This article is more than 5 years old. Information may no longer be current.
Resistant hypertension: A growing issue, novel devices on the horizon
Click Here to Manage Email Alerts
Treatment-resistant hypertension is a common clinical problem associated with higher risk for CV and renal adverse events, poor quality of life and increased health care costs. The exact prevalence is unknown, but resistant hypertension is projected to become more prevalent with the growing elderly population and rising obesity rates.
Keith C. Ferdinand, MD, addresses challenges
and recent advances in the treatment
of resistant hypertension.
Images: Marty Perlman
The introduction of novel catheter-based devices to decrease sympathetic nervous system activity is growing. A 2013 European Society of Cardiology consensus document stated that “renal denervation can be considered as a therapeutic option in patients with resistant hypertension whose blood pressure cannot be controlled by a combination of lifestyle modification and pharmacological therapy according to current guidelines.”
Cardiology Today assembled a group of experts at the American Heart Association Scientific Sessions in November to tackle the topic of resistant hypertension (RHTN). In this round table, they discuss definitions, goals and current treatment options to expand the understanding of RHTN and ultimately improve long-term clinical management.
Editor’s note: This round table took place at AHA 2013 before the January press release from Medtronic about results from the SYMPLICITY HTN-3 trial.
Round Table Participants
Moderator
Carl J. Pepine, MD
Deepak L. Bhatt, MD, MPH
Keith C. Ferdinand, MD
Manesh R. Patel, MD
Elijah Saunders, MD
Current definitions of hypertension
Carl J. Pepine, MD: To begin, let’s discuss current definitions of RHTN for our readers.
Keith C. Ferdinand, MD: RHTN has a variable definition. The JNC 7 mentions persons with a BP that is not controlled, despite three medications, at full doses, and optimally one of those drugs should be a diuretic. The 2008 AHA scientific statement, Resistant Hypertension: Diagnosis, Evaluation and Treatment, has a different definition: Persons who can be on four medications and the BP can be controlled, but it is still considered RHTN.
Manesh R. Patel, MD: I use the JNC 7 definition most. Dr. Ferdinand, how do you differentiate between resistant, refractory and uncontrolled?
Ferdinand: “Uncontrolled” is a general term used for persons without a BP less than goal, which we acknowledge is somewhat arbitrary. As an example, I use ≥140 mm Hg/90 mm Hg. Uncontrolled BP could be related to lack of adherence; use of a combination that is not pharmacologic optimal (beta-blocker with a central agonist); and/or consumption of large amounts of sodium, which overrides the benefits of renin-angiotensin-aldosterone system (RAAS) blockers. We use the JNC 7 definition or we use the AHA definition. “Refractory” usually means you’re doing everything right: adhering; controlling sodium; no white-coat hypertension; and so on. Yet BP remains uncontrolled.
Patel: It is an important distinction because as we think about therapies for RHTN and strategies and health systems to improve BP, we will likely see all of these groups. Yet there is overlap between those that are resistant that may still be refractory.
Pepine: You have touched on an important area for our readers. The AHA consensus document simply modified the decades-old definition popularized by Tarazi and Gifford in the 1970s.
Carl J. Pepine
However, as you note, the definition of RHTN remains operationally difficult and is variable in the literature. Also, much of this was from the pre-RAAS-blocker, statin and MI-reperfusion era. Others include “drugs of different pharmacologic classes,” and some also use “BP >130/80 mm Hg for patients with diabetes or CKD” while others specify patients treated with “[more than] three agents (including a thiazide) who had ≥1 BP measurement above target.” Acelajado and colleagues recently defined RHTN “as a BP that remained uncontrolled after ≥3 visits to a HTN clinic with a minimum 6-month follow-up”. As it was not possible to assure adherence, lack of measurement error, lack of white-coat hypertension and so on, “apparent treatment-resistant hypertension” is becoming popular to define the patient “not reaching goal BP on three drugs.”
Our readers should understand that we all want the RHTN definition to be a useful marker for high risk.
Barriers to care
Pepine: Practically speaking, it seems like these barriers have been around for ages and unless we have new tools or strategies for treatment of RHTN we will not be able to move forward.
Deepak L. Bhatt, MD, MPH: You are absolutely right. Often, advances in science move fields forward. Sort of like with pulmonary hypertension: It has been around forever, but there has been a lot of excitement and innovation because now some drugs are available that can actually make patients feel better. We are in a similar situation with RHTN. We have knowledge about what to do, but it may not be widely disseminated beyond hypertension experts. What has changed now is the potential to bring novel interventional techniques to the care of a subset of RHTN patients.
Elijah Saunders, MD: In our clinic, until we have given a trial of spironolactone on top of standard drugs including a diuretic (unless there is a contraindication), we do not call them resistant.
Pepine: Dr. Saunders, that is a good point that leads us to the pathophysiology of RHTN. Dr. Ferdinand, is the pathophysiology of RHTN different from that of “garden variety” essential hypertension?
Ferdinand: Primary hypertension for the most part is related to obesity, increased sodium, decreased potassium, increased alcohol, psychosocial stress and/or disadvantaged status. The Institute of Medicine says if we change those environmental factors we could decrease the hypertension burden.
With RHTN, the person has similar pathophysiology, but now has remodeling of the arterioles, left ventricular hypertrophy, chronic kidney disease, increased age and perhaps diabetes, and these factors superimpose on the various lifestyle factors already discussed. After 15 to 20 years, it is likely that such a person will become resistant.
Pepine: What about the central nervous system?
Ferdinand: Sympathetic stimulation is clearly a part of it, as Dr. Bhatt noted when he mentioned interventions.
Saunders: There are some patients in whom we cannot detect organ damage, LV hypertrophy, CKD and so forth, and they are also resistant. We all have these patients — you give every drug in the book, they are compliant with a diuretic, spironolactone, etc., but they do not respond. Yet, they don’t have a lot of LV hypertrophy or detectable renal damage. I suspect there may be some pathophysiology at the cellular level that we cannot measure.
Ferdinand: Do you think that sometimes you’re missing white-coat hypertension, patients who do not have true, sustained hypertension? Have you found that persons with RHTN and still have no target organ damage benefit from ambulatory BP monitoring?
Saunders: Rarely, but sometimes. They usually will have target organ damage. But I do see such patients with true RHTN. It’s a group that we need to know more about.
New advances in the field
Pepine: What has stimulated this topic recently is the introduction of devices designed to interrupt the renal sympathetic efferent and afferent nerve activity to alter BP. Although the ESC published a consensus document on this intervention, colleagues tell me that it is not yet widely accepted in Europe.
Patel: I’ve heard the same, because a number of these devices are approved in Europe for clinical use. In many countries, it’s not about restrictions on reimbursement. The real rate-limiting factor, from what European interventionalists tell me, is that referring physicians are skeptical of the data and await results from the SYMPLICITY HTN-3 trial. They want more rigorous data in terms of randomized controlled trial design to show the actual magnitude of benefit and full characterization of potential risks and adverse effects, namely renal artery stenosis. There is concern that prior reports may overestimate the benefit. There hasn’t been the widespread diffusion of this technology in Europe that I would have predicted.
Bhatt: European interventionalists I’ve spoken with are enthusiastic about the procedure and state that in their own experiences the results are good. But, the bottleneck has been that referring physicians, whether hypertension experts or primary care, are nervous about the strength of data and magnitude of effect. They are just waiting for more data.
Ferdinand: The original 32-mm Hg/11-mm Hg BP decreases from the first reports have not been reproduced in more rigorous studies. Another reason is recognizing that if you add low-dose spironolactone, the appropriate diuretic, and appropriate combination of a calcium channel blocker plus RAAS-blocker, a lot of these so-called RHTN patients melt away.
Patel: That’s key. An important fact about RHTN is how rigorous and how standardized the treatment has been. New technologies have a differential acceptance rate, and I think the data in Europe tell us that most patients with hypertension don’t live in the interventional space; they live in primary care. Education and continued data would probably help in those two spaces.
Carl J. Pepine, MD, and Manesh R. Patel, MD, anticipate more data on novel devices for the treatment of resistant hypertension.
Available data on renal denervation
Pepine: Let’s discuss what we know about the available data, short of SYMPLICITY HTN-3. Is it that you disagree that renal denervation is lowering BP or do you disagree with the magnitude of benefit?
Ferdinand: In my opinion, the magnitude of benefit is the question. SYMPLICITY HTN-1, an observational study, suggested a 32-mm Hg/11-mm Hg difference in BP. That was not seen with wider use, and I would suspect that in SYMPLICITY HTN-3, with more rigorous evaluation, we won’t see that same robust BP reduction.
Bhatt: I’m completely blinded to SYMPLICITY HTN-3, so I am speculating, but the points Dr. Ferdinand have made are right on the mark. There is, in my opinion, no question that what has been observed to date is an overestimate of benefit. Some might deal with patients not being optimally treated, not screened for white-coat hypertension, or a variety of other issues.
There are also some basic clinical trial issues. Lack of blinding can introduce bias. If we start a calcium channel blocker, when that patient returns we expect a reduction in BP; if we take the BP and it’s 200 mm Hg/100 mm Hg, we think that has to be an error so we retake the BP. It is human nature to throw out that first “erroneous” measurement, and repeat until a more acceptable value is obtained.
Another issue is statistical regression to the mean. To get into any trial, values must exceed the threshold in the inclusion criteria. If you take patients who were enrolled because of that inclusion criteria and do nothing but measure BP in a week, many will have a lower BP just because it is regressing to the mean BP for that population.
Such issues cannot be accounted for unless trials are blinded and controlled and, of course, randomized, which some of the initial work wasn’t. Potential biases can unintentionally creep into trials and therefore lead to what is an overestimate of the likely true effect of renal denervation.
Pepine: But there is no question that there is an important BP-lowering effect. Is that what I’m hearing?
Bhatt: SYMPLICITY HTN-3 will answer, in the most rigorous fashion, what that effect is. The questions include what the risks are and how much is the BP effect. This is not unique to denervation. Many technologies in cardiology, like the first-in-man drug-eluting stents, had zero restenosis. That being said, if we put it in the context of medical therapies that come on the market for BP reduction, they are sometimes reducing BP by 5 to 8 mm Hg. That is a meaningful difference that improves CV outcomes long term.
Ferdinand: I’m not skeptical that there’s some BP lowering. Also, decreasing the burden of antihypertensive medicines may be beneficial.
Saunders: Another concern is that we’re treating systolic and diastolic BP elevation with the same drugs, yet we know the mechanisms for BP elevation are somewhat different. I’m unsure whether SYMPLICITY HTN-3 will be able to distinguish which patients will respond and whether those with systolic elevation respond relatively better than those with diastolic elevation.
Details of future trials
Bhatt: SYMPLICITY HTN-3 is the US trial designed for FDA approval. It is a trial of more than 500 RHTN patients randomized to renal denervation with the Medtronic Symplicity catheter or a control treatment, which involves renal angiography but not radiofrequency energy to the renal artery. The primary endpoint is 6-month office systolic BP vs. baseline. A secondary endpoint is ambulatory BP monitoring. Additionally, all patients are assessed by ultrasound for renal artery stenosis at 6 months. If non-diagnostic, an MR or invasive angiogram is performed to determine risk for renal artery stenosis. Some case reports have been described, but this is the first systematic controlled evaluation. This should provide characterization of renal denervation on office and ambulatory BP and safety.
At 6 months after the primary endpoint is ascertained, patients initially randomized to controlled treatment can crossover, assuming they still meet inclusion criteria. All patients will be followed for a few years to see about sustained efficacy and complications.
Pepine: What are the differences in SYMPLICITY HTN-4 from previous trials?
Saunders: The patients can come in with BP >140 mm Hg but <160 mm Hg.
Bhatt: The major difference is taking the BP range a little bit lower, which, in fact, I think will make many more patients eligible if the trial is positive.
Special populations
Pepine: Is there any difference in response to renal denervation by sex or race?
Saunders: About 10 patients at my institution are enrolled in SYMPLICITY HTN-3. We are blinded, so we don’t yet know who has responded. However, women seem to be more resistant.
Patel: In the data to date, difference by race has been a limitation. We will not know until SYMPLICITY HTN-3 is unblinded. With this trial, we stratified by race to answer that question.
Other benefits of renal denervation
Pepine: Impressive information from early trials suggests that renal denervation not only lowers BP and pill burden, but reverses metabolic abnormalities, HF, atrial fibrillation and so on. Other than a small signal suggesting ACE inhibitors prevent AF, I am not aware that anything else that controls BP prevents AF.
Ferdinand: Not directly. There have been suggestions, but it is not that clear. African Americans have higher mean BP, early onset, more target organ damage, AF, and so on. I would think that improving the metabolic derangements is real. Small vessel dilation decreases insulin resistance, decreases glucose levels, increases glucose utilization in the muscle, and may help lower lipid levels. In terms of the sympathetic nervous system, at the 2013 Vascular Biology Working Group there was an excellent discussion of how simply laying supine decreases sympathetic discharge, promoting dieresis in HF. So, it’s not unreasonable that blunting sympathetic drive would affect HF symptoms.
Bhatt: An exciting possibility is that renal denervation can affect these other organ symptoms. There have also been beneficial effects on insulin resistance and some small data sets suggest benefits in sleep apnea. Dialing down sympathetic tone, at least in patients at CV risk, sounds appealing.
Pepine: Is there general belief that these are relatively unique effects for the sympathetic denervation, things that do not regularly result with antihypertensive drug therapy?
Ferdinand: Yes, to some extent. We also have baroreflex activation, which may decrease sympathetic discharge, as another way of attacking organ damage.
Elijah Saunders
Saunders: I believe the increased number of companies getting involved with is based upon the notion that there is something beyond BP reduction.
Patel: There are signals of pleiotropic effects, but how much?
Pepine: Surprising to me is the increase in effect over time, as long as a year or more. It would seem to me that if you ablate, like we do for arrhythmias, the pathway is destroyed. So, what is going on here? Is there remodeling of the central nervous system and/or the vasculature?
Bhatt: At this time, it is all speculation. Another point is how much is being ablated: complete vs. partial sympathetic denervation. There is still a lot to be learned.
Patel: A concern had been that re-enervation would attenuate the effects of renal denervation. In 3-year follow-up data that has not been seen. Some data suggest even greater benefit over time. I’m less convinced that’s a real finding; rather, I think it is probably an artificial finding. That is, the patients who are early on enrolled into renal denervation studies probably have the greatest BP at baseline, and perhaps a greater magnitude of benefit as well. In addition, we don’t have great characterization of what the medical therapy changes were in the longer-term follow-up, and some changes in BP could be related to medication changes. I’m not necessarily convinced that there is a growing benefit over time. But, I’m reassured that there is not an attenuation that has been detected, at least at 3 years’ follow-up.
Ferdinand: There could be a placebo effect of simply being in a clinical trial. Participants take their medicines a little better; they adhere to the low-sodium diet a little better; they are less concerned that their BP won’t be controlled because now “they know it is.”
Next steps
Pepine: What is the next step, outside of improving devices?
Patel: Let’s assume that there is a significant reduction felt to be clinically meaningful and the device is approved. The next step will be, like with many innovative and potentially disruptive technologies, getting the therapy to the patients who need it the most while ensuring that those patients were managed in a way similar to those in this study. There are several important questions. Do we set up networks of places where patients with RHTN go through a practice pattern where they have things managed in a consistent way to become eligible for the therapy? Is there information around patients who seem to benefit more or didn’t benefit as much? Are there methods to determine during the procedure whether an adequate amount of ablation was performed?
Ferdinand: Who should get the device is as important as the effectiveness. It would be probably more effective if we were able to treat these patients optimally and identify which do not have true RHTN vs. using the device simply because we can. It will avoid excessive cost and exposure for potential complications. With more widespread use of any intervention, we will have persons who are less skilled, who take a less-appropriate approach to using the intervention, and have more complications. The dollar cost is something I won’t address, but is obvious. There’s no sense just applying a device when we have available low-cost, generic medications, often in combination, to control BP.
Saunders: If these devices are approved, there is a need for strict guidelines, initially. There are going to have to be certain centers where the procedures can be performed and certain physicians who can do it. Patients are going to be asking for it. They have already told me, “If this device is approved, get me off of these drugs.”
Bhatt: It is important to ensure that patients who might be getting this procedure understand there’s a good chance they will still be continued on medicines. Maybe they’ll be able to decrease a medicine causing side effects, but it’s not as though they’re going to come off all medications. Maybe if we start treating patients in the 140 to 150 mm Hg range it will be different. But those who truly have RHTN will still need medications.
Ferdinand: That’s an excellent point. There are two ways of looking at this. One is going back to the coronary atherosclerosis world, when we told people we “fixed your arteries.” We then learned that many patients had diffuse disease and needed statins, BP control and antiplatelet agents, because even if the culprit lesion was “fixed,” they could have another event from another lesion, which at the time of the angiogram may have been mild.
We also need to remember that we did all the right things for persons with type 2 diabetes and obesity, and demonstrated weight loss, decreases in BP and lipids, using the LOOK AHEAD trial as an example. The control group did just as well for CV events, with an increase in statins, antihypertensive agents and aspirin.
We can’t allow patients to buy into a concept that medications are inherently bad. There are no bad medications. We have evidence-based outcomes to show this. AHA Get With The Guidelines registries show that if people take certain medicines when they have specific conditions, they have less MI, less stroke, less hospitalization and may even live longer.
Pepine: Let’s get to the million-dollar question. Assuming SYMPLICITY HTN-3 is positive, and SYMPLICITY HTN-4 extends it to the lower BP level, do we need a real outcome trial?
Patel: This might be the one place where you have the opportunity for therapy in the long run that could potentially be cost-effective. If it does what we think it does — I know there are a lot of “ifs” in this — it’s a compliant therapy. Patients will not have to take medication every day. We have discussed how important medicines are and I agree. But there is a group that either doesn’t have access to BP medications or remains undertreated for other reasons; shouldn’t they also be offered this therapy?
Ferdinand: The outcomes trial would need to have similar or equal BP levels on medications vs. the device, plus or minus medications. I’m going to go on a limb, 15 years from now I don’t think we will know the results.
Bhatt: There are plans for an outcomes trial. In concept, the trialist part of me supports that, but the pragmatist and clinician in me has concerns. If it is similar BP control, I think it’s unlikely, unless these pleiotropic effects are robust, to see any difference in outcomes. If there’s a differential in BP, it’s important to realize that differential BP would need to be maintained over time to translate into CV outcomes. Even if there’s an initial randomization of best medical therapy alone vs. best medical therapy plus device, there’s the potential for add-on therapy in the medical arm where they get the device just because it’s available. A doctor says, “I’m going to do it.” And the patient says, “Do it.” In the arm initially randomized to device, patients may come off medicines, including ACE inhibitors that may have other benefits, at least in patients with HF or diabetes. There’s potential add on and dropout of therapies that would be differential in the two arms that narrow the contrast in BP and could potentially lead to a negative trial.
Pepine: Well, except that the hypothesis I thought that Dr. Ferdinand put forward — you’re going to achieve the same BP, the same BP lowering.
Patel: That’s a critical question to the outcomes trial. We don’t have to design it here today, and obviously there are some interested in doing one, but if you get to the point where this becomes standard, I would say it becomes a strategy trial. Do you go to medicines first vs. this direction?
Saunders: Do we all agree that, if and when this device is approved, it will be limited to specialized centers?
Bhatt: Whether we can limit it might be a separate issue, but whether we should, I’d say yes. I imagine it would roll out similar to transcatheter aortic valve replacement, with initial restriction to a few high-volume sites, when there’s partnership between hypertension experts and interventionalists. A team approach will be necessary to treat these patients. The vast majority of patients referred to the interventional hypertension center will be managed medically, either with addition of spironolactone or other regimen intensification or lifestyle modification. That will advance the management of patients with hypertension tremendously. But a proportion of those patients might benefit from interventional therapies.
Saunders: I agree. Many patients are referred to us from the interventionalist. Then we decide whether to send them back to an interventionalist for intervention. We need the center with two arms involved.
Patel: That’s very much the way we do it, too. It will be interesting to see whether we can use this to set up systems that not only help with the device getting to the patients that need it, but putting in other systems potentially to also inform through electronic health records.
Final comments
Pepine: Please give our readers a brief take-home message.
Bhatt: Renal denervation is extremely exciting, but there has been hyping to date, which is understandable. As noted with initial studies of other advances, there is always excitement; we need that to move the field forward, otherwise things would die. However, we need to see what SYMPLICITY HTN-3 shows. As a properly conducted, randomized, blinded trial, it will give us a precise estimate of treatment effect, which I anticipate will be less than that reported, but I am hoping it will still be enough to be considered clinically relevant.
Deepak L. Bhatt
With that information, the key will be to set up centers of excellence that deal with hypertension. That is, trying to improve the care of all hypertension patients and deciding who is best managed with medicines and who needs intensified therapy with procedural care. If we succeed in setting up such systems, we will have advanced the care of hypertension.
Ferdinand: Data from the National Health and Nutrition Examination Survey (NHANES) 2003-2008 suggest that the prevalence of resistant hypertension is as much as 12% to 13% of all US adults treated for hypertension. That may be an overestimate, but in looking at 78 million people, about 15 million may be potentially available to have a device utilized. I think that’s overkill. We have to continue to look at those five pillars of prevention: sodium; potassium; weight; physical activity; and alcohol. We have to apply optimal medications with the addition of small doses of spironolactone. If we do those things, patient preference aside, we will probably find fewer people who need this device vs. just looking at the numbers of patients with RHTN.
Saunders: The potential to have beneficial effects in the metabolic area and other areas such as HF is interesting. The long-term effect is whether outcomes, because of these additional benefits, are going to be better than with drugs alone. The future, from a research point of view, looks bright and exciting.
Patel: The excitement is justified. We want something more for these difficult-to-treat patients. But I suspect that when we see more data, it will give us, hopefully, our best and most effective therapy. In the future, we need to determine how to set up centers of excellence and proper training so that patients go through the five pillars to get their BP controlled. Finally, how do we do that effectively and fast enough for our patients so that we improve their health?
Disclosure: Bhatt reports research grants from Medtronic (co-principal investigator of SYMPLICITY HTN-3, steering committee of SYMPLICITY HTN-4). Ferdinand reports serving on the speaker’s bureau for AstraZeneca, Forest and Takeda; grant/research support from Lilly; and consulting for AstraZeneca, Daiichi Sankyo, Forest and Novartis. Patel reports research grants from AHRQ, AstraZeneca, Johnson and Johnson, Medtronic, and the NHLBI; he also reports serving on the advisory board for AstraZeneca, Bayer, Genzyme and Janssen Medical. Pepine reports no relevant financial disclosures. Saunders reports serving as a speaker, researcher or advisor for Arbor Pharmaceuticals, Daiichi Sankyo, Forest Laboratories, NHLBI, NIDDK, Novartis and Takeda.