Consensus can be achieved by methods ranging from gentle persuasion to physical coercion. The method chosen by the Pharmaceutical Benefits Advisory Committee* for the Australian Consensus Conference on the Management of Hypertension held in Canberra on 17-18 November 1993 largely followed that of an earlier conference on lipid-lowering strategies.1 A 'non-aligned' panel was selected from general and hospital practice, epidemiology and public health, health economics and clinical pharmacology. The panel was presented with the views of speakers, acting as advocates, in response to pre-arranged questions. These questions ranged from diagnosis, through the best approaches to management, to the outcomes of hypertension. Speakers had a clear brief - to review the evidence on their allotted question and to present it succinctly and persuasively.

The audience of several hundred came from groups with a special interest in cardiovascular medicine, including the pharmaceutical industry. However, general practitioners were the largest single professional group in the audience.

The presentations themselves generated the greatest heat – one speaker confidently asserting that the conference would never have occurred if all classes of antihypertensive drugs were as cheap as diuretics and beta blockers. That this was at least half true was admitted by the Minister for Health, Senator Graham Richardson, who opened the conference. However, he did couple his concern over cost with the fact that the outcomes of detection and management of hypertension in Australia are not yet as good as those in other comparable countries. We still have a long way to go before all people with hypertension requiring treatment are identified, given appropriate management and adequately followed up. Most of this is work for general practitioners.

After a day of presentations and discussion, the panel retired to draw up a draft consensus statement which was presented and debated the next day. Members of the panel had the brief to base their document on the evidence presented and to give, for example, greater weight to data from randomised controlled trials than from non-randomised or uncontrolled studies, which in turn stood ahead of descriptive studies. The responses to the 16 preset questions were later modified by the panel in the light of the second afternoon's debate to produce the final consensus statement.2

Some of the main conclusions are worth emphasis. A diagnosis of hypertension begins with careful measurement of blood pressure repeated on several occasions. (Precise guidelines for measurement also emerged from this conference.) Normotension can be defined as a pressure consistently less than 140 mmHg systolic and 90 mmHg diastolic. The urgency of confirmation, investigation and treatment progressively rises with higher diastolic and/or systolic pressures.

Routine, adequate history-taking and physical examination should establish the possibility of underlying causes, the extent of target organ damage and the presence of associated vascular risks. Recommended investigations in confirmed cases of hypertension are urine analysis and microscopy, measurement of plasma potassium, creatinine, uric acid, fasting glucose and lipid profile, and an electrocardiogram. It was recognised that the evaluation of left ventricular mass by echocardiography is increasingly being used in the initial and ongoing assessment of hypertensive patients, but it was concluded that the overall role of echocardiography has not been established and requires further definition.

The evidence for a reduced risk of stroke and coronary events when hypertension is treated is so strong that treatment should be considered in patients with blood pressure consistently greater than 140 mmHg systolic and 90 mmHg diastolic. Lifestyle modification, including increasing exercise, reducing body weight and sodium and alcohol intake should be primary interventions unless the blood pressure is high enough to warrant urgent drug treatment. Measures to stop smoking, which adds to vascular risk, have a high priority.

If, after a trial of lifestyle modification, blood pressure is no lower, drug treatment should begin. All major classes of antihypertensive drugs are effective in lowering blood pressure. (The present utilisation of antihypertensive drugs is shown in Fig. 1.) If no reason such as a contraindication or co-existing condition exists to favour a specific drug, it is rational to choose a drug that has been shown to provide beneficial effects in major mortality/morbidity trials. At present, such trials have only been performed using diuretics and beta blockers as first-line treatment. Angiotensin-converting enzyme inhibitors are of particular benefit in hypertension with associated heart failure or left ventricular dysfunction and in hypertensive diabetics with microalbuminuria. Preliminary evidence suggests calcium channel blockers may inhibit progression of atheroma.

Failure to respond to apparently adequate medication should lead to renewed consideration of a missed underlying cause, the ingestion of potential antagonists such as non-steroidal anti-inflammatory drugs and, more usually, poor compliance. Combinations of drugs are valuable when monotherapy fails. While most patients started on drug therapy require long-term treatment, those with mild hypertension who successfully modify their lifestyle may be able to reduce or stop medication under careful supervision.

These, in summary, are the main points from the 1993 Consensus Conference. What of the process itself? While the proceedings went smoothly and relatively few points proved very contentious, there are clearly opportunities for manipulation of data and conclusions inherent in this form of consensus process. For example, the selection and wording of the questions may effectively ensure a particular answer. Panel members (on this occasion, I am sure, genuinely unbiased) could be selected by a process which automatically puts a bias in the outcome. So, is this the best way of achieving authoritative guidelines on major issues in therapeutics? Probably not, for evidence-driven guidelines must be revised and modified as new evidence presents – a continuous process, not a set-piece, one-off event such as a consensus conference. Perhaps the proposed development of Cochrane Centres3 charged with exactly this task may provide legislators, prescribers and consumers alike with the best objective data available for treating not only hypertension but also many other common conditions.

Editorial note: The conference proceedings will be published as a supplement to Australian Prescriber later this year.

* with co-sponsors, the Pharmaceutical Benefits Scheme Education Program, the National Heart Foundation of Australia, the High Blood Pressure Research Council of Australia, the Royal Australian College of General Practitioners and the Health Care Committee of the National Health and Medical Research Council.

Fig 1

Community use of antihypertensive drugs. For the definition of defined daily dose (DDD), see 'Monitoring drug use in Australia' Aust Prescr 1993: 16:27-9. The Cyclical dispensing patterns are a result of safety net provisions which were introduced into the Pharmaceutical Benefits Scheme in November 1986.


Anthony J. Smith

Professor of Clinical Pharmacology, University of Newcastle, Mater Misericordiae Hospital, Waratah, N.S.W