RECOMBINANT TISSUE PLASMINOGEN ACTIVATOR IN THE
TREATMENT OF INTRAVENTRICULAR HEMORRHAGE
SECONDARY TO PERIVENTRICULAR ARTERIOVENOUS
MALFORMATION BEFORE SURGERY: CASE REPORT
Krishna Kumar, M.B., M.S.
Department of Surgery, Section of
Neurosurgery, Regina General
Hospital, University of
Saskatchewan, Regina,
Saskatchewan, Canada

Denny D. Demeria, M.D.
Department of Surgery, Section of
Neurosurgery, Regina General
Hospital, University of
Saskatchewan, Regina,
Saskatchewan, Canada

Ashok Verma, M.D.
Department of Radiology, Regina
General Hospital, University of
Saskatchewan, Regina,
Saskatchewan, Canada
Reprint requests:
K. Kumar, M.B., M.S., F.R.C.S. (C),
Medical Office Wing, Regina
General Hospital, Regina, SK S4P
0W5, Canada.
Email: krishna.kumar@rqhealth.ca
Received, September 10, 2002.
Accepted, December 5, 2002.

OBJECTIVE AND IMPORTANCE: Intraventricular hemorrhage (IVH) is known to cause
acute obstructive hydrocephalus, refractory elevated intracranial pressures (ICPs), and
lowered cerebral perfusion pressures, leading to cortical ischemia. Frequent obstruction of external ventricular drains as a result of thrombus is a recurring theme. We
present a case of IVH secondary to periventricular arteriovenous malformation (AVM)
that was not visible at admission angiography and was treated by intraventricular
infusion of recombinant tissue plasminogen activator before surgical intervention.
CLINICAL PRESENTATION: An 11-year-old boy presented with acute onset of headache followed by two seizures, loss of consciousness, decerebration, right temporal
hematoma, IVH, and acute obstructive hydrocephalus.
INTERVENTION: A right external ventricular drain was placed but functioned poorly.
ICP could not be controlled by conventional methods. Five milligrams of recombinant
tissue plasminogen activator was injected into the ventricular system via the external
ventricular drain. This was repeated daily for 4 days. This treatment resulted in
progressive improvement in ICP and clinical status. Once the clot partially cleared,
magnetic resonance imaging and magnetic resonance angiography suggested the
presence of a right periventricular arteriovenous malformation, which was confirmed
by angiography and subsequently resected.
CONCLUSION: Recombinant tissue plasminogen activator is effective in resolving IVH causing obstructive hydrocephalus and uncontrollable ICP posing a life-threatening situation,
secondary to ruptured arteriovenous malformation, before surgical intervention.
KEY WORDS: Arteriovenous malformation, Fibrinolysis, Hydrocephalus, Intraventricular hemorrhage,
Recombinant tissue plasminogen activator
Neurosurgery 52:964-969, 2003

I

ntraventricular hemorrhage (IVH) secondary to trauma, ruptured arteriovenous malformation (AVM), or aneurysm carries a grave prognosis (17). These lesions are notorious for
causing uncontrollable elevated intracranial pressures (ICPs) and
reduced cerebral perfusion pressures (CPPs), leading to ischemia
as a result of recurrent obstruction of external ventricular drains
(EVDs). This, in turn, is responsible for the high morbidity and
mortality of these patients. The literature suggests that in cases of
ruptured AVM or aneurysm, intraventricular fibrinolytic therapy should be initiated after the aneurysm or the AVM has been
surgically or endovascularly secured. Thus far, there is only one
report in the literature of the use of recombinant tissue plasminogen activator (rtPA) when treating an unsecured AVM (13).
Moreover, some authors have postulated that rtPA is contraindicated because of the perceived danger of rehemorrhage in
these cases (16).

964 | VOLUME 52 | NUMBER 4 | APRIL 2003

DOI: 10.1227/01.NEU.0000053028.06474.C6

www.neurosurgery-online.com

We present a case of a ruptured AVM situated in the right
temporal lobe, just medial to the temporal horn. This resulted in
IVH filling the lateral, third, and fourth ventricles, which subsequently caused life-threatening obstructive hydrocephalus and
raised ICP, which was refractory to conventional therapy. This
was treated with intraventricular rtPA to lyse the hematoma,
retain the patency of the EVD, and normalize the ICP, which
allowed for subsequent visualization and resection of the AVM.

CASE REPORT
A previously well 11-year-old boy awoke one morning
screaming; he complained of severe generalized headache and
experienced episodes of vomiting. He became unconscious
while en route to the local hospital and sustained two generalized tonicoclonic seizures. He was noted to have aniscoria,

www.neurosurgery-online.com

TISSUE PLASMINOGEN ACTIVATOR FOR INTRAVENTRICULAR HEMORRHAGE

with the right pupil being greater than the left. He was intubated and transferred to our facility for further investigation.
On presentation to our emergency department, the patient
was decerebrate, with a Glasgow Coma Scale score of 4. Pupils
were now equal and reactive. A funduscopic examination was
within normal limits. Laboratory investigations revealed a
normal coagulation profile. A computed tomographic (CT)
scan of the head revealed extensive IVH filling the entire
ventricular system, with predominance in the right frontal and
temporal horns, with a resultant midline shift of 4 mm (Fig. 1).
The severity of the IVH was a score of 4 according to the Le
Roux classification (10). A pancerebral angiogram demonstrated no evidence of aneurysm, AVM, or arterial spasm (Fig.
2).
The patient was transferred to the operating room, and an
EVD was introduced into the right frontal horn. Thick, bloody
cerebrospinal fluid (CSF) began to flow through the ventriculostomy under high pressure. In the intensive care unit, the
patient was sedated and paralyzed with atracurium. However, the ICP remained elevated, ranging between 20 and 60
mm Hg, despite frequent administration of mannitol. The
EVD required frequent irrigation, because blood products frequently obstructed it. Phenobarbital coma was introduced on
Day 3 after admission, because the ICP peaked at 80 mm Hg
for 1 hour and the CPP decreased to 30 mm Hg. Unfortunately, the phenobarbital coma failed to control the elevated
ICP and precipitated difficulties in maintaining the CPP because of the hypotensive effects of phenobarbital. Therefore, it
was discontinued. Despite these efforts, the patient’s ICP remained elevated to 40 mm Hg. The CPP decreased to 45
mm Hg. The possibility of intraventricular rtPA was considered as a rescue measure to control the ICP by way of clot
lysis. After discussing this with other colleagues, we explained
the situation to the family. The risks and benefits of rtPA were
presented, and the family agreed to its off-label use. Thus, on
Day 4 after admission, 5 mg of rtPA in 5 ml of normal saline

FIGURE 1. A and B, unenhanced axial CT scans of the head at two different levels obtained at admission to hospital, showing blood filling the
lateral, third, and fourth ventricles.

NEUROSURGERY

FIGURE 2. A and B, right vertebral angiograms obtained at admission
(A, lateral view; B, anteroposterior view). C–E, right carotid angiograms
obtained at admission (C, lateral view; D, anteroposterior view; E, oblique
view). There is no evidence of aneurysm or AVM in either the anterior or
posterior circulation.

was administered through the EVD, followed by 1 ml of sterile
water. The catheter was clamped for 1 hour and then opened.
There was immediate flow of bloodstained CSF. The ICP
normalized over the next 12 hours. A CT scan showed a 30%
decrease in the amount of intraventricular blood (Fig. 3). However, the ICP began to increase again. Encouraged by the
positive response to the initial rtPA administration, we decided to repeat administration of rtPA for the next 3 consecutive days. Each day, before administration of the rtPA, a CT
scan was repeated to follow the progress of clot lysis and to
ensure that no new hemorrhage occurred as a complication of
rtPA administration. After the last dose, a CT scan revealed no
midline shift and no ventricular dilation, with only 15% residual blood in the right lateral ventricle compared with the
initial CT scan. The ICP could now be controlled by intermittent venting of the EVD. The atracurium was discontinued,
and sedation was reduced. By Day 10 after admission, the
patient still required EVD venting two to three times per day
to maintain the ICP within the normal range. The EVD was
removed from the right side and reinserted on the left to
prevent the possibility of infection. The patient was extubated
on Day 9 after admission. By Day 15 after admission, the
ventriculostomy was no longer required and was removed. A
repeat CT scan on Day 16 showed normal ventricles, complete
resolution of IVH, and a low-density lesion just medial to the
right temporal horn, secondary to the hemorrhage that was in
this area (Fig. 4). This was the area that was previously obscured by blood products.
To further evaluate the pathogenesis of the hemorrhage,
magnetic resonance imaging (MRI) was performed 1 month
after admission; to our surprise, the MRI scan revealed flow
voids in the right temporal lobe, raising the suspicion of an
AVM (Fig. 5). There was also extensive signal abnormality

VOLUME 52 | NUMBER 4 | APRIL 2003 | 965

KUMAR ET AL.

CPP, and cortical ischemia,
leading to a poor prognosis
(14, 19). Correlation also exists in predicting the extent
of ventricular dilation and
the amount of intraventricular blood on early CT
scans (3, 7, 15). Therefore,
the thrust of intervention in
IVH is to rapidly eliminate
the intraventricular blood,
to divert the ventricular
CSF, to decrease ventricular dilation, and to normalize the ICP (12). Surgical
evacuation of intraventricular blood is hazardous, if
not impossible, when most FIGURE 4. Unenhanced axial CT
of the ventricular system is scan of the head, obtained at Day 16
filled with blood (12). The after admission. An EVD is seen in
EVD usually fails in elimi- the left lateral ventricle. There is
nation of the intraventricu- complete resolution of the IVH. Lowdensity lesions are noted just medial
lar blood, because it is freto the right temporal lobe (arrow).
quently
obstructed
by
thrombus; natural clot lysis
has a latency period of 48 to 72 hours after the initial ictus (16).
It is during this latency period that extrinsic rtPA could accelerate the natural intrinsic process of clot lysis if injected at an
optimum dose. In cases in which the ICP cannot be controlled,
the mortality of such patients approaches 78% (17).
Under these desperate circumstances, various fibrinolytic
agents have been tested in cases of IVH caused by aneurysmal
rupture, trauma, or hypertension. Currently, the contraindications for intraventricular administration of fibrinolytics include unsecured cerebral aneurysms, untreated AVMs, and
uncorrected coagulopathies. In most cases, fibrinolytic therapy is initiated only after the aneurysm has been secured (2, 4,

throughout the right temporal lobe. A repeat cerebral
angiogram was performed,
which confirmed the presence of an AVM in the right
temporal lobe, fed by the parietal branches of the right
posterior cerebral artery
(Fig. 6). The AVM was best
visualized on a right vertebral injection. Subsequently,
the patient underwent craniotomy, and the AVM was
excised. A postoperative angiogram confirmed its total
excision (Fig. 7). The patient
was discharged 7 days after
surgery.
The patient has since been
followed up on an outpatient
basis, with the most recent review being 6 months after resection of his AVM. On examination, he is alert and
oriented. He has a left homonymous hemianopsia, mild
left-sided ataxia, and hyperreflexia with bilateral flexor
plantar responses. Sensory
functions are within normal
limits. He has since returned
to school; however, he requires personal assistance in
language and mathematics.

DISCUSSION
IVH is a significant complication of hemorrhage regardless of cause and carries
a high morbidity and mortality rate (3, 7, 11, 20). This
is because of persistently
raised ICP, which is difficult
to control, despite the use of
EVDs, because of recurrent
blockage by blood clots occupying the ventricular system. This complication is
compounded by the obstruction of the CSF pathways by blood products in
the subarachnoid space. A
direct correlation exists between persistently elevated
ICP, resulting in reduced

FIGURE 3. Unenhanced axial CT scans
of the head. A, an EVD in place in the
right frontal horn. B, CT scan performed
24 hours after the initial dose of rtPA was
administered. C, CT scan after the fourth
dose of intraventricular rtPA. There is
complete clearing of blood from the frontal
horns, with residual blood in the right
temporal horn.

966 | VOLUME 52 | NUMBER 4 | APRIL 2003

FIGURE 5. MRI scans of the
head. A, coronal dual-echo T2weighted scan showing a small
cluster of flow-void abnormalities in the anteromedial aspect of the right temporal lobe
(arrow). B, axial T2-weighted fast field echo revealing similar abnormalities (arrow).

www.neurosurgery-online.com

TISSUE PLASMINOGEN ACTIVATOR FOR INTRAVENTRICULAR HEMORRHAGE

16, 21). There is only one publication
describing the use of rtPA for IVH in a
case of ruptured AVM before surgical
excision (13). The current report may
be the second reported case of the successful use of rtPA before excision of
an AVM in the treatment of IVH associated with uncontrolled elevated ICP.
Our case presents a unique feature regarding how rtPA can help discover
FIGURE 6. Right vertebral angiograms (A, lateral view; B, anteroposterior view; C, oblique view)
the source of IVH when the blood
repeated 5 weeks after admission showing the AVM (arrows) in the right temporal lobe fed by branches
completely obscures the region of the of the posterior cerebral artery.
underlying pathology. This report
supports the observation by Mayfrank
et al. (13) that intraventricular administration of rtPA is not
examined. Doses of 0.5 to 2 ␮g of rtPA induced lysis of 80 to
necessarily contraindicated without previous excision of a
88% of clot at 24 hours. Regional cerebral blood flow, an
ruptured AVM. Indeed, the potential lifesaving benefits outindicator of possible brain ischemia, was decreased to 37% of
weigh the inherent risks of rehemorrhage in such a case.
baseline value at 24 hours after the buildup of IVH compared
The use of rtPA was considered in this case because of the
with the controls. An analysis of various subgroups revealed
desperate situation with which we were confronted when our
that a dose of 0.5 ␮g of rtPA alleviated this, whereas doses of
attempts at controlling ICP and maintaining CPP by conven1 to 2 ␮g were associated with a decrease in regional cerebral
tional methods had failed. We initially used EVD, hyperventilablood flow by 36 to 40%. Higher doses of rtPA (1–2 ␮g)
tion, atracurium, and finally phenobarbital coma. Intraventricuinduced a greater degree of severe periventricular edema,
lar rtPA was chosen in preference to other fibrinolytic agents
choroid plexus edema, and bilateral intraventricular leukocybecause of its significantly higher lytic effects on experimental
tosis. After reviewing the standards of clinical practice and
subarachnoid hemorrhages on a per molar basis (9). We chose a
reviewing the literature, Wang et al. (22) concluded that an
dose of 5 mg of rtPA because the literature has demonstrated that
optimal dosage of 0.1 mg of rtPA per milliliter of blood should
doses between 2 and 5 mg have been used without hemorrhagic
be used. This indicates that the dose of rtPA must be carefully
sequelae (4, 5, 9, 12, 18). The duration of intraventricular fibrinodetermined by the quantity of IVH to efficaciously reduce the
lytic therapy in this case was determined by the guidelines
mass effect and to avoid unnecessary adverse reactions.
suggested in the literature (1, 6, 12), with specific attention being
given to the maintenance of ventricular drainage, CT evidence of
Role of MRI
intraventricular clot resolution, and the ability to normalize the
MRI is moderately effective in detection of AVM as a cause
ICP. In our patient, four doses of rtPA were required to achieve
of
intracerebral hemorrhage or IVH, particularly in a younger
stabilization of his ICP and CPP.
population. In an analysis of a patient population less than 20
years old with intracerebral hemorrhage, Griffiths et al. (8)
rtPA Dosage
found that AVMs were responsible for 65%, aneurysms acThe optimal dosage of rtPA remains a controversial issue.
counted for 5%, and no cause was detected for intracerebral
Until now, few studies have investigated the dose-effect relahemorrhage or IVH in the remaining 30% of the patients. A
tionship of rtPA for IVH treatment. Most of the clinical studies
review of the literature fails to give any guidelines regarding
(9, 12, 18) report the total milligram dose of rtPA without
the stage of clot resolution at which MRI can diagnose the
mentioning the total interventricular clot volume. A review of
underlying cause of hemorrhage, especially when the admisthe literature (4, 5, 17) detailing rtPA treatment and patient
sion CT scan and cerebral angiography are negative. The
outcomes after a dose of 8 mg of rtPA shows that patients with
reason for the AVM not being visualized at admission by CT
moderate IVH (Graeb score, 5–8) had a lower incidence of
good outcome and a higher mortality rate than those with
severe IVH (Graeb score, 9–12). It seems that 8 mg was probably an excessive dose for patients with moderate IVH.
Wang et al. (22), in a recent work, addressed the dose-effect
relationship of rtPA in the treatment of experimental IVH in
the rodent model. After induction of experimental IVH in the
left lateral ventricle, either 0.25 to 2 ␮g of rtPA or Ringer’s
saline was injected, and results were compared with those in
controls and rats receiving injections of saline solution. Residual intraventricular clot volumes, ventricular volumes, reFIGURE 7. Postoperative right vertebral angiograms (A, anteroposterior
gional cerebral blood flow, and brain histopathology were
view; B, lateral view) showing total excision of the AVM.

NEUROSURGERY

VOLUME 52 | NUMBER 4 | APRIL 2003 | 967

KUMAR ET AL.

scan and angiography may have been the compressive effect
of the hematoma on the AVM, preventing its visualization. In
these circumstances, MRI at admission may also not reveal an
underlying AVM because of the absence of flow voids. A few
days after ictus and before cerebral angiography is repeated,
MRI has the greatest potential in disclosing the underlying
cause, as was the case here.

CONCLUSIONS
This case study illustrates that rtPA can be used without
producing hemorrhagic sequelae in cases of IVH secondary to
a ruptured AVM before surgical excision, when other conventional methods have failed. Although this is only the second
such case reported in the literature, it confirms that the presence of an unsecured AVM is not an absolute contraindication
for fibrinolytic therapy, as has been propagated previously.
Although the recent study by Wang et al. (22) concludes that
a dose of 0.1 mg/ml of rtPA should be used, the specific effect
of rtPA alone on neural tissue remains to be identified.

REFERENCES
1. Akdemir H, Selcuklu A, Pasoglu A, Oktem IS, Kavuneu I: Treatment of
severe intraventricular hemorrhage by intraventricular infusion of urokinase. Neurosurg Rev 18:95–100, 1995.
2. Azmi-Ghadimi H, Heary RF, Farkas JE, Hunt CD: Use of intraventricular
tissue plasminogen activator and Guglielmi detachable coiling for the acute
treatment of casted ventricles from cerebral aneurysm hemorrhage: Two
technical case reports. Neurosurgery 50:421–425, 2002.
3. de Weerd AW: The prognosis of intraventricular haemorrhage. J Neurol
222:46–51, 1979.
4. Findlay JM, Grace MG, Weir BKA: Treatment of intraventricular hemorrhage with tissue plasminogen activator. Neurosurgery 32:941–947, 1993.
5. Goh KY, Poon WS: Recombinant tissue plasminogen activator for the treatment
of spontaneous adult intraventricular hemorrhage. Surg Neurol 50:526–532, 1998.
6. Grabb PA: Traumatic intraventricular hemorrhage treated with intraventricular recombinant-tissue plasminogen activator: Technical case report.
Neurosurgery 43:966–969, 1998.
7. Graeb DA, Robertson WD, Lapointe JS, Nugent RA, Harrison PB: Computed
tomographic diagnosis of intraventricular hemorrhage: Etiology and prognosis. Radiology 143:91–96, 1982.
8. Griffiths PD, Beveridge CJ, Gholkar A: Angiography in non-traumatic brain
hematoma: An analysis of 100 cases. Acta Radiol 38:797–802, 1997.
9. Kajimoto Y, Ohta T, Kuroiwa T: Comparison of intrathecally administered
urokinase, tissue type plasminogen activator, and combination of urokinase
and lysine-plasminogen for clot lysis after experimental subarachnoid hemorrhage in dogs. Neurosurgery 40:572–577, 1997.
10. Le Roux PD, Haglund MM, Newell DW, Grady MS, Winn HR: Intraventricular hemorrhage in blunt head trauma: An analysis of 43 cases. Neurosurgery 31:678–685, 1992.
11. Little JR, Blomquist GA Jr, Ethier R: Intraventricular haemorrhage in adults.
Surg Neurol 8:143–149, 1977.
12. Mayfrank L, Lippitz B, Groth M, Bertalanffy H, Gilsbach JM: Effect of
recombinant tissue plasminogen activator on clot lysis and ventricular dilatation in the treatment of severe intraventricular haemorrhage. Acta
Neurochir (Wien) 122:32–38, 1993.
13. Mayfrank L, Rohde V, Gilsbach JM: Fibrinolytic treatment of intraventricular haemorrhage preceding surgical repair of ruptured aneurysms and
arteriovenous malformations. Br J Neurosurg 13:128–131, 1999.
14. Meyer SA, Kessler DB, Van Heertum RL, Thomas CE, Fink ME, Brannigan
T: Effect of intraventricular blood on global cortical perfusion in acute
intracerebral hemorrhage: A single-photon emission computed tomographic
study. Ann Neurol 38:288, 1995 (abstr).

968 | VOLUME 52 | NUMBER 4 | APRIL 2003

15. Mohr G, Ferguson G, Khan M, Malloy D, Watts R, Benoit B, Weir B:
Intraventricular haemorrhage from ruptured aneurysm: Retrospective analysis of 91 cases. J Neurosurg 58:482–487, 1983.
16. Naff NJ: Intraventricular hemorrhage in adults. Curr Treat Options Neurol
1:173–178, 1999.
17. Nieuwkamp DJ, de Gans K, Rinkel GJ, Algra A: Treatment and outcome of severe
intraventricular extension in patients with subarachnoid or intracerebral hemorrhage: A systematic review of the literature. Neurology 247:117–121, 2000.
18. Rhode V, Schaller C, Hassler WE: Intraventricular recombinant tissue plasminogen activator for lysis of intraventricular hemorrhage. J Neurol
Neurosurg Psychiatry 58:447–451, 1995.
19. Rosenberg GA, Saland L, Kyner WT: Pathophysiology of periventricular tissue
changes with raised CSF pressure in cats. J Neurosurg 59:606–611, 1983.
20. Ruscalleda J, Peiro A: Prognostic factors in intraparenchymatous hematoma
with ventricular haemorrhage. Neuroradiology 28:34–37, 1986.
21. Usui M, Saito N, Hoya K, Todo T: Vasospasm prevention with postoperative
intrathecal thrombolytic therapy: A retrospective comparison of urokinase,
tissue plasminogen activator, and cisternal drainage alone. Neurosurgery
34:235–245, 1994.
22. Wang YC, Lin CW, Shen CC, Lai SC, Kuo JS: Tissue plasminogen activator
for the treatment of intraventricular hematoma: The dose-effect relationship.
J Neurol Sci 202:35–41, 2002.

COMMENTS

I

would not recommend knowingly instilling tissue plasminogen activator (rtPA) into the ventricles of patients with known
unsecured, ruptured vascular lesions. In this case, however, it is
likely that the arteriovenous malformation (AVM) was relatively
protected from rebleeding because of some degree of posthemorrhagic thrombosis or focal pressure from the hematoma. Furthermore, it was likely fairly distant from the site of administration of the rtPA, which may have reduced the risk of recurrent
bleeding. However, catastrophic bleeding can occur even with
the administration of small doses of intracranial rtPA. This point
is well documented in the literature. If the initial angiography
had shown an AVM, I would have operated and resected it and
then administrate the rtPA. If not, I would have administered
rtPA, as the authors did. Generally, though, at my institution, my
colleagues and I administer 0.5 to 1 mg twice a day rather than
the doses administered here.
R. Loch Macdonald
Chicago, Illinois

T

he authors have exercised aggressive and thoughtful management of a difficult case with a dismal prognosis. Their
heroic interventions clearly contributed to lifesaving and
eventually to the favorable outcome. This case illustrates several salient features related to acute intraventricular hemorrhage (IVH) from periventricular AVMs. The authors could be
criticized in retrospect for not diagnosing the AVM during the
acute phase of the illness. A contrast-enhanced computed
tomographic scan would likely have revealed the lesion, and
certainly magnetic resonance imaging performed earlier
would have alerted the authors to the AVM. A review of the
initial arteriogram and selected images displayed in the article
reveals a suspicious vascular blush in the inferior temporal
region, where the AVM was subsequently discovered. It is
likely that careful analysis of all angiographic films would
have revealed a hint of the compressed nidus. Nevertheless, it

www.neurosurgery-online.com

TISSUE PLASMINOGEN ACTIVATOR FOR INTRAVENTRICULAR HEMORRHAGE

is not unusual for a small AVM to be angiographically occult
in the setting of acute compression and elevated intracranial
pressure (ICP). A repeat angiogram clearly revealed the lesion.
Ironically, had the authors suspected that they were dealing
with an acutely ruptured periventricular AVM, they would have
been less likely to use intraventricular rtPA thrombolysis. In turn,
they would have been less likely to administer lifesaving treatment! The common wisdom has largely contraindicated intracranial thrombolysis in the setting of untreated AVMs and aneurysms. Yet, another report in the literature also confirms that
thrombolysis may be used without inevitable rebleeding from
the AVM. Periventricular AVMs are particularly prone to rehemorrhage, and the safety of this treatment is particularly notable
in this regard. Nevertheless, I must caution that these two reported cases do not prove the safety of this treatment with high
confidence. It may be advisable to include cases of known AVMs
with otherwise intractable ICP in subsequent Phase I clinical
trials to assess the safety of intraventricular rtPA. The mortality
rates of patients with undrained ventricular casts associated with
intractable intracranial hypertension is high. Hence, judicious,
compassionate use of this treatment should be considered in
such cases, and it is preferable that this information be monitored
in ongoing clinical trials designed to address the safety and
effectiveness of intraventricular thrombolysis.
Issam A. Awad
Denver, Colorado

K

umar et al. present an interesting case of a pediatric patient
who had a diffuse IVH from a ruptured AVM. The initial
angiographic evaluation did not reveal the source of the hemorrhage, and external ventricular drainage was difficult because of
repeated episodes of catheter occlusion by clot. In this difficult
clinical situation, a decision was made to instill rtPA into the
lateral ventricle. This treatment was successful in establishing
reliable external ventricular drainage with subsequent control of
ICP. Hemorrhage from the AVM did not recur.
This interesting case report documents the feasibility of the
intraventricular administration of rtPA for the treatment of IVH
in a patient with a ruptured AVM. I do not think, however, that
this approach can be recommended as a routine treatment. If an
AVM ruptures into the ventricular system and rtPA causes lysis
of clot in the ventricular system, one must assume that this
treatment will increase the risk of recurrent hemorrhage. That

this did not happen in one or two cases is of interest but is not
reassuring.
I wonder if the authors would have administered intraventricular rtPA if the source of hemorrhage had been documented on the initial angiogram. When facing a similar situation and with the knowledge that the source of hemorrhage
is a ruptured AVM, I would elect to proceed with surgical
evacuation of the intraventricular clot rather than administer
rtPA. I have, in fact, used this approach successfully on several
occasions. The authors have successfully treated a young patient in a difficult situation and report their experience in a
clear, concise fashion, but they probably should not use this
treatment in this type of situation again.
Robert E. Harbaugh
Lebanon, New Hampshire

K

umar et al. provide additional evidence supporting the utility of administering intraventricular rtPA to manage severe
IVH. That the authors successfully cleared the ventricles and
saved the patient’s life despite the presence of a periventricular
AVM is noteworthy in several respects. First, it underscores the
need for rapid and aggressive treatment in patients with ICP
secondary to hydrocephalus. Although there are definite risks in
using local thrombolytic therapy in patients with unsecured vascular lesions, these risks are justified when not using thrombolytics would lead to almost certain death or severe, permanent
deficits. This case also underscores that early rebleeding from
AVMs, even in the periventricular region, is less common than
early rebleeding from arterial aneurysms (with or without associated AVMs). One might surmise that early rebleeding from a
ruptured AVM nidus is comparatively rare, especially in angiographically occult cases, in part because the cause of rehemorrhage is less likely to be clot resolution than other factors such as
ongoing thrombosis of critical venous drainage. Whatever the
case, if the patient had presented with a visible AVM, we would
have made every effort to secure the bleeding site endovascularly
before administering rtPA. If this procedure is not possible and
the patient is dying from untreated hydrocephalus, intraventricular rtPA might be indicated in selected cases after careful consideration of all options.
E. Sander Connolly, Jr.
Sean D. Lavine
New York, New York

Customer Service Contact Information
CNS Members: Call Sue Souders at: 301/223-2325.
Non-CNS Members: Call either 301/223-2300 or 1-800/638-3030.
Customer Service fax: 301/223-2400.
Customer Service email: customerservice@lww.com

NEUROSURGERY

VOLUME 52 | NUMBER 4 | APRIL 2003 | 969