Intracranial Venous Thrombosis After Placement of a Lumbar
Mitchell G. Miglis•David N. Levine
Published online: 16 October 2009
? Humana Press Inc. 2009
neuroscience and are often managed in the neurointensive
care unit. Complications are generally rare, and intracranial
We report the case of a 45-year-old woman who
developed a cerebrospinal fluid (CSF) leak after spinal
surgery. Fifteen hours after placement of a lumbar drain she
developed pure alexia and color agnosia caused by left
lateral sinus thrombosis with hemorrhagic infarction in the
posterior inferior left temporal lobe. We review the litera-
ture on the association of IVT with injury to the spinal dura,
and we propose a mechanism whereby the lumbar drain
may facilitate its development.
nearly all cases were associated with post-dural puncture
headache, which occurs in only a minority of cases of dural
puncture. Injury to the spinal dura alters the distribution of
craniospinal elasticity causing profound intracranial CSF
dilation of cerebral veins resulting in both orthostatic head-
ache and venous stasis. We propose that placement of the
lumbar drain and elevation of the head of the bed aggravated
intracranial CSF hypotension and facilitated IVT.
When a lumbar drain is placed for treatment
of a spinal CSF leak, the patient should remain flat in bed.
Any patient with post-dural injury headache that intensifies
after an initial plateau, persists for longer than a week, or
Lumbar drains are frequently used in clinical
loses its orthostatic character should be evaluated for
intracranial sinus or venous thrombosis.
Intracranial sinus thrombosis ? Spinal puncture ?
Post-dural puncture headache ? Pure alexia and color
agnosia without hemianopsia
Lumbar spinal fluid drainage ?
External lumbar cerebrospinal fluid (CSF) drainage is
widely employed in clinical neurosurgery and neurology. It
is used intraoperatively to relax the brain and improve
surgical access to deep structures , and postoperatively
to prevent CSF fistulas . It is employed therapeutically
to induce closure of refractory CSF leaks  and diag-
nostically to predict shunt responsiveness in patients with
suspected normal pressure hydrocephalus . The proce-
dure is generally well tolerated [3, 5], the most common
complications being transient nerve root irritation and post-
dural puncture headache (PDPH) . Infection occurs in
under 0.2% in recent series [3, 5]. Rare serious complica-
tions include symptomatic pneumocephalus in patients
being treated for CSF leaks into an air sinus  and
downward herniation at the tentorium or the foramen
magnum [6, 7]. We report a case in which placement of an
external lumbar drain to facilitate closure of a CSF leak
was complicated by intracranial venous thrombosis (IVT).
A 46-year-old woman was admitted to the orthopedic
service for spinal surgery. She had a long history of neck
M. G. Miglis ? D. N. Levine (&)
Department of Neurology, RIRM-311, New York University
School of Medicine, 400 East 34th Street, New York,
NY 10016, USA
Neurocrit Care (2010) 12:83–87
pain radiating to the right shoulder and numbness in the
right hand. MRI of the cervical spine showed multilevel
cervical spondylosis and cord compression. Her medical
history was significant for migraine headaches and use of
oral contraceptives. She had smoked a pack of cigarettes
every 2 weeks for 3 years but had quit 3 months before
admission. Physical examination revealed neck tenderness,
moderate weakness of the right upper extremity, and absent
biceps and triceps reflexes on the right.
The patient underwent anterior cervical discectomy and
fusion of C4–C7. The surgery was complicated by dural
and spinal cord injury, evidenced by leakage of CSF from
the operative wound, orthostatic cervical and occipital
headache, and weakness and sensory impairment in the left
arm and leg. Repeat MRI of the cervical spine showed
swelling and T2 signal hyperintensity of the spinal cord
from C3 to C6 consistent with contusion and edema. She
was treated with intravenous dexamethasone 6 mg every
6 h to be tapered after eight doses.
When the CSF leak failed to resolve, the patient was
transferred to the neurosurgical service for insertion of a
lumbar drain. On hospital day #9, a 16-gauge lumbar drain
catheter was inserted through a spinal needle in the sub-
arachnoid space at L4-5. The catheter was attached to an
external CSF collection system. The height of the collect-
ing chamber was adjusted to drain 10–15 ml/h. The head of
the bed was elevated 45?, and headache was treated with
Fifteen hours later the patient noticed that she was
unable to read. Examination showed fluent speech with
some difficulty naming objects. She could not read, but she
wrote to dictation normally. Peripheral visual fields were
full to confrontation. The weakness and sensory loss of her
left arm and leg were unchanged. Head CT revealed
intracerebral hemorrhage in the left inferior temporal lobe
(Fig. 1). On the following day, she vomited and developed
severe headache. MRI confirmed hemorrhage with mild
surrounding edema in the left posterior inferior temporal
lobe. Venous time-of-flight MRA revealed lack of flow in
the left transverse sinus and internal jugular vein, consis-
tent with lateral sinus thrombosis (Fig. 2). Anticoagulation
was initially withheld because of the temporal location of
the hemorrhage. By hospital day #13, drainage from the
neck wound had ceased, and the lumbar drain was removed
the following day.
A hematological consultant found no past history of
thrombotic events or miscarriages. Laboratory studies for
hypercoagulability showed only moderate decreases in
protein S, protein C, and marginal decreases in anti-
thrombin III and factor VIII, which may have been the
consequences of recent thrombosis. Platelet count, pro-
thrombin time, partial thromboplastin time, and fibrinogen
level were normal. There was no evidence of prothrombin
gene mutation, Factor V Leiden mutation, or anti-cardio-
Repeat MRI on hospital day #19 showed more extensive
Fig. 1 Non-contrast head CT on hospital day #10. There is acute
hemorrhage in the posterior inferior left temporal lobe
Fig. 2 Magnetic resonance venogram, in antero/posterior projection
on hospital day #11. There is abrupt cessation of flow in the left
transverse sinus, consistent with sinus thrombosis. There is no flow in
the left sigmoid sinus or in the left internal jugular vein
84Neurocrit Care (2010) 12:83–87
causing slight compression of the left temporal horn
(Fig. 3). Anticoagulation with heparin was started on
hospital day #21. On hospital day #25, she complained of
pain in the calf of the paretic left lower extremity. Venous
ultrasound showed occlusion of the peroneal vein. An
inferior vena cava filter was inserted the following day.
During this time the neurological examination changed
very little. Spoken language was normal except for slight
difficulty naming objects. Episodic memory was intact. She
copied complex diagrams accurately. Visual fields by
confrontation were normal. There was severe alexia for
words and numbers. She could not identify single letters
and could read no words. She was able to identify single
numerals with effort but struggled with multidigit numbers.
In contrast, she easily wrote sentences and multidigit
numbers to dictation. There was also severe difficulty
identifying colors. She was almost always incorrect in
naming colors, in selecting the correct color in response to
a dictated color name, or in sorting a variety of colors
according to hue. She was discharged to a rehabilitation
facility on hospital day #37.
In this patient, spinal surgery resulted in dural injury with
CSF leakage and PDPH. Shortly after a lumbar drain was
inserted to treat the CSF leak the patient developed IVT
with venous infarction of the left temporal lobe.
IVT following spinal dural injury has been reported
repeatedly, and the association is unlikely to be coinci-
dental. The number of reported cases in which IVT
developed within days of dural injury is now considerable.
Aidi et al.  reviewed 13 cases reported between 1986
and 1999. We found 16 additional cases [9–18] including
our own. In addition, IVT has occurred after a variety of
forms of spinal dural injury, suggesting a correlation with
the dural injury itself and not with the particular manner in
which the injury occurred. Thus, IVT has been reported
after unintentional dural penetration during epidural anes-
thesia [13, 19–22] or injections , diagnostic lumbar
puncture [8–10, 14, 15, 17, 18, 22], spinal anesthesia
[13, 23], intrathecal injections [11, 12], and myelography
[22, 24, 25]. Finally, when IVT occurs after spinal dural
injury there is a nearly constant association with PDPH,
which was present in at least 26 of the 29 reported cases,
including our own. If the co-occurrence of IVT and dural
puncture were coincidental, the majority of cases of IVT
should occur in patients without PDPH, since PDPH typi-
cally occurs in only 10–15% of patients undergoing lumbar
puncture on a neurological service . Previous reports
[8, 22] have noted a consistent pattern of headache when
IVT complicates dural injury. The initial orthostatic
headache may intensify and may either persist or lose its
orthostatic character after several days, unlike the typical
PDPH, which remains orthostatic and generally disappears
within a week.
The near constant association with PDPH suggests that
the mechanism underlying PDPH may also lead to IVT.
The most likely cause of PDPH is acute dilation of intra-
cranial veins brought on by assuming the erect posture
. When the spinal dura is injured, the distribution of
craniospinal elasticity is altered, so that compliance
increases at the caudal end of CSF space. As a result, the
zero point of CSF pressure moves caudally in the erect
position, producing an abnormally large decrease in intra-
cranial pressure upon sitting or standing. This intracranial
hypotension augments the transmural venous pressure
gradient, causing the intracranial veins to dilate acutely,
over and above the pre-existing dilation in the recumbent
position caused by reduced CSF filling pressure. The acute
dilation stimulates pain receptors in the region of the dural
sinuses, causing the headache. An additional consequence
of such acute venous dilation is venous stasis. Flow
through a blood vessel is the product of its cross-sectional
area and flow velocity. If volume flow is kept constant by
autoregulation, flow velocity will decrease as the vein
dilates. The dural sinuses will also dilate, although to a
lesser degree than the more thin-walled veins. Canha ¨o et al.
 found an average 47% decrease in the velocity of
blood flow in the straight sinus of supine patients shortly
after lumbar puncture.
Fig. 3 MRI on hospital day #19. The FLAIR sequence shows
multiple foci of hyperdense left temporal subacute hemorrhage
surrounded by less hyperdense edema
Neurocrit Care (2010) 12:83–87 85
We suggest that the lumbar drain in our case exacer-
bated venous stasis and contributed to the development of
IVT. The degree of negative intracranial pressure that
develops with a lumbar drain is the difference between the
height of the head and the top of the external CSF col-
lecting chamber, which is exposed to atmospheric pressure.
If the site of the leak being treated is also at or near
atmospheric pressure, the collecting chamber must be
placed below the level of the leak in order for CSF to flow
through the drain instead of the leak. With the head of the
bed elevated, vertical distance is introduced between the
head and the site of the leak, and an even greater distance is
introduced between the head and the collecting chamber of
the lumbar drain. The resulting intracranial CSF hypoten-
sion produces venous stasis. Our patient was being treated
for a spinal CSF leak at the low cervical level, which
caused PDPH but not thrombosis even before insertion of
the lumbar drain. After the drain was inserted the head of
the bed was elevated to 45?. The resulting increase in
intracranial hypotension likely promoted further intracra-
nial venous dilation and stasis, facilitating thrombosis.
Nevertheless, IVT remains very rare compared to
PDPH. This suggests that the degree of venous stasis
induced by spinal CSF leakage is usually insufficient to
produce IVT except in patients who are predisposed to
venous thrombosis for other reasons. Our patient had
multiple prothrombotic risk factors, including use of oral
contraceptives [29, 30], high-dose glucocorticoids [31, 32],
recent surgery, and a history of migraine. At least 22 of the
29 previously reported cases of IVT complicating dural
injury had additional prothrombotic risk factors, including
congenital thrombophilia, the postpartum state, oral con-
traceptive use, high-dose glucocorticoids therapy, and
In summary, IVT may complicate placement of a
lumbar drain that is being used to treat a spinal CSF leak.
We suggest that all such patients be kept flat, particularly
if PDPH is present before the drain is inserted. In any
patient with PDPH, the supine posture will not only
alleviate headache, but will also prevent the acute
venodilation that promotes intracranial venous stasis. If
our formulation of pathophysiology is correct, patients
with lumbar drains for other reasons are not at risk for
IVT unless excessive drainage is allowed to occur, since
the level of the external CSF collecting chamber is not
constrained by the site of a CSF leak and can be adjusted
to avoid overdrainage and profound intracranial CSF
hypotension. Any patient with PDPH after spinal dural
injury of any type, whose headache intensifies after its
initial plateau, loses its orthostatic character, or persists
unabated for longer than a week, should undergo imaging
to rule out IVT.
Monica Lester Family Trust.
This work was supported by a grant from the
1. Michenfelder JD, Gronert GA, Rehder K. Neuroanesthesia.
2. Graf CJ, Gross CE, Beck DW. Complications of spinal drainage
in the management of cerebrospinal fluid fistula. J Neurosurg.
3. Houle PJ, Vender JR, Fountas K, McDonnell D, Fick JR,
Robinson J. Pump-regulated lumbar subarachnoid drainage.
4. Haan J, Thomeer R. Predictive value of temporary external lumbar
drainage in normal pressure hydrocephalus. Neurosurgery. 1988;
5. Grady RE, Horlocker TT, Brown RD, Maxson PM, Schroeder DR.
Neurologic complications after placement of cerebrospinal fluid
drainage catheters and needles in anesthetized patients: implica-
tions for regional anesthesia. Anesth Analg. 1999;88:388–92.
6. Bloch J, Regli L. Brain stem and cerebellar dysfunction after
lumbar spinal fluid drainage: case report. J Neurol Neurosurg
7. Snow RS, Kuhel W, Martin SB. Prolonged lumbar drainage after
the resection of tumors of the skull base: a cautionary note.
8. Aidi S, Chaunu M, Biousse V, Bousser M. Changing pattern of
headache pointing to cerebral venous thrombosis after lumbar
puncture and intravenous high-dose steroids. Headache. 1999;39:
9. Albucher JF, Vuillemin-Azaı ¨s C, Manelfe C, Clanet M, Guiraud-
Chaumeil B, Chollet F. Cerebral thrombophlebitis in three
patients with probable multiple sclerosis. Cerebrovasc Dis.
10. Benzon HT, Iqbal M, Tallman MS, Boehlke L, Russell EJ.
Superior sagittal sinus thrombosis in a patient with postdural
puncture headache. Reg Anesth Pain Med. 2003;28:64–7.
11. Bienfait HP, Gijtenbeck JMM, van den Bent MJ, de Bruin HG,
Voogt PJ, Pillay M. Cerebral venous and sinus thrombosis with
cerebrospinal fluid circulation block after the first methotrexate
administration by lumbar puncture. Neuroradiology. 2002;44:
12. Ergan M, Hansen von Bu ¨neau F, Courthe ´oux P, Viader F,
Prouzeau S, Marcelli C. Thrombose veineuse ce ´re ´brale apre `s
injection intra-the ´cale de glucocorticoı ¨des. Revue du Rhuma-
tisme (Ed. francaise). 1997;64:595–8.
13. Gewirtz EC, Costin M, Marx GF. Cortical vein thrombosis may
14. Gunal DI, Afsar N, Tuncer N, Aktan S. A case of multiple
sclerosis with cerebral venous thrombosis: the role of lumbar
puncture and high-dose steroids. Eur Neurol. 2002;47:57–8.
15. Maurelli M, Bergamaschi R, Candeloro E, Todeschini A, Micieli
G. Cerebral venous thrombosis and demyelinating diseases:
report of a case in a clinically isolated syndrome suggestive of
multiple sclerosis onset and review of the literature. Mult Scler.
16. Milhaud D, Heroum C, Charif M, Saulnier P, Pages M, Blard JM.
Dural puncture and corticotherapy as risks factors for cerebral
venous sinus thrombosis. Eur J Neurol. 2000;7:123–4.
17. Mouraux A, Gille M, Dorban S, Peetters A. Cortical venous
thrombosis after lumbar puncture. J Neurol. 2002;249:1313–5.
18. Sta ¨dler C, Vuadens P, Dewarrat A, Janzer R, Uske A, Bogous-
slavsky J. Thrombose veineuse ce ´re ´brale apre `s ponction lombaire
86 Neurocrit Care (2010) 12:83–87
et ste ´roı ¨des intraveineux chez deux patients souffrant d’une Download full-text
scle ´rose en plaques. Rev Neurol (Paris). 2000;156:155–9.
19. Borum SE, Naul LG, McLeskey CH. Postpartum dural venous
sinus thrombosis after postdural puncture headache and epidural
blood patch. Anesthesiology. 1997;86:487–90.
20. Hubbert CH. Dural puncture headache suspected, cortical vein
thrombosis diagnosed. Anesth Analg. 1987;66:283–6.
21. Ravindran RS, Zandstra GC. Postpartum headache following
regional anesthesia; a symptom of cerebral venous thrombosis.
Can J Anaesth. 1989;36:705–7.
22. Wilder-Smith E, Kothbauer-Margreiter I, La ¨mmie B, Sturze-
negger M, Ozdoba C, Hauser SP. Dural puncture and activated
protein C resistance: risk factors for cerebral venous sinus
thrombosis. J Neurol Neurosurg Psychiatry. 1997;63:351–6.
23. Schou J, Scherb M. Postoperative sagittal sinus thrombosis after
spinal anesthesia. Anesth Analg. 1986;65:541–2.
24. Brugeilles H, Penisson-Besnier I, Pasco A, Oillic P, Lejeune P,
Mercier P. Cerebral venous thrombosis after myelography with
iopamidol. Neuroradiology. 1996;38:534–6.
following myelography: demonstration by magnetic resonance
imaging. Br J Radiol. 1992;65:1134–6.
26. Fishman RA. Cerebrospinal fluid in diseases of the nervous
system. 2nd ed. Philadelphia: W.B. Saunders Company; 1992.
27. Levine D, Rapalino O. The pathophysiology of lumbar puncture
headache. J Neurol Sci. 2001;192:1–8.
28. Canha ¨o P, Batista P, Falca ¨o F. Lumbar puncture and dural sinus
thrombosis—a causal or casual association? Cerebrovasc Dis.
29. Atkinson EA, Fairburn B, Heathfield KWG. Intracranial venous
thrombosis as a complication of oral contraception. Lancet. 1970;
30. BloemenkampKWM, Rosendaal
Enhancement by factor V Leiden mutation of risk of deep-vein
thrombosis associated with oral contraceptives containing a third
generation progestagen. Lancet. 1995;346:1593–6.
31. La Brocca A, Terzolo M, Pia A, Paccotti P, De Giuli P, Angeli A.
Recurrent thromboembolism as a hallmark of Cushing’s syn-
drome. J Endocrinol Invest. 1997;20:211–4.
32. Maxwell SRJ, Moots RJ, Kendall MJ. Corticosteroids: do they
damage the cardiovascular system? Postgrad Med J. 1994;70:
Neurocrit Care (2010) 12:83–8787