Education And Debate

Use and interpretation of ambulatory blood pressure monitoring: recommendations of the British Hypertension Society

BMJ 2000; 320 doi: http://dx.doi.org/10.1136/bmj.320.7242.1128 (Published 22 April 2000) Cite this as: BMJ 2000;320:1128
  1. Eoin O'Brien, professor, cardiovascular medicine (eobrien{at}iol.ie)a,
  2. Andrew Coats, director, clinical cardiologyb,
  3. Patrick Owens, clinical lecturer in medicinec,
  4. James Petrie, professor, clinical pharmacologyd,
  5. Paul L Padfield, consultant physiciane,
  6. William A Littler, professor, clinical cardiologyf,
  7. Michael de Swiet, consultant physiciang,
  8. Fáinsía Mee, representative, Nurses Hypertension Societya
  1. a Blood Pressure Unit, Beaumont Hospital, Dublin 9, Ireland
  2. b Royal Brompton Hospital, London SW3 6NP
  3. c Cardiology Department, Royal Cornwall Hospital, Truro TR1 3LJ
  4. d Clinical Pharmacology Unit, Department of Medicine and Therapeutics, Aberdeen Royal Infirmary, Aberdeen AB9 2ZB
  5. e Department of Medical Sciences, Western General Hospital, Edinburgh EH4 2XU
  6. f Department of Cardiovascular Medicine, Queen Elizabeth Hospital, Birmingham B15 2TH
  7. g Imperial College School of Medicine, Institute of Obstetrics and Gynaecology, Queen Charlotte's and Chelsea Hospital, London W6 0XG
  1. Correspondence to: E O'Brien
  • Accepted 13 March 2000

Over the past 20 years or so, the accuracy of using the conventional Riva-Rocci sphygmomanometer and Korotkoff's sounds to measure blood pressure has been questioned, and efforts have been made to improve measurements with automated devices.1 2 In the same period, the phenomenon of white coat hypertension has been recognised—whereby some patients who apparently have raised blood pressure actually have normal blood pressure when the measurement is repeated away from the medical environment; this has focused attention on methods of measurement that provide profiles of blood pressure rather than rely on isolated measurements made under circumstances that may influence blood pressure.3 These methods have included repeated measurements of blood pressure using the traditional technique, self measurement of blood pressure in the home or workplace, and ambulatory blood pressure measurement using automated devices.2 Ambulatory monitoring is advantageous because it gives multiple measurements throughout the day and night

This paper considers only the ambulatory measurement of blood pressure in adults. Its purpose is not to make a case for or against ambulatory measurement; others have already done so.4 5 Although the results of a number of ongoing, longitudinal studies are forthcoming, there is now firm evidence that ambulatory blood pressure measurement is a more sensitive predictor of cardiovascular outcome than conventional measurement.6 We have not considered the complex issues of health economics that the increasing use of ambulatory measurement raises.7 We realise that this technique is being used more often and that doctors who find ambulatory measurement useful in the day to day management of patients with high blood pressure need recommendations from those who have experience. However, regardless of the technique used to diagnose hypertension it is only one factor in determining a patient's risk profile and must be assessed in relation to concomitant disease, such as diabetes mellitus, and in relation to the degree of target organ involvement as recommended in the British Hypertension Society's guidelines on the management of hypertension.8 9

Summary points

One of the most important indications for ambulatory monitoring is to exclude white coat hypertension

The technique is also valuable in diagnosing and treating elderly patients and is used increasingly in pregnancy

Practices should consider carefully which monitor to buy, taking into account whether it has been independently validated, and should also consider how the data are analysed and presented

Methods

Recommendations on the use of ambulatory measurement have tended to be ambivalent, although firmer proposals are now being made.10 11 Such ambivalence has not assisted doctors wishing to use the technique; however, making recommendations on the basis of incomplete evidence may lead to charges of advocating a technique that is not supported by the evidence. Recognising this, we have based our recommendations on evidence when it is available, and in cases in which it is not we have given advice on the basis of our collective experience of using ambulatory measurement over many years. What seems reasonable today may have to be modified as additional evidence becomes available: such is the essence of scientific reasoning. Where possible we have graded the strength of the evidence on which we have based our recommendations according to the scheme discussed by Shekelle et al.12

Setting up an ambulatory blood pressure measurement service

Which monitor?

A large variety of devices for ambulatory measurement are available, and the number will increase as the technique becomes more widespread.11 A number of factors influence the choice of monitor (box); the most important factor is whether the device has been validated independently according to either the protocol of the British Hypertension Society13 or that of the US Association for the Advancement of Medical Instrumentation,14 or both. table 1 shows the results of independent evaluations of devices using these protocols. (A list of the manufacturers of ambulatory systems can be found on the BMJ's website.)

Table 1.

Results of independent evaluation of 23 ambulatory blood pressure measuring devices15. Devices were evaluated using the protocols of the British Hypertension Society12 and the US Association for the Advancement of Medical Instrumentation13

View this table:

Factors to consider when choosing a monitor

Has the device been validated by the British Hypertension Society or the US Association for the Advancement of Medical Instrumentation?

How much does it cost?

How expensive is the software?

Does the software offer the information that you need?

Are the operating instructions adequate?

How much will maintenance cost?

How expensive are consumables, such as batteries?

Does the practice have adequate computer facilities to support the data analysis?

Is support available from technical or nursing staff within the practice?

Are training facilities available from the manufacturer or supplier?

Is the warranty adequate?

Does the manufacturer or supplier have servicing facilities that are easily accessible?

What type of service is most appropriate?

Doctors may establish their own service to provide ambulatory measurement, refer patients to a hospital service with open access to ambulatory measurement, or refer patients to a blood pressure clinic for full evaluation, which would include ambulatory measurement. Often an open access referral service is used and difficult cases are referred for fuller evaluation to a blood pressure clinic.

Training

The technique of ambulatory blood pressure measurement is specialised and should be approached with care. An understanding of the principles of traditional blood pressure measurement, cuff fitting, monitor functioning, and interpretation of the data from ambulatory measurement is recommended.16 In practice, a nurse who is interested and has experience in caring for patients with hypertension can use the devices after a comparatively brief training. However, the analysis and interpretation of the ambulatory profiles require experience, and this is best learnt from the doctor in charge of the service offering ambulatory measurement.

Using an ambulatory monitor

About 15 to 30 minutes need to be allotted to fitting the monitor and preparing the patient if good results are to be obtained (box). Recommendations for cuff dimensions are shown in table 2. 9 17 Whichever cuff is chosen, the centre of the inflatable bladder should be placed over the brachial artery. The key to successful ambulatory measurement is educating the patient about the process of monitoring (box).

Table 2.

Recommended size of cuffs for measuring blood pressure12

View this table:

Fitting the monitor

This will take about 15 to 30 minutes

Have the patient relax in a quiet room

Enter the patient's details, such as name and identification number, into the monitor

Measure blood pressure in both arms

If the difference in systolic blood pressure is <10 mm Hg use the non-dominant arm for monitoring

If the difference in systolic blood pressure is ≥10 mm Hg use the monitor on the arm with higher pressure

Select the appropriate cuff

Select the frequency of measurement (usually every 30 minutes)

Inactivate the display to prevent the patient becoming distracted by the measurements

Give the patient written instructions

Show the patient how to remove and inactivate the monitor after 24 hours

How to prepare the patient so that monitoring is successful

Explain
  • the procedure

  • the frequency of inflation and deflation

  • how to manually deflate the cuff

  • that in the event of failed measurements the monitor will repeat the measurement

Instruct patients
  • to keep their arm steady during measurement

  • to keep their arm at heart level during measurement

  • to engage in normal activities between measurements

  • to keep the monitor attached at night

  • to place the monitor under a pillow or on the bed at night

Provide
  • a telephone number for patients to call in case of problems or anxiety

  • a diary card for the patient to record:

    their activity level at the time of measurement

    the time they go to bed

    the time they get up

    the time they take their drug treatment

    any symptoms

In clinical practice, measurements are usually made at half hourly intervals so as not to interfere with activity during the day and with sleep at night, but measurements can be made more frequently if necessary. There are a number of ways of analysing the blood pressure values recorded during the 24 hour cycle.18 One simple and popular method is to divide the 24 hours into day (beginning with the first entry on the diary card) and night (beginning with the last entry). Another is to use a fixed time method in which the retiring (2101 to 0059) and rising (0601 to 0859) periods, during which blood pressure is variable, are eliminated; in this method the daytime period is set at 0900 to 2100 and the night-time from 0100 to 0600. In this way variations that may exist between the young and the old and in different cultures are to some extent eliminated from the analysis.18

Analysing the data

Many statistical techniques exist for describing different aspects of ambulatory records, and no one method is ideal.19 20 The important points to consider when analysing data are summarised in the box. The detection of artefactual readings and the handling of outlying values (which may not always be erroneous) have been debated,21 and we believe that if there have been a sufficient number of measurements editing the values is not necessary.

Analysing and presenting the data

Number of measurements necessary

Day: >14 systolic and diastolic blood pressure measurements

Night: >7 systolic and diastolic blood pressure measurements

Causes of poor data

Poor technique

Arrhythmias

Small pulse volume

Inability of automated device to measure blood pressure

Editing data

Restrict editing to physiologically impossible pressures, such as if the diastolic pressure equals the systolic pressure

Displaying the data

Statistics should include:

Mean daytime systolic and diastolic pressures and heart rate

Mean night-time systolic and diastolic pressures and heart rate

Mean 24 hour systolic and diastolic pressures and heart rate

Plot the data (figure)

Devices for ambulatory measurement are usually sold with software packages that present the data in a variety of ways. It would facilitate practice if the graphic presentation of the data was standardised, as is the case for electrocardiograms. The presentation of data should be independent of the type of monitor used. A standardised approach could provide a graphical display of the data from the ambulatory measurement (on screen or on a printout) with a graph showing blood pressure on the vertical axis and time on the horizontal axis (figure); normal values could also be shown.20 22 One program (DABL Cardiovascular 2000, ECF Medical, Dublin, Republic of Ireland) already provides a printed report derived from the data obtained by the monitoring.20

Table 3.

Recommended standards for normal and abnormal pressures during ambulatory measurement. These pressures are only a guide, and lower pressures may be abnormal in patients whose total risk factor profile is high and in whom there is concomitant disease8 9

View this table:
Figure1

Examples of ambulatory blood pressure patterns plotted by the DABL program. The white coat window is the first hour of recording

Deciding what constitutes normal blood pressure and what constitutes abnormal in ambulatory measurement is controversial, and the approaches to defining normal values have been discussed, 23 24 but the values shown in table 3 are commonly used (strength of evidence C).23 We appreciate that these values may be regarded as conservative by some, but lower values cannot be recommended unless evidence from ongoing longitudinal studies indicates otherwise.

Clinical indications for ambulatory blood pressure measurement

Ambulatory measurement provides a large number of measurements over time—usually 24 hours—which can be plotted to give a profile of blood pressure. Although in practice the average daytime (or night-time) blood pressure is used to govern decisions, the clinical use of ambulatory measurement has allowed a number of phenomena in hypertension to be identified that would otherwise not have been (strength of evidence B-C).22 25 Ambulatory measurement is of benefit

  • In excluding white coat hypertension (evidence strength C)

  • In helping with the diagnosis of patients with borderline hypertension (evidence strength D)

  • In deciding on treatment for elderly patients (evidence strength A-C)

  • In identifying nocturnal hypertension (evidence strength C)

  • In assessing patients whose hypertension has been resistant (evidence strength D)

  • As a guide to determining the efficacy of drug treatment over24 hours (evidence strength B)

  • In diagnosing and treating hypertension in pregnancy (evidence strength C-D), and

  • In diagnosing hypotension (evidence strength B-C).

White coat hypertension

From the first use of home and ambulatory monitoring, it became apparent that measuring a patient's blood pressure in a clinic or office could raise their blood pressure above their mean ambulatory pressure as a result of the white coat phenomenon.3 2529 The features of white coat hypertension are summarised in the box. In patients with normal blood pressure daytime ambulatory pressure may be a little higher than blood pressure measured conventionally, but in those with hypertension daytime blood pressure is usually substantially, but unpredictably, lower than blood pressure measured conventionally.

White coat hypertension

Definition

Blood pressure ≥140/90 mm Hg when measured in office

Normal daytime ambulatory pressure <135/85 mm Hg

Prevalence of white coat hypertension

15-30% of general population

Common in elderly people and pregnant women

Risks

Less than from sustained hypertension

Probably small risk when compared with people with normal blood pressure

Possibly a precursor to hypertension

Clinical implications

No clinical characteristics assist in diagnosis

Must be considered in people newly diagnosed with hypertension

Should be considered before drug treatment is prescribed (could lead to fewer drugs being prescribed)

Must be placed in context of the overall risk profile

Should reassure patients, employers, and insurers that risk from white coat hypertension is low or absent

Patients need follow up and monitoring again

Borderline hypertension

Patients whose blood pressure is considered to be borderline may also benefit from ambulatory monitoring, especially young patients in whom lifelong drug treatment may otherwise be prescribed inappropriately and who may be penalised in terms of insurance or employment if the diagnosis of hypertension is misapplied.

Considering treatment in elderly patients

The results of the ambulatory study of the systolic hypertension in Europe trial show that conventional measurement of systolic pressure in elderly people may produce results that are on average 20 mm Hg higher than daytime ambulatory pressure,30 leading to an overestimation of the occurrence of isolated systolic hypertension among elderly patients and probably excessive treatment. Moreover, results from this study also show that ambulatory systolic pressure is a better predictor of cardiovascular risk than pressure measured conventionally.31 A variety of ambulatory patterns are found among elderly people, including a number of hypotensive states associated with baroreceptor or autonomic failure. These blood pressure patterns include white coat hypertension, isolated systolic hypertension, postural hypotension, post-prandial hypotension, daytime hypotension and nocturnal hypertension, drug induced hypotension, and autonomic failure. Since elderly people can be particularly susceptible to the adverse effects of drug treatment given to lower blood pressure, identifying hypotension is particularly important, although its management may be challenging.32

Nocturnal hypertension

Ambulatory measurement is the only non-invasive technique that permits blood pressure to be monitored during sleep. The relevance of nocturnal hypertension is still controversial, but there is increasing evidence that nocturnal blood pressure may provide important information; for example, blood pressure at night is independently associated with end organ damage above the risk associated with daytime values.33 34 It has also been shown that the absence of a night-time drop in blood pressure is associated with target organ involvement, and it may be a useful (although non-specific) clue to the presence of secondary hypertension. (Patients whose blood pressure drops at night are sometimes known as “dippers,” and those whose blood pressure does not drop are sometimes known as “non-dippers.”33)

Hypertension resistant to treatment

Patients are classed as having resistant hypertension when their blood pressure remains consistently above 150/90 mm Hg with conventional measurement despite being treated with three or more drugs. In these patients ambulatory monitoring may indicate that the apparent lack of response is caused by the white coat phenomenon; alternatively, the absence of a night-time drop in blood pressure may suggest secondary hypertension.

Pregnancy

The main use of ambulatory measurement in pregnancy is to identify white coat hypertension; it may occur in nearly 30% of pregnant women.35 Recognising it is important so that pregnant women are not admitted to hospital or given antihypertensive drugs unnecessarily. Normal values for ambulatory blood pressure among pregnant women are available, and the changes in pressure which occur during the different trimesters of pregnancy and in the postpartum period have been defined.36 The evidence that ambulatory measurement may predict pre-eclampsia is not conclusive.37 However, ambulatory blood pressure correlates better with proteinuria than does conventional sphygmomanometry,38 and it is a better predictor of complications of hypertension.39 In addition, women diagnosed with hypertension by ambulatory monitoring have infants with lower birth weights and this association is not found when blood pressure is measured conventionally.40 Moreover, women with white coat hypertension tend to be more likely to have a caesarean section than women with normal blood pressure, suggesting that if ambulatory measurement was used rather than conventional measurement, some caesarean deliveries might be avoided.35

Ambulatory hypotension

Ambulatory measurement may be useful in identifying hypotensive episodes in young patients in whom hypotension is suspected of causing symptoms.32 Ambulatory measurement may also identify drops in blood pressure induced by the drugs used to treat hypertension, which may have untoward effects in patients with compromised arterial circulation, such as those with coronary and cerebrovascular disease.41

As a guide to drug treatment

The role of ambulatory measurement in guiding drug treatment is the subject of much research, and its role in this regard has not been fully established. However, recent reviews of the clinical value of ambulatory measurement have highlighted the potential usefulness of 24 hour recordings of blood pressure in guiding drug treatment.42 43 Furthermore, a recent well controlled study showed that when ambulatory measurement rather than measurement in a clinic was used as the basis for prescribing treatment, significantly less antihypertensive drug treatment was prescribed.4 Ambulatory measurement gives the prescribing doctor an assessment of the patient's response to treatment that conventional measurement cannot provide: the efficacy of treatment without the white coat effect can be ascertained, an excessive effect of drugs and the occurrence of symptoms can be determined, and the duration of the effect of drugs over 24 hours can be shown.

Who should be monitored again?

Ambulatory blood pressure measurement inconveniences patients, and it should be used with discretion. The decision of when to repeat ambulatory measurement is largely one of clinical judgment, which may be influenced by factors such as an excessive variability of blood pressure, an inappropriate response to treatment, an adverse risk factor profile, and the need for careful control of blood pressure, such as in patients with hypertension who have diabetes mellitus or renal disease. As a general rule it is usually not necessary to repeat ambulatory measurement more frequently than once a year. Conventional measurement may be relied on for follow up in patients who do not have evidence of a white coat effect when monitored with ambulatory measurement. The patients for whom monitoring again may be helpful are

  • Patients with white coat hypertension

  • Patients being treated for hypertension who also have evidence of white coat hypertension

  • Elderly patients with hypotension

  • Patients with night-time hypertension, and

  • Patients who have had their drug treatment changed.

Acknowledgments

We acknowledge the helpful comments and criticism we received from members of the executive of the British Hypertension Society, especially Professor G MacGregor, Professor L Ramsay, Professor N Poulter, and Professor B Williams. We thank Professor Thomas Pickering, of the Hypertension Center at the New York Hospital, Cornell Medical Center, USA, and Dr Jan Staessen, of the Katholieke Universiteit Leuven, Belgium, for reading the manuscript and making a number of helpful suggestions.

Contributors: All authors participated in researching and writing the paper. EO'B, chairman of the working party on Blood Pressure Measurement of the British Hypertension Society, will act as guarantor for the paper. AC, JP, PP, WL, MdeS, and FM are members of the working party, and PO was co-opted for the purpose of assisting with the drafting of the recommendations.

Footnotes

  • Funding Travel expenses incurred by the authors attending meetings of the working party were reimbursed by the British Hypertension Society.

  • Competing interests EO'B is director of the blood pressure unit at Beaumont Hospital, which has been contracted by manufacturers from time to time to conduct validation studies of ambulatory blood pressure measurement devices; the results of these studies have been published. Negotiations to distribute commercially the DABL Cardiovascular 2000, of which EO'B is one of the originators, are under way with ECF Medical.

  • Embedded Image Additional information about manufacturers appears on the BMJ's website

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.