Author + information
- Alan H. Gradman, MD∗ ()
- ↵∗Reprint requests and correspondence:
Dr. Alan H. Gradman, 1239 Shady Avenue, Pittsburgh, Pennsylvania, 15232.
The study by Bangalore et al. (1) that appears in this issue of the Journal addresses the most pressing topic in antihypertensive therapy today: determining the optimal thresholds and targets for treatment in various segments of the hypertensive population. The paper is a direct response to the recent publication of the 2014 Evidence-Based Guidelines for the Management of High Blood Pressure in Adults, authored by panel members of the Eighth Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC-8), originally appointed by the National Heart, Lung, and Blood Institute, and dedicated to the hypothesis that appropriate guidelines should be based exclusively on data derived from randomized clinical trials (RCTs) (2). After analyzing available RCT data, the JNC 8 panel concluded that the blood pressure (BP) threshold and target for treatment in patients >60 years of age without diabetes or chronic kidney disease should be 150/90 mm Hg rather than 140/90 mm Hg, as recommended in the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC-7). They considered this recommendation to be strongly evidence-based. It was acknowledged, however, that only “low-quality” evidence supported the conclusion that there was no additional benefit to be gained from “setting a goal [systolic] BP lower than 140 mm Hg in this age group.” The panel’s recommendation for less aggressive treatment in older adults has proved to be the most contentious aspect of the recommendations, even among the panel members themselves, some of whom strongly criticized their own paper in a subsequent publication (3).
Treatment of hypertension in the elderly constitutes one of the great success stories in cardiovascular medicine, and it is important not to endanger the gains that have been achieved. Clinical trials in elderly patients with isolated systolic hypertension, as well as those with systolic/diastolic BP elevation, have documented relative risk reductions for stroke, myocardial infarction (MI), and heart failure equal to or greater than those seen in younger individuals (4). In HYVET (Hypertension in the Very Elderly Trial), drug treatment reduced overall mortality by 28%, stroke by 34%, and new-onset heart failure by a remarkable 72% in patients >80 years of age (5). Because age is the most potent determinant of cardiovascular risk, the same magnitude of risk reduction translates into the prevention of many more clinical events per 1,000 patients treated in elderly compared with younger people with hypertension. If the JNC-8 panel recommendations are implemented, approximately 6 million fewer adults in the United States would be deemed eligible for drug therapy, and treatment intensity would be reduced for an additional 13.5 million individuals (6).
The paper by Bangalore et al. (1) supports a lower BP target, specifically in patients with hypertension and coronary artery disease (CAD). Their data are derived from retrospective analysis of the INVEST (INternational VErapamil SR Trandolapril STudy), which randomized hypertensive patients with CAD to receive atenolol or verapamil-SR (sustained release) as their primary antihypertensive agent. Patients >60 years of age from either treatment group who entered the trial with a baseline systolic blood pressure (SBP) >150 mm Hg were analyzed. Those with an on-treatment BP <140 mm Hg were found to have a lower risk of the combined endpoint of all-cause death, nonfatal MI, and nonfatal stroke than those with on-treatment SBP of 140 to 149 mm Hg or >150 mm Hg. After adjustment for baseline patient variables, cardiovascular mortality and stroke remained significantly less frequent in patients with on-treatment SBP <140 mm Hg than in those with SBP of 140 to 149 mm Hg. The implication is that the appropriate target BP in this patient population should be <140 mm Hg rather than <150 mm Hg, as recommended by the JNC-8 panel.
It is important to recognize that these results cannot be used as evidence against the JNC 8 panel’s selection of 150 mm Hg as the threshold for treatment. All patients in the analyzed cohort entered INVEST with an SBP of >150 mm Hg, and all would have been treated according to the new panel guidelines. Although thresholds for treatment and treatment targets are often thought of as identical, they are not. In HYVET, patients were eligible for inclusion only if they exhibited a sustained SBP >160 mm Hg, and once randomized, the BP target was 150/80 mm Hg. Determining the proper threshold for treatment in the elderly is particularly challenging, because not only HYVET but all the major clinical trials in elderly hypertensive people have randomized patients with a baseline SBP of >160 mm Hg. Thus, there is little, if any, RCT evidence that treatment of older adults with SBP between 140 and 160 mm Hg actually reduces cardiovascular events (7). This information was critical to the JNC-8 panel’s decision to raise the treatment threshold for elderly patients, although many have questioned application of this less stringent guideline at the relatively young age of 60 years. From a public health perspective, there is a compelling need for a well-designed, placebo-controlled trial to determine the optimal threshold for treatment in this large segment of the adult population.
With regard to selection of the optimal BP targets, the study results are of considerable interest, but they are subject to interpretation. Physicians in INVEST were given specific BP goals: <140/90 mm Hg for “uncomplicated” patients and <130/85 mm Hg for those with diabetes and/or renal impairment. Although 57% of patients achieved <140 mm Hg, 43% did not, despite having received the same initial treatments and having had physicians who were given identical instructions regarding BP management. In effect, the authors have compared the prognosis of “responders” to “nonresponders,” using the post-randomization variable of achieved on-treatment BP as a measure of response. There is considerable evidence that response to treatment is itself a function of patient characteristics, known or unknown, that may independently influence prognosis (8). In this study, patients who did not achieve an on-treatment BP <140 mm Hg were older, had a higher baseline SBP, and had a higher prevalence of major comorbidities, including diabetes and renal insufficiency, characteristics known both to adversely affect prognosis and to reduce the response to antihypertensive drug therapy. Although the authors endeavored to statistically correct for many known variables, their data lack the effect of randomization, which serves to equalize all baseline characteristics of a studied population.
In the present study, JNC-7 BP targets were set by protocol, and a predictable heterogeneity of response to treatment was observed. The lower risk exhibited by that portion of the study population that achieved an on-treatment BP <140 mm Hg suggests that treatment should be given to all patients to lower BP into that range. However, the efficacy of using the JNC-7 targets, as was done in this study, must include the sum of all cardiovascular events and adverse experiences that occurred in patients who did and did not achieve those targets. In published analyses of achieved BP data, the authors attempted to identify the optimal on-treatment BP, the “sweet spot” at which cardiovascular risk is minimized. Although such analyses may accurately identify an “ideal” BP, this BP cannot be consistently attained in a real-world population of hypertensive patients. There will always be a spectrum of response, with some patients falling above and some below the ideal BP. To adjudicate between a target of 140 mm Hg and 150 mm Hg, it would be necessary to perform a prospective trial in which the population was randomized to these targets and the overall results were compared.
In the absence of such data, and given the depth and duration of this controversy, it is clear that there is no right answer. As was seen in ACCORD (Action to Control Cardiovascular Risk in Diabetes), more aggressive treatment of nonresponders could be accompanied by an increased number of treatment-related adverse events in this relatively “resistant” segment of the older hypertensive population. In patients with CAD, concern about the J curve and the possibility of an increased risk of coronary events has tempered enthusiasm for aggressive BP reduction, particularly in older patients with low diastolic BP. In the TNT (Treating to New Targets) study, in which 10,001 patients with CAD were analyzed, an on-treatment BP of 146/81 mm Hg was identified as being associated with the lowest risk of a combined endpoint that included MI and stroke (9). However, unlike INVEST, the study was not conducted specifically in hypertensive patients.
My overall sense is that the JNC-8 panel recommendations are reasonable for patients with hypertension and CAD. Retention of the diastolic BP target of <90 mm Hg ensures that patients with significant systolic/diastolic hypertension will be treated regardless of their SBP. For high-risk diabetic patients, the lower target of <140/90 mm Hg is recommended for all patients, regardless of age. For patients with isolated systolic hypertension and an SBP in the 140s, clinical judgment will be required. Patients at increased risk of stroke should probably receive more intensive treatment, because this endpoint, unlike MI, is particularly sensitive to BP reduction. Practitioners should carefully note the corollary recommendation given by the JNC-8 panel that “if pharmacologic treatment for high BP results in lower achieved SBP (e.g., <140 mm Hg) and treatment is not associated with adverse effects on health or quality of life, treatment does not need to be adjusted.” It is worth noting that this corollary recommendation was based not on the results of RCTs but on that imprecise, oft-criticized “expert opinion.” Fortunately, or unfortunately, this will remain an essential component in the clinical care of patients with hypertension and CAD.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Gradman has reported that he has no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Bangalore S.,
- Gong Y.,
- Cooper-DeHoff R.M.,
- Pepine C.J.,
- Messerli F.H.
- James P.A.,
- Oparil S.,
- Carter B.L.,
- et al.
- Bangalore S.,
- Messerli F.H.,
- Wun C.C.,
- et al.,
- for the Treating to New Targets Steering Committee and Investigators