Author + information
- Received March 24, 1997
- Revision received June 17, 1997
- Accepted June 26, 1997
- Published online October 1, 1997.
- Frederic Kontny, MDA,* (, )
- Jon Dale, MD, PhDA,
- Ulrich Abildgaard, MD, PhDB,
- Terje R. Pedersen, MD, PhDA,
- on Behalf of the FRAMI Study Group1
- ↵*Dr. Frederic Kontny, Department of Cardiology, Aker University Hospital, N-0514 Oslo, Norway.
Objectives. The present trial investigated the efficacy and safety of dalteparin in the prevention of arterial thromboembolism after an acute anterior myocardial infarction (MI).
Background. Left ventricular (LV) thrombus formation is associated with increased risk of arterial embolism in patients with an acute MI. Thrombolytic and antiplatelet therapy do not prevent thrombus formation.
Methods. A total of 776 patients were enrolled in a multicenter, randomized, double-blind, placebo-controlled trial of subcutaneous dalteparin (150 IU/kg body weight every 12 h during the hospital period). Thrombolytic therapy and aspirin were administered in 91.5% and 97.6% of patients, respectively. The primary study end point was the composite of thrombus formation diagnosed by echocardiography and arterial embolism on day 9 ± 2.
Results. Of 517 patients with echocardiographic recordings available for end point analysis, thrombus formation or embolism, or both, was found in 59 (21.9%) of 270 patients (59 with thrombus, none with embolism) in the placebo group and 35 (14.2%) of 247 patients (34 with thrombus, 1 with embolism) in the dalteparin group (p = 0.03). The risk reduction of thrombus formation associated with dalteparin treatment was 0.63 (95% confidence interval 0.43 to 0.92, p = 0.02). Analyses of all randomized patients (388 in each group) revealed no significant difference between the placebo and dalteparin groups with respect to arterial embolism (6 vs. 5 patients), reinfarction (8 vs. 6 patients) and mortality rates (23 vs. 23 patients, p = NS for all). Dalteparin was associated with an increased risk of hemorrhage: major in 11 dalteparin group patients (2.9%) versus 1 placebo group patient (0.3%, p = 0.006); minor in 52 dalteparin group patients (14.8%) versus 8 placebo group patients (1.8%, p < 0.001).
Conclusions. Dalteparin treatment significantly reduces LV thrombus formation in acute anterior MI but is associated with increased hemorrhagic risk.
Left ventricular (LV) thrombus formation frequently complicates acute anterior myocardial infarction (MI), even in patients given thrombolytic therapy and aspirin [1–4]. The clinical relevance of thrombus is the association with an increased risk of arterial embolism [5–7]; thus, the prevention of thrombus formation is important. Enhanced thrombin activity, whether related to the acute event itself or to thrombolytic therapy , plays a pivotal role in thrombus formation and provides a rationale for the use of heparin in its prevention. However, clinical trials have failed to demonstrate any additional benefit of such treatment on morbidity or mortality rates when given in conjunction with streptokinase and antiplatelet agents [9, 10].
New antithrombotic compounds are currently undergoing clinical investigation. One group of these readily available compounds for clinical use is the low molecular weight heparins (LMWHs). These compounds are derivatives of unfractionated heparin (UFH). However, the pharmacokinetic properties of LMWHs differ from that of UFH. High dosage-independent bioavailability and a longer half-life provide an antithrombotic efficacy of subcutaneous LMWH that equals intravenous UFH [11, 12]. The bioavailability of dalteparin, an LMWH (molecular weight 5,000 daltons) derived from porcine heparin through polymerization and ultrafiltration, is ∼90% . In comparison, the bioavailability of subcutaneous UFH is highly variable and dose dependent. Only 20% of the drug is absorbed at doses of 5,000 IU [14, 15]. The consequent inadequate antithrombotic response may explain the discouraging clinical results with subcutaneous UFH [16–18]. Furthermore, LMWHs seem to bear less hemorrhagic risk than that observed with UFH.
A recent study demonstrated significant reduction in morbidity and mortality, with low risk of hemorrhage, with dalteparin in unstable angina pectoris. However, there are no reports from randomized trials on dalteparin in acute myocardial infarction.
The objective of the present study was to investigate the efficacy and safety of dalteparin in the prevention of LV thrombus formation and arterial embolism after acute anterior MI.
1.1 Study design and patients
The study was designed as a prospective, randomized, double-blind, placebo-controlled, multicenter trial with parallel groups. Three university hospitals and 10 central and county hospitals in Norway participated in patient recruitment. Informed consent was obtained from all patients, and the study was otherwise also performed in accordance with the Helsinki Declaration.
Thrombus formation may occur very early after an acute MI. Thus, prompt administration of randomized treatment before enzymatic verification of acute MI was encouraged. Delay in study enrolment was omitted through randomization on a 24-h/day basis and with few inclusion criteria: case history and electrocardiographic (ECG) changes (i.e., Q waves or ST segment elevation ≥1 mm in leads I and aVL or ≥2 mm in any two or more contiguous leads of leads V1through V6) suggestive of an evolving anterior Q wave MI. These specific ECG changes were chosen to rule out patients with infarct sites other than anterior and those with a non–Q wave MI, both associated with low risk of thrombus formation. Patients with elevated enzyme levels but without the ECG changes were treated according to local routines. Patients of either gender and all ages were eligible for inclusion. Continued study drug treatment required verification of a first anterior MI with a peak aspartate aminotransferase (AST) value ≥100 U/liter or creatine kinase (CK) twice normal levels or higher. Patients with one or more of the following conditions were regarded as ineligible: previous anterior MI, infarct site other than anterior, time from onset of symptoms to start of study treatment >15 h, ongoing treatment with or indication for heparin or warfarin, systolic blood pressure >210 mm Hg, diastolic blood pressure >115 mm Hg, cerebrovascular events within the previous 2 months, known allergy to trial drug, peptic ulcer, known bleeding disorders, serious liver or renal failure, pregnancy, neoplastic or concomitant life-threatening diseases, alcohol abuse and simultaneous participation in other clinical trials.
1.2 Study end points
The primary end point was the composite of LV thrombus formation diagnosed by echocardiography on day 9 ± 2 and arterial embolism. Secondary end points encompassed reinfarction and total and cardiovascular mortality. Reinfarction was diagnosed by recurrent chest pain resistant to glyceryltrinitrate associated with a significant rise in AST and CK and new ST segment changes on the ECG. Safety variables were minor and major hemorrhage (i.e., cerebrovascular or any other bleeding that required transfusion or surgical intervention), allergic reactions and thrombocytopenia (i.e., platelet count ≤100 × 109/liter or reduction ≥50% from prerandomization values). All randomized patients were followed up for 3 months concerning efficacy variables. The present report includes only in-hospital data.
1.3 LV thrombus on echocardiography
LV thrombuswas defined as an echodense mass within the ventricular cavity with 1) a marked delineation toward the lumen and 2) a structural texture different from the myocardium and that 3) could be separated from the underlying endocardium by an endocardial lining; 4) was visible throughout the entire cardiac cycle; and 5) could be demonstrated in at least two different views. Thrombus shape was classified as protruberantwhen the free margin toward the ventricular lumen was convex. Thrombus mobilityrequired demonstration of pendulations apart from those mediated by myocardial contractions. LV wall motion was graded from 0 to 3: 0 = normal motion; 1 = hypokinesia; 2 = akinesia; 3 = dyskinesia, including signs of LV aneurism. Three different echocardiographs were used: Vingmed CFM (models 700, 725, 750 and 800; all Vingmed Sound, Norway); Irex Meridian (Johnson & Johnson); and Acuson 128 × p/10 (Acuson). Multiple depth, gain and reject settings were used for optimal visualization of intracavitary structures. Recordings from parasternal long- and short-axis, apical two- and four-chamber and subcostal views were stored on separate videocassettes using Panasonic AG 6200 (VHS PAL) and Panasonic AG 7330 (SVHS PAL) video recorders. Before the start of the study, the investigators, all experienced in echocardiography, attended a specially designed echocardiography workshop to ensure homogenous examinations with optimal quality. An end point classification committee with no knowledge of the clinical data, including randomized treatment, successively reviewed all echocardiographic recordings. The local investigators continuously received reports on recording quality from the committee. The committee members were encouraged to make a joint conclusion after independent review of the recordings. An external video audit committee was responsible for final classification in cases of discrepancy.
1.4 Arterial embolism
A standardized clinical examination for recognition of embolic events was performed repeatedly during the hospital period. Ischemic strokewas defined as any focal neurologic deficit of sudden onset that was observed by persons other than the patient, lasted >30 min and cerebral hemorrhage excluded by computed tomography. Suspected arterial emboli to other regions were verified by angiography.
1.5 Study treatment
Dalteparin (Pharmacia, Sweden; 150 IU/kg body weight every 12 h) was injected subcutaneously in the abdominal region. This dosage was based on our own results in a previous dose-finding study . Body weight was adjusted to weight classes of 10 kg (i.e., 40 to 49, 50 to 59, 60 to 69), with an upper limit of 100 kg. For safety reasons, the first study injection was given 8 h after thrombolytic therapy. In patients not receiving such therapy, study treatment was commenced immediately after randomization. The scheduled treatment period was 9 ± 2 days.
1.6 Thrombolytic and concomitant antithrombotic therapy
Intravenous streptokinase (1.5 × 106IU/h) was administered in patients with no contraindications and with symptoms ≤6 h in duration. Soluble aspirin (300 mg) was given orally on admission and 160 mg/day thereafter. No other antithrombotic agents were allowed during the study period. At discharge from the hospital, patients considered at high risk of thromboembolic events (i.e., thrombus or severe wall motion abnormalities) received warfarin; all others continued receiving antiplatelet therapy.
1.7 Laboratory measurements
Analyses of plasma dalteparin (Coatest LMW Heparin, Chromogenix, Sweden), thrombin–antithrombin complex (Enzygnost TAT micro, Behring, Germany) and d-dimer (Asserachrom D-Di, Stago, France) were performed at the core laboratory in blood samples taken before and 4 h after the study injection on the morning of treatment days 2 and 7.
1.8 Statistical considerations
Patient data were recorded on specially designed case record forms. Data processing was performed by Pharmacia AB, Stockholm in cooperation with the steering committee and clinical coordinator. All statistical analyses were performed independently of the sponsor.
1.8.1 Sample size calculation
The major outcome measurewas the composite of LV thrombosis and arterial embolism at day 9 ± 2 (mean ± SD). Because of randomization before definite verification of acute MI and a lower cutoff peak value for AST of ≤100 U/liter, it was calculated that 30% of the randomized patients would not meet the continuation criteria. Because of technical reasons or early death, a further 10% were not expected to have echocardiographic recordings available for end point analyses. The end point incidence among those who continued study participation was estimated to be 28% in the placebo group. According to available published data, a 50% reduction was estimated for sample size calculation. To reach a power (1 − beta) that equaled 90% at an alpha value of 0.05 (two-tailed) for detecting this reduction, 388 patients in each group were needed.
1.8.2 Statistical methods
Analyses of LV thrombus formation were performed in patients with available echocardiographic recordings. However, all randomized patients were retained in all other end point analyses. A two-sided exact Fisher-Irwin test with a 5% significance level was applied for comparison of specific events between the two patient groups. Logistic regression was used for simultaneous analyses of associations between background variables (including treatment) and efficacy variables. Only statistically significant variables were retained in the regression analyses. Results are presented as odds ratios with 95% confidence intervals and p values.
A total of 776 patients (388 in each treatment group) with a suspected first acute anterior MI were randomized. Baseline characteristics of the two study groups were similar (Table 1). Of all randomized patients, 91.5% received streptokinase and 97.6% aspirin. The use of these drugs, as well as other concomitant medical treatment, was equally distributed between the treatment groups.
After randomization, 137 patients (17.7% [68 placebo, 69 dalteparin]) were ineligible for study continuation according to protocol criteria. Of these 137 patients, 112 (81.8%), equally distributed between the treatment groups, had an acute MI and received streptokinase. However, the peak enzyme values did not reach the lower cutoff limit for continuation. Discontinuation from further study drug treatment occurred in 73 patients (9.4%) for reasons shown in Table 2. Thirty-six patients (4.6% [19 placebo, 17 dalteparin]) died before echocardiography. Thus, echocardiography was performed in 530 (68.3%) of 776 patients (277 placebo, 253 dalteparin). For technical reasons, 7 (2.5%) of 277 echocardiographic recordings in the placebo group and 6 (2.4%) of 253 in the dalteparin group were excluded from end point analyses. Thus, analysis of LV thrombus formation was performed in 517 patients (270 placebo, 247 dalteparin) (Fig. 1). The difference in number of patients not available for echocardiographic analyses was not statistically significant (141 of 388 [placebo] vs. 118 of 388 [dalteparin], p = 0.09). There was no difference in baseline characteristics between treatment groups among patients available for analyses. All had a verified first acute anterior MI. No patient received UFH during the study period, and there was no crossover between treatment groups.
2.2 LV thrombus formation and arterial embolism
The combined incidence of LV thrombus formation and arterial embolism was 21.9% (59 of 270) in the placebo group and 14.2% (35 of 247) in the dalteparin group (p = 0.030). LV thrombus formation without embolism was found in 93 (18.0%) of 517 patients: 59 (21.9%) of 270 in the placebo group; 34 (13.8%) of 247 in the dalteparin group (p = 0.022) (Table 3). There was no significant difference in the prevalence of thrombus between the recruiting centers (p = 0.31). During day 0 to 11, arterial embolism occurred in 11 (1.4%) of 776 patients. Five (1.3%) of 388 patients in the placebo group and 4 (1.0%) of 388 in the dalteparin group had a verified ischemic stroke. One additional patient in the placebo group had a stroke, but this event was not further classified. Peripheral arterial embolism occurred in one patient given dalteparin.
Protruberant shape of the thrombus was found in 49 cases (52.7%), and 7 (7.5%) were classified as mobile, with no statistically significant difference between the treatment groups (Table 3).
Irrespective of randomized treatment, the extent of LV wall motion abnormalities was significantly associated with thrombus formation: grade 0= no thrombus; grade 1= 3.6% thrombosis; grade 2= 24.3% thrombosis; grade 3= 31.0% thrombosis (p < 0.001 for linear association). The distribution of LV wall motion abnormalities did not differ between treatment groups.
Time from onset of symptoms to start of study treatment was 12.6 ± 2.4 h in patients with and 12.4 ± 3.3 h in those without thrombolytic therapy. (p = NS). Thrombus was diagnosed in 84 (17.5%) of 479 patients with compared with 9 (23.7%) of 38 without thrombolytic therapy (p = 0.38). However, after adjustment for peak AST value and gender there was a trend toward an increased prevalence of thrombus among patients not receiving thrombolytic therapy (p = 0.08). No other concomitant treatment (i.e., aspirin, beta-adrenergic blocking agents, nitrates, diuretic drugs and angiotensin-converting enzyme inhibitors) influenced thrombus formation.
Mean peak AST was higher in patients with (516 ± 278 U/liter) than in those without thrombus (428 ± 304 U/liter, p = 0.02). Furthermore, there was a nonsignificant trend toward a higher mean peak CK value in patients with than in those without LV thrombus (3,406 ± 2,391 vs. 2,932 ± 2,151 U/liter, respectively, p = 0.09). LV thrombus developed in 16 (11.6%) of 138 women and 77 (20.3%) of 379 men (p = 0.027).
After logistic regression analyses, randomized treatment, peak AST value and gender remained significantly associated with LV thrombus formation (Table 4).
2.3 Reinfarction and mortality
Reinfarction occurred in 14 (1.9%) of 776 patients during days 0 to 11: 8 (2.1%) of 388 in the placebo group; 6 (1.6%) of 388 in the dalteparin group. This difference was not statistically significant.
The total mortality rate during days 0 to 11 was 5.9%, with no difference between the groups. The presumed cause of death was cardiac related in all 23 patients in the placebo group and in 21 of the patients in the dalteparin group. Two patients in the dalteparin group died of intracranial hemorrhage.
2.4 Hemorrhage and other complications
Major hemorrhage occurred in 12 patients (1.6%): 1 (0.3%) of 388 in the placebo group; 11 (2.9%) of 388 in the dalteparin group (p = 0.006). Two fatal and one nonfatal cerebral hemorrhages occurred in the dalteparin group and were verified by computed tomography. In this treatment group, all patients were given thrombolytic therapy, and 10 received aspirin. The event occurred within 48 h after the start of treatment in six patients; further clinical details are shown in Table 5. Minor hemorrhage occurred in 52 of 388 dalteparin-treated patients and in 8 of 388 placebo-treated patients and were also significantly associated with dalteparin treatment (p < 0.001) (Table 5). Two minor bleeding events but no major event was related to interventional procedures. Thrombocytopenia occurred in one patient (0.3%) treated with dalteparin. No allergic reactions were observed.
2.5 Laboratory measurements
The target range of plasma dalteparin levels measured 4 h after the morning injection was 0.7 to 1.4 IU/ml on the basis of our previous results from a pilot study . Blood samples taken on treatment day 7 from 225 patients in the dalteparin group revealed that 156 (69.3%) had values within, 53 (23.6%) had values below, and 16 (7.1%) values above the target range. Mean plasma concentration for the whole group was 0.91 ± 0.40 IU/ml. The plasma levels did not differ significantly between patients with and without LV thrombus (0.80 ± 0.45 vs. 0.92 ± 0.35 IU/ml, respectively, p = NS). The prevalence of thrombus among patients with plasma levels ≤0.7 IU/ml was 21.6% compared with 11.1% among those with plasma levels >0.7 IU/ml (p = NS). Analyses revealed high plasma dalteparin levels (1.1 to 2.0 IU/ml) in samples available from four patients with a major hemorrhage, including two with a cerebral hemorrhage. There was no significant difference in plasma dalteparin levels between patients with and without minor hemorrhage (0.96 ± 0.51 vs. 0.91 ± 0.37 IU/ml, respectively).
Thrombin–antithrombin complex and d-dimer levels measured in 174 patients did not differ between those with and those without LV thrombus. Adjustment for randomized treatment did not alter this result.
The present trial assessed the potential benefits of subcutaneous dalteparin (150 IU/kg body weight every 12 h) in the prevention of LV thrombus formation and arterial embolism after acute anterior MI. The treatment groups were well balanced with respect to baseline characteristics and prognostically important factors, such as infarct size, degree of LV wall motion abnormalities and time to treatment. The use of thrombolytic therapy, aspirin and other concomitant treatment also did not differ between the groups. We found that dalteparin treatment was associated with a statistically significant reduction in the composite end points of LV thrombus formation and arterial embolism (p = 0.030).
3.1 LV thrombosis
The propensity of thrombus formation in acute MI is influenced by several factors, such as infarct site and size, LV dysfunction, including regional akinesia, and clinical evidence of heart failure (see Danzig et al for review). Differences in these as well as other factors, such as diagnostic techniques, time of examination and use of antithrombotic treatment, cause substantial variation in the reported frequency of thrombus from different series.
LV thrombosis occurred in 21.9% of patients in the placebo group in the present trial. A similar frequency (26%) was found in patients randomized to streptokinase in a comparable, Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico II (GISSI-2)–connected study . In our trial, weight-adjusted dalteparin treatment reduced thrombus formation by 37%, which is concordant with the effect of fixed-dose subcutaneous UFH (12,500 IU every 12 h) in thrombus prevention observed by Turpie et al. (11% vs. 32%, p < 0.001). In contrast, this heparin treatment regimen did not prevent thrombus formation in the GISSI-2–connected trial (27% vs. 30%, p = NS). The concomitant use of thrombolytic therapy in the GISSI-2–connected study, as opposed to the study by Turpie et al , may offer one explanation for the lack of heparin effect in the former. Thrombolytic therapy is associated with enhanced thrombin activity, and higher heparin dosages may therefore be needed to achieve adequate anticoagulant effects when conjunctive heparin treatment is given .
Furthermore, irrespective of thrombolysis, therapeutic levels of anticoagulation are often not achieved with subcutaneous UFH due to slow release and limited capacity to saturate heparin clearance sites. This limitation may become critical to the treatment effect when given in conjunction with thrombolytic therapy.
The beneficial effect of dalteparin in the present trial is probably due to the better anticoagulation achieved with a weight-adjusted dosage regimen and to the superior pharmacologic properties of LMWH versus UFH (e.g., better bioavailability, avoidance of neutralization by platelet factor 4, less binding to plasma proteins and endothelium) [24–26].
3.2 Arterial embolism, including stroke
The clinical relevance of LV thrombosis is its relation to arterial embolism. In a meta-analysis of 11 studies , the pooled odds ratio for increased risk of embolization was 5.45 (95% confidence interval 3.02 to 9.83) in patients with versus those without LV thrombus. However, as with thrombus frequency, the reported incidence of arterial embolism, including stroke, varies substantially in different series.
The low incidence of stroke in the present trial (1.2%) is concordant with that observed in the GISSI-2–connected study (0.5%). In contrast, earlier series [2, 27, 28], including that of Turpie et al. , have reported higher stroke rates (1.5% to 10%). Again, differences in the use of thrombolytic and antiplatelet therapy may be related to the variable event rates. Such combined antithrombotic therapy seems to reduce the embolic potential of a thrombus, although it apparently does not prevent thrombus formation itself. Indeed, a significant reduction in stroke rate was achieved with such combined treatment in the Second International Study of Infarct Survival (ISIS-2) trial (0.6% vs. 1.1%) .
Reperfusion by thrombolysis with improvement in left ventricular contractility and blood flow, as well as direct effects on the thrombus surface of thrombolytic agents and aspirin, may influence thrombus morphology and thus reduce the risk of embolism [21, 30]. The occurrence of thrombus mobility, a morphologic characteristic associated with increased risk of embolism, was notably lower in both the present trial (7.5%) and the GISSI-2–connected study (2%) than that reported in several series of patients not receiving thrombolytic and antiplatelet therapy (∼20%) [5, 6, 31–33]. Dalteparin treatment did not seem to influence thrombus mobility or shape in the present trial.
An assumption of a direct relation between thrombus presence and arterial embolism would suggest a lower embolic event rate than that observed in the dalteparin group. However, a probability of 0.42 of achieving no difference between the groups even if a true difference exists still makes it conceivable that prevention of thrombus formation may influence the rate of arterial embolism in a larger population.
3.3 Reinfarction and death
The overall reinfarction rate during days 0 to 11 was 1.9% (2.1% vs. 1.6%, p = NS). Although consistent with the in-hospital event rate in patients randomized to receive streptokinase and aspirin in the ISIS-2 trial (1.8%) , it is somewhat lower than that observed in other large clinical trials (2.6% to 2.8%) [9, 10]. In the present trial, the exclusion of patients at high risk for reinfarction (e.g., those with unstable angina or a non–Q wave myocardial infarction [MI]) by per protocol criteria for study inclusion and continuation may have influenced the event rate.
In one recent study , long-term treatment with LMWH given in conjunction with thrombolytic therapy did reduce the risk of reinfarction after acute MI. This reduced risk was also found with long-term dalteparin treatment in patients with unstable angina or a non–Q wave MI . Neither of these studies have demonstrated any significant effect on mortality.
The overall rate of cerebral hemorrhage (0.42%) is consistent with the Third International Study of Infarct Survival (ISIS-3) results . All events occurred in the dalteparin group. However, the low number of patients does not permit any assessment of a potential relation to this treatment.
Most patients with major hemorrhage in the dalteparin group received streptokinase and aspirin, suggesting the need for reduced dalteparin dosage. In the Fragmin During Instability in Coronary Artery Disease (FRISC) trial , major hemorrhage was reduced from 6.3% to 0.8% by lowering the dalteparin dosage from 150 to 120 IU/kg body weight twice daily. However, thrombolytic therapy was not used because these patients had unstable angina pectoris or a non–Q wave MI.
Compared with the 2.9% incidence of major hemorrhage associated with dalteparin treatment in the present trial, this complication seems to occur in ∼4% of patients treated with intravenous UFH. Dalteparin, even at a dosage of 150 IU/kg body weight twice daily, may thus be associated with lower hemorrhagic risk.
3.5 Study limitations
Early randomization, before establishment of a definite diagnosis of acute MI, has the risk of including patients with either aborted or small infarctions. Such patients are at low risk of thromboembolic complications. Thus, the study protocol included a lower cutoff limit of peak AST ≤100 U/liter in its continuation criteria. The sample size calculation included an estimate that 40% of randomized patients would not have available echocardiographic recordings because of failure to meet the continuation criteria or for other reasons (e.g., death, withdrawal, technical). The actual rate of patients without available recordings was 33.4%. However, although unequal in the two groups, the rates did not differ significantly. Furthermore, except for end points that included LV thrombosis, all statistical analyses were based on intention to treat, thus retaining all 776 randomized patients.
Patients with LV thrombosis remain at risk for embolic events beyond the acute phase. A study period of 11 days, as in the present trial, may have contributed to the low rate of arterial embolism. These conditions, together with the unclear clinical significance of LV thrombi without embolism, restrict the “risk–benefit” assessment of dalteparin treatment.
An observed risk reduction lower than that calculated in the sample size estimation may imply a weakened statistical significance. However, in the present study the observed risk reduction of 37%, as opposed to a calculated reduction of 50%, was still statistically significant. Thus, the difference between the calculated and observed risk reduction should not affect the interpretation of the results of the present study.
Dalteparin (150 IU/kg body weight twice daily) is associated with a significant reduction in LV thrombus formation in patients with an acute anterior MI, even when streptokinase and aspirin are routinely given. Whereas the results of randomized trials with UFH do not appear to justify the routine addition of UFH to aspirin, conjunctive dalteparin treatment may thus improve clinical outcome of patients with acute MI. However, it remains to be determined whether the benefit observed with dalteparin in the present trial could be extended to include the prevention of clinically important events in a larger patient population.
Dalteparin treatment carries an increased risk of hemorrhage. However, this risk appears to be lower than that observed with UFH, and a presumed better tolerance of a reduced dalteparin dosage must be weighed against a possible loss in clinical benefit.
Further investigations should aim at identification of patients at particularly high risk of both thromboembolic and hemorrhagic complications.
We are indebted to the many co-workers (see Appendix A) and, above all, patients for their participation in this trial.
A.1 The FRAMI Study Group
Steering committee.Chair: Jon Dale, MD, PhD; Co-Chair: Ulrich Abildgaard, MD, PhD; Members: Terje Pedersen, MD, PhD, Frederic Kontny, MD, Anders Nesvold, MD. Study coordinator: Frederic Kontny, MD. Safety monitoring committee: Bernt Ly, MD, PhD, Terje Andersson, MD. Participating centers, investigators and study nurses: Aker sykehus, Oslo: Frederic Kontny, Nina Kristiansen. Akershus sentralsykehus, Nordbyhagen: Arfinn Rollag, Frøydis Sjølyst, Tom Finn Olsen. Askim sykehus, Askim: Trond Fløgstad, Marit Busengdal. Diakonissehjemmets sykehus, Bergen: Rune Fanebust, Elin Magnussen. Haugesund sykehus, Haugesund: Kjell Waage, Eva Viland. Hedmark sentralsykehus, Elverum: Hall Schartum-Hansen, Trude Grønvold, Anne G. Moe. Kongsvinger sykehus, Kongsvinger: Erik Anker, Kari Glasø. Moss sykehus, Moss: Per Lem, Jorunn Gregersen. Møre og Romsdal sentralsykehus, Ålesund: Torstein Morstøl, Kirsti Sørdahl. Orkdal Sanitetsforenings sykehus, Orkanger: Arve Tromsdal, Tore G. Salvesen, Oddny Rian. Regionsykehuset, Trondheim: Lars Hegrenæs, Andrea Stoum, Inger Lundhaug. Telemark sentralsykehus, Porsgrunn: Dag Torvik, Helge Flaten. Telemark sentralsykehus, Skien: Kenneth Knutsen, Randi M. Johnson. Pharmacia study coordinator: Per Blom MD, PhD. Data Monitoring and entry center(Pharmacia): Svend Undeland, Carin Fellenius, Claes Göran Brobäck, Grete Handeland, Tor Seim, Kerstin Wicklund, Gudrun Ekman. Data analyses (Medstat Research, Lillestrøm): Leiv Sandvik, PhD. End point classification committee: Jon Dale, MD, PhD, Frederic Kontny, MD. External video audit committee: Jan Erik Nordrehaug, MD, PhD, Karl Arne Johannessen, MD, PhD. Core laboratory (Haematological Research Laboratory, Aker University Hospital): Ulrich Abildgaard, MD, PhD, Mette Lie Larsen, Ellen Røkås.
- aspartate aminotransferase
- creatine kinase
- electrocardiogram, electrocardiographic
- Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico II
- Second International Study of Infarct Survival
- low molecular weight heparin
- left ventricular
- myocardial infarction
- unfractionated heparin
- Received March 24, 1997.
- Revision received June 17, 1997.
- Accepted June 26, 1997.
- The American College of Cardiology
- Johannessen KA,
- Nordrehaug JE,
- von der Lippe G
- Vecchio C,
- Chiarelli F,
- Lupi G,
- Bellotti P,
- Domenicucci S
- Haugland JM,
- Asinger RW,
- Mikell FL,
- Elsperger J,
- Hodges M
- Stratton JR,
- Resnick AD
- Vaitkus PT,
- Barnathan ES
- ISIS-3 (Third International Study of Infarct Survival) Collaborative Study Group
- Barrowcliffe TW,
- Johnson EA,
- Thomas DP
- Kroon C,
- ten Hove WR,
- de Boer A,
- et al.
- Goldhaber S
- Delanty N,
- Fitzgerald DJ
- Dantzig JM,
- Delemarre BJ,
- Bot H,
- Visser CA
- Galvani M,
- Abendschein D,
- Ferrini D,
- Ottani F,
- Rusticali F,
- Eisenberg PR
- Lane DA,
- Denton J,
- Flynn AM,
- Thunberg L,
- Lindahl U
- Hemker HC. The mode of action of heparin in plasma. In: Verstraete M, Vermylen J, Lijnen R, Arnout J, editors. Leuven: Thrombosis and Haemostasis. Leuven University Press, 1987:17–36.
- Lane DA,
- Pejler G,
- Flynn AM,
- Thompson EA,
- Lindahl U
- Kontny F,
- Dale J,
- Nesvold A,
- Lem P,
- Søberg T
- Visser CA,
- Kan G,
- Meltzer RS,
- Dunning AJ,
- Roelandt J
- Johannessen K-A,
- Nordrehaug JE,
- von der Lippe G,
- Vollset SE
- Domenicucci S,
- Bellotti P,
- Chiarella F,
- Lupi G,
- Vecchio C
- Collins R,
- MacMahon S,
- Flather M,
- et al.