Hypertension has been defined by the levels of blood pressure (BP) above which lowering BP will reduce the cardiovascular risk associated with elevated BP. This level has been classically 140/90mm Hg on the basis of actuarial data from the insurance industry. However, we now know that cardiovascular risk rises progressively from levels as low as 115/75mm Hg upward with a doubling of the incidence of both coronary heart disease and stroke for every 20/10mm Hg increment of BP. 1 Accordingly, the concept of prehypertension was introduced for systolic BPs (SBPs) between 120 and 140mm Hg by the Seventh Report of the Joint National Committee on the Prevention, Detection, Evaluation and Treatment of High Blood Pressure (JNC7). 2 These levels of BP are associated with increased cardiovascular risk, 3 but there was until recently no evidence that lowering BP below 120mm Hg decreased cardiovascular risk further. Furthermore, there is little in the form of randomized clinical trials that demonstrates that in uncomplicated hypertension without cardiovascular risk factors or target organ damage, lowering SBP of <160mm Hg reduces cardiovascular risk. 4 However, this has been challenged by a recent systematic review and meta-analysis. 5
The Systolic Blood Pressure Intervention Trial (SPRINT) trial published recently represents a major advance in our knowledge on how to manage hypertensive patients. 6 It demonstrated in hypertensive patients at high risk of cardiovascular disease with SBP >130 and <180mm Hg, who were randomly assigned to be treated to a goal SBP of <140 vs. <120mm Hg, that indeed lowering SBP to <120mm Hg was very beneficial. By 1-year postrandomization, the trial achieved SBPs of 121.4mm Hg in the intensive treatment group and 136.2mm Hg in the standard treatment group, a difference that was maintained throughout the trial. After a median follow-up of 3.26 years, there was a 25% relative risk reduction of the primary endpoint (myocardial infarction or other coronary syndromes, stroke, heart failure, or death from cardiovascular causes) and 27% relative risk reduction in all-cause mortality in the intensive treatment group. Patients in the intensive therapy group took a mean of approximately 3 vs. 2 antihypertensive drugs in the standard therapy group. Patients who entered the study with SBP of <132mm Hg had better outcomes than those who entered the study with SBP >132mm Hg. Importantly, in SPRINT, BP was measured in the office with automated devices (automated office BP or AOBP), unattended, and with a 5-minute rest period.
On the basis of these data, some organizations have already adapted recommendations for management of hypertension. The online text UpToDate recommended shortly after the publication of the SPRINT lowering SBP in high-risk individuals to 125–130mm Hg if BP is measured manually or 120–125mm Hg if it is measured in the office with an automated device (AOBP). 7 Hypertension Canada in its newest Canadian Hypertension Education Program guideline 8 , 9 has recommended for high-risk nondiabetic subjects aged ≥50 years, with systolic BP levels ≥130mm Hg, intensive management to target a systolic BP ≤120mm Hg (measured with unattended AOBP), and caution in extending these lower targets to non-SPRINT patients. This is particularly important for diabetic hypertensive individuals, considering the fact that the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial did not provide evidence in favor of intensive lowering of BP for this group of high-risk patients. 10 Potential reasons for the discrepancy between ACCORD and SPRINT have already been discussed in a previous Editorial in the journal. 11 However, it should be noted that in ACCORD, BP was measured with observed AOBP. Thus, the intensive treatment arm goal BP of <120mm Hg in ACCORD may be more aggressive than the intensive treatment arm goal BP of <120mm Hg in SPRINT measured with unattended AOBP, as discussed below.
These data, and recommendations thereof, beg the question whether we should change our classical definition of hypertension from ≥140/90mm Hg to perhaps ≥130/80mm Hg. As well, should we have a definition that applies to high-risk hypertensive individuals, including or not diabetic persons with hypertension, and a different one for those who are at less cardiovascular risk?
As pointed out already, 11 the method of BP measurement is critically important when discussing BP values. The recommendation of UpToDate 7 is clear in its differentiation of BP measured in the office manually or with an automated device. The Canadian Hypertension Education Program guideline from Hypertension Canada recommends the use of unattended AOBP measurement as the preferred office BP measurement, in addition to ambulatory BP monitoring and home BP measurement. 8 , 9 For the latter, we do not have BP values from SPRINT. A systolic BP of 120mm Hg measured with AOBP may be read as 130mm Hg if the BP is measured manually, which explains the recommendation made in UpToDate. 7 Indeed, when we examine the evidence regarding BP measurement by different methods, it has been demonstrated that unattended AOBP readings are preferable to conventional BP measurements in routine clinical practice because they are more accurate. Since in AOBP measurements the white coat effect is eliminated, BP readings are close and may even be lower than daytime ambulatory BP monitoring and home BP monitoring.
Accordingly, we should qualify any BP values by the methodology used to measure the BP. In SPRINT, the lowering of SBP to <120mm Hg (in fact to a mean of 121.4mm Hg) 6 may correspond therefore to a reduction of SBP to <130mm Hg if measured manually as usually done in previous antihypertensive trials. Manual research quality BP measurement is close to daytime ambulatory BP monitoring and home BP measurements, so the values of BP from SPRINT probably mean that in high-risk hypertensive individuals SBP should be lowered to an office SBP of <130mm Hg if research quality measurement is performed or if daytime ambulatory BP monitoring or home BP measurement is being used. However, if the BP is measured in a busy clinic in general practice without unattended AOBP measurement in a quiet room and following a guideline-directed protocol, but rather as manually measured BP and exposed to the activity of the clinic, it is probable that the <120mm Hg by unattended AOBP from the SPRINT study is equivalent to <135–140mm Hg under such conditions, since there may be a difference of up to 15–20mm Hg between manual BP in usual clinical practice and unattended AOBP. 12–14
This would suggest that in fact even for high-risk patients recommendations of thresholds and goals of BP should not be very different from present ones as long as we are careful to adapt them to the method that is used for measurement of BP, since how you measure BP is evidently going to provide very different values.
This will be particularly important for the elderly to avoid overaggressive treatment that could result in harm. Hypertensive individuals above age 75 years benefited more than younger individuals in the SPRINT trial. 6 Recent data from SPRINT suggest that even elderly individuals who are frail as measured by a variety of scores benefited as much as fit elderly subjects. 15 Although in SPRINT there was no increase in injurious falls, if primary care physicians start lowering BP of the elderly intensively without using the AOBP as used in SPRINT, it is likely that achieved SBP measured manually in a busy clinical practice setting could be equivalent to SPRINT SBPs of <110mm Hg, which was an excluding condition in SPRINT, and could be associated with harm, including falls and hip fractures, etc.
Finally, there is the question of high- vs. low-resource healthcare systems. If maintaining very strict intensive control of BP requires very frequent follow-up and increased medication, in low- and middle-income countries where already cardiovascular drugs are not all present in the list of national essential medicines, 16 where access to physicians or other healthcare professionals is quite limited, and where AOBP is unavailable, it is unlikely that it will be possible to achieve these levels of control of BP. In fact, we have already suggested that in some of these countries BP should be controlled with whichever antihypertensive medications are available and that some lowering of BP should be attempted even if it is impossible to achieve the recommended goals, since any minor lowering of BP will be beneficial for those patients. 17
Which leads us to the question we started with: Is it time, after SPRINT, to introduce a new definition of hypertension? From the previous discussion, it would seem that since AOBP is not yet generally available, the definition of hypertension should remain BPs ≥140/90mm Hg, and that the goal should remain to lower BP in most hypertensive patients to <140/90mm Hg, and in those at higher cardiovascular risk, including chronic kidney disease patients, elderly individuals and those at a Framingham Risk Score of ≥15%, and perhaps as well diabetic subjects, goal BP should be <130/80mm Hg. The values of BP noted above are for guideline-directed manual BP measurement performed in usual clinical practice. Furthermore, any mention of BP values should be accompanied by a detailed identification of the method of BP measurement. However, if BP is measured with guideline-directed unattended AOBP following a 5-minute rest period as in SPRINT, hypertension in high-risk subjects could be defined as BP ≥130/80mm Hg with a threshold for treatment of 130/80mm Hg and a goal SBP of ≤120mm Hg.