More reasons to use automated office blood pressure in clinical practice

In the 1990's, it became evident from longitudinal clinical outcome studies that 24‐hour ambulatory blood pressure monitoring (ABPM) was a significantly better predictor of future cardiovascular events than manual office BP performed carefully according to standard guidelines.1 Thereafter, ABPM became a “gold standard” for evaluating the accuracy of different techniques of BP measurement. Comparative studies also reported that office BP was generally higher than the ambulatory BP, resulting in different thresholds for defining hypertension, 140/90 mm Hg for the office BP and 135/85 mm Hg for the awake ambulatory (A)BP.1 The underlying assumption was that office BP in clinical practice was the same as readings obtained in research studies. After all, guidelines for recording BP had for decades provided detailed instructions on how to measure BP, including having the patient seated with feet on the floor, back supported, legs not crossed, upper arm at heart level, and no conversation.

However, a landmark study2 by Beckett and Godwin in 2005 became the first indication that the “emperor had no clothes” with respect to office BP. These authors reported that mean office BP recorded in 481 patients in routine clinical practice was 10/3 and not 5/5 mm Hg higher than the mean awake ABP. Besides challenging this widely held assumption, Beckett and Godwin's study also provided a solution by reporting that a new technique for the measurement of BP in the office, now referred to as “automated office BP” (AOBP), gave a mean reading which was similar to the awake ABP. The advantages of AOBP over routine office BP were subsequently recognized in the 2011 Canadian Hypertension guidelines3 and became the recommended technique for performing office BP4, 5 in Canada (2016) and in the United States (2019).

The key to obtaining a more accurate office BP using AOBP was having the patient resting alone in a quiet place while several readings were performed using a fully automated, oscillometric sphygmomanometer. Studies in both a research setting6 and routine clinical practice7 showed that the first office systolic BP recorded within 2 minutes of the patient being left alone was already decreased by a mean of 8‐15 mm Hg. The only factor distinguishing this first AOBP reading from a conventional office BP was that the patient was alone and could not speak with office staff. All the other aspects of performing the office BP readings were unchanged. Thus, conversation and the presence of nurses or doctors are likely the most important causes of white coat hypertension.

The 2019 AHA guidelines for BP measurement5 which recommended AOBP for use in clinical practice are supported by a recent meta‐analysis.8 In 19 studies, mean systolic AOBP was only 0.3/1.0 mm Hg different from the mean awake ABP. In 9 studies, mean routine office systolic BP was 14.5 mm Hg higher than the AOBP. Mean office systolic BP performed according to guidelines in research studies was also 7.0 mm Hg higher than the AOBP. Although a post hoc analysis of data from SPRINT9 suggested that AOBP could be performed in the presence of office staff, more recent studies have not supported this hypothesis.10

The use of an automated sphygmomanometer instead of a mercury device was also not a factor in reducing the white coat effect seen with routine office BP. In 27 211 hypertensive patients followed by primary care physicians in Spain, the mean of 2 automated office BP readings was 25/11 mm Hg higher than the awake ABP.11

Elsewhere in this issue, Andreadis et al12 have discussed how AOBP can be implemented into clinical practice. In some respects, the authors have been somewhat cautious in their approach. For example, not all devices require 5 minutes of rest before the first AOBP reading. AOBP readings similar to the awake ABP have been obtained with as little as a 1 minute delay between the patient being left alone and the first of 3 readings recorded 1 minute apart.13 Similarly, with no antecedent rest, the mean of 5 AOBP readings recorded about 30 seconds apart using the BpTRU was also similar to the mean awake ABP.8 This aspect of AOBP is important, since a frequent criticism of AOBP is that it takes too long to perform, which is only true if patients are not rested for 5 minutes before recording a conventional office BP, as recommended in the guidelines.

Others have also stated that AOBP is impractical because limited office space does not permit the patient to rest alone in a quiet place.14 In such circumstances, there would also not be anywhere in the office for patients to sit quietly for 5 minutes before conventional BP readings are obtained. Concerns about space and the time taken to perform AOBP likely explain why a white coat effect is currently so common when office BP is recorded in clinical practice.

AOBP has also been criticized because there are not sufficient clinical outcome data to support its use for diagnosing hypertension.15 Andreadis et al12 address this misconception by clearly stating that when AOBP identifies a patient with possible hypertension, the diagnosis should be confirmed by ABPM or home BP. AOBP is primarily for hypertension screening and is not recommended as a replacement for out‐of‐office BP readings. The ESC/ESH guidelines16 use 140/90 mm Hg as the threshold for diagnosing hypertension with conventional office BP. However, if the white coat effect seen in clinical practice is taken into account, the true value should be 150/95 mm Hg, whereas the available evidence supports having the value of 135/85 mm Hg for AOBP, awake ABP and home BP.8 The AHA threshold equating office BP, awake ABP and home BP at 130/80 mm Hg may not be optimal, but it still better than using 140/90 mm Hg.17

Andreadis et al have made important contributions to our understanding of AOBP and the advantages of its use in clinical practice. Hopefully, the evidence favoring AOBP will lead to its achieving greater acceptance in guidelines which currently still recommend conventional office BP measurement.

CONFLICT OF INTEREST

No conflicts of interest to declare.