Anticoagulant Therapy After Acute Myocardial Infarction

Alexander G.G. Turpie, MD

The use of anticoagulant therapy for patients who have had an acute myocardiil infarction is still con- troversial, mainly because early major studies had conflicting findings, but reanalysis of the data did produce evklence that anticoagulation had clinically and statistically significant beneBts. Now more evi- dence, imludlng the results of a lo-day In-hospital study of low- and high-dose calcium heparin, has been gathered to support using anticoagulants for thesepatlents.

The study used the development of left ventricu- lar mural thrombosis, a frequent complication of acute myocardiil infarction that carries a high risk for systmic embollc complications, to assess clinical outcome: A recluced incidence of mural thrombosis woukl be taken to indicate reduced chances that patients wouid have major systemic emboli. Two- dimensional echocardiography was used to detect thrombi.

In the study, the imklence of left ventricular mu- ral thrombosis was significantly lower in the high- than in the low-dose group. Among patients in the high-dose group in whom a mural thrombosis did develop, plasma heparin concentrations were sig- nificantly lower and activated partial thromboplas- tin times were shorter. These data suggest that monitoring plasma hepatin levels and anticoagulant responsecanensure maximal treatment effective- ness. No slgnificant differences in other out- comes-such as bleeding complieations, nonhe- morrhagic strokes and mortality-were found be- tween the high- and low-dose treatment groups.

(Am J Cardiol1990;65:2OG23C)

From McMaster University, Hamilton, Ontario, Canada. Address for reprints: Alexander G. G. Turpie, MD, Hamilton Gen-

eral Hospital, 237 Barton St. E., Hamilton, Ontario LSL2X2, Canada.

20C THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 65

w ether anticoagulant therapy should be ad- ministered to patients after an acute myocar- dial infarction (MI) remains a controversial is-

sue. Accumulating evidence strongly suggests that ad- ministering anticoagulant therapy to patients who have had acute MI is associated with several potential benefits. These include: (1) decreasing mortality, (2) reducing in- farct extension or recurrent infarction, (3) preventing left ventricular mural thrombosis and systemic embolism, and (4) preventing deep-vein thrombosis and pulmonary embolism. These potential benefits of anticoagulant ther- apy have not been universally recognized, however, large- ly because of the conflicting findings of early studies.

The controversy surrounding the use of anticoagulant therapy after acute MI reflects the inconsistent findings of 3 large randomized and controlled clinical trials of anticoagulant therapy for acute MI that were published in the late 1960s and early 1970~.‘-~ All 3 studies report- ed that anticoagulant therapy resulted in reduced mor- bidity and fewer systemic emboli and venous thromboem- boli. However, only 2 of the studies1+3 cited fewer strokes or pulmonary emboli. Furthermore, only 1 of the 3 trials2 showed a statistically significant reduction in mortality. Thus, these findings provided little clear-cut evidence of which outcomes were improved by using anticoagulants.

The results of a subsequent meta-analysis of all trials of anticoagulant therapy for acute MI were published by Chalmers et al4 in 1977. It provided convincing evidence that anticoagulant therapy does produce a clinically and statistically significant reduction in the major outcomes of MI, including death and thromboembolic events.

However, despite the evidence that anticoagulants are beneficial for treating acute MI, there is currently no consensus on their use. In fact, prescribing patterns vary widely from one region to another. Nowhere is this better illustrated than in the recent International Study of In- farct Survival (ISIS-2) study of the planned use of anti- coagulant therapy.5 This study found that 89% of physi- cians in France planned to use anticoagulant therapy, compared with no physicians in Belgium. In the United States, use of full-dose intravenous heparin was planned for 37% of patients, oral anticoagulant therapy for 5% of patients, and low-dose subcutaneous heparin for 31% of patients.

There are several reasons why anticoagulant therapy has not been more widely adopted for management of acute MI. These include the inconsistent trial results, the small (although statistically significant) reductions in the major outcomes, and the development of alternative treatment regimens-particularly systemic thrombolytic therapy and early angioplasty.

Moreover, major advances in treating acute MI have

been made during the years since the original studies of anticoagulant therapy were conducted, and these ad- vances have resulted in a decrease in control mortality from 20% to only about 10 to 12%. Using today’s metho- dologic standards, many thousands of patients with acute MI would have to be enlisted into a randomized prospec- tive trial in order to demonstrate statistically significant benefits of anticoagulation treatment in reducing major adverse outcomes, such as mortality and systemic embol- ic events.

LEFT VENTRICULAR MURAL THROMBOSIS AS A “SURROGATE” END POINT

A reasonable alternative study strategy would be to use “surrogates” to assess important clinical outcomes. Left ventricular mural thrombosis is one such possible surrogate. Mural thrombosis is a frequent complication of acute MI. (The frequency of mural thrombosis, as determined by autopsy series, is approximately 50%.) Thrombus formation is probably determined by the inter- action of left ventricular systolic dysfunctions, primarily dyskinesia, and endocardial necrosis.

The development of 2-dimensional (2-D) echocardi- ography has provided an accurate means of detecting thrombi in a clinical setting. Numerous studies of mural thrombosis in patients with acute MI have found the incidence of thrombi to be about 20 to 30% in all patients with MI, but considerably higher (40 to 50%) in patients with anterior transmural MI.

The occurrence of left ventricular mural thrombosis carries a high risk of systemic embolic complications, such as stroke, mesenteric embolism, or acute femoral or brachial artery occlusion. The incidence of systemic em- bolic events among patients in whom a mural thrombus develops has been reported to be as high as 37%. Al- though the risk is highest during the first month after an acute MI, thrombi can persist-as can their embolic po- tential. These embolic events can complicate the care of patients who are convalescing after an acute MI.

Because most clinically significant systemic emboli are associated with left ventricular mural thrombosis, it seems reasonable to consider a reduction in left ventricu- lar mural thrombosis as an indicator for a clinically sig- nificant reduction in major systemic embolic events.

METHODS To explore this hypothesis, a double-blind, random-

ized multicenter trial was conducted to compare a high vs low dosage of calcium heparin administered subcutane- ously for 10 days to prevent left ventricular mural throm- bosis in patients with acute anterior transmural infarc- tion.6 High dosage was defined as 12,500 U every 12 hours; low dosage as 5,000 U every 12 hours.

Study end points: The principal outcome measured by the study was left ventricular mural thrombosis, de- tected by 2-D echocardiography 10 days after the pa- tient’s MI. A positive diagnosis consisted of a well-de- fined intracavitary mass in contact with, but distinct from, mural endocardium, visualized in more than 1 view or plane, and synchronous in motion with underlying myocardium (Fig. 1). Presence or absence of left ventric-

FIGURE 1. Two- edtocardiogram showing a well- defhedintracavitarymassincontactwith,butdistinctfrom, themralendocarbun.LA=kftatrium;LV=kftventriek; MV = mitral vaive; RA = right m RV = right ventride.

ular wall motion abnormalities was also noted during echocardiography. Each 2-D echocardiogram was inter- preted independently, and in a blinded fashion, by 3 car- diologists who were experienced in the use of echocardi- ography. Any disagreements were resolved by consensus.

Other outcomes assessed by the study included death of the patient, nonhemorrhagic stroke, mesenteric or pe- ripheral arterial embolism, deep vein thrombosis, pulmo- nary embolism and bleeding complications. All patients were examined daily for evidence of bleeding and other aftereffects during the lo-day treatment period in the hospital, with 2-D echocardiography performed on the tenth day after MI. After discharge from the hospital, the patients were followed for 1 year to keep track of mortali- ty and the development of systemic emboli and venous thromboemboli.

A consecutive series of 233 patients with suspected acute transmural anterior infarction were recruited for the study. Inclusion criteria were (1) typical ischemic pain for at least 20 minutes or longer within 24 hours of assessment, and (2) electrocardiographic evidence of evolving or completed anterior MI.

Patient groups: Patients were excluded if they were age 90 years or older, had undergone cardiopulmonary resuscitation from asystole or secondary ventricular ti- brillation, were unable to begin treatment within 36 hours of pain onset, or if heparin administration was contraindi- cated-e.g., if the patient had active or recent hemor- rhage, peptic ulceration or recent surgery.

Of the 233 original patients, 221 were included in analyses of heparin effectiveness and safety (Table I). Of these, 183 (95 in the high-dose heparin group and 88 in the low-dose heparin group) underwent 2-D echocardiog- raphy on the tenth day.

There were no significant differences in the baseline characteristics of the patients in the 2 treatment groups. Forty-one percent of the patients had received p blockers, and 19% had received nonsteroidal anti-inflammatory drugs. Only 6% had received aspirin at any time during the first 10 days after infarction. Use of these drugs was

THE AMERICAN JOURNAL OF CARDIOLOGY FEBRUARY 2.1990 21c

A SYMPOSIUM: TRROMBOSlS AND ANTlTRROMBOTlC TRERAPY--DlRECTlON FOR THE ’90s

TABLE I Patient Population

Total randomized = 233

Eligible = 221 Ineligible = 12 No infarction 9 Major surgery 1 Refused 1 Delayed entry 1

2-D echocardiography = 183 No 2-D Echocardiography = 38 Died <lO days 24

Refused 8

Too ill 3 Technical 3

Rx36hours=171 Rx 36-72 hours = 12

TABLE II Incidence of Left Ventricular Mural Thrombosis at 10 Days After Myocardial Infarction

Finding on 2-D Echo High-Dose Group Low-Dose Group

Positive 10 (10.5%)* 28 (31.8%)* Negative 85 60

Total patients 95 88

l p = O.OW38. 2D Echo = Zdimensional echocardiography.

TABLE Ill Creatine Kinase (CK) Concentrations

CK U/L (mean f SEM*)

Highdose group 1.097 f 125

Lowdose group 1.920 f 149 Positive 2-D echocardiography 2,282 f 213* Negative 2-D echocardiography 1.819 f 108

* p = 0.028 SEM = standard error of the mea”.

TABLE IV Plasma Heparin Concentrations and Corresponding APlTs

High-Dose Group Low-Dose Group

Mean f SEM plasma 0.18 f 0.017 0.01 f 0.005 heparin (U/ml)

Mean APTT (set) 47.8 f 1.23 34.8 f 0.75

APTT = activated partial thromboplastin time; SEM = standard error of the mean.

distributed evenly among the patients in the 2 treatment groups.

RESULTS The study showed a 10.5% incidence of left ventricu-

lar mural thrombosis among the high-dose heparin group, which was significantly lower than the 3 1.8% inci- dence among the low-dose group (Table II). Of the 10 mural thrombi in patients who received high-dose hepa- rin, 5 were flat and 5 were protuberant; 1 protuberant thrombus was mobile. Of the 28 mural thrombi in pa-

22c THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 65

2 E 0.26

2 0.24 0.22 .

g 0.20 0.16 6 0.16

0.14 .E Li

0.12 8 0.06 0.10

I 0.00 ii 0.04

i!J 0.02 0 n 0 1 2 3 4 5 6 7 8 9 10

Day

10

1

O6 Day

FlGURE3.Mandhf&dpddthrombophdintilIK (APPT)hldghdosegroup:Compaimbehmenpatienbwith

tients treated with low-dose heparin, 15 were flat and 13 were protuberant; none was mobile.

As expected from previous studies, thrombosis and ventricular wall motion abnormalities were associated. Almost 95% of the 183 patients who underwent 2-D echccardiography had abnormal wall motion, which was classified as akinesia in most cases. All patients with mural thrombosis also had wall motion abnormalities, whereas no ventricular thrombi were found in patients with normal wall motion.

Serum creatine kinase levels were measured on hospi- tal admission and again 8 and 16 hours later. The highest concentrations recorded for each patient were used to calculate the mean serum creatine kinase levels. There were no significant differences between the mean creatine kinase concentrations of the high- and low-dose groups.

In contrast, patients with positive results had signifi- cantly higher serum creatine kinase levels than patients with negative results on 2-D echocardiograms (Table III). This finding is consistent with the results of previous studies that showed a correlation between the frequency of left ventricular mural thrombosis and infarct size. However, in the high-dose group alone, there were no significant differences in peak creatine kinase concentra- tions between patients with positive or negative results on 2-D echocardiograms.

Of the 183 patients who underwent echocardiogra- phy, 82 were taking p blockers. Of this subgroup, 28% had a mural thrombus, whereas of the 101 patients not receiving /3 blockers, only 15% had a mural thrombus (p <O.Ol). The high incidence of mural thrombi in the fi- blocking group may be related to the hemodynamic ef- fects of the negative inotropic action of these drugs.

Because all patients were given a constant-dose hepa- rin regimen during the lo-day treatment period, the plas- ma heparin concentrations and the corresponding acti- vated partial thromboplastin times (APTTs) were mea- sured daily at the midpoint between injections. Not surprisingly, the mean plasma heparin concentration in the high-dose group was significantly higher-almost 20 times higher-than that in the low-dose group. Similarly, the mean AP’IT in the high-dose group was also signifi- cantly higher than that in the low-dose group (Table IV).

In the high-dose group the plasma heparin concentra- tions were significantly lower among patients in whom a mural thrombus developed than among those in whom it did not. The APTTs were also significantly different: Patients with negative results on the 2-D echocardiogram had longer APTTs (Fig. 2 and 3). These data clearly show the relation between anticoagulant response and clinical effectiveness in preventing left ventricular mural thrombosis. They also suggest that plasma heparin levels and anticoagulant response should be monitored to achieve maximal therapeutic effectiveness.

other short-term outcomes: During the lo-day in- hospital observation period, there were nonhemorrhagic strokes in 1 of 95 patients in the high-dose heparin group, and in 4 of 88 patients in the low-dose heparin group. This difference was not statistically significant. Four of the 5 patients who had nonhemorrhagic strokes underwent 2-D echocardiography, which showed that 3 of the 4 had a left ventricular mural thrombosis. One patient in the low- dose group had a fatal pulmonary embolism.

There were no cases of transient ischemic attacks or peripheral arterial embolism during the 1 O-day in-hospi- tal period.

No statistically significant differences were found be- tween the high- and low-dose groups in the frequency of bleeding complications. Overall, 6 patients in the high- and 4 in the low-dose groups had hemorrhagic complica- tions. The only major bleeding complication observed was gastrointestinal hemorrhage, which occurred in 1 patient in each group.

Twenty-four patients died during the lo-day treat- ment period: 11 in the high- and 13 in the low-dose heparin groups. This difference was not significant. (As was pointed out earlier, mortality was not the primary outcome measured by this study. Thousands of patients would have to be treated and followed in order to detect any significant decrease in mortality.)

Cardiac arrest was the leading cause of death in both groups, but it appeared to occur earlier among patients in the high-dose group. The other causes of death included myocardial rupture, cardiogenic shock and pulmonary embolism-all of which are common causes of mortality among this patient population.

Study follow-up: As mentioned earlier, the patients

were observed for up to 1 year after infarction. (The course of therapy after the 10 days of treatment with either high- or low-dose calcium heparin was left to the discretion of the patients’ physicians.) Thirty of the 38 patients with positive results on 2-D echocardiograms were treated with anticoagulants for up to 6 months; the remaining patients were treated with antiplatelet agents. In none of the latter did systemic embolism develop dur- ing follow-up.

Overall, 4 patients had a stroke during the follow-up period-2 in each of the original treatment groups. In addition, 4 patients-3 in the original high- and 1 in the low-dose groups-had transient ischemic attacks. These cerebral events occurred between the eleventh day and the tenth month after the patient’s infarction.

None of these patients had positive results on 2-D echocardiograms or had received anticoagulant therapy after the tenth day.

With regard to long-term mortality, a total of 25 patients died during the follow-up period-13 in the high- and 12 in the low-dose groups. Again, this differ- ence was not significant. One patient in each group died of a stroke, and 1 patient in the low-dose group had a fatal pulmonary embolism 1 month after the acute MI. The deaths of the other 22 patients were due to either docu- mented MI or sudden death.

CONCLUSIONS The results of this study support the benefits of in-

hospital anticoagulant therapy for patients with acute MI. More specifically, this study showed that using high- dose subcutaneous calcium heparin (12,500 U every 12 hours) for patients with acute anterior transmural MI significantly decreased the incidence of left ventricular mural thrombosis. The regimen was not associated with increased incidence of bleeding complications. Further- more, the beneficial effects of this anticoagulant regimen appear to be related to plasma heparin concentration and to the corresponding APTT response, suggesting that these parameters should be monitored.

Thus, accumulating evidence now strongly supports the wider use of anticoagulant therapy during the imme- diate post-MI period, to reduce effectively and safely mural thrombosis formation and its potential conse- quences in terms of morbidity, associated thromboembol- ic complications and mortality.

REFERENCES 1. Report of the Working Party on Anticoagulant Therapy in Coronary Throm- bosis to the Medical Research Council. Assessment of short-term anticoagulant administration after cardiac infarction. Br Med J 1969;1:335-342. 2. Drapkin A, Merskey C. Anticoagulant therapy after acute myocardial infarc- tion. JAMA 1972;22:54/-548. 3. Results of a Cooperative Clinical Trial. Anticoagulants in acute myccardial infarction. JAMA 1973;225:724-729. 4. Chalmers TC, Matta RJ, Smith H, Kunzler AM. Evidence favoring the use of anticoagulants in the hospital phase of acute myocardial infarction. N Engl J Med 1977;297:20:1091-1096. 5. ISIS-2 Collaborative Group. Randomized trial of intravenous streptokinase, oral aspirin, both or neither, among 17,187 cases of suspected acute myocardial infarction: ISIS-2. Lancer 1988:2:349-360. 6. Turpie AGG, Robinson JG, Doyle DJ, Mulji AS, Mishkel GJ, Sealey BJ, Cairns JA, Hirsh J, Skingley L, Gent M. Prevention of left ventricular mural thrombosis in acute transmural anterior myocardial infarction: a double blind trial comparing high dose with low dose subcutaneous calcium heparin. N Eng/ .I Med 1989;320(6):352-357.

THE AMERICAN JOURNAL OF CARDIOLOGY FEBRUARY 2,199O 23C