Automated office blood pressure measurement in the management of hypertension - fourth in series.

Background

In 2013, a joint task force of the European Society of Cardiology and the European Society of Hypertension published new guidelines for the management of hypertension (1). A highlight was the statement that “BP can no longer be estimated using a mercury sphygmomanometer in many - although not all - European countries. Auscultatory or oscillometric semi-automated sphygmomanometers are used instead”. A similar statement was made in the 2014 guidelines of the International Society of Hypertension/American Society of Hypertension (2): “The electronic device is preferred (to record BP) because it provides more reproducible results than the older (auscultatory) method and is not influenced by variations in technique or by the bias of observers”. These guidelines recognise several realities:

1. The use of mercury in healthcare settings in the European Community and elsewhere is now severely curtailed.
2. More importantly is the recognition that manual BP measurement, regardless of the type of sphygmomanometer used, is inferior because it is subject to multiple sources of error in routine clinical practice. 

Furthermore, automated devices which have been properly validated for accuracy are now readily available as an alternative to manual BP.

1. Use of semi-automated electronic sphygmomanometers in the office

Most semi-automated sphygmomanometers are basically home BP recorders adapted for use in the office. Some devices such as the Omron HEM 705CP were worked with in major clinical trials such as the ASCOT study (3). Semi-automated recorders generally call on the oscillometric method to record BP with some units also being capable of performing auscultatory readings if, for some reason, the oscillometric mode does not give a satisfactory BP. The devices are semi-automatic since they must be activated by the patient or health professional to take one or more readings. This approach eliminates sources of measurement error related to the observer and should provide a more accurate reading than with manual BP.
However, semi-automated devices still seem to provoke some white coat effect. Several studies (table 1) have compared semi-automated electronic office BP to ABPM or home BP. In each instance, the semi-automated office BP readings performed under research study conditions with strict adherence to BP measurement guidelines were higher than the awake ambulatory BP.

Table 1. Studies comparing semi-automated BP readings with awake ambulatory or home BP (8-11)

2. Automated office BP measurement (AOBP)

Considering that (self-) measurement of BP using a semi-automated device does not eliminate the white coat effect, Gelfer and colleagues at VSM Medtech in Vancouver, Canada postulated that a fully automated oscillometric sphygmomanometer might go one-step further by recording an accurate office BP with little or no white coat effect (8). The company went on to develop the BpTRU which is capable of automatically taking five BP readings with the patient resting quietly alone. An initial ‘test’ reading (not included in the mean of five readings) is added to verify that the cuff has been properly applied and that a valid BP reading is present. The BpTRU was validated (9) according to the BHS and AAMI standards. 
In studies (table 2) comparing AOBP using the BpTRU with awake ambulatory BP in family practice, hypertension clinic, ABPM unit and research setting, the AOBP was similar to the awake ambulatory BP.

Table 2. Studies comparing automated office BP with awake ambulatory BP

The correlation between the awake ambulatory BP and the AOBP is consistently stronger than the comparable manual office BP (table 3). AOBP also substantially reduces the digit preference (rounding off BP readings to the nearest zero value) associated with manual office BP (11,14).
Other aspects of AOBP have now been examined. AOBP readings taken either one or two minutes apart using the BpTRU are similar (15). Unlike manual BP, AOBP is similar in both the physician’s office and in non-treatment settings such as in an ABPM unit or pharmacy (16,17). 

Table 3. Coefficients of correlation for either routine manual BP or automated office BP versus systolic /diastolic awake ambulatory BP in different patient settings.

The elimination of the white coat effect is not simply due to the patient resting quietly alone for five minutes. No period of rest is required before the first AOBP reading is taken. Most of the fall in BP occurs within one minute of the physician leaving the examining room and higher BP levels return as soon as the physician re-enters the room (18). Other fully automated oscillometric sphygmomanometers such as the Omron HEM-907 produce AOBP readings which are comparable to the BpTRU (19).

3. A randomised controlled trial of AOBP in clinical practice

The Conventional versus Automated Measurement of BP in the Office (CAMBO) study was designed to evaluate the impact of AOBP measurement on the management of patients in routine, community-based, clinical practice (14). In this study, 67 practices (555 patients) in eastern Canada were randomised to either ongoing use of manual BP measurement (control group) or AOBP (intervention group) using the BpTRU. In ‘real-world’ practice, AOBP reduced office BP significantly more than the decrease in BP from baseline seen in the control, usual practice group. Introduction of AOBP in the intervention group reduced the routine manual pre-enrollment mean BP of 150/81 mmHg to 136/78 mmHg with the awake ambulatory BP being 132/74 mmHg. The CAMBO study also examined the impact AOBP might have on the prevalence of masked hypertension (normal office BP and hypertension on ABPM). In a sub-study (20) involving 140 patients in the AOBP intervention group and 112 patients in the control group who had no changes in drug therapy during the first three post-enrollment office visits, the prevalence of masked hypertension was significantly lower in AOBP patients than in the control patients who continued to be followed with manual BP measurements. The reduction in masked hypertension with AOBP occurred despite the AOBP (131/76 mmHg) being similar to the awake ambulatory BP (131/74 mmHg) whereas the manual BP was higher (138/79 mmHg).

4. Automated office BP and target organ damage

In a predominantly normotensive population (n=176), Campbell et al (21) noted a significantly stronger correlation between AOBP and the intima media wall thickness of the carotid artery compared to manual BP readings. Andreadis et al (22) examined the correlation between left ventricular mass index and awake ambulatory BP, AOBP and clinic BP recorded by research staff in 90 patients attending a hypertension clinic. The correlation with left ventricular mass index was significantly stronger for both the awake systolic ambulatory BP (r=0.37) and AOBP (r=0.37) compared to the clinic BP (r=0.12).  
The Cardiovascular Health Awareness Program (CHAP) used the BpTRU to screen BP status in older persons residing in 39 communities in the Province of Ontario, Canada (23). Communities were randomised to a pharmacy-based BP screening program using the BpTRU (intervention group) or to a control group. Residents who underwent BP screening with the BpTRU results reported to their own physicians had significantly fewer cardiovascular events during the subsequent year than those who were not screened for BP status. CHAP is the only BP screening study to report that BP screening reduces cardiovascular morbidity and mortality.

5. Automated BP in population surveys

In Canada, the BpTRU has been used in two large-scale surveys assessing the BP status of residents in the community (24,25). In the Ontario Survey on the Prevalence of High Blood Pressure (ON-BP), 2,551 subjects had their BP recorded using the BpTRU. The Canadian Health Measures Survey also included an automated BP by the BpTRU in the 3,943 participants randomly selected from the community. The findings in both surveys were similar with the prevalence of hypertension being unchanged from the early 1990’s but control rates had improved markedly to about 65%. The use of the AOBP technique has several advantages over manual BP, in particular greater accuracy and consistency. Moreover, use of the automated device precludes the need for intensive training of observers in auscultatory BP measurement and repeated testing to ensure consistent manual BP readings during the period of the survey. 

6. Rationale for using AOBP instead of semi-automated electronic sphygmomanometers 

The ESH/ESC guidelines (1) state that when feasible, AOBP should be considered as the preferred electronic sphygmomanometer because it improves reproducibility and provides office BP readings which are close to the awake ambulatory BP and home BP. The feasibility of AOBP is no longer an obstacle to its use in clinical practice. AOBP can be taken at one minute intervals without having the patient resting for five minutes before the first AOBP reading (26). Resting for one minute or perhaps even less before the first AOBP is sufficient. Thus, AOBP readings should take no longer than a proper manual BP, providing that the patient rests for five minutes before the first manual BP reading, as recommended in the guidelines.
The initial recommendation for AOBP stated that the readings should be taken with the patient resting alone in a quiet room. More recent reports suggest that the patient only needs to be left alone in a quiet place which could be a waiting area or other location in the office. AOBP recorded in the office waiting room, an ABPM unit or in a community pharmacy were similar to AOBP readings obtained with the patient alone in the examining room (16,17,27). 
The relatively high cost of devices available for recording AOBP compared to semi-automated home BP sphygmomanometers adapted for professional use has been a limitation of AOBP. The BpTRU, Omron-HEM 907 and Microlife WatchBP Office are indeed relatively expensive (US$ 600-1000), especially if several recorders are needed for one practice. The recent introduction of lower cost ‘basic’ devices specifically designed for AOBP has brought the cost of adopting AOBP into clinical practice much closer to the semi-automated electronic devices which require activation of the recorder by the patient or health professional.
Thus, the time required for recording AOBP, the need for a separate examining room and the cost of AOBP devices are no longer limiting factors in the use of AOBP in clinical practice.


Table 4: Principles of automated office BP measurement

Principles of automated office BP measurement
1.Use fully automated, validated, electronic sphygmomanometer
2.Device should take multiple readings without requiring activation of readings by the patient or health professional
3.Patient should be resting in a quiet place without possibility of conversation, preferably alone in a room

7. Anticipated future Research Involving AOBP

Studies to date have documented the advantages of AOBP over manual and semi-automated electronic sphygmomanometers when used in the office setting. Based upon studies comparing AOBP with the awake ambulatory BP and home BP, it appears that all three methods of measurement should have the same cut-point for diagnosing hypertension, 135/85 mmHg. Ideally, there should be clinical outcome data relating AOBP to cardiovascular end-points in order to confirm the AOBP reading for defining hypertension. However, it should be noted that the relationship between BP and cardiovascular risk is a continuous one with any cut-points separating normal from high BP being somewhat arbitrary. After all, there is still disagreement as to the appropriate value for defining hypertension using ABPM, even after numerous outcome studies conducted over two decades (28).
More population surveys need to be performed using AOBP with sub-sets of the subjects also having manual BP taken in order to achieve a better understanding of the relationship between AOBP readings and manual BP in this setting.
AOBP has already been used in clinical trials examining pharmacotherapy in hypertension (29) and more such studies are anticipated.

Conclusion

Involvement of humans in recording BP is a major source of measurement error and bias. Regardless of the type of sphygmomanometer used, BP readings taken by doctors and nurses tend to be higher due to increased patient anxiety, conversation with the patient, digit preference, single readings and non-standardised measurement technique. The simple act of a patient pushing a button to activate a semi-automated electronic recorder also appears to increase BP, even if the patient is alone.
The replacement of manual sphygmomanometers and semi-automated electronic devices with AOBP eliminates many of the causes of inaccurate office BP measurement and provides a reading which closely approximates the awake ambulatory BP and home BP, gold standards for predicting future cardiovascular risk associated with hypertension. Given that an electronic device is now recommended for office BP measurement, AOBP should become the standard method for obtaining BP readings in clinical practice with only a few simple procedures needed to be followed (table 4). 
Thus, AOBP offers the opportunity to maintain an important role for office BP in the management of hypertension. The combination of ABPM to make a definitive diagnosis of hypertension with AOBP and home BP to detect hypertension and to monitor the progress of patients on drug therapy brings the management of hypertension into the twenty-first century.