In Type 2 Diabetes, How Low Should Blood Pressure Go?

Is Intensive Blood Pressure Control Beneficial in High-Risk Patients?

Numerous clinical trials and epidemiologic analyses have established that lowering blood pressure reduces the risk for macrovascular and microvascular diseases among people with diabetes.^[1] However, optimal systolic blood pressure (SBP) targets have not been agreed upon.

Buckley and colleagues^[2] performed a post hoc analysis of ACCORD-BP participants who were in the standard glucose control arm and met inclusion criteria for SPRINT (additional cardiovascular disease [CVD] risk factors). The intensive SBP control arm (target < 120 mm Hg) was compared with the standard SBP control arm (< 140 mm Hg) on risk for a composite outcome of CVD death, nonfatal myocardial infarction (MI), nonfatal stroke, revascularization, and heart failure, as well as each of those components individually. Adverse events were also compared.

The main analysis focused on SPRINT-eligible ACCORD-BP participants, all of whom had diabetes. A secondary analysis pooled these people with actual SPRINT participants, none of whom had diabetes, to test whether the results differed between those with and those without diabetes.

Intensive Control Shows a Benefit

A total of 1284 individuals met inclusion criteria, about one half of whom (n = 652; 51%) were in the intensive SBP control arm. Characteristics and risk factors in the two groups were essentially identical.

Mean SBP achieved in the intensive SBP control group was 120 mm Hg and 134 mm Hg in the standard SBP control group. Intensive SBP control significantly reduced risk for the composite outcome by 21%. These results were driven by nonfatal MI (31% reduction), and nonfatal stroke and heart failure also trended toward significance.

There was no difference between the groups in cardiovascular or all-cause mortality. Adverse events were about twice as common in the intensive SBP control arm but affected < 5% of the group.

Do Benefits Differ Between Patients With Normoglycemia and Those With Prediabetes?

In a post hoc analysis of the aforementioned SPRINT study,^[3] participants (none of whom had diabetes) were stratified on the basis of fasting plasma glucose level at baseline; normoglycemia was defined as a fasting plasma glucose level < 100 mg/dL and prediabetes as ≥ 100 mg/dL. The main objective was to determine whether the benefits of intensive SBP control reported in SPRINT were equivalent by glycemic status.

As in the original SPRINT study and the above report by Buckley and colleagues,^[2] intensive SBP control (target < 120 mm Hg) was compared with standard SBP control (< 140 mm Hg) on risk for a composite of CVD death, nonfatal MI, nonfatal stroke, revascularization, and heart failure, as well as each of those components individually.

Investigators also examined renal outcomes by presence of chronic kidney disease at baseline, as well as serious adverse events by treatment arm. A total of 3898 participants (42%) had prediabetes, and 5425 (58%) had normoglycemia.

The mean achieved SBP in the intensive SBP control group was 122 mm Hg among patients with normoglycemia and 121 mm Hg among those with prediabetes. In the standard SBP control arm, the mean achieved SBP was 136 mm Hg in both glycemic groups.

The risk reductions in the primary composite associated with intensive SBP control were 31% and 17% among those with prediabetes and normoglycemia, respectively, with a nonsignificant interaction indicating that the risk reductions did not differ. Risk reductions by glycemic status for all-cause mortality were 23% for patients with prediabetes and 29% for normoglycemic patients; the P value for the interaction was nonsignificant.

In fact, no interaction term was significant for any of the individual components of the composite or for other prespecified secondary outcomes, including renal outcomes. The proportions of patients experiencing serious adverse events were essentially identical by treatment arm and glycemic status.

Lessons for Clinical Practice

In a post hoc analysis of the aforementioned SPRINT study,^[3] participants (none of whom had diabetes) were stratified on the basis of fasting plasma glucose level at baseline; normoglycemia was defined as a fasting plasma glucose level < 100 mg/dL and prediabetes as ≥ 100 mg/dL. The main objective was to determine whether the benefits of intensive SBP control reported in SPRINT were equivalent by glycemic status.

As in the original SPRINT study and the above report by Buckley and colleagues,^[2] intensive SBP control (target < 120 mm Hg) was compared with standard SBP control (< 140 mm Hg) on risk for a composite of CVD death, nonfatal MI, nonfatal stroke, revascularization, and heart failure, as well as each of those components individually.

Investigators also examined renal outcomes by presence of chronic kidney disease at baseline, as well as serious adverse events by treatment arm. A total of 3898 participants (42%) had prediabetes, and 5425 (58%) had normoglycemia.

The mean achieved SBP in the intensive SBP control group was 122 mm Hg among patients with normoglycemia and 121 mm Hg among those with prediabetes. In the standard SBP control arm, the mean achieved SBP was 136 mm Hg in both glycemic groups.

The risk reductions in the primary composite associated with intensive SBP control were 31% and 17% among those with prediabetes and normoglycemia, respectively, with a nonsignificant interaction indicating that the risk reductions did not differ. Risk reductions by glycemic status for all-cause mortality were 23% for patients with prediabetes and 29% for normoglycemic patients; the P value for the interaction was nonsignificant.

In fact, no interaction term was significant for any of the individual components of the composite or for other prespecified secondary outcomes, including renal outcomes. The proportions of patients experiencing serious adverse events were essentially identical by treatment arm and glycemic status.

Lessons for Clinical Practice

Almost 20 years ago, the United Kingdom Prospective Diabetes Study^[4] demonstrated that "tight" blood pressure control among patients with diabetes reduced the risk for macrovascular and microvascular diseases, using target levels for tight versus standard SBP control that would not be considered acceptable today.

Multifactorial studies^[5,6] that simultaneously examined SBP control and low-density lipoprotein cholesterol control suggested that very low targets could be beneficial, but then the ACCORD-BP study^[7] found that targeting SBP < 120 mm Hg did not reduce the rate of a composite of fatal and nonfatal CVD events compared with SBP < 140 mm Hg; that finding was supported by a subsequent meta-analysis.^[8] As a result, the American Diabetes Association (ADA) relaxed its guidelines for SBP control in 2013.^[9]

The results of the recent SPRINT study,^[10] which showed a substantial benefit of intensive blood pressure control among people without diabetes, left many wondering why the results would not extend to people with diabetes. Aside from the fact that SPRINT excluded people with diabetes, SPRINT participants had to be at considerable CVD risk (history of CVD, chronic kidney disease, or 10-year Framingham Risk Score ≥ 15%).

The ACCORD criteria were not as selective. Buckley and colleagues^[2] suggest that their results might be due to a more pronounced effect of intensive SBP control when risk is greater (owing to the SPRINT criteria), and that could indeed be part of the explanation. It could also be true that the effects of a difference in SBP of 15 mm Hg do not fully manifest in a mean follow-up of approximately 5 years, an explanation hypothesized in the original ACCORD-BP results.^[7]

It should also be noted that although the initial ACCORD results did not show a benefit of intensive versus standard SBP treatment, further planned analyses found that risk for major CVD was lower in the intensive SBP control group (with or without intensive glycemia treatment) compared with the standard SBP control/standard glycemia group.^[11] Thus, the ACCORD and SPRINT results may not be so different after all.

The results from Bress and colleagues^[3] further suggest that the benefit of intensive SBP control is uniform across different levels of glycemia, at least at levels below the threshold for diabetes. It defies logic that crossing the threshold would cause the benefit to be lost.

The only remaining question, then, is whether tight SBP control can be achieved safely. In addition, there was no difference in rates of adverse events. However, Buckley and colleagues^[2] found adverse event rates to be nearly twice as high in the intensive therapy arm. Although that may sound alarming, absolute rates were < 5%, so the large majority of patients were not affected by intensive therapy.

Current ADA guidelines call for an SBP target of < 140 mm Hg but acknowledge that lower targets, such as < 130 mm Hg, may be appropriate if they can be achieved without undue treatment burden.^[12] How easily lower levels can be achieved is of course an important consideration, but taken together, existing data along with the two studies reviewed here suggest that intensive SBP control targeting levels < 120 mm Hg should probably be attempted in most patients with diabetes.