Effects of Intensive versus Standard Ambulatory Blood Pressure Control on Cerebrovascular Outcomes

Abstract

Background.

Subcortical microvascular disease represented by brain white matter hyperintensity (WMH) on magnetic resonance imaging (MRI), is associated with functional decline in older people with hypertension. The effects of 2 levels of 24-hour average systolic blood pressure (BP) on mobility, white matter disease progression, and cognitive function over 3 years were studied.

Methods.

This trial was a prospective, randomized, blinded endpoints study in patients 75 years or older with systolic hypertension and MRI evidence of WMH lesions. Patients were randomized to a 24-hour mean systolic BP of less than or equal to 130 mmHg (intensive-treatment) versus less than or equal to 145 mmHg (standard-treatment) using antihypertensive therapies. Primary study outcomes were changes in mobility (gait speed) and accrual of WMH volume following 3 years. Changes in cognitive function (executive processing) and adverse events were also evaluated.

Results:

In 199 randomized patients, mean age of the cohort was 80.5 years and 54% were women; the average 24-hour systolic BP was 149 mmHg. Goal BPs were achieved after a median treatment period of 3-4 months; at that time, the mean 24-hour systolic BP was 127.7 mmHg in the intensive-treatment group and 144.0 mmHg in the standard-treatment group for an average difference of 16.3 mmHg. Changes in gait speed were not different between treatment groups (0.40 ± 2.0 versus 0.42 ± 2.7 in the intensive-treatment and standard-treatment groups, respectively, p= 0.91) while changes from baseline in WMH volumes were smaller (0.29%) in the intensive-treatment group compared to the standard-treatment group (0.48%; p = 0.03). Cognitive outcomes were also not different between the treatment groups. Major adverse cardiovascular events were higher in the standard-treatment group compared to the intensive-treatment group (17 versus 4 patients, p = 0.01). Falls, with or without injury, and syncope were comparable in the treatment groups.

Conclusions.

Intensive lowering of ambulatory BP reduction in older patients with hypertension did not result in differences in mobility outcomes but it was associated with reduction in accrual of subcortical white matter disease. Over periods longer than 3 years, reduction in accumulation of white matter disease may be a factor in conserving function.

Introduction

Subcortical microvascular disease is associated with hypertension in older persons.^1,2 Microvascular disease with accrual of white matter hyperintensities³ is accompanied by deterioration in cognitive function,⁴ stroke,⁵ and limited mobility ⁶. For example, when white matter lesion burden increases over time, cognitive function worsens ⁷ and there is also an increase in the risk of falls.⁸

Approximately two-thirds of the population over the age of 75 years have evidence of white matter hyperintensities on brain magnetic resonance imaging.^9,10 Older patients with hypertension and white matter hyperintensity lesions may progress to functional disability.^10,11 Previously, we demonstrated that ambulatory blood pressure was more closely associated with longitudinal progression of white matter hyperintensity lesions than clinic blood pressure. ¹⁰ This finding led us to conduct the INtensive versus Standard Ambulatory Blood Pressure Lowering to Prevent Functional DeclINe In The ElderlY (INFINITY) trial^{12, 13} to evaluate the effects of 2 levels of 24-hour systolic ambulatory systolic blood pressure on the progression of brain white matter hyperintensity lesions and functional effects in patients with systolic hypertension, 75 years and older.

Methods

Data Sharing Plan: Individual patient data will be shared. A research proposal must be approved by an independent review panel and the study sponsor, and researchers must sign a data sharing agreement. Anonymized individual patient level data will be provided in a secure access environment upon approval of a research proposal and signature of the data sharing agreement. Data can be requested 24 months after the primary publication and will be available for a period of 3 years for requests. Proposals for access should be sent to wolfson@uchc.edu and wwhite@uchc.edu. The INFINITY overall results will be posted to the NIH clinical trials website https://clinicaltrials.gov/ct2/show/NCT01650402 by Q4/2019.

Study Design

INFINITY was a prospective, randomized, open-label, and blinded endpoints trial. Details of the design and baseline characteristics of the study population have been published previously ^12,13. The authors designed and participated in the coordination and oversight of the trial. An independent data safety committee monitored the study and had access to unblinded data; members of the data safety committee were chosen by the sponsor and made their recommendations to the sponsor. The authors of this publication, who vouch for the accuracy and completeness of the reported data, drafted the manuscript and had full access to the final study data. The National Institute of Aging and ethical committees at the University of Connecticut School of Medicine, Farmington and collaborative institutions approved the study design and all participants provided written informed consent.

Study Patients

Patients were eligible for enrollment if they were 75 years or older, had a diagnosis of systolic hypertension based on both clinic and ambulatory blood pressure measurements, and visible white matter hyperintensity lesions on screening magnetic brain imaging (MRI) (typically ≥ 0.5% or greater lesion volume corrected for intracranial cavity volume size). Further criteria for inclusion were clinic systolic blood pressures of 150 to 170 mmHg if taking 1 or more antihypertensive drugs or > 170 mmHg on 0 to 1 antihypertensive drug, at screening. Exclusion criteria included uncontrolled diabetes mellitus, a history of stroke, dementia or clinically limited gait, unstable cardiac disorders within 90 days before screening, severe kidney dysfunction (estimated creatinine clearance < 25 ml/min using the Cockcroft and Gault formula based on ideal body weight), medical conditions that limited survival to less than 3 years and contraindications to magnetic resonance imaging (see Supplemental Table 1 for complete inclusion and exclusion criteria).

Study Treatment and Procedures

Patients were randomly assigned to intensive-treatment (24-hour systolic blood pressure ≤ 130 mmHg) or to standard-treatment (24-hour systolic blood pressure of ≤145 mmHg). The goal was to provide a 15 mm Hg difference in ambulatory systolic blood pressure between the treatment groups. The blood pressure level chosen for the standard group was based on the results of the active treatment group in the Hypertension in the Very Elderly Trial (HYVET). ¹⁴ Patients randomized to intensive-treatment received the ACE inhibitor lisinopril and the dihydropyridine calcium channel blocker amlodipine at the time of randomization. If there was intolerance to an ACE inhibitor, the angiotensin receptor blocker losartan was administered. During a 4-month titration period with monthly visits, doses of these agents were increased and if needed, a thiazide diuretic was added to the regimen. If 3 drugs did not result in the goal systolic blood pressure, other drugs could be added as fourth and fifth agents including beta-blockers, aldosterone antagonists, or alpha-1 adrenergic blockers. Patients randomized to standard-treatment were started on the ACE inhibitor lisinopril or losartan depending on clinical history. If systolic blood pressure remained above goal, amlodipine was added as a second agent; the therapeutic pathway was otherwise the same as for intensive-treatment. For those patients initiated on lisinopril whose systolic blood pressure fell below 140 mm Hg, the dose of the initial agent was reduced by 50 percent. If systolic blood pressure remained below goal, antihypertensive therapy was discontinued. Outpatient visits were performed during a 2-week screening period, at the time of randomization and at monthly visits until it was estimated by clinic monitoring that blood pressure goals may have been met. In both treatment groups, confirmation of achieving the goal of randomized therapy was always made using 24-hour ambulatory blood pressure monitoring. After achieving the ambulatory blood pressure goal, visits occurred every 3 months during subsequent study years. Assessments for study outcomes were made at baseline, 18 months, and 36 months. Additional safety laboratory tests were performed at 12 and 24 months.

Study Outcomes

The objectives of INFINITY were to compare the changes in mobility parameters (primarily gait times) by study group. We hypothesized that the intervention group would have less accrual of white matter hyperintensity lesions (calculated from magnetic resonance images) and this would be reflected in the mobility changes. Secondary end points included changes from baseline in cognitive function (executive function, processing speed, and memory). The consistency of effects on study outcomes was explored in a subgroup of those patients whose blood pressure remained within their assigned category during the trial. Safety end points included mortality, major non-fatal cardiovascular events (non-fatal myocardial infarction, non-fatal stroke, hospitalizations for arrhythmia, heart failure, unstable angina, coronary revascularization, transient ischemic attack ascertained by hospital source records), and events of special interest potentially related to hypotension, including syncope and falls.

Blood pressure measurements:

Clinic blood pressures were taken by a physician in triplicate, 1 to 2 minutes apart in the non-dominant arm and averaged using a validated digital device (TM −247, Suntech Medical Instruments, Morrisville, NC) in the morning before taking antihypertensive medication as previously described. ¹³ The 24-hour ambulatory blood pressure was conducted using a validated ambulatory blood pressure recorder (Oscar II, Suntech Medical Instruments, Morrisville, NC).^12,13 Eighty recordings were programmed for the 24-hour period with 15-minute intervals from 6 am to 10 pm, and 30-minute intervals from 10 pm to 6 am. Follow-up ambulatory blood pressure recordings were made between 2 to 4 months post-randomization to evaluate if therapeutic goals had been met and 1 to 2 months later again if the goal was unmet and medications had been adjusted. Clinic blood pressures were measured at all visits and ambulatory blood pressures were performed at 18 and 36 months post-randomization.

Mobility Measures:

At baseline, 18 and 36 months, standardized mobility tests were performed in duplicate to document patients’ ability to maintain stance (side by side, semi-tandem, tandem, unipedal) and change position (supine- to-sit, sit-to-stand, forward reach, 8 meter walk, ascend and descend 4 steps) as previously described ^9,15,16. Excepting functional reach, documentation of timing was made with a digital stopwatch. Detailed descriptions of each standardized mobility tests are in the Supplementary Materials.

Cognitive testing:

Measures of executive functioning and processing speed included the Trail Making Test, Symbol Digit Modalities Test, Digit Span Memory Test, Hopkins Verbal Learning Test, Stroop Color and Word Test, and 2 subtests from the California Computerized Assessment Package, Simple Reaction Time, and Sequential Reaction Time¹². Descriptions of the neuropsychological testing are in the Supplementary Methods.

Brain image acquisition and analyses.

Magnetic resonance imaging of the brain was performed with a 1.5-T Siemens Avanto scanner (Erlangen, Germany) between February 2012 and July 2016 and subsequently with a 1.5-T Siemens Aera scanner until completion of the trial in October 2018. We verified the image analysis reproducibility to be virtually identical between the old and new scanners’ on seven study subjects who were imaged with both scanners. High-resolution structural images were obtained with three-dimensional (3D) T₁-weighted magnetization prepared rapid gradient echo (MPRAGE), T₂-weighted 3D-Fast Spin Echo (T₂), and a 3D fluid attenuated inversion recovery (FLAIR) series without the use of contrast material (Supplemental Figure 1). To account for brain size, the total volume of white matter hyperintensity lesions was expressed as a fraction of the intracranial cavity volume. Further details on brain imaging methods are described in the Supplementary Methods. Quantification of brain imaging data was performed by an investigator blinded to patient identifiers and any clinical data.

Additionally, research personnel performing mobility testing and neuropsychological testing were blinded to treatment assignment.

Statistical Analyses

It was estimated that approximately 140 randomized study patients would be required to complete assessments at 36 months based on our earlier observational studies^9,10 of older persons. Based on results from our cohort in these observational studies in which no intervention was administered, selected participants were evaluated who had 24-hour systolic blood pressures between 140 and 150 mmHg, inclusive, at follow-up, and between 120 and 130 mmHg, inclusive, at follow-up. These data demonstrated significant separation in the progression of white matter hyperintensity lesions and decline in gait speed.¹⁰ In INFINITY, the co-primary endpoints of gait speed and white matter hyperintensity volumes (corrected for intracranial cavity chamber size) showed that 70 patients in each group would yield 85% power to observe a difference of 0.15 meters/second in gait speed and a difference of 0.93% in white matter hypersensitivity volumes. It was expected that 30% of enrolled patients might not complete the trial due to death or disability; hence, the target for enrollment was 200 study participants.

Descriptive statistics (sample means and standard deviations or frequencies and percentages) were calculated for demographics at baseline (Table 1) and compared using t-tests for continuous variables and chi-square tests for categorical variables (age > 80 years, sex, race, smoking status, medication use, diagnosis of depression). Additionally, sample means and standard deviations at all three time points are shown for mobility, cognitive and WMH measures (Table 2). The primary analysis was intention-to-treat and compared mobility measures between the 36-month follow-up and baseline in the intensive- and standard-treatment groups. Linear mixed models for repeated measures were used to examine the relationship between each measure and study group. Models included time point, study group, and the time x study group interaction term. The time points included were baseline and end of study, and time was included in the model as a fixed effect. Similar models were fit for outcomes WMH% and cognitive measures. A sensitivity analysis, using the same statistical methods, was pre-specified to evaluate the outcomes in those patients who met and remained at goal ambulatory blood pressure throughout the trial (for intensive-treatment this was defined as 24 hour systolic blood pressure values of < 135 mmHg and for the standard group, 24-hour blood pressure values of > 140 mmHg).

Table 1.

Characteristics of the Patients at Baseline

Characteristic	Intensive-Treatment (n=99)	Standard-Treatment (n=100)
Age (years)	80.9 (4.4)	80.3 (3.8)
Age > 80 years old, n (%)	46 (46.5)	38 (38.0)
Female sex, n (%)	57 (57.6)	51 (51)
Race or ethnic group, n (%)
Asian	2 (2.02)	1 (1)
Black	7 (7.07)	6 (6)
White	85 (85.86)	89 (89)
Hispanic/Latino	5 (5.05)	4 (4)
Body mass index, kg/m2	27.5 ± 4.8	28.2 ± 5.1
Short Physical Performance Battery for mobility	10.8 ± 1.1	10.7 ± 1.1
Mini-mental state exam score	28.1 ± 1.5	28.3 ± 1.5
Education (years)	14.7 ± 3.0	15.6 ± 3.0
Smoking status, n (%)
Current	2 (2.0)	0 (0)
Ex-smoker	61 (61.6)	58 (58)
Never	31 (31.3)	36 (36)
Clinic Blood Pressures (mm Hg)
Systolic	149.7 ± 15.4	152.0 ± 17.5
Diastolic	73.9 ± 10.4	77.3 ± 10.2
Orthostatic change (Seated-Standing at 3 minutes)	3.0 ± 8.8	3.4 ± 10.7
Ambulatory Blood Pressure (mm Hg)
24-hour systolic	147.8 ±13.4	150.3 ± 13.4
24-hour diastolic	73.2 ± 8.6	74.9 ± 8.6*
Awake systolic	150.2 ± 12.4	152.1 ± 13.6
Awake diastolic	75.0 ± 8.5	76.6 ± 9.0
Sleep systolic	140.4 ± 21.1	143.9 ± 18.1
Sleep diastolic	67.2 ± 11.3	69.3 ± 9.9
Biochemical Studies
Serum creatinine, mg/dl	1.09 ± 0.37	1.09 ± 0.27
Low-density lipoprotein cholesterol, mg/dL	107.8 ± 32.4	101.8 ± 31.4
High-density lipoprotein cholesterol, mg/dL	57.8 ±17.5	57.1 ± 16.0
Serum glucose, mg/dL	104.7 ± 15.4	116.1± 43.4*
Medications, n (%)
Statins	57 (57.6)	58 (58)
Anti-hyperglycemic	14 (14.1)	16 (16)
Aspirin, low dose (< 325 mg/day)	48 (48.5)	47 (47)
Anti-depressants	18 (18.2)	14 (14)
Magnetic Resonance Imaging
Intracranial Cavity Volume (milliliters)	1451.5 ± 145.3	1502.6 ± 161.3
White Matter Hyperintensity (milliliters)	21.1 ± 17.0	20.2 ± 16.3
White matter hyperintensity/intracranial cavity volume (%), median (IQR)	1.17 (0.61 – 1.95)	0.93 (0.59 – 1.80)
Geriatric Depression Index, n (%)
Normal	82 (85.4)	88 (93.6)
Mild	14 (14.6)	6 (6.4)
Comorbidities, n (%)
Chronic Kidney Disease†	22 (22.2)	24 (24.2)
Cardiovascular Disease	39 (40.2)	39 (39)
Diabetes (type 2)	14 (14.1)	18 (18)

^*p < 0.05 between groups at baseline; all other comparisons are non-significant. The body-mass index is the weight in kilograms divided by the square of the height in meters. Race was self-reported. Mean ± SD shown unless otherwise noted.

^†estimated creatinine clearlance < 60 ml/minute and > 30 ml/minute

Table 2.

Changes from Baseline in Mobility and Neuroimaging Parameters

	Baseline		18 Months		End of Study		Change from Baseline to End of Study				P-Value†
	Intensive n=99	Standard n=100	Intensive n=86	Standard n=89	Intensive n=78	Standard n=83	Intensive n=78	Standard n=83	Difference1
Mobility Parameters	Mean ± SD*	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Estimate	95% CI
Gait (8 m) Time (seconds)	7.8 ± 1.9	7.9 ± 2.2	7.8 ± 2.3	7.8 ± 2.3	8.2 ± 2.5	8.2 ± 3.3	0.40 ± 2.0	0.42 ± 2.73	0.04	(−0.68, 0.76)	0.91
Four Stair Descent Time (seconds)	3.7 ± 1.1	3.8 ± 1.5	3.9 ± 1.9	4.1 ± 2	4.3 ± 1.6	4.4 ± 2.6	0.56 ± 1.49	0.70 ± 1.84	0.09	(−0.41, 0.59)	0.72
Four Stair Ascent Time (seconds)	3.7 ± 0.9	3.6 ± 1.0	3.6 ± 1.1	3.8 ± 1.4	4.1 ± 1.4	4.0 ± 1.6	0.40 ± 1.40	0.47 ± 1.24	0.03	(−0.36, 0.42)	0.88
Sit to Stand Time (seconds)	10.7 ± 3.8	10.2 ± 9.6	12.1 ± 3.7	12.3 ± 5.8	12.5 ± 3.9	13.2 ± 5.7	1.88 ± 4.59	2.97 ± 5.11	1.13	(−0.31, 2.57)	0.12
Supine to Sit Time (seconds)	3.0 ± 2.2	2.8 ± 1.2	3.1 ± 1.7	2.8 ± 1.2	3.1 ± 1.4	3.0 ± 1.3	0.02 ± 2.13	0.29 ± 1.37	0.15	(−0.38, 0.68)	0.58
Unipedal Balance (seconds)	12.2 ± 9.9	12.9 ± 10.8	10.5 ± 9.2	12.4 ± 11	10.0 ± 9.1	9.9 ± 8.5	−3.9 ± 9.56	−4.79 ± 8.61	−0.87	(−3.79, 2.04)	0.56
Functional Reach (cm)	27.0 ± 6.8	28.0 ± 7.4	24.7 ± 7.3	25.9 ± 8.1	23.4 ± 7.4	24.9 ± 7.8	−3.93 ± 7.89	−3.44 ± 7.86	−0.49	(−2.9, 1.92)	0.69
White Matter Parameters
White Matter Hyperintensity (%)	1.45 ± 1.16	1.35 ± 1.12	1.55 ± 1.02	1.58 ± 1.33	1.73 ± 1.28	1.74 ± 1.50	0.29 ± 0.39	0.50 ± 0.70	0.19	(0.02,0.37)	0.03
Sensitivity analysis‡	Intensive n=49	Standard n=54	Intensive n=48	Standard n=54	Intensive n=49	Standard n=53	Intensive n=49	Standard n=54	Difference†		P-Value†
	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD
White Matter Hyperintensity (%)	1.26 ± 1.13	1.35 ± 1.18	1.31 ± 0.75	1.74 ± 1.55	1.48 ± 1.10	1.93 ± 1.66	0.23 ± 0.40	0.58 ± 0.71	0.35	(0.13,0.58)	0.0028

SD – standard deviation

^*Means and SD are descriptive statistics

^†Results from linear mixed models (the pre-specified analyses in statistical analysis plan)

^‡Analysis of patients whose ambulatory blood pressure remained < 135 mmHg in the intensive treatment group and > 140 mmHg in the standard treatment group at goal and at 36 months

The proportions of adverse events were compared and confidence intervals were estimated. Fisher’s exact test was used when the number of events was small. All data were analyzed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA.) and p < 0.05 determined significance.

Results

Study Patients

A total of 199 patients were enrolled at the University of Connecticut Health Center between February 2012 through October 2015 (Figure 1). The treatment groups were well balanced with regard to baseline characteristics with the exceptions of small differences in the clinic diastolic blood pressure (3 mmHg lower in the intensive-treatment group) and serum glucose (11 mg/dl lower in the intensive-treatment group) (Table 1). Seventy-five per cent of patients in the intensive-treatment group received more than 1 antihypertensive agent compared to 58 per cent in the standard group. The average number of antihypertensive medications for the intensive group was higher than for the standard group (p=0.003; 2.4 ± 1.2 versus 1.9 ± 1.1 classes of therapy) (Supplemental Table 2). Similarly, the number of classes of antihypertensive medications in the intensive group was higher than for the standard group (Supplemental Table 3). Premature discontinuation from the study was higher in the intensive-treatment blood pressure group than in the standard-treatment group (20.2% of patients versus 14.0%) (Figure 1). The baseline characteristics of those patients who completed the trial versus those who did not complete the trial were comparable (Supplemental Table 4). Vital status was known for 100% of the study population at the end of the trial. Median duration of exposure in the intensive-treatment group was 982.5 days (interquartile range, 1071 to 1115) and 1027.8 days in the standard-treatment group (interquartile range, 1092 to 1122). Median time to reach goal blood pressure in the intensive-treatment group was 3 months (interquartile range, 2.1 to 5.2 months) and 3.8 months (interquartile range, 2.2 to 6.1 months) in the standard-treatment group (p = 0.04).

Figure 1.

Disposition of the Study Population (CONSORT)

Blood Pressure

The 2 treatment strategies resulted in sustained between-group differences in the 24-hour systolic blood pressure (Figure 2). At the time of reaching the assigned goal of therapy, the mean 24-hour systolic blood pressure was 127.7 mmHg in the intensive-treatment group and 144.0 mmHg in the standard-treatment group for an average difference of 16.3 mmHg. Mean 24-hour diastolic blood pressure at that time was 64.6 mmHg in the intensive-treatment group and 72.3 mmHg in the standard-treatment group. Following 3 years of follow-up, the mean 24-hour systolic blood pressure was 130.7 mmHg in the intensive-treatment group and 146.0 mmHg in the standard-treatment group. The systolic blood pressure in the clinic showed similar patterns to the ambulatory pressure – after 3 years of follow-up, the seated clinic systolic blood pressure was 132.6 mmHg in the intensive-treatment group and 145.6 mmHg in the standard-treatment group (Figure 2). Distribution of ACE inhibitors, angiotensin receptor blockers, alpha-blockers and beta-blockers were comparable in the two treatment groups, whereas the use of aldosterone blockers, calcium antagonists, and diuretics were greater in the intensive-treatment group (Supplemental Table 2).

Figure 2.

Mean (± 95% confidence intervals) for systolic blood pressures during the trial. The 24-hour average systolic blood pressure target (shown in the upper panel) in the intensive-treatment group was ≤130 mmHg and the target in the standard-treatment group was ≤145 mmHg. The clinic systolic blood pressure (shown in the lower panel) was recorded but not used to guide therapy.

Clinical and Neuroimaging Outcomes

Mobility outcomes are shown in Table 2. Mobility tests showed small declines in function at 36 months versus baseline, however, there were no differences observed between treatment groups. At baseline, the volume of white matter hyperintensity lesions corrected by intracranial cavity volume was comparable between treatment groups (Table 1). Changes in white matter hyperintensity volume from baseline to 36 months were significantly greater in patients in the standard-treatment group than in the intensive-treatment group (0.48 % versus 0.29 %, p = 0.025) (Table 2, Figure 3). In the sensitivity analysis of those patients who reached blood pressure goal and remained within their pre-specified treatment BP ranges, 24-hour systolic BPs were 130.6 ± 8.7 mmHg at 18 months and 124.8 ± 6.9 at 36 months for the intensive-treatment group and were 147.6 ± 11.1 mmHg at 18 months and 152.0 ± 8.5 mmHg at 36 months for the standard-treatment group. The changes from baseline in white-matter hyperintensity was 0.58% versus 0.23%, in the standard-treatment and intensive-treatment groups, respectively, p<0.01 (Table 2, Figure 3). In patients who changed categories (i.e. from intensive to standard BP levels or vice versa) during the trial, accumulation of white-matter hyperintensity was 0.36%.

Figure 3.

Changes from baseline at 18 and 36 months in white matter hyperintensity volumes corrected for intracranial cavity size (WMH%) by treatment assignment. The upper panel shows the intention-to-treat population. The lower panel shows the pre-specified per protocol population that included only those patients whose blood pressure remained at goal for 3 years in their assigned treatment (intensive-treatment or standard-treatment).

Cognitive parameters were stable over the trial with the exception of sequential choice reaction time (Supplemental Table 5). For this cognitive test, there were significantly lower reaction times in the standard-treatment group compared to the intensive-treatment group. Functional outcomes sensitivity analysis of those patients who reached goal blood pressure and remained within their treatment group for the entire trial were comparable to the intention-to-treat analysis with the exception of sit-to-stand, which showed that the intensive-treatment group was significantly slower, and sequential reaction time was no longer significant. (Supplemental Table 6).

Serious adverse events

There were 4 deaths (4%) in the standard-treatment group and 2 deaths (2%) in the intensive-treatment group. There were a greater number of major non-fatal cardiovascular events in the standard-treatment group compared to the intensive-treatment group (17 versus 4 patients; 18 versus 5 events), risk ratio = 0.24 (95% confidence intervals, 0.08, 0.68), p < 0.01) (Table 3). Falls, with or without injury, or due to mechanical or non-mechanical causes were comparable between treatment groups. Syncope and near syncope events were low in number and not different between treatment groups.

Table 3.

Serious Adverse Events

	Patients (Events)
Types of Events	Intensive-Treatment (n=99)	Standard-Treatment (n=100)
All Serious Events	36 (56)	38 (67)
Deaths	2	4
Cardiovascular	1	1
Non-cardiovascular	1	3
Cardiovascular Events*	4† (5)	17‡ (18)
Arrhythmia	4 (4)	6 (6)
Heart Failure	0 (0)	2 (2)
Stroke	1 (1)	2 (2)
Myocardial Infarction	0 (0)	4 (4)
Other Events∥	0 (0)	4 (4)
Events of Special Interest (pre-specified)
Falls with Injury	4 (5)	5 (5)
Falls without Injury	31 (41)	32# (45)
Mechanical	26 (36)	27 (35)
Non-mechanical	5 (5)	9 (10)
Syncope/pre-syncope	4 (5)	4 (4)

^*risk ratio for serious cardiovascular adverse events = 0.24 (95% confidence intervals, 0.08, 0.68), p<0.01

^†One patient had both an arrhythmia and a separate hospitalization for stroke

^‡One patient had both an arrhythmia and a separate hospitalization for heart failure

^∥Pulmonary embolism, unstable angina with revascularization, severe aortic stenosis, transient ischemic attack

^#Four patients had both mechanical and non-mechanical falls

Discussion

INFINITY showed that lowering the 24-hour mean ambulatory blood pressure to a target goal of ≤130 mm Hg or less compared with the standard goal of ≤145 mm Hg did not result in significantly different changes in mobility outcomes over the 3 years of the trial. However, less accrual of white matter disease was observed in older adults with systolic hypertension. Study patients assigned to the intensive-treatment group also had lower rates of major cardiovascular events, including myocardial infarction, stroke, and hospitalized heart failure.

The association between hypertension and white matter hyperintensity lesion progression is well established and associated with higher risk for cognitive decline^4,9, limited mobility⁶, and incident stroke³. In our earlier longitudinal assessments of older persons, the progression of white matter hyperintensity lesion volume was more strongly related to ambulatory blood pressure than clinic blood pressure⁹, possibly due to the improved reproducibility of ambulatory blood pressure in the elderly¹⁷ and led to the design of INFINITY¹². The premise of INFINITY was that more intensive reduction of ambulatory systolic blood pressure could lead to a slowing of the accumulation of white matter lesions due to microvascular disease that would eventually impact mobility. Prior work has demonstrated that damage to the periventricular white matter may be particularly detrimental to mobility as it impairs the visual-spatial input necessary for effective motor function.^{10, 18} Results of the trial did demonstrate a significant impact of lower 24-hour systolic blood pressure on progression of white matter lesions but was unaccompanied by a detectable impact on mobility and cognitive function. Given that the accrual of white matter hyperintensity volume occurs silently for many years, our inability to demonstrate effects on the functional parameters may be due to our shorter-term follow-up period compounded by the requirement that patients entering INFINITY had intact mobility, thus requiring more than 3 years to show functional deterioration. Our results are consistent with findings from a substudy of 192 patients with a prior history of stroke from the Perindopril Protection Against Recurrent Stroke (PROGRESS) trial.¹⁹ In PROGRESS, patients were randomized to active treatment with perindopril plus indapamide or placebo and MRIs were graded with a visual rating system from no WMH to severe WMH and calculation of WMH volumes was performed. In those patients receiving active drug therapy, there was a reduction in the accrual of volume of white matter hyperintensities measured by magnetic resonance imaging (by inspection with observationally-based quantitation) following an 11 mmHg lowering of clinic systolic blood pressure over 36 months. Of note, the patient population in PROGRESS was substantially younger than patients in INFINITY by 20 years and there was no functional evaluation. Our results are consistent with those from a larger study over 4 years which failed to demonstrate an effect of intensive reduction of the clinic blood pressure on the gait speed of older adults ²⁰. Effects of white matter hyperintensity volume on cognition appears to slow processing⁹ and other measures of executive function²¹ but does not necessarily result in cognitive decline¹⁰ as seen in our results. Recent results from the Systolic Blood Pressure Intervention Trial (SPRINT) did not show reductions in the incidence of probable dementia in patients with hypertension treated to a clinic systolic blood pressure goal of less than 120 mm Hg after 3.3 years of follow-up.²² However, SPRINT did demonstrate that targeting a systolic blood pressure of less than 120 mmHg, compared with less than 140 mmHg, was associated with a smaller increase in cerebral white matter lesion volume.²³ Finally, SPRINT did show a significant benefit of intensive-treatment of blood pressure on the risk of mild cognitive impairment, a measure that was not assessed in INFINITY.

While the sample size of our study and the intermediate duration of our trial are study limitations, we believe that our results have substantial clinical relevance. In a period of just 3 years, patients with prior evidence of microvascular disease of the white matter had a relative reduction of about 40 per cent in the buildup of white matter hyperintensity volumes in the intensive-treatment group which would likely be amplified by duration and stage of hypertension. Thus, lower ambulatory systolic blood pressure could benefit patients with cerebrovascular disease over the long-term as it is likely that white matter hyperintensity lesions accumulate over decades of older age making effective blood pressure control a key medical intervention capable of conserving function.

Although not powered for major cardiovascular outcomes, patients in the intensive-treatment group did have significantly fewer events than patients in the standard-treatment group. While these results are consistent with the findings of SPRINT²⁴ where intensive clinic blood pressure reduction led to reductions in mortality, there were too few deaths in INFINITY to make any meaningful comparisons with SPRINT. Notably, the non-fatal cardiovascular events were remarkably low in number in the intensive-treated group. In contrast, there were comparable rates of serious falls and syncope between the treatment groups suggesting that a target mean 24-hour ambulatory systolic blood pressure of 130 mmHg is safe in older persons with hypertension.

In conclusion, targeting a 24-hour ambulatory systolic blood pressure of ≤130 mm Hg, as compared with ≤145 mm Hg in older persons with systolic hypertension did not result in changes in mobility or cognitive parameters during a 3-year period. There was, however, less progression of white matter disease and lower rates of cardiovascular events in these older patients randomized to intensive-treatment. These data suggest that 3 years was likely to be too short in duration for reductions in accrual of white matter hyperintensity to translate to functional benefit.

Clinical Perspectives.

What is new?

• The effects of 2 levels of 24-hour ambulatory systolic blood pressure (intensive treatment of ≤ 130 mmHg versus standard treatment of ≤145 mmHg) on functional and cerebrovascular outcomes in older persons with hypertension were evaluated.

• After 3 years, intensive-treatment levels of ambulatory blood pressure did not show lessening of mobility or cognitive decline compared to standard-treatment levels, however, less accrual of white matter disease was observed in those with intensive-treatment.

• There were significantly less non-fatal cardiovascular events in the intensive-treatment group than in the standard-treatment group.

What are the clinical implications?

• Treatment of systolic hypertension guided by ambulatory blood pressure monitoring demonstrated that lower systolic blood pressure levels resulted in significantly less accumulation of small vessel disease in the white matter of the brain.

• While this structural finding was not accompanied by differences in mobility and cognitive outcomes, it is possible that over periods of time longer than 3 years, benefit on functional outcomes would occur.

• As intensive-treatment of ambulatory blood pressure in older people was well tolerated and accompanied by lower numbers of cardiovascular events, a 24-hour average systolic blood pressure ≤ 130 mmHg is a reasonable treatment goal.

Non-standard Abbreviations and Acronyms

HYVET

Hypertension in the Very Elderly Trial

INFINITY

INtensive versus Standard Ambulatory Blood Pressure Lowering to Prevent Functional DeclINe In The ElderlY

PROGRESS

Perindopril Protection Against Recurrent Stroke

SPRINT

Systolic Blood Pressure Intervention Trial

WMH

white matter hyperintensity