Author + information
- Received February 23, 2005
- Revision received May 12, 2005
- Accepted May 15, 2005
- Published online September 6, 2005.
- Eigil Fossum, MD, PhD⁎,⁎ (, )
- Andreas Moan, MD, PhD†,
- Sverre E. Kjeldsen, MD, PhD⁎,‡,
- Richard B. Devereux, MD, FACC§,
- Stevo Julius, MD, ScD‡,
- Steven M. Snapinn, PhD∥,
- Jonathan M. Edelman, MD∥,
- Ulf de Faire, MD, PhD¶,
- Frej Fyhrquist, MD, PhD#,
- Hans Ibsen, MD, PhD⁎⁎,
- Krister Kristianson, PhD††,
- Ole Lederballe-Pedersen, MD, PhD‡‡,
- Lars H. Lindholm, MD, PhD§§,
- Markku S. Nieminen, MD, FACC#,
- Per Omvik, MD, PhD∥∥,
- Suzanne Oparil, MD, FACC¶¶,
- Hans Wedel, PhD##,
- Björn Dahlöf, MD, PhD⁎⁎⁎,
- LIFE Study Group
- ↵⁎Reprint requests and correspondence:
Dr. Eigil Fossum, Department of Cardiology, Ullevaal University Hospital, N-0407 Oslo, Norway
Objectives We conducted a subgroup analysis in the Losartan Intervention For Endpoint reduction in hypertension (LIFE) study to determine whether aspirin interacted with the properties of losartan, an angiotensin-II receptor antagonist.
Background Negative interactions between angiotensin-converting enzyme inhibitors and aspirin have been reported. There are no data reported from clinical trials about possible interactions between angiotensin-II receptor antagonists and aspirin.
Methods The LIFE study assigned 9,193 patients with hypertension and left ventricular hypertrophy (LVH) to losartan- or atenolol-based therapy for a mean of 4.7 years, with 1,970 (21.4%) taking aspirin at baseline. The primary composite end point (CEP) included cardiovascular death, stroke, and myocardial infarction (MI). The present cohort was stratified by aspirin use at baseline.
Results Blood pressures were reduced similarly in the losartan with aspirin (n = 1,004) and atenolol with aspirin (n = 966) groups. The CEP was reduced by 32% (95% confidence interval 0.55 to 0.86, p = 0.001) with losartan with aspirin compared to atenolol with aspirin, adjusted for Framingham risk score and LVH. The test for treatment versus aspirin interaction, excluding other covariates, was significant for the CEP (p = 0.016) and MI (p = 0.037).
Conclusions There was a statistical interaction between treatment and aspirin in the LIFE study, with significantly greater reductions for the CEP and MI with losartan in patients using aspirin than in patients not using aspirin at baseline. Further studies are needed to clarify whether this represents a pharmacologic interaction or a selection by aspirin use of patients more likely to respond to losartan treatment.
The Losartan Intervention For Endpoint reduction in hypertension (LIFE) study demonstrated that losartan-based therapy reduced cardiovascular end points significantly better than atenolol-based therapy for a similar reduction in blood pressure (BP) in patients with hypertension and left ventricular hypertrophy (LVH) (1–5).
There are reports of negative interactions between aspirin and angiotensin-converting enzyme inhibitors in preventing cardiovascular end points (6,7). These reports are not consistent, and the interaction may be dose dependent (8–10). There are no data from large clinical trials regarding potential interactions between AT1 antagonists and aspirin. Thus, we tested the hypothesis that aspirin does not interact negatively with the beneficial properties of losartan.
The LIFE study was an investigator-initiated, prospective, multinational, double-blind, double-dummy, randomized, active-controlled, parallel-group study. The primary objective was to evaluate the long-term effects of once-daily losartan- versus atenolol-based antihypertensive therapy in 55- to 80-year-old patients with hypertension and electrocardiographically (ECG) documented LVH on the incidence of cardiovascular death, stroke, and myocardial infarction (MI). The study protocol, design, organization, clinical measures, end point definitions, statistical considerations, baseline characteristics, LVH criteria, exclusion criteria, and main outcome have been published (1–5,11,12). In the present analysis, patients are stratified according to use of aspirin at baseline.
All cardiovascular end points and BPs were analyzed using the intention-to-treat approach. The difference between treatment groups with respect to clinical events was assessed by a Cox regression model with degree of LVH and the Framingham risk score (13) at baseline as covariates. This adjusted analysis with these covariates was chosen a priori as the primary analysis to account for any potential difference in key risk predictors at baseline. The interaction between treatment and aspirin was tested by including indicators for treatment group, aspirin status, and the product of the two in the Cox regression model. Treatment effects were measured by hazard ratios (relative risks [RRs]) and their 95% confidence intervals (CIs). The risk reduction for losartan versus atenolol was calculated as 100 × (1 − RR). Event rates over time are presented as Kaplan-Meier curves. Differences between groups in changes in ECG measures of LVH were analyzed with the Wilcoxon rank-sum test.
The p values for the comparison of baseline characteristics for patients with and without aspirin use at baseline are from a chi-square test for categorical variables and ANOVA for continuous variables. The present analyses are exploratory, and the p values are thus unadjusted for multiplicity. All tests were performed at two-sided 5% significance levels.
At baseline, 1,970 patients (21.4% of the main study population) used aspirin: 1,004 and 966 patients in the losartan and atenolol groups, respectively (Table 1).Patients with aspirin use at baseline assigned to losartan- or atenolol-based treatment were similar in demographic characteristics, severity of hypertension, prevalence of coexisting cardiovascular conditions, Framingham risk score, and ECG-LVH criteria. Compared with the cohort without aspirin use at baseline, the patients using aspirin were more likely to have a history of any vascular disease (60% vs. 16%), stroke (25% vs. 3%), and diabetes (17% vs. 12%) (p < 0.001 for all), as shown in Table 2.Among patients with and without aspirin at baseline, 46% and 56%, respectively, were women (p < 0.001). More than 90% in the aspirin and the non-aspirin cohorts were Caucasian.
Aspirin use throughout the study
The use of aspirin throughout the study is shown in Table 3to illustrate the number of patients who remained on aspirin therapy or switched group at different time points during the trial.
Study drug at end point or termination of follow-up
Mean follow-up time was 4.7 years. There were 74% and 68% of patients who remained on study therapy throughout the entire follow-up time in the losartan with aspirin and atenolol with aspirin groups, respectively, compared with 78% and 74% in the groups without aspirin.
Blood pressure, heart rate, and LVH
In the losartan and atenolol groups, systolic and diastolic BP levels at the last visit were 144.0/80.2 versus 145.7/79.3 mm Hg in the aspirin cohort (p = 0.060 for systolic BP and p = 0.051 for diastolic BP) and 145.0/81.9 versus 146.4/81.6 mm Hg in the non-aspirin cohort (p < 0.001 for systolic BP, p = 0.33 for diastolic BP), respectively. Systolic and diastolic BPs were reduced by 30.8/16.8 and 28.9/16.9 mm Hg in the losartan with aspirin and atenolol with aspirin groups, respectively (p = 0.067 for change in systolic BP and p = 0.77 for diastolic BP, respectively). In patients without aspirin use, BPs were reduced by 29.9/16.6 and 28.8/16.6 mm Hg in the losartan and atenolol groups, respectively (p = 0.018 for change in systolic BP and p = 0.95 for diastolic BP, respectively).
Among aspirin patients, heart rate changed from 73.6 to 71.9 beats/min, a 1.6-beat decrease in the losartan group, and from 73.5 to 65.6, a 7.8-beat decrease in the atenolol group. Among non-aspirin patients, heart rate went from 74.1 to 71.9 beats/min, a 2.1-beat decrease in the losartan group, and from 73.6 to 65.6, a 7.9-beat decrease in the non-aspirin group.
At the end of the study, Cornell voltage-duration product and Sokolow-Lyon voltage were reduced significantly in the losartan with aspirin cohort as compared with the atenolol with aspirin cohort, respectively (−271 vs. −37 mm × ms, p = 0.001 and −4.5 vs. −2.6 mm, p < 0.001) and also in the losartan without aspirin cohort as compared with the atenolol without aspirin cohort (−295 vs. −146 mm × ms, p < 0.001 and −4.6 vs. −2.7 mm, p < 0.001).
End points in losartan- and atenolol-treated patients taking aspirin at baseline
The primary composite end point (Table 4)occurred in 128 patients in the losartan group and in 180 patients in the atenolol group. The adjusted RR was 0.68 (95% CI 0.55 to 0.86, p = 0.001). Cardiovascular mortality occurred in 56 and 76 in the losartan and atenolol groups, respectively (RR 0.73, 95% CI 0.52 to 1.03, p = 0.074); stroke (non-fatal and fatal) occurred in 61 and 94, respectively (RR 0.63, 95% CI 0.45 to 0.86, p = 0.004); and MI (non-fatal and fatal) occurred in 44 and 58, respectively (RR 0.75, 95% CI 0.51 to 1.11, p = NS). Figure 1shows the Kaplan-Meier curves for the primary end point, cardiovascular mortality, stroke, and MI, and illustrates the interaction and time effect. The test for interaction (excluding other covariates) was significant for both the primary end point (stronger effect among patients with aspirin use at baseline, p = 0.016 for interaction; Fig. 1A) and MI (losartan rate lower than atenolol rate among patients with aspirin use at baseline, vice versa in patients without aspirin use at baseline, p = 0.037 for interaction; Fig. 1D). Table 5shows the effect of losartan relative to atenolol among patients not taking aspirin at baseline.
In this subgroup analysis of the LIFE study, there was a greater reduction in cardiovascular death, stroke, and MI with losartan-based compared to atenolol-based treatment in patients using aspirin than in patients not using aspirin at baseline. Blood pressure reductions were similar with both therapies.
The present analyses were performed on the basis of previous reports about aspirin and angiotensin-converting enzyme inhibitor interactions (6,7). The pharmacologic rationale for this interaction is the common bradykinin-prostaglandin pathway (9–11,14,15). There is limited information about possible aspirin interactions with AT1 antagonists, except for small studies (14) and experimental data (15–17). However, possible common pathways for aspirin and AT1 (16,17) and AT2 (18) receptors have been suggested, and the AT2 receptors, which are stimulated during AT1-blockade, may be involved in bradykinin production (18). Thus, a pharmacologic rationale for a possible AT1 antagonist–aspirin interaction exists. Furthermore, because the exact signaling pathways and function of the other AT receptors are still in part unknown (18), interactions may be present even though the mode of action is unknown. Because of the common use of both aspirin and AT1 antagonists and the lack of data about possible interactions from large-scale clinical trials, the present study was undertaken.
The present findings may be explained by a pharmacologic interaction, as discussed, or a selection by aspirin use of patients more likely to respond to losartan treatment. Significantly more of the patients taking aspirin at baseline had a history of vascular disease and diabetes compared with the patients not taking aspirin. Because of the relatively short duration of the study, it is not unexpected that the differences between losartan and atenolol are more clearly shown in the patients at highest risk of end points. Owing to the post-hoc design of the study, it cannot be concluded that losartan is equal to atenolol in patients not taking aspirin, as suggested in Table 5. Importantly, the main findings from the LIFE trial, a significant reduction in strokes and new-onset diabetes, were maintained in the patients not taking aspirin. The equal rate of new-onset diabetes in the aspirin cohort may, among other factors, be related to the high prevalence at baseline.
In accordance with previous publications regarding aspirin–angiotensin-converting enzyme inhibitor interaction (6–8), the data have been analyzed according to aspirin use at baseline. Although some patients started taking aspirin during the trial, Table 3shows that there was good separation between the groups throughout the trial.
A first limitation may be that the present study was a post-hoc subgroup analysis. Thus, there was no randomization to aspirin and aspirin use was not blinded. The patients using aspirin had a history of significantly more cardiovascular disease than did the patients not using aspirin. Second, a predominantly white population was studied. Third, the patients were high-risk hypertensive patients with ECG-LVH, and the outcome should be interpreted within this context. Moreover, because the aspirin dosage was not reported, we cannot assess a possible dose-dependent interaction, which has been an issue in the aspirin–angiotensin-converting enzyme inhibitor discussion (8–10). Strengths of the study were the large number of patients (n = 1,970) and the well-balanced distribution of the two groups with respect to baseline characteristics, prior medical history, and BP reduction.
Losartan-based antihypertensive therapy combined with aspirin was more effective than an atenolol-based treatment with aspirin in reducing the primary composite end point of cardiovascular morbidity and mortality and MI in a large subset of the LIFE study participants. The data show a statistical positive interaction between losartan and aspirin. Further studies are needed to clarify whether this represents a pharmacologic interaction or a selection by aspirin use of patients more likely to respond to losartan treatment.
We acknowledge Sigrid Helle Berg for dedicated work with the study and Paulette A. Lyle for manuscript assistance.
Merck and Co., Inc., has supported the trial by an unrestricted grant and has reviewed the manuscript. Dr. Fossum has received speakers’ honoraria from Merck. Drs. Kjeldsen, Devereux, Julius, de Faire, Fyhrquist, Ibsen, Lederballe-Pedersen, Lindholm, Nieminen, Omvik, Oparil, Wedel, and Dahlöf are members of the LIFE Steering Committee and have received grant support from Merck and Co., Inc., the sponsor of the LIFE study. Drs. Moan, Snapinn, Edelman, and Kristianson are or have been employees of Merck and may own stock or stock options in the company. The presentation of data is the intellectual property of the LIFE Steering Committee.
- Abbreviations and Acronyms
- blood pressure
- confidence interval
- Losartan Intervention For Endpoint reduction in hypertension
- left ventricular hypertrophy
- myocardial infarction
- relative risk
- Received February 23, 2005.
- Revision received May 12, 2005.
- Accepted May 15, 2005.
- American College of Cardiology Foundation
- Dahlöf B.,
- Devereux R.,
- de Faire U.,
- et al.
- Dahlöf B.,
- Keller S.E.,
- Makris L.,
- Goldberg A.I.,
- Sweet C.S.,
- Lim N.Y.
- Al-Khadra A.S.,
- Salem D.N.,
- Rand W.M.,
- Udelson J.E.,
- Smith J.J.,
- Konstam M.A.
- Okin P.M.,
- Roman M.J.,
- Devereux R.B.,
- Kligfield P.
- Anderson K.M.,
- Wilson P.W.,
- Odell P.M.,
- Kannel W.B.
- Nawarskas J.J.,
- Townsend R.R.,
- Cirigliano M.D.,
- Spinler S.A.
- Zhu B.Q.,
- Sievers R.E.,
- Browne A.E.,
- et al.