A preview of this full-text is provided by Springer Nature.
Content available from Neurosurgical Review
This content is subject to copyright. Terms and conditions apply.
ORIGINAL ARTICLE
Prognostic variables and outcome in relation to different bleeding
patterns in arteriovenous malformations
Nazife Dinc
1
&Sae-Yeon Won
1
&Johanna Quick-Weller
1
&Joachim Berkefeld
2
&Volker Seifert
1
&Gerhard Marquardt
1
Received: 15 January 2019 /Revised: 12 February 2019 /Accepted: 25 February 2019 / Published online: 5 March 2019
#Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract
Subarachnoid hemorrhage as bleeding pattern occurs rarely in ruptured arteriovenous malformations (AVM). The aim of the
present study is to evaluate different bleeding patterns in hemorrhages due to an AVM and their impact on outcome in terms of
risk and treatment stratification. We evaluated 158 patients with ruptured AVMs who were admitted to our neurosurgical
department from 2002 to 2017. We compared different bleeding patterns, such as intracerebral hemorrhage (ICH), subarachnoid
hemorrhage (SAH), or both (ICH + SAH) and evaluated predictive variables for outcome in the last follow-up. ICH was found in
48.7% of patients, isolated SAH in 20.9% of patients, and both in 30.4% of patients. The mean parenchymal blood volume was
34.5cm
3
. 38.6% of the patients had AVM-associated aneurysms,mostly located pre-nidal (77%). AVMs with ruptured aneurysms
often resulted in ICH with SAH component (p< 0.001) and SAH pattern occurred more often infratentorial (p= 0.003). In a
multivariate logistic regression model, poor clinical state on admission (WFNS IV-V) (p< 0.01), eloquence, infratentorial
location (p= 0.05), and presence of ICH with SAH component (p= 0.02) were the most relevant predictors of a poor outcome
after AVM rupture. Bleeding pattern in AVM hemorrhage depends on angioarchitectural and hemodynamic features and SAH
component predicts outcome negatively. AVM-associated pre-nidal aneurysms may harbor a higher risk for rupture and SAH
pattern when located infratentorial. Treatment decisions and risk stratifications should be considered in view of these findings.
Keywords Arteriovenous malformation .Subarachnoid hemorrhage .Hemorrhage pattern
Introduction
Outcome in hemorrhage due to an arteriovenous malforma-
tion (AVM) is infrequently devastating and depends on bleed-
ing pattern. While recent studies suggest that AVM hemor-
rhage is not as devastating as previously presumed compared
to aneurysmal subarachnoid hemorrhage (SAH), which is
more often lethal, AVM rupture tends to result in more neu-
rological deficits [1]. In addition to several negative prognos-
tic factors such as older age [2], poor initial admission state
[3], hematoma volume [4], and infratentorial location [5,6], a
subarachnoid component seems to impact outcome signifi-
cantly. Intraparenchymal hematoma is the most common
pattern in AVM-associated hemorrhage [7]. SAH due to an
AVM is described to amount 9% and literature therefore is
rare. The objective of our single-center study is to assess
how far subarachnoid component in AVM-associated hemor-
rhage impacts the final outcome negatively.
Materials and methods
Study design
The study was performed in accordance with our institutional
ethical review board. A retrospective review of patients with
ruptured brain arteriovenous malformation (AVM) who were
registered and evaluated at our neurosurgical department from
2002 to 2017 and entered into our prospectively maintained
database was carried out. The onset of hemorrhage was con-
sidered as the day at which the patient was admitted to our
emergency service. The diagnosis AVM was confirmed with
magnetic resonance imaging (MRI) or digital subtraction an-
giography (DSA). Different bleeding patterns such as
*Nazife Dinc
nazife.dinc@kgu.de
1
Department of Neurosurgery, Goethe University Hospital Frankfurt,
Schleusenweg 2-16, 60528 Frankfurt, Germany
2
Department of Neuroradiology, Goethe University Hospital,
Frankfurt, Germany
Neurosurgical Review (2019) 42:731–736
https://doi.org/10.1007/s10143-019-01091-7
Content courtesy of Springer Nature, terms of use apply. Rights reserved.