Intended for healthcare professionals

  1. Research
  2. Effects of intensive...
  3. Effects of intensive blood pressure treatment on orthostatic hypertension: individual level meta-analysis
CCBY Open access
Research

Effects of intensive blood pressure treatment on orthostatic hypertension: individual level meta-analysis

BMJ 2025; 388 doi: https://doi.org/10.1136/bmj-2024-080507 (Published 25 March 2025) Cite this as: BMJ 2025;388:e080507

Linked Editorial

Orthostatic hypertension in hypertensive patients: should we bother?
  1. Stephen P Juraschek, associate professor1,
  2. Jiun-Ruey Hu, interventional cardiology fellow2,
  3. Jennifer L Cluett, assistant professor1,
  4. Carol Mita, librarian3,
  5. Lewis A Lipsitz, professor1 4,
  6. Lawrence J Appel, professor5,
  7. Nigel S Beckett, consultant geriatrician6,
  8. Barry R Davis, professor7,
  9. Rury R Holman, professor8,
  10. Edgar R Miller 3rd, professor5,
  11. Kenneth J Mukamal, professor1,
  12. Ruth Peters, professor9 10,
  13. Jan A Staessen, professor11 12 13,
  14. Addison A Taylor, professor14,
  15. Jackson T Wright Jr, professor15,
  16. William C Cushman, professor16
  1. 1Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA
  2. 2Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, CA, USA
  3. 3Countway Library, Harvard University, Boston, MA, USA
  4. 4Hebrew SeniorLife, Hinda and Arthur Marcus Institute for Aging Research and Harvard Medical School, Boston, MA, USA
  5. 5Johns Hopkins University, Baltimore, MA, USA
  6. 6Department of Ageing and Health, Guy's and St Thomas' NHS Foundation Trust, London, UK
  7. 7Department of Biostatistics and Data Science, Coordinating Center for Clinical Trials, The University of Texas School of Public Health, Houston, TX, USA
  8. 8Diabetes Trials Unit, Radcliffe Department of Medicine, University of Oxford, Oxford, UK
  9. 9The George Institute for Global Health, Sydney, NSW, Australia
  10. 10The School of Population Health, University of New South Wales, Sydney, NSW, Australia
  11. 11Alliance for the Promotion of Preventive Medicine (APPREMED), Mechelen, Belgium
  12. 12Department of Cardiovascular Medicine, Shanghai Key Laboratory of Hypertension, Shanghai Institute of Hypertension, State Key Laboratory of Medical Genomics, National Research Centre for Translational Medicine, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China
  13. 13Biomedical Research Group, Faculty of Medicine, University of Leuven, Leuven, Belgium
  14. 14Michael E DeBakey VA Medical Center and Baylor College of Medicine, Houston, TX, USA
  15. 15Case Western Reserve University, University Hospitals Cleveland Medical Center, Cleveland, OH, USA
  16. 16University of Tennessee Health Science Center, Memphis, TN, USA
  1. Correspondence to: S P Juraschek sjurasch{at}bidmc.harvard.edu
  • Accepted 4 February 2025

Abstract

Objective To determine the effects of intensive blood pressure treatment on orthostatic hypertension.

Design Systematic review and individual participant data meta-analysis.

Data sources MEDLINE, Embase, and Cochrane CENTRAL databases through 13 November 2023.

Inclusion criteria Population: ≥500 adults, age ≥18 years with hypertension or elevated blood pressure; intervention: randomized trials of more intensive antihypertensive drug treatment (lower blood pressure goal or active agent) with duration ≥6 months; control: less intensive antihypertensive drug treatment (higher blood pressure goal or placebo); outcome: measured standing blood pressure.

Main outcomes Orthostatic hypertension, defined as an increase in systolic blood pressure ≥20 mm Hg or diastolic blood pressure ≥10 mm Hg after changing from sitting to standing.

Data synthesis Two investigators independently abstracted articles. Individual participant data from nine trials identified during the systematic review were appended together as a single dataset.

Results Of 31 124 participants with 315 497 standing blood pressure assessments, 9% had orthostatic hypotension (that is, a drop in blood pressure after standing of systolic ≥20 mm Hg or diastolic ≥10 mm Hg), 17% had orthostatic hypertension, and 3.2% had both a rise in systolic blood pressure and standing blood pressure ≥140 mm Hg at baseline. The effects of more intensive treatment were similar across trials with odds ratios for orthostatic hypertension ranging from 0.85 to 1.08 (I2=38.0%). During follow-up, 17% of patients assigned to more intensive treatment had orthostatic hypertension, whereas 19% of those assigned less intensive treatment had orthostatic hypertension. Compared with less intensive treatment, the risk of orthostatic hypertension was lower with more intensive blood pressure treatment (odds ratio 0.93, 95% confidence interval 0.90 to 0.96). Effects were greater among non-black versus black adults (odds ratio 0.86 v 0.97; P for interaction=0.003) and adults without diabetes versus those with diabetes (0.88 v 0.96; P for interaction=0.05) but did not differ by age ≥75 years, sex, baseline seated blood pressure ≥130/≥80 mm Hg, obesity, stage 3 kidney disease, stroke, cardiovascular disease, standing systolic blood pressure ≥140 mm Hg, or pre-randomization orthostatic hypertension (P for interactions ≥0.05).

Conclusions In this pooled cohort of adults with elevated blood pressure or hypertension, orthostatic hypertension was common and more intensive blood pressure treatment modestly reduced the occurrence of orthostatic hypertension. These findings suggest that approaches generally used for seated hypertension may also prevent hypertension on standing.

Study registration Prospero CRD42020153753 (original proposal).

Introduction

Orthostatic hypertension, an elevation in blood pressure after standing, is an emerging risk factor for several adverse health outcomes, including cardiovascular disease, stroke, kidney disease, and cognitive impairment.123 Orthostatic hypertension also seems to be an important predictor of all cause mortality among older adults.4 Although individual cohort studies have observed that orthostatic hypertension disproportionately affects adults with hypertension, the effects of blood pressure treatment on the occurrence of orthostatic hypertension have not been systematically examined.

In a recent, individual participant data meta-analysis of hypertension trials with standing blood pressure assessments, we examined the effect of more intensive blood pressure treatment on orthostatic hypotension.5 However, we did not examine the effect of treatment on orthostatic hypertension, which was also collected during these trials. Current recommendations for the treatment of orthostatic hypertension focus on agents that are not considered first line for seated hypertension (that is, thiazide diuretics, dihydropyridine calcium channel blockers, or angiotensin converting enzyme inhibitors/angiotensin receptor blockers).6 Whether more intensive treatments generally used for seated hypertension might be efficacious for orthostatic hypertension is unknown, but it could have implications for the formulation of treatment strategies to tackle this emerging hypertensive phenotype.

The objectives of this study were to use the individual participant data combined from the nine hypertension trials identified by the systematic review above to determine the prevalence of orthostatic hypertension among adult participants of hypertension treatment trials and the effect of more intensive blood pressure treatment (that is, a lower blood pressure treatment goal or active therapy versus either a higher blood pressure treatment goal or placebo) on orthostatic hypertension and to assess for effect modification by demographic characteristics.

Methods

Search strategy and eligibility criteria

This post hoc study focuses on orthostatic hypertension, but the search strategy of our original systematic review, focused on orthostatic hypotension, was described elsewhere.57 In brief, our review was registered in the PROSPERO registry on 28 April 2020 (CRD42020153753) and initial searches included MEDLINE/PubMed), Embase, and the Cochrane Central Register of Controlled Trials databases without language restrictions. A research librarian (CM) prepared our search strategy, which focused on hypertension, blood pressure treatment, standing blood pressure (particularly orthostatic hypotension), and randomized trials. Duplicate records were removed in EndNote, and two independent investigators (SPJ and JLC) screened abstracts with Covidence, with discrepancies adjudicated by consensus. This search was updated to include citations through 13 November 2023 (supplementary methods 1; supplementary figure A). The search ultimately entailed 1127 unique abstracts and 70 unique trials. Only one of the eligible trials was excluded owing to inability to share individual participant data. Because our original search included all trials with standing blood pressure measurements, we were able to use the outcomes of this search to examine orthostatic hypertension as a post hoc analysis.

The original systematic review was guided by the following PICO (population, intervention, comparison, outcomes) criteria.8 Population: trials of at least 500 adults (ages 18 years or older) with elevated blood pressure or hypertension (based on seated measurements). Intervention: at least six months of randomized antihypertensive drug treatment (blood pressure goal or active agent). Comparison: at least two blood pressure goals (one less than the other) or active therapy versus placebo. Outcome: orthostatic hypotension measured after randomization. Although orthostatic hypotension was the outcome of the original search, all these trials also had the relevant data for calculation of orthostatic hypertension. We excluded trials of pregnant women or children, animal experiments (non-human trials), reviews, observational studies, and studies without direct measures of orthostatic hypotension (for example, based on self-report or claims). We pooled trials together overall and by type—that is, those comparing two treatment goals (a lower versus a higher goal) or placebo controlled trials.

In addition to our systematic review above, we also attempted to contact investigators of trials of antihypertensive drug treatment in adults with elevated blood pressure or hypertension included in a recent meta-analysis focused on cardiovascular disease,9 asking about the availability of standing blood pressure measurements. This process led to the inclusion of one trial not identified through our original search.10 One trial was not able to provide us with data owing to data sharing restrictions.11 All trials identified had both pre-randomization and post-randomization orthostatic blood pressure assessments, which could be used to derive orthostatic hypertension. Risk of bias characterization was updated to reflect orthostatic hypertension as the primary outcome of this systematic review (supplementary table A).12

Treatment assignment

Similarly to our previous work, we chose a priori to examine pooled effects by categories of trial design: trials of blood pressure treatment goal (that is, one goal lower than the other goal) and trials of an active antihypertensive agent versus placebo. More intensive treatment included patients assigned a lower blood pressure treatment goal and those assigned to active antihypertensive treatment, and less intensive treatment included those assigned a higher blood pressure treatment goal and those assigned to placebo.

Orthostatic hypertension

We determined the difference between standing minus seated blood pressure for each trial at all available visits (a visit being a clinical session whereby a participant interacted with a study team and blood pressure was measured). We defined orthostatic hypertension as standing minus seated systolic blood pressure of ≥20 mm Hg or diastolic blood pressure of ≥10 mm Hg, the definition used in SPRINT and our previous work.1314151617 Seated blood pressure varied by trial protocol—for example, based on one measurement or based on the average of two or three measurements (sometimes with the first measurement excluded). Standing blood pressure similarly varied according to trial protocols but often included only a single measurement (see table 1). We defined standing systolic hypertension as a standing systolic blood pressure ≥140 mm Hg. This was incorporated into a recently updated definition of orthostatic hypertension—that is, a change in systolic blood pressure of ≥20 mm Hg and a standing systolic blood pressure of ≥140 mm Hg (the new consensus definition).1819 Orthostatic hypotension was defined as standing minus seated systolic blood pressure of ≤−20 mm Hg or diastolic blood pressure of ≤−10 mm Hg.20 Baseline orthostatic hypertension or standing systolic hypertension was based on the seated and standing blood pressures measured in the visit in closest proximity and before or during the randomization visit.

Table 1

Characteristics of included trials

View this table:

Other covariates

We obtained the following covariate information from each trial: age, sex (women, men), race (black, non-black; this was not universally available), pre-randomization seated and standing systolic and diastolic blood pressure, baseline creatinine or estimated glomerular filtration rate or chronic kidney disease status, body mass index, diabetes status, previous stroke, and history of cardiovascular disease. We defined obesity as body mass index ≥30 and stage 3 chronic kidney disease as estimated glomerular filtration rate <60 mL/min/1.73 m2 on the basis of the 2021 CKD-EPI race-free, creatinine equation21 or self-reported history of kidney disease (SHEP trial only). Differences in the definitions of diabetes, stroke, and cardiovascular disease between studies were described elsewhere.57

Statistical analysis

Pre-randomization and post-randomization visit data from all trials were appended into a single analytic dataset before pooled analyses. Analyses were restricted to blood pressure, body mass index, and estimated glomerular filtration rate measures between the 0.01st and 99.99th centiles of all measurements (baseline and follow-up) from all nine trials to account for biologically implausible outliers (particularly relevant for stratified analyses; see supplementary table B for values corresponding to these thresholds and supplementary table C for values corresponding to 0.1st and 99.9th centiles). We summarized population characteristics via means and proportions overall, by orthostatic hypertension status, by trial type, and according to each study. We used kernel density plots (bandwidth 5) to visually examine the distribution of systolic and diastolic blood pressure in seated and standing positions and the difference between positions (standing minus seated) according to the pre-randomization visit and follow-up visits among participants assigned to a lower blood pressure treatment goal or active therapy and among those assigned to a higher blood pressure treatment goal or placebo across all studies. We compared characteristics between participants with and without a baseline orthostatic hypertension assessment.

We plotted the proportion of orthostatic hypertension detected during study visits grouped according to a series of time intervals: month 0/pre-randomization, after randomization to ≤1 month, >1 to ≤6 months, >6 to ≤12 months, >12 to ≤24 months, >24 to ≤36 months, >36 to ≤48 months, and >48 months. We plotted proportions overall according to assignment (that is, a lower treatment goal/active therapy or a higher treatment goal/placebo) via generalized estimating equations (Poisson family, log link, robust variance estimator, exchangeable correlation matrix) without adjustment. We tabulated the number of measurements and number of individual participants at risk, determining the proportion with orthostatic hypertension at any time during each time period. We also examined the proportions, changes in proportion, and odds of orthostatic hypertension over time, using generalized estimating equations (binomial family, logit link, robust variance estimator, exchangeable correlation matrix) adjusted for study. We used generalized estimating equations to account for repeated measurements within participants as they are able to generate valid variance estimates even when the within group correlation structure is mis-specified. Models included interaction terms to assess for differences at different time points. We also examined the relation between baseline orthostatic hypertension and follow-up orthostatic hypertension via generalized estimating equations (binomial family, logit link, robust variance estimator, exchangeable correlation matrix), using an interaction term with randomized treatment assignment to assess whether this relation differed by treatment.

In addition, we compared the effect of more intensive treatment (that is, a lower treatment goal or active therapy) versus higher treatment goals or placebo on the odds of orthostatic hypertension during follow-up visits, using generalized estimating equations (binomial family, logit link, robust variance estimator, exchangeable correlation matrix). We did these analyses for individual trials and pooled by trial type (that is, the five blood pressure treatment goal trials and the four placebo controlled trials) and overall. We repeated this as a sensitivity analysis using a Poisson family log link.

We repeated models using alternate definitions of orthostatic hypertension (described above) both for 0.01st to 99.99th centiles of blood pressure values and with truncation at 0.1st and 99.9th centiles. We also determined mean systolic or diastolic blood pressure before and after randomization and treated orthostatic change in systolic or diastolic blood pressure as a continuous outcome (these models used a normal family, identity link; by contrast, all models with alternate definitions of orthostatic hypertension as dichotomous outcome variables used a binomial family logit link).

Moreover, we did subgroup analyses examining orthostatic hypertension in the following pre-specified strata: age (≤75 or >75 years), sex (men or women), race (non-black or black), pre-randomization seated blood pressure (systolic blood pressure ≥130 mm Hg or diastolic blood pressure ≥80 mm Hg, no or yes), diabetes (no or yes), previous stroke (no or yes), stage 3 chronic kidney disease (<60 or ≥60 mL/min per 1.73 m2; in SHEP, kidney disease was self-reported), body mass index (<30 or ≥30), history of cardiovascular disease (no or yes), standing systolic blood pressure before randomization (<140 or ≥140 mm Hg), and pre-randomization orthostatic hypertension (no or yes). We used interaction terms to compare effects across strata. We repeated these analyses using the recent consensus definition for orthostatic hypertension,1819 as well as with truncation at the 0.1st and 99.9th centiles of continuous covariates.

We did all analyses for the nine trials as well as by trial type—that is, the five trials that compared two blood pressure treatment goals or the four placebo controlled trials. We used a two stage meta-analysis with a random effects model weighted by the inverse variance in sensitivity analyses and evaluated heterogeneity between studies via the I2 statistic.22 We examined heterogeneity by trial design and overall. Although our a priori intention was to pool studies, this plan was subject to evaluation of heterogeneity (both its magnitude and direction of effect).23 Small study effects were assessed via Egger’s test and funnel plots.24 We used Stata 15.1 for all statistical analyses. We considered a two tailed P value of <0.05 without adjustment for multiple comparisons to be statistically significant. A dummy dataset and analytic codes are available at https://doi.org/10.7910/DVN/RHUF9F; the code is also available in supplementary methods 2.

Patient and public involvement

The original systematic review was initiated without patient or public involvement. However, a patient of SPJ with a history of orthostatic hypertension was a motivation for this work and reviewed this manuscript at the time of the revision request. This patient’s feedback was incorporated into the manuscript.

Results

Population characteristics

Of the 31 124 participants with 315 497 measurements contributing to this individual participant data meta-analysis, the mean age was 67.6 (standard deviation (SD) 10.4) years with 25.1% over the age of 75 years; 47.4% of participants were women (table 2). Differences between participants with and without orthostatic assessments at baseline are found in supplementary table D. Before randomization, the mean seated systolic blood pressure was 152.6 (SD 21.3) mm Hg and the mean seated diastolic blood pressure was 80.9 (11.5) mm Hg. After standing, systolic blood pressure was 152.3 (SD 21.2) mm Hg and diastolic blood pressure was 83.9 (12.1) mm Hg, with a mean postural change in systolic blood pressure of −1.9 (SD 11.4) mm Hg and in diastolic blood pressure of 2.4 (7.4) mm Hg (see supplementary table E for similar results based on all available pre-randomization visits). Before randomization, 8.7% of participants had orthostatic hypotension, 16.7% had orthostatic hypertension, and 1.9% had standing hypertension. The distribution of systolic and diastolic blood pressure became narrower with treatment and shifted to the left (supplementary figures B and C). Little change occurred in the distribution of orthostatic changes before and after treatment, regardless of assignment.

Table 2

Participants’ characteristics

View this table:

Proportion of orthostatic hypertension over time

The proportion of participants with orthostatic hypertension increased initially but then decreased over time in both arms, with a greater reduction in the more intensive treatment group (fig 1). We observed a similar pattern with respect to the relative odds of orthostatic hypertension in that a significant increase occurred within the first month in the lower goal/active therapy group compared with the higher goal/placebo group (supplementary table F). However, these initial increases did not persist over time in the lower goal/active therapy group, whereas the odds in the standard group remained elevated compared with pre-randomization. Having orthostatic hypertension before randomization was associated with having orthostatic hypertension during follow-up (odds ratio 2.53, 95% confidence interval (CI) 2.39 to 2.68), and this relation did not differ by randomized treatment assignment (P for interaction=0.61).

Fig 1
Fig 1

Proportion of participants with orthostatic hypertension by study month. Follow-up visits were grouped together (month 0/before randomization, <1 month, 1-6 months, 6-12 months, 12-24 months, 24-36 months, 36-48 months, and >48 months). Proportions were estimated with generalized estimating equations, using unadjusted Poisson family, log link. This model accounts for correlated within person measurements. Numbers below figure represent measurements contributing to each proportion by time period. Note that some trials had multiple visits before randomization, which contributed to these models and may account for differences in proportion of orthostatic hypertension estimated by this approach, versus descriptive estimate based on single visit in table 2. In addition, number of unique participants at risk in each time period is reported, with percentage with orthostatic hypertension at any time during this time period. These proportions differ from those in figure. BP=blood pressure; CI=confidence interval; OHTN=orthostatic hypertension

Aggregate effects on orthostatic hypertension

Although all trials, except AASK, showed a lower odds of orthostatic hypertension in either the lower goal or active arms, only SHEP and SPRINT had statistically significant results (fig 2). Pooling the five trials comparing blood pressure treatment goals showed that a lower (more intensive) treatment goal was associated with lower odds of orthostatic hypertension (odds ratio 0.95, 95% CI 0.92 to 0.99). Similarly, pooling together the placebo controlled trials showed that active therapy lowered the odds of orthostatic hypertension (odds ratio 0.87, 95% CI 0.83 to 0.93). We found moderate heterogeneity across the nine trials (I2=38.0%). When we pooled the nine trials together, a lower goal or active therapy reduced the odds of orthostatic hypertension compared with a higher goal or placebo (odds ratio 0.93, 95% CI 0.90 to 0.96). A sensitivity analysis using Poisson regression did not meaningfully change our findings (prevalence ratio 0.94, 95% CI 0.91 to 0.97). In sensitivity analyses using a two stage analysis, results were similar, both overall and by trial design (supplementary figures D-F).

Fig 2
Fig 2

Effects of blood pressure (BP) treatment (either lower blood pressure treatment goal or active therapy versus higher blood pressure treatment goal or placebo) on occurrence of orthostatic hypertension at visit level, using generalized estimating equations to account for clustering by participant. Pooled effects are organized according to five blood pressure treatment goal trials and four placebo controlled trials and overall. Size of each point estimate and pooled effect is weighted by number of follow-up visits with orthostatic hypertension assessments. I2=38.0% (determined on basis of two stage meta-analysis, used to assess trial heterogeneity). CI=confidence interval

We examined alternative definitions of orthostatic hypertension. With nearly all definitions examined, a lower (more intensive) blood pressure treatment goal or active therapy was associated with a lower odds ratio of orthostatic hypertension compared with a higher blood pressure treatment goal or placebo (table 3). We also examined the effect of treatment on orthostatic change as a continuous outcome variable (supplementary table G). Whereas trials of blood pressure treatment goal tended to increase orthostatic change in systolic blood pressure (that is, a trend toward an increase in systolic blood pressure with standing), placebo controlled trials tended to decrease orthostatic change in systolic or diastolic blood pressure. We also repeated the principal analyses using the recent consensus definition and the systolic change alone with similar results (see supplementary figures G-I). Finally, we examined alternate definitions of orthostatic hypertension with truncated centile ranges with similar results (supplementary table H).

Table 3

Effect of more intensive treatment on orthostatic hypertension, alternate definitions

View this table:

Stratified analyses

More intensive treatment (that is, a lower blood pressure treatment goal or active therapy) was associated with a lower odds of orthostatic hypertension than a higher blood pressure treatment goal or placebo in strata of age, sex, seated systolic blood pressure of ≥130 mm Hg or diastolic blood pressure of ≥80 mm Hg, estimated glomerular filtration rate <60 mL/min/1.73 m2, body mass index, history of cardiovascular disease, pre-randomization standing systolic blood pressure of ≥140 mm Hg, and pre-randomization orthostatic hypertension (table 4). We found evidence for greater reduction in the occurrence of orthostatic hypertension among non-black participants and participants without diabetes. Results were similar when we defined orthostatic hypertension by using the consensus definition (supplementary table I) and with truncated centiles (supplementary table J).

Table 4

Effect of more intensive treatment on orthostatic hypertension, stratified by pre-specified subgroups (all 9 trials)

View this table:

Discussion

In this meta-analysis of individual participant data from 31 124 adults with elevated blood pressure and hypertension, more intensive treatment (that is, a lower blood pressure treatment goal or active therapy) modestly reduced the occurrence of orthostatic hypertension on the basis of measurements from 315 497 visits. This effect was consistent regardless of trial type or definition of orthostatic hypertension. Moreover, these effects were generally consistent across demographic characteristics and medical comorbidities.

Comparison with other studies

Blood pressure is highly regulated by the autonomic nervous system in healthy adults such that blood pressure remains relatively constant across body positions. Whereas substantial focus has been directed toward falls in blood pressure on standing (that is, orthostatic hypotension), little attention has been given to orthostatic hypertension.2526 Recent epidemiological evidence has identified increases in blood pressure on standing as potentially pathologic, linking orthostatic hypertension with a range of adverse events, including cardiovascular disease, stroke, kidney disease, and cognitive impairment.123 These long term associations with adverse health outcomes contribute to a growing belief that orthostatic hypertension may represent a form of unrecognized or masked hypertension that may require monitoring of adults in the standing position and adjusting drug treatment accordingly.27

This study confirms that orthostatic hypertension is common among adults with hypertension, but it also shows that more intensive blood pressure treatment might attenuate orthostatic hypertension over time. This is an important observation. Our previous work showed that more intensive treatment caused a net increase in the difference in blood pressure in response to standing.5 However, this study suggests that this effect may be short term and dissipate with chronic treatment. This observation, if replicated, may be important for treating clinicians, who might be dissuaded from treating hypertension because of short term orthostatic hypertension. Physiologic mechanisms for this observation are unclear. The short term increase may be secondary to autonomic over-response to valsalva, cardioacceleration after leg muscle contraction, or mobilization of excess lower extremity fluid.2829 In the long term, we speculate that the resolution of orthostatic hypertension may be related to healthy remodeling of the vasculature with tighter blood pressure control.30 Further study of mechanisms should be evaluated in future work, particularly with repeat standing measurements. This may also be related to measurement error and additionally to more controlled blood pressure in general, such that blood pressure and fluctuation in blood pressure measurement is also lower,3132 reducing risk for orthostatic hypertension.

Our study used a definition of orthostatic hypertension that mirrored the one for orthostatic hypotension, used in previous work.1314151617 However, discussion is ongoing as to how orthostatic hypertension should be defined.25 Unlike with orthostatic hypotension, which focuses on changes in blood pressure alone, recent guidelines have advocated for a definition that includes both an increase in systolic blood pressure on standing and an elevated standing threshold (systolic blood pressure ≥140 mm Hg).18 This definition was proposed for a general population, not necessarily a hypertensive population. When we used this definition, the prevalence of orthostatic hypertension was substantially lower in our population. Nevertheless, the effects of treatment were even more pronounced. This is due in part to our observation that antihypertensive agents, particularly the longer acting agents used in many of the trials in our study, lower blood pressure in all body positions.533 As a result, standing hypertension would also be reduced with more intensive treatment. From the perspective of studying mechanisms of injury related to orthostatic hypertension, we caution against the use of this joint definition as it may make identifying and evaluating treatment response to the rise in blood pressure, which itself may be pathologic and occur below the 140 mm Hg threshold, more difficult.17 Some authors have also questioned whether diastolic blood pressure should be included in definitions of orthostatic hypertension, as diastolic blood pressure usually increases with standing.34 However, given that systolic and diastolic blood pressure are known to be correlated in both seated and standing positions, their change would be expected to correlate as well, and thus some patients with a rise in diastolic blood pressure would also have a rise in systolic blood pressure. Whether thresholds of change in systolic blood pressure or diastolic blood pressure are optimal for identifying risk with cardiovascular disease should be the focus of subsequent work. Nevertheless, the effects of treatment on orthostatic increases in systolic and diastolic blood pressure as defined in this study, using thresholds that mirrored those for orthostatic hypotension, were quite consistent.

We did not identify compelling evidence that the effects of treatment differed by demographic or medical characteristics. Although a strong interaction between black and non-black populations was apparent, this information was not uniformly collected by trials outside of the US, reducing our sample for this analysis. We also observed that effects were attenuated among adults with diabetes. Mechanisms are beyond the scope of this study, but we speculate that this lack of effects among black adults and those with diabetes may reflect known challenges in achieving blood pressure control. Additional research should probe these associations further.

Limitations and strengths of study

Our study has limitations. Firstly, we identified only nine trials, which differed with respect to their interventions, frequency of follow-up, duration, blood pressure measurement procedures, and study populations. These differences might have influenced our results. Despite these differences, our findings were relatively consistent and our sample was sufficiently large that additional trials are not likely to alter our pooled observation. Secondly, generalizability to clinical practice may be limited owing to the strict entry criteria used by these trials, differences in prescribing regimens that might not reflect real world drug choices, and careful monitoring and drug titration protocols that may affect titration patterns. Thirdly, subgroup analyses relied on covariate definitions that were based on self-report and at times differed in definition across studies. Any resulting misclassification could weaken contrasts across subgroups, reducing our ability to detect differences. Moreover, we pre-specified our subgroups. Whether associations might differ across different categories (for example, younger age) should be examined in dedicated studies. Fourthly, we did not examine antihypertensive drug class in this study, which should be a focus of subsequent work. Fifthly, assessments of orthostatic hypertension were based on seated-to-standing protocols, which may not be interchangeable with supine-to-standing maneuvers.293335 In the case of orthostatic hypotension, some authors have proposed modified thresholds for identifying adults with orthostatic hypotension (that is, a drop in systolic blood pressure of 15 mm Hg or diastolic blood pressure of 7 mm Hg).36 In our own work, we have observed a net increase in blood pressure with standing from the seated position.29 Thus, whether a seated-to-standing protocol should have a higher or lower threshold compared with supine-to-standing to establish orthostatic hypertension remains unclear. Moreover, aside from SYST-EUR, none of the major outcome trials examined in our study measured supine blood pressure.33 The manner by which starting position might underestimate or overestimate orthostatic hypertension should be the focus of future work. Sixthly, although both lower treatment goals and active therapy lowered the occurrence of orthostatic hypertension, the effect of active therapy was greater in magnitude. Whether different goals might alter the observed effect is beyond the scope of this study. Seventhly, temporal effects of treatment on orthostatic hypertension should be interpreted cautiously, as different populations contributed to visits at different time points. We attempted to include all available data to preserve the trials’ randomized contrasts. However, as some trials (for example, ACCORD) assessed orthostatic blood pressure after starting, some of these participants did not have a pre-randomization assessment, which could influence estimates of orthostatic hypertension at baseline. This would not affect the pooled contrast overall but could affect the proportion with orthostatic hypertension over time. Finally, our analysis did not examine the effects of treatment on clinical events among adults with orthostatic hypertension. This has been questioned in previous work and represents an important focus for subsequent research.14

This study has notable strengths. Firstly, this systematic review and meta-analysis represents one of the largest data collections of orthostatic hypertension in the context of drug treatment for hypertension. Secondly, doing an individual participant meta-analysis allowed for greater harmonization of data and examination of under-represented subgroups that were not feasible within individual trials. Thirdly, whereas data on orthostatic hypertension have been presented from various trials with respect to outcomes, to our knowledge this is the only patient level meta-analysis of the risk of orthostatic hypertension in treated hypertensive patients. Finally, the associations between treatment and orthostatic hypertension across trials were similar, suggesting that the effects of more intensive hypertension treatment on orthostatic hypertension are quite reproducible.

Implications

Our study potentially has clinical implications. Orthostatic hypertension has received increasing attention as a novel and distinct presentation of hypertension. This carries the suggestion that distinct pharmacologic strategies are needed for its treatment. Some authors have suggested that β blockers may be more effective for treating orthostatic hypertension by blunting the β adrenergic response to standing, and evidence also shows that peripheral α blockers may be effective.6 However, neither of these classes is preferred for initial treatment of hypertension on the basis of the experience from hypertension outcome trials.37 Nevertheless, if either is superior to other classes in reducing orthostatic hypertension, it might be of value as add-on therapy in the presence of residual orthostatic hypertension. Although more work is needed to evaluate specific drug classes with respect to orthostatic hypertension, our data provide reassurance that focusing on seated blood pressure control and treating seated hypertension among adults with hypertension can modestly reduce orthostatic hypertension. At this time, no trials are assessing whether treating standing blood pressure levels to some goal provides additional benefit to the traditional seated approach.

Conclusions

In this large, individual participant data meta-analysis of blood pressure treatment trials, more intensive blood pressure treatment, especially with active treatment (versus placebo), modestly reduced the occurrence of orthostatic hypertension regardless of its definition or baseline demographic and medical characteristics. Future research should examine orthostatic hypertension in relation to clinical outcomes as well as whether specific classes of antihypertensive drugs or lower treatment goals might better prevent orthostatic hypertension and its sequelae.

What is already known on this topic

  • Orthostatic hypertension (an extreme increase in blood pressure after standing) is a pathologic form of higher standing blood pressure, predicting adverse health outcomes in observational studies

  • Current recommendations to treat orthostatic hypertension are based on a few small trials of agents not considered first line for hypertension treatment

What this study adds

  • Using data from nine randomized trials of >30 000 participants, this individual level meta-analysis shows that more intensive blood pressure treatment reduces the occurrence of orthostatic hypertension

  • Many of the included trials used first line antihypertensive agents, suggesting that common approaches for seated hypertension may also be used to treat orthostatic hypertension

  • Although orthostatic hypertension is common among adults with hypertension, it may be treated using standard approaches recommended for seated hypertension

Ethics statements

Ethical approval

The Institutional Review Board at Beth Israel Deaconess Medical Center determined this research to be human subjects exempt research.

Data availability statement

The data associated with this paper were used with institutional agreements between the NHLBI BioLINCC repository of institutions that conducted the original trials. The data use agreements do not permit public sharing of trial data. However, most of the data used in these analyses may be obtained with data use agreements involving the NHLBI BioLINCC or by request to the corresponding authors of the original trials. A dummy dataset and corresponding analytic codes are available at https://doi.org/10.7910/DVN/RHUF9F.

Acknowledgments

We thank the BioLINCC repository for data from the following studies: ACCORD BP, SPRINT, and SHEP. We thank a patient of SPJ for her review and contributions to our manuscript.

Footnotes

  • Contributors: SPJ contributed to oversight, conceptualization, resources, data extraction, data analysis, data interpretation, primary draft, and critical review of the manuscript. J-RH contributed to data extraction and critical review of the manuscript. JLC contributed to data extraction and critical review of the manuscript. CM contributed to search strategy, support, and critical review of the manuscript. LAL contributed to data interpretation and critical review of the manuscript. LJA, NSB, BRD, RRH, ERM, RP, JAS, AAT, and JTW contributed to primary data acquisition, data interpretation, and critical review of the manuscript. KJM contributed to conceptualization, data interpretation, and critical review of the manuscript. WCC contributed to project oversight, conceptualization, data interpretation, primary draft, and critical review of the manuscript. SPJ had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. SPJ is the guarantor. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.

  • Funding: SPJ was supported by National Institutes of Health/National Heart, Lung, and Blood Institute (NIH/NHLBI) K23HL135273 and R01HL153191. RRH is an Emeritus UK National Institute for Health Research senior investigator. KM was supported by K24AG065525. The funders had no input on the design and conduct of the study; the collection, management, analysis, and interpretation of the data; the preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.

  • Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/disclosure-of-interest/ and declare: support from NIH/NHLBI; WCC is a co-investigator for a hypertension trial funded by Recor and is the principal investigator for two hypertension trials with George Medicine; no other relationships or activities that could appear to have influenced the submitted work.

  • Transparency: The lead author (the manuscript’s guarantor) affirms that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

  • Dissemination to participants and related patient and public communities: An abstract of this research was presented at the AHA Hypertension Conference in September 2023 and will be publicly available through PubMed Central for patients and clinicians throughout the world.

  • Provenance and peer review: Not commissioned; externally peer reviewed.

http://creativecommons.org/licenses/by/4.0/

This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http://creativecommons.org/licenses/by/4.0/.

References

  1. ↵
    1. Pasdar Z,
    2. De Paola L,
    3. Carter B,
    4. Pana TA,
    5. Potter JF,
    6. Myint PK
    . Orthostatic hypertension and major adverse events: a systematic review and meta-analysis. Eur J Prev Cardiol2023;30:1028-38. doi:10.1093/eurjpc/zwad158 pmid:37202364
    OpenUrlCrossRefPubMed
  2. ↵
    1. Rouabhi M,
    2. Durieux J,
    3. Al-Kindi S,
    4. Cohen JB,
    5. Townsend RR,
    6. Rahman M
    . Orthostatic Hypertension and Hypotension and Outcomes in CKD: The CRIC (Chronic Renal Insufficiency Cohort) Study. Kidney Med2021;3:206-215.e1. doi:10.1016/j.xkme.2020.10.012 pmid:33851116
    OpenUrlCrossRefPubMed
  3. ↵
    1. Strumia M,
    2. Vidal JS,
    3. Cestac P,
    4. Sallerin B,
    5. Hanon O,
    6. Rouch L,
    7. S.AGES Investigators
    . Orthostatic hypotension and orthostatic hypertension are both associated with lower cognitive function: The S.AGES cohort. J Am Geriatr Soc2023;71:3721-30. doi:10.1111/jgs.18571 pmid:37655948
    OpenUrlCrossRefPubMed
  4. ↵
    1. Kostis WJ,
    2. Sargsyan D,
    3. Mekkaoui C,
    4. et al
    . Association of orthostatic hypertension with mortality in the Systolic Hypertension in the Elderly Program. J Hum Hypertens2019;33:735-40. doi:10.1038/s41371-019-0180-4 pmid:30804464
    OpenUrlCrossRefPubMed
  5. ↵
    1. Juraschek SP,
    2. Hu JR,
    3. Cluett JL,
    4. et al
    . Effects of Intensive Blood Pressure Treatment on Orthostatic Hypotension : A Systematic Review and Individual Participant-based Meta-analysis. Ann Intern Med2021;174:58-68. doi:10.7326/M20-4298 pmid:32909814
    OpenUrlCrossRefPubMed
  6. ↵
    1. Hoshide S,
    2. Parati G,
    3. Matsui Y,
    4. Shibazaki S,
    5. Eguchi K,
    6. Kario K
    . Orthostatic hypertension: home blood pressure monitoring for detection and assessment of treatment with doxazosin. Hypertens Res2012;35:100-6. doi:10.1038/hr.2011.156 pmid:21918522
    OpenUrlCrossRefPubMed
  7. ↵
    1. Juraschek SP,
    2. Hu JR,
    3. Cluett JL,
    4. et al
    . Orthostatic Hypotension, Hypertension Treatment, and Cardiovascular Disease: An Individual Participant Meta-Analysis. JAMA2023;330:1459-71. doi:10.1001/jama.2023.18497 pmid:37847274
    OpenUrlCrossRefPubMed
  8. ↵
    Robinson KA, Saldanha IJ, Mckoy NA. Frameworks for Determining Research Gaps During Systematic Reviews. 2011. https://www.ncbi.nlm.nih.gov/books/NBK62478/.
  9. ↵
    1. Brunström M,
    2. Carlberg B
    . Association of Blood Pressure Lowering With Mortality and Cardiovascular Disease Across Blood Pressure Levels: A Systematic Review and Meta-analysis. JAMA Intern Med2018;178:28-36. doi:10.1001/jamainternmed.2017.6015 pmid:29131895
    OpenUrlCrossRefPubMed
  10. ↵
    1. UK Prospective Diabetes Study Group
    . Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. BMJ1998;317:703-13. doi:10.1136/bmj.317.7160.703 pmid:9732337
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Brenner BM,
    2. Cooper ME,
    3. de Zeeuw D,
    4. et al.,
    5. RENAAL Study Investigators
    . Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and nephropathy. N Engl J Med2001;345:861-9. doi:10.1056/NEJMoa011161 pmid:11565518
    OpenUrlCrossRefPubMedWeb of Science
  12. ↵
    1. Higgins JPT,
    2. Altman DG,
    3. Gøtzsche PC,
    4. et al.,
    5. Cochrane Bias Methods Group,
    6. Cochrane Statistical Methods Group
    . The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ2011;343:d5928. doi:10.1136/bmj.d5928 pmid:22008217
    OpenUrlFREE Full Text
  13. ↵
    1. Townsend RR,
    2. Chang TI,
    3. Cohen DL,
    4. et al.,
    5. SPRINT Study Research Group
    . Orthostatic changes in systolic blood pressure among SPRINT participants at baseline. J Am Soc Hypertens2016;10:847-56. doi:10.1016/j.jash.2016.08.005 pmid:27665708
    OpenUrlCrossRefPubMed
  14. ↵
    1. Rahman M,
    2. Pradhan N,
    3. Chen Z,
    4. et al
    . Orthostatic Hypertension and Intensive Blood Pressure Control; Post-Hoc Analyses of SPRINT. Hypertension2021;77:49-58. doi:10.1161/HYPERTENSIONAHA.120.15887 pmid:33161773
    OpenUrlCrossRefPubMed
  15. ↵
    1. Ghazi L,
    2. Cohen JB,
    3. Townsend RR,
    4. et al
    . Orthostatic hypotension, orthostatic hypertension, and ambulatory blood pressure in patients with chronic kidney disease in CRIC. J Hypertens2024;42:329-36. doi:10.1097/HJH.0000000000003604 pmid:37889527
    OpenUrlCrossRefPubMed
  16. ↵
    1. Petriceks AH,
    2. Appel LJ,
    3. Miller ER 3rd.,
    4. et al
    . Associations of Hypertension and Orthostatic Hypotension With Subclinical Cardiovascular Disease. J Gerontol A Biol Sci Med Sci2024;79:glae234. doi:10.1093/gerona/glae234 pmid:39292998
    OpenUrlCrossRefPubMed
  17. ↵
    1. Dooley SW,
    2. Larbi Kwapong F,
    3. Col H,
    4. et al
    . Orthostatic and Standing Hypertension and Risk of Cardiovascular Disease. Hypertension2025;82:382-92. doi:10.1161/HYPERTENSIONAHA.124.23409 pmid:39633562
    OpenUrlCrossRefPubMed
  18. ↵
    1. Jordan J,
    2. Biaggioni I,
    3. Grassi G,
    4. Fedorowski A,
    5. Kario K
    . When Blood Pressure Increases with Standing: Consensus Definition for Diagnosing Orthostatic Hypertension. Blood Press2023;32:2161871. doi:10.1080/08037051.2022.2161871 pmid:36597193
    OpenUrlCrossRefPubMed
  19. ↵
    1. Jordan J,
    2. Biaggioni I,
    3. Kotsis V,
    4. et al
    . Consensus statement on the definition of orthostatic hypertension endorsed by the American Autonomic Society and the Japanese Society of Hypertension. Clin Auton Res2023;33:69-73. doi:10.1007/s10286-022-00897-8 pmid:36173501
    OpenUrlCrossRefPubMed
  20. ↵
    1. Freeman R,
    2. Wieling W,
    3. Axelrod FB,
    4. et al
    . Consensus statement on the definition of orthostatic hypotension, neurally mediated syncope and the postural tachycardia syndrome. Clin Auton Res2011;21:69-72. doi:10.1007/s10286-011-0119-5 pmid:21431947
    OpenUrlCrossRefPubMed
  21. ↵
    1. Levey AS,
    2. Stevens LA,
    3. Schmid CH,
    4. et al.,
    5. CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration)
    . A new equation to estimate glomerular filtration rate. Ann Intern Med2009;150:604-12. doi:10.7326/0003-4819-150-9-200905050-00006 pmid:19414839
    OpenUrlCrossRefPubMedWeb of Science
  22. ↵
    1. Higgins JPT,
    2. Thompson SG
    . Quantifying heterogeneity in a meta-analysis. Stat Med2002;21:1539-58. doi:10.1002/sim.1186 pmid:12111919
    OpenUrlCrossRefPubMedWeb of Science
  23. ↵
    Deeks JJ, Higgins JPT, Altman DG, McKenzie JE, Veroniki AA; the Cochrane Statistical Methods Group. Cochrane Handbook for Systematic Reviews of Intervention. Chapter 10: Analysing data and undertaking meta-analyses. 2024. https://training.cochrane.org/handbook/current/chapter-10.
  24. ↵
    1. Egger M,
    2. Davey Smith G,
    3. Schneider M,
    4. Minder C
    . Bias in meta-analysis detected by a simple, graphical test. BMJ1997;315:629-34. doi:10.1136/bmj.315.7109.629 pmid:9310563
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Palatini P
    . Orthostatic Hypertension: A Newcomer Among the Hypertension Phenotypes. Hypertension2023;80:1993-2002. doi:10.1161/HYPERTENSIONAHA.123.21537 pmid:37417253
    OpenUrlCrossRefPubMed
  26. ↵
    1. Jordan J,
    2. Ricci F,
    3. Hoffmann F,
    4. Hamrefors V,
    5. Fedorowski A
    . Orthostatic Hypertension: Critical Appraisal of an Overlooked Condition. Hypertension2020;75:1151-8. doi:10.1161/HYPERTENSIONAHA.120.14340 pmid:32223382
    OpenUrlCrossRefPubMed
  27. ↵
    1. Huang JF,
    2. Zhang DY,
    3. An DW,
    4. et al.,
    5. ANTI-MASK Investigators
    . Efficacy of antihypertensive treatment for target organ protection in patients with masked hypertension (ANTI-MASK): a multicentre, double-blind, placebo-controlled trial. EClinicalMedicine2024;74:102736. doi:10.1016/j.eclinm.2024.102736 pmid:39091669
    OpenUrlCrossRefPubMed
  28. ↵
    1. Nobrega ACL,
    2. O’Leary D,
    3. Silva BM,
    4. Marongiu E,
    5. Piepoli MF,
    6. Crisafulli A
    . Neural regulation of cardiovascular response to exercise: role of central command and peripheral afferents. Biomed Res Int2014;2014:478965. doi:10.1155/2014/478965 pmid:24818143
    OpenUrlCrossRefPubMed
  29. ↵
    1. Juraschek SP,
    2. Appel LJM,
    3. M Mitchell C,
    4. et al
    . Comparison of supine and seated orthostatic hypotension assessments and their association with falls and orthostatic symptoms. J Am Geriatr Soc2022;70:2310-9. doi:10.1111/jgs.17804 pmid:35451096
    OpenUrlCrossRefPubMed
  30. ↵
    1. Schiffrin EL
    . Vascular remodeling in hypertension: mechanisms and treatment. Hypertension2012;59:367-74. doi:10.1161/HYPERTENSIONAHA.111.187021 pmid:22203749
    OpenUrlFREE Full Text
  31. ↵
    1. Raber I,
    2. Belanger MJ,
    3. Farahmand R,
    4. et al
    . Orthostatic Hypotension in Hypertensive Adults: Harry Goldblatt Award for Early Career Investigators 2021. Hypertension2022;79:2388-96. doi:10.1161/HYPERTENSIONAHA.122.18557 pmid:35924561
    OpenUrlCrossRefPubMed
  32. ↵
    1. Webb AJS,
    2. Fischer U,
    3. Mehta Z,
    4. Rothwell PM
    . Effects of antihypertensive-drug class on interindividual variation in blood pressure and risk of stroke: a systematic review and meta-analysis. Lancet2010;375:906-15. doi:10.1016/S0140-6736(10)60235-8 pmid:20226989
    OpenUrlCrossRefPubMedWeb of Science
  33. ↵
    1. Grobman B,
    2. Turkson-Ocran RN,
    3. Staessen JA,
    4. et al
    . Body Position and Orthostatic Hypotension in Hypertensive Adults: Results from the Syst-Eur Trial. Hypertension2023;80:820-7. doi:10.1161/HYPERTENSIONAHA.122.20602 pmid:36744469
    OpenUrlCrossRefPubMed
  34. ↵
    1. Johansson M,
    2. Fedorowski A,
    3. Jordan J,
    4. Engström G,
    5. Nilsson PM,
    6. Hamrefors V
    . Orthostatic blood pressure adaptations, aortic stiffness, and central hemodynamics in the general population: insights from the Malmö Offspring Study (MOS). Clin Auton Res2023;33:29-40. doi:10.1007/s10286-022-00911-z pmid:36473959
    OpenUrlCrossRefPubMed
  35. ↵
    1. Breeuwsma AC,
    2. Hartog LC,
    3. Kamper AM,
    4. et al
    . Standing orthostatic blood pressure measurements cannot be replaced by sitting measurements. Hypertens Res2017;40:765-70. doi:10.1038/hr.2017.39 pmid:28298654
    OpenUrlCrossRefPubMed
  36. ↵
    1. Shaw BH,
    2. Garland EM,
    3. Black BK,
    4. et al
    . Optimal diagnostic thresholds for diagnosis of orthostatic hypotension with a ‘sit-to-stand test’. J Hypertens2017;35:1019-25. doi:10.1097/HJH.0000000000001265 pmid:28129252
    OpenUrlCrossRefPubMed
  37. ↵
    1. Whelton PK,
    2. Carey RM,
    3. Aronow WS,
    4. 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. Hypertension2018;71:1269-324. doi:10.1161/HYP.0000000000000066 pmid:29133354
    OpenUrlFREE Full Text
Back to top