Contributions to Hypertension Public Policy and Clinical Practice: A Review of Recent Reports

The clinical practice of hypertension continues to change. This reflects new research on the basic mechanisms of this condition, a broader understanding of therapeutic strategies, and a growing realization that hypertension is of critical importance worldwide as a cause of cardiovascular events and premature mortality. The definitions of hypertension are part of this change. During the past 40 or 50 years, the systolic blood pressure (SBP) thresholds for diagnosing hypertension and setting treatment targets have ranged from 160 mm Hg all the way down to a potential value of 120 mm Hg, assuming that the recent Systolic Blood Pressure Intervention Trial (SPRINT)1 will influence future hypertension practice.

There is uncertainty regarding the best way to measure blood pressure (BP) in clinical practice: Should we remain with the conventional sphygmomanometer and stethoscope, or should we now depend on the newer automated oscillometric devices? Has ambulatory BP monitoring (ABPM) become the new gold standard for diagnosis? Or should we start to rely more on home BPs?

There is a belief among some experts that BP, although important in itself, is also a surrogate measure of underlying arterial disease, so might it not be more appropriate to diagnose hypertension, and judge the success of its treatment, by depending on measures of vascular disease such as pulse wave velocity? And, from a public health point of view, might it not be possible to make a huge population‐wide impact on the problem of hypertension by introducing major lifestyle changes into our communities, particularly reductions in sodium intake?

We anticipate that SPRINT1 and other recent clinical trials and analyses will form the basis for new hypertension practice guidelines being prepared in the United States, Europe, and other major centers throughout the world. Certainly, there will be a call for the international organizations engaged in hypertension to provide guidance and advice on how best to manage hypertension, including countries and communities with severely limited resources. It is interesting, even as this important work gets underway, to look at recent publications by distinguished hypertension experts, many published within the pages of this Journal, which can give some sense of where the practice of hypertension will be heading during the next several years.

BP and Public Health Perspectives

The importance of hypertension has recently been highlighted by the World Health Organization (WHO).2 Of striking interest, for the first time a noncommunicable disease—hypertension—has been recognized as a primary cause of major illness and premature mortality across the globe. Among major efforts being made to address this problem is the Global Standardized Hypertension Treatment Project,3, 4 which aims in an immediate fashion to recommend simple and achievable ways to control BP across a wide range of economic and social conditions.

The World Hypertension League (WHL) is one of the leading organizations involved in addressing global needs for broad and achievable strategies for hypertension management. The membership of the WHL comprises almost 100 national hypertension societies, associations, and partners. In addition, the WHL represents over 50,000 healthcare providers directly and over 100,000 indirectly through its communication network. This provides a powerful base for WHL's authoritative voice dealing with major aspects of hypertension policy, diagnosis, and management.

This Journal is the official journal of the WHL and so its reach is now global in scope and plays a major part in the world's hypertension literature. The Journal of Clinical Hypertension has provided the forum for several authoritative policy statements and position papers by the WHL dealing with critical aspects of public policy and clinical practice in hypertension.5, 6 Clearly, it is difficult to make progress in this work unless there are reliable data to guide plans for addressing hypertension globally, and so the report by the WHL setting out standards for the uniform reporting of hypertension in adults using population survey data stands as a powerful recommendation to individual countries regarding such core indicators as BP distribution, prevalence of hypertension, awareness of this condition, antihypertensive drug treatment, and control of hypertension.7 Hopefully, this standardized approach to collecting data will provide critical information that will allow public health authorities as well as practitioners to optimize the care of hypertensive patients.

One of the most important trends in hypertension, in common with other conditions, is the growing proportion of patients aged 80 years and older. There has been a rapid increase in the number of older people with hypertension,8 and, compared with younger age groups, these individuals have a substantially greater risk of major cardiovascular outcomes.9 In a major analysis of these trends in the United States, public health experts at the University of Alabama at Birmingham have reported that the critical areas of awareness, treatment, and control of hypertension, as well as the use of appropriate multidrug treatment regimens, have all improved progressively during the period from 1988 to 2010.10 It will be very interesting to see further reports in this group of patients aged 80 years and older to determine whether even in the few years since 2010 there has been even more progress. Major clinical trials such as SPRINT1 and the Hypertension in the Very Elderly Trial (HYVET)11 have emphasized the powerful cardiovascular protective benefits of antihypertensive therapy in this age group.

BP: Measurement

All of us who have worked in the field of hypertension know how inconsistent BP readings can be, even when they are taken carefully within the same patient. This problem is magnified when we consider how best to bring hypertension diagnosis and care into communities with limited resources where BP measurements are often made by nonprofessional health workers. This concern has led to a policy statement by the WHL on noninvasive BP devices.12 The WHL expert panel that issued this statement emphasized the importance of achieving consistency in measurements and the fact that the traditional auscultatory method for measuring BP is difficult to teach and perform.13 Thus, this panel has recommended a concerted effort to validate semiautomated and fully automated oscillometric devices that are reliable and—of critical importance in many areas of the world—affordable.12

There is by now a strong awareness of the so‐called white‐coat effect, in essence the increase in BP produced simply by the fact that a professional person such as a doctor or nurse is measuring a patient's BP. Dr Martin Myers and his colleagues have been highly influential in developing methods for measuring BP in the clinic or office that minimize this effect and so provide BPs that are representative for each patient. In an expert commentary,14 Dr Myers has discussed practical ways to minimize the white‐coat effect or “human factor.” He points out that the most effective method for achieving a meaningful BP reading is to use automated devices in a setting where patients are “nonobserved”—in other words, without a doctor, nurse, or other professional person in the room.4 In fact, SPRINT based its method for measuring BP on this principle of nonobserved automated readings, a strong acknowledgement of the recognition of this approach. Of course, it will become necessary for us to find ways of translating BPs measured in this fashion into values that correspond to those normally obtained by “conventional” office or clinic readings. It is a matter of concern that there will be confusion in interpreting BP values from clinical trials such as SPRINT until there is broad agreement on exactly how BPs should be measured.

Nighttime BP may be particularly predictive of future cardiovascular events. Clearly, to obtain meaningful BPs during the hours of sleep it is necessary to use techniques such as ABPM. It has been demonstrated, using ABPM, that cardiovascular outcomes such as silent cerebral infarcts15 or left ventricular hypertrophy16 are best predicted by nocturnal BP values. A more recent study by Chinese investigators has emphasized that nighttime BP is strongly associated with the risk for hemorrhagic strokes, and that patients whose BPs fail to show the normal fall or dipping pattern at night are particularly susceptible to these major events.17 These data, of course, raise the issue of how best to identify such patients and how to most effectively protect them; inevitably, there has been much interest in dosing antihypertensive drugs late in the evening in order to provide maximum BP reduction and cardiovascular protection during the nighttime hours.

Heart rate is usually measured with BP but hasn't received as much attention as a predictor of outcomes. Even so, there is a good basis for believing that high heart rates, perhaps indicative of excess sympathetic activity, might be independently associated with events.18, 19 In trying to better understand this relationship, it is interesting to note that a recent publication showed that nighttime heart rate was related to important inflammatory markers such as leukocyte counts and C‐reactive protein levels.20 These mechanisms that might possibly underlie the increased risk of adverse outcomes deserve further exploration.

Markers of Cardiovascular Outcomes in Hypertension

There are several clinical findings that are produced by hypertension and can serve to predict major outcomes. For instance, findings such as left ventricular hypertrophy, increased intima‐media thickness of the carotid vessels, microalbuminuria, and reduced glomerular filtration rate (GFR) are highly suggestive of future cardiovascular outcomes. Recently, there has been a focus on the variability of BP from visit to visit during the treatment of hypertension, termed visit‐to‐visit variability (VVV); likewise, one of the best markers of arterial stiffness is pulse wave velocity, most commonly measured between the carotid and femoral arteries (cfPWV) or between the brachial and ankle vessels (baPWV).

It has now been well described that an increase in VVV across several clinic visits is associated with increased fatal and nonfatal coronary and stroke outcomes.21, 22 Dr Paul Muntner and colleagues23 have reported the effects of differences in VVV across five to seven visits during the Antihypertensive and Lipid‐Lowering Therapy to Prevent Heart Attack Trial (ALLHAT). In particular, these investigators looked at the effects of the primary drugs being examined in ALLHAT, namely chlorthalidone, amlodipine, and lisinopril. Compared with chlorthalidone, amlodipine had a lower VVV (as measured by the standard deviation of the mean BPs across the visits), whereas lisinopril had a higher VVV. Potentially, these findings could explain some of the differences in outcomes between these agents during the ALLHAT trial, and so might be a useful tool for further determining the effectiveness of therapeutic interventions.

A similar approach looking at the variability of BP across visits has been applied by Dr John Kostis and colleagues to the Systolic Hypertension in the Elderly Program (SHEP) trial.24 Performing a 17‐year follow‐up of mortality from the trial (using the United States National Death Index), these investigators showed that VVV was associated with mortality and appears to be particularly predictive of adverse events in actively treated patients (with chlorthalidone); it was less predictive in the placebo group. The investigators suggested that the variability between visits could simply be accounted for by a lack of consistency by patients in taking their medication, although a finding of increased VVV could also reflect the effects of large‐vessel stiffness.25

There have been differing estimates of the effects of VVV on renal function. In one trial, it was shown that in patients with nondiabetic chronic kidney disease there was a clear relationship between increases in VVV and declines in renal function.26 However, in a later study, the same authors in a study of patients with diabetic CKD followed over 12 consecutive visits found that there was not a clear relationship between VVV and a decline in renal function.27 It is becoming progressively more apparent that there are differences between diabetic and nondiabetic patients in their vascular responses to BP, and so the apparent inconsistency between studies in nondiabetic and diabetic patients should not be too surprising.

Pulse wave velocity (PWV) is now regarded as a useful and relatively easy‐to‐measure index of arterial stiffness; stiff arteries tend to produce increased PWV and have been associated with declining renal function.28 However, one recent trial looked at both cfPWV and baPWV and showed that while both of these indices increased as GFR declined, these relationships were not significant.29 Interestingly, the pulse pressure (difference between systolic and diastolic BPs), which can also serve as an estimate of large artery stiffness, was significantly associated with declining renal function in the large 903‐patient cohort that was studied. Clearly, we are still searching for useful intermediate or surrogate predictors to help guide evaluation of renal function, but it seems at this time that neither VVV nor PWV are necessarily the most useful measurements of changes in the kidney.

It is believed that increased PWV as a predictor of arterial pathology is influenced by changes in the arterial wall including intimal thickening calcium deposition, endothelial dysfunction, and increased laying down of collagen.30 To explore the relationship between arterial stiffness and coronary disease, investigators studied patients scheduled for coronary artery bypass grafting.31 The PWV was measured with an oscillometric device, which demonstrated a significantly greater PWV in the coronary disease patients when compared with healthy volunteers. This study also found that PWV was predictive not only of the presence of coronary disease but also gave some quantitative estimate of its severity.

An obvious question is whether therapy for hypertension can reduce PWV and provide potential evidence for a reduction in arterial stiffness. It is known that during exercise there is an increase in PWV, which perhaps is greater in people with hypertension or at risk for hypertension than in healthy controls.32 In a new report that studied untreated young patients with hypertension, the greater increase in PWV that they experienced, compared with healthy controls, could be prevented by the angiotensin receptor blocker, valsartan.33 There is growing literature documenting the effects of different therapies on arterial stiffness, and the possibility that PWV could be a powerful guide in determining the effectiveness of hypertension treatment deserves further exploration.

It is noteworthy that PWV as a routine measure of arterial stiffness in hypertension has now been recommended by a consensus guideline panel.34 There has been a desire to simplify the measurement of PWV, and the possibility that sophisticated oscillometric devices could be a substitute for the more demanding methods used previously is attractive. One such approach tested an oscillometric device in each of the four extremities and showed that such measurements in each of the extremities individually, or combined into a sum of all four, has a close relationship with PWV obtained by direct arterial measurements in patients undergoing cardiac catheterization.35 It is likely that an increasing range of relatively easy‐to‐use devices will become available for clinical use in the next few years.

Another marker of cardiovascular change in hypertension, known for many years but not often considered recently, has been electrocardiographic evidence of left atrial (LA) enlargement. It has been clearly demonstrated that evidence for LA enlargement that occurs before left ventricular enlargement is an early sign of hypertrophic heart disease.36 Not surprisingly, a recent study of hypertensive patients observed that LA enlargement was more common than LV enlargement.37 This study also noted that an increase in LA size (by echocardiography) was closely associated with heart failure with preserved ejection fraction, which is a common form of heart failure associated with hypertension. The same observers noted that this relationship was exaggerated in patients who had evidence of metabolic syndrome. This report speculated that evidence for LA enlargement could be an important early indicator of heart failure in hypertension. This could be a very simple and useful clinical tool for picking up early evidence of target organ involvement in patients with hypertension.

Salt: Policy Issues

Dealing with salt science and salt policy has become a controversial and difficult task. There is general agreement that excess levels of salt are associated with higher BP and an increased risk of cardiovascular events, but it is not clear at which threshold salt intake becomes a meaningful health problem. The issue has been further confused by the fact that at the low end of dietary salt intake, at levels well below 5 g daily, there may be a tendency for an increase in adverse outcomes. This apparent danger of low salt intake, however, may be a manifestation of reverse causality and simply reflect the fact that people with advanced and potentially fatal illness may be consuming markedly reduced food intakes and thus ingesting low levels of salt. Other problems that add to the complexity of this area are related to the fact that measuring salt intake in individual people is difficult; the standard approach for measurement is a 24‐hour urine collection for sodium, but this technique is often inadequately performed. In addition, even when this procedure is properly conducted, the measurements of salt can vary quite widely from day to day despite the fact that dietary salt has supposedly been reasonably constant. Nevertheless, there is a strong belief among public health scholars that, on a population basis, a general reduction in salt intake would be associated with a BP reduction across the population and so presage a reduction in cardiovascular outcomes.

Because of the perceived lack of rigor in salt studies,38, 39 the WHL in collaboration with other major organizations such as the American Heart Association, British Heart Society, International Society of Nephrology, the World Action On Salt and Health, the Canadian Stroke Network, and the World Stroke Organization published a strong statement on salt research, calling for greater rigor and globally accepted standards for future studies. In particular, the statement drew attention to improved methods for quantifying salt intake in individuals and populations, the measurement and ascertainment of clinical endpoints, and statistical methodology, including concerns regarding the validity of observational and cross‐sectional studies.40 This statement was supported by a commentary by Dr Lawrence Appel, who argued that salt remains one of the major underlying issues in hypertension.41, 42 Dr Appel argued that dealing with excess dietary salt may be even more important than the use of drugs for the treatment of hypertension since a decrease in dietary salt when applied to large populations could reduce the progression from prehypertension to hypertension that occurs with age.

Consistent with this thinking, the WHL, in collaboration with the International Society of Hypertension (ISH), published a call to governments, nongovernment organizations, and the food industry to reduce the salt content of processed foods. This approach was first articulated by the WHO, which made the point that this approach could be particularly cost‐effective in low‐ and middle‐income countries.43 The exhortation by the WHL and ISH in many respects has become a major thrust of these organizations in their plans to deal with the global issue of hypertension.44

In addition to these efforts by the WHL and the ISH, the WHO has also recommended a 30% reduction in dietary salt by 2025.45 In keeping with this plan, the WHL, in a follow‐up policy statement dealing with targets and timelines for reducing salt in food, noted that several countries––including Great Britain and several South American countries––had already set salt reduction targets into their national health policies for processed foods. The WHL went on to argue that the key to success is not only a reduction in the amount of sodium in processed foods but also clear and accurate labeling that can serve to guide individual consumers as well as health workers. The WHL also made the valid point that this campaign would be most successful if multiple countries could harmonize their efforts, thus making it far easier to share experiences and develop strategies.46

Original Research on Salt and Hypertension

It has been believed that reducing sodium intake47 and increasing potassium intake48 will decrease BP. However, a recent study based on 7000 adults without previous hypertension included in the National Health and Nutrition Examination Survey 2001–2006 survey found no association between BP and either dietary sodium or potassium.49 This finding does not necessarily mean that efforts to change sodium and potassium might not affect BP, but it seems to indicate a heterogeneity in the population, possibly suggesting that individuals susceptible to excess salt or inadequate potassium have already developed a manifestation of hypertension, or alternatively, that those who, for whatever reason, have become hypertensive may then be more responsive to sodium and potassium.

In considering the role of dietary salt in hypertension, it would be of interest to focus on patients with prehypertension since they appear to be susceptible to early evidence of coronary atherosclerosis.50 More recently, a longitudinal study in prehypertensive patients with a high salt diet (≥6 g/d) were more than 50% more likely to become hypertensive and twice as likely to have a cardiovascular event as people with a lower salt diet.51 It should be noted that most patients in that study were 60 years and older, suggesting that older people may be particularly vulnerable to the effects of salt intake. This is particularly interesting since it is known that hypertensive patients have a different taste perception of salt, perhaps explaining why they tend to eat greater quantities.52 In fact, it has been shown in patients 70 years and older, when asked to indicate a preference for bread with high, normal, or low salt content, that they were more likely than normotensive individuals to express a preference for the high salt choice.53 This suggests that salt intake is not just a chance event, and physicians should consider the possibility of salt‐seeking behavior in their older hypertensive patients.

It is generally believed that in Asian countries there is a tendency to greater amounts of salt in the usual diet. In India, for instance, it is known that as many as 40% of people ingest 12 g or more of salt daily.54 Could this be an explanation of why people in South Asia are so vulnerable to cardiovascular disease? This may be further exaggerated according to a study which showed that hypertensive patients in India had significantly greater average 24‐hour sodium intakes (measured by urine collection) than healthy controls.55 Once again, this raises the possibility that hypertensive patients may simply have a “salt preference” when compared with people with normal BPs.

Clinical Practice and Guidelines

In early 2014, the panelists appointed to the Eighth Joint National Committee (JNC 8) published their recommendation for BP targets in the treatment of hypertension. This recommendation was controversial since it indicated that for people 60 years and older, clinical trial evidence best supported a systolic treatment target of <150 mm Hg.56 This recommendation was strongly in variance with that of other guidelines, including that of the American Society of Hypertension/ISH Clinical Practice Guidelines, which concluded that the well‐established target of 140 mm Hg should remain for all adults with hypertension.57 Clearly, these differences between guidelines were important and there were strong disagreements among experts from different countries who contributed to a special issue of this Journal (April 2014) and contrasted their own national recommendation with those of JNC 8. A summary of these arguments was published in the same issue.58 Of course, the debate on whether 140 mm Hg or 150 mm Hg should become the target for SBP soon became moot because the SPRINT trial, published only a year later, seemed to make it clear that even in older patients (including many aged older than 75 years) the desired office‐based target should be <130 mm Hg (by usual office measurements). Clearly, this is a dramatic difference from the previous recommendations and one that almost certainly in the months to come (late 2016, early 2017) will affect the way practitioners manage hypertension.

Perhaps the most neglected patients with hypertension are children and adolescents. In general, hypertension in these young people is diagnosed according to a complex nomogram that takes into account age, height, and weight and gives a diagnosis of hypertension if a child's BP exceeds the 95th percentile. However, in a statement on pediatric hypertension, the US Preventive Services Task Force (USPSTF) stated: “Current evidence is insufficient to assess the balance of benefits and harms” of screening for high BP in asymptomatic children to prevent heart disease.59 This seemed like a harsh position to take, in essence suggesting a potential abandonment of children with high BPs simply because there has been inadequate clinical outcomes evidence to support any kind of intervention. Even so, some pediatric experts were supportive of this statement, arguing that we really do not have adequate information about hypertension in children and what kind of complications this condition might lead to in adulthood.60 The expert article went on to call for detailed longitudinal studies to assess the long‐term consequences of BP abnormalities in childhood. In a sense, we already do have some important data on pediatric hypertension. For instance, it is known that key surrogate endpoints such as left ventricular hypertrophy, increased carotid intima‐media thickness, and albuminuria can be demonstrated in children with hypertension. Bearing in mind the very high cost of long‐term randomized studies, not to mention the logistical challenges of multiyear clinical trials, may make such research close to impossible and support further consideration of studies utilizing surrogate outcomes. Other experts, also weighing in on the USPSTS recommendation, were concerned with the nihilistic approach of this statement.61 They pointed out that the cost of BP screening in children is relatively trivial since it is now regarded as a routine part of interactions between pediatricians and children, and that failing to obtain an adequate BP data base could result in many young patients with hypertension being completely overlooked.

One of the major factors in childhood hypertension is obesity, including in children younger than 10 years.62 A study of obese children aged 2 to 9 years, compared with normal‐weight children, found higher SBPs, higher diastolic BPs, and a significantly higher prevalence of the lowest quintile of high‐density lipoprotein cholesterol.63 Despite the negative attitude of the USPSTF, these findings should also add to the consideration of careful lifestyle changes in such children even in the absence of randomized controlled clinical trial evidence. This difficult and controversial area will likely be a battleground for some time to come, because surely it is in children and young adults where there might be the greatest opportunity to favorably affect the lifelong trajectory of vascular disease.

Treatment‐Resistant Hypertension and Renal Denervation

Renal denervation is a technique based on the premise that ablating the renal nerve will result in reduced sodium retention and reduced release of renin by the kidneys and a reduction in systemic sympathetic activity. This procedure is usually accomplished by advancing a catheter into the renal arteries and emitting radiofrequency or ultrasound energy across the walls of the arteries to destroy the renal nerve fibers that track closely to the outside walls of these vessels. Early reports of denervation in patients with uncontrolled hypertension indicated powerful BP reductions.64 However, subsequent experience has been less impressive. An important paper published in this Journal showed a sharp heterogeneity in BP results in patients subjected to this technique.65 About half the patients had strong results (averaging 20 mm Hg reductions in SBP), but other patients had far poorer results, including some patients whose pressures actually rose (presumably because these patients discontinued previous antihypertension medications). The big question arises: How can we predict which patients are going to get good results? Currently, ongoing clinical trials with more advanced catheters, and using more aggressive techniques, may expand the benefits of this procedure. We should know more within the next 12 months.

One of the concerns with renal denervation is the possibility that putting heightened energy across the wall of the renal artery could cause damage to the artery wall. One method for examining renal arteries noninvasively is to use the innovative technique of 64‐detector spiral computed tomography to perform renal arteriography.66 Investigators were able to measure the diameter, length, presence of atherosclerosis, and plaque in renal arteries before and after renal denervation.67 There was no change in patients undergoing denervation in any of these measurements 12 months after the procedures, although, interestingly, a control group of patients not subjected to renal denervation had evidence for increased plaque burden. These investigators concluded that this procedure could reduce BP without meaningful damage to renal arteries. In another look at renal function in patients undergoing renal denervation, it was shown that this procedure was effective in decreasing BP in patients with chronic kidney disease.68 In a follow‐up study, it was demonstrated that success in controlling BP with renal denervation might be an important determinant of renal function: Patients whose BP was controlled by denervation actually had significantly higher estimated GFR at 3 monthly time points up to 1‐year postprocedure (averaging 19 mL/min/1.73 m²) when compared with patients whose BPs were not controlled.69 It is not clear whether the improvement in BP might have been the cause of the improved renal function, or whether these were two independent but desirable effects of the intervention.

A Brief Summary

It is interesting that the emphasis in hypertension during the past couple of years has swung from a focus on pharmaceutical studies to the public health and clinical practice aspects of this condition. The declaration by the WHO that a noncommunicable condition such as hypertension is a major cause of premature mortality and cardiovascular disease globally has provided a major impetus. The WHL, often utilizing the pages of this Journal, has taken the lead in setting worldwide standards for many aspects of hypertension care, ranging from a call for improved BP measurement techniques to a focus on improving research methodologies in salt science as well as a call for public agencies and food manufacturers to start systematically reducing the sodium content of processed foods. Important clinical studies are leading the way to new hypertension practice guidelines, which, almost certainly, will call for more aggressive control of BP in all adult patients, regardless of ethnicity and age. For those of us committed to the field of hypertension, the near future promises important new developments in an area of medicine that is of critical importance throughout the world.

Disclosures

None.