Accepted Manuscript Top 10 Landmark Studies in Hypertension Elizabeth W. Edwards, MD, MBA Donald J. DiPette, MD Raymond R. Townsend, MD Debbie L. Cohen, MD PII:

S1933-1711(14)00453-7

DOI:

10.1016/j.jash.2014.03.333

Reference:

JASH 502

To appear in:

Journal of the American Society of Hypertension

Received Date: 3 March 2014 Revised Date:

9 March 2014

Accepted Date: 10 March 2014

Please cite this article as: Edwards EW, DiPette DJ, Townsend RR, Cohen DL, Top 10 Landmark Studies in Hypertension, Journal of the American Society of Hypertension (2014), doi: 10.1016/ j.jash.2014.03.333. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Title: Top 10 Landmark Studies in Hypertension Short title: Top 10 Hypertension Studies Elizabeth W. Edwards, MD, MBA1 Donald J. DiPette, MD,1 Raymond R. Townsend, MD,2 Debbie L. Cohen, MD2 Department of Medicine, University of South Carolina School of Medicine, Columbia, SC; 2 Renal, Electrolyte, and Hypertension Division, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA.

RI PT

1

Acknowledgements: None

SC

Word Count: 6364 Number of references: 22

M AN U

Tables: 1 Figures: 0

Corresponding author:

TE D

Debbie L. Cohen, MD

Renal, Electrolyte, and Hypertension Division, Perelman School of Medicine at the University of Pennsylvania

Tel: 215-615-0794

AC C

Fax: 215-615-0349

EP

1 Founders Building, 3400 Spruce St, Philadelphia, PA, 19104

Email: [email protected]

Financial support: none

1

ACCEPTED MANUSCRIPT Abstract: (200) The field of hypertension has evolved considerably over the last 70 years, from a belief that elevated blood pressure was an inevitable consequence of aging and beneficial to maintaining perfusion to overwhelming proof of the cardiovascular morbidity and mortality associated with elevated blood pressure. The authors

RI PT

reviewed the literature on hypertension and selected 10 studies pivotal in changing physicians’ attitudes regarding the management, treatment, and outcomes of hypertensive patients. Four studies cover treatment initiation and blood pressure goals, two studies compare pharmacologic antihypertensive agents, and the final

SC

four address the approach to blood pressure control in special populations (diabetes mellitus and chronic

M AN U

kidney disease). The authors readily acknowledge the many other contributions to the field of hypertension not profiled here.

AC C

EP

TE D

Key words: blood pressure, clinical trials, hypertension

2

ACCEPTED MANUSCRIPT Introduction Management of hypertension has evolved considerably over the last 70 years. In the mid 1940s a reasonable systolic blood pressure was 100 plus the patient’s age. Physicians assumed elevated blood pressure was the body’s normal adaptation to stiffening and sclerotic arteries and thus a beneficial

RI PT

compensatory mechanism to maintain organ perfusion.1 To imagine the cardiovascular unawareness just seven decades ago is startling. Most physicians, including those prominent in the cardiovascular field, largely dismissed early data the American Society of Actuaries published in 1928 linking higher blood pressure to

SC

premature death.2 At the time, cardiovascular disease was considered an inevitable consequence of aging.

M AN U

Furthermore, even as the medical community became aware of the cardiovascular consequences of hypertension, the first relatively tolerable pharmacologic antihypertensive agents, β-blockers and diuretics, were not developed until the 1950s.

Hypertension affects about one billion people worldwide. Data from the recent report of the national health and nutrition examination survey (NHANES) conducted in the United States from 2009 to 2010

TE D

estimates that approximately 30% of U.S. adults have hypertension but only about half of those actually control their elevated blood pressure.3 Approximately 20% of individuals with hypertension do not know they

EP

have the condition, increasing their risk for heart attack and stroke, the first and third leading causes of death in the U.S. Substantial differences exist between demographic subgroups with adult men aged 18 to 44,

AC C

Mexican Americans, foreign-born, and those without health insurance having a lower prevalence of control of their hypertension than their counterparts.4 A recent predictive model showed that for every 10% increase in hypertension treatment an additional 14,000 deaths per year in the adult population ages 25 to 79 years could be prevented.5

Hypertension is now a major and well-known risk factor for cardiovascular disease. Starting at 115/75 mm Hg each increase of 20/10 mm Hg in blood pressure doubles the risk of cardiovascular disease.6 Treating a blood pressure above 140/90 mm Hg has been shown to reduce cardiovascular complications markedly. The 3

ACCEPTED MANUSCRIPT first studies investigating the benefits of treating blood pressure were prospectively designed and rapidly demonstrated that pharmacologically lowering extremely high blood pressures dramatically reduced target organ damage.7, 8, 9 The favorable results of these initial studies prompted greater and more in-depth questions regarding the optimal blood pressure at which to initiate pharmacologic treatment; the optimal

RI PT

blood pressure goals; risk reduction in various age, gender, racial, and ethnic groups and those with concomitant disease states such as diabetes and chronic kidney disease; and what pharmacologic agents are best utilized and when. Over the ensuing years and to the present, multiple evidence-based studies have

SC

answered these questions. The intent of this paper is to review 10 landmark clinical studies (table 1) the

M AN U

authors believe have had the greatest impact in changing physicians’ attitudes regarding hypertension. We chose 10 landmark studies in three focus areas: 1. Treatment initiation and goals, 2. Comparison of pharmacologic antihypertensive agents, and 3. Approach in special populations (patients with diabetes mellitus and those with chronic kidney disease). The first focus area examines whether to treat hypertension, when to initiate therapy, which patients benefit from lowering blood pressure, and what the optimal blood

TE D

pressure goal should be. The studies profiled include the Veterans Administration (VA) Cooperative Study (reported in 3 parts) ;7, 8, 9 the Hypertension, Detection, and Follow-up Program (HDFP) study;10, 11 Systolic

EP

Hypertension in the Elderly Program (SHEP) and its European counterpart (SYST-EUR) (counted together because of their similarity);12, 13 and the Hypertension Optimization Trial (HOT).14 The second focus area

AC C

discusses antihypertensive agents, classes, and combinations. The two studies in this focus area are The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT)15 and The Avoiding Cardiovascular events through Combination therapy in Patients Living with Systolic Hypertension (ACCOMPLISH) trial.16 The third focus area examines whether blood pressure treatment and goals should differ for patients with diabetes mellitus (United Kingdom Prospective Diabetes Study Group (UKPDS)18 and Action to Control Cardiovascular Risk in Diabetes (ACCORD)19 and chronic kidney disease (Modification of Diet in Renal Disease Study Group (MDRD)20 and African American Study of Kidney Disease and Hypertension Trial 4

ACCEPTED MANUSCRIPT (AASK).21 We acknowledge that there are many other contributions to the field of hypertension that are not profiled.

1. Treatment initiation and goals

RI PT

Whether to initiate treatment for non-malignant hypertension was a controversial question until the mid 1960s when the VA Cooperative Study Group answered that question with a resounding yes.7 The VA Cooperative Study Group conducted a landmark trial – the first randomized, placebo-controlled, double-blind,

SC

multi-institutional pharmacologic efficacy trial in cardiovascular medicine. The study randomized 523

M AN U

hypertensive men with diastolic blood pressures between 90 mm Hg and 129 mm Hg to treatment with hydrochlorothiazide, reserpine, and hydralazine, or placebo; results were published in three separate parts: the first included diastolic blood pressures between 115 and 129 mm Hg and the second and third included diastolic blood pressures between 90 and 114 mm Hg. The third part was a sub-analysis of the data and studied the influence of age, prior cardiovascular or renal abnormalities, and prior diastolic blood pressure in

TE D

the 90 to 114 mm Hg group.9 The first component of the study, published in 1967, reported the results on the 143 men with diastolic blood pressures between 115 and 129 mm Hg who received active pharmacologic

EP

treatment or placebo. Because of the striking advantage of treating these 143 men, researchers terminated the trial early and published their results: that active antihypertensive treatment favorably influenced the

AC C

clinical course of these patients.7 After 24 months of active treatment, blood pressure was reduced by 43/29.7 mm Hg in the active-treatment group whereas there was no significant change in mean blood pressure readings in the placebo group. Twenty-seven severe, complicating events occurred in placebo-treated patients as compared to two in the active group. Four deaths occurred in the placebo-treated group and none in the actively treated patients. Other complications in the placebo group included grade III or IV hypertensive retinopathy, congestive heart failure, increasing azotemia, cerebrovascular thrombosis, transient ischemic attacks, cerebral hemorrhage, myocardial infarction, and severely elevated blood pressure. Severe 5

ACCEPTED MANUSCRIPT complications in the active-treatment group were one cerebrovascular thrombosis and one case of multiple drug toxicity. This report was the first to establish clearly beneficial effects of pharmacologic treatment for severe hypertension. The results of the second component of the VA Cooperative Study were published in 1970.8 This study

RI PT

reported on the 380 men with diastolic blood pressures between 90 and 114 mm Hg who received active pharmacologic treatment or placebo. Fifteen percent of the patients dropped out of the study due to developing more severe hypertension or intolerance to therapy. The blood pressure in the treatment group

SC

decreased on average by 27.2/17.4 mm Hg while the blood pressure in the placebo group increased on

M AN U

average by 4.2/1.2 mm Hg. Morbidity attributed mainly to heart failure, stroke, and renal failure was reduced from 55% in the placebo group to 18% in the treatment group over a 5-year period. Deaths were reduced from 35 in the control group to 9 in the treatment group. Twenty patients in the placebo group developed serious diastolic hypertension of 125 mm Hg or higher compared to none in the treatment group. Pharmacologic treatment was more effective in preventing heart failure and stroke than coronary events but the benefit was

TE D

related to the level of pre-randomization blood pressure. The treated group benefitted, but not as dramatically as the 115 to 129 mm Hg group. This report focused on patients with mild to moderate diastolic

EP

hypertension and together with the first component of the VA Cooperative Study, provided early clinical evidence of the beneficial impact of lowering blood pressure. It established the repeatedly observed

AC C

phenomenon that in addition to target organ preservation, drug treatment of hypertension also decreased the likelihood of a patient’s progressing to a higher category of blood pressure and therefore a higher level of cardiovascular risk.

Investigators published a third report examining the mild to moderate hypertension group (those with diastolic blood pressures between 90 and 114 mm Hg) in 1972.9 This sub-analysis examined the influences of age, cardiovascular or renal abnormalities on blood pressure in more detail. This report showed that those

6

ACCEPTED MANUSCRIPT subjects with initial diastolic blood pressures less than 105 mm Hg and no prior cardiovascular, central nervous system, or renal abnormalities; and age less than 50 years had only a minor benefit of treatment. In summary, the greatest determinant of effectiveness of treatment across the spectrum of diastolic blood pressures was the initial diastolic blood pressure. Those in the 115 to 129 mm Hg group immediately

RI PT

enjoyed a dramatic benefit of treatment. There was also clear benefit in the 105 to 114 mm Hg group. The only group left that exhibited less benefit was the 90 to 104 mm Hg group. Within this group, those with prior cardiovascular or renal abnormalities or age greater than 50 years benefitted from treatment whereas those

SC

without prior cardiovascular or renal abnormalities or age less than 50 showed little benefit. However, the

would benefit this group to be determined.

M AN U

number of patients in this group was small and the period of follow up short, leaving whether treatment

Thus, the VA Cooperative Study left several questions unanswered. These unanswered questions included whether treating mild hypertension (diastolic blood pressures between 90 and 104 mg Hg) in those without prior cardiovascular or renal abnormalities or age less than 50 would be beneficial. In addition, all

TE D

participants were men. Thus, there was a need to determine the effectiveness of antihypertensive therapy on the general population including women, and particularly in younger individuals with milder hypertension and

EP

no prior risk factors. The findings of the VA Cooperative Study also brought into question the use of placebo therapies in hypertension trials. The National Heart Lung and Blood Institute (NHLBI) therefore sponsored The

AC C

Hypertension Detection and Follow-up Program (HDFP) trial which extended the findings of the VA Cooperative Study and published its report in 1979.10, 11 The HDFP randomized 10,500 patients with mild diastolic hypertension with diastolic blood pressures ranging between 90 and 104 mm Hg to two groups: stepped care and referred care. The stepped-care group received a standardized program of pharmacologic antihypertensive therapy. In this arm medications were increased in a stepwise fashion to achieve a diastolic blood pressure goal of less than 90 mm Hg. Antihypertensive agents consisted primarily of the thiazide-like diuretic chlorthalidone (or triamterene or 7

ACCEPTED MANUSCRIPT spironolactone), reserpine (or methyldopa), hydralazine, and guanethidine. The diastolic blood pressure goal was less than 90 mm Hg for most participants; for those entering the study who already had a diastolic blood pressure less than 100 mm Hg, the goal was less than 85 mm Hg. Investigators also counseled participants about a low sodium intake, weight loss, lowering cholesterol, and smoking cessation. The referred-care group

RI PT

was referred to the participants’ regular community medical care sources to receive usual treatment. Deaths from cardiovascular disease were 26% lower and five-year mortality was 20% lower in the stepped-care group compared to the referred-care group. In addition the stepped-care group had a significant

SC

reduction in cardiovascular events. The committee determined no deaths were attributable to drug toxicity

M AN U

and that drug side effects rarely prevented achievement of the blood pressure goal in the stepped-care group. This trial established the efficacy of antihypertensive therapy for adults with mild blood pressure elevations and this favorable outcome was despite the fact that the control group was not a true placebo group with most control participants receiving some antihypertensive treatment.

HDFP investigators also published a separate report categorizing their findings based on race, gender,

TE D

and age.11 Throughout the trial, for black men, black women, white men, and white women and for all people aged 30 to 49, 50 to 59, and 60 to 69 at entry, control of blood pressure was consistently better for stepped-

EP

care than for referred-care participants. There was no difference in favorable outcomes in the stepped-care group between race, gender, and age.

AC C

Both the VA Cooperative Study and the HDFP study investigated the effect of lowering blood pressure on participants with elevated diastolic blood pressures. Previous studies had all used diastolic blood pressures as targets and until as recently as 1991, skepticism remained about the effectiveness of treating patients with isolated systolic hypertension (systolic blood pressure greater than 160 and diastolic blood pressure less than 90 mm Hg). The Systolic Hypertension in the Elderly Program (SHEP)12 trial tackled this conundrum. SHEP investigators recruited individuals over age 60 with a systolic blood pressure greater than or equal to 160 mm Hg and a diastolic blood pressure less than or equal to 90 mm Hg. The study recruited and 8

ACCEPTED MANUSCRIPT randomized 4736 patients to active pharmacologic treatment or to placebo. The treatment group received chlorthalidone (at lower doses than in previous trials) as well as atenolol and reserpine if necessary to achieve the systolic blood pressure goal of less than 160 mm Hg. Interestingly, 33% of the placebo group also received antihypertensive agents. The average achieved blood pressure in the treatment group was 143/68 mm Hg; the

RI PT

average blood pressure in the placebo group was 155/72 mm Hg. Over the course of the five years of the study, mean blood pressure dropped 28/9 mm Hg in the treatment group and 15/4 mm Hg in the placebo group. Subjects in the treatment group had a significant reduction of all strokes, fatal and nonfatal, by 36%

SC

(RR 0.64, 95% CI, 0.50 to 0.82). Along with this reduction in stroke, there was a 25% reduction (RR 0.75, 95% CI

M AN U

0.60 to 0.94) in all coronary heart disease, a 32% reduction (RR 0.68, 95% CI, 0.58 to 0.0.79) in all cardiovascular disease, and a 13% reduction (RR 0.87, 95% CI 0.73 to 1.05) in total mortality. In addition, transient ischemic attacks and episodes of congestive heart failure also dropped. Similar to HDFP, the treatment group enjoyed significantly better outcomes in spite of the fact that 33% of the placebo group received antihypertensive treatment. The SHEP study also showed that a low dose of diuretic lowers blood

TE D

pressure effectively and results in better outcomes. This was the first trial to demonstrate that there is significant benefit to lowering systolic blood pressure in the elderly population. Before this trial, many felt that

EP

the value of treating isolated systolic blood pressure was unproven and potentially could result in harm such as predisposing to falls and fractures; after SHEP, the medical community could no longer ignore isolated

AC C

systolic hypertension even in elderly populations. The European Working Party on High Blood Pressure in the Elderly (SYST-EUR)13 study posed a similar question to the SHEP trial: would treatment of isolated systolic hypertension in the elderly, but with a different class of medication, namely calcium channel blockers, reduce cardiovascular complications? The answer was once again a resounding yes. This was another large placebo-controlled trial that supported the findings of previous studies.

9

ACCEPTED MANUSCRIPT In SYST-EUR all 4695 study participants initially began placebo treatment.13 Systolic blood pressures were in the range of 160 to 219 mm Hg and diastolic blood pressures less than 95 mm Hg on entry to the study. The average age of participants was 70 years. After three months, investigators randomized participants to active treatment or to continue with placebo. In SYST-EUR active treatment consisted of

RI PT

nitrendipine, enalapril, and hydrochlorothiazide. The treatment goal was to reduce systolic blood pressure by at least 20 mm Hg to less than 150 mm Hg. The fall in blood pressure was greater with active therapy than with placebo (23/7 mm Hg versus 13/2 mm Hg). After two years of follow up, those in the treatment group

SC

had 42% fewer total strokes (95% CI, 17 to 60%, P=0.03), 44% fewer non fatal strokes (95% CI, 14% to 63%,

M AN U

P=0.007), 26% fewer coronary deaths (95% CI, 15% to 54%, P=0.17), and 33% fewer non-fatal cardiac endpoints (95% CI, 3% to 53%, P=0.03). Similar to SHEP, all-cause mortality was not different between the active and placebo groups. The study was terminated early because of the 42% total stroke reduction in the active-treatment arm. Investigators concluded that among elderly patients with isolated systolic hypertension, antihypertensive treatment starting with the calcium channel blocker nitrendipine reduces the rate of

TE D

cardiovascular outcomes. Investigators inferred it was the lowering of blood pressure that was beneficial, not the actual drug. Both the SHEP and SYST-EUR trials demonstrated that treating isolated systolic hypertension

EP

in the elderly with diuretics or calcium channel blockers reduces cardiovascular outcomes. Together, SHEP and SYST-EUR established the safety of drug treatment in older persons.

AC C

The Hypertension Optimal Treatment (HOT) Study Group14 acknowledged the ample evidence that treating diastolic blood pressure greater than 90 mm Hg in any age, gender, or ethnic group had cardiovascular benefits. However, the question remained as to what is the optimal diastolic blood pressure goal to achieve once therapy begins. To answer this question the HOT trial randomized 18,790 patients from 26 countries, aged 50 to 80 years (mean 61.5 years) with a baseline diastolic blood pressure between 100 and 115 mm Hg (mean 105 mm Hg) to three different target diastolic blood pressures: less than or equal to 80 mm Hg or less than or equal to 85 mm Hg or less than or equal to 90 mm Hg.14 All participants received the calcium 10

ACCEPTED MANUSCRIPT channel blocker felodipine as baseline therapy with the addition of other agents, according to a five-step regimen, to achieve the assigned diastolic blood pressure targets. The aim was to evaluate the relationship between these three levels of achieved diastolic blood pressure and cardiovascular morbidity and mortality in hypertensive patients.

RI PT

The results demonstrated that the lowest incidence of major cardiovascular events occurred at an average diastolic blood pressure of 82.6 mm Hg; the lowest risk of cardiovascular mortality occurred at 86.5 mm Hg. Further reductions did not result in further benefit but were not harmful. Interestingly, in patients

SC

with diabetes there was a 51% further reduction (P for trend 0.005) in major cardiovascular events in the less

2. Comparison of pharmacologic agents

M AN U

than or equal to 80 mm Hg group compared to the less than or equal to 90 mm Hg group.

The second focus area of this review is data related to the comparison of differing pharmacologic antihypertensive classes. To address this focus area, the authors selected two studies. The first of these two

TE D

landmark studies, the ALLHAT study, compared different pharmacologic antihypertensive classes in order to determine if one drug class is better than another in improving outcomes in hypertensive individuals. The

EP

ALLHAT study is the largest randomized, double-blind, antihypertensive trial to date.15 The second study selected was the ACCOMPLISH study which examined potential differences in outcomes with different

AC C

combination (two-drug) therapy.16

The ALLHAT study was conducted between 1994 and 2002. At the time, prescriptions for newer, more expensive, classes of antihypertensives were increasing. ALLHAT sought to identify which class of antihypertensive medication, if any, worked best to lower cardiovascular risk. The antihypertensive arm of the study initially compared four classes of agents: diuretics (chlorthalidone), ACE-inhibitors (lisinopril), calcium channel blockers (amlodipine), and α-blockers (doxazosin) in the reduction of cardiovascular events and mortality in patients with hypertension. Importantly, the doxazosin arm of the study was stopped prematurely 11

ACCEPTED MANUSCRIPT because of an increased incidence of congestive heart failure and a higher risk of combined cardiovascular outcomes including stroke.15 Following the discontinuation of the doxazosin arm, the remaining three arms continued to completion. 33,357 participants with hypertension and at least one other coronary disease risk factor were

RI PT

randomly assigned to receive either chlorthalidone, amlodipine, or lisinopril, and followed for an average of 4.9 years. The study population was comprised of 35% African Amerians and 19% Hispanics. African Americans were specifically included as most prior hypertension studies had included very few African

SC

American participants. One-third of participants also had diabetes. The primary outcome of the study was

M AN U

combined fatal coronary heart disease or nonfatal myocardial infarction and stroke which occurred in 2956 participants. There was no difference in the primary outcome between the three drug classes. Likewise, allcause mortality did not differ between the three groups. While cumulative event rates for the primary outcome were identical between the three groups, there was significantly less heart failure reported in the patients randomized to the diuretic arm (P fewer non-fatal then randomized to 59 years strokes, 27% fewer active treatment coronary deaths, with nitrendipine 33% fewer non(enalapril and HCTZ fatal cardiac if necessary to endpoints achieve goal SBP, compared to 150) or placebo. placebo. No difference in allcause mortality. Study terminated early because of significant stroke reduction in treatment group. DBP 100 – Lowest incidence 18,790 patients 115 mmHg, of major CV events randomized to 3 at average DBP of different target DBP: age 50 – 80 82.6; lowest risk of ≤80, ≤85, or ≤90. All years participants CV mortality at received felodipine 86.5; further reductions did not plus other agents to achieve DBP goal. result in further benefit but were not harmful. 33,357 subjects HTN and at No difference in randomized to least one primary outcome chlorthalidone, other or all-cause lisinopril, coronary mortality between amlodipine or heart disease 3 groups. doxazosin. risk factor, Cumulative event Doxazosin arm 1/3 had rates identical stopped early due to diabetes between 3 groups. increased CHF and significantly less stroke. CHF in diuretic arm. Comparisons were similar in subjects participants with

endpoints.

Indicated treating isolated systolic hypertension in older patients with calcium channel blockers reduces stroke and cardiac endpoints.

RI PT

necessary to achieve goal SBP < 160) or placebo.

Indicated optimal DBP not established but worthwhile to maintain DBP < 90 mmHg.

Showed no difference in CHD or nonfatal MI and stroke between CCB, ACE-I, or diuretics.

ACCEPTED MANUSCRIPT

9. United Kingdom Prospective Diabetes Study Group (UKPDS), 1999

1148 people Type II randomized to tight diabetes in BP control (goal BP < the UK 150/85) or less tight control (goal BP