Recombinant Tissue Plasmhogen Activator in Acute Thrombotic and Embolic Stroke Gregory J. del Zoppo, MD," Klaus Poeck, MD,? Michael S. Pessin, MD,$ Samuel M. Wolpert, MD,S Anthony J. Furlan, MD,§ Andreas Ferbert, MD,? Mark J. Alberts, MD,"Justin A. Zivin, MD, PhD,' Lawrence Wechsler, MD,"" Otto Busse, MD, I? Ralph Greenlee, Jr, MD,$$ Lawrence Brass, MD,§§ J. P. Mohr, MDJIIl Edward Feldmann, MD,## Werner Hacke, MD,""" Carlos S. Kase, MD,t?t Jose Biller, MD,S$$ Daryl Gress, MD,§§§ and Shirley M. Otis, MD"

An open angiography-based, dose rate escalation study on the effect of intravenous infusion of recombinant tissue plasminogen activator (rt-PA) on cerebral arterial recanalization in patients with acute focal cerebral ischemia was performed at 16 centers. Arterial occlusions consistent with acute ischemia in the carotid or vertebrobasilar territory in the absence of detectable intracerebral hemorrhage were prerequisites for treatment. After the @-minute rt-PA infusion, arterial perfusion was assessed by repeat angiography and computed tomography scans were performed at 24 hours to assess hemorrhagic transformation. Of 139 patients with symptoms of focal ischemia, 80.6% (112) had complete occlusion of the primary vessel at a mean of 5.4 f 1.7 hours after symptom onset. No dose rate response of cerebral arterial recanalization was observed in 93 patients who completed the rt-PA infusion. Middle cerebral artery division (M2)and branch (M3)occlusions were more likely to undergo recanalization by 60 minutes than were internal carotid artery occlusions. Hemorrhagic infarction occurred in 20.2% and parenchymatous hematoma in 10.6% of patients over all dose rates, while neurological worsening accompanied hemorrhagic transformation (hemorrhagic infarction and parenchymatous hematoma) in 9.6%of patients. All findings were within prospective safety guidelines. No dose rate correlation with hemorrhagic infarction, parenchymatous hematoma, or both was seen. Hemorrhagic transformation occurred significantly more frequently in patients receiving treatment at least 6 hours after symptom onset. No relationship between hemorrhagic transformation and recanalization was observed. This study indicates that site of occlusion, time to recanalization, and time to treatment are important variables in acute stroke intervention with this agent. del Zoppo GJ, Poeck K, Pessin MS, Wolpert SM, Furlan AJ, Ferbert A, Alberts MJ, Zivin JA, Wechsler L, Busse 0, Greenlee R Jr, Brass L, Mohr JP, Feldmann E, Hacke W, Kase CS, Biller J, Gress D, Otis SM. Recombinant tissue plasminogen activator in acute thrombotic and embolic stroke. Ann Neurol 1792;32:78-86

Studies using angiography during the early hours of stroke have shown a high frequency of atherothrombotic or embolic arterial occlusions El-51. Recanalization can follow intraarterial catheter-directed infusion of the thrombolytic agents urokinase (u-PA) and streptokinase proximal to a cerebral arterial occlusion [I, 3, 4, 61, and significant reduction in infarct volume [4] and improvement in clinical outcome [3] may follow

recanalization. That experience provides a practical background for the evaluation of the relatively fibrinspecific agent recombinant tissue plasminogen activator (rt-PA) in acute thromboembolic stroke. Little information exists regarding the risks of any thrombolytic agent in acute atherothrombotic or embolic stroke. While early studies of stroke patients who received intravenous thrombolytic agents suggested an

From the *Scripps Clinic and Research Foundation, La Jolla, CA; tKLinikum, Rheinisch Westfalische Technische Hochschule, AaFederal Republic Of Germany; Center, Boston, MA; §The Cleveland Clinic Foundation, Cleveland, OH; 'Duke University Medical Center, Durham, NC; 'University of California, San Diego, San Diego, CA; **Universityof Pittsburgh, Pittsburgh. PA; ttKlinikum I Minden, Minden. Federal ReDublic of Germany; tiuniversity of Texas-Southwestern Medical Center, Dallas, TX; §§Yale University School of Medicine, New Haven, CT;'I ilNeurologicalInstitute, Columbia University, New York, NY; #%ode Island Hospital, Providence, RI;***University of Heidelberg, Heidelberg, Federal Republic of Germany; tttBoston University Medical Center, Boston, MA; $*$University of Iowa College of Medicine, Iowa City, IA; and. §§§Massachusetts General Hospital, Boston, MA.

Received Oct 8, 1991, and in revised form Dec 19. Accepted for publication Jan 16, 1992. Address correspondence to Dr del Zoppo, Department of Molecular and Experimental Medicine, scripps clinic and Research Foundation, 10666 No,.& T~~~~~ pines~ ~La ~ d ~CA , u92037, ~ ,

78 Copyright 0 1972 by the American Neurological Association

increased risk of intracerebral hemorrhage, serious shortcomings in study design have justified a reappraisal of the efficacy and safety of this approach {7]. Limited animal model studies support this reappraisal IS, 91. Before a study of the clinical efficacy of rt-PA in acute cerebral ischemia can be undertaken, it is necessary to determine the dose rate that would optimize cerebral arterial recanalization and minimize the risk of intracerebral hemorrhagic complications. Accordingly, we have performed a prospective, open, multicenter safety and recanalization study of the intravenous infusion of rt-PA at increasing dose rates in patients with acute, symptomatic, angiographically documented, cerebrovascular arterial occlusion.

Materials and Methods The study protocol and the informed consent procedures for participation in this study were approved by the Institutional Review Board or Ethics Commission in each of the institutions participating in this multicenter study. The rt-PAIAcute Stroke Study Group consisted of 16 major institutional centers in North America and the Federal Republic of Germany (see Appendix).

Patient Selection Between June 1987 and May 1990, all patients with symptoms consistent with acute focal cerebrovascular ischemia arriving at the study group institutions were screened for participation in this study. Entry was confined to patients 21 to 80 years old with no significant prior impairment of neurological function. All patients were required to have symptoms of acute focal ischemia, occurring no longer than 8.0 hours from the expected time of initial rt-PA infusion. Clinical exclusion criteria included major neurological deficit(s), implying a large ischemic event (e.g., a combination of severe hemiplegia, decreased consciousness, or forced-eye deviation), or minimal neurological deficits indicating small, minor, or transient tissue injury; malignant hypertension or a systolic/diastolic blood pressure higher than 200/ 120 mm Hg; a history of new neurological deficit(s) within 6 weeks of the presenting event; previous intracranial hemorrhage; presumed septic embolism; any condition known to be associated with an increased risk of hemorrhage after the use of thrombolytic agents [lo); known sensitivity to contrast agents; serious advanced illnesses with a reduced life expectancy; and any condition that the investigator thought would pose a significant hazard to the patient receiving rt-PA. Computed tomographic (CT) exclusion criteria included highdensity lesion($ consistent with recent hemorrhage of any degree; evidence of significant mass effect or midline shift; evidence of a lacunar infarct that may have been responsible for the admitting symptoms; or intracranial tumor, arteriovenous malformation, or aneurysm. Evidence of a lowattenuation defect on the initial CT did not preclude study participation. Angiographic inclusion criteria required unequivocal complete occlusion(s) of an extracranial or intracranial artery (Thrombolysis in Myocardial Infarction trial [TIMI) grade 0 111)) in the appropriate carotid or vertebrobasilar territory consistent with the patient’s clinical presenta-

tion. Angiographic exclusion criteria included arterial stenosis as the sole lesion; suspicion of arterial dissection; or coexistent aneurysm, arteriovenous malformation, or any nonatherosclerotic arteriopathy (e.g., vasculitis). Any significant neurological deterioration following the entry angiogram, but prior to the administration of rt-PA, dismissed the patient from the study.

Study Protocol After completing informed consent requirements, patients underwent baseline noncontrast CT. Cerebral angiography was then performed on patients not excluded by the screening CT. Patients with an angiographically documented TIMI grade 0 perfusion of brain artery(ies) received a single, preassigned, intravenous dose of rt-PA over 60 minutes without bolus. At completion of the infusion, the status of the occluded artery(ies) was determined by repeat angiography. Noncontrast CT was performed in all treated patients 24 hours after entry to the study and prior to discharge. CT was performed if neurological deterioration occurred at any point. Neurological evaluations were performed at presentation; just prior to and following the rt-PA infusion; at 24, 48, and 72 hours; and at 7 and 14 days. Otherwise, patients were managed in accord with standard local practice.

Recombinant Tissue Plasminogen Activator Infusion The agent employed for this study was a two-chain forn of rt-PA (duteplase), supplied by the Burroughs Wellcome Company (Research Triangle Park, NC) 112, 13). Duteplase differs from alteplase by (1) the substitution of methionine for valine 245, and (2) a specific activity of 310,000 IU/mg, compared to 580,000 IU/mg for alteplase by clot lysis assay (Burroughs Wellcome, data on file, 1988). rt-PA dose is expressed here in million international units (MIU) of thrombolytic activity.

Outcome Events The primary outcome events for this study were (1) recanalization of angiographically documented cerebral arterial occlusion($ at 60 minutes (efficacy), and (2) intracerebral hemorrhage with associated neurological deterioration within 24 hours of rt-PA infusion (safety). The recanalization of each occluded artery was assessed according to published TIMI perfusion grades El 1). Recanalization status per patient was determined as no perfusion (TIMI grades 0 or l), partial perfusion, or complete perfusion (grade 3) of all documented arterial occlusions. Partial perfusion referred either to TIMI grade 2 perfusion of a single obstructed artery, or grade 2 or 3 perfusion in some but not all of several previously obstructed arteries. For separate analysis, the primary vessel referred to the intracranial occlusion-bearing artery most probably responsible for the initial focal symptoms. The presence of hemorrhage on the 24-hour CT was classified as either (1) hemorrhagic infarction (HI) or (2) parenchymatous hematoma (PH). H I was defined as patchy areas of hlgh attenuation with indistinct margins and a speckled or mottled appearance, whether separate or coalescent, within an infarction. P H was defined as a homogeneous region of high attenuation with or without associated mass effect. Hemorrhages that

del Zoppo et al: rt-PA in Acute Thrombotic Stroke 79

could not be classified as H I or P H were categorized as “indeterminant.”

Ancilkny Medications A normal prothrombin time (PT) and activated partial thromboplastin time (aPTT) were required prior to the infusion of rt-PA. Systemic heparin was forbidden within the 24 hours following rt-PA infusion, and heparin use was confined to that necessary for cathether maintenance during angiography (< 4 IU/ml or 2,000 IU todpatient). Oral anticoagulants were not allowed prior to or within 72 hours of rt-PA treatment, and volume expanders, hydroxyethyl starch, or dextrans were prohibited. Prospective Guidelines From the outset, all investigators used prospectively derived definitions of CT and angiography outcomes and hemorrhage types. The decision to advance to the next dose rate was made according to the anticipated recanalization frequency versus the maximum acceptable incidence of rt-PA-associated hemorrhagic transformation (HI or PH) with neurological deterioration based on literature sources f14-16}. The dose rate was advanced by 0.08 to 0.09 MIU/kg per 60minute increments per cohort if recanalization was less than 70% in the previous cohort. However, if the probability of hemorrhagic transformation with neurological deterioration was greater than 20% at a given dose (via a hypothesis test with a type I error of 0.05 and a type I1 error of 0.2), study termination at that dose level was considered. For dose rate groups C (0.29 MIU/kg) and beyond, a maximum cohort size of 15 subjects was anticipated. At the study outset, the logistics and safety of the serial angiographic procedures were tested in small cohorts, and at a duteplase dose rate considered to be ineffective (group A). This accounts for the cohort sizes of groups A, B, and C. Neuroradiology Core Facility Angiograms and CTs were interpreted at each center, and then adjudicated at the Neuroradiology Core Facility. Copies of all CTs and of selected angiographic films for each patient were sent to the core facility where an assessment of the presence or absence and type of hemorrhagic transformation and of infarction size was made. Infarction volume was determined from the 24-hour CT utilizing a cursor tracer and the Sigma-Scan Measurement Program (Jandel Scientific, Corte Madera, CA). The volume of infarction was computed from the cross-sectional area of the traced infarct zones on each serial scan slice multiplied by the scan slice thickness. Unblinded comparison between angiograms made before and after rt-PA infusion was performed to determine the arterial perfusion grade at each dose rate. Safety Committees An internal safety committee of three study investigators reviewed CTs for hemorrhagic transformation. An external safety committee of three clinical scientists not otherwise involved in this study reviewed the same neuroradiographic data and assigned the relationships of neurological deterioration to hemorrhagic transformation, and of each hemorrhagic

80 Annals of Neurology

Vol 32

No l July 1992

rl-PA ACUTE STROKE STUDY GROUP TRIAL

Inclusion

Exclusion

COMPLETION OF ANGIOGRAPHY No arterial occlusion ARTERIAL OCCLUSIONS DOCUMENTED

Angiographicexclusion’ Extracerebral procedure complications’” INITIATION OF tt-PA INFUSION

104

Procedural exclusions+ COMPLETION OF tt-PA INFUSION

Total

93

46

Fig I . Patient entry following angiography. Asterisk indicates that angiographic exclusions included aneurysm, suspected dissection, and other $ndings; double asterisk, seaere puncture site hemorrhage; dagger, see text. rt-PA = recombinant tisscle phsminogen activator. event to the study agent. This committee was also responsible for the confirmation of study termination.

Data Collection and Statistical Methods Collection of case report forms on all patients receiving rt-PA was coordinated by the sponsor. Limited information regarding demography, associated illness, CTs, and baseline angiography findings on all patients undergoing angiography but excluded from treatment was collected at the Scripps Clinic and Research Foundation (SCRF). The data bases of the sponsor, the Neuroradiology Core Facility, and the SCRF have been merged for this analysis. Data are expressed either in literal form o r as the mean standard deviation. Statistical comparisons were made employing Fisher’s exact test (two-tailed p or 2p), the MantelHaenzsel test, or the signed rank test, where appropriate.

*

Results Patient Characteristics

Of 139 patients with stable symptoms of focal cerebral ischemia undergoing angiography, 27 had no definable occlusion (Fig 1). Consequently, 112 patients (80.6%) had complete occlusion of the primary vessel within 8.0 hours of symptom onset. Eight patients were dismissed because of angiographic exclusions (eg., suspected dissection, aneurysm) or an extracerebral complication during the angiographic procedure. Thus, 104 patients received intravenous rt-PA, 93 of whom completed the assigned dose rate infusion and were eligible for assessment of efficacy. The 11 patients excluded from the efficacy analysis were 6 patients who did not complete the assigned rt-PA infusion (including 1 patient with a possible allergic reaction), 2 patients who presented with an aneurysm or an arterial dissection, 2 patients who were entered beyond the 8-hour period, and 1 patient who received saline solution in error. No differences in major demographic characteristics between rt-PA-treated and angiography-excluded pa-

Table 1. Patient Characteristics rt-PA Treated

Not Treateda

104 52/52 61.0 2 12.4 5.4 1.7 105.8 ? 14.7

n Gender (M/F) Age (yr) A(T-0) [A(A-O)} (hr) MAP (mm Hg) Antithrombotic agents Aspirin Dipyridamole Heparin Oral anticoagulants Cardiovascular history Previous cerebrovascular events Extracranial disease Cardiac arrhythmias Valvular dysfunction Congestive heart failure Ischemic cardiac disease Hypertension Tobacco use

35 25/10 58.4 f 12.5 C5.2 f 1.51 108.5 f 16.8

*

18 6

5 1

9 2

0 1

21 8 43 7 12 38 46 54

7 2 2 1 2 7 19 16

2p Value 0.02 7 NS NS NS

NS NS NS NS NS NS 0.0003 NS NS NS NS NS

aPatients not treated were those without a cerebrovascular occlusion or with another exclusion (e.g., suspicion of arterial dissection, aneurysm, severe puncture site hemorrhage) that prevented entry into the study.

K-PA = recombinant tissue plasminogen activator; A(T-0) = interval from symptom onset to treatment; A(A-0) = interval from symptom onset to angiography (untreated patients); MAP = mean arterial pressure on admission; 2p = two-tailed p; NS = not statistically significant.

Table 2. Probable Stroke Mechanism

extracranial internal carotid artery (ICA) occlusions,

Assigned Cause (rt-PA Treated)

n

Cardioembolic Internal carotid artery atherothrombosis 2 local embolism Angiographic complication Oral contraceptives Unknown Total

53 24 3 3 10 93

K-PA = recombinant tissue plasminogen activator.

tients were seen, except for gender ratios (Table 1). Patients with a history of any cardiac arrhythmia were significantly more likely to have had a cerebrovascular occlusion. No significant difference in the time from symptom onset to treatment [A(T-0)] for patients receiving rt-PA, and from symptom onset to angiography EA(A-O)J for treatment-excluded patients was observed (see Table 1). The mean A(T-0) was 5.4 1.7 hours and mean A(A-0) was 5.2 1.5 hours. Approximately 83% of 9 3 treated patients had a presumed embolic event from either a cardiac or a carotid artery source (Table 2).

*

*

Distribution of Cerebmascular Occlusions

The distribution of vascular occlusions for each dose rate is shown in Table 3. Among 93 patients completing the study, 23 had angiographically documented

including 12 patients with tandem ipsilateral middle cerebral artery (MCA) occlusions and 11 patients in whom the ipsilateral MCA was not seen because contralateral angiography was not performed. In addition, 2 patients had intracranial ICA occlusions together with a posterior cerebral artery (PCA) occlusion or an anterior cerebral artery (ACA) occlusion. Sixty-two patients had MCA stem, MCA division, and/or MCA branch occlusions, while both MCA and ACA occlusions were present in 3 patients, and isolated ACA, PCA, and combined basilar arterial and PCA occlusions were found in 1 patient each. Proximal and/or distal MCA occlusions comprised 50.0 to 86.7% of the dose rate cohorts, while ICA (+- MCA) occlusions contributed 13.3 to 50.0% to each cohort. Recanalization

No relationship of dose rate to recanalization (partial or complete) was observed among patients completing the rt-PA infusion (Fig 2). Among patients without an ICA occlusion at presentation (n = 68), a trend for more frequent recanalization was apparent across all dose rates (2p = 0.089). There was no difference in mean A(T-0) between patients with recanalization at 60 minutes (5.5 & 1.7 hours) and those without recanalization (5.6 1.7 hours). When each anatomical site of vascular occlusion was considered, a significant increase in the proportion of division or branch MCA occlusions exhibiting recanalization was seen, compared with ICA occlusions when all dose rates were

*

del Zoppo et al: rt-PA in Acute Thrombotic Stroke 81

Table 3. Vascular Distribution of Occlusions (per Patient)

ICA

Dose Rate Group

4

C D

E F G H I

Total =

Intracranial Occlusion

7 7 15 15

-

9

-

2

-

1 3 3

-

1

14

-

15 7

1

-

-

-

1

-

-

-

-

-

4

-

-

1 1

-

2

-

5

-

-

-

1

-

1

1 1

-

-

-

1

1

1

-

10

93

-

11

-

1

-

Other

-

1

2

9 4 3 4 4

2 1

1 2 1

-

3

-

1 2

2

-

1 3

1 1

2 15

2

2

33

14

-

-

3

3 2

-

-

-

-

-

1

-

1

-

-

-

-

-

1

1

1

I

1

1

internal carotid artery; MCA = middle cerebral artery.

-

I

oNo rzzm Partial

Dose-Rate A

B

C

D

E

F

G

H

I

4

7

7

15

15

9

14

15

(7)

N

MCA

n

A B

ICA

2

~

~

~~

~

Fig 2. RecanalizationfOl(0wingrecombinant tissue phsminogen activator infusion at each dose rate. Complete, partial, and no recanaIizations are indicated as percent of each dose rate cohort N . Actual numbers of patients in each recanalization class are indicated in their respective bar.

considered (Table 4). The overall frequency of extracranial ICA recanalization was 8.0%, whereas the overall vascular recanalization frequency of MCA stem and of more distal MCA occlusions in groups A through I was 26.1% (12/46 vessels) and 38.1% (43/113 vessels), respectively. When only the primary or most probable symptom-related vessel was considered, the outcome for distal MCA vessels was the same. No dose rate relationship to recanalization by vascular occlusion location was apparent.

deterioration was observed (Fig 3). The cumulative incidence of H I and P H for all dose rates was 30.8% (32 of 104), of which 10 patients (9.6%) had associated clinical deterioration. These findings remained within the prospectively established safety limits. Neurological deterioration was more often associated with P H (6/11) than with H I (4/21) ( 2 p = 0.056): Of the 10 patients who developed neurological deterioration, only 2 patients displayed any recanalization at 60 minutes. One patient in group D had both an H I and a PH, as well as a large (211-cm3) infarction following partial recanalization of MCA stem and division occlusions. In 2 patients, P H occurred in regions distant from the infarction. Patients with hemorrhagic transformation received rt-PA significantly later (6.1 1.5 hours) than did the remaining patients (5.3 2 1.7 hours; 2 p = 0.006). There was no relationship of HI, PH, or both to arterial recanalization, the presence of infarction on baseline CT, infarction size determined from the predischarge CT, prestudy antiplatelet agent use, or entry blood pressure.

*

Angiography

One patient had a new neurological deficit during post-i-t-PA angiography in the territory contralateral to the entry symptoms, as confirmed by follow-up CT.

Hemowhagic Transfomnation

Hemorrhagic transformation was classified as H I in 2 1 patients with P H in 11 among the 104 patients who received rt-PA. (Indeterminant hemorrhages were ultimately classified as HI in 2 patients and as PH in 2.) No dose rate relationship to hemorrhagic transformation (HI and PH) at 24 hours with or without clinical 82 Annals of Neurology Vol 32 No 1 July 1992

Mortality

Thirteen (12.5%) of 104 rt-PA-treated patients died in the hospital (Table 5). Mortality was associated with P H in 2 patients and with H I in 2 patients, while large infarcts with cerebral edemdherniation (n = 4 ) and cardiopulmonary causes (n = 5) accounted for the re-

Table 4 . Recanalization of Occluded Vessels by Location Proportion of RecanalizationINo Recanalization"

MCA

Dose Rate Group

MIUIkg

ICA

Stem (M,)

Division (M2)

Branch (M3)

A

0.12 0.20 0.29 0.37 0.43 0.51 0.59 0.67 0.75

0: 3 0: 1

0 :3 0:4 2:1 5:7 1:8 0:2 0 :5 3:4 1 :o 12 : 34 0.116

0:2 0:2 0:2 6:5 4:1 1:3 0:2 2:4 1:s 14:26 0.018

2:6 3:6 1:2 8:4 9:s) 1:5 0:2 4:7 1:3 29 :44 0.005

B C D E F G H I Total 2 p Valueb

0:1 1:4 0:4 0:2 0:5 1:2 0 :1 2:23

"All values represent the proportion of recanalized vessels to nonrecanalized vessels at 60 minutes of infusion of recombinant tissue plasminogen activator. bComparison of proportion of MCA patency with ICA patency for dose rate groups A through I (see text). Although group A dose rate was not considered to have an effect on recanalization, its addition to the totals here does not alter the outcome of this analysis. ICA

=

internal carotid artery; MCA = middle cerebral artery.

-

100

0

60

Table 5 . Mortality % Hemorrhagic

Transformation PH

Causation

HI

40 20 0 Dose-Rate A N 4

B 7

C 8

D

E

F

G

H

I

16

17

10

19

16

(7)

Fig 3. Hemowhagic transformationfollowing infusion of recombinant tissue plasminogen activator at each dose rate. Hemorrhagic infarction (HI),parenchymatous hematoma (PH), and no hemowhage are indicated as percent of each dose rate cohort N.Figures refer to the actual number of patients exhibiting hemowhage of the type represented by each bar. The vertical bars ctt each dose rate represent the number of hemorrhagic events occuwing in patients witb complete recanalization,partial recanalization, and no recanalization at 60 minutes, as depicted in Figure 2.

Intracerebral hemorrhage HI PH Cerebral edema With HI Without H I Medical Congestive heart failure Myocardial infarction Pneumonia Pulmonary embolism

0.05-0.58

20.7-29.6

0.517.0

8.5-21.1,

3.0-35.0

9.9-33.2

1 2 53.6

1 4 2 1. 1

2b

'Relative to rt-PA treatment. bA single patient died of a pulmonary embolism 35 days after rt-PA treatment.

HI = hemorrhagic infarction; P H = parenchymatous hematoma; rt-PA = recombinant tissue plasminogen activator.

hemorrhage followed an infusion of 0.59 MIU/kg in 1 patient. mainder. Recanalization did not occur in 3 of the 5 patients with cerebral edema.

Peripheral Complications Serious or life-threatening non-central nervous system hemorrhages occurred infrequently. A 77-year-old woman with a hepatic tumor and subcapsular hemorrhage who received 0.67 MIU/kg of rt-PA, and separate patients with severe epistaxis and hemorrhages at catheter insertion sites who received 0.5 1 MIU/kg, required blood transfusions. Severe gastrointestinal

Discussion Early safe pharmacological recanalization of cerebral arterial occlusions offers a potential therapeutic approach to focal cerebral ischemia. In anticipation of clinical efficacy trials on the systemic infusion of rt-PA in acute atherothrombotic and embolic stroke, a dose rate study on recanalization and safety was undertaken in patients presenting within 8 hours of the onset of stable focal ischemic symptoms. This study was prematurely terminated, however, before completion of dose del Zoppo et al: rt-PA in Acute Thrombotic Stroke 83

rate group I (0.75 MIU/kg) because of the withdrawal of duteplase from clinical trials following a patent suit. Occlusion of a cerebral artery in the symptomatic territory was documented in 80.6% of patients undergoing angiography a mean of 5.4 hours from symptom onset. This compares with carotid territory occlusion frequencies of 76% at 6 hours f171, 59% at 24 hours 151, and 41% at 1 week f181 following the onset of focal cerebral ischemia, in several recent angiographic studies. The earlier report of spontaneous recanalization by Dalal and colleagues f 3 } , and the finding of 95% patency in patients (56/59)with an acute cerebral embolism reexamined within a median of 20 days (3-47 days) of the acute event support these observations f203. It was anticipated at the outset of this study that a dose rate of 0.12 MIU/kg (group A) would have no effect on recanalization. Coronary artery patency has been documented 90 minutes following the intravenous infusion of 0.24 MIU/kg of duteplase during acute myocardial infarction (Burroughs-Wellcome Company, data on file, 1989). Although the rt-PA dose rates chosen for this study escalated through those known to achieve coronary artery reperfusion, the recanalization response to increased dose rates appeared flat. Heterogeneity in sites of cerebral vascular occlusion among the dose rate groups, limited cohort size, and differing susceptibility of occlusions at various anatomical sites may have contributed to the lack of a dose rate effect. In view of thrombus size and the association of thrombus with arteriosclerosis, proximal ICA occlusions may have been more resistant to systemic rt-PA infusion than were MCA division and branch occlusions, as suggested by two pilot studies of intraarterial infusion u-PA and streptokinase 11, 41. The lower overall per-vessel recanalization in this study may also reflect the restriction of blood flow proximal to an occlusion. Drug delivery at the thrombus surface depends on proximity of the occlusion to the nearest bifurcation (in the anterograde direction), as well as retrograde flow via collateral vessels. Restricted flow in an occluded segment has been seen during regional or systemic perfusion of a thrombolytic agent 143. Additionally, with the infusion scheme used in this study the perfusion status at 60 minutes may not have reflected the maximum incidence of arterial patency. Reperfusion of MCA stem occlusions beyond the 60-minute angiogram was noted in a limited number of patients undergoing later angiography or serial transcranial Doppler studies. Delayed recanalization is also confounded by an unknown but definable incidence of spontaneous recanalization f l 8 , 21-24]. It is also possible that the rate of recanalization with this agent may depend on the vascular territory involved. These considerations suggest that to optimize recanalization, future studies of thrombolytic agents in acute 84 Annals of Neurology Vol 32 No 1 July 1992

stroke will need to consider the location of the cerebrovascular occlusion. Limited data with which to judge acceptable boundaries for hemorrhagic transformation were available at the inception of this study, as no comparable angiography- and CT-based “natural history” study had been performed. Nonetheless, estimates from four CT-based studies of nonanticoagulated patients with predominantly embolic stroke suggested a 5 to 43% incidence of hemorrhagic transformation 114, 15, 20, 25, 261. Neurological deterioration related to hemorrhagic transformation ranged from 4.2 to 10.7% in three studies 114, 16, 271. Among three reports, including a total of 344 patients with cerebral embolism 120, 25, 271, P H developed in 2 to 8.6%. From these considerations a maximum acceptable incidence of hemorrhagic transformation with associated neurological deterioration of 20% was applied to each dose rate. Overall, the cumulative incidence of hemorrhagic transformation with neurological deterioration was 9.6% and did not exceed 14% for any dose rate group except group I. Here, the number of patients was artificially limited by early study termination. H I was infrequently accompanied by neurological deterioration. When deterioration was seen in patients with H I (4/21 patients), arterial recanalization occurred in only 2. The incidence of H I at all dose rates remained well within that suggested by both postmortem 128, 291 and prospective CT-based series 114, 201. In short, systemic administration of this rt-PA did not significantly increase the incidence of anticipated hemorrhage with neurological deterioration at dose rates known to achieve coronaty artery reperfusion. The hypothesis that H I results from the downstream movement of emboli in cerebral arteries with reperfusion leakage from ischemia-injured capillaries 1281 implies that recanalization of large cerebral arteries may produce significant CT-detectable hemorrhages. In the present study, H I and PH were not associated with early detectable recanalization. This observation and the work of Ogata and colleagues suggest that major cerebral arterial reperfusion is not required for hemorrhagic transformation f301. Limited experimental work would seem to support this observation 131, 321. In at least 1 patient (group C) a significant P H followed angiographically recorded contrast extravasation from small, penetrating lenticulostriate arteries, consistent with unchecked hemorrhage. Certainly, the unopposed action of an agent capable of degrading fibrin on its formation would promote continuing hemorrhage. That rt-PA did not itself cause all hemorrhagic transformation in this population is reflected by the occurrence of H I in dose rate group A. Although unproved, progressive ischemic injury to vessels distal to an obstruction may also contribute to the risk of hemorrhage. This mechanism may explain the significant association

of hemorrhagic transformation when treatment commenced beyond 6 hours. Other potential contributors to intracerebral hemorrhagic risk, including use of antiplatelet agents before the study, history of compensated hypertension, and the presence of early infarction on baseline CT, were not found more often in subjects with than those without hemorrhagic transformation. The PHs seen were generally well circumscribed and, with two exceptions, were confined to the region of infarction. Nonetheless, no relationship between the incidence or type of hemorrhage to CT infarction size at 24 hours was seen. This observation contradicts the conclusions of earlier retrospective studies 127J, but is supported by one postmortem study 1293. This acute intervention study suggests a number of issues for future consideration. Angiography very early in the course of cerebral ischemia is feasible and safe, and provides important diagnostic and vascular outcome data for correlation with clinical outcome. Despite the high frequency of arterial occlusions at study entry, location of an occlusion cannot be precisely predicted from clinical presentation alone. However, if neurological outcome is dependent on recanalization, and if occlusion location is pertinent to recanalization efficacy as indicated here, knowledge of the vascular status will be needed to judge treatment efficacy. In view of the relative resistance of ICA occlusions at all dose rates, such patients may be excluded from future systemic infusion trials and vessel-specific dose rates may be required. Certainly, it is not possible to extrapolate effective dose rates for recanalization from comn a q artery reperfusion data to the cerebral arterial circulation. Hemorrhagic transformation per se was not a limiting feature in the outcome of this study. Indeed, the majority of patients had either silent hemorrhagic transformation or improved in the face of a documented hemorrhagic event. A dose rate of intravenous rt-PA that safely and predictably recanalizes MCA and other major cerebral arterial occlusions is yet to be determined. In the absence of a definable recanalization response to rt-PA dose rates and since hemorrhagic transformation remained within acceptable limits at each dose rate, the effect of dose rates higher than that at which the study was terminated could be investigated. This study suggests that the proper conditions required to assess neurological improvement and recanalization include time to recanalization, occlusion location, and the contribution of spontaneous recanalization. We think that future studies should therefore emphasize early vascular diagnosis for early treatment, ideally with angiographic correlation, and a placebo-controlled format.

Appendix Participants in the n-PAIAcute Stroke Study Group include: Klinikum, Rheinisch Westf;ilische Technische

Hochschule, Aachen, Federal Republic of Germany (FRG): Klaus Poeck, MD, Andreas Ferbert, MD, Hartmut Briickmann, MD, Rolf Schneider, E. Bernd Ringelstein, and Armin Thron, MD; Duke University Medical Center, Durham, NC: Mark J. Alberts, MD, Linda Gray, MD, James Davis, MD, Larry Goldstein, MD, and Michael Brothers, MD; Tubs-New England Medical Center, Boston, MA: Michael S . Pessin, MD, L. Dana DeWitt, MD, Louis R. Caplan, MD, Conrad0 J. Estol, MD, Samuel Wolpert, MD, Eddie S. K. Kwan, MD, and Mary Anderson, MD; University of California, San Diego, CA: Justin A. Zivin, MD, PhD, John F. Rothrock, MD, Patrick D. Lyden, MD, Wayne M. Clark, MD, Ken P. Madden, MD, PhD, and John Hesselink, MD; University of Pittsburgh, Pittsburgh, PA: Lawrence Wechsler, MD, Charles Jungreis, MD, Oscar M. Reinmuth, MD, Benjamin H. Eidelman, MD, Neil A. Busis, MD, Gordon Banks, MD, Michael Giuliani, MD, Joseph A. Horton, MD, William L. Hirsch, MD, David W. Johnson, MD, and Susan S . Kemp, MD; The Cleueland Clinic, Cleueland, OH: Anthony J, Furlan, MD, Marc I. Chimowitz, and Charles Lanzieri, MD; Klinikum I Minden, Minden, FRG: Otto Busse, MD, and Dr Chahine; University of TexasSouthwesternMedical Center, Dallas, T X : Ralph Greenlee, Jr, MD, Phillip Purdy, MD, D. Hal Unwin, MD, Hunt Batjer, MD, and Michael Devous, MD; Yale University School of Medicine, New Haven, CT: Lawrence Brass, MD, James Abrams, MD, and Pierre B. Fayad, MD; Neurological Institute, Columbia University, New York, N Y : Jay Mohr, MD, and Jackie Bello, MD; Rhode Island Hospital, Prouidence, RI: Edward Feldmann, MD, and Richard Haas, MD; University of Heidelberg, Heidelberg, FRG: Werner Hacke, MD, Riidiger von Kummer, MD, Markus Miiller-Kiippers, MD, Klaus Sostor, MD, and Michael Forsting, MD; Boston University Medical Center, Boston, MA: Carlos Kase, MD, Shripad Tilak, MD, and Philip A. Wolf, MD; The University of Iowa College of Medicine, Iowa City, IA: Jose Biller, MD, Anthony Ryals, MD, E. Eugene Marsh, 111, MD, Betsy B. Love, MD, David Gordon, MD, Harold P. Adams, Jr, MD, Steve H . Cornell, MD, and Dan Loes, MD; Scripps Clinic and Research Foundztion, La Jolla, CA: Gregory J. del Zoppo, MD, and Shirley Otis, MD; Massachusetts General Hospital, Boston, MA: Daryl Gress, MD, Ken Davis, MD, and J. Philip Gstler, MD. The centers are listed in descending order of patient accrual. The Neuroradiology Core Facility: Tubs-New England Medical Center: Samuel M. Wolpert, MD, Chief, Section of Neuroradiology. External Review Committee: Henry Barnett, MD, Bruce Coull, MD, and Mark Dyken, MD. Internal Review Committee: Anthony J. Furlan, MD, Mark J. Alberts, MD, Carlos Kase, MD, and Michael S. Pessin, MD. Data Retrieual and Publications Committees: Gregory del Zoppo et al: rt-PA in Acute Thrombotic Stroke 85

J. del Zoppo, MD, Andreas Ferbert, MD, Anthony J. Furlan, MD, Werner Hacke, MD, Michael S. Pessin, MD, and Samuel Wolpen, MD. This study received support from the Burroughs-Wellcome Company (Research Triangle Park, NC). We are grateful for the involvement of Richard Kent, MD, Judith Littlejohn, MD, Rose Snipes, MD, and Jacqueline Zinn.

References 1. del Zoppo GJ, Ferbert A, Otis S, et al. Local intra-arterial fi-

brinolytic therapy in acute carotid territory stroke: a pilot study. Stroke 1988;19:307-313 2. Fieschi C, Bozzao L. Transient embolic occlusion of the middle cerebral and internal carotid arteries in cerebral apoplexy. J Neurol Neurosurg Psychiatry 1969;32:236-240 3. Hacke W, Zeumer H , Ferbert A, et al. Intra-arterial thrombolytic therapy improves outcome in patients with acute vertebrobasilar occlusive disease. Stroke 1988;19:1216-1222 4. Mori E, Tabuchi M, Yoshida T,Yamadori A. Intracarotid urokinase with thromboembolic occlusion of the middle cerebral artery. Stroke 1988;19:802-812 5. Solis OJ, Roberson GR, Taveras JM, et al. Cerebral angiography in acute cerebral infarction. Rev Interam Radial 1977;2:19-25 6. Zeumer H, Freitag HJ, Grzyka U, Neunzig HP. Local intraarterial fibrinolysis in acute venebrobasilar occlusion. Technical developments and recent results. Neuroradiology 1989;31: 336-340 7 del Zoppo GJ, Zeumer H, Harker LA. Thrombolytic therapy in acute stroke: possibilities and hazards. Stroke 1986;17:595-607 8 del Zoppo GJ. Relevance of focal cerebral ischemia models experience with fibrinolytic agents. Stroke 1990;2l(supp1 IV):IV-l55-IV-160 9 Zivin JA, Fisher M, DeGirolami U, et al. Tissue plasminogen activator reduced neurological damage after cerebral embolism. Science 1985;230:1289-1292 10 NIH Consensus Conference. Thrombolytic therapy in treatment. Br Med J 1980;280:1585-1587 11. TIMI Study Group. Special report: the Thrombolysis in Myocardial Infarction (TIMI) trial. N Engl J Med 1985;312: 932-936 12. Armstrong PW, Baigrie RS, Daly PA, et al. Tissue plasminogen activator: Toronto (TPAT) placebo-controlled randomized trial in acute myocardial infarction. J Am Coll Cardiol 1989;13: 1469- 1476 13. Sleight P. ISIS-3 Trial. Results of SK vs APSAC vs tPA. In: Proceedings of the 40th annual scientific session of the American College of Cardiology. March 5, 1991 14. Hornig CR, Dorndorf W, Agnoli AL. Hemorrhagic cerebral infarction: a prospective study. Stroke 1986;17:179-185

86 Annals of Neurology Vol 32 No 1 July 1992

15. Lodder J. CT-detected hemorrhagic infarction; relation with the size of the infarct, and the presence of midline shift. Acta Neurol Stand 1984;70:329-335 16. On BR, Zamani A, Kleefield J. Funkenstein H H . The clinical spectrum of hemorrhagic infarction. Stroke 1986;17:630-637 17. Fieschi C, Argentino C , Lenzi GL, et al. Clinical and instrumental evaluation of patients with ischemic stroke within the first six hours. J Neurol Sci 1991;1989:311-321 18. Irino T, Taneda M, Minarni T. Angiographic manifestations in post-recanalized cerebral infarction. Neurology 1977;27:471475 19. Dalal PM, Shah PM, Aiyar RR. Arteriographic study of cerebral embolism. Lancet 1965;2:358-361 20. Okada Y,Yamaguchi T, Minematsu K, et al. Hemorrhagic transformation in cerebral embolism. Stroke 1989;20598-603 21. Dalal PM, Shah PM, Sheth SC, Deshparde CK. Cerebral emboLism: angiographic observations on spontaneous clot lyses. Lancet 1965;1:61-64 22. Fisher CM. The natural history of middle cerebral artery trunk occlusion. In: Austin GM, ed. Microneurosurgical anastomoses for cerebral ischemia Springfield, I L Charles C Thomas, 1976: 146-154 23. It0 Z, Suzuki A, Hen R, Menura K. Prognostic factors in spontaneous recanalization of middle cerebral artery occlusion. In: Ito 2, Kutsuzawa T, Yasui N , eds. Cerebral ischemia-an update. Amsterdam: Excerpta Medica, 1983:159-166 24. Taneda M, Shimada N , Tsuchiya T. Transient neurological deficits due to embolic and reopening of the cerebral arteries. Stroke 1985;16:522-524 25. Fisher M, Zito JL, Siva A, DeGirolami U. Hemorrhagic infarction: a clinical and CT study. Stroke 1984;15:192 26. Weisberg LA. Nonseptic cardiogenic cerebral embolic stroke: clinical-CT correlations. Neurology 1985;35:896-899 27. Cerebral Embolism Study Group. Immediate anticoagulation of embolic stroke: brain hemorrhage and management options. Stroke 1984;15:779-789 28. Fisher CM, Adams RD. Observations on brain embolism with special reference to the mechanism of hemorrhagic infarction. J Neuropathol Exp Neurol 1951;10:92-94 29. Fisher CM, Adams RD. Observations on brain embolism with special reference to hemorrhage infarction. In: Furlan AJ, ed. The h e m and stroke. Exploring mutual cerebrovascular and cardiovascular issues. New York Springer, 1987:17-36 30. Ogata J, Yutani C, Imakita M, et al. Hemorrhagic infarct of the brain without a reopening of the occluded arteries in cardioembolic stroke. Stroke 1989;20:876-883 31. del Zoppo GJ, Copeland BR, Harker LA, et al. Experimental acute thrombotic stroke in baboons. Stroke 1986;17:12541265 32. Lyden PD, Zisvin JA, Clark WA, et al. Tissue plasminogen activator-mediated thrornbolysis of cerebral emboli and its effect on hemorrhagic infarction in rabbits. Neurology 1989;39: 703-708

Recombinant tissue plasminogen activator in acute thrombotic and embolic stroke.

An open angiography-based, dose rate escalation study on the effect of intravenous infusion of recombinant tissue plasminogen activator (rt-PA) on cer...
928KB Sizes 0 Downloads 0 Views