Report of the Canadian Hypertension Society Consensus Conference
Report of the Canadian Hypertension Society Consensus Conference
Canadian Medical Association Journal 1993; 149: 289-293
[en bref]
~ Members of the organizing committee.
Preparation for the 1992 consensus conference began with the striking of four panels (Appendix 1), each charged with the review and revision of existing guidelines in its area of responsibility according to new evidence. Virtually all the specialists invited were able to participate. Each panel included at least one member who had participated in the earlier consensus conference on the subject.
The panels were charged with doing extensive reviews of the research to determine the current state of evidence concerning the measurement of blood pressure and the diagnosis and treatment of hypertension. The literature review of the levels of blood pressure at which the initiation of antihypertensive therapy does more good than harm for various groups was based on the most recent CHS report on pharmacologic treatment, augmented by formal MEDLINE searches. In addition, searches were conducted on the role of echocardiography, self-assessment of blood pressure and ambulatory blood pressure monitoring in diagnosis.
The goal of the search was to find solid evidence from which to derive recommendations that could be implemented by Canadian general and family practitioners the physicians who detect and manage most cases of hypertension. Journal articles were rated according to the methodologic strength of the studies they described (Tables 1 to 4). This approach was synthesized from a number of sources.[6-9]
The recommendations were graded according to the level of the evidence supporting them[10] (Table 5). Thus, grade A recommendations are based on very strong evidence and should be given the highest priority by practitioners. For the most part, grade D recommendations concern practical issues of implementation and are based on expert opinion, clinical experience and the common sense of the panelists. Recommendations for further research were made according to each committee s perceptions of the importance to clinical management of a topic and the lack of satisfactory evidence to date; thus they were not graded. Similarly, recommendations that the evidence was insufficient to warrant clinical use of a procedure (such as home blood pressure monitoring) were not graded.
Each panel prepared a series of draft reports that were circulated among panel members and selected organizers. There were frequent communications between the panels chairpeople and the organizers to establish the level of progress and to discuss issues of potential overlap between groups. Final draft reports from each group were circulated to all panelists before the Montreal conference.
All the proposed recommendations were carefully discussed during the 2-day conference. Each panel met in committee and in plenary sessions. Observers from the pharmaceutical industry considered the final draft recommendations in each of two independent committees (Appendix 1) and provided comments during the plenary sessions, although they did not vote on the recommendations. The organizers participated in the deliberations of at least two of the panels and also engaged in the discussions of the observers from the pharmaceutical industry. There was ample opportunity for debate between the members of each expert panel and the other panelists and observers during the plenary sessions. Finally, a secret vote on each recommendation was conducted among the entire group of expert panelists.
The recommendations were presented and discussed in two public meetings, first in May 1992 at the International Heart Health Conference, in Victoria, and then in September 1992 at the annual meeting of the Royal College of Physicians and Surgeons of Canada, in Ottawa. Both presentations provided vigorous debate and commentary but no new information.
Particularly important was that none of the grades of recommendations was regarded by the panelists as superseding the need for clinical judgement. Thus, the circumstances of individual patients may take precedence over the recommendations, even those given a grade of A. Furthermore, the recommendations were based on evidence available in the spring of 1992, and new evidence should be considered as it emerges. To facilitate this process the key studies supporting each recommendation graded A through C are cited with the recommendation.
Recommendations of this type are of value only if they are accepted and applied to the population for whom they were intended. Their publication is the next step in the education of the physicians and other health care professionals who are responsible for detecting and managing hypertension. The 1992 recommendations will be included in this series of articles, which deals with the topics of diagnosis, pharmacotherapy, hypertension in elderly people, and diabetes and hypertension.
Both the report of the JNC V and our report on diagnosis deal with issues such as ambulatory blood pressure monitoring and the use of echocardiography for the detection of left ventricular hypertrophy. The US report leans toward a more liberal use of these techniques in the detection and evaluation of hypertension than we felt was appropriate from the available information. It suggests selective use of both techniques, whereas our recommendations are for further evaluation and standardization before their use is formally endorsed.
In the JNC V report there is a shift from the wide range of initial monotherapies approved in the Joint National Committee s fourth report[14] (JNC IV) to a recommendation that diuretics and ß-blockers are the drugs of choice, unless these are contraindicated or there are special reasons to give other agents. This conclusion is based on the fact that these are the only classes of drugs that have been shown in long-term controlled clinical trials to reduce rates of illness and death precisely the rationale of the report of the 1988 Canadian consensus conference on pharmacotherapy[2] and a guiding principle for the 1992 conference. Thus, the US recommendations on the initial drug treatment of hypertension are now in concert with the Canadian ones.
The reversal of the JNC IV recommendation was not entirely welcomed by the authors of the accompanying editorial,[12] Weber and Laragh, who have interpreted it as being somewhat regressive and narrowly legalistic and as placing too much reliance on the benefits observed in cardiovascular end points in studies using diuretics and ß-blockers. The authors of the report of the 1988 Canadian consensus conference[2] and those of the report on pharmacologic treatment from the 1992 Canadian consensus conference (the third article in this series) have encountered similar criticisms, often from physicians who find the putative adverse effects of diuretics and ß-blockers on lipids and other metabolic factors to be convincing reasons for choosing a-blockers, angiotensin-converting enzyme (ACE) inhibitors or calcium entry blockers as the routine drugs of choice, even though no long-term studies have been done to show that these drugs reduce the rates of cardiovascular illness and death.
There are subtle differences between the therapeutic algorithm proposed by the JNC V and that of the Canadian consensus conference. Most important, the Canadian report is more explicit in its recommendation of alternative monotherapy when the combination of a low-dose diuretic and a ß-blocker is unsuccessful. Weber and Laragh[12] have affirmed our perspective: At the very least, physicians should be encouraged to substitute differing drug types if their first selections do not work adequately. Sequential monotherapy will increase the number of patients controlled with one drug instead of two or more.
With regard to the management of hypertension in the elderly we largely agree with the JNC V a matter of little surprise, since both sets of recommendations relied heavily on the same limited database of large-scale trials of antihypertensive drugs in older patients.
In the management of diabetic patients the JNC V is more tolerant than we are of the role of diuretics. It expresses caution about the use of most drugs in diabetic patients and merely warns the reader to recognize that hypokalemia induced by diuretics may reduce glucose tolerance. Our recommendations, on the other hand, conclude that diuretics should be avoided, since there is evidence that death rates may increase in diabetic patients treated with thiazides.
The special report on primary prevention[13] reiterates much of what was stated in the report of the CHS consensus conference on nonpharmacologic management of hypertension[5] but from the perspective of prevention rather than treatment. The authors of the special report encourage the concept of lifestyle modification through education; in fact, many of their conclusions are based on extrapolations of the effects of nonpharmacologic therapy in patients with confirmed hypertension rather than on conclusive trials of primary prevention. We agree with Weber and Laragh[12] that it is premature to conclude that this approach will be as effective or as cost-effective as the authors might wish.
In summary, the recommendations of the JNC V and those of the 1992 Canadian consensus conference are much closer than those of the 1988 and 1989 reports.[2,14] In both Canada and the United States there will be those with a healthy cynicism about diuretics and ß-blockers as the first choice in therapy. However, the onus is clearly on the supporters of a-blockers, ACE inhibitors and calcium entry blockers to demonstrate that these medications are at least as successful in reducing rates of illness and death as diuretics and ß-blockers.[15,16] We expect JNC VI and the next Canadian consensus conference on pharmacotherapy to have explicit answers to these concerns.