Author + information
- Received June 21, 2000
- Revision received November 20, 2000
- Accepted January 12, 2001
- Published online May 1, 2001.
- Michael Mundt Ottesen, MD∗,* (, )
- Lars Køber, MD, DMSc∗,
- Stig Jørgensen, MD†,
- Christian Torp-Pedersen, MD, DMSc, FACC∗,
- on behalf of the TRACE Study Group
- ↵*Reprint requests and correspondence: Dr. Michael Ottesen, Department of Cardiology P, Projektgruppen, Gentofte University Hospital of Copenhagen, DK-2900, Hellerup, Denmark
The aim of this study was to evaluate the consequences, measured as mortality and in-hospital stroke, of the use of thrombolytic therapy among patients with acute myocardial infarction (AMI), who do not fulfill accepted criteria or who have contraindications to thrombolytic therapy (i.e., overutilization) and among patients who are withheld thrombolytic treatment despite fulfilling indications and having no contraindications (i.e., underutilization).
The implementation of treatment with thrombolysis in clinical practice is not in accordance with the accepted criteria from randomized studies. The consequence has been over- and underutilization of thrombolytic therapy among patients with AMI in clinical practice. The outcome of overutilization of thrombolytic therapy has not been described previously.
We examined 6,676 consecutive patients admitted to the hospital with an AMI and recorded characteristics, in-hospital complications and long-term mortality.
Overall, 41% of the patients received thrombolytic therapy. Thrombolytic therapy was underutilized in 14.3% and overutilized in 12.9% of the patients. The use of thrombolytic therapy was associated with reduced mortality in every subgroup examined, including patients without an accepted indication, with an accepted indication and in patients with prior stroke. The risk ratio of in-hospital stroke was not increased in connection with thrombolytic therapy, not even in patients with prior stroke (relative risk = 0.237, 95% confidence interval: 0.031 to 1.810, p = 0.17).
With the large benefit known to be associated with thrombolytic therapy and the favorable result of thrombolytic therapy in patients with contraindications observed in this study, we conclude that a formal evaluation of thrombolytic therapy in wider patient categories is warranted.
Intravenous thrombolytic therapy has received worldwide acceptance as one of the cornerstones in the treatment of patients with acute myocardial infarction (AMI). In randomized studies, thrombolytic therapy alone is shown to reduce mortality by approximately 25% and combined with aspirin is shown to reduce mortality by 42% (1).
There is a consensus that patients presenting within 12 h of onset of symptoms with ST-segment elevation or bundle branch block (2)will benefit from thrombolytic therapy and that patients not fulfilling these criteria may even be harmed by the treatment. A substantial number of patients who fulfill the accepted indications do not receive the treatment due to contraindications. The contraindications originate from clinical trials that established the therapy and where safety was of primary concern. The rationale for the contraindications was based on controversial evidence, empiricism, rational considerations and experience, but not on earlier clinical trials. Nonetheless, the contraindications are accepted by official guidelines (3,4).
In this situation, clinical practice will necessarily demonstrate the appropriate use, underutilization and overutilization of thrombolytic therapy. Underutilization has received much attention (5–7), but the consequences of giving thrombolytic therapy to patients who either do not fulfill accepted indications or who have accepted contraindications have not received attention. We have, therefore, found it important to perform a comprehensive comparison of the consequences of underutilization and overutilization of thrombolytic therapy in a large series of consecutive patients admitted to the hospital alive with an AMI. The analyses will focus on survival and in-hospital stroke.
The study population consisted of consecutive patients over 18 years old and admitted with an AMI to one of 27 Danish coronary care units from 1990 to 1992. All study departments had complete regional uptake. The presence of an AMI required chest discomfort or electrocardiographic changes suggestive of infarction or ischemia and was accompanied by an increase in cardiac enzymes to twice the upper normal value of the local hospital laboratory. Data were collected prospectively as part of the screening procedure for the TRAndolapril Cardiac Evaluation study (TRACE) (8), which was approved by the regional ethical committees of Denmark.
A medical history was obtained for every patient, and an echocardiography was recorded on videotape for evaluation of left ventricular systolic function, estimated as wall motion index. In-hospital complications, including stroke and survival, were systematically recorded for all patients.
Data concerning electrocardiograms (ECG) were collected by the study personnel as a summary of ECG on days 1 to 3. ST-segment elevations were defined as at least one or more ST-segment elevations of 2 mm in chest leads or 1 mm in limb leads. ST-segment depressions were defined as at least one or more ST-segment depressions of 1 mm in chest or limb leads.
No strict criteria for thrombolytic therapy were applied at the time of this study. Treatment with thrombolytic therapy was left to the judgment of the attending physician. The following indications were used, in general: admission <12 to 24 h from onset of symptoms consistent with AMI and new changes in the ECG (ST-segment elevation or bundle branch block). ST-segment depression or T-wave inversions represented relative criteria. Retrospectively, indications were categorized according to the recommendations given by the Fibrinolytic Therapy Trialist Collaborative Group (FTT). Indications for treatment with thrombolysis were admission within 12 h of onset of relevant clinical symptoms and ECG changes consisting of pathological ST-segment elevation or bundle branch block. Patients not having indications were marked as having “no indications.”
Conditions considered as contraindications were gastrointestinal ulcer, known anemia, pregnancy, malignant neoplasm or “other risk of bleeding” (i.e., known aortic aneurysm, recent operation or trauma, hemorrhagic diathesis and ongoing anticoagulation treatment), history of stroke and severe uncontrolled hypertension.
Mortality status of all patients was obtained on June 26, 1998. Survival data of 39 patients could not be obtained after discharge.
All tests of statistical significance were two-tailed, and p values of <0.05 were considered significant. Logistic multivariate regression analyses were performed to examine the association between baseline characteristics and risk of stroke or death at day 30 using backward selection procedure. Long-term mortality was examined using Cox proportional hazard regression analysis. All analyses were performed with SAS statistical package programs (Version 6.12, SAS Institute, Cary, North Carolina).
The study population consisted of 6,676 consecutive patients admitted to the hospital with an enzymatic-verified AMI. Due to missing data concerning thrombolytic therapy, 84 (1.3%) patients were excluded, leaving 6,592 patients for analyses. Figure 1illustrates a chart classifying the patients according to indications and contraindications for thrombolytic therapy and treatment with thrombolysis.
A total of 2,690 patients (40.8%) received thrombolytic therapy. Baseline characteristics according to indications for thrombolytic therapy or the presence of contraindications are listed in Table 1. The fibrinolytic agents used were streptokinase for 2,563 (95.3%) patients, alteplase for 89 (3.3%), while 38 (1.4%) patients received other fibrinolytic drugs.
Underutilization of thrombolytic therapy
An indication for thrombolytic therapy was present for 2,781 patients (42.2%) who were registered as having “no contraindications.” Of these patients, 1,839 (66.1%) received thrombolytic therapy (Table 2). Therefore, thrombolytic therapy was underutilized in 942 cases.
Overutilization of thrombolytic therapy
There were 3,811 (57.8%) patients who were registered as having “no indications” or at least one contraindication for the use of thrombolytic therapy. Of these patients, 851 received thrombolytic therapy. Thrombolytic therapy was, thus, overutilized in 851 patients. In principle, the 3,811 patients with “no indications,” or at least one contraindication for the use of thrombolytic therapy, can be divided into three group with dissimilar reasons for not receiving thrombolytic therapy: 1) 2,777 patients had “no indications”, 2) 414 patients had contraindications, and 3) 620 patients had “no indications” as well as at least one contraindication (Fig. 1). The most common reason for “no indication” was hospital admission later than 12 h after onset of symptoms, explaining 60.9% of such classifications. There were 1,850 patients (54.5%) who had no ST-segment elevations or bundle branch block, and symptoms were not consistent with infarction in 814 patients (24.0%).
Contraindications were infrequent (16%), but 20.9% of the patients with at least one contraindication were treated with thrombolysis. The most common reason for contraindication was a history of prior stroke, which was present in 540 (52.2%) patients, and the other reasons for contraindication were that 20.3% of the patients had cancer, 17.8% gastrointestinal ulcer, 9.1% were on anticoagulation treatment, 7.1% had high risk of bleeding and 4.8% had known anemia. There were 101 patients who had two contraindications, and eight patients had three contraindications. No cases of uncontrolled hypertension were registered.
Patients with prior stroke are assumed to have an excessive high risk of an in-hospital stroke in connection with thrombolytic therapy. Of 540 patients with prior stroke, 99 (18.3%) received thrombolytic therapy (Table 2).
Thrombolytic therapy and mortality
The odds ratio of mortality associated with thrombolytic therapy in univariate analysis was 0.502 (95% confidence interval: 0.466 to 0.540, p < 0.0001). Thirty-day mortality for patients receiving thrombolytic therapy versus those not treated was 7.9% and 15.9% (Fig. 2). Mortality of patients who received thrombolytic therapy and had no contraindication was 7.7%, while the mortality was 10.6% for those who received thrombolytic therapy despite contraindications (p = 0.12).
The independent relative risk of mortality at 30 days and four years was addressed in a multivariate Cox regression model with backward selection (Table 3). Parameters for the logistic regression analysis were chosen on the grounds of being of clinical importance or statistically significant in univariate analysis. Twenty-nine parameters were included in the logistic regression analysis: age, gender, hypertension, diabetes mellitus, atrial fibrillation, thrombolytic therapy, time from onset of symptoms until admission, Killip class ≥3, risk of bleeding (i.e., known aortic aneurysm, recent operation or trauma, hemorrhagic diathesis and ongoing anticoagulation treatment), chest pain on admission, ST-segment depression, ST-segment elevations without depression, bundle branch block, anemia, cancer, use of medication for gastrointestinal ulcer, current smoker, body mass index, anterior Q-wave infarction, time of day, wall motion index, congestive heart failure, ventricular fibrillation, ventricular tachycardia, weight, history of stroke, AMI, angina pectoris or gastrointestinal ulcer.
The analyses were performed in subgroups, and thrombolytic therapy was associated with a favorable outcome, both short- and long-term. In the subgroups where thrombolytic therapy was underutilized, 30-day and four-year mortality was 16.3% and 45.6%, respectively. Among patients where thrombolytic therapy was overutilized, 30-day and four-year mortality were 8.1% and 30.0%, respectively.
In-hospital stroke in relation to thrombolytic therapy
The in-hospital incidence of stroke among patients treated with thrombolytic therapy versus those not treated was 0.7% and 1.7%, respectively (p = 0.001). In-hospital stroke among patients where thrombolytic therapy was overutilized was 1.0% versus 1.7% for patients not treated (p = 0.26). Among patients where thrombolytic therapy was underutilized, the risk of stroke was 0.8% versus 1.0% when thrombolytic therapy was overutilized (p = 0.65).
Patients with a history of prior stroke had an in-hospital stroke incidence of 2.0% when thrombolytic therapy was used and 4.1% when thrombolytic therapy was not used (p = 0.49). Importantly, patients who received thrombolytic therapy had a more favorable risk profile (Table 2).
The independent risk ratio of in-hospital stroke in association with thrombolytic therapy was addressed in multivariate Cox regression models with backward elimination (Table 4); characteristics included are listed in Thrombolytic therapy and mortality section. There was no subgroup among which thrombolytic therapy was associated with a significant excess risk of developing an in-hospital stroke.
To our knowledge, this is the first study that addresses the consequences of overutilization of thrombolytic therapy in clinical practice. The study demonstrated that thrombolytic therapy was associated with a reduced risk of mortality without increasing the risk of stroke in patients either without standard indications or with accepted contraindications. Specifically, the risk of in-hospital stroke did not appear to be increased even in patients with a history of prior stroke.
Incidence of stroke associated with use of thrombolytic therapy
The findings we present are in line with a study from two nationwide surveys suggesting that patients with AMI and prior cerebrovascular events had adverse outcome with thrombolytic therapy due to older age and less favorable risk profiles, whereas thrombolytic therapy might be beneficial in selected patients with AMI and nonrecent prior cerebrovascular event (9). Even an increase in the incidence of nonfatal intracranial hemorrhage can be justified from a decision model derived from pooling of randomized trials of intravenous streptokinase. It is stated that an achieved mortality reduction by thrombolytic therapy can justify a higher rate of nonfatal intracranial hemorrhage (10). Among our patients treated with thrombolytic therapy, 0.8% had an in-hospital stroke as compared with 1.7% among those not treated with thrombolytic therapy. Comparable to our data are observations from the National Registry of Myocardial Infarction (11)where a stroke frequency of 0.9% among patients who received thrombolysis was found. In Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico-2 (12)and the Global Use of Strategies to Open Occluded Coronary Arteries-I study (13), 1.1% and 1.4% of the patients treated with thrombolysis had an in-hospital stroke, respectively. Thus, it is rather reassuring that the implementation of thrombolytic therapy in clinical practice has not resulted in an increased incidence of in-hospital strokes, despite the substantial use of the treatment among patients with contraindications.
Selecting for thrombolytic therapy and mortality
In our population, 40.8% received thrombolytic therapy, but as many as 34% of patients with AMI did not receive thrombolysis, even though indications for treatment with thrombolytics were fulfilled. In a recent European study, the corresponding figure was 20% (14). Numerous other studies have, in agreement, shown that only 33% to 50% (5,15)of patients eligible for thrombolytic therapy do receive the treatment. The 35-day mortality among all our patients treated with thrombolysis was 8.2%. Among patients who fulfilled standard indications without contraindications, the 35-day mortality was 8.1%. This is comparable to the 35-day mortality of 9.6% in the thrombolytic studies according to FTT. The corresponding mortality among (2) patients belonging to the group with “indications” who were not treated with thrombolytic therapy was 18.2%.
In general, mortality among patients with AMI who do not receive thrombolytic therapy is twice as high as it is among patients treated with thrombolytic therapy (16,17); in clinical practice, this is partly due to selection bias. It is tempting to use the results we have presented to conclude that contraindications and lack of indications should be ignored and that thrombolytic therapy should be given to many more patients. But the correct conclusion is that physicians are able to select those patients who are beyond the standard indications but who will benefit from thrombolytic therapy. Whether thrombolytic therapy, in general, will benefit part of the rejected high-risk population remains to be shown.
With the large benefit known to be associated with thrombolytic therapy and the favorable result of thrombolytic therapy in patients with contraindications observed in this and other studies, a randomized study of thrombolytic therapy among patients with some of the current contraindications is warranted, in particular, in patients with a risk of bleeding based on events some time in the past. The benefits of thrombolysis could very well outweigh the risks in many patients with a risk of bleeding or stroke.
☆ Supported by Roussel UCLAF, Romainville, France, and Knoll AG, Ludwigshafen, Germany.
- acute myocardial infarction
- Fibrinolytic Therapy Trialist Collaborative Group
- TRAndolapril Cardiac Evaluation study
- Received June 21, 2000.
- Revision received November 20, 2000.
- Accepted January 12, 2001.
- American College of Cardiology
- Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group
- Ryan T.J,
- Anderson J.L,
- Antman E.M,
- et al.
- Task Force on the Management of Acute Myocardial Infarction of the European Society of Cardiology
- Rogers W.J,
- Bowlby L.J,
- Chandra N.C,
- et al.
- Gore J.M,
- Granger C.B,
- Simoons M.L,
- et al.
- Stevenson R,
- Ranjadayalan K,
- Wilkinson P,
- Roberts R,
- Timmis A.D