JACC Vol. 14, No. 4 October 1989:877-84

877

Combined Tissue-Type Plasminogen Activator and Prostacyclin Therapy for Acute Myocardial Infarction

ERIC J. TOPOL, MD, FACC,* STEPHEN G. ELLIS, MD, FACC,”

ROBERT M. CALIFF, MD, FACC,t BARRY S. GEORGE, MD,$ DAVID C. STUMP, MD,9

ERIC R. BATES, MD, FACC,” ELIZABETH G. NABEL, MD,* JOSEPH A. WALTON, MD,*

RICHARD J. CANDELA, MD,S KERRY L. LEE, PHD,I EVA M. KLINE, RN,*

BERTRAM PITT, MD, FACC* AND THE THROMBOLYSIS AND ANGIOPLASTY IN MYOCARDIAL

INFARCTION (TAMI) 4 STUDY GROUPIE

Ann Arbor, Michigan; Durham, North Carolina; Columbus, Ohio and Burlington, Vermont

Current limitations of recombinant tissue-type plasminogen activator (r&PA) therapy for acute myocardial infarction include failure to achieve recanalization in 25% of patients, reocclusion and reperfnsion injury. Iloprost, a stable ana- logue of prostacyclin (PGI,), has been demonstrated to facilitate thrombolysis and reduce myocardial stunning in experimental models. To evaluate combined therapy, r&PA (100 mg 3 h) and Iloprost (2 nglkg per min for 48 h) were administered to 25 patients and then r&PA alone (same dose) was given to an additional 25 patients with evolving myocardial infarction. At 90 min after drug administration, infarct-related vessel patency was observed in I1 (44%) of 25 who received rt-PA plus Iloprost compared with 15 (60%) of 25 who received r&PA alone (p = 0.26). At 1 week, reocclusion had occurred in 3 (14%) of 21 patients who received combined therapy

compared with 6 (26%) of 23 patients treated with rt-PA alone (p = 0.46).

Ejection fraction increased significantly from baseline to 7 days for r&PA alone whereas it decreased with combined therapy (&PA alone: 47.3 -t 11.5% at baseline to 50.4 + 9.8% at 7 days; rt-PA plus Iloprost: 51.3 + 10.1% at baseline to 49.0 + 9.4% at 7 days; difference between groups p = 0.05). At 4 h after therapy, fibrinogen decreased 33% for r&PA plus Iloprost compared with a 52% for r&PA alone (p = 0.001). Fibrinogen degradation products increased 60% more for r&PA alone than for rt-PA plus Iloprost. Thus, the combination of &PA plus Iloprost at the doses employed did not improve immediate or follow-up coronary artery patency or left ventricular functional recovery com- pared with that achieved with &PA alone.

(J Am Co11 Cardiol1989;14:877-84)

There is abundant experimental evidence that intravenous prostacyclin administration has a beneficial effect in myocar- dial infarction. Prostacyclin (PGI,) has been demonstrated to

From the *Division of Cardiology, Department of Internal Medicine, University of Michigan Medical Center, Ann Arbor, Michigan; tDivision of Cardiology, Department of Medicine, Duke University Medical Center, Durham, North Carolina; #Riverside Methodist Hospital, Columbus, Ohio and IDivision of Hematology, Department of Medicine, University of Ver- mont, Burlington, Vermont. I’Coinvestigators and collaborating institutions are listed in the Appendix. This study was supported in part by Berlex Laboratories, Cedar Knoll, New Jersey and Grants HL-35058 and HL-01489 from the National Institutes of Health, Bethesda, Maryland. It was designed by the TAM1 investigators in collaboration with Berlex Laboratories.

Manuscript received January 17. 1989; revised manuscript received March 13, 1989, accepted April 13, 1989.

Address: Eric J. Topol, MD, Division of Cardiology, Bl-F245, University of Michigan Medical Center, 1500 E. Medical Center Drive, Ann Arbor, Michigan 48109-0022.

01989 by the American College of Cardiology

reduce infarct size (l-6), inhibit oxygen free radical forma- tion (7,8) and neutrophil function (9), thereby reducing reperfusion injury and myocardial stunning (10). Besides these important actions at the tissue and myocyte levels, prostacyclin is both the most potent endogenous platelet inhibitor known, capable of producing active platelet disag- gregation, and a relatively strong systemic and coronary artery vasodilator (11). These latter properties account for its acceleration of thrombolysis (12), potential to sustain infarct-related vessel patency, reduction in preload and afterload and improvement in collateral and infarct-related vessel blood flow (1,2,11).

Intravenous tissue-type plasminogen activator @-PA) has been shown to be an effective coronary thrombolytic agent, achieving recanalization of the infarct-related artery in 75% of patients (13-15), improving cardiac function (16-

0735-10971891$3.50

878 TOPOL ET AL. JACC Vol. 14, No. 4 PROSTACYCLIN AND rt-PA FOR MYOCARDIAL INFARCTION October 1989:877-84

19) and decreasing mortality rate in the acute phase of infarction (19). However, approximately 25% of patients (20,21) fail to achieve coronary artery recanalization within 90 min of administration of rt-PA, and the drug requires >45 min on average to be effective (13,14,22). Reocclusion is not infrequent, ranging from 13% to 33% without mechanical intervention (13,15,22-25), and improvement in left ventric- ular function has been modest (15-19,25). Thus, several of the major deficiencies of coronary thrombolytic therapy with rt-PA might hypothetically be compensated for or avoided by combined treatment with prostacyclin. In this pilot study, we evaluated the combination of rt-PA and a stable analogue of prostacyclin, Iloprost, in patients with evolving myocar- dial infarction. Our goal was to demonstrate the potential for this combination to facilitate infarct-related vessel patency, reduce the frequency of reocclusion and recurrent ischemic events and improve cardiac function.

Methods Study patients. Patients at two clinical sites were enrolled

from September 24, 1987 to March 1, 1988. Entry criteria included chest discomfort of 220 min but <6 h duration, electrocardiographic (ECG) ST segment elevation rl mm in two or more contiguous leads, age 575 years, no previous myocardial infarction in the same territory and no contrain- dication to thrombolytic therapy, as detailed in previous Thrombolysis and Angioplasty in Acute Myocardial Infarc- tion (TAMI) protocols (25,26). The protocol was approved by the institutional review boards at the participating insti- tutions, and patients gave informed consent before enroll- ment.

Protocol. For a pilot study, we planned to treat 25 patients with ‘combined &PA and Iloprost. After the pa- tient’s enrollment, one intravenous line was started in each forearm and the medications were reconstituted. Predomi- nantly single chain rt-PA (13) (Genentech, Inc.) was admin- istered in a dose of 1 mglkg body weight for the first h, with a maximal dose of 80 mg, and 10% of this dose was given as a bolus. The balance of the 100 mg dose was infused at a constant rate over the next 2 h. Iloprost was initially infused intravenously at a dose of 0.5 ng/kg per min. The dose was increased at 5 min intervals to 1 ng/kg per min and to 2 ng/kg per min, and was then kept constant for 48 h, provided that the patient’s blood pressure was stable and >85 mm Hg and no serious side effects occurred. For the treatment of hypo- tension, intravenous hydration was used before tapering or discontinuing the Iloprost. One hour before discontinuation of Iloprost, the infusion was gradually tapered to avoid the effect of sudden cessation of the drug.

After initiation of therapy, patients were transferred to the cardiac catheterization laboratory for angiographic de- termination of patency of the infarct-related vessel and baseline left ventricular function. After arterial access had

been obtained, heparin, 5,000 U, was administered intrave- nously. A single contrast injection into the infarct-related artery was performed at 60 min after therapy if possible. The primary end point of infarct-related vessel patency was determined at 90 min after therapy, utilizing up to four contrast injections of the vessel in multiple orthogonal hemiaxial views. Angiography of the noninfarct-related cor- onary arteries and the left ventricle was also performed.

If the infarct-related vessel was demonstrated to be patent (TIM1 grade 2 or 3 flow) (15,22,24), no coronary angioplasty was performed during the immediate cardiac catheterization. Patients whose infarct-related vessel was occluded at 90 min after therapy (TIM1 grade 0 or 1 flow) underwent “rescue” coronary angioplasty (26).

After transfer to the coronary care unit, patients were treated with oral aspirin (5 grains/day) and heparin (1,000 U/h) for at least 72 h, the dose adjusted to maintain the partial thromboplastin time at 2 to 2.5 times the normal value. Lidocaine was given prophylactically as a 75 mg bolus and 1 to 2 mgimin infusion during the first 24 h. Beta- adrenergic blockers, calcium channel blockers and nitrates were not given unless specifically required to treat ischemia, hypertension or rhythm disturbances.

At 7 days, patients returned to the cardiac catheterization laboratory for repeat left ventriculography and coronary angiography. If an interim event of recurrent ischemia had occurred, defined as ~20 min of chest discomfort, unre- lieved by nitroglycerin and accompanied by ECG changes, patients underwent emergency coronary angiography. In these patients, a third cardiac catheterization was performed before hospital discharge. After 25 patients were treated with rt-PA and Iloprost, the decision was made to treat an additional 25 patients with rt-PA alone but otherwise employ the identical protocol. The decision was based on the con- cern regarding the negative data obtained for combined therapy, as described later, which included a low infarct- related vessel patency rate, high abrupt closure frequency after rescue coronary angioplasty and minimal recovery of left ventricular function.

Angiographic studies. Cineangiographic films were for- warded to a core angiographic laboratory for blinded review by a single observer to determine infarct-related vessel TIM1 flow grade (22), percent diameter stenosis by quantitative angiography (27) and left ventricular function. Technically inadequate studies due to inadequate opacification or exces- sive extrasystoles were deleted from analyses. Quantitative percent diameter and absolute diameter stenosis were deter- mined with use of a computerized automatic edge detection algorithm (27). Global ejection fraction was determined by the area-length method (28) and regional wall motion of the infarct and noninfarct zones by the centerline chord method (29).

Assays of the hemostatic and fihrinolytic system. Blood samples were collected on 0.01 M citrate at baseline and 4,

JACC Vol. 14, No. 4 TOPOL ET AL. October 1989:877-X4 PROSTACYCLIN AND rt-PA FOR MYOCARDIAL INFARCTION

879

Table 1. Baseline Characteristics of the 50 Patients

No. of patients Age (yr) Gender (no.) Weight (kg) Time from chest pain to

therapy (h)

rt-PA t Iloprost

2s 51.0 ? 11.5 l9M, 6F

79.2 2 3.3 3.4 ? 0.3

r&PA Alone

2s 56.6 2 8.1 19M. 6F

81.0 + 2.8 2.1 5 0.2

Infarct-related vessel (no.) LAD 10 (40%) 1 I (44%) LCX 3 (12%) I (4%) RCA 12 (48%) I? (48%) SVG 0 1 (4%)

SBP at entry (mm Hg) 128.2 + 25.4 132.8 + 18.9 HR at entry (beatsimin) 74.0 + 18.2 75.2 ? 17.6

F = female; HR = heart rate; LAD = left anterior descending coronary artery; LCx = left circumflex coronary artery; M = male: RCA = right coronary artery; r&PA = recombinant tissue-type plasminogen activator: SBP = systolic blood pressure: SK = saphenous vein graft.

12, 24 and 48 h after the initiation of rt-PA therapy. The 4 h sample also contained 200 KIU/ml of aprotinin to minimize in vitro fibrinolytic artifact. Samples were processed as previously described (30). Briefly, after immediate centrifu- gation for harvesting of plasma, samples were kept frozen at -20°C until assayed for fibrinogen by a coagulation rate assay (31) and for fibrinogen degradation products by agglu- tination inhibition of tanned red blood cells (32). The rt-PA plasma level at 4 h after therapy was measured by antigenic assay (30).

Data analysis. Data are expressed as mean values t SD, unless otherwise stipulated. The differences for infarct- related vessel patency, reocclusion and side effects were compared by chi-square test or, if the expected cell size was <5, Fisher’s exact test. Serial studies of left ventricular function by contrast ventriculography were evaluated by paired t test for intragroup comparison and unpaired t test for comparison between groups. The changes in fibrinogen degradation products were compared with use of a Wilcoxon signed-rank test.

Results Patient characteristics. The baseline demographic, angio-

graphic and hemodynamic variables are presented in Table 1. One additional patient received r&PA and Iloprost but was subsequently deleted from the analysis because of a lack of confirmation of myocardial infarction by serum enzyme or ECG criteria. Angiography in this patient demonstrated normal coronary arteries.

The two treatment groups were comparable except for a 0.7 h delay in therapy in the combined rt-PA and Iloprost group, which was accounted for in part by the protracted preparation time required to reconstitute this medication (2

Patency (95% C.1.l

??I-PA 8 lloprost 44% (24,65)

50 1

&j t-PA 60% (39,79)

40

m

g 30

z 0.

20

s

10

0 0 1 2 3

TIMI Grade

Figure 1. Infarct-related vessel patency by Thrombolysis in Myo- cardial Infarction (TIMI) grade in the two treatment groups. The 95% confidence limits ((2.1.) for the two treatment regimens are demonstrated to be overlapping.

mg vials). Infarct-related vessel patency at 90 min occurred in 1 I (44%) of the 25 rt-PA plus Iloprost-treated patients and 15 (60%) of the 25 patients treated with rt-PA alone (p = 0.26) (Fig. 1).

Clinical course and outcome (Tables 2 and 3). Early infarct-related vessel patency was established in 21 patients who received rt-PA plus Iloprost therapy and in 23 patients who were treated with rt-PA alone. Patency was demon- strated during the course of the four serial coronary angio- grams in three patients receiving combined therapy com- pared with one who received rt-PA alone. Reocclusion was subclassified as occurring after pharmacologic reperfusion after mechanical (balloon angioplasty) intervention and ac- cording to when it occurred (Table 2). There were no significant differences for reocclusive episodes between the two patient groups, particularly when abrupt closures in the cardiac catheterization laboratory were considered. After discharge from the laboratory, the decreased frequency of reocclusion in patients receiving combined therapy was related to pharmacologic reperfusion.

There were no in-hospital deaths. The adverse events possibly related to drug therapy and bleeding complications are presented in Table 3. The patients treated with combined rt-PA and Iloprost had an excess of severe nausea and vomiting (p = 0.001). Significant hypotension led to tapering of Iloprost dosage in five patients and discontinuation in an additional three patients. As expected, there was a more significant overall decrease in blood pressure and reflex tachycardia in the patients who received combined therapy. The transfusion requirement was similar for the two groups, and the combined rt-PA and Iloprost therapy did not lead to an increased frequency of bleeding complications.

Left ventricular function. Global and regional left ven- tricular function results at baseline and 7 days are shown in

880 TOPOL ET AL. JACC Vol. 14, No. 4 PROSTACYCLIN AND rt-PA FOR MYOCARDIAL INFARCTION October 1989:877-84

Table 2. Clinical and Angiographic Outcomes in the Two Treatment Groups

&PA = lloprost rt-PA Alone (n = 25) (n = 25)

Coronary artery patency First 90 min injection 11 (44%) 15 (60%) Final diagnostic injection 14 (56%) 16 (64%) At end of procedure* 21 (84%) 23 (92%)

Abrupt closure in 5 (24%) 2 (9%) laboratoryt

Reocclusion$ 3 (14%) 6 (26%) Reocclusion after lysis only 1114 (7%) 4116 (25%) Reocclusion after PTCA 2/7 (29%) 217 (29%) Emergency CABG 1 (4%) 1(4%) Death 0 0

*Denotes that the patient was discharged from the cardiac catheterization laboratory with a patent infarct-related vessel (TIM1 grade 2 or 3 flow pattern). tDenotes the transient cessation of infarct-related vessel flow (TIM1 grade 0) after rescue angioplasty, requiring repeat balloon dilation. fDefined as an angiographically confirmed event after the patient was discharged from the cardiac catheterization laboratory. CABG = coronary artery bypass grafting; FTCA = percutaneous transluminal coronary angioplasty; rt-PA = recombinant tissue-type plasminogen activator.

Figures 2 and 3, respectively. In 25 patients treated with rt-PA plus Iloprost, baseline ejection fraction was 51.3 + lO.l%, infarct zone wall motion, expressed as standard deviation (SD) per chord, was -2.73 + 0.92 SD/chord and noninfarct zone wall motion was 1.31 -+ 1.09 SD/chord. At follow-up study in 23 patients, ejection fraction was 49.0 t 9.4%, infarct zone wall motion was -2.58 2 0.93 SD/chord and noninfarct zone wall motion was 0.61 -+ 1.08 SD/chord. Ejection fraction in 23 paired studies decreased 1.8 + 5.3 percentage points (p = 0.12), and infarct zone wall motion did not significantly improve (0.21 -t 0.90 SD/chord, p =

Table 3. Side Effects in the 50 Patients

r&PA = Iloprost &PA Alone (n = 25) (n = 25) p Value

Flushing 3 (12%) I (4%) Headache 8 (32%) 4 (16%) Bleeding (~2 U) 5 (20%) 5 (20%) Nausea/vomiting* 19 (76%) 7 (28%) 0.001 Antiemetic use 13 (54%) 4 (18%) 0.08 Vomiting I6 (67%) 7 (30%) 0.02 SBP (4 h)

mm Hg 113 + 19 122 k 21 <O.OOl % change 12.5% 1 7%1

HR (4 h) beats/min 85 + 12 802 15 % change 13% t 6% t

*Defined as protracted nausea and vomiting lasting >I h documented in the medical record. r&PA = recombinant tissue-type plasminogen activator; t = increase; 4 = decrease; SBP = systolic blood pressure.

??t-PA and lloprost 60 6

8 60 5

E =

40 4

I c 30 3

s z 20 2

S 5 10 1

0 0

-1

Figure 2. Effects of rt-PA and Iloprost versus rt-PA alone on recovery of global left ventricular function. The baseline and 7 day ejection fractions (left y axis) are plotted, as is the change (A) in ejection fraction (right y axis). Note the significant difference (4 in recovery of ejection fraction between the two interventions (p = 0.05).

0.28). Noninfarct zone wall motion regressed significantly (0.69 t 0.74 SD/chord, p < 0.001).

For 24 patients who received rt-PA alone, baseline global ejection fraction was 47.3 + 11.5%, infarct zone wall motion was -3.03 ? 0.56 SD/chord and noninfarct zone wall motion was 1.52 + 1.9 SD/chord. At the 7 day repeat study, the

Figure 3. Regional wall motion of the infarct zone expressed as standard deviation (SD) per chord using the centerline chord method (29). The patients receiving rt-PA alone had significant improvement in regional wall motion; those receiving rt-PA plus Iloprost had a slight but statistically insignificant improvement. The absolute im- provement (A) from baseline to 7 day study of wall motion was three-fold higher for patients receiving &PA alone compared with those receiving the combination.

2 1 ??t-PA and lloprost ??t-PA

1 0.76

p - 0.26

JACC Vol. 14. No. 4 TOPOL ET AL. October 1989:877-84 PROSTACYCLIN AND n-PA FOR MYOCARDIAL INFARCTION

881

ventriculograms obtained in six of these patients were not technically adequate for analysis and one patient refused repeat cardiac catheterization. At follow-up study in 18 patients, ejection fraction was 50.4 ? 9.8%. infarct zone wall motion was -2.13 2 1.05 SD/chord and noninfarct zone wall motion was 0.72 ? 1.67. In 18 paired studies, ejection fraction increased 2.3 ? 7.9 percentage points (p = 0.22). infarct zone wall motion improved 0.76 ? 0.89 SD/chord (p = 0.002) and noninfarct zone wall motion regressed 1.09 t- 1.39 (p = 0.004). The recovery in ejection fraction, as measured by the difference from baseline to follow-up study, was reduced for combined rt-PA and Iloprost compared with rt-PA alone (p = 0.05) and infarct zone wall motion recovery was also diminished with combination therapy (p = 0.05).

When on1.v patients in whom pharmacologic reperfusion Cut-PA plus Iloprost or rt-PA alone) was successful were considered, the same trends for a decreased recovery of left ventricular function in the combined rt-PA and Iloprost treatment group were observed. For patients with an open infarct-related vessel at early catheterization and paired, technically adequate ventriculographic studies, the median ejection fraction at 90 min of therapy was 48.5% for 14 patients receiving rt-PA plus Iloprost and 48.5% for 16 patients who received rt-PA alone. At 7 days, the ejection fraction was 48.5% and 52.0%, respectively. Regional wall motion of the infarct zone at 90 min for the two groups was -2.96 SD/chord and -3.07 SD/chord for combined therapy and rt-PA alone groups, respectively. At follow-up study, the median values were -2.89 SD/chord and -2.05 SD/ chord, respectively. This represents no improvement for the combined treatment group but a 33% recovery of regional function in the patients receiving rt-PA alone.

Quantitative coronary angiography. Technically ade- quate paired quantitative coronary angiographic data with- out total occlusion were available in 13 patients who re- ceived rt-PA plus Iloprost and 14 patients treated with rt-PA alone. Patients with an occluded infarct-related artery at initial cardiac catheterization (6 patients) or those with at least one inadequate study for quantitative analysis (11 patients) were excluded. The absolute mean diameter in- creased from 0.68 of: 0.64 mm at baseline to 0.84 ? 0.54 mm at 7 days for rt-PA plus Iloprost compared with 0.93 t 0.44 to 0.87 + 0.67 mm, respectively, for rt-PA alone. The baseline and follow-up quantitative infarct-related vessel stenosis rates were not significantly different between the groups (p = 0.24 and 0.89. respectively). Over the 7 days, percent diameter stenosis decreased from 78.9 t 17.3% to 71.1 z 18.0% for patients treated with rt-PA plus Iloprost and increased slightly from 71 .O 5 13.2% to 72.9 2 19.9% for patients receiving rt-PA alone.

Effects on hemostasis and fihrinolysis. The effects of the treatment regimen on fibrinogen and fibrinogen degradation products are summarized in Table 4. Of note, the baseline fibrinogen level was higher in the patients receiving rt-PA

Table 4. Effect on Hemostatic Variables in the Two Treatment Groups

No. of patients

Fibrinogen (mgidl)

IT-PA t lloprost

25

rt-PA Alone

19*

Baseline

4 h (% change)

12 h (‘ST change)

FDP

2.70 + 0.52 3.45 ? 1.07

1.79 + 0.65 (33%)** 1.65 -+ 0.89 (52%)t

2.09 _c 0.58 (22%) I .60 2 0.98 (53%)

Baseline 0 0

4 h titer (median) 0: lOO$

I.! h titer (median) 5$ 33$

ri-PA antigen (4 h) &/ml) 547.4 + 689.9 406 + 314.3

*Six patients in the &PA group who subsequently received urokinase

were deleted from this analysis. tp = 0.001 I. $p < 0.005 by Wilcoxon

analysis. FDP = fibrinogen degradation products; IT-PA = recombinant tissue

plasminogen activator.

alone. In addition, the decline in fibrinogen and development of fibrinogen degradation products over 12 h was signifi- cantly greater in this group. The plasma rt-PA levels at 4 h after initiation of therapy were highly variable, but were not found to be lower in the patients treated with &PA alone; peak samples were not obtained.

Discussion Lack of enhancement of thrombolysis with Iloprost. Al-

though this study was relatively small, sequential and a pilot study by design, our findings raise several issues that must be considered in the development of new adjunctive phar- macologic strategies for the treatment of acute myocardial infarction. First, we tested an agent, Iloprost, which holds much promise for enhancing thrombolysis, sustained infarct- related vessel patency and ventricular functional recovery (9,10.12). The combination of rt-PA and this prostacyclin analogue led to disappointing results, with an overall low coronary artery patency rate and less recovery of ventricular function than that obtained with treatment with rt-PA alone. As these findings were investigated, we noted substantially less degradation of the hemostatic proteins by rt-PA plus Iloprost. Unfortunately, peak plasma r&PA levels (50 to 60 min after therapy) were not obtained in the current study. Recently, Kerins et al. (33) demonstrated that under exper- imental steady state plasma conditions of exogenously ad- ministered rt-PA, an infusion of Iloprost led to an abrupt 31% reduction in the plasma rt-PA level, which promptly reverted to steady state levels after cessation of Iloprost. Their study corroborates the clinical findings that we ob- served, and together these studies support the possibility of an in vivo pharmacokinetic interaction. The precise mecha- nism for such an interaction has not been elucidated, but likely involves Iloprost’s effect of increased hepatic blood flow and potential secondary accelerated clearance of rt-PA.

882 TOPOL ET AL. JACC Vol. 14, No. 4 PROSTACYCLIN AND H-PA FOR MYOCARDIAL INFARCTION October 1989:877-84

Nonetheless, our results point out the importance of per- forming careful experimental pharmacokinetic testing of new drug combinations in this setting before their introduction to patients with acute myocardial infarction.

Lack of improvement in ventricular function. Second, at the maximal dose generally tolerated of Iloprost (2 nglkg per min) in our clinical study, there was no evidence of improve- ment in ventricular function, even in patients with a patent infarct-related vessel throughout their hospitalization. Pre- vious experimental studies (9,10,34) suggest that at doses of Iloprost ranging from 5 to 20 ng/kg per min, reperfusion injury and myocardial stunning can be markedly abated by inhibition of neutrophil activity. The current study repre- sents the first clinical study of thrombolytic therapy coupled with an agent purported to reduce reperfusion injury. In our study, the two agents were administered simultaneously, providing an ample concentration of Iloprost to be present before reperfusion was achieved, and the infusion was continued for 48 h. These negative results in terms of no potentiation of ventricular function and our inability to replicate the data obtained in the experimental setting may be due to inadequate doses of Iloprost administered relative to the studies cited or to inability to inhibit reperfusion injury in the clinical setting. With respect to the potential of inadequate dosing, it is unlikely that most patients can tolerate even higher doses of Iloprost, because several patients in the current study experienced either significant hypotension or severe nausea and vomiting. Furthermore, we observed a hemodynamic effect of Iloprost with respect to lowering systolic blood pressure. The decrease in after- load in the acute phase may have led to a higher initial ejection fraction, thus reducing the chance to demonstrate recovery of left ventricular function.

Lack of prevention of reocclusion. Third, reocclusion occurred despite the maintenance infusion of Iloprost. Al- though experimental studies have demonstrated Iloprost’s potent platelet disaggregative properties (35,36), and reoc- elusion after thrombolysis may be partially a platelet- mediated event (37), several patients in our study experi- enced abrupt coronary artery closure while in the cardiac catheterization laboratory. These reocclusive episodes oc- curred in the setting of concomitant Iloprost therapy, intra- venous heparin infusion, aspirin administration and recent or ongoing rt-PA therapy. Recently, Yasuda and Gold and colleagues (38,39) studied combined rt-PA and the platelet (7E3) monoclonal antibody to the glycoprotein IIAlIIIB receptor. These investigators demonstrated that, in the experimental model, reocclusion could be abolished at cer- tain dose combinations. Similarly, acceleration of coronary artery recanalization was shown with combined throm- bolytic and platelet antagonist therapy compared with rt-PA alone. Although we were unable to determine the extent of in vivo platelet inhibition in our patients, previous clinical studies (40) have demonstrated marked disaggregation with

the doses of Iloprost employed. However, in vitro experi- ments (41) with rt-PA and Iloprost have demonstrated less antiplatelet effect with the combination compared with rt-PA alone. Certainly, in the current study, the stimulus for reocclusion was not substantially reduced by this prostacy- clin analogue. From such experiments and our clinical observations, it appears that the process of thrombolysis may actually serve a platelet-stimulating role (42,43). Fur- thermore, the infarct-related vessel patency rate was not high or accelerated compared with that with rt-PA alone. In both treatment groups, a low patency rate was demonstrated compared with that in several clinical trials testing the same dose of this predominantly single chain rt-PA preparation. However, because of the small sample size, the confidence intervals for patency are especially wide (Fig. 1).

Conclusions. Recent large scale clinical trials (17-20) have emphasized the fundamental value of thrombolytic therapy for improved survival and ventricular function and reduction in infarct size. In the Second International Study of Infarct Survival (ISIS-2) (44), the importance of early treatment with aspirin in conjunction with streptokinase for mortality reduction was clearly demonstrated. The role of other adjunctive pharmacologic agents to coronary throm- bolysis is the focus of many ongoing clinical trials. Our results suggest that at the maximal clinically tolerated dose of Iloprost, ventricular functional recovery is not aug- mented. Although the potential rt-PA and Iloprost in vivo pharmacokinetic interaction may be overridden by increased rt-PA dosing, this may not be achieved without compromis- ing patient safety (45). The negative findings of our pilot study should not be generalized to other prostacyclin ana- logues or other thrombolytic enzymes (46). More experimen- tal and clinical investigation will be necessary for coupling thrombolytic therapy and prostacyclin or other agents that antagonize platelet function or potentially limit reperfusion injury.

Appendix The TAM1 Study Group

University of Michigan: Eric J. Topol, MD (Principal Investigator), Steuhen G. Ellis. MD. Josenh A. Walton. MD. Eric R. Bates. MD. Elizabeth G. ‘Nabel, MD, Eva Kline: RN, BSN, Laura Gorman, RN; BSN, Barbara Schumacher, BS, Bertram Pitt, MD. Satellite Center: Foote Hospital, Jack- son, Michigan: Gregory Baumann MD, John Maino, MD, Mary Ann Men- eleson. MD. Constance Dovle. MD. Patricia Lamb. MD.

Riverside Methodist f&pit& Barry S. George, MD, Richard J. Candela, MD, Joanne Dillon, RN, BS, Ronald D. Frazier, MD, Ramona Masek, RN, BS, Ann Pickel, RN.

Duke University Biostatistical Core Laboratory: Kerry L. Lee, PhD, Lynne Aronson, BS, Jane M. Boswick, MPH, Joy Miller, RN, Kristina Sigmon, MA, Lynn Harrelson, MS, Robert M. Califf, MD.

University of Michigan Core AngioeraDhic Laboratory: Darrell Debowey, MS, John Kt&kel, B&-Stephen G. Ellis, MD, Eric J. Topol, MD.

Data Monitoring Committee: Mark Hlatkv. MD. Daniel B. Mark, MD, MPH, Kerry L. Lee, PhD.

University of Vermont Core Hematology Laboratory: David Stump, MD, Dagnija Thornton, BS.

JACC Vol. 14. No. 4 October 1989:877-84

TOPOL ET AL. 883 PROSTACYCLIN AND n-PA FOR MYOCARDIAL INFARCTION

References 19.

1. Chiariello M, Golino P, Cappelli-Bigazzi M, Ambrosio G, Tritto I. Salvatore M. Reduction in infarct size bv the urostacvclin analoaue 20.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

, . . iloprost (ZK 36374) after experimental coronary artery occlusion- reperfusion. Am Heart J 1988:115:499-504.

Smith EF Ill, Gallenkamper W. Beckmann R, Thomsen T. Mannesmann G, Schror K. Early and late administration of a PGI?-analogue, ZK 36 374 (Iloprost): effects on myocardial preservation, collateral blood flow and infarct size. Cardiovasc Res 1984;18:163-73.

Melin JA, Becker LC. Salvage of ischemic myocardium by prostacyclin during experimental myocardial infarction. J Am Coil Cardiol 1983: 2r279-86.

Jugdutt BI, Hutchins GM, Bulkley BH, Becker LC. Dissimilar effects of prostacyclin, prostaglandin E, and prostaglandin E?, on myocardial infarct size after coronary occlusion in conscious dogs. Circ Res 1981; 49:685-700.

Lefer AM, Ogletree ML, Smith JB, et al. Prostacyclin: a potentially valuable agent for preserving myocardial tissue in acute myocardial ischemia. Science 1978:200:52-4.

Ribeiro LGT, Brandon TA. Hopkins DG, Reduto LA, Taylor AA, Miller RR. Prostacyclin in experimental myocardial ischemia: effects on hemo- dynamics, regional myocardial blood flow, infarct size and mortality. Am J Cardiol 1981;47:835-40.

Fantone JD, Kinnes DA. Prostaglandin E, modulation of superoxide production by human neutrophils. Biochem Biophys Res Commun 1983; 113:506-12.

Thiemermann C. Steinhagen-Thiessen E. Schror K. Inhibition of oxygen- centered free radical formation by the stable prostacyclin-mimetic Ilo- prost (ZK 36 374) in acute myocardial ischemia. J Cardiovasc Pharmacol 1984;6:365-6.

Simpson PJ, Mickelson J, Fantone JC, Gallagher KP, Lucchesi BR. Iloprost inhibits neutrophil function in vitro and in vivo and limits experimental infarct size in canine heart. Circ Res 1987:60:666-73.

Farber NE, Pieper GM, Thomas JP, Gross GJ. Beneficial effects of Iloprost in the stunned canine myocardium. Circ Res 1988;62:204-15.

Bugiardini R, Galvani M, Ferrini D, et al. Myocardial ischemia during intravenous prostacyclin administration: hemodynamic findings and pre- cautionary measures. Am Heart J 1987;133:23~40.

Schumacher WA, Lee EC, Lucchesi BR. Augmentation of streptokinase thrombolysis by heparin and prostacyclin. J Cardiovasc Pharmacol 1985; 7:739-46.

Top01 EJ, Morris DC, Smalling RW, et al. A multicenter, randomized, placebo controlled trial of a new form of intravenous recombinant tissue-type plasminogen activator (Activase) in acute myocardial infarc- tion. J Am Co11 Cardiol 1987:9:1205-13.

Collen D, Top01 EJ. Tiefenbrunn AJ, et al. Coronary thrombolysis with recombinant human tissue-type plasminogen activator: a prospective, randomized, placebo-controlled trial. Circulation 1984;lO: 1012-7.

Passamani E, Hodges M, Herman M, et al. The Thrombolysis in Myocardial Infarction (TIMI) phase II pilot study: tissue plasminogen activator followed by percutaneous transluminal coronary angioplasty. J Am Coll Cardiol 1987;10:51B-64B.

Guerci AD, Gerstenblith G, Brinker JA, et al. A randomized trial of intravenous tissue plasminogen activator for acute myocardial infarction with subsequent randomization to elective coronary angioplasty. N Engl J Med 1987:317:1613-8.

O’Rourke M. Baron D, Keogh A, et al. Limitation of myocardial infarction by early infusion of recombinant tissue-type plasminogen activator. Circulation 1988;77:131 l-5.

National Heart Foundation of Australia Coronary Thrombolysis Group. Coronary thrombolysis and myocardial salvage by tissue plasminogen activator given up to 4 hours after onset of myocardial infarction. Lancet 1988:1:203-8.

21.

22.

23.

24.

25.

26.

27.

28

29

30

31

32

33

Van de Werf F. Lessons from the European Cooperative Recombinant Tissue-Type Plasminogen Activator @t-PA). J Am Coll Cardiol 1988; 12:14A-9A.

Top01 EJ, Califf RM, Kereiakes DJ. George BS. Thrombolysis and Angioplasty in Myocardial Infarction (TAMI) trial. J Am Coll Cardiol 3987:10:65B-74B. Califf RM. Top01 EJ, George BS, et al. Characteristics and outcome of patients in whom reperfusion with intravenous tissue-type plasminogen activator fails: results of the Thrombolysis and Angioplasty in Myocardial Infarction (TAMI) I trial. Circulation 1988;77: 1090-9.

Williams DO, Borer J, Braunwald E, et al. Intravenous recombinant tissue-type plasminogen activator in patients with acute myocardial infarction: a report from the NHLBI Thrombolysis in Myocardial Infarc- tion trial. Circulation 1986;73:338-46.

Gold HK, Leinbach RC, Garabedian HD. et al. Acute coronary reocclu- sion after thrombolysis with recombinant human tissue-type plasminogen activator: prevention by a maintenance infusion. Circulation 1986:73: 347-52.

Chesebro JH, Knatterud G, Roberts R. et al. Thrombolysis in Myocardial Infarction (TIMI) trial, phase I: a comparison between intravenous tissue plasminogen activator and intravenous streptokinase. Circulation 1987; 76: 142-54.

Top01 EJ, Califf RM, George BS, et al. A randomized trial of immediate versus delayed elective angioplasty after intravenous tissue plasminogen activator in acute myocardial infarction. N Engl J MEd 1987:317:581-8.

Top01 EJ, Califf RM, George BS, et al. Coronary arterial thrombolysis with combined infusion of recombinant tissue-type plasminogen activator and urokinase in acute myocardial infarction. Circulation 1988:77: 1100-7. Mancini GBJ, Simon SB. McGillem MJ, LeFree MT, Friedman HZ, Vogel RA. Automated quantitative coronary arteriography: morphologic and physiologic validation in vivo of a rapid digital angiographic method. Circulation 1987;75:452-60.

Sander H, Dodge HT. The use of single plane angiocardiograms for the calculation of left ventricular volume in man. Am Heart J 1968;75:325-34.

Sheehan FH, Bolson EL, Dodge HT, Mathey DG, Schafer J, Woo HW. Advantages and applications of the centerline method for characterizing regional ventricular function. Circulation 1986;74:293-305.

Stump DC, Top01 EJ, Chen AB, Hopkins A, Collen D. Monitoring of hemostasis parameters during coronary thrombolysis with recombinant tissue-type plasminogen activator. Thromb Haemostas 1988;59:133-7.

Clauss A. Gerinnungsphysiologische Schnellmethode zur Bestimmung des Fibrinogens. Acta Haematol 1957;37:237-46. Merskey C. Lalezari P, Johnson AJ. A rapid, simple, sensitive method for measuring fibrinolytic split products in human serum. Proc Sot Exp Biol Med 1969:131:871-5.

Kerins DM, FitzGerald GA, Fitzgerald DJ. Interaction of a prostacyclin analog and tissue plasminogen activator @t-PA) during coronary throm- bolysis (abstr). Clin Res 1989;37:4A.

34. Schror K, Darius H, Matzky R, Ohlenforf R. The antiplatelet and cardiovascular action of a new carbacyclin derivative (ZK 36374) equiv- alent to PGIz in vitro. Arch Phannacol 1981:316:252-j.

35. Romson JL. Haack DW, Abrams GD, Lucchesi BR. Prevention of occlusive coronary artery thrombosis by prostacyclin infusion in the dog. Circulation 1981:64:90614.

36. Schror K, Darius H, Ohlendorf R, Matzky R, Klaus W. Dissociation of antiplatelet effects from myocardial cytoprotective activity during acute myocardial ischemia in cats by a new carbacyclin derivative (ZK 36375). J Cardiovasc Pharmacol 1982;4:554-61.

37. Johns JA. Gold HK. Management of coronary reocclusion following successful thrombolysis. In: Top01 EJ, ed. Acute Coronary Intervention. New York: Alan R. Liss, 1987:95-106.

38. Yasuda T, Gold HK, Fallon JT, et al. Monoclonal antibody against the platelet glycoprotein (GP) IIb/IIIa receptor prevents coronary artery reocclusion after reperfusion with recombinant tissue-type plasminogen activator in dogs. J Clin Invest 1988;81:128~91.

884 TOPOL ET AL. PROSTACYCLIN AND rt-PA FOR MYOCARDIAL INFARCTION

JACC Vol. 14, No. 4 October 1989:877-84

39. Gold HK, Caller BS, Yasuda T, et al. Rapid and sustained coronary artery recanalization with combined bolus injection of recombinant tis- sue-type plasminogen activator and monoclonal antiplatelet GPIIb/IlIa antibody in a canine preparation. Circulation 1988;77:67@-7.

40. Tremoli E, Madema P, Mannucci L, Colli S, Paoletti R. In vitro effects of Iloprost on platelet aggregation in normal and hypercholesterolemic subiects. In: Grvalewski RJ. Stock G. eds. Prostacvclin and Its Stable Analogue Ilopros; New York: Springer-Verlag, 1987:47-51.

41. Shebuski RJ, Ohlstein EH. Attenuation of platelet responsiveness to prostacyclin (PGI,) after tissue-type plasminogen activator (abstr). Cir- culation 1987;76(suppl IV):IV-338.

44. ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Randomized trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. J Am Coll Cardiol 1988;12:3A-l3A.

45. Braunwald E, Knatterud GL, Passamani ER, Robertson TL. Announce- ment of protocol change in Thrombolysis in Myocardial Infarction trial (letter). J Am Coll Cardiol 1987;9:467.

42. Fitzgerald DJ, Louis R, Wright F, FitzGerald FA. Functional significance 46. Kawaguchi H, Yasuda H. Effect of various plasminogen activators on of platelet activation following coronary thrombolysis (abstr). Circulation prostacyclin synthesis in cultured vascular cells. Circ Res 1988; 1987;76(suppl IV):IV-151. 6331029-35.

43. Badimon L, Lassila R, Badimon J, Vallabhajosula S, Chesebro JH, Fuster V. Residual thrombus is more thrombogenic than severely dam- aged vessel wall (abstr). Circulation 1988;78(suppl II):II-119.