- Email: [email protected]
What does STOP-2 tell us about management of hypertension?
CORRESPONDENCE
COMMENTARY
CORRESPONDENCE
What does STOP-2 tell us about management of hypertension? Sir—Lennart Hansson and colleagues (Nov 20, p 1751), 1 reporting the Swedish Trial of Old Patients with Hypertension (STOP-2), state that “old and new antihypertensive drugs were similar in prevention of cardiovascular disease” and that “older and newer antihypertensive drugs are equally useful”. Since STOP-2 was not designed to compare individual drugs, some important differences may have been concealed. At the time of the study’s design and preparation, -blockers as part of a conventional strategy could have been considered adequate. However, the only trial to have permitted direct comparison of a -blocker with a diuretic2 and a reanalysis of older trials according to the primary treatment strategy in the active group3 have demonstrated the superiority of lowdose diuretics over -blockers in preventing coronary heart disease. The controls in STOP-2 received one of three -blockers or a diuretic as the first option, but we are not told how many patients were initially treated with one of them. Moreover, pindolol, a drug implicated in coronary risk4 and not merely the absence of a protective effect, was one of the -blockers used. Diuretics were in bad company in this trial, and the physicians should wait for the results of trials designed to compare drugs and not strategies before concluding that ability to lower blood pressure is sufficient to ascribe the property “antihypertensive” to any drug. Flávio Danni Fuchs Serviço de Cardiologia, Hospital de Clínicas de Porto Alegre, Ramiro Barcelos, Porto Alegre, RS, Brazil (e-mail: [email protected]) 1
2
Hansson L, Lindholm LH, Ekbom T, et al, for the STOP-Hypertension-2 Study Group. Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality and morbidity: the Swedish Trial in Old Patients with Hypertension-2 study. Lancet 1999; 354: 1751–56. MRC Working Party. Medical Research Council Trial of Treatment of Hypertension in Older Adults. BMJ 1992; 304: 405–12.
THE LANCET • Vol 355 • 19 February, 2000
3
4
Psaty BM, Smith NL, Siscovick DS, et al. Health outcomes associated with antihypertensive therapies used as first-line agents. JAMA 1997; 277: 739–45. Strandberg TE, Salomaa UV, Naukkarinen VA, et al. Cardiovascular morbidity and multifactorial primary prevention: fifteen-year follow-up of the Helsinki Businessmen Study. Nutr Metab Cardiovasc Dis 1995; 5: 7–15.
Sir—STOP-21 was randomised but both patients and doctors were aware of the treatment administered. 2 Information on adverse events was collected by asking patients whether or not new symptoms had occurred since their last visit. This approach probably invalidates the assessment of adverse events presented in table 4.1 The 25% ankle oedema in the calciumantagonist group and the 30% cough in the angiotensin-converting enzyme (ACE) inhibitor group, both much higher than frequencies in blinded studies,4 exemplify how expectation of investigators and selective reporting by patients may colour the outcome. More importantly, the open design may have biased the reported incidences of non-fatal events, the diagnosis of which rests on symptoms or clinical signs open to interpretation, such as congestive heart failure or non-Q-wave infarction. The design2 does require all events to be ascertained by a blinded committee, but that does not rule out the possibility that prior knowledge of the treatment allocation results in selective overreporting or underreporting of events, especially in an era in which calcium antagonists have been subjected to controversy4 while ACE inhibitors were proved to be of benefit in patients with overt heart failure. Another obstacle to interpreting STOP-2 is the combination of randomised treatment with second-line antihypertensive drugs. A large proportion of the patients allocated to the newer drugs also received or crossed over to conventional treatment, because -blockers were associated with calcium antagonists and thiazides with ACE inhibitors. At the last visit, 46% of patients were on more than one antihypertensive drug. Figure 5
contrasts ACE inhibitors partly combined with diuretics, with calcium antagonists partly combined with -blockers—not simply ACE inhibitors with calcium antagonists, as the legend suggests. Moreover, only 61–66% of patients were still taking their randomised treatment at the last visit, which may have further obscured true differences beween first-line drugs. The research question which STOP-2 set out to resolve will remain unsettled until the results of properly designed double-blind trials such as ALLHAT5 become available. Ji G Wang, Jerzy Gasowski, Robert Fagard, *Jan A Staessen Study Co-ordinating Centre, Laboratory of Hypertension, Campus Gasthuisberg, University of Leuven, Herestraat 49, B-3000 Leuven, Belgium (e-mail: [email protected]) 1
2
3
4
5
Hansson L, Lindholm LH, Ekbom T, et al, for the STOP-Hypertension-2 Study Group. Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality and morbidity: the Swedish Trial in Old Patients with Hypertension-2 study. Lancet 1999; 354: 1751–56. Hansson L, Hedner T, Dahlof B. Prospective randomized open blinded end-point (PROBE) study: a novel design for intervention trials. Blood Press 1992; 1: 113–19. Johnson BF, Eisner GM, McMahon FG, Jain AK, Rudd P, Sowers JR. A multicenter comparison of adverse reaction profiles of isradipine and enalapril at equipotent doses in patients with essential hypertension. J Clin Pharmacol 1995; 35: 484–92. McMurray J. Calcium-antagonist controversy: the long and short of it? Lancet 1997; 349: 585–86. Elliott WJ, Davis BR, Cutler JA, et al. Rationale and design for the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Am J Hypertens 1996; 9: 342–60.
Sir—STOP-21 was designed to detect a 25% difference in cardiovascular mortality between patient groups receiving conventional or newer antihypertensive drugs. The study was reported as negative but we do not believe it is possible to conclude that there was a similar effect for the two treatments on the primary combined endpoint, fatal cardiovascular events.
651
CORRESPONDENCE
A difference of less than 25% could still exist. One problem could be the heterogeneity of the treatment populations. Patients with both systolic-diastolic hypertension and isolated systolic hypertension were included but while the groups were relatively homogeneous in terms of systolic pressure, the large disparity in diastolic pressure could confound the findings. At the end of the study, some hypertensive patients could have had a normal or raised diastolic blood pressure (if they had previously had systolic-diastolic hypertension) but others could have experienced a large decrease in diastolic pressure if they previously had isolated systolic hypertension. The SHEP study 2 reported, in patients with isolated systolic hypertension, that the lowering of systolic blood pressure was associated with a diminished cardiovascular risk whereas those who experienced the greatest reduction in diastolic pressure had a much greater risk. Analysis of STOP-2 by Cox regression with adjustment for diastolic, not systolic, pressure at inclusion, might be inconclusive. The major factor characterising STOP-2 patients was the greater pulse pressure at the start of the study (close to 100 mm Hg). At the end of the period, there was a modest degree of systolic hypertension in most treated patients, whatever drug or drug combination they were on. A haemodynamic pattern involving a pulse pressure on treatment of almost 70 mm Hg is a marker for increased cardiovascular risk3 and could at least partly explain, together with the higher prevalence of diabetes mellitus and myocardial infarction (as suggested by the authors), the higher frequency of endpoints in STOP-Hypertension-2 compared with STOP-Hypertension. As also discussed by the authors, conventional antihypertensive drugs may not be best suited to achieving diastolic/systolic pressures <90/<140 mm Hg. A similar finding was observed in the HOT study, where normalisation of diastolic pressure, by more aggressive therapy, was not associated with a similar normalisation of systolic and pulse pressures.4 Drugs to be used for elderly patients with isolated systolic hypertension should primarily reduce systolic pressure without decreasing diastolic pressure, thus promoting a normal pulse pressure and reversing mechanisms associated with increased aortic rigidity. Apparently, neither conventional nor the newer agents achieve this goal. This could be explained by inadequate doses;
652
another possibility is the development of drugs acting predominantly on large-artery walls.5 *Michel E Safar, Jacques Blacher, Jean-Jacques Mourad, Gérard London, Edward D Frohlich *Department of Internal Medicine, Hôpital Broussais, 75014 Paris, France; Hôpital Saint Michel, Paris; and Alton Ochsner Medical Foundation, New Orleans, LA, USA (e-mail: [email protected]) 1
2
3
4
5
Hansson L, Lindholm LH, Ekbom T, et al, for the STOP-Hypertension-2 Study Group. Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality and morbidity: the Swedish Trial in Old Patients with Hypertension-2 study. Lancet 1999; 354: 1751–56. Somes GW, Pahor M, Shorr RI, Cushman WC, Applegate WB. The role of diastolic blood pressure when treating isolated systolic hypertension. Arch Intern Med 1999; 159: 2004–09. Franklin SS, Khan SA, Wong ND, Larson MG, Levy D. Is pulse pressure useful in predicting risk for coronary heart disease? The Framingham Heart Study. Circulation 1999; 100: 354–60. Safar ME, Blacher J, Staessen JA. Hypertension Optimal Treatment (HOT) trial. Lancet 1998; 352: 573. O’Rourke M, Frohlich ED. Pulse pressure: is this a clinically useful risk factor? Hypertension 1999; 34: 372–74.
Sir—No placebo group was included in STOP-2,1 but in this open label trial, where the patients were randomly assigned to one of the three therapeutic options and eight different drugs were used, no definitive answer about the usefulness of a specific antihypertensive drug class can be given. -blockers or diuretics were given as additional drugs in the calcium antagonist and the ACE-inhibitor groups, and this makes a comparison between conventional and newer drugs impossible. A direct comparison is only allowed for the group treated with ACE inhibitors plus diuretics vs those on -blockers plus diuretics, since there is no overlap in intention to treat. Of special interest are the baseline data on the diabetic patients. Despite higher prevalences of previous myocardial infarction (17 vs 7 vs 6), stroke (18 vs 10 vs 8), and atrial fibrillation (22 vs 10 vs 13) in the conventional than in the ACE inhibitor and in the calcium antagonist groups, respectively, the primary endpoints did not differ significantly in the three groups. This randomisation bias could even count as further evidence for the proven effect of -blockers and diuretics. It is surprising that a blood-pressure decrease of over 34/16 mm Hg could be achieved by a combination of only two drugs. In the HOT2 and CAPPP3 trials, where thrice-daily regimens were
allowed, blood pressure fell by 28/22 mm Hg and 13/11 mm Hg, respectively, on average. By contrast, the decrease in blood pressure achieved in HOPE4 was only 2/3 mm Hg with 10 mg ramipril. *Andrea Siebenhofer, Robert Zweiker, Gernot A Brunner, Peter Mrak, Thomas R Pieber Department of Internal Medicine, Karl Franzens University of Graz, 8036 Graz, Austria (e-mail: [email protected]) 1
2
3
4
Hansson L, Lindholm LH, Ekbom T, et al, for the STOP-Hypertension-2 Study Group. Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality and morbidity: the Swedish Trial in Old Patients with Hypertension-2 study. Lancet 1999; 354: 1751–56. Hansson L, Zanchetti A, Carruthers SG, et al, for the HOT Study Group. 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. Hansson L, Lindholm LH, Niskanen L, et al. Effect of angiotensin-converting enzyme inhibition compared with conventional therapy on cardiovascular morbidity and mortality in hypertension: the Captopril Prevention Project (CAPPP) randomised trial. Lancet 1999; 353: 611–16. The Heart Outcomes Prevention Evaluation Study Investigators. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on death from cardiovascular causes, myocardial infarction, and stroke in high-risk patients. N Engl J Med 2000; 342: 145–53.
Sir—STOP-21 shows for the first time that the efficacy of dihydropyridine calcium antagonists is similar to that of -blockers, diuretics, and ACE inhibitors in preventing cardiovascular events. Though the differences were small and not significant, the lowest total mortality, cardiovascular mortality, and stroke rates were associated with felodipine and isradipine. However, in view of the controversy over the safety of calcium antagonists, we find the significantly higher myocardial infarction rate compared with that seen for ACE inhibitors in STOP-2 disturbing. One possible explanation could be the dose frequency (once daily) for isradipine.2 Since there is a circadian variation in the onset of acute cardiovascular diseases it would be interesting to know whether these myocardial infarctions occurred predominantly in the early morning when no protection could be expected from short-acting calcium channel blockers.3 The antihypertensive effect of felodipine is associated with persistent stimulation of the sympathetic nervous system.4 Were there any differences in heart rate between the treatment groups? In view
THE LANCET • Vol 355 • 19 February, 2000
COMMENTARY
CORRESPONDENCE
What does STOP-2 tell us about management of hypertension? Sir—Lennart Hansson and colleagues (Nov 20, p 1751), 1 reporting the Swedish Trial of Old Patients with Hypertension (STOP-2), state that “old and new antihypertensive drugs were similar in prevention of cardiovascular disease” and that “older and newer antihypertensive drugs are equally useful”. Since STOP-2 was not designed to compare individual drugs, some important differences may have been concealed. At the time of the study’s design and preparation, -blockers as part of a conventional strategy could have been considered adequate. However, the only trial to have permitted direct comparison of a -blocker with a diuretic2 and a reanalysis of older trials according to the primary treatment strategy in the active group3 have demonstrated the superiority of lowdose diuretics over -blockers in preventing coronary heart disease. The controls in STOP-2 received one of three -blockers or a diuretic as the first option, but we are not told how many patients were initially treated with one of them. Moreover, pindolol, a drug implicated in coronary risk4 and not merely the absence of a protective effect, was one of the -blockers used. Diuretics were in bad company in this trial, and the physicians should wait for the results of trials designed to compare drugs and not strategies before concluding that ability to lower blood pressure is sufficient to ascribe the property “antihypertensive” to any drug. Flávio Danni Fuchs Serviço de Cardiologia, Hospital de Clínicas de Porto Alegre, Ramiro Barcelos, Porto Alegre, RS, Brazil (e-mail: [email protected]) 1
2
Hansson L, Lindholm LH, Ekbom T, et al, for the STOP-Hypertension-2 Study Group. Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality and morbidity: the Swedish Trial in Old Patients with Hypertension-2 study. Lancet 1999; 354: 1751–56. MRC Working Party. Medical Research Council Trial of Treatment of Hypertension in Older Adults. BMJ 1992; 304: 405–12.
THE LANCET • Vol 355 • 19 February, 2000
3
4
Psaty BM, Smith NL, Siscovick DS, et al. Health outcomes associated with antihypertensive therapies used as first-line agents. JAMA 1997; 277: 739–45. Strandberg TE, Salomaa UV, Naukkarinen VA, et al. Cardiovascular morbidity and multifactorial primary prevention: fifteen-year follow-up of the Helsinki Businessmen Study. Nutr Metab Cardiovasc Dis 1995; 5: 7–15.
Sir—STOP-21 was randomised but both patients and doctors were aware of the treatment administered. 2 Information on adverse events was collected by asking patients whether or not new symptoms had occurred since their last visit. This approach probably invalidates the assessment of adverse events presented in table 4.1 The 25% ankle oedema in the calciumantagonist group and the 30% cough in the angiotensin-converting enzyme (ACE) inhibitor group, both much higher than frequencies in blinded studies,4 exemplify how expectation of investigators and selective reporting by patients may colour the outcome. More importantly, the open design may have biased the reported incidences of non-fatal events, the diagnosis of which rests on symptoms or clinical signs open to interpretation, such as congestive heart failure or non-Q-wave infarction. The design2 does require all events to be ascertained by a blinded committee, but that does not rule out the possibility that prior knowledge of the treatment allocation results in selective overreporting or underreporting of events, especially in an era in which calcium antagonists have been subjected to controversy4 while ACE inhibitors were proved to be of benefit in patients with overt heart failure. Another obstacle to interpreting STOP-2 is the combination of randomised treatment with second-line antihypertensive drugs. A large proportion of the patients allocated to the newer drugs also received or crossed over to conventional treatment, because -blockers were associated with calcium antagonists and thiazides with ACE inhibitors. At the last visit, 46% of patients were on more than one antihypertensive drug. Figure 5
contrasts ACE inhibitors partly combined with diuretics, with calcium antagonists partly combined with -blockers—not simply ACE inhibitors with calcium antagonists, as the legend suggests. Moreover, only 61–66% of patients were still taking their randomised treatment at the last visit, which may have further obscured true differences beween first-line drugs. The research question which STOP-2 set out to resolve will remain unsettled until the results of properly designed double-blind trials such as ALLHAT5 become available. Ji G Wang, Jerzy Gasowski, Robert Fagard, *Jan A Staessen Study Co-ordinating Centre, Laboratory of Hypertension, Campus Gasthuisberg, University of Leuven, Herestraat 49, B-3000 Leuven, Belgium (e-mail: [email protected]) 1
2
3
4
5
Hansson L, Lindholm LH, Ekbom T, et al, for the STOP-Hypertension-2 Study Group. Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality and morbidity: the Swedish Trial in Old Patients with Hypertension-2 study. Lancet 1999; 354: 1751–56. Hansson L, Hedner T, Dahlof B. Prospective randomized open blinded end-point (PROBE) study: a novel design for intervention trials. Blood Press 1992; 1: 113–19. Johnson BF, Eisner GM, McMahon FG, Jain AK, Rudd P, Sowers JR. A multicenter comparison of adverse reaction profiles of isradipine and enalapril at equipotent doses in patients with essential hypertension. J Clin Pharmacol 1995; 35: 484–92. McMurray J. Calcium-antagonist controversy: the long and short of it? Lancet 1997; 349: 585–86. Elliott WJ, Davis BR, Cutler JA, et al. Rationale and design for the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Am J Hypertens 1996; 9: 342–60.
Sir—STOP-21 was designed to detect a 25% difference in cardiovascular mortality between patient groups receiving conventional or newer antihypertensive drugs. The study was reported as negative but we do not believe it is possible to conclude that there was a similar effect for the two treatments on the primary combined endpoint, fatal cardiovascular events.
651
CORRESPONDENCE
A difference of less than 25% could still exist. One problem could be the heterogeneity of the treatment populations. Patients with both systolic-diastolic hypertension and isolated systolic hypertension were included but while the groups were relatively homogeneous in terms of systolic pressure, the large disparity in diastolic pressure could confound the findings. At the end of the study, some hypertensive patients could have had a normal or raised diastolic blood pressure (if they had previously had systolic-diastolic hypertension) but others could have experienced a large decrease in diastolic pressure if they previously had isolated systolic hypertension. The SHEP study 2 reported, in patients with isolated systolic hypertension, that the lowering of systolic blood pressure was associated with a diminished cardiovascular risk whereas those who experienced the greatest reduction in diastolic pressure had a much greater risk. Analysis of STOP-2 by Cox regression with adjustment for diastolic, not systolic, pressure at inclusion, might be inconclusive. The major factor characterising STOP-2 patients was the greater pulse pressure at the start of the study (close to 100 mm Hg). At the end of the period, there was a modest degree of systolic hypertension in most treated patients, whatever drug or drug combination they were on. A haemodynamic pattern involving a pulse pressure on treatment of almost 70 mm Hg is a marker for increased cardiovascular risk3 and could at least partly explain, together with the higher prevalence of diabetes mellitus and myocardial infarction (as suggested by the authors), the higher frequency of endpoints in STOP-Hypertension-2 compared with STOP-Hypertension. As also discussed by the authors, conventional antihypertensive drugs may not be best suited to achieving diastolic/systolic pressures <90/<140 mm Hg. A similar finding was observed in the HOT study, where normalisation of diastolic pressure, by more aggressive therapy, was not associated with a similar normalisation of systolic and pulse pressures.4 Drugs to be used for elderly patients with isolated systolic hypertension should primarily reduce systolic pressure without decreasing diastolic pressure, thus promoting a normal pulse pressure and reversing mechanisms associated with increased aortic rigidity. Apparently, neither conventional nor the newer agents achieve this goal. This could be explained by inadequate doses;
652
another possibility is the development of drugs acting predominantly on large-artery walls.5 *Michel E Safar, Jacques Blacher, Jean-Jacques Mourad, Gérard London, Edward D Frohlich *Department of Internal Medicine, Hôpital Broussais, 75014 Paris, France; Hôpital Saint Michel, Paris; and Alton Ochsner Medical Foundation, New Orleans, LA, USA (e-mail: [email protected]) 1
2
3
4
5
Hansson L, Lindholm LH, Ekbom T, et al, for the STOP-Hypertension-2 Study Group. Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality and morbidity: the Swedish Trial in Old Patients with Hypertension-2 study. Lancet 1999; 354: 1751–56. Somes GW, Pahor M, Shorr RI, Cushman WC, Applegate WB. The role of diastolic blood pressure when treating isolated systolic hypertension. Arch Intern Med 1999; 159: 2004–09. Franklin SS, Khan SA, Wong ND, Larson MG, Levy D. Is pulse pressure useful in predicting risk for coronary heart disease? The Framingham Heart Study. Circulation 1999; 100: 354–60. Safar ME, Blacher J, Staessen JA. Hypertension Optimal Treatment (HOT) trial. Lancet 1998; 352: 573. O’Rourke M, Frohlich ED. Pulse pressure: is this a clinically useful risk factor? Hypertension 1999; 34: 372–74.
Sir—No placebo group was included in STOP-2,1 but in this open label trial, where the patients were randomly assigned to one of the three therapeutic options and eight different drugs were used, no definitive answer about the usefulness of a specific antihypertensive drug class can be given. -blockers or diuretics were given as additional drugs in the calcium antagonist and the ACE-inhibitor groups, and this makes a comparison between conventional and newer drugs impossible. A direct comparison is only allowed for the group treated with ACE inhibitors plus diuretics vs those on -blockers plus diuretics, since there is no overlap in intention to treat. Of special interest are the baseline data on the diabetic patients. Despite higher prevalences of previous myocardial infarction (17 vs 7 vs 6), stroke (18 vs 10 vs 8), and atrial fibrillation (22 vs 10 vs 13) in the conventional than in the ACE inhibitor and in the calcium antagonist groups, respectively, the primary endpoints did not differ significantly in the three groups. This randomisation bias could even count as further evidence for the proven effect of -blockers and diuretics. It is surprising that a blood-pressure decrease of over 34/16 mm Hg could be achieved by a combination of only two drugs. In the HOT2 and CAPPP3 trials, where thrice-daily regimens were
allowed, blood pressure fell by 28/22 mm Hg and 13/11 mm Hg, respectively, on average. By contrast, the decrease in blood pressure achieved in HOPE4 was only 2/3 mm Hg with 10 mg ramipril. *Andrea Siebenhofer, Robert Zweiker, Gernot A Brunner, Peter Mrak, Thomas R Pieber Department of Internal Medicine, Karl Franzens University of Graz, 8036 Graz, Austria (e-mail: [email protected]) 1
2
3
4
Hansson L, Lindholm LH, Ekbom T, et al, for the STOP-Hypertension-2 Study Group. Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality and morbidity: the Swedish Trial in Old Patients with Hypertension-2 study. Lancet 1999; 354: 1751–56. Hansson L, Zanchetti A, Carruthers SG, et al, for the HOT Study Group. 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. Hansson L, Lindholm LH, Niskanen L, et al. Effect of angiotensin-converting enzyme inhibition compared with conventional therapy on cardiovascular morbidity and mortality in hypertension: the Captopril Prevention Project (CAPPP) randomised trial. Lancet 1999; 353: 611–16. The Heart Outcomes Prevention Evaluation Study Investigators. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on death from cardiovascular causes, myocardial infarction, and stroke in high-risk patients. N Engl J Med 2000; 342: 145–53.
Sir—STOP-21 shows for the first time that the efficacy of dihydropyridine calcium antagonists is similar to that of -blockers, diuretics, and ACE inhibitors in preventing cardiovascular events. Though the differences were small and not significant, the lowest total mortality, cardiovascular mortality, and stroke rates were associated with felodipine and isradipine. However, in view of the controversy over the safety of calcium antagonists, we find the significantly higher myocardial infarction rate compared with that seen for ACE inhibitors in STOP-2 disturbing. One possible explanation could be the dose frequency (once daily) for isradipine.2 Since there is a circadian variation in the onset of acute cardiovascular diseases it would be interesting to know whether these myocardial infarctions occurred predominantly in the early morning when no protection could be expected from short-acting calcium channel blockers.3 The antihypertensive effect of felodipine is associated with persistent stimulation of the sympathetic nervous system.4 Were there any differences in heart rate between the treatment groups? In view
THE LANCET • Vol 355 • 19 February, 2000