What Should Our Blood Pressure Goal Be in Patients With Diabetes?

Hypertension and type 2 diabetes mellitus frequently coexist, and the incidence and prevalence of both have increased in recent years, primarily because of the obesity epidemic and the aging of the population. It is well‐known that hypertensive patients with diabetes have a 2‐ to 3‐fold increase in the risk of cardiovascular morbidity and mortality compared with hypertensive patients without diabetes. ¹ Therefore, it seemed logical to recommend more aggressive treatment of risk factors in persons with diabetes, especially low‐density lipoprotein (LDL) cholesterol and blood pressure (BP) levels. I will not address the current state of evidence for LDL cholesterol goals in diabetic patients, virtually all of whom should probably be treated with lipid‐lowering therapy, but I will discuss how aggressive I believe we should currently be in lowering BP in patients with hypertension and diabetes. This issue is particularly timely because of the presentation (at the European Society of Cardiology meeting) and publication (in The Lancet) in September 2007 of the BP component of the Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation (ADVANCE) trial. ²

Observational studies have shown a continuous relationship between BP level and both macrovascular (major cardiovascular) and microvascular (primarily retinopathy and nephropathy) outcomes in diabetic patients, across a broad range of BP levels, beginning in the normal range (<120/80 mm Hg). ¹ , ³ The risk at each level of BP is substantially increased in persons with diabetes, however. Virtually every practice guideline since 2000 that has dealt with treatment of hypertension in diabetic patients, including the Seventh Report of the Joint National Committee on the Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7), ⁴ has recommended initiating antihypertensive drug treatment in persons with diabetes who have systolic BP levels ≥130 mm Hg or diastolic BP levels ≥80 mm Hg and treating to a BP goal of <130/80 mm Hg. But what is the evidence that reducing BP to these levels is safe and will reduce macrovascular and microvascular events to the degree suggested by epidemiology or at least to a clinically meaningful degree more than reduction of BP to <140/90 mm Hg, the level recommended for most hypertensive patients?

Results of randomized clinical trials of BP lowering in patients with hypertension and diabetes provide the primary evidence to answer the question of what BP goals should be. In the Systolic Hypertension in the Elderly Program (SHEP), there were 583 participants with type 2 diabetes. Major cardiovascular disease (CVD) events were reduced by 34% in the actively treated group, which initiated treatment with the thiazide‐type diuretic chlorthalidone in older participants with isolated systolic hypertension. ⁵ The systolic BP entry criterion was 160 to 219 mm Hg and the treatment goal was <160 mm Hg with a ≥20 mm Hg reduction from baseline. The achieved mean systolic BP was 146 mm Hg in the actively treated group and 155 mm Hg in the placebo group in the SHEP diabetes subgroup. ⁶

In the Hypertension Optimal Treatment (HOT) study, hypertensive patients were randomized to 1 of 3 BP goals: ≤90, ≤85, and ≤80 mm Hg. ⁷ Although there were no significant differences in CVD events in the entire group of 18,790 participants with hypertension in the HOT study, in the subgroup with diabetes (n=1501), major CVD events were reduced by 51% (P=.005) in participants randomized to the lower diastolic BP goal compared with the ≤90 mm Hg goal. ⁷ , ⁸ In this diabetic subgroup, the achieved mean BP level in the group with the ≤80 mm Hg goal was 144/81 mm Hg, compared with 148/85 mm Hg in the group with the ≤90 mm Hg goal. ⁸

In the United Kingdom Prospective Diabetes Study (UKPDS), ⁹ 1148 patients with hypertension and type 2 diabetes were randomized to goal BP levels <150/85 mm Hg (tight BP control) or <180/105 mm Hg (less tight BP control). After 8.4 years median follow‐up, diabetes‐related end points were reduced by 24% (P=.005), deaths related to diabetes were reduced by 32% (P=.019), strokes were decreased by 44% (P=.013), and microvascular end points were reduced by 37% (P=.009) in the group with the lower BP goal compared with the less intensively treated group. Average BP level over 9 years was 144/82 and 154/87 mm Hg in the tight and less tight BP control groups, respectively; a difference of only 10/5 mm Hg was associated with these large differences in outcome.

In the Systolic Hypertension in Europe (Syst‐Eur) trial, participants with isolated systolic hypertension (similar entry criteria to SHEP but with a goal systolic BP level <150 mm Hg) were randomized to the dihydropyridine calcium channel blocker nitrendipine or placebo; in a post hoc analysis of the relatively small group of participants with diabetes (n=492), there were significant reductions in CVD mortality (by 76%), all CVD events (by 69%), and stroke (by 73%). ¹⁰ The mean systolic BP was reduced from 175 to 153 mm Hg in the active therapy group compared with 162 mm Hg in the placebo group; again, a 9‐mm Hg lower systolic BP level was associated with a large difference in outcome.

In the Appropriate Blood Pressure Control in Diabetes (ABCD) trial, 470 participants with hypertension and type 2 diabetes were randomized to moderate control of BP (diastolic BP 80–89 mm Hg) or intensive control of BP (diastolic BP 75 mm Hg) to compare the incidence and progression of diabetic nephropathy, retinopathy, CVD, and neuropathy. ¹¹ , ¹² The achieved mean BP level was 132/78 mm Hg in the intensive control group and 138/86 mm Hg in the moderate control group. There were no differences in any microvascular end points for the 2 BP goal groups. Although the intensive therapy group had a lower mortality rate, 5.5% compared with 10.7% (P=.037), there were no statistically significant differences in rates of myocardial infarction, cerebrovascular events, or heart failure to account for the mortality difference. ¹² The mortality difference seen in the ABCD trial should be interpreted cautiously, however, because the small sample size was small, mortality was not a primary prespecified outcome, there was a marginal significance level, the multiple comparisons performed, and the absence of benefit for CVD or renal outcomes.

Until recently, therefore, the HOT study and UKPDS have provided the strongest clinical trial evidence to support BP level goals in patients with diabetes and hypertension. They suggest a BP goal of <150/85 mm Hg (UKPDS) and diastolic BP ≤80 mm Hg (HOT study). The achieved mean systolic BP in the more intensively treated groups for both studies was 144 mm Hg, however, and in SHEP it was 146 mm Hg. Therefore, the findings of these trials are consistent with systolic BP goals of <145 mm Hg in patients with diabetes, and no trials, including the ABCD trial, reported a confirmed benefit with goals lower than this. Taken together, they do nevertheless support a diastolic BP goal of ≈80 mm Hg.

There is new evidence supporting a somewhat lower systolic BP goal for patients with diabetes. It comes from the ADVANCE trial, ² which was designed to assess the effects of diabetes on macrovascular and microvascular disease (separately and in combination) using a factorial design to test glycemia and BP questions. The approach in this trial to increase the use and effectiveness of treatment for lowering BP in patients with diabetes has been stopped and reported. In the BP portion of the ADVANCE study, 11,140 participants with type 2 diabetes and a broad range of BP values were randomized (double blind) to receive a fixed‐dose combination of the diuretic indapamide (up to 1.25mg/d) plus the angiotensin‐converting enzyme (ACE) inhibitor perindopril (up to 4 mg/d) or placebo, irrespective of initial BP level or the use of other antihypertensive drugs. These are half the usual recommended maximum dosages, in part so that the ACE inhibitor (perindopril, up to 4 mg/d) could be continued in an open‐label fashion or added if the investigator felt that it was indicated. A confounding aspect of the trial was that by the final visit, open‐label ACE inhibitor use had been added in 50% of the treatment group (in addition to the blinded ACE inhibitor) and in 60% of the control group.

During the 4.3 years of follow‐up, an average BP level of 135/75 mm Hg was achieved in the group assigned to the ACE inhibitor/diuretic combination, compared with an average of 140/77 mm Hg in the placebo group (5/2 mm Hg difference). The more intensively treated group with the lower BP experienced a 9% lower rate of macrovascular and microvascular events combined (P=.04), as well as lower rates of cardiovascular (by 18%) and all‐cause (by 14%) mortality (P=.03 for both). Neither the macrovascular (−8%) nor the microvascular (−9%) components of the primary outcome were separately significant, although they were about the same magnitude. Although the primary outcomes were marginal or not significant, the mortality benefit was impressive because of the size of the trial.

The ADVANCE trial can be interpreted as a BP trial, supporting a systolic BP goal of 135 mm Hg, or a trial of the benefits of using a diuretic plus an ACE inhibitor in patients with diabetes, although the primary difference in the randomized groups was the use of a diuretic. There is really no way to resolve which interpretation is correct with this design, although both are reasonable to apply to the diabetic hypertensive population. There was no heterogeneity for the primary or other outcomes across a variety of subgroups, including those with or without a history of hypertension or above or below a systolic BP level of 140 mm Hg; within most subgroups the treatment effect was not independently significant. I am not certain, therefore, that this trial alone gives a compelling reason to add 2 medications in “normotensive” diabetics. Certainly, based especially on the findings of the Antihypertensive and Lipid‐Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) ¹³ and the Heart Outcomes Prevention Evaluation (HOPE), ¹⁴ it is reasonable to prescribe a thiazide‐type diuretic and an ACE inhibitor in most patients with diabetes and hypertension; the results of ALLHAT suggest, however, that the diuretic should have priority over the ACE inhibitor if only 1 of the 2 is chosen, unless there is a more compelling indication for the ACE inhibitor (eg, low ejection fraction heart failure). To achieve a systolic BP goal in hypertensive diabetics of <130 mm Hg or even <135 mm Hg, as suggested in the ADVANCE trial, however, at least 2 antihypertensive drugs will be needed for most individuals.

Relying primarily on the available evidence from clinical trials, the 2004 hypertension guidelines of the US Department of Veterans Affairs and the US Department of Defense recommended a goal BP in patients with diabetes of <140/80 mm Hg. ¹⁵ Should all practice guidelines for patients with diabetes and hypertension now recommend BP goals of <135/80 mm Hg? Perhaps they should. Within the next few years, however, the results of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial should clarify whether treating to an even more aggressive systolic BP goal will reduce CVD events in patients with type 2 diabetes mellitus. ⁶

The ACCORD trial is designed to test the effects on preventing major CVD events of intensive glycemic control, treatment to increase high‐density lipoprotein cholesterol and lower triglycerides levels (in the context of good LDL cholesterol and glycemic control), and intensive BP control (in the context of good glycemic control) in 10,251 participants with type 2 diabetes. The BP trial is an unmasked, open‐label randomized trial with a sample size of 4733 participants. In the intensive BP group (goal systolic BP <120 mm Hg), as compared with a group with a goal of <140 mm Hg, antihypertensive therapy has been initiated or changed to the combination of a thiazide‐type diuretic and an ACE inhibitor, angiotensin receptor blocker, or β‐blocker, with drugs from other classes added to achieve the BP goal. The study is ongoing.

At present, I believe that the data from SHEP, the HOT study, and UKPDS are consistent with a systolic BP goal of ≈145 mm Hg; the ADVANCE trial findings would suggest lowering that goal to ≈135 mm Hg, or at least <140 mm Hg, but if the ACCORD BP trial demonstrates a reduction in major CVD outcomes, it would support recommending a systolic BP goal in patients with diabetes mellitus of <120 mm Hg. If the achieved systolic BP levels in the ACCORD study are ≈125 to 130 mm Hg, then perhaps there will be some debate about the specific goal level, but it should provide definitive evidence for lower goals than currently proven. If the ACCORD study results do not show CVD benefits for the lower BP goal, guidelines should be changed to reflect the evidence from ADVANCE supporting goal levels of systolic BP ≈135 mm Hg, or perhaps <140 mm Hg in diabetic hypertensive patients.

Why is it important not to recommend intensifying antihypertensive medication treatment to reduce BP below what we have definitively proven in trials? Although I do not believe it is likely, it is possible that treating to lower BP levels may be harmful (remember the J‐curve?). If it is neither beneficial nor harmful, however, then adding additional drugs is a waste of patients' and payers' resources and time and perhaps more importantly may contribute to reduced adherence to evidence‐based drug treatment regimens. This is especially important in patients with diabetes, who are often on a number of medications to improve glycemia or medications which have been proven to reduce cardiovascular risk, such as statins and aspirin. Prescribing additional, possibly unnecessary, antihypertensive medications may result in patients' choosing to not fill or take prescriptions for medications with more evidence for safety and benefit. Perhaps the ACCORD results will show us that the goal should be lower than 135 to 140 mm Hg and that it is worth the additional cost and effort to reduce CVD outcomes in our diabetic patients.