In his original commentary,1 Dr Bakris provides a careful review of the origins of, the tradition behind, and the evidence base of blood pressure (BP) control goals in diabetes. As both his article and my own commentary discuss, there are few data that specifically address the issue of ideal treatment targets. The Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial is indeed unique in this capacity.2 Dr Bakris provides an excellent summary and admits that past guidelines3 were derived from a lower level of evidence. This is certainly the case but does not fully address the potential negative impact of reverting to BP targets of <140 mm Hg.
The Socratic Method cannot in and of itself resolve this conflict. However, Socrates used an approach—by answering questions to stimulate critical thought and ensure that all consequences of an outcome were considered—that we should embrace. Are the data from ACCORD and other studies overwhelmingly in favor of a categoric BP target of 140 mm Hg (vs 135 mm Hg or 133 mm Hg or even 130 mm Hg)? And what are the possible unintended consequences of applying a global standard (whatever the target) in the setting of an imperfect and complex dataset? In my original discussion, I discuss the same datasets considered by Dr Bakris. Those data support that modest, but incremental, benefits are observed when mean BP values approach 130 mm Hg as was seen in ACCORD (with achieved systolic BP in the standard group of 133 mm Hg) and in Action in Diabetes and Vascular Disease: Preterax and Diamicron-MR Controlled Evaluation (ADVANCE) where BP averaged 135 mm Hg in the more intensive treatment arm.2,4 With these data in mind, one can readily accept the argument that mean values <140/90 mm Hg are adequate for protection from a majority of cardiovascular (CV) events. Yet, in both studies, patients treated to lower targets (119 mm Hg in ACCORD and 135 mm Hg in ADVANCE) had lower rates of macroalbuminuria. In ACCORD (as noted by Dr Bakris) lower systolic BP was also associated with a significant reduction in the risk of stroke. To accept a higher target is to discard the potential importance of these clinical events and the protection afforded by lower BPs.
Furthermore, to accept the arguments of Dr Bakris, one would also require even better predictive methods for specific complication—somehow knowing in advance who, among the millions of adults affected by diabetes, would be at highest risk for stroke and progression of renal disease, as these are the only groups for whom the lower target may be most beneficial. While this is admittedly purist and is supported by the data from ACCORD, the African American Study of Kidney Disease (AASK)5 and other studies showed, in an evidence-based fashion, that a significant fraction of patients in the minority would be exposed to undue risk.
Noting that substantial time, effort, and resources are required by both clinician and patient to achieve BP values well below 130 mm Hg (as in the ACCORD intensive cohort), most studies (including ACCORD in the standard treatment group) support that reasonable BP control to values well below 140 mm Hg can be safely achieved with simple two-drug therapy and standardized treatment milestones (eg, any BP >140 mm Hg).2 As such, what undue risk is avoided with 140 mm Hg as a minimally acceptable standard (which the data do generally support) or must we consider that to achieve values in the mid to low 130-mm Hg range (arguably not <130 mm Hg) a lower target is reasonable to protect even more individuals and can be safely and readily achieved?
As future groups gather to consider the evidence and reassess targets, we must consider (as has been the case with A1c targets and low-density lipoprotein cholesterol)2 whether treatment standards are intended to identify a minimally acceptable threshold—in which case a BP target of <140 mm Hg is adequate—or whether such targets represent a marker of optimized care where a majority of complications (including stroke and microvascular disease) are significantly reduced. Dr Bakris provides detailed support for such a standard. However, in our clinical world, treatment decisions must be individualized and acquiescence to a higher target for many components of risk in diabetes may lead to poorer control of all risk factors. Far too much progress has been made in diabetes, with evidence from recent studies suggesting that our current approach to diabetes care has resulted in lower rates of serious diabetes-related complications.6 Indeed, the beneficial effects on renal disease and mortality of lower BP and glucose control appear to be additive.7 These outcomes6 have been achieved as life expectancy for patients with diabetes has increased. To revert to higher targets for optimized care at this point could be considered regressive. The evidence base from ACCORD, African-American Study of Kidney Disease and Hypertension (AASK), ADVANCE, and many other trials support a marginal BP target of <140 mm Hg, but do not alone provide evidence to completely alter optimized treatment targets, nor does it provide adequate guidance for and protection of patients at highest risk for other complications such as stroke and renal disease. If such a change is to be effected, it must be done with clear language to support the care of patients at risk for complications beyond CVD. Maintaining a BP target well below 140 mm Hg with guidance on rational, safe, and simple approaches to combination antihypertensive therapy, as was done in ACCORD, should remain the standard of care.2