OUP user menu

DECOPI (DEsobstruction COronaire en Post-Infarctus): a randomized multi-centre trial of occluded artery angioplasty after acute myocardial infarction

Philippe Gabriel Steg, Christophe Thuaire, Dominique Himbert, Didier Carrié, Stéphane Champagne, Damien Coisne, Khalife Khalifé, Pierre Cazaux, Damien Logeart, Michel Slama, Christian Spaulding, Ariel Cohen, Ashok Tirouvanziam, Jean-Michel Montély, Rosa-Maria Rodriguez, Eric Garbarz, William Wijns, Isabelle Durand-Zaleski, Raphaël Porcher, Lionel Brucker, Sylvie Chevret, Claude Chastang
DOI: http://dx.doi.org/10.1016/j.ehj.2004.10.019 2187-2194 First published online: 10 December 2004


Aim To determine whether late recanalization of an occluded infarct artery after acute myocardial infarction is beneficial.

Methods and results Two hundred and twelve patients with a first Q-wave myocardial infarction (MI) and an occluded infarct vessel were enrolled. After coronary and left ventricular contrast angiography, patients were randomized to percutaneous revascularization (PTCA, n=109), carried out 2–15 days after symptom onset or medical therapy (n=103). The primary endpoint was a composite of cardiac death, non-fatal MI, or ventricular tachyarrhythmia. The majority had single-vessel disease and less than one-third had involvement of the left anterior descending artery. The use of pharmacological therapy was high in both groups. At six months, left ventricular ejection fraction was 5% higher in the invasive compared with the medical group (P=0.013) and more patients had a patent artery (82.8% vs 34.2%, P<0.0001). Restenosis was seen in 49.4% of patients in the PTCA group. At a mean of 34 months of follow-up, the occurrence of the primary endpoint was similar in the medical and PTCA groups (8.7% vs 7.3% respectively, P=0.68), but the overall costs were higher for PTCA. The secondary endpoint combining the primary endpoint with admission for heart failure was also similar between groups (12.6% vs 10.1% in the medical and PTCA groups, respectively, P=0.56).

Conclusions Systematic late PTCA of the infarct vessel was associated with a higher left ventricular ejection fraction at six months, no difference in clinical outcomes, and higher costs than medical therapy. These results must be interpreted with caution given the small size and low risk of the population.

See page 2177 for the editorial comment on this article (doi:10.1016/j.ehj.2004.10.002)

Keywords Acute myocardial infarction; Recanalization; Percutaneous transluminal coronary angioplasty


In acute myocardial infarction (MI), the benefit of early opening of the infarct-related artery in reducing mortality is established. To be effective, treatment must be initiated within 12 h of symptom onset, but patients often seek medical attention too late and either do not receive reperfusion therapy or reperfusion therapy fails to successfully recanalize the artery. Experimental and observational clinical data suggest that late opening of the infarct vessel, beyond the time window compatible with myocardial salvage, may be beneficial through other mechanisms.1–6 In particular, the contrast between the magnitude of the survival benefit provided by early reperfusion, and the mild-to-moderate improvement in late ventricular function and limitation of infarct size, suggests that patency of the infarct vessel per se may play an important role. This is the core of the 'open artery hypothesis', delineated by Braunwald in the 1980s.7–10 While substantial experimental evidence supports the hypothesis,2,11–13 only four randomized clinical trials have been carried out enrolling a total of only 264 patients and with conflicting results.14–17 Therefore the DEsobstruction COronaire en Post-Infarctus (DECOPI) trial was set up to evaluate the clinical benefit of late re-establishment of coronary patency by percutaneous transluminal coronary angioplasty (PTCA) after MI, measured by a combined primary endpoint of cardiac death, non-fatal MI or ventricular tachyarrythmias. Additionally, the trial assessed its impact in terms of angiographic outcomes and cost.


Enrolment started in July 1998 at 16 university hospitals in France and Belgium and finished in December 2001. To be eligible for inclusion, patients had to be aged 20–75 years, have a first Q-wave MI, no spontaneous or low-level recurrent ischaemia (defined as ST depression >1 mV at <90 W workload, and/or a fall in blood pressure, and/or ventricular arrhythmia on exertion occurring despite anti-anginal therapy), and angiographic demonstration of total occlusion of the infarct-related artery (TIMI grade 0–1 flow) located on a proximal segment with a reference luminal of at least 2.0 mm on an angiogram performed at least 48 h after symptom onset.

The exclusion criteria were spontaneous or low-level ischaemia, left main coronary artery stenosis >50%, coronary lesions requiring coronary artery bypass grafting (CABG) or technically contraindicated for PTCA in the judgment of the investigator, life-threatening co-morbidity, allergy to aspirin or thienopyridine, women of childbearing potential, and patients in whom long-term follow-up was uncertain. Written informed consent was obtained from all patients before randomization. Standard post-MI therapy was recommended in both groups. Patients were then randomized to either PTCA or medical therapy.

Patients randomized to revascularization were scheduled to undergo PTCA, which could be performed immediately or within a few days but no later than 15 days after symptom onset even if spontaneous recanalization had occurred between the 'qualifying angiogram' and the procedure. The use of stents for PTCA was encouraged. PTCA success was defined as TIMI flow grade 3 at the end of PTCA with a residual stenosis of <50%. Glycoprotein IIb/IIIa antagonists were given at the physician's discretion, and a thienopyridine was administered for ⩾1 month if a stent was placed. Creatine kinase measurements were collected 12 and 24 h after PTCA. Patients randomized to medical therapy were given standard treatment for MI.

Revascularization by PTCA or CABG of the infarct-related artery was strongly discouraged but was allowed in patients who, despite proper anti-anginal therapy, had evidence of recurrent ischaemia (i.e., dynamic ST-segment or T-wave changes or ischaemia on a stress test).

Follow-up at six months included a stress test performed on medical therapy, and coronary angiography with contrast left ventriculography. Subsequent follow-up comprised clinical evaluation using standardized questionnaires and telephone interviews every six months for up to two years and yearly thereafter.

The primary endpoint was the incidence over time of the composite endpoint of cardiac death, non-fatal MI (using WHO criteria18), or ventricular tachyarrhythmia (ventricular fibrillation or tachycardia >30 s, or of shorter duration but associated with syncope). The principal secondary endpoint was the combination of the primary endpoint plus hospital admission for heart failure. Costs were computed from the viewpoint of the healthcare system over the duration of follow-up. Volume of resources was drawn from the case report form and was collected prospectively. Unit costs for hospital resources (bed days, drugs, blood products and any invasive or costly procedure/treatment) were estimated from the accounting system (not from charges) and ambulatory costs from the payers' reimbursement schedule. The study protocol was approved by the ethics committee of Hôpital Pitié Salpétrière.

Determination of sample size

Assuming a relative reduction of 33% in the combined incidence of the primary endpoint by PTCA over medical therapy and an event rate of 15% in the medical arm, 720 patients would be required to demonstrate a significant difference between groups (α=5%, β=10%, two-tailed test). However, owing to limited funding and lagging enrolment, an interim analysis was performed based on 14 June 2001 as the reference date for analysis. A total of 200 patients had been randomized, with a total of 14 patients fulfilling the criteria for the primary endpoint, six in the medical arm and eight in the PCTA arm. Based on a Bayesian analysis of these data, it was computed that there was a less than 0.001 chance of demonstrating any benefit from PTCA over medical therapy with the completed sample size. In December 2001, in the light of these results and owing to the difficulties in recruitment, the Data Safety and Monitoring Board recommended stopping enrolment and completing follow-up.

Statistical analysis

Data were analysed centrally in a blinded fashion. The effect of PTCA over the medical-treatment group on the distribution of time to event (non-fatal MI, the primary endpoint, admission for heart failure, total mortality, MI, any admission, or revascularization) was estimated using Cox models in which only the randomized group was introduced as a binary covariate; accordingly, a non-adjusted hazard ratio (HR) was used as the measure of treatment benefit. Proportional hazards assumption was checked using time-varying co-efficients. Cardiovascular deaths were considered in a competing risks framework, and analysed through the use of cumulative incidence curves and the Gray test.19 Continuous (left ventricular ejection fractions, LVEFs) and binary (six-month angiographic outcomes) endpoints were compared between randomized groups using the non-parametric Wilcoxon sum rank test or Fisher's exact test, respectively. Differences from baseline values were tested by the non-parametric Wilcoxon signed-ranks test. All P-values were two-sided. Analysis was performed using the SAS 8.2 (SAS Inc, Cary, NC, USA) and Splus2000 (MathSof, Inc, Seattle, WC, USA) software packages.


Overall, 212 patients were enrolled in the study; 109 were randomized to PTCA and 103 to medical treatment (Fig. 1). Four patients in the medical arm withdrew their consent after randomization, three of whom were treated medically. Two patients in each arm had major protocol deviations. The results are reported by intention to treat.

Fig. 1

DECOPI study profile. PTCA, percutaneous transluminal coronary angioplasty.

Baseline characteristics were similar in the two treatment arms (Table 1). Diabetes was slightly but not statistically significantly more common in patients who underwent PTCA (10.9% vs 19.3%, P=not significant). Two-thirds of patients had single-vessel disease and the left anterior descending (LAD) artery was involved in less than one third of patients.

View this table:

Table 1. Patient baseline characteristics

N (%) Median [Q1–Q3]Medical therapy (n=103)PTCA (n=109)
Age (years)56 [50–66]58 [50–66]
Men85 (82.5)95 (87.2)
Diabetes11 (10.9)21 (19.3)
Current or previous smoker71 (70.3)82 (75.2)
Hypertension33 (32.7)33 (30.3)
Hypercholesterolaemia47 (47.0)52 (47.7)
Family history of coronary artery disease37 (36.6)33 (30.3)
Delay between myocardial infarction and randomization (days)6 [4–9]7 [4–9]
Anterior myocardial infarction30 (29.4)29 (26.6)
Treatment with thrombolysis17 (17.0)10 (9.2)

Angiographic characteristics
Delay between myocardial infarction and qualifying angiogram (days)5 [4–7]5 [4–8]
Single-vessel disease63 (62.4)77 (70.6)
Two-vessel disease32 (31.7)23 (21.1)
Three-vessel disease6 (5.9)9 (8.3)
Baseline left ventricular ejection fraction50 [44–58]51 [45–58]
TIMI 0 flow in infarct-related artery88 (87.1)91 (83.5)
TIMI 1 flow in infarct-related artery13 (12.9)18 (16.5)

Infarct-related artery
– Left anterior descending30 (29.7)29 (26.6)
– Right coronary artery48 (47.5)64 (58.7)
– Left circumflex23 (22.8)16 (14.7)

PTCA, percutaneous transluminal coronary angioplasty; Q1, first quartile; Q3, third quartile.

Angioplasty of the infarct artery was performed at a median of eight (Q1=5; Q3=11) days after the qualifying MI. Glycoprotein IIb/IIIa antagonists were used in 9.4% of the patients and 80.4% of the patients underwent placement of at least one stent. TIMI 3 flow in the infarct artery was achieved in 82.2% of the patients, TIMI 2 in 4.7%, and 13.1% had TIMI 0–1 flow. With regard to creatinine kinase, 16.7% of the patients had values above the upper limit of normal after PTCA, and three patients (2.8%) had values >3 the upper limit of normal.

At six months, 166 (78%) patients underwent repeated angiography. In these patients, LVEF was higher in the invasive (median 59%) compared with the medical group (median 54%; P=0.0128). Patency of the infarct artery was higher in the PTCA compared with the medical therapy group (82.8% vs 34.2%, P<0.0001, Table 2). The 34.2% patency rate in the medical therapy group was the result of successful crossover angioplasty in 8.9% and spontaneous recanalization in 25.3%. Likewise, the 82.8% patency rate in the PTCA group was the net result of persistent occlusion after failed baseline PTCA in 4.6% of the patients and reocclusion of the infarct vessel in 12.6%. In the 94 patients with both measurements and in whom the infarct artery was patent at six months, LVEF increased by 5.7±10.1% (from 52.4±11.3% to 58.1±12.1%; P<0.0001 by the Wilcoxon signed-ranks test), whereas there was no significant increase in the 62 patients with a closed artery at six months (baseline value 49.9±10.3%, six-month value 51.5±13.3%; P=0.16) (Table 3). Patients with a patent infarct artery at six months experienced a similar increase in LVEF regardless of the randomization assignment (medical 5.1±9.6% vs angioplasty 5.0±10.4%; P=0.99 by the Wilcoxon rank sum test) (Table 3). Likewise, patients with an occluded infarct artery at six months had a similar and much lower increase in LVEF in both treatment arms (medical 1.5±8.9% vs angioplasty 1.8±10.6%; P=0.38) (Table 3). At six months, restenosis of the infarct artery (>50% stenosis) was seen in 49.4% of the patients in the PTCA group, but target vessel revascularization in 26.1%.

View this table:

Table 2. Angiographic outcomes at six months follow-up

Characteristics N (%)Medical therapy (n=103)PTCA (n=109)P-value
Not assessed patientsN=24N=220.59
– Early death (within the first 6 months)46
– Non-assessed patients2016

Angiogram performed N=79N=87
Patent vessel27 (34.2)72 (82.8)<0.0001
Crossover angioplasty (successful)7 (8.9)NA
Spontaneous recanalisation20 (25.3%)NA
Persistent occlusion52 (65.8)NA
Failed initial PTCA, persistent occlusionNA4 (4.6)
Re-occlusionNA11 (12.6)
Left ventricular ejection fraction54 (42.5–62.5)59 (50–67)0.0128

NA, not applicable; PTCA, percutaneous transluminal coronary angioplasty.

View this table:

Table 3. Comparison of angiographic left ventricular ejection fraction changes according to randomization and patency of the infarct-related artery at six-month angiography in patients with both measurements

Randomised armMedical therapyPTCATotal
Mean±standard deviation6-month patent6-month occluded6-month patent6-month occluded6-month patent6-month occluded
a Wilcoxon signed-ranks test. LVEF: left ventricular ejection fraction.
Number of pairs265068129462
Month 0 LVEF, %50.4±13.049.8±11.153.2±10.650.4±6.452.4±11.349.9±10.3
Month 6 LVEF, %55.5±13.251.3±13.559.2±11.652.2±12.858.1±12.151.5±13.3
Absolute variation from month 05.1±9.61.5±8.95.0±10.41.8±10.65.7±10.11.6±9.2

The use of pharmacological therapy during follow-up was high and was similar in both groups: after two years, 83.3% of patients were on aspirin, 81.0% on beta-blockers, 57.8% on angiotensin-converting enzyme inhibitors, and 82.0% on statins. After a mean follow-up of 34 months, the primary endpoint was similar in both groups, occurring in nine (8.7%) and eight (7.3%) patients from the medical and PTCA groups, respectively (P=0.68; Table 4). The three-year cumulative incidence of cardiovascular death was estimated at 5.6% in the PTCA group vs 8.2% in the medical group (P=0.63). Conversely, there were five non-fatal MIs, namely three (2.7%) in the intervention group and two (1.9%) in the medical group (P=0.70). None of the patients developed non-fatal, non-MI-related (i.e., within 28 days of MI) ventricular tachyarrhythmias.

View this table:

Table 4. Patients (crude ratio) fulfilling criteria of primary and secondary endpoints

N (%) median [Q1–Q3]Medical therapy (n=103)PTCA (n=109)Measure of treatment differencea (95% CI)P-value
a I.e. hazard ratio (HR).
b Heart failure, angina, or arrhythmia.
c Hierarchical combined endpoint: one patient with a validated non-fatal MI died of cardiovascular cause several months later.
Follow-up (months)35 [23–47]36 [26–48]

Primary endpoints
Cardiovascular deaths7 (6.8)6 (5.5)0.765 (0.255–2.295)0.63
Non-fatal myocardial infarction2 (1.9)3 (2.7)1.414 (0.236–8.464)0.70
Non-fatal, non-MI-related VT/VF00
Primary endpointc9 (8.7)8 (7.3)0.82 (0.316–2.127)0.68

Secondary endpoint
Primary endpoint+admissions for heart failure13 (12.6)11 (10.1)0.787 (0.352–1.756)0.56

Other endpoints
Total mortality9 (8.7)8 (7.3)0.773 (0.298–2.006)0.60
All myocardial infarctions3 (2.9)4 (3.7)1.233 (0.276–5.512)0.78
Any sustained VT/VF1 (0.97)1 (0.92)0.837 (0.052–13.437)0.90
Admissions for cardiac causesb26 (25.2)17 (15.6)0.595 (0.323–1.097)0.096
Revascularization33 (32.0)30 (27.5)0.868 (0.530–1.424)0.57
Admissions for heart failure5 (4.8)3 (2.7)0.561 (0.134–2.349)0.43
Ischaemia at six-month stress test16 (21.6)17 (21.5)0.995 (0.544–1.822)1.00

PTCA, percutaneous transluminal coronary angioplasty; VT, ventricular tachycardia; VF, ventricular fibrillation.

The probability of survival did not differ between the two groups (HR 0.77; 95% CI 0.30–2.00; P=0.60; Fig. 2). The hospital admission rate for heart failure (4.8% vs 2.7%) and the primary endpoint plus admission for heart failure (12.6% vs 10.1%) were non-significantly increased in the medical therapy group compared with the intervention group (Table 4). The rate of all ventricular tachyarrhythmias was <1% in both groups. Functional status at the end of follow-up did not differ significantly between the groups; slightly higher proportions of patients in the medical group had NYHA class >I dyspnoea (26.6%) and Canadian Cardiovascular Society class >1 angina (10.6%) compared with the PTCA group (18.8% and 7.0%, respectively; P=not significant for both). The proportion of patients with ischaemia at the six-month stress test was identical in both groups (Table 4).

Fig. 2

Estimated probability of infarct-free, ventricular tachycardia-free survival, using Kaplan–Meier estimator.

The median lengths of stay in the coronary care unit and in the hospital were similar in both groups (Table 5). The rate of re-admission for any cause for patients who received medical therapy was increased compared with the intervention arm (49.5% vs 38.5%) but the difference was not significant. The initial mean cost was higher for PTCA than medical therapy (with a mean difference of €2225 per patient; 95% CI 1185–3264; P<0.0001), with no difference in follow-up costs, resulting in higher overall costs for the invasive approach (mean difference of €2406 per patient; 95% CI 780.7–4032; P<0.0001).

View this table:

Table 5. Cost analysis

N (%) median [Q1–Q3]Medical therapy (n=103)PTCA (n=109)Measure of treatment differencea(95% CI)P-value
a Either hazard ratio or difference in means.
Median length of stay in CCU (days)4 [2–5]4 [3–6]0.927 (0.108–1.745)0.01
Median length of hospitalisation (days)10 [8–13]11 [8–14.5]0.403 (−1.9 to +2.7)0.10
Any readmission51 (49.5)42 (38.5)0.748 (0.497–1.125)0.16

Treatment costs (n=100)(n=108)
Initial cost, median (€)6511 [5391–8805]9266.5 [7483–11435]2225 (1185–3264)<0.0001
Follow-up cost, median (€)3612.5 [2969–4865]3675 [2910–5132]286.4 (−951.42 to +1524)0.80
Total cost, median (€)10,809.5 [8632–15302]13,484 [11350–17460]2406 (780.7–4032)<0.0001

CCU, coronary care unit; PTCA, percutaneous transluminal coronary angioplasty.

In a post-hoc analysis, patients surviving up to six months and who underwent repeat coronary angiography were categorised on the basis of six-month patency of the infarct artery, and angiographic and clinical outcomes, censored for the first six months were analysed over the total duration of follow-up (Table 6). In this analysis, total (1.0% vs 9.1%; P=0.032) and cardiovascular mortality from six-month angiography and beyond (0% vs 6.1%; P=0.011) were significantly lower in patients with a patent compared to an occluded infarct vessel (Table 6). Likewise LVEF was higher in the patent group. Finally, there was a non-significant trend towards a lower incidence of the primary endpoint in the group with a patent vessel.

View this table:

Table 6. Post-hoc analysis in patients surviving up to six months after hospital discharge and who underwent repeat coronary angiography

N (%) median [Q1–Q3]Occluded IRA (n=66)Patent IRA (n=100)Measure of treatment differencea (95% CI)P-value
a Either hazard ratio or difference in means.
LVEF, median (Q1–Q3)53 (44–60)59 (49–57)5.53 (1.50–9.55)0.004
All-cause mortality (%)6 (9.1)1 (1.0)0.099 (0.012–0.822)0.032
Cardiovascular death (%)4 (6.1)00.01
Primary endpoint (%)4 (6.1)2 (2.0)0.315 (0.058–1.723)0.18

IRA, infarct-related artery; LVEF, left ventricular ejection fraction.


This is the largest single randomized trial of late re-opening of the occluded artery following MI. In contrast to earlier studies, patients received optimal medical therapy and 80% of patients undergoing PTCA received a stent, although only around 10% of these received glycoprotein IIb/IIIa antagonists. While patency of the infarct artery was seen in over 80% of patients in the angioplasty arm and only 40% in the medical arm, this did not translate into statistically significant differences in clinical outcomes. Furthermore, overall costs were higher in the angioplasty group (although this finding may be less relevant in countries with a shorter length of hospitalisation such as the United States). The study demonstrated a benefit in terms of left ventricular function of 5% at six months in favour of the PTCA arm, which would be expected to translate into clinical benefit in a larger population followed for a longer time. However, in this cohort, possibly because of the high use of evidence-based therapies, clinical outcomes were excellent in both study arms.

This low risk is also possibly related to investigator bias at enrolment: baseline LVEF was 50%, two-thirds of patients had single-vessel disease, and the LAD was the target vessel in <30% of patients (although that rate is similar to that seen in pre-trial screening data from the ongoing Open Artery Trial [OAT]).9 Indeed, the primary endpoint occurred at an annualized rate of only 3.1% in the medical arm. Spontaneous recanalization within six months occurred in almost one-third of medically treated patients. By contrast, in the PTCA group failure to recanalize occurred in 13% of patients, re-occlusion within six months in an additional 12%, and restenosis in 49%. Despite these differences, at six-month follow-up, the median LVEF was superior in the PTCA group compared with the medical group by 5%, and there were trends for reduced dyspnoea and cardiac hospitalisation in the PTCA group. In a hypothesis-generating post hoc analysis, we found that patency of the infarct-related artery at six months – seen in 83% of patients in the angioplasty arm and only 40% of those in the medical arm – was a strong predictor of beneficial outcome in terms of reduced mortality and improved ejection fraction. Also, paired analyses of LVEF as a function of patency at six months concur in demonstrating that persistent patency is crucial in determining LVEF changes.

The rate of restenosis (49%) was high, but this study was not designed to examine this issue. Absolute rates of restenosis must be interpreted as a function of the definition applied.20 For the sake of simplicity, we used a common and practical definition but also a sensitive definition (residual stenosis >50% of reference diameter). The target vessel revascularization rate (a clinically relevant index of restenosis) was 26.1% in the PTCA arm, which is in line with the expected rate in the 'uncoated stent' era.

The four previously available randomized trials of recanalization of occluded infarct artery were much smaller than DECOPI and yielded conflicting and inconclusive results.9,14–17 These trials found a substantial rate of spontaneous late recanalization of the infarct artery in the conservative arm and highlighted the problem of re-occlusion in the angioplasty arm. One trial, TOAT (The Open Artery Trial),15 found somewhat unexpectedly that recanalization of occluded arteries approximately one month after acute MI adversely affected remodelling and was associated with and increased risk of an adverse clinical event. In contrast, neither the trial by Horie17 nor the DECOPI trial found evidence of worse outcomes in the angioplasty arm, and in the trial by Horie17 there was no adverse impact of angioplasty on left ventricular remodelling. In fact, both the end-diastolic and the end-systolic left ventricular volumes were smaller in the PTCA compared with the no-PTCA group. A recent randomized trial in stable single vessel disease patients after MI21 has suggested a long-term survival benefit of routine PTCA, but in that trial over two-thirds of the patients had a patent infarct-related artery at baseline and therefore differ substantially from the DECOPI population.

Because recruitment and event rates were lower than planned, the study is markedly underpowered, and the wide confidence intervals for the primary endpoint show that there could be real benefit (or harm) of PTCA. In fact, there are non-significant trends for better outcomes in the PTCA arm of the trial in terms of hospital admissions for cardiac causes and angiographic outcomes. Although the proportion of patients with angiographic follow-up were similar between groups, 22% of the patients did not undergo repeat coronary angiography and left ventricular angiography at six months, therefore the interpretation of follow-up patency and ejection fraction data should be cautious. In the post-hoc analysis an open infarct-related artery was associated with improved outcomes and ejection fraction. The relatively low rate of use of glycoprotein IIb/IIIa antagonists and the absence of use of drug-eluting stents (which were not available at the time of the study) suggest that improvements in PCI technique and adjunctive therapy may result in better immediate and six-month results than in the trial.

In conclusion, systematic late PTCA of the infarct vessel was associated, in this study, with a higher LVEF at six months, no difference in clinical outcomes, and higher costs than medical therapy. These results must be interpreted with caution given the small size and low risk of the population. Post hoc analyses of clinical and angiographic data suggest the benefit of an open artery. Furthermore, recent advances in PCI technique such as drug-eluting stents may help to increase the patency rates and reduce the re-occlusion or restenosis rate. Whether a strategy of routine PTCA would be beneficial in a higher risk population is currently being tested in a large international mortality trial: the OAT trial, which plans to enrol 3200 patients at high risk because of an impaired ejection fraction or proximal occlusion of a large artery.

DECOPI Study Group

Principal investigator: Ph.G. Steg

Co-ordinator: L. Brucker

Steering Committee: J.P. Bassand, C. Chastang, A. Cohen, N. Danchin, P. Guéret, D. Himbert, J.P. Monassier, J. Puel, A. Vahanian

Critical Event Committee: C. Bauters, O. Dubourg, R. Frank, S. Weber

Data Safety and Monitoring Board: M. Bertrand, R. Gourgon, M. Komajda

Statistical analysis: S. Chevret, R. Porcher, S. Gourdain

Economic analysis: I. Durand-Zaleski

F. Albert, P. Allouch, S. Alsagheer, N. Amabile, S. Battaglia, H. Benamer, M. Benmaklouf, J.P. Bertinchant, F. Boccara, J. Boschat, K. Bougrini, D. Carrié, S. Cattan, P. Cazaux, S. Champagne, B. Charbonnier, B. Citron, D. Coisne, J.L. Colas, J.-P. Collet, P. Couppié, P.D. Crochet, S. Dinanian, C. D'Ivernois, E. Donal, X. Dujardin, J.P. Elkaïm, M. Etchegoin, L.J. Feldman, G. Finet, E. Garbarz, P. Hacot, M. Hanssen, P. Henry, L. Jacquemin, M. Jean, K. Khalifé, A. Lafont, B. Ledermann, F. Ledru, A.Leenhardt, D.Logeart, J.Machecourt, M.Madjoub, L.Maillard, J.P.Maroni, N.Meneveau, G. Montalescot, J.M. Montély, F. Paganelli, G. Rangé, G. Rioufol, R.M. Rodriguez, C. Saunier, L. Schiro, M. Slama, C. Spaulding, F. Tarragano, E. Teiger, C. Thuaire, A. Tirouvanziam, W. Wijns.


This study was supported by Caisse Nationale d'Assurance Maladie des Travailleurs Salariés, Paris; the Fondation pour la Recherche Médicale, Paris; the Programme Hospitalier de Recherche Clinique/DRC AP-HP; and an unrestricted grant from Guidant-ACS, France.


  • Funding: This study was supported by Caisse Nationale d'Assurance Maladie des Travailleurs Salariés, Paris; the Fondation pour la Recherche Médicale, Paris; the Programme Hospitalier de Recherche Clinique/DRC AP-HP; and an unrestricted grant from Guidant-ACS, France.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
View Abstract