Drugs 25 (Suppl. 2): 326-330 (1983) 0012-6667/83/0300-0326/$02.50/0 " ADIS Press Australasia Ply Ltd (Inc. NSW). All rights reserved.
Clinical Evidence that ~-Adrenoceptor Blockers Prevent More Cardiovascular Complications than Other Antihypertensive Drugs D.G. Beevers, J.H. Johnston, H. Larkin and P. Davies MRC Blood Pressure Unit, Western Infirmary, Glasgow, and Department of Statistics, University of Birmingham, Birmingham
Summary
Using a computer file from the Glasgow Blood Pressure Clinic, an analysis was made of the results of treatment in 920 (453 male and 467 female) consecutive patients below the age of 65 years who received antihypertensive medication. Vascular complications had developed prior to treatment in 242 patients. is-Blocking drugs (mainly oxprenolol and propranolol) were used alone or in combination with other drugs at some stage in 416 patients, whereas 504 patients never received /i-blockers. There was a highly significant tendency both for men and women who had received is-blockers to suffer fewer heart attacks and strokes than those treated with other types of therapy. This trend was found in patients presenting initially with or without previous vascular complications, and was present both in patients aged 45 to 54 and 55 to 64 years of age. There was no significant difference in the height of the blood pressure prior to starting therapy in the 2 therapeutic sub-groups, and no difference in the average blood pressure while receiving treatment. On the basis of this uncontrolled study of the results of treating patients in a busy blood pressure clinic, we conclude that is-adrenoceptor blockers have advantages over other antihypertensive drugs. Formal clinical trials are needed to confirm this, but our data suggest that all hypertensive patients should receive a ,a-blocker as part of their antihypertensive regimen (provided that specific contraindications do not exist).
1. Introduction The ,B-adrenoceptor blockers were originally introduced for the treatment of established coronary heart disease, and recently they have been shown to be of help in preventing the recurrence of myocardial infarction (BHAT, 1981; Norwegian Multicenter Study, 1981; Hjalmarson et aI., 1981). However, it is still not known whether ,B-blockers have added benefits over other drugs in preventing coronary heart disease and other vascular complications in hypertensive patients. Two small uncontrolled studies suggest that this might be so (Lambert, 1972; Stewart, 1976), and formal clinical trials are currently underway to resolve this (MRC, 1977). In the meantime, useful information can be obtained by conducting an analysis of the results
of treatment in large blood pressure clinics where some patients have received ,B-blockers and others have not. This approach has the advantage that it is conducted in all patients, not only in the minority of cases who are suitable or willing to enter clinical trials. We have therefore investigated the outcome of treatment in 920 consecutive patients attending the Glasgow Blood Pressure Clinic.
2. Methods The Glasgow Blood Pressure Clinic (1972) is a multicentre computerised clinic which was established to facilitate mounting of pooled research projects using large numbers of patients.
327
Advantages of Il-Blockers
2.1 Patients
2.2 Diagnoses
The present analysis was conducted in 920 consecutive male (453) and female (467) patients aged 35 to 64 years, who attended the clinics at the Western and Royal Infirmaries, the MRC Blood Pressure Unit and the Renfrew Blood Pressure Clinic (Beevers et aI., 1976). Patients were included only if their pretreatment blood pressure readings were available, they had received antihypertensive drug therapy, and the outcome of treatment was known. For this reason, special efforts were made to contact general practitioners to ensure that patients who had stopped attending had not done so on account of vascular complications of hypertension. Patients who had surgical treatment for hypertension were excluded. Of the 920 cases studied, 105 had some evidence of intrinsic renal disease or renal artery disease, but they did not undergo surgery.
The underlying diagnoses as well as the diagnoses of the cardiovascular complication of hypertension were based on the opinion of the clinician managing the patient at the time. As these were busy routine clinics, many different physicians (both senior and junior) had seen the patients, and no formal criteria of diagnosis could be employed. Similarly, in patients who died, the cause of death was based on the death certificate completed by the doctor looking after the patient at the time. The diagnostic categories for vascular complications were: 1. stroke including transient cerebral ischaemic attack 2. ischaemic heart disease, either first episode of angina or first heart attack 3. other vascular diseases, which included heart failure and claudication. 2.3 Treatment Groups
o~
NO PREVIOUS COMPLICATIONS
Blocker
~
other druqs
BP mm Hg 225
BP mm Hg 225
200
200
115
175
150
150
125
125
100
100
15
Before
Treated MEN
Before
Treated
75
WOMEN
The allocation of patients into different therapeutic regimens was based on the clinician's wishes at the time. For the purpose of this analysis, patients were categorised into those who received: a) a (j-blocker (usually propranolol or oxpren0101) without a diuretic b) a (j-blocker with a diuretic c) a diuretic with or without other drugs but excluding a (j-blocker d) patients who received neither a diuretic nor a (j-blocker but who received other drugs (mainly methyldopa). 2.4 Assessment The quality of control of blood pressure was calculated by taking the mean of blood pressure readings within each 6-month period of attendance at the clinic and then calculating a grand mean of these 6-monthly means. Thus, the quality of control is characterised by means ofa single measure of blood pressure whilst receiving drug therapy. The average duration of follow-up in the whole series was 44.4 months.
3. Results Fig. 1. Systolic and diastOlic blood pressures both before and during treatment in 678 previously uncomplicated hypertensives, in relation to the type of antihypertensive therapy .
Of the 920 patients included in the study, 242 (98 female) had experienced a vascular complica-
Advantages 01 (I-Blockers
328
Table I. Patients [number (%)] developing their first vascular complication while receiving antihypertensive medication
Treatment
Number
Developed vascular complication no
a) b) c) d)
(I-Blocker. no diuretic /l-Blocker. with diuretic Diuretic. no I'I-blocker Other drugs
yes
Stroke
Other vascular disease
10 10 20 7
3 7 13 5
1 8 14 4
98 208 291 81
84 183 244 65
14 25 47 16
a + b) /l-Blocker with or without diuretic
306
267
39 (12.7)
20 (6.5)
10 (3.3)
9 (2 .9)
+ d) Other drugs b + c) Diuretic with
372
309
63 (16.9)
27 (7 .3)
18 (4.8)
18 (4.8)
499
427
72 (14.4)
30 (6.0)
20 (4.0)
22 (4.4)
179
149
30 (16.7)
17 (9.5)
8 (4.5)
5 (2.8)
678
576
c
or without
blocker
a
+ d) Other drugs
Total
p-
102
tion of hypertension prior to receiving antihypertensive drug therapy; this group was analysed separately. 678 patients were treated for hypertension without previous clinical evidence of complications of hypertension. Of these, 306 were treated
PREVIOUS COMPLICATIONS
D f3 Blocker
~ other drugs
BP mm Hg
BP mm Hg
rn
rn
200
200
m
m
75
Before
Treated MEN
Before
Treated
75
WOMEN
Fig. 2. Systolic and diastolic blood pressures both before and during therapy in 242 previously complicated hypertensive patients. in relation to the type of antihypertensive therapy.
(14.3) (12.0) (16.1) (19.8)
Angina or myocardial infarction
47
(10.2) (4 .8) (6.9) (8.6)
28
(3.1) (3.4) (4 .5) (6.2)
(1.0) (3.8) (4 .8) (4 .9)
27
with a i3-blocker and 372 did not receive a 13blocker. A vascular complication of hypertension developed in 12.7% of patients receiving a i3-blocker and 16.9% of patients not receiving a i3-blocker. This reduction in complications in the patients on ,a-blockers was reflected in fewer strokes, heart attacks and other vascular complications (table I). This effect was not explained by these patients either having milder hypertension in the first place or having better control of blood pressure from their drugs (fig. 1). Furthermore, the advantageous effect of {:Jblockade was observed separately in patients aged 45-54 and 55-64 years when starting treatment, as well as in men and women separately. The number of complications developing in patients aged 35 to 44 years was too few for confident statistical analysis. Also, our data show no evidence that patients who received a diuretic in their antihypertensive regimen fared badly; 14.4% of patients on thiazides developed further complications, whereas 16.7% of patients who did not receive a thiazide developed complications. The main difference between the groups was apparently due to P-blockade with or without diuretic. Of the 242 patients who received antihypertensive medication only after they had developed a vascular complication of hypertension, 69 (28.5%) developed a further complication. Table II shows that 16.5% of the group on It-blockers and 38.3% of the group not on It-blockers developed further vascular complications. Again, this difference could not be accounted for by differences in the severity or the quality of control of blood pressure (fig. 2). There was also no evidence that thiazide diuretics conferred any disadvantage in this group of patients.
329
Advantages of /l-Blockers
Table II. Number (%) of previously complicated hypertensives who developed a second vascular complication while receiving antihypertensive medication
Treatment
Number
Developed further vascular complication no
a) b) c) d)
/l-Blocker, no diuretic /l-Blocker, with diuretic Diuretic, no /l-blocker Other drugs
a + b) /l-Blocker with or without diuretic c
+ d)
Other drugs
b + c) Diuretic with or without (Jblocker a
+ d)
Other drugs
Total
yes
18 (16.5)
8 (7.3)
9 (8.3)
1 (0.9)
(16.3) (16.7) (37.2) (41.9)
(7.0) (7.6) (12.7) (S.4)
133
82
51 (38.3)
15 (11.3)
21 (15.8)
15 (11.3
168
119
49 (29.2)
18 (10.7)
21 (12.5)
10 (6.0)
74
54
20 (27.0)
5 (6.8)
9 (12.2)
6 (8.1)
242
173
69
developed ischaemic heart disease
14 (32.6%) 5(7.9%)
7 (16.3%) 3 (4.8%)
4(6.1%)
1 (2.3) 0(0) 10 (9.8) 5 (16.1)
91
Patients with previous stroke (106)
9 (17.6%)
3 (7.0) 6 (9.1) 15 (14.7) S (19.3)
109
Patients with previous ischaemic heart disease (117)
No (J-blocker (51) (J-Blocker (66)
3 5 13 2
7 11 38 13
developed further stroke
developed myocardial infarction
Other
36 55 64 18
Table III. Results of treatment in patients with previous stroke or ischaemic heart disease
No {J-blocker (43) {J-Blocker (63)
Stroke
43 66 102 31
Among the patients who had previous complications of hypertension, 117 had had previous ischaemic heart disease (either first episode of angina or first myocardial infarction), and 66 ofthese received a fj-blocker (table III). Myocardial infarction or reinfarction developed in 6.1 %, whereas of the 51 patients who did not receive a fj-blocker, 17.6% developed infarction or reinfarction. There was also a minimal reduction in the development of stroke in these patients. 106 patients had had a stroke prior to antihypertensive therapy; of these, 63 received fj-blockers and 43 did not. Second strokes developed in 7.9% of the patients on fJblockers and 32.6% of the patients not on fJ-blockers. Furthermore, only 4.8% of patients who received fj-blockers who had had I stroke developed
Drug regimen
Angina or myocardial infarction
developed stroke
4 (7.8%)
4 (6.1%)
23
30
16
ischaemic heart disease whereas 16.3% of those patients who did not receive a ,B-blocker developed ischaemic heart disease.
4. Discussion These data demonstrate that hypertensive patients who received a fj-blocker as part of their antihypertensive medication fared better with fewer heart attacks and strokes than similar patients who did not receive a i9-blocker. The effect was most impressive in patients who had had previous heart attacks and strokes, and was manifest by a reduction in the occurrence of a second episode of heart disease as well as in second stroke. Benefits were not due to differences in the severity of the hypertension prior to therapy or to differences in the quality of control of blood pressure. It must be stressed, however, that this study is not a formal clinical trial and that these trends need confirmation in a large scale project. This project reflects the situation in busy dinics and inevitably many objections can be made. Firstly, no allowance has been made for other coronary risk factors; secondly, no account has been taken of patient compliance with hypertensive regimens. This point is partly answered by the similarity of the quality of blood pressure control in patients both on or otT fj-blocker therapy. One advantage of this study is that it was possible to check the hypothesis that the drug treatment of hypertension using a thiazide diuretic may be harmful. It is theoretically possible that the
Advantages of /l-Blockers
beneficial antihypertensive effect of thiazide diuretics might be offset by their harmful metabolic side effects, and thiazides might actually be causing as much coronary heart disease as they prevent (Bloxham and Beevers, 1979). This analysis is reassuring as no adverse effect of thiazides was seen. Points raised in this study need confirmation in formal trials, and the MRC Trial of Hypertension (1977) should go some way towards this. In the MRC trial, patients are randomly allocated to either placebo or active treatment. The active treatment group is then sub-divided into those who receive a thiazide without a {3-blocker and those who receive a /1-blocker without a thiazide. However, the problem with the MRC trial is that if the {3-blocker group fare better, this might be due to a beneficial effect of {3-blockade or, conversely, to a harmful effect ofthiazides. In theory, therefore, it would be useful to have a third group of treated hypertensives who receive neither a thiazide nor a {:t-blocker, but this would not be feasible. Previous studies of data from this project have stressed 2 other points, namely that the quality of control of blood pressure, the achieved blood pressure, is of overriding importance in predicting outcome, and is more important than the severity of the hypertension in the first place (Beevers et at, 1978). This underlines the need for good quality of blood pressure control in hypertensive patients. We have also found that the quality of control of systolic and mean arterial blood pressure is just as important as the quality of control of diastolic pressure. Our data suggest that a .8-receptor blocker should be used in the antihypertensive regimen of all patients, unless of course there are specific contraindications. Thiazide diuretics can be added safely in the second step of therapy. The implications of this are great. It is possible that more convincing evidence of coronary prevention might be
330
obtained if fj-blockers were used routinely. However, a very important consideration is the large difference in cost of (3-blockers and thiazides, the latter being considerably less expensive. For this reason, the prospective clinical trials currently underwayare urgently needed.
References Beevers, D.G.; Duncan, S.; Nelson, C.S. and Padfield, P.L.: A blood pressure clinic in a health centre. Postgraduate Medical Journal 52: 683-686 (1976). Beevers, D.G.; Johnston, 1.; Devine, B.L.; Dunn, F.G.; Larkin, H. and Titterington, D.M.: Relation between prognosis and blood pressure before and during treatment of hypertensive patients. Clinical Science and Molecular Medicine 55(Suppl. I): 333-336 (1978). BHAT (The Beta Blocker Heart Attack Trial). Journal of the American Medical Association 246: 2073-2074 (1981). Bloxham, C.A. and Beevers, D.G.: The effect of thiazide diuretics on coronary risk factors. Postgraduate Medical Journal 55(Suppl. 3): 9-13 (\ 979). Glasgow Blood Pressure Clinic. Journal of the Royal College of Physicians of London 7: 87-92 (1972). Hjalmarson, A.; Elmfcldt, 0 .; Herlitz, J.; Holmberg, S.; Malek, \.; Nyberg, G.; Ryden, L.; Swedberg, K.; Vedin, A.; Waagstein, F.; Waldenstrom, A.; Waldenstrom, J.; Wedel , H.; Wilhelmsen, L. and Wilhelmsson, c.: Effect on mortality of metaprolol in acute myocardial infarction. Lancet 2: 823-827 (1981). Lambert, D.M.D.: Beta-blockers and life expectancy in ischaemic heart disease. Lancet I: 793-794 (1972). MRC (Medical Research Council). A randomised controlled trial for mild to moderate hypertension, design and pilot trial ex~ perience. British Medical Journal I: 1437-1441 (1977). Norwegian Multicenter Study: Timolol-induced reduction in mortality and re-infarction in patients surviving acute myocardial infarction. New England Journal of Medicine 304: 801807 (1981). Stewart. \'McD.G. : Compared incidence of fir1>! myocardial infarction in hypertensive patients under treatment containing propranolol or excluding beta-blockade. Clinical Science and Molecular Medicine 51(Suppl.): 509-511 (1976). Author's address: Dr D. G. Beevers. Department of Medicine, Dudley Road Hospital, Birmingham BIB TQH (England).