Hypertension Guidelines: A Major Reappraisal Critically Examines the Available Evidence


  • Michael A. Weber MD,

    1. From the Cardiology Division, State University of New York, Downstate College of Medicine, New York, NY;
    2. the Division of Clinical Pharmacology, Department of Medicine, University of Miami Miller School of Medicine, Miami, FL
    Search for more papers by this author
  • Barry J. Materson MD

    1. From the Cardiology Division, State University of New York, Downstate College of Medicine, New York, NY;
    2. the Division of Clinical Pharmacology, Department of Medicine, University of Miami Miller School of Medicine, Miami, FL
    Search for more papers by this author

Michael A. Weber, MD, Professor of Medicine, SUNY Downstate College of Medicine, 308 E. 38th Street, Suite 201, New York, NY 10016
E-mail: michaelwebermd@cs.com

In November 2009 the authors of the European Hypertension Guidelines—a project of the European Society of Hypertension and the European Society of Cardiology—published a lengthy and detailed reappraisal of the evidence underlying the guidelines’ recommendations.1 Remarkably, this rigorous exercise occurred only 2 years after publication of the previous guidelines, highlighting a desire not only to integrate the results of recent research but also to better clarify which recommendations are based on hard evidence and which on judgment and tradition. This fine piece of scholarship by our European colleagues deserves close attention.

The Guidelines Process

The creation of practice guidelines for hypertension, as with other disciplines, is largely based on available published evidence that can inform clinicians on the optimal strategies for managing their patients. Because the available evidence is far from complete in many essential areas, the recommendations made by guidelines committees inevitably reflect the experience and judgment of the experts writing these recommendations.

In the United States, the last such report on hypertension management was the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) published in 2003.2 Even though it is now over 6 years later, the follow-up report—presumably JNC 8—has not yet been completed. The reason for this delay is the desire of the current Committee to put out recommendations based on exhaustive scrutiny of available evidence. This approach, while reducing subjective judgment in the writing of recommendations, may be limited by the deficiencies in our knowledge of hypertension. Furthermore, there is a risk of leaving practitioners without the benefit of expert opinion in the many areas not supported by high-level evidence. And, as noted in the reappraisal, even different sets of high-level evidence can be conflicting.

The European Approach

Our colleagues in Europe, however, have tackled the problem of providing responsible and authoritative guidelines in a rather different fashion. The European Society of Hypertension/European Society of Cardiology (ESH/ESC) guidelines on hypertension, which were most recently published in 2007,3 have now been amended barely 2 years later by a detailed update.3 This most recent document, reflecting the work of a large group of accomplished experts, has focused its attention on a rigorous reevaluation of all available hypertension clinical research.

Perhaps the most interesting aspect of this current paper—which its authors have appropriately termed a “reappraisal”—is its highly critical exploration of both contemporary and older clinical evidence. To put it bluntly, many of the statements and assumptions in previous guidelines are now being questioned, making it clear that essential pieces of information needed to guide the management of clinical hypertension are still not available. The European group is to be praised for its willingness to challenge and question its previous recommendations. At the same time, their insights and questions help explain why the task of the JNC 8 committee in the United States has become so difficult and prolonged.

The American Approach

A major difference between the United States and the European guidelines for hypertension is that the American recommendations for diagnosing hypertension are based primarily on blood pressure values. True, the JNC 7 report states that the presence of such concomitant conditions as diabetes or chronic kidney disease reduces the blood pressure level at which hypertension should be diagnosed and treated. But though the JNC 7 committee attempted to avoid creating disease-based silos it was unable to find a satisfactory way to combine blood pressure levels with risk factors. Even so, the JNC committee performed field testing of its recommendations to confirm their applicability to the clinical setting. The European guidelines, however, base clinical decision making on a far wider range of concomitant cardiovascular risk factors, including evidence for previous events as well as risk factors known to increase the probability of cardiovascular outcomes.

Indeed, the Europeans, in assessing total cardiovascular risk in their patients, even take into account subclinical evidence of cardiovascular disease. The reappraisal document emphasizes the view that the presence of such findings as micro- or macroalbuminuria or left ventricular hypertrophy should be taken into account when assessing patients with hypertension and determining the need for treatment. Since these so-called intermediate findings have predictive power for future fatal and nonfatal events, it is possible to see the logic for their inclusion in the decision-making process; still, evidence-based purists could argue that since we do not yet have clinical trial evidence regarding this strategy’s effect on patient outcomes, it should not yet be formally recommended.

When to Start Treatment: Is 140/90 Still Valid?

In recent years the major guidelines on hypertension have agreed that drug therapy should be started in patients whose blood pressures were at 140/90 mm Hg or higher despite appropriate lifestyle management.2,3 For patients at higher risk of cardiovascular outcomes, including those with diabetes, chronic kidney disease, or coronary disease, 130/80 mm Hg has been specified as the threshold for the diagnosis and treatment of hypertension.

When these values were first proposed it was acknowledged that they were not based on substantive clinical evidence, but repetition has the effect of converting a reasonable suggestion into an entrenched belief. Now, the publication of the reappraisal takes a further challenging look at these blood pressure criteria, identifying gaps in our knowledge and the types of future research that will be needed.

A Critical Gray Area

We simply do not have clear evidence on what to do with patients whose untreated systolic blood pressures are in the range between 140 and 159 mm Hg. This is a hard admission to make because this range—usually classified as Stage 1 hypertension— probably includes more than half of all individuals regarded as having hypertension. Unfortunately, no authoritative clinical trial has recruited untreated patients with baseline blood pressures in this range and demonstrated an outcomes benefit of treatment. The principal studies that have demonstrated benefits of antihypertensive treatment have studied patients with blood pressures higher than this range4,5 or patients at high risk already receiving therapies in whom it was not possible to determine their original untreated values.6,7 A summary of the major statements on this topic in the reappraisal paper is shown in Table I. Their recommendations appear to us to be well-considered and practical.

Table I.   Treatment Initiationa
  1. Abbreviations: BP, blood pressure; DBP, diastolic blood pressure; SBP, systolic blood pressure. aModified from Mancia et al,1 Box 3.

1. Although trial evidence is scanty, it appears reasonable to recommend that, in stage 1 hypertensive patients (SBP 140–159 mm Hg) at low and moderate risk, drug therapy should be started after a suitable period with lifestyle changes. Prompter initiation of treatment is advisable if grade 1 hypertension is associated with a high level of risk, or if hypertension is grade 2 or 3.
2. In patients with high normal BP (SBP 130–139 mm Hg or DBP 85–89 mm Hg) uncomplicated by diabetes or previous cardiovascular events, no trial evidence is available of treatment benefits except for a delay in the onset of hypertension (crossing the 140/90 mm Hg cutoff).
3. Initiation of hypertensive drug therapy in diabetic patients with high normal BP is presently unsupported by prospective trial evidence. For the time being, it appears prudent to recommend treatment initiation in high normal BP diabetic patients if subclinical organ damage (particularly microalbuminuria or proteinuria) is present.
4. In general, early BP-lowering treatment, before organ damage develops or becomes irreversible or cardiovascular events occur, appears a prudent recommendation because in high-risk hypertensive patients, even intense cardiovascular drug therapy, though beneficial, is unable to lower total cardiovascular risk below the high-risk threshold.

Starting Therapy in Special Groups

Guidelines include advice about initiating treatment in special groups of hypertensive patients. In patients with diabetes, for instance, they recommended that treatment be started at blood pressures of 130/80 mm Hg or higher.2,3 In reality, though, only one outcomes trial in hypertension included diabetic patients with baseline blood pressures <160 mm Hg, and even then provided inconclusive data on major events.8 On the other hand, there were some nonhypertensive patients with diabetes included in the Heart Outcomes Prevention Evaluation (HOPE) trial, but statistical details of their outcomes benefits during active therapy were not fully reported.9 Another more recent trial demonstrated a reduction in endpoints during treatment of patients with diabetes, but their baseline values were 140 mm Hg or above, not 130 mm Hg.10 Due to these data gaps, the useful ESH/ESC recommendations (Table I [item 3] and Table II [item 2]) are based on a combination of evidence and expert opinion.

Table II.   Blood Pressure Goals of Treatmenta
  1. Abbreviations: BP, blood pressure; DBP, diastolic blood pressure; SBP, systolic blood pressure. aModified from Mancia et al,1 Box 4.

1. On the whole, there is sufficient evidence to recommend that SBP be lowered below 140 mm Hg (and DBP below 90 mm Hg) in all hypertensive patients, both those at low moderate risk and those at high risk. Evidence is missing only in elderly hypertensive patients in whom the benefit of lowering SBP below 140 mm Hg has never been tested in randomized trials.
2. The recommendation of previous guidelines to aim at a lower goal SBP (<130 mm Hg) in diabetic patients and in patients at very high cardiovascular risk (previous cardiovascular events) may be wise, but it is not consistently supported by trial evidence. In no randomized trial in diabetic patients has SBP been brought down to below 130 mm Hg with proven benefits, and trials in which SBP was lowered to below 130 mm Hg in patients with previous cardiovascular events have given inconsistent results.
3. Despite their obvious limitations and a lower strength of evidence, post-hoc analyses of trial data indicate a progressive reduction of cardiovascular events incidence with progressive lowering of SBP down to about 120 mm Hg and DBP down to about 75 mm Hg, although the additional benefit at low BP values is small. A J-curve phenomenon is unlikely to occur until lower values are reached, except perhaps in patients with advanced atherosclerotic artery disease. Great caution should be applied in considering outcomes data from post-hoc analyses.

The same lack of hard evidence applies to elderly hypertensive patients, for none of the outcomes trials in this group have included people with initial systolic blood pressures below 160 mm Hg. There is also a lack of clear information in hypertensive patients at high risk due to previous cardiovascular disease.11,12 In patients with prior coronary events it has been difficult to establish baseline blood pressure values because their post-event care had often required the use of drugs that had blood pressure-lowering properties. In patients with previous cerebrovascular events, the Perindopril Protection Against Recurrent Stroke (PROGRESS) trial demonstrated cardiovascular and stroke benefits of combined angiotensin-converting enzyme (ACE) inhibitor and diuretic treatment, including patients without a history of hypertension.13 Still, a later analysis indicated that these benefits occurred primarily in patients with blood pressures >140 mm Hg.14 Moreover, the recent Prevention Regimen for Effectively Avoiding Second Strokes (PROFESS) trial, which included some nonhypertensive patients, failed to show clear outcomes benefits in post-stroke patients treated with an angiotensin receptor blocker.15 Therefore, ESH/ESC declined to make firm recommendations on this issue (Table I [item 4)], but they did state that is was “prudent” to begin early blood pressure lowering treatment “before organ damage develops or becomes irreversible.” It will be interesting to see whether or how JNC 8 chooses to address this issue.

Choosing Blood Pressure Targets

It is logical to argue that if a certain blood pressure level is proposed as a starting point for treatment, then reducing blood pressure below that level should be the goal of treatment. And, indeed, for most hypertensive patients the major guidelines use 140/90 mm Hg for that dual purpose.2,3 However, in considering available data, some studies may speak only to the starting points for treatment, and others only to goals.

For patients with uncomplicated hypertension, a critical review of several trials found a significant outcomes advantage of active treatment over placebo when systolic blood pressures were reduced below 140 mm Hg by the administered treatment.16 It must be noted, however, that the baseline blood pressure was at least 160 mm Hg in these studies, so strictly speaking they provide no evidence of benefit when initial systolic values between 140 and 159 mm Hg (stage 1 hypertension) are reduced below 140 mm Hg.

Special Groups

Data to support current guidelines recommendations for treatment targets in special groups are meager. Trials in the elderly have shown the benefits of reducing blood pressure in these patients.17–19 But none of these trials achieved systolic blood pressures below 140 mm Hg, and it may be most realistic (and evidence-based) to suggest a target of 150 mm Hg for older patients.

Similarly, it is not possible to specify a precise target in people with diabetes. Because they are at a high risk of cardiovascular events, effective antihypertensive therapy in diabetic patients produces a greater absolute reduction in events than in nondiabetic patients.20,21 But there is no direct evidence from reported trials to support the treatment goal of <130/80 mm Hg recommended by current guidelines. As well, for patients with coronary artery disease there are no authoritative studies to support a target of <130/80 mm Hg. Again, in the absence of definitive evidence, the ESH/ESC decided to rely on expert opinion in its practical recommendations displayed in Table II.

Patients with chronic kidney disease appear to get protection from further deterioration of renal function by drugs—most typically blockers of the renin-angiotensin system—that reduce both blood pressure and albuminuria. At least one study has shown that these two beneficial actions are independent of each other,22 indicating that both should be pursued. But, as with patients with diabetes, there is no evidence, pro or con, regarding the value of a blood pressure target below 130/80 mm Hg.

Data From Post Hoc Analyses

In high risk hypertensive patients in the Valsartan Antihypertensive Long-Term Use Evaluation (VALUE) trial,23 those who achieved systolic blood pressures <140 mm Hg, as compared with those who didn’t, had significantly fewer fatal and nonfatal cardiovascular and stroke events, regardless of the type of therapy used. More recently, other post-hoc analyses of large outcomes trials in high risk subjects—not all of them patients with hypertension—have also indicated that lower achieved blood pressures (<140/90 mm Hg or even lower) are associated with the best cardiovascular protection, particularly for stroke.24,25

But while these findings are of interest, such analyses do not provide the type of data that would be produced by trials in which event rates are compared among hypertensive patients randomized to differing blood pressure targets. Indeed, it is very possible that the apparent association between low blood pressures and low event rates in the post-hoc analyses simply reflect the fact that patients with relatively less arterial disease are more likely to have both a better cardiovascular prognosis as well as a greater blood pressure response to therapy. Ultimately well designed prospective trials will be required to resolve these issues. The major conclusions of the reappraisal paper on blood pressure targets are listed in Table II.

A Practical Approach

A look at the currently available evidence indicates that patients with blood pressures above 160 mm Hg (stage 2 hypertension) get meaningful cardiovascular protection if their blood pressures are reduced to below 140/90 mm Hg. As already discussed, it is not certain that further blood pressure reduction would provide additional outcomes benefits.

But there are no firm data to indicate that patients starting with blood pressures between 140/90 and 159/99 mm Hg (stage 1) will benefit from antihypertensive therapy. Even so, this lack of evidence has not deterred guidelines committees from recommending treatment in such patients.

The European guidelines have attempted to use a scientifically-based approach, reasoning that lifestyle modifications should be the primary treatment in stage 1 hypertensive patients who do not have concomitant cardiovascular risk factors or disease; whereas drug therapy should be introduced more rapidly in stage 1 patients with a history of previous events or with such markers of vascular disease as microalbuminuria, left ventricular hypertrophy, or findings of arterial stiffness or lesions.

The American approach is more empirical and recommends drug therapy for all stage 1 hypertensive patients.2 This, we believe, is an appropriate position for the JNC Committee to take. Epidemiologic data clearly indicate a heightened probability of major events in stage 1 hypertension as compared with lower blood pressures.26 As well, it is not unreasonable to extrapolate from the positive results in patients with more severe forms of hypertension. After all, there is no evidence to suggest harm from following this practice. Fortunately, modern drugs are largely free of the side effects that adversely affected quality of life with older drugs, so adding to the feasibility of this strategy.

Still, it must be acknowledged that definitive data to guide these decisions are still missing. Appropriately designed studies to help define optimal starting points for treatment, as well as optimal treatment goals, are urgently needed. One such trial, evaluating the effects on cardiovascular event rates of different achieved blood pressure goals in patients with diabetes is expected to publish its results in 2010.27

Choosing Drugs

The debate over which class of drug is most appropriate for initiating therapy has by now become largely moot. There are 3 major reasons for this: First, there is a belief among experts that the primary benefits of antihypertensive therapy are mediated by blood pressure reduction, regardless of how this is achieved; second, a variety of trials have indicated similar—albeit not identical—effects on clinical outcomes of the different drug classes; and, third, there is a growing recognition that many, if not most, patients will require more than one drug for blood pressure control, thus obviating the need to depend on a single agent. Furthermore, common concomitant conditions in hypertensive patients, including diabetes, angina, chronic kidney disease, and heart failure, will mandate the selection of particular drug classes.

Accordingly, the European guidelines acknowledge that diuretics, β-blockers, calcium channel blockers, ACE inhibitors, and angiotensin receptor blockers can all be considered as first line drugs.3 The US guidelines (last published in 2003), however, still retain a preference for thiazide diuretics based on their apparent equivalence to other drug classes in cardiovascular protection and their claimed lower cost.2 These assumptions have long been challenged and it seems unlikely that future guidelines will repeat this recommendation.

The new reappraisal document from the ESH/ESC authors has thoroughly evaluated new evidence pertaining to the individual drug classes. Their conclusions are shown in Table III. Despite many interesting studies that have been published recently, the basic recommendations regarding drug selection have not changed and clinicians remain free to select those drugs they believe would be most appropriate for managing their individual patients.

Table III.   Choice of Antihypertensive Drugsa
  1. Abbreviations: ACE, angiotensin-converting inhibitor; BP, blood pressure; ESH/ESC, European Society of Hypertension/European Society of Cardiology. aModified from Mancia et al.1, Box 5.

1. Large-scale metaanalyses of available data confirm that major antihypertensive drug classes, that is, diuretics, ACE inhibitors, calcium antagonists, angiotensin receptor antagonists, and β-blockers do not differ significantly in their overall ability to reduce BP in hypertension.
2. There is also no undisputable evidence that major drug classes differ in their ability to protect against overall cardiovascular risk or cause-specific cardiovascular events such as stroke and myocardial infarction. The 2007 ESH/ESC guidelines conclusion that diuretics, ACE inhibitors, calcium antagonists, angiotensin receptor antagonists, and β-blockers can all be considered suitable for initiation of antihypertensive treatment, as well as for its maintenance, can thus be confirmed.
3. Because the percentage of patients responsive to any drug classes is limited, and patients responsive to one drug are often not those responsive to another drug, keeping the number of drug options large increases the chance of BP control in a larger fraction of hypertensives. This is of crucial importance because cardiovascular protection by antihypertensive treatment substantially depends on BP lowering per se, regardless of how it is obtained.
4. Each drug class has contraindications as well as favorable effects in specific clinical settings. The choice of drugs should be made according to this evidence. The traditional ranking of drugs into first, second, and third-line choices, with an average patient as reference, has now little scientific and practical justification and should be avoided.
5. Drugs acting via direct renin inhibition are the only new classes of antihypertensive agents that have recently become available for clinical use.

ESH/ESC took head-on the contentious issue of the use of β-blockers for the treatment of hypertension. Their thoughtful discussion deserves reading. They do not support the overall negative recommendations on this heterogeneous group of drugs, some of which may be of particular value in hypertension.

Combination Treatment

The use of combinations of antihypertensive drugs is growing, particularly with the increasing clinical availability of fixed dose (2 or more drugs in one tablet or capsule) products. The updated European statement acknowledges that progressing too rapidly to combination therapy, or initiating treatment with 2 drugs, could cause some patients who might be controlled with just one well selected drug to get more drug exposure than required. On the other hand, from the public health perspective, the experts argue that early or initial combination treatment would sharply increase blood pressure control rates among the large numbers of hypertensive patients currently not achieving desirable blood targets and so provide overall greater cardiovascular protection.

Both the United States and European hypertension guidelines have embraced this strategy and recommend initiating treatment with a 2-drug combination in those patients in whom a single drug is unlikely to achieve the target blood pressure. Importantly, the Unites States Food and Drug Administration has agreed with these concepts, and a number of 2-drug fixed-dose combinations have recently been approved for initial therapy by this agency. Although there is a lack of objective evidence to confirm the outcomes benefits of this approach, the arguments that early blood pressure responses are predictive of long-term control rates,23 and that the use of modern combinations as initial therapy appears to be safe and well tolerated, support these new trends.

The most common combinations now prescribed use ACE inhibitors or angiotensin receptor blockers together with low doses of thiazides. However, the recent Avoiding Cardiovascular Events Through Combination Therapy in Patients Living With Systolic Hypertension (ACCOMPLISH) study comparing amlodipine with hydrochlorothiazide, each combined with a renin-angiotensin system blocker, reported significantly greater outcomes benefits with the calcium channel blocker combination despite similar blood pressure effect.28

Combining ACE inhibitors and angiotensin receptor blockers has never been proposed as an optimal hypertension treatment (despite some benefits demonstrated in patients with chronic heart failure or nephropathy). The recent Ongoing Telmisartan Alone and in Combination With Ramipril Global Endpoint Trial (ONTARGET) findings indicated that this type of combination in maximum doses might have some deleterious renal effects when given to patients at high cardiovascular risk,29 thus making it an even less attractive alternative. It can be noted, though, that the combination of an angiotensin receptor blocker with a renin inhibitor is effective for blood pressure reduction and so could be considered for appropriately selected hypertensive patients.

Resolving Uncertainties

There is a clear need to address remaining questions regarding hypertension management. In particular, future research should be directed to the following areas.

1. How to incorporate non–blood pressure risk factors into hypertension treatment decisions

Although experts acknowledge the heightened probability of cardiovascular events when more than one risk factor exists, guidelines in the United States and Europe differ in how to deal with this situation. The JNC recommendations for initiating antihypertensive treatment are based primarily on blood pressure values regardless of concomitant risk factors (although, of course, appropriate management of lipid and glucose abnormalities and other disorders is strongly recommended). In contrast, the European guidelines take non–blood pressure findings into account when deciding on whether to initiate treatment for hypertension. A recent report from an authoritative group of experts in the United States has also noted that the diagnosis, staging, and treatment of hypertension should be influenced by concomitant cardiovascular findings and risk factors.30 Clearly, it would be of practical value to develop an evidence-based formula to integrate these factors so that physicians (and recommendations writers) can be guided more specifically in their decision making.

2. What is the best strategy for dealing with stage 1 hypertension?

Because clinical outcomes trials in hypertension have not provided definitive data in patients with untreated blood pressures in the range 140–159/90–99 mm Hg, there is uncertainty about whether these patients would benefit from treatment. Even though the European experts take the view that the presence of previous events or risk factors compels the rapid initiation of drug therapy in such patients, there are still many people with stage 1 hypertension in whom the elevated blood pressure is their predominant finding.
In seeking evidence for such patients there is a fundamental research problem: since guidelines already recommend treatment for stage 1 hypertension, a placebo-controlled trial to measure the potential outcomes benefits of active therapy would likely be regarded as unethical. As well, the relatively low event rates in this milder form of hypertension would require a costly long-term trial of large numbers of patients. Thus, the recent reappraisal report from Europe1 suggests that such a trial use intermediate endpoints (measurable changes in such findings as microalbuminuria, left ventricular mass, arterial stiffness or carotid wall thickness or plaque volume) as surrogates for hard endpoints.

3. Do patients with diabetes, chronic kidney disease or evidence of cardiovascular or stroke disease benefit from more aggressive blood pressure criteria?

At present, the guidelines recommend a blood pressure of 130/80 mm Hg as the diagnostic and treatment-goal blood pressure threshold for patients with these conditions. But, as discussed earlier, there is no evidence to support this particular blood pressure value. The anticipated Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial in patients with diabetes will compare the relative outcomes merits of treatment to targets of <120 or <130–139 mm Hg.27 If <120 mm Hg is superior in outcomes prevention to <140 mm Hg, then it could reasonably be inferred that lower is better for such patients; but if there is not a meaningful difference, we might not even be confident that <130 mm Hg is more protective than <140 mm Hg. So, depending on the results of this important trial, it may be necessary to do additional focused research in patients with diabetes, and in any case to further explore optimal blood pressure criteria in patients with kidney and cardiovascular diseases.

4. Do lifestyle strategies, even when they reduce blood pressure, provide clinical outcomes benefits?

It has been assumed that reducing blood pressure, however it is achieved, will provide outcomes benefits in hypertension. Nevertheless, given the practical difficulty of maintaining long-term blood pressure reductions with lifestyle strategies, it would be valuable to test (either by major endpoints or by intermediate measures) whether patients with prehypertension or stage 1 hypertension, when randomized to treatment with lifestyle measures or to drug therapy, have similar clinical outcomes rates when achieving the same blood pressure reductions.

Making Progress

The reappraisal report has made a valuable contribution to our understanding of hypertension and its management. It is comforting to recognize that, for the most part, our contemporary strategies for dealing with hypertension have been very reasonable, even if not fully supported by hard evidence. Obviously we need more data to help guide our decision making, particularly in hypertensive patients with conditions like diabetes, chronic kidney disease, and cardiovascular disease.

The rigorous writing of the European authors not only puts their own guidelines into perspective, but should also prove helpful to the experts in the United States who have undertaken the task of producing the forthcoming hypertension recommendations for this country.