- Email: [email protected]
Effect of losartan on sudden cardiac death in people with diabetes: data from the LIFE study
RESEARCH LETTERS
compared with nine (2%) of 500 in those without atrial fibrillation. The respective figures for the atenolol group were 14 (13%) of 105 and 16 (3%) of 504. Our results suggest losartan affords better protection against cardiac death from arrhythmias for patients with diabetes mellitus than does atenolol. Importantly, our analyses were exploratory and require confirmation.
investigation was supported in part by the Leukaemia Research Fund. Imperial College London and Hammersmith Hospital received reimbursement from Novartis Pharma for the costs incurred in entering patients in the trial. The sponsor had no role in study design, data collection, data interpretation, data analysis, or writing of the report. 1
2
3
4
5
Druker BJ, Talpaz M, Resta DJ, et al. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001; 344: 1031–37. Kantarjian H, Sawyers C, Hochhaus A, et al. Hematologic and cytogenetic responses to imatinib mesylate in chronic myelogenous leukemia. N Engl J Med 2002; 346: 645–52. Allan NC, Richards SM, Shepherd PC. UK Medical Research Council randomised, multicentre trial of interferon-alfa n1 for chronic myeloid leukaemia: improved survival irrespective of cytogenetic response. The UK Medical Research Council’s Working Parties for Therapeutic Trials in Adult Leukaemia. Lancet 1995; 345: 1392–97. Shah N, Nicoll J, Nagar B, et al. Multiple BCR-ABL kinase domain mutations confer polyclonal resistance to the tyrosine kinase inhibitor imatinib (STI571) in chronic phase and blast crisis chronic myeloid leukemia. Cancer Cell 2002; 2: 117–25. O’Brien SG, Guilhot F, Larson RA, et al. Imatinib compared with interferon and low dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med 2003; 348: 994–1004.
Lancet 2003; 362: 619–20 See Commentary page 591
Prevention of sudden cardiac death in patients with coronary heart disease is an important goal for treatment. In placebo-controlled long-term studies, several blockers, including atenolol, have shown consistent reductions in postinfarction mortality, and up to 50% reduction in risk of sudden cardiac death.1 In patients with diabetes mellitus, especially if associated with autonomic neuropathy, a rise in heart rate and fall in heart rate variability have been suggested to be important risk factors for sudden cardiac death and increased all-cause mortality.2 Moreover, blockade has been suggested to be more effective than angiotensin-converting enzyme inhibition in prevention of sudden cardiac death in a model of hypertension and left ventricular hypertrophy.3 blockade inhibits the proarrhythmic effects of both neural and humoral sympathetic stimulation and inhibits the vagal withdrawal that accompanies ischaemia.4 In a prespecified subgroup of patients with diabetes with left ventricular hypertrophy, workers on the Losartan Intervention For Endpoint reduction in hypertension (LIFE) study reported a major reduction of all-cause mortality—especially cardiovascular mortality—with a losartan-based versus an atenolol-based antihypertensive treatment.5 Against this background, we postulated post hoc that the losartan-based treatment might have a better effect on sudden cardiac death than atenolol-based therapy, and we aimed to test this hypothesis. As part of the main LIFE study, patients with diabetes, hypertension, and electrocardiogram-documented left ventricular hypertrophy were assigned once-daily losartanbased or atenolol-based antihypertensive therapy.5 In both groups, about 85% received a diuretic as supplementary treatment. Here, we used Cox regression analysis with baseline Framingham risk score and electrocardiogram-left ventricular hypertrophy as covariates to compare effects of losartan and atenolol on cardiovascular morbidity and mortality (defined as cardiovascular death, stroke, or myocardial infarction). An independent end-point committee, masked to the type of treatment the patients received, classified all endpoints including sudden cardiac death. We studied 1195 men and women (53% women) with diabetes, with a mean age of 67 years (SD 7). Average blood pressure after placebo run-in was 177/96 mm Hg. Patients were followed up for at least 4 years (mean 4·7 years [SD 1·1]).
Department of Haematology, Hammersmith Hospital, London W12 0NN, UK (D Marin MD, S Marktel MD, R Szydlo PhD, M Bua MD, N Foot, E Olavarria MD, E Kanfer FRCP, Prof J M Goldman FRCP, Prof J F Apperley FRCP); International Bone Marrow Transplant Registry, Medical College of Wisconsin, Milwaukee, WI, USA (J P Klein); and Department of Haematology, Western General Hospital, Edinburgh, UK (P Shepherd MRCP) Correspondence to: Prof John M Goldman (e-mail: [email protected])
Effect of losartan on sudden cardiac death in people with diabetes: data from the LIFE study Lars H Lindholm, Björn Dahlöf, Jonathan M Edelman, Hans Ibsen, Knut Borch-Johnsen, Michael Hecht Olsen, Steven Snapinn, Kristian Wachtell, for the LIFE study group In the Losartan Intervention For Endpoint reduction in hypertension (LIFE) study, a major reduction of all-cause mortality—especially cardiovascular mortality—in patients with diabetes with left ventricular hypertrophy was reported for treatment with losartan. We postulated post hoc that losartan might have a better effect on sudden cardiac death than atenolol, and we aimed to test this hypothesis. 44 patients with diabetes died of sudden cardiac death; significantly fewer deaths arose in the losartan group (14) than in the atenolol group (30; p=0·027). In the losartan group, five (6%) of 86 patients with diabetes and atrial fibrillation during the trial died of sudden cardiac death
Cardiovascular death
Losartan (n=586)
Atenolol (n=609)
Rate*
n
Rate*
Adjusted hazard ratio (95% CI)†
p
n
13·6
38 (6%)
21·8
61 (10%)
0·63 (0·42–0·95)
0·028
Coronary heart disease death Sudden cardiac death (overall) <1 h 1–24 h Non-sudden cardiac death >24 h
7·5 5·0 2·1 2·9
21 (4%) 14 (2%) 6 (1%) 8 (1%)
13·2 10·7 3·9 6·8
37 (6%) 30 (5%) 11 (2%) 19 (3%)
0·59 (0·34–1·01) 0·49 (0·26–0·92) 0·58 (0·22–1·58) 0·43 (0·19–0·99)
0·052 0·027 0·290 0·047
2·5
7 (1%)
2·5
7 (1%)
1·01 (0·35–2·89)
0·987
Non-coronary cardiovascular death
6·1
17 (3%)
8·6
24 (4%)
0·71 (0·38–1·32)
0·273
*Per 1000 patient-years of follow-up. †Adjusted for amount of left ventricular hypertrophy and Framingham risk score at randomisation.
Coronary deaths of patients with diabetes treated with losartan or atenolol
THE LANCET • Vol 362 • August 23, 2003 • www.thelancet.com
619
For personal use. Only reproduce with permission from The Lancet
RESEARCH LETTERS
8 Atenolol Losartan
Proportion of patients (%)
7
p=0·027
6 5 4 3 2 1 0 0
12
36
48
60
525 533
198 234
Study month
Number at risk Atenolol 609 Losartan 586
24
592 573
575 566
550 552
Sudden cardiac death by month of follow-up for patients with diabetes (n=1195) treated with atenolol or losartan
Significantly fewer cardiovascular deaths arose in the losartan group than in the atenolol group, and there was a similar tendency for coronary deaths (table). When analysed separately, significantly fewer sudden cardiac deaths happened in the losartan group than in the atenolol group (figure), irrespective of whether deaths arose within 1 h or 24 h of symptom onset (table). These findings of reduced risk in losartan-treated patients were independent of other risk factors for sudden cardiac death. Difference in regression of left ventricular hypertrophy between the two treatment groups did not account for the difference in sudden cardiac death. The last serum potassium value before sudden death was closely similar in patients in the losartan group (mean 4·26 mmol/L [SD 0·42]) and in the atenolol group (4·23 mmol/L [0·40]); change from baseline was 0·01 mmol/L in both groups. In those patients in the LIFE trial who did not have diabetes (n=7998), there were equal numbers of sudden cardiac deaths (n=67) in each of the two treatment groups. In patients with diabetes, 19 (10%) of those with atrial fibrillation (n=191) at baseline or during the trial died of sudden cardiac death. In those without atrial fibrillation (n=1004), 25 (2%) died of sudden cardiac death. The corresponding figures for the losartan group were five of 86 (6%) versus nine of 500 (2%), and in the atenolol group 14 (13%) of 105 versus 16 (3%) of 504. Study drug discontinuation—particularly discontinuation of atenolol— did not account for these differences. Results of our post-hoc analysis showed a close to 50% risk reduction in sudden death in losartan-treated patients with diabetes compared with those treated with atenolol. This finding was not so in patients without diabetes. Our results should be viewed against the background that blockers are known to offer protection against cardiovascular death and sudden cardiovascular death, particularly in patients with active coronary heart disease. A possible explanation for the enhanced outcome in patients in the losartan group is that this drug has a better anti-
620
arrhythmic property than atenolol, which accords with our findings in patients with diabetes with and without atrial fibrillation. The number of sudden cardiac deaths (n=44) in the present analyses consisted of 44% of all cardiovascular deaths and 26% of all deaths in patients with diabetes from the LIFE study during follow-up. From experimental studies and analyses of the LIFE study, losartan compared with atenolol is known to reduce the risk of episodes of atrial fibrillation. Losartan would possibly favour an antiarrhythmic effect, by leading to regression of left ventricular hypertrophy and possibly less atrial fibrosis. Abnormalities in QT-dispersion and heart rate variability are known to be associated with left ventricular hypertrophy, possibly by induced repolarisation abnormalities. A modulation of sympathetic nerve activity might be implicated as well. Inhibition of angiotensin-converting enzyme, and angiotensin 2 antagonism, might inhibit the sympathetic nervous system by blunting of the facilitating effect of angiotensin 2 on sympathetic tone. Importantly, our analyses were exploratory and require confirmation. Contributors Study data are in a Merck database. Merck provided authors with free access to all data. The authors designed the analyses, interpreted the data, and wrote the paper. Merck reviewed the paper. All authors have commented on the manuscript.
Conflict of interest statement LHL chairs the Working Group on High Blood Pressure 2001–2004 for The Swedish Council on Technology Assessment in Health Care, and has previously received a research grant from AstraZeneca for the ALPINE study. BD has served as a consultant to many pharmaceutical companies, including Merck, and has attended speaking engagements and been compensated for travel and time spent on research and lectures by these companies. MHO and HI have been supported partly by grants from Merck. KW has received research grants and speaking honoraria from Merck. JME and SS (statistician) are employees of Merck.
Acknowledgments This study was funded by Merck. The sponsor reviewed the report and employs two of the authors (JE and SS). 1
2 3
4 5
Yusuf S, Wittes J, Friedman L. Overview of results of randomised clinical trials in heart disease, I: treatment following myocardial infarction. JAMA 1988; 260: 2088–93. Singh N. Diabetes, heart rate and mortality. J Cardiovasc Pharmacol Ther 2002; 7: 117–29. Dellsperger KC, Martins JB, Clothier JL, Marcus ML. Incidence of sudden cardiac death associated with coronary artery occlusion in dogs with hypertension and left ventricular hypertrophy is reduced by chronic beta-adrenergic blockade. Circulation 1990; 82: 941–50. Reiter MJ. Beta-adrenergic blocking drugs as antifibrillatory agents. Curr Cardiol Rep 2002; 4: 426–33. Lindholm LH, Ibsen H, Dahlöf B, et al. Cardiovascular morbidity and mortality in patients with diabetes in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet 2002; 359: 1004–10.
Department of Public Health and Clinical Medicine, Umeå University Hospital, SE 901 85 Umeå, Sweden (Prof L H Lindholm MD); Sahlgrenska University Hospital/Östra, Göteborg, Sweden (B Dahlöf MD); Merck Research Laboratories, West Point, PA, USA (J M Edelman MD, S Snapinn PhD); Glostrup University Hospital, Glostrup, Denmark (M Hecht Olsen MD, H Ibsen MD, K Wachtell MD); and Steno Diabetes Centre, Gentofte, Denmark (K Borch-Johnsen MD) Correspondence to: Prof Lars H Lindholm (e-mail: [email protected])
THE LANCET • Vol 362 • August 23, 2003 • www.thelancet.com
For personal use. Only reproduce with permission from The Lancet.
compared with nine (2%) of 500 in those without atrial fibrillation. The respective figures for the atenolol group were 14 (13%) of 105 and 16 (3%) of 504. Our results suggest losartan affords better protection against cardiac death from arrhythmias for patients with diabetes mellitus than does atenolol. Importantly, our analyses were exploratory and require confirmation.
investigation was supported in part by the Leukaemia Research Fund. Imperial College London and Hammersmith Hospital received reimbursement from Novartis Pharma for the costs incurred in entering patients in the trial. The sponsor had no role in study design, data collection, data interpretation, data analysis, or writing of the report. 1
2
3
4
5
Druker BJ, Talpaz M, Resta DJ, et al. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001; 344: 1031–37. Kantarjian H, Sawyers C, Hochhaus A, et al. Hematologic and cytogenetic responses to imatinib mesylate in chronic myelogenous leukemia. N Engl J Med 2002; 346: 645–52. Allan NC, Richards SM, Shepherd PC. UK Medical Research Council randomised, multicentre trial of interferon-alfa n1 for chronic myeloid leukaemia: improved survival irrespective of cytogenetic response. The UK Medical Research Council’s Working Parties for Therapeutic Trials in Adult Leukaemia. Lancet 1995; 345: 1392–97. Shah N, Nicoll J, Nagar B, et al. Multiple BCR-ABL kinase domain mutations confer polyclonal resistance to the tyrosine kinase inhibitor imatinib (STI571) in chronic phase and blast crisis chronic myeloid leukemia. Cancer Cell 2002; 2: 117–25. O’Brien SG, Guilhot F, Larson RA, et al. Imatinib compared with interferon and low dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med 2003; 348: 994–1004.
Lancet 2003; 362: 619–20 See Commentary page 591
Prevention of sudden cardiac death in patients with coronary heart disease is an important goal for treatment. In placebo-controlled long-term studies, several blockers, including atenolol, have shown consistent reductions in postinfarction mortality, and up to 50% reduction in risk of sudden cardiac death.1 In patients with diabetes mellitus, especially if associated with autonomic neuropathy, a rise in heart rate and fall in heart rate variability have been suggested to be important risk factors for sudden cardiac death and increased all-cause mortality.2 Moreover, blockade has been suggested to be more effective than angiotensin-converting enzyme inhibition in prevention of sudden cardiac death in a model of hypertension and left ventricular hypertrophy.3 blockade inhibits the proarrhythmic effects of both neural and humoral sympathetic stimulation and inhibits the vagal withdrawal that accompanies ischaemia.4 In a prespecified subgroup of patients with diabetes with left ventricular hypertrophy, workers on the Losartan Intervention For Endpoint reduction in hypertension (LIFE) study reported a major reduction of all-cause mortality—especially cardiovascular mortality—with a losartan-based versus an atenolol-based antihypertensive treatment.5 Against this background, we postulated post hoc that the losartan-based treatment might have a better effect on sudden cardiac death than atenolol-based therapy, and we aimed to test this hypothesis. As part of the main LIFE study, patients with diabetes, hypertension, and electrocardiogram-documented left ventricular hypertrophy were assigned once-daily losartanbased or atenolol-based antihypertensive therapy.5 In both groups, about 85% received a diuretic as supplementary treatment. Here, we used Cox regression analysis with baseline Framingham risk score and electrocardiogram-left ventricular hypertrophy as covariates to compare effects of losartan and atenolol on cardiovascular morbidity and mortality (defined as cardiovascular death, stroke, or myocardial infarction). An independent end-point committee, masked to the type of treatment the patients received, classified all endpoints including sudden cardiac death. We studied 1195 men and women (53% women) with diabetes, with a mean age of 67 years (SD 7). Average blood pressure after placebo run-in was 177/96 mm Hg. Patients were followed up for at least 4 years (mean 4·7 years [SD 1·1]).
Department of Haematology, Hammersmith Hospital, London W12 0NN, UK (D Marin MD, S Marktel MD, R Szydlo PhD, M Bua MD, N Foot, E Olavarria MD, E Kanfer FRCP, Prof J M Goldman FRCP, Prof J F Apperley FRCP); International Bone Marrow Transplant Registry, Medical College of Wisconsin, Milwaukee, WI, USA (J P Klein); and Department of Haematology, Western General Hospital, Edinburgh, UK (P Shepherd MRCP) Correspondence to: Prof John M Goldman (e-mail: [email protected])
Effect of losartan on sudden cardiac death in people with diabetes: data from the LIFE study Lars H Lindholm, Björn Dahlöf, Jonathan M Edelman, Hans Ibsen, Knut Borch-Johnsen, Michael Hecht Olsen, Steven Snapinn, Kristian Wachtell, for the LIFE study group In the Losartan Intervention For Endpoint reduction in hypertension (LIFE) study, a major reduction of all-cause mortality—especially cardiovascular mortality—in patients with diabetes with left ventricular hypertrophy was reported for treatment with losartan. We postulated post hoc that losartan might have a better effect on sudden cardiac death than atenolol, and we aimed to test this hypothesis. 44 patients with diabetes died of sudden cardiac death; significantly fewer deaths arose in the losartan group (14) than in the atenolol group (30; p=0·027). In the losartan group, five (6%) of 86 patients with diabetes and atrial fibrillation during the trial died of sudden cardiac death
Cardiovascular death
Losartan (n=586)
Atenolol (n=609)
Rate*
n
Rate*
Adjusted hazard ratio (95% CI)†
p
n
13·6
38 (6%)
21·8
61 (10%)
0·63 (0·42–0·95)
0·028
Coronary heart disease death Sudden cardiac death (overall) <1 h 1–24 h Non-sudden cardiac death >24 h
7·5 5·0 2·1 2·9
21 (4%) 14 (2%) 6 (1%) 8 (1%)
13·2 10·7 3·9 6·8
37 (6%) 30 (5%) 11 (2%) 19 (3%)
0·59 (0·34–1·01) 0·49 (0·26–0·92) 0·58 (0·22–1·58) 0·43 (0·19–0·99)
0·052 0·027 0·290 0·047
2·5
7 (1%)
2·5
7 (1%)
1·01 (0·35–2·89)
0·987
Non-coronary cardiovascular death
6·1
17 (3%)
8·6
24 (4%)
0·71 (0·38–1·32)
0·273
*Per 1000 patient-years of follow-up. †Adjusted for amount of left ventricular hypertrophy and Framingham risk score at randomisation.
Coronary deaths of patients with diabetes treated with losartan or atenolol
THE LANCET • Vol 362 • August 23, 2003 • www.thelancet.com
619
For personal use. Only reproduce with permission from The Lancet
RESEARCH LETTERS
8 Atenolol Losartan
Proportion of patients (%)
7
p=0·027
6 5 4 3 2 1 0 0
12
36
48
60
525 533
198 234
Study month
Number at risk Atenolol 609 Losartan 586
24
592 573
575 566
550 552
Sudden cardiac death by month of follow-up for patients with diabetes (n=1195) treated with atenolol or losartan
Significantly fewer cardiovascular deaths arose in the losartan group than in the atenolol group, and there was a similar tendency for coronary deaths (table). When analysed separately, significantly fewer sudden cardiac deaths happened in the losartan group than in the atenolol group (figure), irrespective of whether deaths arose within 1 h or 24 h of symptom onset (table). These findings of reduced risk in losartan-treated patients were independent of other risk factors for sudden cardiac death. Difference in regression of left ventricular hypertrophy between the two treatment groups did not account for the difference in sudden cardiac death. The last serum potassium value before sudden death was closely similar in patients in the losartan group (mean 4·26 mmol/L [SD 0·42]) and in the atenolol group (4·23 mmol/L [0·40]); change from baseline was 0·01 mmol/L in both groups. In those patients in the LIFE trial who did not have diabetes (n=7998), there were equal numbers of sudden cardiac deaths (n=67) in each of the two treatment groups. In patients with diabetes, 19 (10%) of those with atrial fibrillation (n=191) at baseline or during the trial died of sudden cardiac death. In those without atrial fibrillation (n=1004), 25 (2%) died of sudden cardiac death. The corresponding figures for the losartan group were five of 86 (6%) versus nine of 500 (2%), and in the atenolol group 14 (13%) of 105 versus 16 (3%) of 504. Study drug discontinuation—particularly discontinuation of atenolol— did not account for these differences. Results of our post-hoc analysis showed a close to 50% risk reduction in sudden death in losartan-treated patients with diabetes compared with those treated with atenolol. This finding was not so in patients without diabetes. Our results should be viewed against the background that blockers are known to offer protection against cardiovascular death and sudden cardiovascular death, particularly in patients with active coronary heart disease. A possible explanation for the enhanced outcome in patients in the losartan group is that this drug has a better anti-
620
arrhythmic property than atenolol, which accords with our findings in patients with diabetes with and without atrial fibrillation. The number of sudden cardiac deaths (n=44) in the present analyses consisted of 44% of all cardiovascular deaths and 26% of all deaths in patients with diabetes from the LIFE study during follow-up. From experimental studies and analyses of the LIFE study, losartan compared with atenolol is known to reduce the risk of episodes of atrial fibrillation. Losartan would possibly favour an antiarrhythmic effect, by leading to regression of left ventricular hypertrophy and possibly less atrial fibrosis. Abnormalities in QT-dispersion and heart rate variability are known to be associated with left ventricular hypertrophy, possibly by induced repolarisation abnormalities. A modulation of sympathetic nerve activity might be implicated as well. Inhibition of angiotensin-converting enzyme, and angiotensin 2 antagonism, might inhibit the sympathetic nervous system by blunting of the facilitating effect of angiotensin 2 on sympathetic tone. Importantly, our analyses were exploratory and require confirmation. Contributors Study data are in a Merck database. Merck provided authors with free access to all data. The authors designed the analyses, interpreted the data, and wrote the paper. Merck reviewed the paper. All authors have commented on the manuscript.
Conflict of interest statement LHL chairs the Working Group on High Blood Pressure 2001–2004 for The Swedish Council on Technology Assessment in Health Care, and has previously received a research grant from AstraZeneca for the ALPINE study. BD has served as a consultant to many pharmaceutical companies, including Merck, and has attended speaking engagements and been compensated for travel and time spent on research and lectures by these companies. MHO and HI have been supported partly by grants from Merck. KW has received research grants and speaking honoraria from Merck. JME and SS (statistician) are employees of Merck.
Acknowledgments This study was funded by Merck. The sponsor reviewed the report and employs two of the authors (JE and SS). 1
2 3
4 5
Yusuf S, Wittes J, Friedman L. Overview of results of randomised clinical trials in heart disease, I: treatment following myocardial infarction. JAMA 1988; 260: 2088–93. Singh N. Diabetes, heart rate and mortality. J Cardiovasc Pharmacol Ther 2002; 7: 117–29. Dellsperger KC, Martins JB, Clothier JL, Marcus ML. Incidence of sudden cardiac death associated with coronary artery occlusion in dogs with hypertension and left ventricular hypertrophy is reduced by chronic beta-adrenergic blockade. Circulation 1990; 82: 941–50. Reiter MJ. Beta-adrenergic blocking drugs as antifibrillatory agents. Curr Cardiol Rep 2002; 4: 426–33. Lindholm LH, Ibsen H, Dahlöf B, et al. Cardiovascular morbidity and mortality in patients with diabetes in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet 2002; 359: 1004–10.
Department of Public Health and Clinical Medicine, Umeå University Hospital, SE 901 85 Umeå, Sweden (Prof L H Lindholm MD); Sahlgrenska University Hospital/Östra, Göteborg, Sweden (B Dahlöf MD); Merck Research Laboratories, West Point, PA, USA (J M Edelman MD, S Snapinn PhD); Glostrup University Hospital, Glostrup, Denmark (M Hecht Olsen MD, H Ibsen MD, K Wachtell MD); and Steno Diabetes Centre, Gentofte, Denmark (K Borch-Johnsen MD) Correspondence to: Prof Lars H Lindholm (e-mail: [email protected])
THE LANCET • Vol 362 • August 23, 2003 • www.thelancet.com
For personal use. Only reproduce with permission from The Lancet.