Harsha Vittal, MD
University of Nevada School of Medicine
March 2001
Over the last thirty years, the benefits of treating hypertension have been illustrated through several studies showing a significant decrease in cardiovascular events with anti-hypertensive medication. However, the benefits of pharmacological therapy must be weighed against the risk of those medications. The Atherosclerosis Risk in Communities Study (ARIC) is an ongoing, prospective study of clinical and subclinical atherosclerotic vascular disease. Recently, one analysis of the ARIC study showed a 28% increase in new-onset type II diabetes in hypertensives on beta-blocker therapy. In contrast, there was no increase in new-onset diabetes among people taking diuretics, ACE inhibitors, or calcium-channel antagonists. On the surface, this is an alarming result to physicians and patients who use beta-blockers appropriately in the treatment of hypertension and related cardiovascular diseases. This review examines the findings of the ARIC study and compares the results to other studies addressing the same concerns.
Background and Significance
Since there are a number of choices in antihypertensive therapy, guidelines have been established to aid physicians and patients in this decision. The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC VI) provides clinical guidance for primary care physicians in the evaluation and treatment of hypertension. This report is based largely on evidence-based medicine and the consensus of “thought leaders” in the field of hypertension. The JNC VI guidelines recommend the initiation of beta-blockers or diuretics as first-line agents to treat uncomplicated hypertension in the absence of compelling indications to use another agent (i.e. ACE inhibitors in patients with diabetic nephropathy).
The rationale behind this recommendation is that only diuretics and beta-blockers have been shown to significantly reduce cardiovascular events (i.e. stroke, myocardial infraction, congestive heart failure) in multiple randomized, placebo-controlled clinical trials conducted over many years. Newer agents (such as calcium channel blockers, ACE-inhibitors, angiotensin receptor blockers, and alpha blockers) have been shown to lower blood pressure as effectively as diuretic and beta-blocker therapy, but generally there have been no long-term studies powered to look at hard clinical endpoints with these agents.
In past clinical studies, there has been some evidence that both beta-blockers and thiazide diuretics have negative effects on carbohydrate metabolism and perhaps increase the incidence of diabetes. A number of studies have shown that diabetics not on insulin may become more hyperglycemic when given a beta-blocker, and there are indications that the larger non-diabetic population may be at higher risk of developing diabetes when treated with beta-blockers. This effect may be due to increased insulin sensitivity which has been seen with these agents. In the case of thiazide diuretics, many retrospective studies have observed impairment of glucose tolerance, increased incidence of overt diabetes, and worsening of glycemic control. This effect does appear to be dose related, and many investigators have claimed that it is almost non-existent with lower doses (equivalent of HCTZ 12.5 mg qd). One of the potential benefits of the newer anti-hypertensive agents (calcium antagonists, ACE-inhibitors, etc) is that they do not seem to have any negative effects on glucose metabolism. Whether or not this translates into clinically important differences in outcomes in unknown.
The Atherosclerotic Risk in Communities (ARIC) Study
A recently published analysis of the ARIC study addressed this question by monitoring over 12,000 non-diabetic subjects (both hypertensive and normotensive). This was a purely observational trial accomplished through prospective record review. Subjects were categorized into one of the following regimens: no medications, ACE-inhibitor, calcium channel antagonist, beta-blocker, thiazide diuretics, other single agents, or multidrug therapy. The main outcome, new-onset type II diabetes, was monitored at a 3 year and 6 year follow-up period. In this study, diabetes was defined as a fasting blood sugar greater than 126 mg/dl, a nonfasting blood sugar of greater than 200 mg/dl, or the initiation of insulin or oral hypoglycemic therapy.
Of the 12,000 subjects in the study, 3800 initially had the diagnosis of hypertension. At six years, 1146 total new cases of diabetes were found with 569 of these cases from the hypertension group. This corresponds to a relative risk of 2.43 of patients with hypertension developing diabetes at the six-year mark (as compared to normotensives).
Furthermore, the investigators analyzed the data to determine if the choice of antihypertensive therapy had any effect on the incidence of diabetes. In order to avoid the potential for confounding, a number of variables were adjusted in this analysis (Table 1).
When compared to hypertensive subjects taking no blood pressure medications, this adjusted analysis showed that subjects taking thiazide diuretics, ACE-inhibitors, or calcium-channel antagonists did not have a significant increase in the incidence of new-onset diabetes. On the other hand, subjects taking beta-blockers had a 28% increased risk of developing diabetes compared to those hypertensive subjects who did not take any medications (Table 2).
Implications for the Treatment of Hypertension
Certainly, these results raise doubts about current recommendations for initial anti-hypertensive therapy. However, there are a number of limitations of this study. First and most importantly, this was not a randomized, controlled study. Subjects were prescribed these medications according to the usual practice of their physicians. Even though the researchers painstakingly tried to account for possible confounders, it is virtual impossible to do so in a prospective cohort study. Physicians choose individual anti-hypertensive medications for a number of different reasons, and it is certainly possible that there was an unidentified confounding influence on the choice of beta-blocker for patients who were at risk for the development of diabetes.
Also, the types of beta-blockers used by the subjects were not addressed. A number of studies have shown a greater than six times increase in onset of diabetes in patients taking propranolol compared to the 28% increase found in the ARIC study for all beta-blockers. Furthermore, it has been suggested that carvedilol, a beta-blocker shown to prolong survival in the setting of congestive heart failure, may actually improve glycemic control. These differences may be attributed to the variations in selectivity of each medication (ie: cardioselective v. non-cardioselective), but there may be other differences between these agents. Further studies evaluating specific beta-blocking agents and their effects on glucose tolerance need to be conducted to illustrate the differences among the various formulations.
As far as the other classes of medications are concerned, the ARIC study confirmed that thiazide diuretics, at contemporary doses, do not appear to increase the incidence of diabetes. Historically, thiazide diuretics had been shown to have a hyperglycemic effect, but that was found at higher doses (50 mg to 200 mg) than the doses currently used for hypertension treatment (12.5 mg to 25 mg).
ACE-inhibitors do not seem to have any hyperglycemic effect. In fact, the ARIC results stand in sharp contrast to the results of the recently published Heart Outcomes Prevention Evaluation (HOPE) study. In HOPE, subjects at high risk for cardiovascular disease were randomized to the ACE-inhibitor, ramipril, or placebo. Compared to those on placebo, patients randomized to ramipril had a highly significant decrease in cardiovascular events. Furthermore, patients on ramipril were 34% less likely to develop diabetes over the approximately two years of follow-up. (Table 3) Also, the MICRO-HOPE substudy showed that ramipril therapy decreased the incidence of cardiovascular events by 25% in patients previously diagnosed with diabetes.
What Does This All Mean?
Beta-blockers have consistently been shown to reduce the incidence of stroke, heart attack, and congestive heart failure in numerous well-conducted prospective randomized-controlled trials. As such, the JNC VI guidelines recommend beta-blockers (or thiazide diuretics) as first line treatment of uncomplicated hypertension. Furthermore, due to beneficial effects on the sympathetic nervous system and the renin-angiotensin system, beta-blockers have been shown to prolong survival in patients with coronary artery disease, prior myocardial infarction, and compensated congestive heart failure.
Like all medications, beta blockers are not without adverse effects. Due to problems inherent in the study design, the results of the ARIC study must be viewed with caution. However, these results do seem to indicate that, unlike other agents used to treat hypertension, beta-blockers may increase the risk of new-onset diabetes in susceptible individuals. Whether or not this actually confers a true clinical advantage to other agents is unknown. Newer agents, like ACE-Inhibitors have either neutral or beneficial effects on carbohydrate metabolism and the incidence of diabetes; however, we are lacking studies which demonstrate that these agents in fact decrease events in uncomplicated hypertensives.
For the primary care physician, the well-documented benefits of beta-blockers must be weighed against the risks of initiating beta-blocker therapy. Given the results of the ARIC study, it may be reasonable to avoid beta-blockers in patients with uncomplicated hypertension who may be at increased risk for developing diabetes.
References
- Gress TW, Nieto FJ, Shahar E, Wofford MR, Brancati FL. Hypertension and antihypertensive therapy as risk factors for type 2 diabetes mellitus. Atherosclerosis Risk in Communities Study. N Engl J Med. 2000 Mar 30;342(13):905-12.
- The sixth report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Arch Intern Med 1997;157:2413-46
- Jacob S, Rett K, Wicklmayr M, Agrawal B, Augustin HJ, Dietze GJ. Differential effect of chronic treatment with two beta-blocking agents on insulin sensitivity: the carvedilol-metoprolol study. J Hypertens 1996;14:489-94.
- Yusuf S, Sleight P, Pogue J, Bosch J, Davies R, Dagenais G. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. The Heart Outcomes Prevention Evaluation Study Investigators. N Engl J Med. 2000 Jan 20;342(3):145-53.
- Bristow MR, Gilbert EM, Abraham WT, Adams KF, Fowler MB, Hershberger RE, Kubo SH, Narahara KA, Ingersoll H, Krueger S, Young S, Shusterman N. Carvedilol produces dose-related improvements in left ventricular function and survival in subjects with chronic heart failure. MOCHA Investigators. Circulation. 1996 Dec 1;94(11):2807-16.
|
