(Stroke. 2000;31:841.)
© 2000 American Heart Association, Inc.
Original Contributions |
From the Division of Neurosurgery, Walter Reed Army Medical Center (N.J.N.), Washington, DC; Neurosciences Critical Care Division (J.R.C., M.A.W., A.B., J.A.U., D.F.H.) and Department of Emergency Medicine (P.M.K.), The Johns Hopkins Medical Institutions, Baltimore, Md; Departments of Neurosurgery and Neurology (J.B., S.T.), Mount Sinai Medical Center, New York, NY; Department of Neurological Surgery (R.B.), Medical College of Virginia, Richmond; and Department of Neurology (E.S., B.P.), University Hospital, Innsbruck, Austria.
Correspondence to Daniel F. Hanley, MD, Division of Neurosciences Critical Care, Meyer 8-140, Johns Hopkins Medical Institutions, 600 N Wolfe St, Baltimore, MD 21287-7840. E-mail: dhanley{at}jhmi.edu
| Abstract |
|---|
|
|
|---|
MethodsTwelve pilot IVH patients who required external ventricular drainage were prospectively treated with intraventricular urokinase followed by the randomized, double-blinded allocation of 8 patients to either treatment or placebo. Observed 30-day mortality was compared with predicted 30-day mortality obtained by use of a previously validated method.
ResultsTwenty patients were enrolled; admission Glasgow Coma
Scale score in 11 patients was
8; 10 patients had pulse pressure
<85 mm Hg. Mean±SD ICH volume in 16 patients was 6.21±7.53
cm3 (range 0 to 23.88 cm3), and mean±SD
intraventricular hematoma volume was 44.26±31.65
cm3 (range 1.31 to 100.36 cm3). Four patients
(20%) died within 30 days. Predicted mortality for these 20 patients
was 68.42% (range 3% to 100%). Probability of observing
4 deaths
among 20 patients under a 68.42% expected mortality is 0.000012.
ConclusionsIntraventricular urokinase may significantly improve 30-day survival in IVH patients. On the basis of current evidence, a double-blinded, placebo-controlled, multicenter study that uses thrombolysis to treat IVH has received funding and began January 1, 2000.
Key Words: intraventricular hemorrhage outcome urokinase
| Introduction |
|---|
|
|
|---|
A prevailing concept has been that the initial ventricular hemorrhage almost invariably correlates with an acute elevation of intracranial pressure (ICP). The alteration of consciousness that often develops after initial ictus and that sometimes persists during the subacute stage is thought to be the result of intracranial hypertensioninduced ischemic encephalopathy. Similarly, direct compression of rostral brain stem and thalamus could play a pathogenic role.9 Several studies have identified ventricular dilation, intraventricular hematoma volume, and increased ICP to be indicators of poor prognosis in patients with IVH.10 For this reason, treatment strategies for IVH have centered on ICP management.3 Currently, use of external ventricular drainage (EVD) to accelerate recovery and reduce the risk of development of chronic hydrocephalus is widely recommended. This approach to treatment of IVH patients has become the standard of care when clinical and radiographic manifestations of acute obstructive hydrocephalus are identified. However, ICP elevations are not universally present in IVH patients who manifest acute and subacute neurological deterioration. In addition, normalization of ICP with EVD by means of intraventricular catheter (IVC) does not always lead to improvement in neurological deficits, even after an initially elevated ICP is corrected.11
Clear rationale exists in favor of the benefits of removal of IVH blood. Blood and blood derivatives have long been recognized as proinflammatory agents in neurological disease.12 13 14 15 Case studies that use thrombolysis for removal of blood components from the subarachnoid and ventricular spaces in SAH and ICH have been safely and successfully conducted.16 More recently, Tuhrim et al17 have validated that IVH volume is an independent risk factor for death. In a large consecutive clinical series, Coplin et al18 demonstrated that the chance of poor outcome is 66.7% in untreated patients compared with 31.8% in urokinase-treated patients.
Outcome after ICH can be described as a function of specific clinical features that include location of the hematoma. However, analysis of functional and survival outcome has been limited by the size of the patient series, which limits the application of complex cohort outcome analysis. Furthermore, until scientifically generated therapies such as the amelioration of secondary neuronal damage are designed, mortality remains the primary end point in ICH and IVH therapy trials.
The objective of the present study is to analyze the initial experience accumulated in 4 university-based neurointensive care units (Johns Hopkins Hospital [JHH], Mount Sinai Hospital [MSH], Medical College of Virginia [MCV], and University Hospital, Innsbruck lsqb]INN]) with the use of intraventricular urokinase (UK) in the treatment of IVH and to compare the 30-day mortality found with this therapy with the predicted mortality obtained with the use of a previously validated method.19
| Subjects and Methods |
|---|
|
|
|---|
Criteria for Patient Enrollment in the Study
All patients were screened and enrolled if an IVH with or
without ICH had occurred within 48 hours before admission and was
diagnosed by clinical and brain CT criteria and if an IVC was
clinically required for treatment of obstructive hydrocephalus. In
addition, when intracerebral aneurysms or
arteriovenous malformations were suspected, they had to be excluded by
appropriate diagnostic studies. Exclusion criteria were
pregnancy and age <18 years.
Treatment Protocol
Patients enrolled in the present study were admitted to a
Neurointensive Care Unit or Intensive Care Unit (ICU) with staff
experienced in the acute care of patients with IVH and IVCs. The UK
preparation was Abbokinase (Abbott Laboratories), a sterile,
lyophilized preparation intended for intravascular injection. A vial of
UK contains 250 000 IU as powder, which is reconstituted with sterile
normal saline or Ringers lactate to yield a solution that contains
25 000 IU of UK per milliliter. Patients received a first dose that
ranged from 5000 to 25 000 IU of UK no sooner than 12 hours and no
later than 48 hours after the IVH. Patients randomized to placebo
received only vehicle solution. In addition, CSF outflow and ICP
waveform morphology had to be optimal and consistent with
intraventricular location of the catheter tip
before intraventricular administration of the drug.
ICP and cerebral perfusion pressure were monitored before,
during, and after infusion of the drug. After injection, the IVC was
closed for 1 hour to prevent drainage of the study drug away from the
clot and to allow adequate time for drug-clot interaction. The IVC was
reopened within the initial hour only if needed to control medically
intractable ICP elevations (ICP >20 mm Hg refractory to
hyperventilation and mannitol administration). After 1 hour of closure,
the IVC was reopened with an appropriate drainage gradient. Subsequent
ICP measurement and management followed routine protocols of each ICU.
The study drug was administered every 12 hours until the IVC was
removed based on the patient tolerance to IVC closure for 24 hours (ie,
no sustained ICP elevation >20 mm Hg). Neuroradiological
evaluations were performed, with head CT scans obtained on alternate
days, and urgent studies were performed if an acute neurological
deterioration was recognized. Intraventricular and
intraparenchymal hematoma volumes were studied independently by
use of standard computerized volumetric analysis following the
method reported by Steiner et al.20 In addition to the
radiological monitoring, clinical surveillance was performed with daily
Glasgow Coma Scale (GCS) assessment and ICP, blood pressure, and CSF
analysis that following the specific ICU standards. In the
presence of fever without obvious source, CSF analysis was
performed to investigate the possibility of bacterial or chemical
ventriculitis. The JHH Internal Review Board approved this
protocol.
Statistical Analysis
Predicted mortality in ICH patients with an IVH was obtained by
use of the model designed by Tuhrim and coworkers.17 By
use of logistic regression analysis, 30-day survival
probability may be calculated with the following formula:
![]() |
85
mm Hg) or 1 (>85 mm Hg); IVH is either 0 (absent) or 1
(present); GCS score can assume values of 0 (>8) or 1 (
8); ICH
(intraparenchymal hemorrhage size) is measured in cubic
centimeters; Hydro is expressed as 0 if hydrocephalus is absent or 1 if
present; and IVH represents the size of the
intraventricular hemorrhage in cubic
centimeters. After calculated mean 30-day mortality and actual
mortality rate were obtained in the entire cohort, these were compared
with the exact hypothesis test for binomial random variables to
calculate the probability of obtaining the number of deaths observed or
fewer under the predicted mortality rate. Statistical significance was
accepted only if P<0.05. Values are mean±SD. | Results |
|---|
|
|
|---|
|
Demographics
Between December 1993 and July 1998, informed consent from 20
patients was obtained in the participating university-based referral
centers (JHH, 9 patients; MCV, 3 patients; MSH, 7 patients; and INN, 1
patient). Twelve patients were men, 8 were women; mean±SD age was
57±13.1 (range 35 to 80) years.
Clinical Features
In the JHH cohort, the cause of the IVH was associated with a
history of hypertension in 8 patients and with warfarin anticoagulation
in 1 (immediately reverted on admission). GCS on admission and before
administration of intraventricular UK in the cohort
was 8.7±3.89 (range 3 to 15); 11 patients had GCS
8, and 9 had GCS
>8. Ten patients had pulse pressure <85 mm Hg on admission; the
remaining 10 had pulse pressure >85 mm Hg. Throughout the study
period, no clinical or laboratory evidence of meningitis or
ventriculitis was detected. During the study period, only 1 patient,
from the JHH group, had an IVC exchange, which was due to presumptive
ventriculitis, a diagnosis that was eventually excluded. Only 1 episode
of IVC obstruction due to blood clots (MCV patient 2) occurred.
Duration of IVC placement in 7 of 12 treated patients was 5.14±2.27
days (range 3 to 8 days).
ICP was continuously recorded before and 1 hour after study drug administration in 2 cases. Mean ICP before administration was 7±5.7 mm Hg (range 0 to 22 mm Hg). One hour after administration, mean ICP was 10.2±6.5 mm Hg (range -5 to 28 mm Hg). In another 3 patients, mean ICP throughout the treatment period was 10±6.1 mm Hg (range -5 to 27 mm Hg), with only 4 documented episodes of elevated ICP (20, 27, 22, and 20 mm Hg), 3 of which were in the same patient. These ICP elevations responded to hyperventilation, and CSF drainage was not required to control them.
Four deaths occurred in the entire cohort. In the JHH group, 1 patient (JHH patient 7) died from cardiac arrest. No attempt to reverse this event was made, in accordance with the patients "do not resuscitate" status. A second patient from the same group (JHH patient 2) and MSH patient 7 died after care was withdrawn due to patient neurological condition and preexisting medical problems complicating the neurological disease. In the MCV cohort, MCV patient 2 experienced severe ICP elevation after IVH extension, which led to death.
Laboratory Features
Laboratory evidence of bleeding diathesis was found in only 1 case
(JHH patient 5) and was due to warfarin anticoagulation after left
internal carotid artery angioplasty and stent placement. No other
abnormalities relevant to the acute neurological disease were
identified in the study cohort throughout during the study. In the
prospectively controlled JHH group, baseline coagulation
parameters were as follows: international normalized ratio
(INR), 1.1±0.4 (range 0.8 to 2.0); platelet count,
240.4x103±77.9x103/cm3
(range 119 to
328x103/cm3); and partial
thromboplastin time (PTT), 24.4±3.4 (range 20.6 to 32.0)
seconds in 9 patients. During the study period, INR was 0.91±0.11
(n=7; range 0.8 to 1); platelet count,
261.9x103±59.9x103/cm3
(n=8; range 189 to
367x103/cm3); and PTT,
23.94±4.68 (n=7; range 19 to 33) seconds. At the completion of the
study period, INR was 0.99±0.11 (n=8; range 0.8 to 1.2); platelet
count,
273.0x103±51.7x103/cm3
(n=9; range 211 to
332x103/cm3); and PTT,
23.8±5.1 (n=9; range 19.0 to 35) seconds.
Neuroradiological Features
IVH mean volume was 44.26±31.65 (range 1.31 to 100.36)
cm3. An intraparenchymal component was
present in 16 patients, with a mean ICH volume of 6.21±7.53 (range
0 to 23.88) cm3. The treatment effect on IVH is
depicted in Figures 1
and 2
.
|
|
Treatment
Of the initial 12 patients treated with
intraventricular UK as part of the open-label study
or as a novel form of therapy, the mean number of administered doses of
UK in 9 was 6.56±3.75 (range 2 to 14) doses. The dose of UK
administered ranged from 5000 to 12 500 IU, depending on each
centers treatment protocol. Three patients received 10 000 IU of
UK/dose, 1 received only 1 dose of 5 000 IU followed by successive
doses of 10 000 IU, and 8 patients received 12 500 IU. Radiological
and clinical evidence of IVH enlargement occurred in 1 case (MCV
patient 2). No clinical- or laboratory-based evidence of central
nervous system or systemic bleeding was encountered in the
remaining patients enrolled in the open-label or randomized study
cohorts throughout the study period.
Thirty-Day Outcome
We obtained a predicted 30-day mortality of 68.42% (range 3% to
100%) for the entire cohort. Four of 20 patients died within 30 days,
for an observed 30-day mortality of 20%. Probability of observing
4
deaths among 20 patients under a 68.42% expected mortality is
0.000012. The entire study group was divided into an open-label
study/novel-therapy cohort (n=12) and a randomized study cohort (n=8)
for separate analysis of 30-day outcomes.
3 deaths among 12 patients under a 58% expected
mortality is 0.022. | Discussion |
|---|
|
|
|---|
In the last decade, recognition of the proinflammatory role that certain blood components have on neuronal tissue led to a growing interest in inflammation as a mechanism of secondary brain injury. A blood component identified to play a role in the development of acute and chronic brain injury as well as degeneration is thrombin, as was recently independently demonstrated by Nishino and coworkers21 22 and Lee and coworkers.12 13 14 15 23 24 After first developing an animal model of IVH, Pang et al25 have shown that blood and its products produce inflammation and fibrosis of the ependymal lining.
Pang et al25 25 also demonstrated the effect of intraventricular UK in a canine model of IVH. They tested the hypothesis that intraventricular thrombolytic agents hasten the lysis of the intraventricular blood cast and accelerate the return to normal neurological status.25 26 27 These investigators found that clot dissolution was accelerated in the treated group (3 to 6 days) when compared with controls (38 to 65 days) without an added risk of hemorrhagic complications. The clinical correlate to this radiographic improvement was a more-rapid return to normal consciousness and a sustained reduction in ventricular size.
Although IVH is not common, a poll of referral neurointensive care units suggests that IVH is a high-morbidity condition that occurs up to several times per year in region referral centers. A large referral university hospital is anticipated to evaluate and treat 10 to 12 IVH patients per year that require EVD (JHH; unpublished data, 1999). Since 1990, 7 independent case series on the use of thrombolytic agents in IVH have been published.10 28 29 30 31 32 33 A review of these studies shows that 59 ICH and SAH patients were treated with intraventricular thrombolysis. Seventeen patients received UK and 42 received recombinant tissue plasminogen activator (rtPA). Although different thrombolytic regimens and doses were used, these studies reported intraventricular clot lysis within 4 days, no episodes of IVC obstruction, and 5 episodes of CNS infections. Good neurological outcome was reported in 42 of 59 patients. Twenty-two of 59 patients had long-term follow-up: 8 patients developed communicating hydrocephalus that required permanent CSF diversion, 5 patients developed bacterial meningitis, and no episodes of intraventricular or intraparenchymal hematoma enlargement or systemic bleeding occurred. Although these patients were retrospectively reported and did not have matched controls, the resolution process of the intraventricular clots contrasts with the clot lysis rate of untreated patients with IVH as we currently understand it.
On the basis of experimental and clinical evidence of blood and thrombin-induced neurotoxicity and the inherent mechanical injury that IVH produces on brain parenchyma either directly or through the development of acute hydrocephalus, the use of IVCs has become standard care in these patients. Traditionally, the rationale for the use of EVD has been the control of increased ICP produced by acute obstructive hydrocephalus, when present. However, several aspects of this approach deserve analysis: (1) The outcome of IVH patients treated with EVD is not improved despite corrections of ICP by means of CSF drainage.11 (2) Obstruction of the IVC by blood that leads to catheter exchange and its associated risk of brain injury often hampers the efficacy of IVCs.34 35 (3) Although EVD can correct elevated ICP, it does not accelerate the clot lysis that is necessary for restoration of normal CSF dynamics.36 (4) The duration of EVD and the presence of blood and its degradation products is probably responsible for the risk of ventriculitis and the development of delayed communicating hydrocephalus, respectively.37 38 39 We therefore postulate that if the lysis of intraventricular blood is accelerated, distention of the ventricular wall (and surrounding brain tissue) may decrease, better ICP control may be achieved, faster evacuation of the hematoma will take place, and the risk of catheter obstruction with blood clots will decrease or disappear. This final situation was observed only once during the present study.
In our series, 1 instance occurred of intraventricular hematoma enlargement during therapy with intraventricular UK (6.25%). Schwartz et al40 have recently reported 2 cases of worsening IVH during administration of UK for IVH; they recommended exhaustive screening of candidates for this form of therapy. It is unclear whether worsening of IVH was a complication of thrombolytic therapy, because spontaneous enlargement of the blood clot occurs in 38% of ICH patients within the first 20 hours.41 Furthermore, in a cohort of 17 IVH patients, Naff and coworkers36 reported a nearly 15% increase in the initial clot volume in 47.1% of the patients within the initial 48 to 72 hours after hemorrhage.
In summary, our prospective pilot investigation allows us to conclude that the administration of low doses of intraventricular UK in patients with IVH is safe with careful screening. Our data strongly suggest that we can reduce the 30-day mortality of these patients. Sufficient retrospective and prospective evidence supports use of intraventricular UK or other similar thrombolytics for IVH such that a multicenter, randomized, and double-blind study has been recently funded by the US Food and Drug Administration Orphan Drug Program. The trial began patient accrual in January 2000.
Limitations of the Study
The small sample size in the present study limits the
widespread generalization of these preliminary results to all patients
with primary or secondary IVH. However, the study is large enough to
produce results that strongly suggest beneficial effects of
intraventricular administration of UK and to serve
as positive evidence toward the design of a larger study to help
establish the safety and efficacy of this intervention in the treatment
of IVH.
Addendum
Because UK is no longer available for clinical use, rtPA was
chosen for the phase II trial that started January 2000. This decision
was based on experimental evidence of accelerated
intraventricular clot resolution, available
clinical reports of successful intraventricular
administration in IVH patients, and the lack of antigenicity of rtPA as
opposed to streptokinase.10 16 29 31 42 Although the
optimal dose and administration interval of rtPA in the treatment of
IVH are unknown, 3 mg of intraventricular rtPA q 12
hours was chosen for the following reasons: (1) The proposed dose is
within the range of dosages reported in the clinical series previously
cited. (2) The cardiology literature in
coronary thrombolysis (American Hospital
Formulary Service drug information) suggests an approximate
efficacy ratio of 1 mg rtPA to 5000 to 10 000 IU of UK. When this
ratio is applied to the indication of
intraventricular thrombolysis, 3 mg
of rtPA is an approximation consistent with the experience we
report with 12 500 IU of UK. (3) The 12-hour dosing interval for rtPA
administration remains unchanged from prior protocols. Depending on the
phase II study results with rtPA, future dose-determination studies may
be indicated to evaluate higher doses and shorter intervals.
| Acknowledgments |
|---|
Received November 17, 1999; accepted January 20, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
H. Hallevi, K. C. Albright, J. Aronowski, A. D. Barreto, S. Martin-Schild, A. M. Khaja, N. R. Gonzales, K. Illoh, E. A. Noser, and J. C. Grotta Intraventricular hemorrhage: Anatomic relationships and clinical implications Neurology, March 11, 2008; 70(11): 848 - 852. [Abstract] [Full Text] [PDF] |
||||
![]() |
P Longatti, A Fiorindi, F Di Paola, S Curtolo, L Basaldella, and A Martinuzzi Coiling and neuroendoscopy: a new perspective in the treatment of intraventricular haemorrhages due to bleeding aneurysms J. Neurol. Neurosurg. Psychiatry, December 1, 2006; 77(12): 1354 - 1358. [Abstract] [Full Text] [PDF] |
||||
![]() |
NINDS ICH Workshop Participants Priorities for Clinical Research in Intracerebral Hemorrhage: Report From a National Institute of Neurological Disorders and Stroke Workshop Stroke, March 1, 2005; 36(3): e23 - e41. [Abstract] [Full Text] [PDF] |
||||
![]() |
P.L. Longatti, A. Martinuzzi, A. Fiorindi, L. Maistrello, and A. Carteri Neuroendoscopic Management of Intraventricular Hemorrhage Stroke, February 1, 2004; 35 (2): e35 - e38. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. D Hill Case fatality rates after hospital admission for stroke: Covariates were not measured BMJ, May 15, 2003; 326(7398): 1085 - 1085. [Full Text] |
||||
![]() |
S. A. Mayer Ultra-Early Hemostatic Therapy for Intracerebral Hemorrhage Stroke, January 1, 2003; 34(1): 224 - 229. [Abstract] [Full Text] [PDF] |
||||
![]() |
L.B. Morgenstern, A.M. Demchuk, D.H. Kim, R.F. Frankowski, and J.C. Grotta Rebleeding leads to poor outcome in ultra-early craniotomy for intracerebral hemorrhage Neurology, May 22, 2001; 56(10): 1294 - 1299. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. I. Qureshi, S. Tuhrim, J. P. Broderick, H. H. Batjer, H. Hondo, and D. F. Hanley Spontaneous Intracerebral Hemorrhage N. Engl. J. Med., May 10, 2001; 344(19): 1450 - 1460. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2000 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |