Smokers With ST-Segment Elevation Myocardial Infarction and Short Time to Treatment Have Equal Effects of PCI and Fibrinolysis


Thomas Rasmussen, MD1, Henning Kelbæk, MD1, Jan Kyst Madsen, MD2, Per Thayssen, MD3, Klaus Rasmussen, MD4, Leif Thuesen, MD4, Lars Køber, MD1

Abstract: Objectives. The purpose of this study was to examine the effect of primary percutaneous coronary intervention (PCI) compared to fibrinolysis in smokers and non-smokers with ST-segment elevation myocardial infarction (STEMI). Smokers seem to have less atherosclerosis but are more prone to thrombotic disease. Compared to non-smokers, they have higher rates of early, complete reperfusion when treated with fibrinolysis for MI. Methods and Results. In the Second Danish Multicenter Trial in Acute Myocardial Infarction (DANAMI-2), a total of 1572 patients with STEMI were randomized to either fibrinolysis or PCI (1129 patients were enrolled at 24 referral hospitals and 443 patients at 5 invasive treatment centers). The primary endpoint for this substudy was death by any cause. Secondary endpoints were a composite of death by any cause, clinical re-infarction or disabling stroke. Follow-up was 3 years. The effect of PCI is reported according to time to treatment and smoking status. Data on smoking habits were available for 1534 patients (895 smokers and 639 non-smokers). Smokers with short time to treatment (<3 hours) benefited equally from PCI and fibrinolysis with a trend toward higher mortality in the PCI group (mortality [hazard ratio, 1.64 (0.79-3.41); P=.18], composite endpoint [hazard ratio, 1.06 (0.65-1.71); P=.82]). In non-smokers with short time to treatment PCI was superior to fibrinolysis (mortality [hazard ratio, 0.46 (0.22-0.93); P=.02], combined endpoint [hazard ratio, 0.45 (0.26- 0.79); P=.004]). Patients with >3 hours to treatment all showed a tendency toward a superior effect of PCI irrespective of smoking habits. Conclusions. PCI and fibrinolysis are equally beneficial in smokers with STEMI and short time to treatment.

J INVASIVE CARDIOL 2012;24(8):401-406

Key words: percutaneous coronary intervention, ST-segment elevation myocardial infarction, smoking, fibrinolysis


Primary percutaneous coronary intervention (PCI) has shown to be superior to fibrinolytic therapy in decreasing major adverse cardiovascular events after ST-segment elevation myocardial infarction (STEMI). This has been demonstrated in patients admitted directly to hospitals with angioplasty facilities and in those who were admitted to referral hospitals and immediately transferred to invasive centers. Therefore, PCI is now the gold standard treatment of patients with STEMI.1-7 Various factors around the world, though, challenge the dilemma that an invasive strategy is not feasible in all patients with STEMI. The invasive strategy requires a facility and operators for PCI who conduct enough procedures annually to remain proficient and these conditions are not present throughout the world. In rural regions, which lack invasive centers or have long distances to such facilities, an invasive approach is limited or even impossible.6,7

In a community where a complete invasive strategy for every patient with STEMI is not feasible, identifying patients most likely to benefit from an invasive approach is very important. Risk stratification of patients at admission is one way to differentiate patients. An earlier study (DANAMI-2) used the Thombolysis in Myocardial Infarction (TIMI) risk score to identify a group of high-risk patients with significantly reduced mortality with primary angioplasty.8 There might still be other groups of patients who could benefit more from PCI than others compared to fibrinolysis.

Smoking is known as a major risk factor in the development of acute myocardial infarction (MI). The mechanism by which smokers develop MI seems to be different to non-smokers. The infarct lesions often have a greater thrombotic component with relatively less atherosclerotic burden,9-11 and smokers are younger when they develop their MI.9-15 They appear to have less multi-vessel disease and fewer infarctions are located in the territory of the left anterior descending coronary artery.9 Active smokers seem to be in a state of relative hypercoagulation with increased hematocrit and fibrinogen levels, impaired endothelial function, a tendency to vasospasm, and enhanced production of platelet thrombus. All these factors predispose smokers to the formation of thrombus in the setting of MI.9,10 Randomized trials have shown that smokers with MI have higher rates of early, complete reperfusion when treated with fibrinolysis than non-smokers.9,10,14 This suggests that smokers could benefit relatively more from fibrinolysis than non-smokers. This has, to our knowledge, not been well examined, and therefore our goal was to investigate whether an interaction between smoking status and the effect of primary PCI compared to fibrinolysis in patients with STEMI could be found. In the present study, we report a 3-year clinical outcome for smokers vs non-smokers in the Second Danish Multicenter Trial in Acute Myocardial Infarction (DANAMI-2) after STEMI randomized to either PCI or fibrinolysis.


The present study included 1572 patients from the DANAMI-2 multi-center study, which was a randomized trial to determine whether patients with STEMI could benefit from PCI compared to fibrinolysis. The DANAMI-2 study has been described in detail previously.1,2,16 In brief, a total of 4278 STEMI patients were screened for enrollment in the study, with 1572 patients randomized to either PCI or fibrinolysis with intravenous alteplase. Of these, 1129 patients were enrolled at 24 referral hospitals without angioplasty capability and 443 patients at 5 invasive centers.

The study was approved by the National Ethics Committee of Denmark and all patients provided written informed consent.

Criteria for eligibility. In brief, the criteria for inclusion were age ≥18 years, presence of symptoms for >30 minutes but <12 hours, and a cumulative ST-segment elevation a minimum of 4 mm in at least two contiguous leads.

The criteria for exclusion were contraindications to fibrinolysis, left bundle branch block, MI or fibrinolysis in the last 30 days, pulseless femoral arteries, previous coronary artery bypass graft (CABG), known renal failure (defined as a serum creatinine above 2.83 mg/dL or 250 µmol/L), diabetes treated with metformin, non-ischemic heart disease, non-cardiac disease with life expectancy <12 months and high risk during transfer because of cardiogenic shock (severe heart failure with persistent systolic blood pressure ≤65 mm Hg), sustained life-threatening arrhythmia, or a need for mechanic ventilation.

Smoking status. Smoking status was classified according to self-report. Patients who were currently smoking at the time of admission were considered smokers and patients who had never smoked or ceased smoking previously were considered non-smokers.

Treatment. A detailed description of the revascularization treatment has been published previously.1 The fibrinolysis group was treated with aspirin, b-blocker, accelerated plasminogen activator (alteplase), and intravenously unfractionated heparin. Alteplase was administered as a bolus of 15 mg followed by infusion with 0.75 mg/kg over 30 minutes and subsequent infusion of 0.5 mg/kg over 60 minutes.

Angioplasty was preceded by treatment with aspirin, b-blocker, and intravenous unfractionated heparin. Platelet glycoprotein IIb/IIIa administration was left to the discretion of the treating physician. Stenting of the culprit lesion was attempted in all patients unless the vascular diameter was <2.0 mm. All stents were bare-metal stents. After stent implantation, ticlopidine or clopidogrel was administered for 1 month according to the guidelines at the time of the trial. Judgment of the angiograms was carried out by an independent core laboratory (Cardialysis).

Endpoints and definitions. In the present paper, the primary endpoint was death by any cause within 3 years. Secondary endpoints were a composite of death by any cause, clinical re-infarction, or disabling stroke within 3 years. A procedure-related re-infarction was not included as a secondary endpoint.

Detailed definitions of the endpoints are available elsewhere.1 A re-infarction was defined as an increase in creatine kinase and MB isoenzyme activity and either symptoms of chest pain or presence of electrocardiographic changes. A clinical re-infarction was defined as an increase in the creatine kinase MB in a patient with normalized levels after the index infarction or as an increase of at least 50% compared to the previous level. Disabling stroke was defined as a fatal stroke or a stroke causing a clinically significant mental or physical disability, ranging from slight handicap (ie, patients could not engage in all previous activities, but could still take care of himself or herself) to severe (ie, patients in a bedridden state with requirement for constant care and nursing).2

Statistical analysis. All analyses were based on intention-to-treat. Categorical variables are expressed as percentages and continuous variables are reported as medians and interquartile ranges or mean ± standard deviation. Differences in baseline characteristics for smokers and non-smokers were assessed using chi-square test for discrete variables and Kruskal-Wallis test for continuous variables. A P-value <.05 was considered statistically significant. Median time to treatment was 3 hours, and this was prespecified as the cut-off for subgroup analyses of benefit from treatment. Cumulated event rates were illustrated with the use of Kaplan-Meier curves. The curves were compared for differences with the log-rank test. Hazard ratios were analyzed with Cox regression analysis. Interaction analyses were performed by the Cox proportional hazard model including interaction variables. As patients were randomized to angioplasty or fibrinolysis, multi-variable adjustments were not performed when the effect of treatment allocation was tested. All calculations were performed using SAS, version 9.1 (SAS Institute).

Post new comment

  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

More information about formatting options

Enter the characters shown in the image.