50
who underwent suboccipital or translabyrinthine
resection of a CPA tumor, 5 (4.6%) developed
transverse sinus thrombosis and symptomatic
papilledema [ 61 ]. Three patients were treated
with shunting and two were treated with acet-
azolamide and dexamethasone, and all patients’
symptoms resolved. No patients were treated
with anticoagulation. The number of patients
who had asymptomatic thrombosis was not
divulged. In a more recent series by Moore et al.
[ 62 ], 5 out of 43 patients (11.6%) who underwent
resection of acoustic neuroma developed trans-
verse or sigmoid sinus thrombosis. All who
developed thrombosis underwent tumor removal
via translabyrinthine approach. Four were asymp-
tomatic and diagnosed on routine postoperative
imaging, and the fifth was diagnosed 5 weeks
postoperative after developing ataxia and a cere-
bellar infarction. Those who developed thrombo-
sis had larger tumors (3.24 cm vs 1.8 cm,
P < 0.001) and underwent a longer operation
(10 h vs 8 h, P = 0.07), though surgical duration
fell just short of reaching significance. All were
treated with systemic anticoagulation for
6 months, and none developed intracranial hem-
orrhage. A Cochrane review of two small pro-
spective studies found that systemic
anticoagulation for venous sinus thrombosis had
a nonsignificant absolute reduction in the risk of
death and dependency of 13%, suggesting antico-
agulation in these patients is safe [ 63 ]. However,
some reserve systemic anticoagulation for
patients with clot propagation or symptom pro-
gression [ 58 , 59 , 61 ]. In those patients who are
clinically deteriorating despite systemic antico-
agulation, chemical or mechanical thrombec-
tomy are options, particularly when a preexisting
ICH is present [ 59 ].
Venous Thromboembolism
Patients undergoing craniotomy are known to be
at moderate to high risk for developing postop-
erative deep vein thrombosis (DVT) and pulmo-
nary embolism (PE). A meta-analysis by Kimmell
and Jahromi [ 64 ] of 4,844 patients who under-
went craniotomy found an incidence of venous
thromboembolism (VTE) of 3.5%, including
2.6% with DVT and 1.4% with PE. Significant
predictors of VTE on multivariate analysis in this
study included craniotomy for tumor, transfer
from acute care hospital, age ≥ 60 years, depen-
dent functional status, tumor involving the CNS,
sepsis, emergency surgery, surgical duration
≥ 4 h, postoperative urinary tract infection, post-
operative pneumonia, ≥ 48 h of postoperative
mechanical ventilation, and return to the operat-
ing room. Patients were assigned one point for
each of these factors for a cumulative VTE score
ranging from 0 to 12, with a higher score associ-
ated with a higher risk of VTE and mortality, as
well as a longer hospital stay. Specifically, those
with a score of ≥5 were 20 times more likely to
develop a VTE compared with patients with a
VTE score of 0. While chemical VTE prophy-
laxis is commonplace following most general
surgical and orthopedic procedures, its use still
remains infrequent by neurosurgeons and otolar-
yngologists who perform craniotomies.
Numerous prospective trials and meta-analyses
have failed to definitively demonstrate the safety
of heparin products following craniotomy [ 65 – 72 ].
One such meta-analysis by Hamilton et al. [ 73 ]
found an absolute risk reduction of symptomatic
and asymptomatic VTE of 9.1% with heparin
prophylaxis, compared to an absolute risk
increase of 0.7% for ICH and 2.8% for minor
bleeding. This translates to a likelihood of pre-
venting 13 symptomatic and asymptomatic VTE
for every ICH caused when using prophylactic
heparin. However, when broken down by symp-
tomatic or proximal VTE and PE, which are more
clinically relevant, this ratio changes to 5 to 1 for
proximal VTE and PE and 2 to 1 for symptomatic
VTE. Currently, the American College of Chest
Physicians only recommends the use of heparin
in craniotomy patients who are “very high risk,”
with a risk of VTE ≥10, specifically in patients
with a malignancy [ 74 ].Major Neurological ComplicationsAnsari et al. defined major neurological compli-
cations in their systematic review as arterial orJ.C. Sowder et al.