Dr Matthew Lockyer explains how his practice devised a guideline which provides a framework for assessing new abnormal blood pressure
I have always been intrigued by hypertension. It seems, in many ways, to be doctor-created illness in that we decide arbitrarily that a certain level of a physiological measurement is abnormal. This may not cause symptoms, but places the individual at higher risk of cardiovascular damage.
Typically, the medical profession, having created this disease, then proves very bad at defining it, poor at diagnosing it, and even worse at treating it. And even when we do treat it, we keep changing our minds about best management.
The well-known 'rule of halves' tells us that half of all hypertension is undiagnosed; of the diagnosed patients, half are untreated; and of the treated patients, half are on inadequate therapy. Even if your practice population are doing well, the new British Hypertension Society (BHS) guidelines, soon to be published, are expected to contain lower target blood pressures (BPs), so we must now reassess patients who were previously 'well controlled'.
New, well-designed trials, such as the UK Prospective Diabetes Study (UKPDS)1 and the Hypertension Optimal Treatment (HOT) trial2, among others, support vigorous treatment of BP to lower levels.
They herald a return to multi-layered drug regimens – out of vogue for much of the 1990s – and lowered thresholds for initiating drug treatment, in addition to lifestyle measures.
Considering how often I check BP in the surgery, I always find the under-lying messages of poor performance and treatment rather dispiriting.
In our practice we have attempted to improve our performance over the last 10 years by the use of a guideline for doctors and practice nurses on initial diagnosis and investigation of hypertension. This will now be revised for the third time to take account of the new BHS guidelines.
To help make sense of BP detection and assessment, I remember the words of a charismatic physician at medical school: 'Hypertension on its own is not a diagnosis,' he drilled into us. I continue to adhere to his teaching.
Hypertension is either primary (essential or idiopathic) or secondary. Secondary causes of hypertension are rare in primary care – about 5% of new cases is a figure I've seen quoted – but you must not forget to look for them. Remember that the 'PRICE' of missing them can be high (Table 1).
|Table 1: Secondary causes of hypertension|
|Iatrogenic (e.g. steroids, NSAIDs)|
|Endocrine (e.g. Conn's syndrome, Cushing's syndrome, phaeochromacytoma)|
Primary, or essential, hypertension is by far the more common diagnosis. This needs to be qualified as mild, moderate or severe. Finally, the presence or absence of end-organ damage needs to be assessed.
There is no debate about the value of treatment for severe and moderate hypertension, or where end-organ damage occurs. Diabetes seems to be becoming a cardiovascular diagnosis, with BP control as important as bloodhsugar control. The area of difficulty remains diagnosing and managing patients with mild hypertension.
Our practice nurses have been using our guideline for BP checks on new patients, routine screening or cases referred by the GPs.
The purpose of the guideline is to improve our detection of hypertension at all grades, and to ensure that baseline investigations for diagnosis and risk assessment are performed consistently in patients with persistently abnormal BP.
The guideline has been designed by the doctors and practice nurses in collaboration, and laminated copies are provided in the clinic rooms for reference.
The advantages of the guideline, which seems to work well, are that it provides a clear framework for the nurses to assess new abnormal BP, enabling referral of the patient to the doctor as a 'package' complete with multiple measurements, and results of investigations.
This allows the doctor to spend more time assessing the patient for secondary causes of hypertension and end-organ damage. It is easy to forget to check for heart murmurs, cardio-megaly, renal masses or retinopathy. The doctor also has more time to decide on and explain treatment to the patient. In the case of lifestyle measures, the nurses have often prepared the ground and the doctor is reinforcing an existing message.
We have updated our guideline as new evidence has come to light. The first revision took account of lowered treatment thresholds in the elderly, and the recommended emphasis on lifestyle measures over drug treatment in younger hypertensive patients.
We also tried to exclude white coat hypertension by lending patients practice home BP monitors. This proved less successful than we had hoped, mainly because it was difficult to ensure that patients were followed up and investigated in accordance with the guideline when they had to wait to borrow a meter and return with the results.
Recently published studies have shown improvements in outcome with surgery-recorded BPs. Some believe that the white coat hypertension response may itself be pathological. We have therefore stopped using home monitors in diagnosis for the time being, although we still use them to assess treatment especially when changing drug regimens. Our latest revision of the guideline (below) incorporates the new evidence.
|Latest revision of the guideline|
Our guideline is concerned only with diagnosis and initial investigation. Those tricky decisions for management remain – but at least we now have a sound foundation.
This guideline is not offered as a gold standard – we have no idea whether our detection and management is better or worse than that of anyone else. Rather, it is our response to changing advice. It would be very interesting to hear the views of other practitioners on this topic.
- UK Prospective Diabetes Study (UKPDS) Group. Tight BP control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. BMJ 1998; 317: 703-13.
- Hansopp L et al. Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. Lancet 1998; 351: 1755-62.