Ambulatory blood pressure measurement in general practice.

u NTlL recently, a~ilbulatory blood pressure measurement was largely tlie preserve of physicians working in specialized centres. ?'lie reasons for this included the fact that the first clevices recorded blood pressure invasively arid were associated wit11 so~tic risk ant1 were tl~ercfore of liniitcd clinical application. l'lie early setiii-automated non-invasive devices of the 1960s had to be fitted by trained perso~incl arid thus their application was confined to research. With the advent of more user friendly and less expensive automated devices, tliis technique lias beconie a niore attractive propositio~i to the general practitioner. T11e issues of what constitutes a rior~iial result in ambulatory blood pressure measurenlent, tlie role of atiibulatory blood pressure niotiitoring in clinical practice and the prognostic iniportalice of tlie technique are beconling more clearly defined.' Clearly, if the equip~nent used for the procedure does not measure blood pressure accurately, it has no place in tlie diagnosis and man5gement of hypertension. Thus, the major initial consideration to be taken into account by the general practitioner iri selecting an ambulatory blood pressure system is its accuracy and reliability. Although increasing numbers of ambulatory blood pressure nionitors come on the market each year, there is a t present no obligation on manufacturers to comply witli the few recomniended standards that are available for these s ~ s t e m s . ~ There is no standard for automated blood pressure devices in the United Kingdom, although the British Hypertension Society has published a protocol for evaluatirig automated devices with special reference to ambulatory monitoring systen~s.~ In tlie United States of America, the Association for the Advance~ile~it of Medical instrumentation lias produced a detailed standard for automated and semi-automated devices4 which is shortly to be updated. In most subjects, mean 24 hour ambulatory blood pressure values are lower tlia~i blood pressure values measured in the clinic and the difference appears to be greater with increasing blood pressure levels measured in the cl i~i ic .~ In the past, anibulatory blood pressure levels were studied in relatively small groups of 'normal' subjects who were often selected from blood pressure clinics on the basis of blood pressure readings on conventional measurement and were not, therefore, representative of tile population. For this reason, the Allied Irish Bank study was set up at Beaumont Hospital, Dublin witli the object of establishing reference values for arribulatory blood pressure levels in a sample of 815 liealtliy bank e~iiployees aged 17 to 79 years." Mean 24 hour ambulatory blood pressure averaged 118/72 nlmHg (systolic/diastolic) while the mean daytime and nighttime levels averaged 124/78 mniHg and 106/61 mmHg respectively. Taking the mean and two standard deviations as the upper limit of normal yielded an upper limit of 24 hour ambulatory blood pressure of 139/87 mmHg, and of daytime and night-time blood pressures of 147/94 mmHg and 127/76 mmHg respectively. A review of studies on non-invasive ambulatory blood pressure monitoring in healthy and apparently normotensive subjects produced broadly similar results.' Although tlie exact relevance of these reference values to end organ effects, morbidity and ~iiortality is not clear, they are nonetheless of practical use in the interpretation of ambulatory blood pressure results and represent an iniportant step forward in the development of the clinical application of the technique. The evaluation and management of hypertension in general practice is generally along the guidelines published by the Britisli Hypertension SocietyR and tlie World Health Organizat io~i ,~ neither o r whicll advise 011 tlie clinical use of a~nbulalory blood pressure Incasurernent. Tlius, there is a need for guidelines on the diagnosis and treatment of hypertension based on ambulatory blood pressure nieasurenlent, siniilar to those for blood pressure nleasuremerit in the clinic. As this is a relatively expensive investigation, priority should be given to those cases where the procedure is niost likely to alter the doctor's management of tlie patient. In the context of general practice, this will mainly be in the area of diagnosis and evaluation of mild to moderate Iiypertension and to a lesser extent in the follow up of treatment. In tlie majority of hypertensive patients tlie only abnormal finding is an elevation of, blood pressure with no evidence of target organ damage as deterniined by physical examination, urinalysis, fundoscopy, electrocardiograpli or echocardiograph. Management is largely determined by what is regarded as the pat ie~i~ 's true' blold pressure. At present, patients in whom diastolic pressures remain greater than 100 mmHg on repeated measurement (perhaps every two weeks) over three to four months are offered treatment on the basis that tlie discrimination of a high risk group can be improved by repeated measurements of blood pressure in the c l i n i ~ . ~ This is because patients diagnosed as having hypertension on measurement in the clinic have a tendency for blood pressure to fall to normal levels on repeated measurement. Since tliis phenomenon does not occur with ambulatory blood pressure ~neasurement , '~ the subject's 'true' blood pressure level can be established on the basis of a single 24 hour recording, thereby obviating the need for multiple surgery visits over a prolonged period. There is general agreement that the decision to initiate drug treatment in a patient diagnosed as hypertensive on the basis of measurements taken in the clinic will be greatly strengthened if the level of the mean daytime blood pressure on ambulatory measurement also remains persistently outside the limits defined as normal for this technique.' However, a more difficult management problem is presented when a diagnosis of 'white coat' hypertension is made, that is where the elevation in blood pressure is transient and confined to the period while tlie patient is in the surgery or hospital setting." The observation that blood pressure nieasurenient may trigger an alerting reaction and a pressor response in a patient lias been made by several workers.12-l4 Julius and colleagues reported tliat about 30% o f subjects with borderline hypertension had high blood pressure readings in the clinic but normal readings at home.I5 Pickering and colleagues found tliat 22% of 292 patients in wliom borderline hypertension had been diagnosed had normal ambulatory blood pressures.I6 The technique of ambulatory blood pressure measurement will enable the general practitioner to identify many patients witli white coat 'hypertension. While there are as yet no results from coritrolled prospective morbidity and niortality studies on which to base clear guidelines, it is generally agreed that these patients d o not require treatment with antihypertensive drugs, at least in the early stages.' Although the benefits to the patient in terms of saved drug costs and lack of side effects from such an