Blood Pressure Target in Diabetics: How Low Is Too Low?

Marwan Saad; Negar Salehi; Zufeng Ding; Jawahar L. Mehta


Eur Heart J. 2019;40(25):2044-2046. 

The prevalence of hypertension in patients with diabetes is substantial, reaching up to three times higher compared with patients without diabetes.[1] The high prevalence of hypertension in patients with diabetes is multifactorial. Excess caloric intake, obesity, and sedentary lifestyle are major risk factors for the development of metabolic syndrome and insulin resistance, while oxidative stress and inflammation play a pivotal role at the cellular level.[2] In patients at early stages of diabetes, hypertension can occur as a result of hyperinsulinaemia-induced increase in body fluid volume, and at later stages as a result of vascular remodelling as well as diabetic nephropathy and subsequent chronic stimulation of the renin–angiotensin–aldosterone system (RAAS).[3]

Patients with type 2 diabetes are at higher risk of cardiovascular complications including coronary artery disease, cerebrovascular disease, peripheral artery disease, and heart failure compared with non-diabetics.[4] Furthermore, it is well established that patients with both hypertension and diabetes are at higher risk of atherosclerotic cardiovascular disease and chronic kidney disease compared with normotensive diabetics.[5]

The optimal blood pressure (BP) target to reduce adverse cardiovascular outcomes, and in particular in patients with diabetes, remains unclear. While the 2017 American College of Cardiology/American Heart Association (ACC/AHA) Hypertension Guidelines recommend a systolic BP (SBP) target of <130 mmHg in almost all hypertensive patients,[6] the 2018 European Society of Cardiology/European Society of Hypertension (ESC/ESH) Guidelines recommend a BP target of <140/90 mmHg (and ≤130/80 mmHg only if further tolerated).[7] For patients with diabetes, the target BP is even less clear, with ongoing societal discrepancy within the USA.[8] Shin et al., utilizing data from the National Health and Nutrition Examination Survey (NHANES) from 2005 to 2014, demonstrated that the application of different BP thresholds from the ACC/AHA (130/80 mmHg) and the American Diabetes association (ADA) (140/90 mmHg) would lead to the reclassification of >10% of adults with diabetes in the USA as having or not having hypertension.[8]

In view of the potential J-curve relationship between BP and outcomes, one of the ongoing dilemmas continues to be whether intensive BP control ought to be recommended over the standard control, with older studies providing conflicting evidence especially in the diabetic population.[9,10] The ACCORD Study in 2010 addressed this question in 4733 patients with type 2 diabetes at high risk for cardiovascular events randomly assigned to intensive BP therapy (SBP <120 mmHg) vs. standard BP therapy (SBP <140 mmHg).[11] The study found similar rates of the primary composite outcome of non-fatal myocardial infarction, non-fatal stroke, or death from cardiovascular causes between both groups, however with lower rates of non-fatal stroke in the intensive therapy group at the expense of serious adverse events related to anti-hypertensive therapy. A few years later, the ACCORD study results were challenged by the SPRINT study.[12] SPRINT examined the difference in outcomes of myocardial infarction, stroke, heart failure, and death from cardiovascular causes in >9000 patients assigned to intensive vs. standard BP control (with targets for the two arms similar to those in the ACCORD trial). The trial included populations at high risk of cardiovascular disease; it did not, however, include patients with diabetes. The trial was prematurely terminated due to significantly lower rates of fatal and non-fatal major cardiovascular events and all-cause death in the intensive therapy arm. The results of the SPRINT trial formed the basis of the current recommendations of lower BP target in all individuals by the ACC/AHA Guidelines.

The plausible causes of discrepancy between the results of the ACCORD and SPRINT trials include a lower separation in achieved systolic BP in ACCORD vs. SPRINT, as well as lower achieved SBP in a higher percentage of standard-group participants in ACCORD, causing a lower event rate in that group compared with SPRINT.[13] An even more important factor is the difference in the patient population in the two studies. The intense inflammation, oxidative stress, and up-regulation of the RAAS probably contributed to a more difficult control of BP, as well as a higher rate of adverse cardiovascular outcomes at any level of BP in diabetic patients compared with non-diabetics.[3,13] It is important to note that both ACCORD and SPRINT demonstrated a largely similar relationship between SBP and adverse cardiovascular outcomes despite the differences in patient population.

The study by Böhm and colleagues in this issue of the European Heart Journal[14] examined the impact of achieved SBP and diastolic BP (DBP) levels on cardiovascular outcomes in an at high cardiovascular risk population with and without diabetes, through an analysis of nearly 31 000 patients from the ONTARGET and the TRANSCEND trials—patients being treated with the angiotensin receptor blocker telmisartan or the angiotensin-converting enzyme blocker ramipril. These two major trials recruited patients with a history of coronary artery disease, peripheral artery disease, transient ischaemic attack, stroke, or diabetes mellitus complicated by end-organ damage, but without symptomatic heart failure at entry. Patients were followed up for a median of 4.6 years. This analysis showed that: (i) mean SBP achieved after treatment was ~10 mmHg higher in diabetics, as shown in the authors' Take Home figure; (ii) regardless of the achieved SBP or DBP, diabetic patients had worse outcomes compared with non-diabetics at all levels of BP, with even higher event rates if diabetes is accompanied by end-organ damage; (iii) diabetes, SBP, and DBP all influenced outcomes, and the interaction between diabetes and BP (low and high, SBP and DBP) was additive; and (iv) compared with a SBP of 120 to <140 mmHg and a DBP of 70 to <80 mmHg, worse cardiovascular outcomes were observed with either high BP (SBP ≥140 or DBP ≥80 mmHg) or low BP (SBP <120 or DBP <70 mmHg) (Figure 1), an effect that was consistent when taking non-diabetics as the reference group, or even within the diabetic population only.

Figure 1.

Hazard ratios of outcomes in patients with diabetes and achieved systolic and diastolic BP (reference population are patients with no diabetes and SBP from 120 to <140 mmHg and DBP from 70 to <80 mmHg).

The results of this study should be interpreted with caution in view of its limitations, mainly being an observational analysis that carries an inherent risk of bias. However, with its large patient sample, comprehensive analysis, and long-term follow-up, the study challenges the current practice guidelines. The worse outcome in patients with diabetes at all BP levels compared with those without diabetes reflects the additive impact of diabetes and hypertension and explains why an adequate BP control in such patients is crucial. However, the J-curve relationship between BP and cardiovascular outcomes, with worse outcomes observed at low BP (SBP <120 or DBP <70 mmHg), remains a significant concern for intensive BP control, and supports previously reported data.[15] Intensive BP control, especially DBP, can be linked to impaired coronary flow, particularly in the presence of significant coronary artery disease.[16] In elderly patients with increased arteriolar stiffness, intensive lowering of DBP and widening of pulse pressure can be detrimental.[17] This requires caution in the BP targets in this population. Whether the observed outcome in the current study is related to reverse causality, although this is less likely, cannot be completely excluded.

In our opinion, physicians should implement the Guidelines' recommendations for BP control cautiously in their patients, taking into account patient age and baseline risk factors such as diabetes and chronic kidney disease, and tailor the desired magnitude of BP reduction based on drug tolerability rather than following strict BP targets, in order to achieve the BP goal without imposing an increased cardiovascular risk that may be linked to aggressive BP control.