Abstract
Following a symposium on hypertension sponsored by the National Heart, Lung, and Blood Institute in San Francisco in April, 2000, a panel was convened to discuss the management of high‐risk hypertensive patients with comorbid diseases. The roundtable consisted of Nancy Houston Miller of Stanford University School of Medicine; Dr. Marc Jaffe of the Kaiser Permanente Medical Group of Northern California; and Dr. Ray Gifford, Jr., Emeritus Chief of Hypertension and Nephrology at the Cleveland Clinic. The panel was chaired by Dr. Marvin Moser of the Yale University School of Medicine.
DR. MOSER: There is little doubt that most hypertensive patients have other diseases, especially if they are older than 60–65 years of age. Most comorbidities, for example diabetes or hyperlipidemia, are related to cardiovascular disease (CVD).
First of all, Marc, what's your guess as to how many hypertensive patients have other diseases that influence their care or their risk? What do you think? Is it 10%, 50%, or more?
DR. JAFFE: With advancing age, patients who have hypertension as their only cardiovascular risk tend to be an increasing minority. When you look carefully at glucose intolerance, diabetes, or hyperlipidemia, the chances of having at least one of these comorbidities in a hypertensive subject are at least 50%.
DR. MOSER: Ray, how do you define a high‐risk patient? The Joint National Committee (JNC) divided risk into categories—groups A, B, and C—in order to clarify treatment priorities. Define what these risk group categories mean.
DR. GIFFORD: Risk group A (Table I) was without any other risk factors at all; (the group was defined as young persons with blood pressure [BP] >140/90 mm Hg, but otherwise no cardiovascular risk factors). Risk group B includes patients who have hypertension and dyslipidemia, who are cigarette smokers, or who have other risk factors. Something as simple as age 60 puts you in group B, because age is a risk factor for CVD in addition to hypertension. Risk Group C is anyone with hypertension who has diabetes, evidence of target organ damage, or heart disease. Being a man or a postmenopausal woman, even if she is less than 60 years of age, puts you in a higher risk category.
Table I.
Risk Stratification of Hypertensive Patients
| Risk Group A |
| • No risk factors |
| • No target organ disease or clinical evidence of cardiovascular disease |
| Risk Group B* |
| • At least one other risk factor, not including diabetes (e.g., smoking history, hyperlipidemia, etc.) |
| • No target organ disease or clinical evidence of cardiovascular disease |
| Risk Group C* |
| • Target organ disease or evidence of clinical cardiovascular disease and/or diabetes |
| • With or without other risk factors |
| *These groups should be treated sooner than low‐risk patients who may be followed for 6 or more months on lifestyle modifications before beginning specific antihypertensive drug therapy. The more risk factors present, the greater the urgency to begin therapy. Most patients in Groups B and C, especially those with stage 2 or 3 hypertension, should be started on medication and lifestyle modification at the same time. |
DR. MOSER: You're saying that a man…
DR. GIFFORD: …is at higher risk than a woman at any level of BP.
DR. MOSER: Okay, so men can't really be classified as low‐risk. If a 48‐year‐old male has a BP of 155/95 mm Hg, he's a stage 1 hypertensive but should be treated as a group B subject, because he's a male. Changing the designation may make a difference in how long you would try nondrug therapy.
DR. GIFFORD: The risk is higher than for women of the same age.
DR. MOSER: Higher than women. Okay, and over 60, the difference disappears?
DR. GIFFORD: Yes, it does. Because most women are postmenopausal by the time they're 60 and the postmenopausal state puts them at a higher risk. Table I lists some of the factors that influence risk categories.
DR. MOSER: What we're saying is that the majority of hypertensives are probably at moderate to high risk of having a cardiovascular event.
DR. GIFFORD: I would think so, yes.
DR. MOSER: What about other factors, Nancy? What else helps us to categorize people? These [categories] are important because their presence or absence helps to determine how quickly patients should be put on antihypertensive drug therapy rather than just being continued on non pharmacologic therapy. Using risk groups provides a guide to determine how quickly to change therapy. Table II outlines JNC‐VI recommendations for initiating various therapies for different risk categories.
Table II.
Risk Stratification and Treatment*
| Risk Group A (No risk Factors; no TOD/CCD)** | Risk Group B (At least one risk factor, not including diabetes; no TOD/CCD) | Risk Group C (TOD/CCD and/or diabetes; with or without other risk factors) | |
| Blood pressure stages (mm Hg) | |||
| High‐normal (130–139/85–89 mm Hg) | Lifestyle modification | Lifestyle modification | Drug therapy‡ |
| Stage 1 (140–159/90–99 mm Hg) | Lifestyle modification (up to 12 months) | Lifestyle modification† (up to 6 months) | Drug therapy |
| Stages 2 and 3 (≥160/≥100) | Drug therapy | Drug therapy | Drug therapy |
| For example, a patient with diabetes and a blood pressure of 142/94 mm Hg plus left ventricular hypertrophy should be classified as having stage 1 hypertension with target organ disease (left ventricular hypertrophy) and with another major risk factor (diabetes). This patient would be categorized as stage 1, Risk Group C, and would be recommended for immediate initiation of pharmacologic treatment. | |||
| *Lifestyle modification should be adjunctive therapy for all patients recommended for pharmacologic therapy; **TOD=target organ disease; CCD=clinical cardiovascular disease; †for patients with multiple risk factors, clinicians should consider drugs as initial therapy, plus lifestyle modifications; ‡for those with heart failure, renal insufficiency, or diabetes | |||
MS. MILLER: I think the JNC‐VI did a good job of identifying risk factors and placing people in appropriate subsets. This is one of the first guidelines that I've seen that has attempted to do something like this and looked at multiple risk factors to determine whom and when to treat.
DR. MOSER: Marc, what about elevated homocysteine and uric acid levels? If these were present, would that change your categorizing a patient who was a nonsmoker and nondiabetic and who didn't have left ventricular hypertrophy? For example, if that patient had a uric acid of 7.6 mg/dL and a high homocysteine level, would you conclude that he or she should be treated earlier than if these findings were not present? Is risk increased, or are these factors not that important? Recent data suggest that elevated homocysteine levels correlate with decreased blood vessel compliance.
DR. JAFFE: Well, when you're assessing someone's risk and estimating morbidity and mortality mostly from cardiovascular‐related events, we have to prioritize. Although there have been observational studies to suggest that elevated homocysteine levels increase risk, some recent reviews suggest that this might not be the case. I would say that the jury is probably still out concerning this finding. Although the presence of an elevated serum uric acid level may be present in those patients with a metabolic syndrome that frequently goes along with hypertension (i.e., elevated triglyceride levels, a low high‐density lipoprotein [HDL] cholesterol, central obesity, and increased risk of coronary artery disease), hyperuricemia in and of itself is not a powerful predictor of risk.
DR. MOSER: So it may go along with other risk factors, but you're not certain about hyperuricemia as an independent risk factor. That's probably consistent with the data. But the syndrome of low HDLs, low‐density lipoproteins (LDLs), high triglycerides, and obesity certainly makes you think of a higher‐risk patient. Of course, we haven't mentioned the presence of congestive heart failure (CHF), diabetes, angina, a previous myocardial infarction (MI), or stroke; all of these put people in class C and indicate a different strategy. So, Dr. Gifford, we have this patient in group C: BP of 155/95 mm Hg (grade or stage 1 hypertension), but she is a diabetic or has had a previous episode of angina. How does the presence of these other factors change what you do? After all, if diabetes wasn't present, you might follow a patient like this for several or more months, just on lifestyle modifications.
DR. GIFFORD: It would certainly change my approach. I would begin pharmacologic therapy immediately, along with nonpharmacologic therapy, because of the high risk. Even if this patient had pressures of 140–145/90–95 mm Hg, I would do that. If there is target organ involvement, such as CHF or renal failure, I would give drug therapy in addition to nonpharmacologic therapy, even if the BP was high‐normal (i.e., 130–140/85–89 mm Hg).
DR. MOSER: Nancy, Marc, do you agree that a person with a BP of 135–140/85–90 mm Hg who is a diabetic should be placed on specific medication in addition to lifestyle changes, without waiting for 3–6 months?
MS. MILLER: I do agree. I think those patients are at high risk for coronary disease, and I would do everything possible to lower their BP and get their lipids under control, to an LDL goal of 100 mg/dL, in order to try and reduce their overall risk of CVD.
DR. MOSER: Marc, do you agree?
DR. JAFFE: Yes, I agree. I would just say that in high‐risk subjects, we should start aggressively with multiple modalities to reduce risk. Later, if the medication regimen has proved effective, the program can always be modified. Unfortunately, what we experience is that we tend not to be particularly effective with behavioral changes alone and may miss the opportunity to move aggressively with high‐risk patients if we wait too long.
DR. MOSER: We all agree that nonpharmacologic intervention should be part of a regimen, but that, in this high‐risk group, it should be used along with antihypertensive drug therapy. This approach is not universally accepted; many doctors are reluctant to use medication in people with stage I or high‐normal BP, even if they have diabetes or evidence of target organ damage—especially if only systolic BP elevations are present. How do we convince them? Do we have good data in this group that we reduce risk by lowering BP?
DR. GIFFORD: Well, we can't reduce macrovascular risk by treating diabetes. Yet we treat the diabetes. So, we aggressively treat those risk factors when we know we can improve the outlook: that's hypertension.
DR. MOSER: But do we have data? Let's take a nondiabetic but a person at high risk because he or she has hyperlipidemia, is a smoker, and is over the age of 65. Blood pressures are about 135–140/85–90 mm Hg—in other words, below the hypertension cutoff point but high‐normal. Should we use medication here? What did the JNC‐VI say about this group?
DR. GIFFORD: JNC‐VI didn't recommend treating this group with medication first.
DR. MOSER: Okay, so in those patients, we use lifestyle modifications for 3–6 months or so, and then it's a judgment call.
DR. GIFFORD: Right.
DR. MOSER: Everyone agree with that?
DR. GIFFORD: According to the JNC‐VI, we might never use medication in the high‐normal group, except in high‐risk patients. But, as I noted before, I would [use medication] in an otherwise high‐risk patient, such as a diabetic with renal failure, and I gather that Marc and Nancy agree.
DR. MOSER: Are there specific factors or disease entities that will determine what medications you start with? For example, if you have a patient with asthma, you are probably not going to use a β blocker, or you would use it cautiously and in small doses, in a patient with angina or post‐MI. What other medical conditions or comorbidities will determine how therapy is started once this has been decided upon? And, as we've noted, most people with hypertension and other risk factors require medication.
DR. JAFFE: In the patient with known systolic dysfunction or CHF, the initial choice of an angiotensin‐converting enzyme (ACE) inhibitor, when tolerated, makes sense. This is usually given with a diuretic. In those patients with isolated systolic hypertension, a diuretic is a good choice for initial therapy.
DR. MOSER: What about people with prostatic hypertrophy? Are you going to use an α blocker? These agents are effective in relieving symptoms.
DR. JAFFE: This is a difficult question. As physicians, we're interested in alleviating symptoms first and foremost, and clearly α blockers have a role in diminishing the signs and symptoms of pros‐tatism. But there is no evidence that the initial use of an α blocker confers the same cardioprotective effects as other, more well established therapies, such as thiazide diuretics and β blockers.
DR. MOSER: In fact, the ALLHAT trial reported that there was an increase in CHF and other cardiovascular events, such as angina and stroke, with an α blocker when compared to a diuretic. So, if you're going to treat someone with prostatic hypertrophy, that's fine. If you're going to treat someone with hypertension and prostatic hypertrophy, perhaps we should use the drugs that work in hypertension. It should be noted that in the ALLHAT study, the use of the α blocker has been discontinued. I believe that α blockers should probably be used as a third‐step agent in resistant patients or as an add‐on for prostate disease. Ray, Nancy, do you agree with that?
DR. GIFFORD: Yes, I agree with that.
MS. MILLER: I agree.
DR. MOSER: For many years, physicians were advised that certain drugs should not be used in diabetes and should be avoided in patients with hyperlipidemia. Specifically, the diuretics and β blockers were singled out, because they had been shown in some studies to increase insulin resistance or to increase lipid levels, at least short‐term. In the case of the β blockers, a reduction in HDL levels and an increase in triglycerides were noted. These seemed to be changes that were not ideal in a hypertensive patient who already may have had evidence of dyslipidemia. Have those fears been confirmed? Where do we stand with the use of diuretics and β blockers with these particular comorbidities? Nancy?
MS. MILLER: I think that in most cases, we're actually finding that what we thought earlier was the case really may not be of clinical importance. Even in those groups in which β blockers, for example, were supposed to be contraindicated (i.e., patients over the age of 65 and/or diabetics), the use of these agents has shown beneficial effects; importantly, these patients can usually tolerate these medications. I believe the same thing is true in relationship to the diuretics. Diuretics do not cause permanent problems in the management of dyslipidemia or in glucose control, and their use in hypertensive diabetics or patients with hyperlipidemia results in a decrease in morbidity and mortality. More research is turning out to favor these drugs.
DR. MOSER: Marc and Ray, would you use a diuretic, or a β blocker, or a combination of the two in a patient with hyperlipidemia, or in a patient with hyperglycemia or diabetes?
DR. GIFFORD: Yes, I would, and I have. If a patient has diabetes and proteinuria, then I would probably favor an ACE inhibitor. But if it's just a matter of treating the BP in a patient with diabetes who doesn't have evidence of nephropathy, a diuretic and a β blocker are good choices and have proved to be effective.
DR. MOSER: What about diuretics in patients with hyperlipidemia? We were once told that's a contraindication.
DR. GIFFORD: Yes, that's true, but the more you look at the long‐term studies, the short‐term effect that diuretics have on lipids doesn't seem to make any difference as far as overall benefits of treatment.
DR. MOSER: And we have good data from the SHEP study, for example, that shows that people with hyperlipidemia did just as well with regard to reduction of cardiovascular events as the people with normal lipids. We also have data from a recent cohort study showing that the use of diuretics had no adverse effects on lipids or glucose. Marc, do you hesitate to use a β blocker or diuretic in a diabetic or a hyperlipidemic patient?
DR. JAFFE: No, I don't hesitate at all. I agree with Dr. Gifford. What we're interested in doing is preventing end‐organ damage. Good outcome‐oriented research has demonstrated that whatever small, short‐term changes there might be in some of the metabolic parameters, these pale in comparison to the benefits achieved with the use of these agents in reducing death and microvascular and macrovascular disease, which really are the major problems for our patients with diabetes.
DR. MOSER: All right, Ray, let's discuss some comparative studies. Data indicate that when you compare a β blocker‐ or diuretic‐based treatment program to an ACE‐based regimen in type 2 diabetics, morbidity and mortality appear to be reduced equally (UKPDS trial). When you compare these two regimens in the elderly (a high‐risk group), the data are also equivalent (STOP 2 studies). What about ACE inhibitors and calcium‐channel blockers (CCBs)? Do we have any comparative data as to which one might be preferred in patients with diabetes or in the elderly? Did the STOP 2 study or one of the diabetic trials find any difference?
DR. GIFFORD: In the ABCD and FACET studies in patients with diabetes, the ACE inhibitor appeared to be more effective than the calcium antagonist with regard to preventing coronary outcomes. There were fewer coronary events in the ACE‐inhibitor group than in the calcium‐antagonist group. But I don't know that that's definitive. I use calcium antagonists in elderly patients and especially those who have angina.
DR. MOSER: Rather than a β blocker.
DR. GIFFORD: Not “rather than” but equally. I think they reduce angina as much as or maybe more than a β blocker, and sometimes I use them together, if the angina is severe and doesn't respond to one of them alone.
DR. MOSER: One additional comment. In the STOP 2 trial in the elderly, the ACE group experienced fewer episodes of CHF or MI than did the CCB group—data consistent with other trials. An important point is that all the trials—including the STOP 2, UKPDS, and the HOT trials—were really trials of at least two drugs. I believe that we all have concluded that the majority of our patients are not going to achieve normotensive BP levels on monotherapy. How does this work in high‐risk patients with comorbidities? Are we going to make their lives more complicated by using two different classes of drugs?
DR. GIFFORD: You don't have to if you use two drugs that are made up in one tablet. I think you enhance compliance by using a combination tablet. And most of the drugs can be given in combination tablets, usually a diuretic and a β blocker, or a diuretic and an ACE inhibitor, or a diuretic and an angiotensin receptor blocker (ARB).
DR. MOSER: What about side effects, Marc, when you give someone two different drugs at the same time, do you run into more or fewer side effects?
DR. JAFFE: Generally, combining medications that work synergistically, for example a β blocker with a diuretic, allows you to use a smaller dose of each agent. When both agents are used at lower doses, there tend to be fewer side effects; these medicines are well tolerated.
DR. MOSER: Nancy, you deal with heart failure a great deal, and that's certainly one of the complications of hypertension. I guess about 70% of people who go into heart failure have or have had hypertension. In that group, we're not using one, two, or even three medications; we're often using four. Are the results better when you add four drugs instead of two or one?
MS. MILLER: It appears that the results are better in many cases. But here you are dealing with a symptomatic disease. Adding drugs doesn't seem to worsen compliance at all, because these patients recognize that they've got to be on medications because of symptoms. They also recognize how critical it is to lower the BP, as well as treat symptoms and correct the physiologic changes that occur in heart failure.
Probably the most difficult population of patients with heart failure to deal with are the uncontrolled diabetic, obese, hypertensive, middle‐aged females. They've had uncontrolled diabetes and uncontrolled hypertension for a number of years. And we're now having to treat those patients very aggressively with multiple drugs. Our emphasis must shift to controlling their BP and diabetes before they go into heart failure.
DR. MOSER: Patients with heart failure who already are on a diuretic and digitalis require ACE inhibitors; morbidity and mortality is lowered when these are added. Now we have good evidence that people who are on ACE inhibitors, a diuretic, and digitalis may do even better if a β blocker or an α‐β blocker is added; morbidity and mortality are reduced still further.
Nancy, you've spent a great deal of your time evaluating compliance. Do you think combination tablets really help adherence to therapy? It seems to me that perception of illness and adherence to therapy is improved with one or two pills a day rather than with two or four. From the patient's point of view—and from the doctor's point of view—titration is certainly easier; goal BP is reached more quickly. Your comments?
MS. MILLER: I've had a great deal of experience dealing with patients who are on low doses of two drugs, and I agree with Dr. Jaffe. Patients on lower doses of two drugs seem to have fewer side effects and are able to tolerate the medications better than higher doses of a single agent.
DR. MOSER: Ray? Do you think that combination pills really do help in achieving goals more quickly and encouraging better adherence?
DR. GIFFORD: I do. This comes from my own observations of patients. The problem with combination pills may be that, if all the medication is combined in one pill—and the patient forgets to take that one pill—it's 100%. If you're taking two pills and fail to take one of them, it's 50%. So that's a problem. But I believe that in people who have side effects to medication, combining the two ingredients in one pill tends to make them less susceptible to side effects. I don't know whether that's the dosage or just maybe their perception. When I used to use placebos (when we thought it was ethical to use placebos), I would give patients who complained of side effects one tablet of a placebo, and they'd get the same side effects. They'd come back and complain, “I'm weak, I'm tired, oh, it's awful.” And you give them two placebos, they're twice as tired. And three placebos, they're three times as tired. So that's one reason I started using combination tablets, because the more medication you get into people with one tablet, the better off you and they are.
DR. MOSER: Perception of illness is different. Marc?
DR. JAFFE: Yes, I think we also need to remember that combination pills have an advantage for the prescribing physician. I do believe that having a combination pill may make it easier for the prescribing physician to prescribe two different medications at once. I think that we need to think about compliance, not only in terms of our patients taking the medications, but in terms of our colleagues being able to easily prescribe combinations of medicines that are well tolerated and effective.
DR. MOSER: A quick summary. There is no doubt that hypertensives have many comorbidities, whether lipid elevations, glucose intolerance, or target‐organ changes, such as left ventricular hypertrophy, CHF, a previous stroke, or a transient ischemic attack. More than half of our hypertensive patients have some comorbid condition. These patients should be treated earlier than patients without other risk factors. But—other than specific instances such as 1) heart failure, where you'd use an ACE inhibitor and a diuretic and probably a β blocker, or 2) specific diseases, like diabetes and nephropathy, where you'd probably favor an ACE inhibitor/diuretic or an ARB/diuretic—these patients generally respond to most of the other antihypertensive agents with lowering of BP and reduction of risk. Comorbidity should not be viewed as a burden but an opportunity to treat patients more vigorously. Contrary to what many people might believe, the high‐risk patients respond well, and reduction of risk in these individuals is greater than in lower‐risk patients, especially in diabetics. Left ventricular hypertrophy can be reversed and heart failure prevented from occurring in the first place by adequate lowering of BP. Finally, the use of combination therapy may help to increase the numbers of patients who achieve goal pressures. Dr. Gifford, a closing comment?
DR. GIFFORD: I agree that it's important to aggressively treat hypertension, even if a patient doesn't have a comorbidity, because you can prevent comorbidity by aggressive treatment of the hypertension. But if a patient does have other risk factors, it becomes even more important that we lower the BP aggressively.
DR. MOSER: Marc?
DR. JAFFE: I think we're seeing a shift in prescribing patterns. It used to be that physicians were cautious; we were trained to be cautious—perhaps too cautious—in avoiding perfectly good medications in those people who needed them most. Now, I see compelling evidence that the medicines that are really the cornerstones of treatment for most patients (namely, JNC‐VI primary recommended therapies, β blockers, and thiazide diuretics) are not only safe but also well tolerated and extremely effective in those patients with comorbidities.
DR. MILLER: I would have to say that I think we are seeing data from one trial after another that [indicate], irrespective of which agent we use, the lower we can drive the BP, the better off we are going to be in preventing overall cardiovascular events. I would have to agree with Dr. Gifford that it's lowering the BP that makes the difference.