Skip Navigation

European Journal of Heart Failure 2002 4(4):501-506; doi:10.1016/S1388-9842(02)00099-5
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Otterstad, J. E.
Right arrow Articles by Ford, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Otterstad, J. E.
Right arrow Articles by Ford, I.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2002 European Society of Cardiology

The effect of carvedilol in patients with impaired left ventricular systolic function following an acute myocardial infarction

How do the treatment effects on total mortality and recurrent myocardial infarction in CAPRICORN compare with previous beta-blocker trials?

Jan Erik Otterstada,* and Ian Fordb

a Division of Cardiology, Vestfold Central Hospital N-3116 Toensberg, Norway
b Robertson Centre for Biostatisics, University of Glasgow Glasgow, Scotland, UK

* Corresponding author. Tel.: +47-333-42634; fax: +47-333-10623. E-maill address:j-otte{at}online.no


   Abstract
Top
 Abstract
1. Introduction
 2. CAPRICORN
 3. What about the...
 4. Limitations in the...
 5. Treatment effects in...
 6. Clinical implications
 References
 
In previous beta-blocker trials, post-myocardial infarction (MI) patients were essentially treated with a beta-blocker or placebo. In the CAPRICORN trial, patients were selected on the basis of a left ventricular (LV) ejection fraction (EF) <40% following the index MI and randomised to carvedilol or placebo, in addition to modern secondary prophylaxis with ACE inhibitors, aspirin and statins. In 1959 patients with a mean LVEF of 33%, treatment with carvedilol over a mean follow-up period of 15 months reduced total mortality from 15.3% with placebo to 11.9% with carvedilol [relative risk reduction (RRR) =23%, absolute risk reduction (ARR) =3.4%]. The incidence of recurrent MI was reduced from 5.8 to 2.3% (RRR 41%, ARR 2.3%). The number needed to treat (NNT) to prevent one death was 28 for the entire study period and 43 for 1 year of treatment. The results of the CAPRICORN trial are compared with three previous beta-blocker post-MI trials: the Gothenburg metoprolol trial (GMT), the Norwegian timolol trial (NTT) and the beta-blocker heart attack trial (BHAT). The RRRs for total mortality were 36% in the GMT and NTT, and 27% in BHAT. The respective NNTs for total mortality were 32, 18 and 38. NNT for 1 year of treatment was 25 in NTT and 80 in BHAT. The RRR for recurrent MIs were 28% in NTT and 16% in BHAT. The reduction of mortality and recurrent MIs in CAPRICORN is within the range of previous post-MI beta-blocker studies. In post-MI patients with LVEF<40%, add-on treatment with a beta-blocker should be given >48 h after initiation with an angiotensin-converting enzyme inhibitor (ACEI) and then with a slow dose escalation as applied in CAPRICORN.

Key Words: CAPRICORN study • Beta-blocker studies post myocardial infarction

Received October 11, 2001; Revised December 17, 2001; Accepted February 22, 2002


   1. Introduction
Top
 Abstract
 1. Introduction
2. CAPRICORN
 3. What about the...
 4. Limitations in the...
 5. Treatment effects in...
 6. Clinical implications
 References
 
The beneficial effect of beta-blocker therapy following a myocardial infarction (MI) was established in studies conducted in the 1970s and 1980s [1–4]. At that time revascularisation therapy was not performed and few patients were on aspirin, ACE inhibitors were not available and there was no systematic evaluation of left ventricular (LV) function. In addition, patients who were included in these studies were mostly at relatively low risk, since those with overt heart failure, and hence many with LV dysfunction, were not included. In the SAVE trial there was a significant effect on mortality with the use of captopril vs. placebo in asymptomatic post-MI patients with LVEF <40% [5]. Subsequently, several studies have verified that beta-blocker therapy, in addition to ACE inhibitors, in patients with chronic heart failure and LVEF<30–40%, has a significant effect on cardiac morbidity and mortality [6–9]. In the recently published CAPRICORN (carvedilol post-infarct survival control in LV dysfunction) study [10], the non-selective beta-blocker carvedilol was added to an ACE inhibitor in post-MI patients with an LVEF<40%. Almost all of these patients were treated with an ACE inhibitor prior to inclusion. This study may be regarded as a logical extension of the SAVE trial, with the exception that both symptomatic and asymptomatic patients were included.

In this review we summarise the design and main results of CAPRICORN, then give a short overview of the main results of three of the most important long-term beta-blocker studies after acute MI [1–4] and evaluate the treatment effect in CAPRICORN vs. the previous studies. Finally, the clinical implications of the CAPRICORN study are discussed, with emphasis on the best approach for establishing beta-blocker therapy in post-MI patients with LV dysfunction.


   2. CAPRICORN
Top
 Abstract
 1. Introduction
 2. CAPRICORN
3. What about the...
 4. Limitations in the...
 5. Treatment effects in...
 6. Clinical implications
 References
 
2.1. Design
This was a multi-centre double-blind, randomised, placebo-controlled trial comprising 1959 post-MI patients with LVEF<40%. The subjects were randomised to placebo or carvedilol, 3–21 (mean 10) days following the index MI, 975 to carvedilol and 984 to placebo. A prerequisite for inclusion was that patients had been given an ACE inhibitor (unless contraindicated) for at least 48 h, with a stable dose for the last 24 h, prior to randomisation. The initial dose of carvedilol was 6.25 mg b.i.d. and if tolerated the dose was doubled every 3–10 days, provided heart failure and adverse effects were absent, heart rate <50 and systolic blood pressure <80 mmHg. The target dose of 25 mg b.i.d. was to be reached within 2–4 weeks. The study duration was event-driven and the primary end-point was total mortality. Due to a much lower mortality than anticipated, a ‘co-primary’, combined end-point was introduced: total mortality or cardiovascular hospitalisation. The necessary number of events to achieve sufficient statistical power was 633. Secondary end-points were sudden death and hospitalisations for heart failure. In addition, non-fatal MI and the combined outcome of total mortality and non-fatal myocardial infarction were reported.

The endpoints emphasised by the CAPRICORN investigators (death, hospitalisation for heart failure and cardiovascular hospitalisation) are the types of outcome that would be expected to be reported in a modern heart failure study. However, CAPRICORN was primarily a post-MI trial with the restriction to subjects with evidence of LV dysfunction. For this reason, it is important in making comparison to previous post-MI trials to do this on the basis of endpoints that would be expected to be reported in such trials, namely death and MI. In fact, previous beta-blocker trials did not report in detail on the incidence of hospitalisation for heart failure and cardiovascular hospitalisation.

2.2. Results
The mean age at inclusion was 63 (25–90) years; approximately 75% were men, 30% had a recurrent index MI, 22% had diabetes mellitus, 12% had been previously revascularised and 57% had an anterior index MI. Treatment included i.v. nitrates in 73%, thrombolysis/primary PTCA in 46% and intravenous diuretics in 34% of these patients. At inclusion, 86% were on aspirin and 98% on an ACE inhibitor.

The target dose of 25 mg b.i.d. was reached in 74% of patients randomised to carvedilol and 84% of those randomised to placebo.

During an average follow-up of 15 months, 707 co-primary end-points occurred and 267 (13.6%) patients died. Total mortality was 116/975 (11.9%) in the carvedilol group and 151/984 (15.3%) in the placebo group (RRR=23%; 95% CI=2–40%). The effect on the combined outcome of total mortality and cardiovascular hospitalisation was not significant. The CAPRICORN investigators have pointed out that there were a similar number of cardiovascular hospitalisations for reasons other than MI or heart failure in the early phase of follow-up in both groups. There was a significant reduction of the combined end-point of total mortality and non-fatal MI (P=0.002). The incidence of non-fatal MI was 3% with carvedilol and 6% with placebo (RRR=41%; 95%CI=10–61%).


   3. What about the previous beta-blocker studies?
Top
 Abstract
 1. Introduction
 2. CAPRICORN
 3. What about the...
4. Limitations in the...
 5. Treatment effects in...
 6. Clinical implications
 References
 
The three most important long-term studies of beta-blocker treatment post-MI are probably the Gothenburg metoprolol trial (GMT) [1], the Norwegian timolol trial (NTT) [2] and the beta-blocker heart attack trial (BHAT) with propranolol [3,4].

The GMT was published in 1981 and is the shortest study, with an overall follow-up of only 3 months. A total of 1395 patients with definite MI (58%) or ‘suspected’ MI (42%) were included. The age, percentage of recurrent MI and of patients with clinical/radiological evidence of heart failure prior to inclusion are given in Table 1. Treatment was started early, within 48 h following the onset of symptoms, and the starting dose was 15 mg of metoprolol or placebo given intravenously. After 15 min, patients were given a dose of 50 mg orally, repeated every 6th h, and after 48 h the dose was increased to 100 mg b.i.d. for 3 months. The mortality in the placebo group was 62/697 (8.9%) vs. 40/698 (5.7%) in the group treated with metoprolol (RRR=36%, P<0.030).


View this table:
[in this window]
[in a new window]

  		Table 1 Baseline risk factors in the three beta-blocker trials: Gothenburg metoprolol trials (GMT) [1], the Norwegian timolol trial (NTT) [2] and the beta-blocker heart attack trial (BHAT) [3], as expressed by age, percentage of recurrent myocardial infarction (MI), anterior MI, clinical/radiological signs of heart failure prior to randomisation, and cardiomegaly assessed radiologically

 
The NTT was also published in 1981 and all 1884 patients included had a definite MI according to the WHO criteria. Treatment was started 7–28 days after the index MI with a dose of 5 mg b.i.d. orally, or matching placebo. After 2 days the target dose of 10 mg b.i.d. was given and mean follow-up was 17 (12–33) months. As evident from Table 1, these patients were at relatively high risk since they were older; 34% had some evidence of heart failure prior to randomisation (pulmonary rales, third heart sound or radiological evidence of pulmonary congestion) and 23% had cardiomegaly on a chest X-ray. They also had a relatively high incidence of anterior and recurrent MIs. The mortality in the placebo group was 152/939 (16.2%) vs. 98/945 (10.4%) in the timolol group (RRR=36%, P=0.0001). The incidence of recurrent MI was reduced from 20.1% with placebo to 14.4% with timolol (P=0.0006).

BHAT conducted in the USA was published in 1982 and comprised 3837 patients with a documented MI. Patients were included 5–21 (mean 14) days following the index MI. The starting dose was propranolol or matching placebo at a dose of 20 mg orally, followed by 40 mg every 8th h. After a minimum of six doses (~48 h), a blood sample was drawn for determination of serum propranolol. If the level in propranolol-treated patients was less than 20 ng/ml, they were assigned to 80 mg t.i.d. (18%); otherwise, they were assigned to 60 mg t.i.d. (82%). As can be observed from Table 1, these patients were younger with a low percentage of anterior and recurrent MIs, and only 14% had evidence of heart failure prior to inclusion, whereas 34% had cardiac enlargement as judged radiologically. Mean follow-up was 25 months and the mortality was 188/1921 (9.8%) in the placebo group vs. 138/1916 (7.2%) with propranolol (RRR=27%, P<0.005). The incidence of ‘definite’ recurrent MI was 5.3% in the placebo group and 4.4% in the group randomised to propranolol (RRR=16%).


   4. Limitations in the comparison between previous studies vs. CAPRICORN
Top
 Abstract
 1. Introduction
 2. CAPRICORN
 3. What about the...
 4. Limitations in the...
5. Treatment effects in...
 6. Clinical implications
 References
 
As noted above, end points such as hospitalisation for heart failure or cardiovascular hospitalisation cannot be used for comparison with the earlier studies, since they were not reported in detail. Furthermore, the abrupt dose escalation in the previous studies vs. the more gradual dose increase in CAPRICORN indicates that a comparison on the development of heart failure in general may be inappropriate. However, a direct comparison on very simple and unchangeable end-points, such as total mortality and recurrent MI, may still be justified. Although new criteria for the diagnosis of acute MI have been suggested [11], it seems likely that the criteria for recurrent MI have been similar in the previous studies and CAPRICORN.

Another problem with such a comparison is different patient selection. In the earlier studies, patients with heart failure and, accordingly, a large number with LV dysfunction were not included. Although 34% of the patients in NTT had some evidence of heart failure prior to inclusion, they had been temporarily stabilised with a diuretic and were without symptoms or signs of heart failure when randomised. By definition, the patient population in CAPRICORN should be at higher risk, but the concomitant medical treatment with the addition of revascularisation procedures will have had a favourable influence on the risk. The exact importance of this additional treatment is difficult to assess in the comparison with earlier trials, where beta-blockers were the only treatment given. In order to evaluate the risk level in the earlier studies and CAPRICORN, we have estimated placebo mortality per year in the timolol trial, BHAT and CAPRICORN in Fig. 1. It has been calculated on the basis of the number of deaths divided by the average number of months of follow-up and multiplied by 12. This also represents a problem, since the risk is anticipated to be highest in the first year in a post-MI study. This risk will furthermore depend on baseline risk factors (Table 1) and the time point of inclusion in relation to the index MI, as well as the duration of follow-up. The time from MI to inclusion is quite similar in CAPRICORN when compared with NTT and BHAT. The follow-up time is nearly identical in CAPRICORN and NTT, but longer in BHAT. In all three studies the patients were in a stable condition when included. Since the GMT trial only ran for 3 months and the patients were included at an earlier stage post-MI, it has not been included in these calculations. It seems apparent that the risk in the timolol trial was very close to that in CAPRICORN. In comparison, the BHAT population had a relatively low risk at baseline (Table 1), and the longer follow-up time also tends to reduce the average risk as calculated in Fig. 1. These differences in placebo mortality must be taken into consideration when assessing the absolute treatment effects as reflected from ARR and NNT.


Figure 1
View larger version (28K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Bar chart showing the mean follow-up in months (bars to the left) in the Norwegian timolol trial (NTT) [2], beta-blocker heart attack trial (BHAT) [3] Gothenburg metoprolol trial (GMT) [1] and CAPRICORN [10]. The yearly percentage mortality in the placebo groups is shown (bars to the right) for the NNT, BHAT and CAPRICORN trials (GMT not included due to short-term follow-up).

 

   5. Treatment effects in the earlier trials vs. CAPRICORN
Top
 Abstract
 1. Introduction
 2. CAPRICORN
 3. What about the...
 4. Limitations in the...
 5. Treatment effects in...
6. Clinical implications
 References
 
With the above-mentioned reservations in mind, we have compared the ARR, RRR and NNT for total mortality and recurrent MIs. Since the follow-up time differs between studies, we also calculated NNT per year of follow-up. As previously published [12], NNT is derived from the difference in end-point rates in the placebo vs. the treatment group. The number of patients enrolled in the treatment group divided by this difference is NNT for the entire study period. The NNT for 1 year follow-up (NNT-1-year) is derived from the number included in the treatment group divided by the difference in the overall end-point rates divided by the average number of months follow-up and then multiplied by 12.

The mortality in the placebo and treatment arms, as well as ARR and RRR, in all four studies are summarised in Table 2. The ARR and RRR were greatest in the timolol trial. However, the treatment results in CAPRICORN are well within the range of the previous beta-blocker trials. This is further supported by the NNT and NNT-1-year data shown in Fig. 2.


View this table:
[in this window]
[in a new window]

  		Table 2 Mean follow-up (months), total mortality in the placebo and treatment groups, absolute (ARR) and relative risk reduction in the GMT [1], NTT [2], BHAT [3] and CAPRICORN [10] trials

 


Figure 2
View larger version (36K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Bar chart of number-needed-to-treat (NNT) to postpone one death during the entire study for the GMT [1], BHAT [3] NTT [3] and CAPRICORN [10] trials. In addition, NNT adjusted for 1-year follow-up is shown to the right (not for the GMT due to a short follow-up period). NA, not available. Abbreviations as in Fig. 1.

 
The incidence and RR of recurrent MIs are shown in Table 3 (the GMT trial is not included since such data were not provided). Again, the timolol data are very strong, but the treatment effect for recurrent MIs in CAPRICORN is within the range of the two previous studies.


View this table:
[in this window]
[in a new window]

  		Table 3 Incidence and risk reduction of recurrent myocardial infarction in the GMT [1], NTT [2] (fatal and non-fatal), BHAT [3] (non-fatal, definite) and CAPRICORN [10] (non-fatal) trials

 
The patient tolerance, expressed as the withdrawal rate (deaths not included) in the treatment and placebo arms in the four studies, is depicted in Fig. 3, demonstrating a satisfactory tolerance for carvedilol when compared with the previous studies. It is noteworthy that even with a withdrawal rate as high as 29% with timolol, the treatment results (based on intention to treat) were remarkably strong in that study.


Figure 3
View larger version (32K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Percentage of withdrawals in the treatment and placebo arms in the NTT [2], BHAT [3], GMT [1] and CAPRICORN [10] trials. Abbreviations as in Fig. 1.

 
To summarise, it seems fair to conclude that the reduction of total mortality and recurrent MIs in CAPRICORN can be regarded as comparable to the earlier beta-blocker studies. This is important, since carvedilol was given in addition to modern secondary prophylaxis, and approximately 50% of patients had been treated with thrombolysis or primary PTCA for their index MI. An additional 11% had been through revascularisation procedures prior to inclusion in the study. The tolerability of carvedilol added to other prophylactic strategies was quite satisfactory. This is also remarkable, since all CAPRICORN patients were selected on the basis of proven LV dysfunction.


   6. Clinical implications
Top
 Abstract
 1. Introduction
 2. CAPRICORN
 3. What about the...
 4. Limitations in the...
 5. Treatment effects in...
 6. Clinical implications
References
 
Before CAPRICORN there were no proper guidelines or evidence-based experience for post-MI patients with LV dysfunction to direct treatment with an ACE inhibitor combined with beta-blocker. Earlier post-MI studies, such as SAVE, AIRE and TRACE [5,13,14] had taught us to start with an ACE inhibitor as soon as the patient was stabilised and then to titrate the dose to a predefined goal. However, considerable confusion still existed on how to introduce a beta-blocker in this setting, since the recent positive trials with beta-blockers in heart failure had not included post-MI patients in the acute phase [6–9]. The previous beta-blocker post-MI trials had not selected patients with proven LV dysfunction. Only in the smallest of these trials, GMT, with a very short follow-up period, was treatment started early with the initial dose given i.v. However, the design of that study did not allow us to evaluate the additional effect of early i.v. treatment vs. the oral approach, as applied in the two larger trials with a much longer period of follow-up.

So far, we have no information on the type of ACE inhibitor and the dosage that was applied in CAPRICORN. Furthermore, there is no information on the extent to which the ACE inhibitor dose was changed after randomisation. Such information will be highly valuable for future clinical practice in this patient group.

Can the CAPRICORN results be regarded as a class effect? The results in CAPRICORN are valid for carvedilol in the doses applied. But we cannot rule out the possibility that other beta-blockers, such as metoprolol and bisoprolol, may have similar effects, since these two drugs have recently been documented to be of significant value in patients with LV dysfunction and chronic heart failure [7,8]. However, we do not know if the results of CAPRICORN are reproducible with other beta-blockers, and, in particular, which doses to use. Whether we believe in a class-effect or not, CAPRICORN supports a strategy of treating post-MI patients with LV dysfunction with an ACE inhibitor first. When stabilised, a beta-blocker should be added with a slow dose escalation aiming for a target dose similar to that which has been proven in trials showing an effect of beta-blocker treatment on mortality and morbidity in heart failure.


   References
Top
 Abstract
 1. Introduction
 2. CAPRICORN
 3. What about the...
 4. Limitations in the...
 5. Treatment effects in...
 6. Clinical implications
 References
 

  1. Hjalmarson Å., Herlitz J., Målek L., et al. Effect on mortality of metoprolol in acute myocardial infarction. Lancet (1981) ii:823–827.
  2. Norwegian Multicenter Study Group. Timilol-induced reduction in mortality and reinfarctions in patients surviving acute myocardial infarction. New Engl J Med (1981) 304:801–807.[Abstract]
  3. Beta-blocker Heart Attack Trial Research Group. A randomised trial of propranolol in patients with acute myocardial infarction. I: mortality results. J Am Med Assoc (1982) 247:1707–1714.[Abstract/Free Full Text]
  4. Beta-blocker Heart Attack Research Group. A randomised trial of propranolol in patients with acute myocardial infarction. II: morbidity results. J Am Med Assoc (1983) 250:2814–2819.[Abstract/Free Full Text]
  5. SAVE Investigators. Pfeffer M.A., Braunwald E., Moye L.A., et al. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. Results of the Survival and Ventricular Enlargement Trial. N Engl J Med (1992) 327:669–677.[Abstract]
  6. Packer M., Bristow M., Cohn J., et al. The effect of carvedilol on morbidity and mortality in heart failure. N Engl J Med (1996) 334:1349–1355.[Abstract/Free Full Text]
  7. CIBIS-II Investigators. The Catdiac Insufficiency Bisoprolol Study II (CIBIS II): a randomised trial. Lancet (1999) 353:9–13.[CrossRef][Web of Science][Medline]
  8. The MERIT-HF Study Group. Effect of metoprolol CR/XL in chronic heart failure (MERIT-HF). Lancet (1999) 353:2001–2007.[CrossRef][Web of Science][Medline]
  9. Packer M., Coats A.J.S., Fowler M.B., et al. Effect of carvedilol on survival in severe chronic heart failure. New Engl J Med (2001) 344:1651–1658.[Abstract/Free Full Text]
  10. The CAPRICORN Investigators. Effect of carvedilol on outcome after myocardial infarction in patients with left ventricular dysfunction: the CAPRICORN randomised trial. Lancet (2001) 357:1385–1390.[CrossRef][Web of Science][Medline]
  11. The Joint European Society of Cardiology/American College of Cardiology Committee. Myocardial infarction redefined—a consensus document of the Joint European Society of Cardiology Committee for the Redefinition of Myocardial Infarction. Eur Heart J (2000) 21:1502–1513.[Abstract/Free Full Text]
  12. Otterstad J.E., Sleight P. The HOPE study: comparison with other trials of secondary prevention. Eur Heart J (2001) 22:1307–1310.[Abstract/Free Full Text]
  13. The Acute Infarction Ramipril Efficacy (AIRE) Study Investigators. Effect of ramipril on mortality and morbidity of survivors of acute myocardial infarction with clinical evidence of heart failure. Lancet (1993) 342:821–828.[Web of Science][Medline]
  14. Kober L., Torp-Pedersen C., Carlsen J.E., et al. A clinical trial of the angiotensin-converting-enzyme inhibitor trandolapril in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med (1995) 333:1670–1676.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 CirculationHome page
J. V. Linseman and M. R. Bristow
Drug Therapy and Heart Failure Prevention
Circulation, March 11, 2003; 107(9): 1234 - 1236.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Otterstad, J. E.
Right arrow Articles by Ford, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Otterstad, J. E.
Right arrow Articles by Ford, I.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?