Diagnosing Hypertension—Implications of the 2011 NICE Guidance

Abstract

The accurate diagnosis of hypertension is a difficult issue. The 2006 National Institute for Health and Clinical Excellence (NICE) guidance for the management of hypertension in adults in primary care recommended the use of clinic measurements taken on multiple occasions. However, this process was both protracted and potentially inaccurate. As a result, some patients may have been inappropriately treated with anti-hypertensive drugs causing both unnecessary side effects and excess costs to the National Health Service (NHS). Others may not have been treated at all resulting in additional cardiovascular events. Given this, NICE has recently updated its guidance in this area to include an increasing role for ‘out-of-office’ techniques, including ambulatory and home monitoring.

The GP curriculum and the measurement of blood pressure

GP curriculum statement 15.1: Cardiovascular problems lists the learning objectives for a GP to manage the scientific aspects of blood pressure (BP) measurement. In particular, GPs should be able to:

Describe and be able to implement the key national guidelines that influence health care provision for cardiovascular problems

Describe the key research findings that influence management of cardiovascular problems

The diagnosis of hypertension can be something of a dilemma. A mildly elevated BP may be attributed to ‘a stressful day’ by the patient, while the doctor may be tempted to collude with them by putting it down to ‘the white coat’ effect requiring no further treatment.

There is now mounting evidence to show that clinic blood pressure monitoring (CBPM) is inadequate. At best, it only provides a snapshot of a continuously changing variable and accuracy is often further compromised by unvalidated equipment and a haphazard procedure. As a result, interest has been growing in out-of-office techniques, such as home and ambulatory BP monitoring methods (see Box 1). However, while these methods offer potential improvements in the accuracy of diagnosis and monitoring of patients with hypertension, there has been some controversy over their routine use due to issues of cost, resources and patient preference.

Box 1.

Out-of-office techniques—what are they?

Home monitoring (HBPM)

The patient uses an electronic device to measure their own BP at home. For each BP measurement, two consecutive readings are taken at least 1 minute apart with the patient seated. BP measurements are taken twice daily, ideally in the morning and evening. BP measurement continues for at least 4 days and ideally for 7 days.

Ambulatory monitoring (ABPM)

A cuff connected to a portable monitor is worn continuously by the patient for a period of 24 hours. The cuff should inflate at least twice an hour during the day (8:00 to 22:00) and once an hour overnight (22:00 to 8:00). The average daytime BP (calculated from at least 14 daytime readings) is used to diagnose hypertension.

Previous NICE guidance

The previous NICE guidance on hypertension management in adults in primary care was published in 2006. It did not routinely recommend the use of automated ambulatory BP monitoring or home monitoring devices in primary care because there was an inadequate evidence base at the time of writing. Instead, the guideline was based on readings taken in the doctor's office. Box 2 describes the method recommended by NICE for measuring office BP in 2006.

Box 2.

The procedure for measuring BP in the office

The patient should sit with their arm extended and supported in a relaxed, quiet and warm environment

A reading should be taken on both arms: that with the highest should be used for subsequent measurements

If the first reading is greater than 140/90 mmHg, a second reading should be taken later in the consultation

To diagnose hypertension, the patient should come back for at least two more appointments during which the BP is checked twice each time. Readings should be taken 1 month apart unless the patient has severe hypertension in which case they should be reassessed earlier

BP measurement devices must be appropriately validated, maintained and regularly recalibrated according to manufacturers' instructions

Source: NICE. Hypertension—management in adults in primary care (2006). Retrieved from www.nice.org.uk/nicemedia/pdf/CG034NICEguideline.pdf

Advantages of out-of-office BP monitoring techniques

Following the guidance issued by NICE in 2006, a minimum of three clinic visits over a period of 2 months were needed to make a diagnosis of hypertension. This was time consuming for both the doctor and the patient. Moreover, the purpose of diagnosing and subsequently treating hypertension is to reduce the risk of cardiovascular disease, including myocardial infarction, stroke and peripheral vascular disease. However, if clinic BP measurements are inaccurate, then this may weaken the relationship between BP and cardiovascular risk.

Since 2006, new evidence has become available to show that out-of-office BP measurement techniques, such as ambulatory and home monitoring, are better at predicting cardiovascular outcomes than the conventional clinic method. In the new 2011 guidance, NICE reviewed 3 pooled analyses of prognostic studies and 11 individual prognostic studies which compared the ability of clinic, home or ambulatory BP measurements to predict cardiovascular outcomes.

They concluded that ABPM was the best prognostic indicator, followed by HBPM then CBPM. There are a number of reasons why this may be the case. Firstly, BP is a continuous variable that is, by definition, changing all the time. Sustained high pressures on the arterial walls result in damage and possible cardiovascular disease. Consequently, prognosis is likely to be most accurately predicted by the method of measurement with the highest frequency of sampling as this will provide the best representation of the true arterial load.

Secondly, some patients have higher BP in the presence of a doctor or nurse than they would outside of the medical environment: this is known as the ‘white coat’ effect. Where this effect causes the BP to cross the diagnostic threshold, then this is termed ‘white coat hypertension’. Conversely, others may have normal clinic BP and higher readings elsewhere: this is termed ‘masked’ hypertension (see Fig. 1). Both these effects can lead to inappropriate clinical decision making, resulting in poorer outcomes for patients and increased costs. Out-of-office techniques can enable identification of these issues and a corresponding improvement in the targeting of treatment.

Figure 1.

BP categories.

Evidence for white coat and masked hypertension

There is increasing evidence that the prevalence of both white coat and masked hypertension is significant. Moreover, these syndromes carry very different prognoses with corresponding implications for treatment. A meta-analysis by Fagard and Cornelissen (2007) indicated that the incidence of cardiovascular events is not significantly different between those with white coat hypertension and those with true normotension, whereas the outcome is worse in patients with masked or sustained hypertension.

This implies that masked hypertension requires treatment while white coat hypertension does not. Since neither of these conditions can be diagnosed by clinic monitoring alone, overtreatment is possible for white coat hypertension and undertreatment for masked hypertension unless home or ambulatory monitoring is used.

Effect of poor nocturnal dipping

Blood pressure is subject to diurnal variation, with lower pressure values occurring at night and higher values during the day (Millar-Craig et al., 1978). O'Brien et al. (1988) showed that stroke is more common in patients whose BP drops by a smaller amount during the night. A ‘non-dipper’ was defined as someone in whom the mean overnight BP was less than 10/5 mmHg below the mean daytime BP. This effect was pronounced with 23.8% of ‘non-dippers’ experiencing a stroke during the follow-up period compared to 2.9% of ‘dippers’. The hypothesis to explain these findings is that target organ damage and prognosis are worse when the BP load is persistent throughout 24 hours than when it is limited to daylight hours. Ambulatory monitoring is currently the only method of identifying whether a patient falls into the dipper or non-dipper category thereby allowing improved cardiovascular risk stratification and possible corresponding modification of treatment.

Clinical vs research office readings

The comparative studies outlined above use office readings taken in the research environment. However, there is evidence that those casually taken in a routine GP appointment may be even more inaccurate. In a study by Myers et al. (1995), the physician's mean BP obtained during a routine visit (146/87 mmHg) was higher than the same physician's mean BP reading taken for research purposes (140/83 mmHg) or when BP was measured by a research nurse (137/78 mmHg) in accordance with BP measurement guidelines. These improvements were achieved by allowing the patient to sit alone in a quiet room for 5 minutes before three readings were taken with the first being discarded, and the average of the second two readings used as the result. In this study, however, readings taken by doctors were still higher than those taken by non-medically qualified members of staff. These issues can be eliminated by out-of-office techniques, which do not require a health professional to be present while the BP is being measured.

NICE guidance 2011

In the light of this evidence, the new NICE guidance 2011 advocates the routine use of ambulatory monitoring with the option of the patient self-monitoring at home if this cannot be tolerated. The guidance states that:

If the initial BP measured in the clinic is 140/90 mmHg or higher, a second measurement should be taken during the consultation

If the second measurement is significantly different from the first, then a third measurement is also required

If the lower of the last two readings is above 140/90 mmHg, a diagnosis of hypertension must be considered

This should be confirmed using 24 hour ABPM. If this cannot be tolerated, then HBPM would be a suitable alternative

However, if the person has severe hypertension (clinic BP of 180/110 mmHg or greater) and evidence of target organ damage, anti-hypertensive drug treatment should be started immediately without waiting for the results of ABPM or HBPM.

Equivalence between methods

Data from trials at the population level (Mancia et al., 1995) have shown that BPs measured using home or ambulatory techniques are lower than those measured in the clinic environment. Therefore, the thresholds for diagnosing hypertension by the office method need to be adjusted downwards for readings obtained using the home or ambulatory methods. However, the degree of adjustment required has been much debated.

The most recent comprehensive study in this area by Head et al. (2010) showed that an office BP reading of 160/100 mmHg taken by a non-medically trained member of staff corresponds to an average daytime ambulatory BP of 152/96 mmHg. If the reading of 160/100 mmHg had been obtained by a doctor, this would correspond to an average daytime ambulatory BP of 142/90 mmHg (i.e. clinic readings taken by doctors are higher than those taken by non-medically qualified members of staff). Meanwhile, an office BP measurement of 140/90 mmHg corresponded to an ambulatory reading of 136/87 mmHg if taken by a non-medical member of staff and 129/81 mmHg if taken by a doctor.

NICE (2011) has used these comparative data to define three grades of hypertension in their new hypertension guidance:

Stage 1 hypertension: initial clinic BP 140/90 mmHg or higher and subsequent ABPM daytime average or HBPM average BP 135/85 mmHg or higher

Stage 2 hypertension: initial clinic BP 160/100 mmHg or higher and subsequent ABPM daytime average or HBPM average BP 150/95 mmHg or higher

Severe hypertension: clinic BP 180/110 mmHg or higher

Benefits of adopting the new guidance

NICE (2011) Analysis by NICE has shown ABPM to be more cost effective than both HBPM and CBPM in all age/gender subgroups. This is mainly achieved through the improved diagnosis of white coat hypertension resulting in reduced unnecessary prescribing of anti-hypertensive medications: these benefits have been shown to outweigh the increased equipment costs of ABPM, particularly in the longer term. In the majority of sub groups, ABPM is also associated with higher quality-adjusted life years (QALYs): this is due to more accurate diagnosis of true hypertension and hence improved targeting of treatment. If the reduction in side effects in those avoiding unnecessary anti-hypertensive treatment thanks to ABPM had also been taken into account, then the QALY gains with this modality of measurement would have been even greater.

However, it should be noted that referral for ABPM on the basis of a raised clinic blood pressure does not promote the improved diagnosis of masked hypertension as patients with this condition will have normal blood pressures in the clinic environment. Therefore, the excess cardiovascular events in this group will not be reduced by the new NICE guidance. Unfortunately, addressing under-diagnosis here is difficult as this would require everyone to undergo ABPM, which is clearly impractical. However, there is increasing interest in improving access to self-monitoring equipment outside of the clinic environment in order to screen more effectively for this condition.

Problems of adopting the new guidance

Applying the 2011 NICE hypertension guidance, large numbers of people will require ambulatory BP monitoring as it recommends this after a single raised clinic blood pressure reading. Despite favourable economic analysis by NICE, ABPM machines retail for £1000–£2000 each (British Hypertension Society, 2011) and most GP practices cannot afford to buy sufficient equipment to meet the needs of their patients. Meanwhile, Alternative providers of this service are few and far between. This raises the question of who should fund the capital outlay required to provide an ABPM service? Perhaps, the new era of commissioning may bring greater incentives to invest in ABPM equipment in order to realize lower prescribing costs.

Cost aside, ABPM can be problematic for patients. Little et al. (2002) showed that the ‘problem’ score for ambulatory monitoring was higher than for the other methods (i.e. home and clinic) due to overall discomfort along with disturbance of daily activities and sleep. Nonetheless, patients thought that all methods were ‘worth the trouble’ in order to get accurate readings.

There is also a training cost associated with ABPM through GP practices. Practice nurses and/or health care assistants need to learn how to fit the monitors, while doctors need to be able to interpret the results. The values recorded by the ambulatory monitors can be accessed by connecting them to a computer with the appropriate software installed. This generates a printout which is similar to that shown in Fig. 2. The mean ambulatory daytime pressure is the key indicator and this should be below 135/85 mmHg. In patients with evidence of poor nocturnal dipping (i.e. a difference between the mean nocturnal and daytime BPs of less than 10/5 mmHg), earlier treatment may be considered since these patients are likely to be at a higher risk of stroke.

Figure 2.

An example of an ambulatory BP monitoring printout. Dr P. Marazzi/SPL.

Practical advice

Clinic monitoring

Despite increasing interest in out-of-office techniques, clinic measurements will continue to have an important role to play in the diagnosis of hypertension. Accuracy can be improved by allowing the patient to sit alone in a quiet room for 5 minutes before taking any measurements (Beckett and Goodwin, 2005). In patients with an arrhythmia, an electronic sphygmomanometer may not be accurate and manual measurement using a mercury device is preferred. It is important not to talk to the patient while taking the measurement as conversation has been shown to increase BP (Myers, 2006). The same study also demonstrated that a more accurate result can be achieved by the clinician leaving the room while the reading is being taken. Clearly, not all these suggestions may be practical within a 10 minute consultation but can be considered when appropriate.

Home monitoring

Some patients may be keen to purchase their own device for home monitoring and this can provide a useful adjunct to clinic measurements. It is important to advise patients to purchase a validated device (Box 3). Devices should be calibrated on a yearly basis. Patients should be advised to sit down and relax for 5 minutes before taking a reading on an outstretched and well-supported arm. The protocol in Box 1 should be followed. If the patient is already taking anti-hypertensive medications, then morning readings should be taken after tablets and evening readings before. It is helpful to give patients a printed proforma on which to record their readings and to specify the time frame in which they should return for review. If there is a significant discrepancy between the patient's own readings and those taken in the clinic, then it may be worth asking them to bring their machine in for comparison with the surgery's device.

Box 3.

A selection of validated devices for home monitoring

Lloyds Pharmacy BP11. Includes 99 reading memory. £9.99

Boots Upper Arm BPM. Includes 14 reading memory. £39.99

Microlife BPA100. £49.89

Omrom M4-1. Fully automatic upper arm £79.95

Proton Healthcare 868BF. Includes 180 reading memory. £139.99

Source: British Hypertension Society. Automatic blood pressure measurement devices suitable for use in the clinic and also at home for self monitoring. Retrieved from www.bhsoc.org/bp_monitors/automatic.stm [date last accessed 04.11.2011]

Note: All prices include Value Added Tax (VAT)

ABPM

Skilled fitting of an ABPM device is essential for patient comfort and an accurate reading. The cuff is placed on the non-dominant arm unless the patient has been shown to have a significant discrepancy in the BP measured bilaterally. If so, the cuff should be placed on the arm with the highest BP. The cable is then threaded across the patient's shoulders and down to the monitor on the patient's opposite hip (which can be placed in a pocket or attached to a belt). Overnight, the monitor should be placed under the pillow while the cuff remains in situ.

The patient should be advised against driving while wearing the monitor as an inflating cuff can cause distraction and the patient may not be covered by insurance if an accident occurs. Likewise, the patient should remove the device while taking a bath or shower and this should be timed in the 30 minute interval between readings. Most devices will give a 5 second warning sound before inflation and patients should be instructed to sit down or stand still when this occurs in anticipation of a reading being taken.

Conclusions

NICE has now made the significant step of recommending the use of either ambulatory or home BP monitors in the initial diagnosis of hypertension. However, the practical arrangements to support the increasing role of these devices remain unclear. It may take some time for implementation of the new guidance while capacity is put into place. Nonetheless, it seems certain that out-of-office techniques will play an increasing role in the diagnosis and management of hypertension in the future, and for this reason, it is important for GPs and GPs in training to understand their benefits.

Key points

Clinic BP measurement is unable to identify white coat or masked hypertension

There is now substantial research evidence to show that out-of-office techniques such as ambulatory and home monitoring are better at predicting cardiovascular outcomes than the conventional office method

The prognostic significance of white coat hypertension is similar to normotension, while that of masked hypertension is nearer to sustained hypertension

Hypertension should be diagnosed at lower levels using out-of-office techniques: an ABPM daytime average or HBPM average of 135/85 mmHg should be used to diagnose stage 1 hypertension, while 150/95 mmHg should be used for stage 2 hypertension

The NICE guidance for hypertension 2011 gives out-of-office techniques an increasing role in the diagnosis and monitoring of hypertension. It is therefore important for GPs to be familiar with their use

References

Beckett

Goodwin

The BP TRU automatic blood pressure monitor compared to 24 hour ambulatory blood pressure monitoring in the assessment of blood pressure in patients with hypertension. BMC Cardiovascular Disorders (2005) 5 (1): p. 18. Retrieved from www.biomedcentral.com/content/pdf/1471-2261-5-18.pdf [last accessed 27.09.2011].

British Hypertension Society. Ambulatory blood pressure measurement devices. Oscillometric mode (2011) Retrieved from www.bhsoc.org/bp_monitors/ambulatory.stm [last accessed 29/08/2011].

Fagard

R.H.

Cornelissen

V.A.

Incidence of cardiovascular events in white-coat, masked and sustained hypertension versus true normotension: a meta-analysis. Journal of Hypertension (2007) 25 (11): p. 2193–8.

Head

G.A.

Milhailidou

A.S.

Duggan

K.A.

Definition of ambulatory blood pressure targets for diagnosis and treatment of hypertension in relation to clinic blood pressure: prospective cohort study. British Medical Journal (2010) 340: p. c1104. Retrieved from www.bmj.com/content/340/bmj.c1104.long [last accessed 04.05.2011].

Little

Barnet

Barnsley

Marjoram

Fitzgerald-Barron

Mant

Comparison of acceptability of and preferences for different methods of measuring blood pressure in primary care. British Medical Journal (2002) 325: p. 258. Retrieved from www.bmj.com/content/325/7358/258.long [last accessed 04.05.2011].

Mancia

Sega

Bravi

Ambulatory blood pressure normality: results from the PAMELA study. Journal of Hypertension (1995) 13: p. 1377–90.

Millar-Craig

Bishop

Raftery

Circadian variation of BP. Lancet (1978) 1: p. 795.

Myers

M.G.

Reeves

R.A.

Joynes

C.D.

Prevalence of white coat effect in treated hypertensive patients in the community. American Journal of Hypertension (1995) 8: p. 591–7.

Myers

M.G.

Automated blood pressure measurement in routine clinical practice. Blood Pressure Monitoring (2006) 11 (2): p. 59–62.

10.

NICE. Hypertension—management in adults in primary care (2006) Retrieved from www.nice.org.uk/nicemedia/pdf/CG034NICEguideline.pdf [last accessed 27.09.2011].

11.

NICE. Hypertension—clinical management of primary hypertension in adults (2011) Retrieved from www.nice.org.uk/nicemedia/live/12167/53228/53228.pdf [last accessed 05.08.2011].

12.

O'Brien

Sheridan

O'Malley

Dippers and non dippers. Lancet (1988) 2: p. 397.

13.

RCG Curriculum statement 15.1: Cardiovascular problems. Retrieved from www.rcgp-curriculum.org.uk/PDF/curr_15_1_Cardiovascular_problems.pdf [last accessed 27.09.2011].