Mortality Rates in Treated Hypertensive Men With Additional Risk Factors Are High But Can Be Reduced

Transcription

1 AJH 1998;11:14 22 Mortality Rates in Treated Hypertensive Men With Additional Risk Factors Are High But Can Be Reduced A Randomized Intervention Study Björn Fagerberg, John Wikstrand, Göran Berglund, Ola Samuelsson, and Stefan Agewall, for the Risk Factor Intervention Study Group* The aim was to examine the feasibility and efficacy of a multifactorial risk factor intervention program in hypertensive patients at high cardiovascular risk. Treated hypertensive men, aged 50 to 72 years, with at least one of the following: serum cholesterol concentration > 6.5 mmol/l, diabetes mellitus, or smoking were randomized to multifactorial risk factor intervention (n 253) or usual care (n 255). The specific intervention was based on group meetings to encourage a lipid lowering diet and smoking cessation. Cholestyramine, nicotinic acid, fibrates, and later statins were used either as single drug therapy or in combination, following agreed guidelines in patients in whom the nonpharmacological intervention was judged to be insufficient. Usual care was given according to clinical practice. The median follow-up time was 6.6 years. Sixty-four patients (25.1%) died in the usual care group, compared with 41 patients (16.2%) in the intervention group (P.016; 95% confidence interval, relative risk 0.42 to 0.92). The overall risk for fatal and nonfatal cardiovascular events was 29% lower in the intervention group than in the usual care group (P.041). Relative to usual care, the intervention program lowered mean in-trial serum concentrations of total cholesterol (6.3%, P <.0001), LDL cholesterol (9.1%, P <.0001), and blood glucose (0.2 mmol/l, P <.05). Among smokers, at entry, cotinine-adjusted quit rates were 28% in the intervention group and 11% in the usual care group (P.012) after 3 years. This study illustrates the very high cardiovascular risk in hypertensive patients 50 to 72 years of age with additional risk factors. The results indicate, however, that the gloomy prognosis may be improved by a dedicated risk factor intervention program. Am J Hypertens 1998;11: American Journal of Hypertension, Ltd. KEY WORDS: Hypertension, hypercholesterolemia, diabetes mellitus, smoking, mortality. Received December 27, Accepted July 28, From the Departments of Medicine (BF, SA), Sahlgrenska University Hospital, Göteborg University, Gothenburg Nephrology (OS), Sahlgrenska University Hospital, Göteborg University, Gothenburg, and the Wallenberg Laboratory for cardiovascular research (JW), Sahlgrenska University Hospital, Göteborg University, Gothenburg, and the Department of Medicine (GB), Malmö University Hospital, Lund University, Malmö, Sweden. JW is also senior medical advisor to Astra-Hässle Cardiovascular Research Laboratory. * The RIS group also consists of Ove K. Andersson, Marianne Hartford, Thomas Hedner, Hans Herlitz, Susanne Ljungman, Bengt Persson, Bengt Widgren, and Marian Wysocki. This study was supported by grants from the Swedish Medical Research Council (B92-19X A, B93-19X B, B94-19X B and 03A, B95-19X B), the Swedish Heart and Lung Foundation, the Swedish Hypertension Society, King Gustaf V and Queen Viktoria Foundation, Astra Hässle Cardiovascular Research Laboratories, Mölndal, Sweden. Address correspondence and reprint requests to Björn Fagerberg MD, Department of Medicine, Göteborg University, Sahlgrenska University Hospital, S Gothenburg, Sweden; bjorn.fagerberg@medfak.gu.se 1998 by the American Journal of Hypertension, Ltd /98/$19.00 Published by Elsevier Science, Inc. PII S (97)

2 AJH JANUARY 1998 VOL. 11, NO. 1, PART 1 MULTIPLE RISK INTERVENTION 15 The coexistence of hypercholesterolemia, diabetes mellitus, and smoking considerably increases the risk for cardiovascular disease in hypertensive patients. 1,2 Therefore multifactorial risk factor intervention has been advocated in the management of patients with high blood pressure and concomitant risk factors. 3 However, the multiple risk factor intervention trials conducted so far, mainly in middle aged men, have either not included hypertensives, 4 or the results have not shown convincingly that concomitant risk factors can be reduced and prognosis improved in patients with hypertension. 5 9 The multifactor Primary Prevention Trial in Göteborg, Sweden, contained a hypertensive cohort of 47 to 54-year-old men. 10,11 Analysis of the results from the first 10 years of treatment revealed that hypercholesterolemia and smoking were common and that these factors, regardless of blood pressure level, were associated with a substantial excess mortality. 11 This emphasized the potential importance of a multiple risk factor intervention approach, 11,12 which had to be evaluated in a controlled trial. Accordingly, the primary aim of this study was to develop and examine the feasibility and efficacy of a multifactorial risk factor intervention program directed towards hypercholesterolemia, tobacco smoking, and diabetes mellitus in hypertensive patients at high cardiovascular risk. METHODS Design The design of the Risk Factor Intervention Study (RIS) has been described in detail elsewhere. 13 The study was a randomized, open, parallel group trial of multiple risk factor intervention against usual care lasting for 6 years. Follow-up examinations were planned after 1 year to investigate the effect of the initial (mainly nonpharmacological) program, and again after 3 and 6 years. 13,14 The study started October 21, 1987, and closed June 1, The study was conducted at one hypertension outpatient clinic. All patients were seen biannually by the 10 physicians who worked at the clinic. A specific team consisting of one physician (B.F.), one dietician, and three nurses, was responsible for encouraging the life style changes. The study protocol was approved by the ethics committee of the Faculty of Medicine, Göteborg University. Patients The inclusion criteria were, except for treated hypertension, 11 male gender, age 50 to 72 years, and one or more of the following: serum cholesterol concentration 6.5 mmol/l, diabetes mellitus, or tobacco smoking. 13 Diabetes mellitus was defined as fasting blood glucose concentration 7.0 mmol/l on two occasions. 13 The serum cholesterol level was based on the mean of two values measured before inclusion in the trial using an enzymatic technique. 15 The exclusion criterion was malignant or other serious disease. Eligibility for participation in the study was evaluated in all available patients originally belonging to two cohorts of hypertensive men who were representative samples from the general population. 10,11,16 Consent was obtained from every patient after he had been given oral and written information. Randomization was done by an assigned physician, who was not responsible for patient care, after stratification for serum cholesterol (below or above 7.3 mmol/l), current smoking (no or yes), and history of cardiovascular disease (no or yes) using computer generated random tables. As previously described, 14 the groups were well balanced. The mean age was 66.4 (6.2) years; systolic and diastolic blood pressures were 155 (19) and 91 (8) mm Hg, respectively; total serum cholesterol was 6.7 (1.2) mmol/l; LDL cholesterol was 4.6 (1.1) mmol/l. Diabetes mellitus, smoking, or serum cholesterol 6.5 mmol/l were observed in 22%, 29%, and 74% of patients, respectively. Eight percent of the patients had suffered from acute myocardial infarction and 6% had a history of previous stroke. Follow-Up Procedures and Examinations All patients had our routine diagnostic work-up. 10 At entry, prior or present cardiovascular disease as well as diabetes mellitus were documented. 11 The patient answered a questionnaire on smoking habits. 11 Data on blood pressure, serum concentrations of cholesterol and triglycerides, two as well as one year prior to inclusion in the study were also collected from the standardized case record forms at the Hypertension Clinic. At entry and at the biannual visits body weight, blood pressure, 12-lead electrocardiogram, blood glucose, serum cholesterol, and triglyceride concentrations were measured as previously described. 13 Glycosylated hemoglobin (HbA 1c ) was monitored biannually in patients with diabetes mellitus. 13 Once yearly, cardiovascular or other events as well as medication and smoking habits were recorded. 11 Smoking habits were examined with a questionnaire 11 and also after 3 and 6 years by measuring the urinary concentration of cotinine (a nicotine metabolite that is a very sensitive biochemical marker of smoking habits 17 ), which was determined by radioimmunoassay (Diagnostic Products Corp., Los Angeles, CA). Patients who denied smoking were still defined as smokers if both the urinary cotinine and the urinary cotinine/creatinine concentrations were above the decision limits (2.0 mol/l and 1.0 mmol/mol, respectively) unless snuff, nicotine chewing gum, or other nicotine preparations were used.

3 16 FAGERBERG ET AL AJH JANUARY 1998 VOL. 11, NO. 1, PART 1 Intervention For the intervention group the aims were to obtain a serum concentration of total cholesterol 6.0 mmol/l, to help current smokers to quit, to discourage nonsmoking patients from starting to smoke, and to reduce the HbA 1c level to 6.0% in diabetic patients. In addition, the aim in both groups was to reduce diastolic blood pressure to 90 mm Hg, and to a similar level so that any difference in end point rates could not be attributable to differences in blood pressure between the two groups. The special intervention team, consisting of one physician, one dietician, and the nurses at the hypertension clinic, was responsible for implementing the nonpharmacological treatment. The physicians at the hypertension clinic were responsible for motivating the patients to participate in this program and thereafter to reinforce the induced changes in diet and smoking habits and, whenever necessary, to initiate or change pharmacological therapy. As previously described in detail, 13 the nonpharmacological intervention program was based on one information meeting followed by five weekly 2 h sessions with the participation of patients and spouses. Behavioral treatment principles were used to change eating habits. Basic nutrition, along with the purchase and preparation of food were discussed using a slide series, a specially developed text book, and and an exhibition of foods and beverages. The dietary recommendations were similar to those of established guidelines. 18,19 Overweight patients individually defined a realistic target weight that they would aim to achieve. All diabetic patients were systematically taught self-monitoring of blood glucose. A follow-up visit with the nurses at the hypertension clinic took place 4 months after entry and thereafter in connection with one of the biannual appointments with the responsible physician to confirm the management strategy. The patients also completed a short questionnaire that evaluated their present eating habits. Cholestyramine, nicotinic acid, fibrates, and later statins were used either as single drug therapy or in combinations following agreed guidelines 19 in patients for whom the nonpharmacological intervention was judged to be insufficient. The smoking cessation program started with the information meeting and an invitation to the patient and his spouse to see the physician in the intervention team. At this visit smoking habits, symptoms of diseases secondary to nicotine usage, psychological and social factors, and motivation for quitting smoking were discussed. The cessation program consisted of five weekly 1 h meetings under the responsibility of a physician using behavioral treatment principles. Nicotine gum was offered to reduce symptoms of nicotine abstinence. In the usual care group, the physician responsible for the individual patient treated hypercholesterolemia, diabetes mellitus, and smoking according to normal clinical practice. Drug treatment at entry, after 3.3 years, and at the final examination is shown in Table 1. -Blockers were used significantly more in the usual care group than in the intervention group at entry and at the end of the study. Lipid lowering agents were used more often in the intervention group. Assessment of End Points and Data Access The mean number of follow-up visits was in the usual care group and in the intervention group. All possible, expected follow-up visits were carried out by 90% of the patients in the usual care group and by 92% of the patients in the intervention group (P NS). One patient in the former group and three patients in the latter group never attended the follow-up visits. The numbers of patients with available data are given in Figure 1. No patient was lost to follow-up of major clinical events, as vitality status and history of past and current diseases were obtained by direct contact with the patient or a relative. Photocopies of medical records covering the clinical events were obtained for classification (see later here). The maximum follow-up time was 7.8 years and the median follow-up time was 6.6 years. The diagnostic criteria have been described previously. 11 All events were independently coded according to defined rules 20 by two doctors who did not participate in the study and who had no knowledge of patient identity or group assignment. The occurrence of a new major Q/QS item (Minnesota code 21 1:1 or 1:2 in a 12 lead resting electrocardiogram read by two independent technicians) in patients without prior myocardial infarction and without other clinical signs of myocardial infarction was defined as a silent myocardial infarction, provided that two independent clinicians also agreed that the observed changes indicated myocardial damage. Statistical Analysis Power calculations before the study showed that 1200 patients had to be recruited in order to discover an intervention effect of 30% after 3 years on the combined end point of myocardial infarction, including sudden death, stroke, and other fatal and nonfatal cardiovascular diseases (time to first event). 11 In the usual care group an end point probability of 20% was assumed ( 0.05, 0.20). Because 1200 patients could not be recruited, but rather only 508 patients, the follow-up period was prolonged to 6 years. All data were analyzed according to the intentionto-treat principle. Mean in-trial data were calculated using all available data recorded during the biannual

4 AJH JANUARY 1998 VOL. 11, NO. 1, PART 1 MULTIPLE RISK INTERVENTION 17 TABLE 1. DRUG TREATMENT AT ENTRY, AFTER 3.3 YEARS, AND AT THE END OF THE STUDY At Entry After 3.3 Years After 6.6 Years Drug Usual Care (n 255) Intervention (n 253) Usual Care (n 227) Intervention (n 235) Usual Care (n 170) Intervention (n 192) Diuretics 124 (49) 116 (46) 73 (32) 72 (31) 62 (36) 69 (36) -Blocker 207 (81) 178 (70) 167 (74) 163 (69) 127 (75) 124 (65)* Calcium antagonist 35 (14) 36 (14) 71 (31) 57 (24) 72 (42) 77 (40) ACE inhibitor 11 (4) 9 (4) 41 (18) 39 (17) 34 (20) 43 (22) Other antihypertensive drugs 66 (26) 62 (25) 38 (17) 51 (22) 18 (11) 15 (8) Insulin treatment 5 (2) 13 (5) 5 (2) 15 (6)* 9 (5) 8 (4) Oral hypoglycemic drugs 34 (13) 19 (8)* 30 (13) 19 (8) 16 (9) 18 (9) Aspirin 8 (3) 10 (4) 21 (9) 22 (9) 41 (24) 45 (23) Statins (9) 3 (2) 34 (18) Nicotinic acid 0 2 (1) 0 2 (1) 0 0 Fibrate 0 2 (1) 2 (1) 20 (9) 5 (3) 19 (10)* Cholestyramine (2) 58 (25) 4 (2) 47 (25) Any type of lipid lowering drug 0 4 (2) 7 (3) 100 (43) 11 (6) 98 (51) * Values are numbers (%). *P.05; P.01; P.001 in comparison between the groups. ACE, angiotensin converting enzyme. follow-up visits. For each of these variables the mean change from baseline was calculated as the mean intrial value minus the baseline value. Unpaired Student s t test with 95% confidence interval and Fisher s exact test were used to compare continuous and categorical variables, respectively. Kaplan-Meier survival curves were calculated and treatment group differences were assessed by the log rank test. Relative risk and 95% confidence interval were calculated with the Cox proportional hazard model. Only two-sided tests were performed and P.05 was regarded as statistically significant. RESULTS Altogether, 105 patients died during follow-up, 64 patients (25.1%) in the usual care group and 41 patients (16.2%) in the intervention group (Table 2). The Kaplan-Meier probability of survival was 83% in the intervention group and 68% in the usual care group (Figure 2; P.016). The intervention program was associated with a significant 44% reduction in the risk of cardiovascular death (P.021 [Cox regression]; Table 2). In comparison with the usual care group, fatal and nonfatal stroke occurred less often in the intervention group (Table 3). Although there was a trend toward fewer fatal and nonfatal coronary events in the intervention group, the difference was not statistically significant. However, the overall risk for fatal and nonfatal cardiovascular events (time to first event) was 29% lower in the intervention group as compared with that in the usual care group (P.041; Table 3). The aim was to achieve similar blood pressure in the two groups, and there was only a 1 mm difference in diastolic blood pressure between the groups (which was lower in the intervention group) during follow-up as shown in Table 4. There was a net increase in heart rate in the usual care group (Table 4). The mean in-trial serum concentrations of total cholesterol, LDL cholesterol, and triglycerides were lower in the intervention than in the usual care group (P.0001, P.0001, and P.007, respectively; see Table 4). The intervention program was associated with net reductions in serum levels of total cholesterol (6.3%), LDL cholesterol (9.1%), and triglycerides (7.5%; Table 4). Patients defined as hypercholesterolemic at entry showed net reductions of 8.8% in total cholesterol and 11.1% in LDL cholesterol. Mean in-trial total cholesterol below 6.0 mmol/l was found in 114 (46%) patients in the intervention group and in 78 (31%) patients in the usual care group (P.0007). Among the patients who were enrolled as smokers (n 148), the total mortality rate was 30.4% compared with 16.7% in nonsmokers (P.0007). More smokers died in the usual care group (n 26; 35.6%) than in the intervention group (n 19; 25.3%), thereby making it

5 18 FAGERBERG ET AL AJH JANUARY 1998 VOL. 11, NO. 1, PART 1 FIGURE 1. Diastolic blood pressures (upper panel) and serum cholesterol concentrations (lower panel) at each biannual visit during the trial in the usual care and intervention groups, respectively. The numbers of patients with available data are given at the bottom of each panel. difficult of compare in-trial quit rate between the groups. After 3 years, 21 (28%) initial smokers in the intervention group and 8 (11%) such men in the usual care group were alive and did not smoke (P.012). At the end of the study, 160 patients were defined as nonsmokers in the intervention group and 135 patients in the usual care group (P NS). Mean values and changes in serum HbA 1c did not differ between the groups. Six patients (11%) in the intervention group and 3 patients (5%) in the usual care group had a mean in-trial HbA 1c below 6% (P NS). The reductions in mean in-trial blood glucose concentration and body weight were significantly greater in the intervention group in comparison with the usual care group (Table 4). DISCUSSION The present results demonstrate the high mortality in a population of older, treated hypertensive patients with additional risk factors and a relatively high frequency of overt cardiovascular disease before entering the study. Thus, the annual total mortality rate was 3.7% compared with rates from 0.2% to 0.7% in previous multiple risk intervention studies of middle aged men. 4 9 The most important conclusion of the study is that the gloomy prognosis may be improved by a dedicated risk factor intervention program. This may be achieved without adversely affecting quality of life, as evaluated during the first 3 years of the study by using a validated questionnaire, as previously reported. 22 We believe, although it is hard to prove scientifically, that the preventive effect is due to the entire multiple risk approach, including an education in a healthier life style. When interpreting these results, possible confounding factors should be considered. In any intervention study they might relate to differences in baseline characteristics between the groups, or to differences in management outside those defined in the intervention program. However, the groups were well matched with respect to risk factor levels at entry; aspirin was used to the same degree in both groups; and drug treatment for hypertension was comparable. The observation that -blockers were given marginally less often in the intervention group would not likely explain the observed difference in mortality, as such treatment may reduce mortality. 16 There was no difference in the systolic blood pressure, whereas the diastolic blood pressure was on average 1 mm lower in the intervention group. Furthermore, the follow-up procedures for the two groups were comparable, as all patients were seen biannually at the same unit and by the same staff. The recording of fatal events was complete and the patients adhered to the follow-up program to a similar degree in the two groups. However, an overall beneficial effect of the special care such as that seen in the intervention group, resulting in an improved prognosis, cannot be excluded. Although not addressed in the design of this study, it may be questioned whether the dysplipidemia among our patients observed at entry only was an expression of the clustering of risk factors seen in hypertension or whether it also was an effect of ongoing antihypertensive treatment consisting mainly of 1 -selective -blockade or thiazides. 23 Most of the patients were recruited from the Gothenburg Primary Preventive Trial and had been treated mainly with -blockers and thiazides for 10 years or more before entering the present study. 11 Follow-up data after 5 to 10 years of treatment in the former trial showed that

6 AJH JANUARY 1998 VOL. 11, NO. 1, PART 1 MULTIPLE RISK INTERVENTION 19 TABLE 2. MORTALITY AND CAUSES OF DEATH No (%) of Patients Cause of Death Usual Care (n 255) Intervention (n 253) Relative Risk (95% CI) Myocardial infarction 14 9 Sudden death 9 8 All coronary deaths 23 (9.0) 17 (6.7) 0.72 ( ) Stroke 7 3 Heart failure* 7 0 Aortic aneurysm 1 3 Pulmonary embolus 1 0 Other cardiovascular deaths 3 1 All cardiovascular deaths 42 (16.5) 24 (9.5) 0.56 ( ) Cancer Suicide 1 1 Trauma 1 0 Other noncardiovascular deaths 5 4 All noncardiovascular deaths 22 (8.6) 17 (6.7) 0.74 ( ) All deaths 64 (25.1) 41 (16.2) 0.62 ( ) CI, confidence interval. Relative risk, calculated by Cox regression analysis. * Three of the seven patients with fatal heart failure had suffered from a previous myocardial infarction. serum cholesterol actually decreased not only in the entire group but also among patients with single treatment with -blockade or thiazides. 24 Thus, it seems unlikely that previous antihypertensive drug treatment caused the elevated serum cholesterol level that was one inclusion criterion in the present study. Our intervention program was directed towards elevated serum lipid levels, smoking, and poor metabolic control in diabetic patients. The nonpharmacological part of the intervention program to reduce the concentration of serum cholesterol was successful in FIGURE 2. Kaplan-Meier curve for total mortality. Number of patients at risk at the beginning of each year is given for the intervention and usual care groups, respectively. comparison with previous multifactorial trials in unselected populations as indicated by results of the first year of treatment, that was mainly based on changes in life style. 5 9,25 The reduction in total cholesterol was obtained from dietary changes in up to half of the patients. This was associated with a small but consistent net weight loss as evidence of a lasting change in life style. In comparison to usual care management, the intervention program achieved an overall 6.3% reduction in total cholesterol and a net decrease of 8.8% among those who were hypercholesterolemic at entry. Irrespective of treatment method, most multifactorial trials have showed net reductions in total cholesterol of only 0.2% to 2%. 5 8 In the Oslo study, 4 showing a favorable outcome on the incidence of coronary heart disease, total cholesterol was lowered by 13%, whereas with the comprehensive intervention in the Helsiniki Businessmen Study, 9 in which cholesterol was lowered by 6.3%, an unexpected increase in morbidity was observed. Cotinine measurements revealed that 15% of the current tobacco users in the present study claimed to be nonsmokers, a result that accords well with previously published data. 17 About 30% of the patients were defined as smokers at entry. There was a higher mortality among these smokers in the usual care group than in the intervention group, which complicated the comparison of quit rates between the two study groups. However, the difference between the intervention and usual care group in the quit rate was 17% after 3 years in the present study, which com-

7 20 FAGERBERG ET AL AJH JANUARY 1998 VOL. 11, NO. 1, PART 1 TABLE 3. ALL FIRST CARDIOVASCULAR EVENTS DURING FOLLOW-UP No (%) of Patients Event Usual Care (n 255) Intervention (n 253) Relative Risk (95% CI) Major coronary Sudden death 7 6 Fatal MI 10 7 Nonfatal MI Silent MI* 8 9 Cumulative coronary events 50 (19.6) 44 (17.4) 0.86 ( ) Stroke Fatal 4 3 Nonfatal Cumulative stroke events 29 (11.4) 16 (6.3) 0.53 ( ) Other fatal cardiovascular events 5 3 All cumulative nonfatal and fatal cardiovascular events 84 (32.9) 63 (24.9) 0.71 ( ) CI, confidence interval. Relative risk, calculated by Cox regression analysis. * Only counted for patients without previous myocardial infarction. Cardiovascular events include sudden death, fatal and nonfatal myocardial infarction (MI), silent myocardial infarction, fatal and nonfatal stroke, and other cardiovascular deaths. Only one event (the first) is counted in each patient. pares favorably with the results in most other multifactorial trials with quit rates of 1.9% to 8%. 4,5,6,8 In the Multiple Risk Factor Intervention Trial (MRFIT) there was a similar quit rate as in the present study. 7 Among the patients with diabetes mellitus the intervention program did not improve glucose metabolism in comparison with usual care management (see HbA1 c levels). The observation of a small net reduction in blood glucose concentrations in the whole intervention group may be an effect of weight reduction. These findings, together with the net reduction in serum triglycerides, indicate a benefical effect on the insulin resistance that is characteristic of hypertensive populations. 23 Results from a recently published substudy of the present study indicate that the intervention program also was associated with a reduced tendency of LDL cholesterol to form complexes with arterial proteoglycans, a mechanism that may be important in modifying atherogenesis. 26,27 Such changes in lipoprotein particles may also be related to insulin resistance. 26,27 In the latter respect we also observed an improved glucose tolerance test in the intervention group. 26 Thus, the present results suggest that the combination of lipid lowering measures, smoking cessation, and a modestly improved glucose metabolism has a more powerful, synergistic effect in this category of hypertensive patients at high cardiovascular risk than what may be anticipated by just adding the changes in risk observed after single risk factor interventions. 3,28 A similar observation of a discrepancy between expected and observed effects relates to the incidence of cerebrovascular diseases; hypercholesterolemia is regarded as a weak and inconsistent risk factor for ischemic stroke 29 and, in spite of this, the incidence of fatal and nonfatal ischemic stroke decreased markedly in the present study (in which lipid lowering was a very important component of the treatment program). Recent cholesterol lowering trials using statins have also shown unexpected reductions in the incidence of stroke. 30,31 Our interpretation of the results is that the observed reductions in mortality and morbidity may be attributed to two distinguishing aspects of the present study. First, the patients included were at much higher cardiovascular risk than were those in previous multifactorial trials. 4 9 In the other trials 4 9 the patients were on average 15 to 20 years younger and were usually symptom-free, and the proportions of hypertensive participants were much lower. In the present study, patients with preexisting cardiovascular disease as well as patients with diabetes mellitus were accepted, and all were hypertensive. This is important, as preventive measures are in general much more effective in patients at high than at low cardiovascular risk. 30,31 Second, as discussed above, we used a comprehensive intervention program that improved not only lipid levels but also glucose metabolism and smoking cessation rates. The study design does not allow an analysis that would clarify the relative importance of each separate treatment aspect in the intervention program. Most of our patients were recruited after systematic screening of two hypertensive cohorts that were originally representative population samples of middle-

8 AJH JANUARY 1998 VOL. 11, NO. 1, PART 1 MULTIPLE RISK INTERVENTION 21 TABLE 4. MEAN IN-TRIAL VALUES AND CHANGES FROM BASELINE IN FACTORS RELATED TO CARDIOVASCULAR DISEASE Usual Care (n 252) Intervention (n 248) Difference Between Groups in Changes from Baseline Mean In-Trial (SD) Change From Baseline, Mean (95% CI) Mean In-Trial (SD) Change from Baseline, Mean (95% CI) Mean (SD) 95% CI Body weight (kg) 81.5 (11.9) 0.6 ( 1.1 to 0.2) 81.3 (14.0) 2.2 ( 2.7 to 1.6) to 0.8 Blood pressure (mm Hg) Systolic 156 (15) 1 ( 1.6 to 2.6) 154 (13) 1 ( 3.0 to 0.8) to 1.2 Diastolic 88 (6) 3 ( 3.5 to 1.6) 87 (6) 4 ( 4.8 to 3.2) to 0.2 Heart rate (beats/min) 65 (9) 2 (0.6 to 3.0) 64 (9) 0 ( 1.0 to 1.0) to 0.2 Cholesterol Total (mmol/l) 6.4 (1.0) 0.2 ( 0.3 to 0.1) 6.0 (0.9) 0.7 ( 0.8 to 0.6) to 0.3 (mg/dl) ( 11 to 5) ( 30 to 22) to 13 LDL (mmol/l) 4.4 (1.0) 0.2 ( 0.3 to 0.1) 4.0 (0.9) 0.7 ( 0.8 to 0.6) to 0.3 (mg/dl) ( 11 to 5) ( 29 to 21) to 12 HDL (mmol/l) 1.16 (0.33) 0.08 ( 0.11 to 0.05) 1.20 (0.33) 0.06 ( 0.09 to 0.03) to 0.06 (mg/dl) ( 4.2 to 1.9) ( 3.6 to 1.2) to 2.3 Triglycerides (mmol/l) 2.17 (1.17) 0.24 (0.13 to 0.35) 1.93 (0.77) 0.02 ( 0.08 to 0.11) to 0.08 (mg/dl) (12 to 31) ( 7 to10) to 7 HbA 1c (%) 7.7 (1.2) 0.2 ( 0.1 to 0.5) 7.7 (1.5) 0.1 ( 0.3 to 0.4) to 0.4 Blood glucose (mmol/l) 5.8 (2.0) 0.0 ( 0.10 to 0.19) 5.6 (2.2) 0.2 ( 0.37 to 0.01) to 0.01 LDL, Low density lipoprotein; HDL, high density lipoprotein; CI, confidence interval. aged men. 10,11,16 They represent a high risk population of men and our results are therefore only applicable to such patients. The main conclusion from this and other studies is that only reducing the blood pressure in patients with hypertension who have other cardiovascular risk factors is not enough, as the risk remains high in many patients even if blood pressure control is acceptable. The results from the present study corroborate earlier data on the very high cardiovascular risk in elderly hypertensive patients with additional risk factors, and give support to international guidelines recommending a multifactorial treatment approach in hypertensive patients. The summary is that the gloomy prognosis may be affected favorably by such an intervention program. ACKNOWLEDGMENTS We are indepted to the participants in the study for their trust and cooperation; to the staff at the Hypertension Unit for their excellent care (Ann-Louise Eriksson, Maija-Lisa Holmasaari, Agneta Hallén, Tuija Kaikkonen); to Lars Wilhelmsen, M.D. and Dag Elmfeldt, MD, for help with patient recruitment: to Göran Lindstedt, MD, for help with determination of cotinine concentrations; to Anders Odén, PhD, for statistical assistance; to Stig Attvall, MD, and Bert Andersson, MD, for end point classification; to Caroline Schmidt, Anna Frödén, and Christina Claesson for electrocardiogram reading and data processing; and to Madis Suurküla, MD and Milan Lomsky, MD, for final reading of the electrocardiograms. REFERENCES 1. Wilhelmsen L, Wedel H, Tibblin G: Multivariate analysis of risk factors for coronary heart disease. Circulation 1973;48: Kannel WB, Sorlie P: Hypertension in Framingham, in Paul O (ed): Epidemiology and Control of Hypertension. Georg Thieme Verlag, Stuttgart, 1975, pp Zanchetti A, Chalmers JP, Arakawa K, et al: The 1993 guidelines for the management of mild hypertension: memorandum from a WHO/ISH meeting. Blood Pressure 1993:2: Hjermann I, Velve Byre K, Holme I, Leren P: Effect of diet and smoking intervention on the incidence of coronary heart disease. Lancet 1981;ii: Kornitzer M, DeBacker G, Dramaix M, et al: Belgian

9 22 FAGERBERG ET AL AJH JANUARY 1998 VOL. 11, NO. 1, PART 1 heart disease prevention project: incidence and mortality results. Lancet 1983;i: World Health Organization European Collaborative Group: European collaborative trial of multifactorial prevention of coronary heart disease: final report on the 6-year-results. Lancet 1986;i: Multiple Risk Factor Intervention Trial Research Group: Multiple risk factor intervention trial. Risk factor changes and mortality results. JAMA 1982: 248: Wilhelmsen L, Berglund G, Elmfeldt D, et al: The multifactorial primary prevention trial in Gothenburg, Sweden. Eur Heart J 1986;7: Strandberg T, Salomaa VV, Naukkarinen VA, et al: Long-term mortality after 5-year multifactorial primary prevention of cardiovascular diseases in middle-aged men. JAMA 1991;266: Andersson O, Berglund G, Hansson L et al: Organization and efficacy of an out-patient hypertension clinic. Acta Med Scand 1978;203: Samuelsson O: Hypertension in middle-aged men: management, morbidity and prognostic factors during long-term hypertensive care. Acta Med Scand 1985; 702(suppl): Samuelsson O, Wilhelmsen L, Andersson OK, et al: Cardiovascular morbidity in relation to change in blood pressure and serum cholesterol levels in treated hypertension. JAMA 1987;258: Agewall S, Fagerberg B, Samuelsson O, et al. Multiple cardiovascular risk factor intervention in treated hypertensive men: what can be achieved. Nutr Metab Cardiovasc Dis 1993;3: Agewall S, Wikstrand J, Samuelsson O, et al: The efficacy of multiple risk factor intervention in treated hypertensive men during long-term follow up. J Intern Med 1994;236: Klose S, Borner K: Enzymatische bestimmung des gesamtcholesterins mit dem Greiner Selective Analyzer (GSA-II). J Clin Chem Clin Biochem 1978;15: Wikstrand J, Warnold I, Olsson G, et al: Primary prevention with metoprolol in patients with hypertension: mortality results from the MAPHY Study. JAMA 1988; 259: Jarvis MJ, Tunstall-Pedoe H, Feyerabend C, et al: Comparison of tests used to distinguish smokers from nonsmokers. Am J Publ Health 1987;77: National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults: Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Arch Intern Med 1988;148: Drug Information Committee: Treatment of hyperlipidemia. Uppsala, Sweden: National Board of Health and Welfare, Drug Information Committee, Keys A (ed): Coronary heart disease in seven countries. Circulation 1970;41(suppl I):I-1 I Rose GA, Blackburn H: Classification of the electrocardiogram for population studies, in Cardiovascular Survey Methods. World Health Organization Monograph Series No. 56. World Health Organization, Geneva, 1968, pp Agewall S, Wikstrand J, Dahlöf C, Fagerberg B: A randomized study of quality of life during multiple risk factor intervention in treated hypertensive men at high cardiovascular risk. J Hypertens 1995;13: Reaven GM, Lithell H, Landsberg L. Hypertension and associated abnormalities the role of insulin resistance and the sympathoadrenal system. N Engl J Med 1996; 334: Samuelsson O, Hedner T, Berglund G, et al: Diabetes mellitus in treated hypertension: incidence, predictive factors and the impact of non-selective beta-blockers and thiazide diuretics during 15 years treatment of middle-aged hypertensive men in the Primary Prevention Trial Göteborg, Sweden. J Human Hypertens 1994; 8: Ramsay LE, Yeo WW, Jackson PR: Dietary reduction of serum cholesterol concentration: time to think again. Br Med J 1991;303: Fagerberg B, Wiklund O, Agewall S, et al: Multifactorial treatment of hypertensive men at high cardiovascular risk and low-density lipoprotein cholesterol affinity to human arterial proteoglycans. Eur J Clin Invest 1996;26: Williams KJ, Tabas I. The response-to-retention hypothesis of early atherosclerosis. Arterioscl Thromb 1995;15: Gould AL, Rossouw JE, Santanello NC, et al: Cholesterol reduction yields clinical benefit. A new look at old data. Circulation 1995;91: You RS, Donnan GA: Serum risk factors for stroke. Cardiovasc Risk Fact 1995;5: The Scandinavian Simvastatin Survival Study group: Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344: Sacks FM, Pfeffer MA, Moye LA, et al: The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med 1996;335: