not associated with OH events (P > 0.11), while load-dependent PWV was associated with more OH events per participant (P = 0.004) regardless of treatment group (Central Illustration). From the 25th to 75th percentile of load-dependent PWV (−0.2 vs 0.9 m/s), the number of OH events per participant increased from 0.28 (0.19, 0.41) events to 0.37 (0.26, 0.53) events. Higher total PWV was associated with greater numbers of SAEs (P = 0.001), largely driven by the strong association of load-dependent PWV with SAEs (P < 0.001) (Central Illustration, Figure 1). From the 25th to the 75th percentile of total PWV (10.2 vs 14.9 m/s), the number of SAEs per participant increased from 0.46 (0.40, 0.53) events to 0.51 (0.45, 0.59) events. From the 25th to 75th percentile of load-dependent PWV (−0.2 vs 0.9 m/s), the number of SAEs per participant increased from 0.42 (0.36, 0.48) events to 0.55 (0.48, 0.63) events. We did not detect an association of structural PWV with the number of OH or SAE events in our analyses. As a sensitivity analysis, we also repeated the primary analyses with total PWV, structural PWV, and load-dependent PWV binned into quartiles (Supplemental Table 4) instead of modeling these variables as splines. The results using quartiles were consistent with our analyses using splines, and higher load-dependent PWV was associated with both greater rates of OH events (rate ratio: 1.92 [1.21, 3.06] for 4th vs 1st quartile) and greater rates of SAEs (rate ratio: 1.72 [1.45, 2.05] for 4th vs 1st quartile). Higher baseline MAP was associated with greater rates of SAEs (P = 0.002) and lower rates of OH (P = 0.040).

Central Illustration.

Balancing Cardiovascular Benefits and Serious Adverse Event Risk With Intensive Blood Pressure Control

In this secondary analysis of 642 SPRINT participants, baseline pulse wave velocity was partitioned into structural and load-dependent components by adjusting to a 120/80 mm Hg reference blood pressure. While intensive blood pressure goals reduce cardiovascular events, they are often limited by concerns for serious adverse events. Our findings demonstrate that higher load-dependent stiffness—reflecting blood pressure-driven arterial loading—is independently associated with a significantly increased risk of adjudicated orthostatic hypotension and total serious adverse events. Importantly, these associations remained consistent regardless of whether patients were randomized to intensive (<120 mm Hg) or standard (<140 mm Hg) blood pressure targets. Identifying elevated load-dependent pulse wave velocity at baseline may serve as a valuable clinical tool to recognize vulnerable patients who require enhanced surveillance and personalized management strategies during intensive antihypertensive therapy to minimize treatment-related complications (adapted from Pewowaruk et al²⁵). BP = blood pressure; SPRINT = Systolic Blood Pressure Intervention Trial; other abbreviations as in Figure 1.

Figure 1.

Pulse Wave Velocity Components and Risk of Adverse Clinical Events

Restricted cubic splines (with 95% CIs) showing the relationship between the exposures of baseline total pulse wave velocity (A and B), structural pulse wave velocity (C and D), and mean arterial pressure (E and F) with the clinical outcomes for all participants regardless of the treatment group. A, C, and E, The frequency of orthostatic hypotension events; B, D, and F, the frequency of SAEs per participant. The negative binomial regression models were covariate-adjusted for age, sex, race, smoking status, mean arterial pressure, number of blood pressure medications, serum creatinine, and prior cardiovascular disease and were also offset for the duration of follow-up for each participant. Continuous variables were modeled as restricted cubic splines to allow for nonlinear effects, and a log-rank test comparing models with and without each exposure was used to calculate P values. MAP = mean arterial pressure; OH = orthostatic hypotension; PWV = pulse wave velocity; SAEs = serious adverse events.

There was a large degree of overlap in the associations of total and structural PWV with SAE and OH events by treatment group (Figure 2) and frailty status (Figure 3). We detected heterogeneity in the association between load-dependent PWV and risk of SAEs across treatment arms. For participants with baseline load-dependent PWV between −1 m/s and 1 m/s, the risk of any SAEs was lower in the intensive vs standard group, whereas participants in this group with baseline load-dependent PWV >2 m/s were at higher risk for SAEs compared to their counterparts in the standard treatment group. We did not detect heterogeneity in the association of SAEs with other types of PWV.

Figure 2.

Arterial Stiffness and Adverse Events Stratified by Treatment Group

Restricted cubic splines (with 95% CIs) showing the relationship by treatment group between the exposures of baseline total pulse wave velocity (A and B), structural pulse wave velocity (C and D), and load-dependent (load-dep.) stiffness (E and F) with the clinical outcomes. G and H, The interaction between baseline mean arterial blood pressure and treatment group on the frequency of orthostatic hypotension and SAEs. Negative binomial regression models included an exposure*treatment group interaction and were covariate-adjusted for age, sex, race, smoking status, mean arterial pressure, number of blood pressure medications, serum creatinine, and prior cardiovascular disease; models were also offset for follow-up duration. Continuous variables were modeled as restricted cubic splines to allow for nonlinear effects, with P values calculated using a log-rank test comparing models with and without the interaction term. Abbreviations as in Figure 1.

Figure 3.

Arterial Stiffness and Adverse Events Stratified by Frailty Status

Restricted cubic splines (with 95% CIs) showing the relationship by degree of frailty between the exposures of baseline total pulse wave velocity (A and B), structural pulse wave velocity (C and D), and load-dependent (load-dep.) stiffness (E and F) with the clinical outcomes. G and H, The interaction between baseline mean arterial blood pressure and frailty status on the frequency of orthostatic hypotension and serious adverse events. Splines were calculated at the median frailty index for each categorical frailty level (fit, prefrail, frail). Negative binomial regression models included an exposure∗frailty index interaction and were covariate-adjusted for age, sex, race, smoking status, mean arterial pressure, number of BP medications, serum creatinine, and prior cardiovascular disease; models were also offset for follow-up duration. Continuous variables were modeled as restricted cubic splines to allow for nonlinear effects, though the interaction term was limited to linear terms to minimize overfitting; P values were calculated using a log-rank test comparing models with and without the interaction. Abbreviations as in Figure 1.

Discussion

In the current analysis, we investigated the association between load-dependent PWV and SAEs using data from a large, prospective, randomized controlled trial. We found that, although the functional relationship between baseline PWV and SAE risk varied across randomized treatment groups, a higher load-dependent PWV at baseline was associated with higher risk of SAEs during follow-up in both groups. We also found the functional relationships for total and structural PWV with the risk of SAEs varied across subgroups defined by frailty index, but there were no significant associations with load-dependent stiffness. The prevalence of hypertension increases with age, and the number of adults with optimal BP control drops significantly with age.³³ However, many prescribers are tentative to aggressively treat hypertension due to concerns about increasing adverse events and causing harm.^10,34, 35, 36, 37, 38, 39

Multiple independent studies showed that aggressive BP goals can improve CVD and cognitive impairment events without significant adverse events, even in older populations.^{3,10,37,40,41} A recent study utilizing data from SPRINT and employing a novel methodology to examine the implications of patient preferences on the benefits vs harms of intensive BP treatment showed that virtually all participants aged 65 years and older (85% to 100%, depending on preference weights) experienced net clinical benefit favoring intensive SBP targets (difference between benefits and harms).⁴² Still, many providers remain hesitant about aggressive BP treatment in older adults due to inconsistent guideline recommendations, concerns about overtreatment, side effects, which are further challenged by other social determinants of health including race/ethnicity, socioeconomic factors, and access to health care. Importantly, no prior study has attempted to improve prediction of SAEs using distinct mechanistic components of arterial stiffness, such as load-dependent vs structural stiffening, to identify which patients may be at highest risk for treatment-related harms.

Even major medical professional societies disagree on optimal BP targets for middle-aged and older adults.6, 7, 8 This disagreement underscores that a “one-size-fits-all” approach to BP targets is archaic, unpredictable, and, at times, unsafe. In order to optimize hypertension treatment, providers and prescribers need new and better ways to determine which individuals are least likely to be harmed from intensive BP targets.^10,37,43,44 Load-dependent stiffness is a novel, noninvasive strategy that has been shown to be associated with adverse events and may be able to improve BP control in hypertensive adults by using arterial metrics to better predict who can tolerate lower BP goals.^12,15, 16, 17^,45,46 In addition to having the potential to identify ideal candidates for intensive BP goals, load-dependent PWV could also serve as a guide to help prescribers identify which patients may benefit from closer follow-up after starting or increasing medications.

OH and SAEs are largely believed to be the result of reduced cardiac output from several distinct but interrelated changes that occur with aging, including increased left ventricular stiffness, decreased intravascular volume and stroke volume, baroreceptor insensitivity, and autonomic dysfunction leading to alterations in vascular tone.47, 48, 49 In some, but not all individuals, these hemodynamic abnormalities are manifested as lightheadedness and falls, activity reduction (decreased functional capacity and lower metabolic equivalents), and worse quality of life.^3,10,50 Our results indicate that load-dependent arterial stiffness also plays a role in the development of OH and SAEs. For the same initial drop in MAP when standing, individuals with high load-dependent PWV would have a larger decrease in total PWV when standing. Total PWV is tightly linked to pulse pressure, so the dynamic drop in PWV after standing could lead to a greater reduction in pulse pressure with standing and subsequent cerebral perfusion. Ultimately, this mechanism could lead to a higher incidence of OH among individuals with high load-dependent PWV.

It has also been postulated that older adults with high levels frailty are at increased risk for SAEs that result from treating hypertension, though conflicting evidence remains.^35,36,51, 52, 53, 54, 55, 56 Our analysis found associations with total arterial stiffness; however, the relationship between load-dependent PWV and the number of SAEs was not evident for those with greater frailty scores. In fact, the observed relationships were stronger among participants with lower frailty scores, and the number of OH events was similar based on frailty status (Figure 3). While prior studies have found that total arterial stiffness has been associated with frailty and injurious falls,^57,58 our findings suggest that load-dependent stiffness is most dominant component and could improve our ability to predict adverse outcomes regardless of frailty status. These unexpected findings could be due to the high degree of individual variability arterial stiffness mechanisms and possibly relates to how stiffness components change over time with and without antihypertensive therapy, which have been previously shown to be highly variable.¹⁹ Higher load-dependent PWV may also be associated with greater baroreceptor sensitivity, which could ultimately lead to autonomic dysfunction, and ultimately, OH and falls.^59,60

Our current findings add considerably to this growing body of literature of how to improve BP care in older adults. Identifying and quantifying arterial stiffness mechanisms may help identify individuals who are ideal candidates for aggressive BP targets and who are at the highest risk of having SAEs. Specifically, using baseline load-dependent PWV to individualize BP treatment has the potential to improve CVD outcomes related to hypertension while minimizing SAEs. Our results also showed stronger associations of load-dependent PWV with SAEs and OH compared to MAP regardless of treatment intensity or frailty status, further supporting that load-dependent PWV provides unique physiological information beyond BP. Recently in a 2024 scientific statement on OH in adults with hypertension,⁶⁰ the need for improved screening and diagnosis was clearly outlined. Our mechanistic approach to arterial stiffness may provide deeper insight into why certain individuals experience SAEs, including OH, despite achieving guideline-directed SBP targets and others tolerate lower goals without problems. This type of noninvasive stiffness testing could be used as a novel risk stratification tool to help identify hypertensive adults who would benefit from the most aggressive BP treatment goals, ultimately decreasing CVD events and medication side effects in this at-risk population.

Study limitations

As has been previously noted, this study is from a large randomized controlled trial with a diverse patient population that prioritized enrollment of older adults, >75 years old, and had centrally adjudicated SAEs. However, there are still several limitations need to be taken into account. While this analysis suggests that load-dependent stiffness could improve our ability to predict SAEs over time, other factors that affect arterial stiffness (ie, vascular remodeling, endothelial dysfunction) were not directly measured. Additionally, this study only evaluated carotid to femoral PWV and did not evaluate other methods of determining arterial stiffness. The interaction analyses between baseline PWV and treatment intensity were exploratory in nature. Due to the number of comparisons made, there is a risk of false positive findings. Therefore, while these interactions provide valuable insights for future research, they are intended to be hypothesis-generating only.

It also should be pointed out that the SPRINT trial recruited participants with hypertension at high CVD risk but without diabetes. Many adults with hypertension also have comorbid diabetes due to their common link with the metabolic syndrome, which could ultimately impact arterial stiffness and its mechanisms.⁶¹

The patient-specific models used assume that load-dependent stiffness and structural stiffness are independently additive, which is supported in the literature.²⁶ Finally, the Bramwell-Hill equation, which allows for the patient specific-models, assumes that an artery is an ideal elastic tube with uniform properties.

Conclusions

Our findings reveal that load-dependent stiffness is independently associated with OH and SAEs, irrespective of antihypertensive treatment intensity or frailty status. This suggests that incorporating load-dependent PWV in clinical practice could help identify individuals who are more susceptible to harm from BP treatment. These results highlight the need for a more nuanced, individualized approach to hypertension management—one that incorporates vascular biomechanics and extends beyond SBP targets alone to determine treatment decisions. Determining load-dependent arterial stiffness could improve our ability to identify adults who are at the greatest risk of SAEs and identify those who will benefit more from close follow-up prior to initiating therapy. Integrating load-dependent stiffness into clinical decision-making has the potential to refine risk stratification, guide follow-up, and ultimately improve the safety metrics with BP treatment.

Funding support and author disclosures

This SPRINT ancillary study was funded by a grant from the National Heart, Lung, and Blood Institute (NHLBI R01 HL107241 –MAS). AtCor Medical graciously provided the SphygmoCor devices that were utilized for this study. Drs Pewowaruk and Gepner have submitted patent applications related to methods of arterial stiffness calculation. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.