Specific Lobar Affection Reveals a Rostrocaudal Gradient in

Functional Outcome in Spontaneous

Intracerebral Hemorrhage
Stefan T. Gerner, MD; Joji B. Kuramatsu, MD; Sebastian Moeller, MD; Angelika Huber, MD;
Hannes Lücking, MD; Stephan P. Kloska, MD; Dominik Madžar, MD; Jochen A. Sembill, MD;
Stefan Schwab, MD; Hagen B. Huttner, MD

Background and Purpose—Several studies have reported a better functional outcome in lobar intracerebral hemorrhage
(ICH) compared with deep location. However, among lobar ICH, a correlation of hemorrhage site—involving the specific
lobes—with functional outcome has not been established.
Methods—Conservatively treated patients with supratentorial ICH, admitted to our hospital over a 5-year period (2008–
2012), were retrospectively analyzed. Lobar patients were classified as isolated or overlapping ICH according to affected
lobes. Demographic, clinical, and radiological characteristics were recorded and compared among lobar ICH patients
using above subclassification. Functional outcome—dichotomized into favorable (modified Rankin Scale, 0–3) and
unfavorable (modified Rankin Scale, 4–6)—was assessed after 3 and 12 months. Multivariate regression analysis was
performed to identify predictors for favorable outcome.
Results—Of overall 553 patients, 260 had lobar ICH. In isolated lobar ICH, median hematoma-volume decreased from
rostral (frontal, 22.4 mL [7.3–55.5 mL]) to caudal (occipital, 7.1 mL [5.2–16.4 mL]; P=0.045), whereas the proportion
of patients with favorable outcome increased (frontal: 23/63 [36.5%] versus occipital: 10/12 [83.3%]; P=0.003). Patients
with overlapping lobar ICH had larger ICH volumes than isolated lobar ICH (overlapping, 48.9 mL [22.6–78.5 mL] versus
15.3 mL [5.0–44.6 mL]; P<0.001) and poorer clinical status on admission (Glasgow Coma Scale and National Institutes
of Health Stroke Scale). Correlations with anatomic aspects provided evidence of a rostrocaudal gradient with increasing
gray/white-matter ratio and decreasing hematoma-volume and rate of hematoma enlargement from frontal to occipital
Downloaded from http://ahajournals.org by on November 7, 2023

ICH location. Multivariate analysis revealed affection of occipital lobe (odds ratio, 3.75 [1.38–10.22]) and affection of
frontal lobe (odds ratio, 0.52 [0.28–0.94]) to be independent predictors for favorable outcome and unfavorable outcome,
respectively.
Conclusions—Among patients with lobar ICH radiological and outcome characteristics differed according to location. Especially
affection of the frontal lobe was frequent and associated with unfavorable outcome after 3 months.   (Stroke. 2017;48:587-
595. DOI: 10.1161/STROKEAHA.116.015890.)
Key Words: cerebral hemorrhage ◼ frontal lobe ◼ hemorrhage ◼ neuroimaging ◼ prognosis

S pontaneous intracerebral hemorrhage (ICH) is associated

with substantial mortality and only one fourth of patients
manage to regain functional independence after 3 months.1–3
To better estimate the prognosis after ICH, several studies
investigated potential predictors for outcome. In addition to
clinical status, demographic parameters and initial hemor-
rhage volume and the location of hematoma is associated with
outcome.3–5 In supratentorial ICH, comparison of patients with
lobar and deep ICH revealed outcome benefits for patients
with lobar ICH.6–11 However, within lobar ICH, a specific
comparison on affected lobe and clinical outcome is lacking.
Given various lobes harboring different functional aspects,
involvement of different lobes by ICH is potentially linked to
outcome.9,11,12 This study investigated frequency and clinical
and radiological characteristics of specifically affected sites
among patients with lobar ICH to analyze location-specific
impact on functional outcome.
Methods
Patient Selection
Between 2008 and 2012, all consecutive patients with atraumatic ICH
admitted to the Department of Neurology of the university hospital

Received August 15, 2016; final revision received December 8, 2016; accepted December 28, 2016.
From the Department of Neurology (S.T.G., J.B.K., S.M., A.H., D.M., J.A.S., S.S., H.B.H.) and Department of Neuroradiology (H.L., S.P.K.), University
Hospital Erlangen, Germany.
Guest Editor for this article was Gregory W. Albers, MD.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.
116.015890/-/DC1.
Correspondence to Stefan T. Gerner, MD, Department of Neurology, University Hospital Erlangen, Schwabachanlage 6, 91054 Erlangen, Germany.
E-mail stefan.gerner@uk-erlangen.de
© 2017 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.116.015890

587
 588  Stroke  March 2017
of Erlangen, Germany, were retrospectively analyzed. A total of 627
patients with supratentorial ICH who did not undergo surgery for
hematoma evacuation were reviewed. ICH caused by the secondary
cause (n=74), that is, hemorrhagic transformation of ischemic stroke,
intracranial aneurysm, tumor, cavernoma, or arteriovenous malfor-
mation were excluded.
Data Collection
Demographic parameters (age and sex), the preexisting condition
(modified Rankin Scale [mRS] before ICH, history of stroke, or car-
diovascular risk factors), intake of antithrombotic medication (oral
anticoagulation and antiplatelet drugs), clinical status on admission
(National Institutes of Health Stroke Scale [NIHSS],13 Glasgow
Coma Scale [GCS],14 and ICH-score4), and in-hospital measures
(need for mechanical ventilation/external ventricular drainage, length
of stay, occurrence of pneumonia/sepsis, and in-hospital mortality)
were extracted from institutional electronic databases and medical
charts. Laboratory data on admission (count of leucocytes/thrombo-
cytes, hemoglobin, and international normalized ratio) were provided
by an institutional laboratory database.
Imaging
Diagnosis of ICH was made either by multislice computed tomog-
raphy (Siemens SOMATOM Volume Zoom or Siemens Definition
AS+) or 1.5 Tesla MRI (Siemens MAGNETOM Sonata or Siemens
MAGNETOM Aera). Two neuroradiologists (H.L. and S.P.K.)
blinded to clinical data classified ICH location and assessed hema-
toma characteristics. Specifically, we defined deep ICH as hema-
toma located in the basal ganglia, internal or external capsula, or
thalamus. Lobar ICH was defined as ICH originating at the cor-
tex and cortical–subcortical junction. In case of deep ICH extend-
ing into the lobes, the adjudicated consensus categorization of the
neuroradiologists was scored (usually categorized according to
Downloaded from http://ahajournals.org by on November 7, 2023
the location the ICH most likely originated).15 The neuroradiolo-
gists further subclassified lobar ICH according to anatomic borders
(frontal lobe: sulcus centralis, and sulcus lateralis; parietal lobe:
sulcus centralis, sulcus parietooccipitalis, and sulcus lateralis; tem-
poral lobe: sulcus lateralis; occipital lobe: sulcus parietooccipita-
lis and tentorium cerebelli). Isolated lobar ICH was scored if the
hematoma respected the anatomic borders of the affected lobe. If
>1 lobe was affected, overlapping lobar ICH was classified accord-
ing to a consensus decision of the 2 lobes mostly affected, that id,
frontotemporal, frontoparietal, temporoparietal, temporooccipital,
and parietooccipital. Moreover, lobar ICH was reassigned to 1 of
the 3 categories according to its location on a rostrocaudal axis.
Further neuroradiological analyses included correlation to anatomic
aspects, such as volume and gray/white-matter (GW) ratio of each
brain lobe affected.16
Hematoma volume was assessed by the ABC/2 formula17 and
adjusted to imaging modality.18 The occurrence of intraventricular
blood was documented, and its extent was measured by assessing the
Graeb score.19 According to an institutional protocol, follow-up imag-
ing was routinely performed 24±6 hours after admission. Hematoma
growth was documented if a volume increase of >33% compared with
the initial ICH volume was observed.20
Outcome
Functional outcome was assessed by physicians certified and trained
in outcome assessment, 90 days and 1 year after index ICH using the
mRS.21 Patients were followed up either as outpatient visits (n=44)
or contacted by either mailed questionnaires (n=374) or, if no reply
was received within 4 weeks, semistructured telephone interviews
(n=114). Outcome was dichotomized into favorable, defined as an
mRS score of 0 to 3, and unfavorable outcomes, defined as an mRS
score between 4 and 6. As shown in Table I in the online-only Data
Supplement, there were no significant differences in outcome mea-
sures among patients with either of above ways of follow-up assess-
ment. Sixteen patients were lost to follow-up.
Statistical Analysis
Statistical analyses were performed by using SPSS 21 (IBM
SPSS Statistics 21, http://www.spss.com) and R 2.12.0 (http://
www.r-project.org). Significance level was set at 2-sided α=0.05.
Comparison of dichotomized data, presented as total number and
frequency in brackets, was undertaken by Pearson χ2 and either
Fisher exact test for 2 samples or Fisher–Freeman–Halton test
for >2 samples. Normal distribution of continuous variables was
excluded by the Kolmogorov–Smirnov test. Data were presented
as median and interquartile range and compared using the Mann–
Whitney U test (if 2 samples) and the Kruskal–Wallis test (if >2
samples). Stepwise multivariate regression analysis was calcu-
lated to identify predictors for favorable outcome (mRS, 0–3) after
90 days and contained parameters that reached a statistical trend
(P=0.10) in previous univariate testing. Propensity score matching
(caliper 0.1, ratio 1:1, nearest neighbor approach20) was performed
to compare outcome among patients with lobar and deep ICH,
adjusted for differences in baseline parameters.
Results
Overall Cohort
Altogether, 553 patients with atraumatic supratentorial ICH—
293 (53%) with main deep and 260 (47%) with main lobar loca-
tion—remained for final analysis. Patients with lobar ICH were
older (74 years [66–81] versus 70 years [60–78]; P=0.001),
had a worse medical condition before ICH onset (ie, premorbid
mRS), and a better clinical status on admission (NIHSS and
GCS) compared with deep ICH (Table 1). Despite larger ICH
volumes in lobar patients (main lobar, 27.0 mL [9.0–59.9 mL]
versus main deep 11.9 mL [4.7–32.1 mL]; P<0.001), there were
no outcome differences among both locations in this unadjusted
analysis (mRS at 3 months, lobar 4 [3–6] versus deep 4 [3–6];
P=0.305; mRS at 1 year, lobar 4 [2–6] versus deep 4 [3–6];
P=0.621). In-hospital mortality was 23% in both groups, and
there were no differences in the proportion of patients with
DNT/DNR orders (do-not-treat/do-not-resuscitate; Table 1). To
specifically compare outcome among patients with deep versus
lobar ICH, we adjusted for above baseline confounders using
a propensity score–matched analysis (after propensity score
matching baseline parameters were evenly balanced among
patients with deep and lobar ICH; Table II in the online-only
Data Supplement). There were significant differences in unfavor
of deep ICH both at 3 and 12 months (mRS at 3 months=0–3:
deep 46/190 [24.2%] versus lobar 78/183 [42.6%]; P<0.001;
Figure 1; Table III in the online-only Data Supplement).
Frequency of Various Lobar ICH Locations
Focusing on lobar ICH patients only, 139 (53.5%) patients
had ICH, which was solely allocated in 1 lobe, referred to as
isolated lobar ICH, and 121 (46.5%) patients had overlapping
ICH affecting >1 lobe. The frequency of distribution of the
various isolated and overlapping lobar ICH locations and their
volumes is presented in Figure 2.
Isolated Lobar ICH
Among patients with isolated lobar ICH, the frontal lobe was
most frequently affected (67/260; 25.8%), whereas isolated
occipital ICH was relatively rare (12/260;4.6%; P<0.001;
Figure 2; Table 2). There was a decrease in average hematoma
volume after the rostrocaudal axis from frontal to occipital
 Gerner et al   Outcome of Specific Lobar-Located ICH    589

Table 1. Comparison of Patients With Main Deep Versus Main Lobar ICH
Overall Patients With Supratentorial ICH (n=553) Main Deep (n=293) Main Lobar (n=260) P Value
Age, y* 70 (60–78)† 74 (66–81) 0.001
Female sex‡ 127 (43.3%) 130 (50.0%) 0.117
Previous comorbidities
 Premorbid mRS* 0 (0–1)† 1 (0–2)† 0.026†
 Hypertension‡ 248 (85.2%)† 194 (75.8%)† 0.005†
 Diabetes mellitus‡ 73 (25.2%) 61 (23.8%) 0.716
 Dyslipidemia‡ 83 (28.6%) 83 (32.4%) 0.335
 Coronary artery disease‡ 19 (7.0%) 29 (11.9%) 0.054
 Atrial fibrillation‡ 54 (18.9%) 38 (14.7%) 0.191
 Previous ischemic stroke‡ 56 (19.4%) 35 (13.8%) 0.085
 Previous hemorrhagic stroke‡ 11 (3.8%)† 35 (13.8%)† <0.001†
 Antiplatelet medication‡ 77 (26.9%) 83 (33.1%) 0.120
 Oral anticoagulation‡ 18 (20.7%) 16 (17.6%) 0.598
Admission status
 Glasgow Coma Scale* 13 (5–15)† 13 (6–15)† 0.010†
 NIHSS* 17 (8–28)† 12 (4–24)† <0.001†
 ICH score* 1 (1–3) 2 (0–3) 0.450
Imaging characteristics
 Initial ICH volume, mL* 11.9 (4.7–32.1)† 27.0 (9.0–59.9)† <0.001†
 Initial intraventricular hemorrhage‡ 189 (64.7%)† 100 (38.8%)† <0.001†
  Initial Graeb score* 2 (0–6)† 0 (0–3)† 0.001†
 Left hemispheric hemorrhage‡ 148 (50.5%) 149 (57.3%) 0.110
Downloaded from http://ahajournals.org by on November 7, 2023

 HE on follow-up CT‡ 44 (17.6%) 24 (11.6%) 0.073

Initial laboratory values
 INR* 1.03 (0.99–1.13) 1.03 (0.98–1.11) 0.510
 Hemoglobin, mmol/L* 8.5 (7.8–9.3)† 8.3 (7.5–9.0)† 0.019†
 Thrombocytes, 103/μL* 220 (179–266) 227 (185–268) 0.133
 Leukocytes, 10 /μL*
3
8.5 (6.6–11.3) 9.2 (7.2–11.5) 0.110
Clinical parameter during hospital stay
 Length of stay, d* 11 (6–19)† 9 (5–14)† 0.006†
 Mechanical ventilation‡ 147 (50.2%)† 73 (28.1%)† <0.001†
 External ventricular drain‡ 108 (36.9%)† 36 (13.8%)† <0.001†
 Sepsis‡ 27 (10.2%)† 11 (4.9%)† 0.030†
 Pneumonia‡ 121 (45.8%)† 80 (35.9%)† 0.026†
 DNT/DNR orders 34 (11.6%) 40 (15.4%) 0.192
Outcome
 In-hospital mortality‡ 67 (22.9%) 60 (23.1%) 0.953
 mRS at 3 mo* 4 (3–6) 4 (3–6) 0.305
  mRS 0–3‡ 86 (30.1%) 89 (36.2%) 0.135
  mRS 4–6‡ 200 (69.9%) 157 (63.8%) 0.135
 mRS at 12 mo* 4 (3–6) 4 (2–6) 0.621
  mRS 0–3‡ 103 (37.2%) 95 (38.8%) 0.708
  mRS 4–6‡ 174 (62.8%) 150 (61.2%) 0.708
CT indicates computed tomography; DNT/DNR, do-not-treat/do-not-resuscitate; HE, hemorrhage enlargement defined as volume increase
>33% on follow-up imaging; ICH, intracerebral hemorrhage; INR, international normalized ratio; mRS, modified Rankin Scale; and NIHSS,
National Institutes of Health Stroke Scale.
*Median (interquartile range).
†Significant differences.
‡No. (valid percentage).
 590  Stroke  March 2017
(ICH volume: frontal, 22.4 mL [7.3–55.5 mL] versus occipi-
tal, 7.1 mL [5.2–16.4 mL]; P=0.039) and correspondingly
a significant increase in median NIHSS and GCS levels on
admission (GCS: frontal, 13 [8–15] versus occipital 15 [15–
15]; P=0.005; NIHSS: frontal, 10 [4–22] versus occipital,
2 [1–4]; P<0.001). Demographic and previous medical his-
tory characteristics were not significantly different (Table 2).
Functional outcome of patients with lobar ICH patients is
shown in Figure 3A. After 3 and 12 months, respectively, there
was a significant difference in the proportion of patients with
favorable outcome when comparing frontal versus occipital
Downloaded from http://ahajournals.org by on November 7, 2023
ICH (mRS at 3 months=0–3: isolated occipital 10/12 [83.3%]
versus isolated frontal 23/63 [36.5%]; P=0.003; mRS at 1
year=0–3: isolated occipital 10/12 [83.3%] versus isolated
frontal 27/63 [42.9%]; P=0.010).
Overlapping Lobar ICH
Figure 2 and Table IV in the online-only Data Supplement
show clinical and radiological characteristics of patients
with overlapping lobar ICH, that is, frontotemporal (n=9;
3.5%), frontoparietal (n=46; 17.7%), temporoparietal (n=40;
Figure 1. Distribution of modified Rankin
Scale (mRS) comparing main lobar and
main deep intracerebral hemorrhage (ICH).
Presented are the mRS after 3 and 2 mo
comparing main lobar and main deep ICH
after propensity score matching (matching
parameter: age, premorbid mRS, arterial
hypertension, history of ICH, and initial
ICH volume). The bold line separates
favorable (mRS, 0–3) and unfavorable out-
come (mRS, 4–6).
15.4%), parietooccipital (n=23; 8.8%), and temporooccipi-
tal (n=3;1.2%). Median ICH volume ranged from 10.0 mL
in temporooccipital ICH to 91.8 mL in frontoparietal ICH.
On admission, patients had a poor clinical status, especially
if hemorrhage was in frontoparietal (GCS, 4 [3–11]; NIHSS,
30 [17–32]) or temporooccipital location (GCS, 5 [3–15];
NIHSS, 26[1–32]). Figure 3B shows outcome data of overlap-
ping ICH revealing significantly higher rates of unfavorable
outcome in frontoparietal than in temporoparietal ICH (mRS
at 1 year=4–6: frontoparietal 35/43 [81.4%] versus temporo-
parietal 24/39 [61.5%]; P=0.046; Figure 3B).
Rostrocaudal Gradient in Lobar ICH
Given above unadjusted data revealing worse favorable out-
comes alongside the rostrocaudal axis from frontal toward
occipital ICH, we correlated these findings with anatomic
aspects of the human brain. As shown previously, there are
differences both in volumes and GW ratios among the lobes
of the human brain, such that the frontal lobe has relatively the
largest size, whereas the GW ratio is highest in temporal and
occipital lobes.16 In a first step, we investigated the frequency
Figure 2. Illustration of frequency and
initial intracerebral hemorrhage (ICH)
volume in lobar ICH classified according
to affected lobes. This figure illustrates
the brain segmented into its 4 lobes. For
better understanding, isolated ICH was
located within the affected lobe, over-
lapping ICH at the border between the
2 lobes that were mainly affected. The
diameter of red circles indicating ICH
correlates with the initial hematoma vol-
ume. F indicates frequency; and V, initial
median ICH volume.
 Gerner et al   Outcome of Specific Lobar-Located ICH    591

Table 2. Admission Status, Demographic, Neuroradiological, and Treatment Characteristics of Patients With Isolated Lobar ICH
Patients With Isolated Lobar ICH Isolated Frontal Isolated Parietal Isolated Temporal Isolated Occipital
(n=139) (n=67) (n=37) (n=23) (n=12) P Value
Age, y* 71 (60–77) 74 (67–81) 74 (66–79) 71 (68–78) 0.417
Female sex† 36 (53.7%) 18 (48.6%) 10 (43.5%) 5 (41.7%) 0.772
Previous comorbidities
 Premorbid mRS* 0 (0–2) 1 (0–2) 1 (0–2) 0.5 (0–2) 0.466
 Hypertension† 46 (68.7%) 28 (75.7%) 17 (73.9%) 8 (66.7%) 0.871
 Diabetes mellitus† 11 (16.4%) 11 (29.7%) 6 (26.1%) 3 (25.0%) 0.384
 Dyslipidemia† 22 (32.8%) 10 (27.0%) 8 (34.8%) 5 (50.0%) 0.532
 Coronary artery disease† 5 (8.1%) 1 (2.8%) 3 (13.0%) 1 (8.3%) 0.475
 Atrial fibrillation† 10 (14.9%) 6 (16.2%) 6 (26.1%) 1 (8.3%) 0.579
 Previous ischemic stroke† 13 (19.7%) 6 (16.2%) 3 (13.0%) 1 (8.3%) 0.855
 Previous hemorrhagic stroke† 7 (10.6%) 6 (16.2%) 2 (8.7%) 4 (33.3%) 0.178
 Antiplatelet medication† 24 (36.4%) 11 (29.7%) 7 (31.8%) 3 (25.0%) 0.869
 Oral anticoagulation† 3 (11.5%) 1 (9.1%) 3 (30.0%) 0 (0%) 0.497
Admission status
 Glasgow Coma Scale* 13 (8–15)‡ 14 (12–15)‡ 15 (12–15)‡ 15 (15–15)‡ 0.017‡
 NIHSS* 10 (4–22)‡ 7 (4–17)‡ 8 (2–14)‡ 2 (1–4)‡ 0.003‡
 ICH score* 1 (1–3)‡ 1 (0–2)‡ 0 (0–2)‡ 0 (0–0)‡ 0.011‡
Imaging characteristics
 Initial ICH-volume, m* 22.4 (7.3–55.5)‡ 15.0 (4.2–48.8)‡ 9.4 (4.0–26.2)‡ 7.1 (5.2–16.4)‡ 0.045‡
 Initial intraventricular hemorrhage† 23 (34.8%) 12 (32.4%) 5 (21.7%) 2 (16.7%) 0.527
Downloaded from http://ahajournals.org by on November 7, 2023

  Initial Graeb score* 0 (0–4) 0 (0–2) 0 (0–0) 0 (0–0) 0.357

 Left-hemispheric hemorrhage† 34 (50.7%) 21 (56.8%) 12 (52.2%) 8 (66.7%) 0.771
 HE on follow-up CT† 7 (12.7%) 2 (6.2%) 2 (9.1%) 0 (0%) 0.701
Initial laboratory values
 INR* 1.01 (0.99–1.10) 1.04 (1.00–1.12) 1.07 (0.99–1.20) 1.02 (0.95–1.13) 0.540
 Hemoglobin, mmol/L* 8.2 (7.5–9.1) 8.2 (7.8–8.9) 8.8 (8.0–9.5) 8.5 (8.3–9.0) 0.200
 Thrombocytes, 10 /μL* 3
220 (173–264) 234 (198–262) 224 (168–302) 266 (202–309) 0.465
 Leukocytes, 10 /μL*
3
9.2 (7.1–11.5) 8.9 (6.9–10.4) 8.8 (6.8–11.6) 9.5 (7.1–11.5) 0.819
Clinical parameter during hospital stay
 Length of stay, d* 10 (6–15) 10 (7–13) 11 (7–17) 7 (5–10) 0.059
 Mechanical ventilation† 20 (29.9%) 8 (21.6%) 4 (17.4%) 0 (0%) 0.113
 External ventricular drain† 12 (17.9%) 4 (10.8%) 2 (8.7%) 0 (0%) 0.380
 Sepsis† 1 (1.7%) 1 (3.0%) 0 (0%) 1 (8.3%) 0.383
 Pneumonia† 23 (38.3%) 8 (24.2%) 8 (34.8%) 1 (8.3%) 0.155
 In-hospital mortality† 12 (17.9%) 6 (16.2%) 1 (4.3%) 0 (0%) 0.206
CT indicates computed tomography; HE, hemorrhage enlargement defined as volume increase >33% on follow-up imaging; ICH, intracerebral hemorrhage; INR,
international normalized ratio; mRS, modified Rankin Scale; and NIHSS, National Institutes of Health Stroke Scale.
*Median (interquartile range).
†No. (valid percentage).
‡Significant differences.

of occurrence of isolated lobar ICH in relation to the size of of isolated occipital ICH was lower than expected (in fron-
the various lobes affected (Table V in the online-only Data tal ICH: expected n=56, observed n=67; in occipital ICH:
Supplement). This analysis revealed that the prevalence of expected n=13, observed n=12; see Table V in the online-only
isolated frontal ICH was higher, whereas the prevalence Data Supplement for formula of calculation).
 592  Stroke  March 2017
Figure 3. Distribution of short (3 mo)- and long (1 y)-term functional outcome of patients with main lobar intracerebral hemorrhage (ICH).
A, Isolated lobar ICH. B, Overlapping lobar ICH. mRS indicates modified Rankin Scale; score 0 to 6: from no deficit to death. A favorable
functional outcome (mRS, 0–3 vs mRS, 4–6) is shown as a bold line. P values are provided for comparison of different locations. N.S. indi-
cates not significant.
Downloaded from http://ahajournals.org by on November 7, 2023
In a next step, we correlated to volume of lobar ICH in
relation to both the volumes of the lobes affected and the
GW ratio (Figure 4). This analysis revealed that absolute
ICH volumes were largest if the frontal lobe was affected
(30.6 mL), decreasing to 26.8 mL in temporoparietal affec-
tion and to 19.4 mL in occipital affection (Figure 4). This
rostrocaudal gradient for absolute ICH volume was inversed
on a relative ICH volume analysis. Given these associations,
we studied the impact of absolute versus relative ICH vol-
ume on functional outcome (Figure I in the online-only Data
Supplement), which revealed a stronger association of abso-
lute than relative ICH volume with functional outcome at
3 months (area under the curve [95% confidence interval]:
absolute ICH volume, 0.752 [0.667–0.836]; P<0.001 ver-
sus relative ICH-volume, 0.706 [0.618–0.795]; P<0.001). In
addition, we correlated the absolute ICH volume to the GW
ratios of the various lobes affected. This analysis revealed
an inversely proportional association between GW ratio and
absolute ICH volume, that is, decreasing GW ratio was asso-
ciated with increasing ICH volume (Figure 4). The correla-
tion of clinical aspects with above radiological parameters
provided evidence that GCS levels were increasing from
frontal to occipital location, linked to decreasing absolute
ICH volume (Figure 4). With regard to increasing GW ratios
from frontal to occipital location, the frequency of hemor-
rhage growth declined (frontal, 15 [16.0%] versus occipital
1 [3.3%]) and the proportion of patients with unfavorable
outcome decreased (frontal, 89 [76.1%] versus occipital 14
[41.2%]; Figure 4).
Predictors for Functional Independence
For prediction of functional independence (mRS, 0–3) after 3
months in patients with ICH multivariate logistic regression
analysis was undertaken. The affection of the occipital lobe
was—both in lobar and deep ICH—along with initial GCS
a significant positive independent predictor for favorable
outcome after 3 months (odds ratio, 3.751 [1.376–10.221];
P=0.010). Analogously, affection of the frontal lobe turned
out to be the second strongest negative predictor of favor-
able outcome next to intraventricular hemorrhage (affection
of frontal lobe odds ratio, 0.516 [0.283–0.942]; P=0.031;
intraventricular hemorrhage odds ratio, 0.394 [0.251–0.617];
P<0.001; Table VI in the online-only Data Supplement).
Discussion
The present study for the first time systematically investigated
the ICH location–specific impact on functional outcome in
lobar ICH. As a key finding, we demonstrated that among
patients with lobar affection, there is a rostrocaudal gradient
with largest and clinically most severe ICH in frontal location
and clinically best prognosis in occipital ICH. Several impli-
cations need attention.
Why do frontal bleedings show worse outcome measures?
The aspect of rostrocaudally decreasing absolute hematoma
volumes, with largest ICH volumes in frontal and smallest in
occipital location, may account for the observed outcome dif-
ferences. Several studies have already proven hematoma size
to be the main predictor of mortality and outcome in ICH.3,4,22
A possible explanation for larger hemorrhage volumes in more
 Gerner et al   Outcome of Specific Lobar-Located ICH    593
Downloaded from http://ahajournals.org by on November 7, 2023

Figure 4. Rostrocaudal classification of lobar intracerebral hemorrhage (ICH) and comparison of clinical and radiological characteristics.
Lobar ICH was classified according to its location on a rostrocaudal axis (frontal location: isolated frontal, overlapping frontotemporal,
and overlapping frontoparietal lobar ICH; temporoparietal location: isolated parietal, isolated temporal, and overlapping temporoparietal
lobar ICH; occipital location: isolated occipital, overlapping temporooccipital, and overlapping parietooccipital lobar ICH). Clinical and
radiological characteristics are presented for each location and observed gradients in these parameters exemplified as blue triangles.
Anatomic characteristics of the affected lobe(s)16 (lobe volume, gray/white-matter [GW] ratio) were assessed separately according to sex
and affected site in each single patient. Relative ICH volume was defined as absolute ICH volume related to the volume of the affected
lobe(s). GCS indicates Glasgow Coma Scale; NIHSS, National Institutes of Health Stroke Scale; Potentially affected lobe volume, volume
of affected lobe(s); and Unfavorable outcome, mRS [3 m]=4 to 6. *No. (valid percentage); †median (interquartile range).

frontal ICH location might be the corresponding arterial pres-
sure gradient; transcranial ultrasound studies suggested higher
peak and mean velocities in the anterior circulation compared
with the posterior circulation that may be related—in case
of vessel rupture and consequent ICH—to larger final hema-
toma volumes in rather rostral locations.23–26 Furthermore,
we observed a higher—but not significantly increased—rate
of hematoma enlargement in frontal compared with occipital
ICH. Interestingly, both aspects—that is, larger volumes and
more frequent ICH growth—may be correlated to anatomic
aspects, such as GW ratio.16 Although the frontal lobe is the
largest and thus statistically may be affected more often than
other lobes, its GW ratio is lowest indicating a relatively higher
proportion of cortex in the occipital lobe.16 Hence, future stud-
ies on ICH location–specific outcome differences should corre-
late findings with radiological and histoanatomic findings.12,27
 594  Stroke  March 2017
Another explanation for the reported outcome differences
among isolated lobar ICH patients refers to assessment of
outcome measures applied in this study. As the mRS score
represents deficits in motor function in a more pronounced
way than other neurological deficits,21 other outcome mea-
sures, such as cognitive or behavioral scores, may have
reflected clinical impairments more adequately.28,29 Hence,
the over-representation of frontal motorcortical involve-
ment within the mRS—while under-representing for
instance occipitally located visual impairments—may also
have played a role in the different observed outcome.30,31
Recent publications managed to develop a functional map
of the brain by multimodal magnetic resonance imag-
ing12,32 and consequently improved our understanding of
functional brain topography that was based on anatomic
studies before.27 Hence, on a more general note, studying
ICH location–specific outcome differences in lobar bleed-
ings should incorporate dedicated lesion-specific outcome
assessments to compare and weigh long-term functional
deficits in these patients.
Besides above-mentioned shortcomings, there are some
additional limitations; however, there are also strengths in
the present study. Contrary to findings of Izumihara et al33
who reported adverse postoperative outcome in parietal lobe
affection, our findings were based on conservatively treated
patients only without any influence of surgical approaches on
outcome. In this regard, one has to keep in mind that vari-
ous previous studies of ICH used different definitions of deep
and lobar ICH.10,11,15 Nonetheless, the retrospective and single-
Downloaded from http://ahajournals.org by on November 7, 2023
center design as well as the small sample size in each group
undermined further statistical adjusting to minimize volume-
related effects on outcome. Furthermore, we may have missed
specific ICH causes that were unmasked on follow-up imag-
ing after hospital discharge as these imaging findings were
not assessed. Moreover, we only obtained functional outcome
data using the mRS score and did not assess other scores on
quality of life or cognition (eg, FUNC score3), which would be
valuable to obtain in future studies.
In conclusion, our findings indicate that among lobar ICH,
there is a rostrocaudal gradient with more severe ICH occur-
ring in rather frontally located bleedings. ICH volume and
anatomic-functional aspects may account for these findings.
Acknowledgments
We thank Dr Petra Burkardt (University Hospital of Erlangen,
Germany) for helping with data management.
Disclosures
None.
References
1. González-Pérez A, Gaist D, Wallander MA, McFeat G, García-Rodríguez
LA. Mortality after hemorrhagic stroke: data from general practice (The
Health Improvement Network). Neurology. 2013;81:559–565. doi:
10.1212/WNL.0b013e31829e6eff.
2. van Asch CJ, Luitse MJ, Rinkel GJ, van der Tweel I, Algra A, Klijn CJ.
Incidence, case fatality, and functional outcome of intracerebral haemor-
rhage over time, according to age, sex, and ethnic origin: a systematic
review and meta-analysis. Lancet Neurol. 2010;9:167–176. doi: 10.1016/
S1474-4422(09)70340-0.
3. Rost NS, Smith EE, Chang Y, Snider RW, Chanderraj R, Schwab K, et al.
Prediction of functional outcome in patients with primary intracerebral
hemorrhage: the FUNC score. Stroke. 2008;39:2304–2309. doi: 10.1161/
STROKEAHA.107.512202.
4. Hemphill JC 3rd, Bonovich DC, Besmertis L, Manley GT, Johnston SC.
The ICH score: a simple, reliable grading scale for intracerebral hemor-
rhage. Stroke. 2001;32:891–897.
5. Kuramatsu JB, Sauer R, Mauer C, Lücking H, Kloska SP, Kiphuth IC,
et al. Correlation of age and haematoma volume in patients with sponta-
neous lobar intracerebral haemorrhage. J Neurol Neurosurg Psychiatry.
2011;82:144–149. doi: 10.1136/jnnp.2010.208439.
6. Castellanos M, Leira R, Tejada J, Gil-Peralta A, Dávalos A, Castillo
J; Stroke Project, Cerebrovascular Diseases Group of the Spanish
Neurological Society. Predictors of good outcome in medium to large spon-
taneous supratentorial intracerebral haemorrhages. J Neurol Neurosurg
Psychiatry. 2005;76:691–695. doi: 10.1136/jnnp.2004.044347.
7. Hårdemark HG, Wesslén N, Persson L. Influence of clinical factors, CT
findings and early management on outcome in supratentorial intracere-
bral hemorrhage. Cerebrovasc Dis. 1999;9:10–21.
8. Juvela S. Risk factors for impaired outcome after spontaneous intracere-
bral hemorrhage. Arch Neurol. 1995;52:1193–1200.
9. Lee JY, King C, Stradling D, Warren M, Nguyen D, Lee J, et al. Influence
of hematoma location on acute mortality after intracerebral hemorrhage. J
Neuroimaging. 2014;24:131–136. doi: 10.1111/j.1552-6569.2012.00766.x.
10. Samarasekera N, Fonville A, Lerpiniere C, Farrall AJ, Wardlaw
JM, White PM, et al; Lothian Audit of the Treatment of Cerebral
Haemorrhage Collaborators. Influence of intracerebral hemorrhage loca-
tion on incidence, characteristics, and outcome: population-based study.
Stroke. 2015;46:361–368. doi: 10.1161/STROKEAHA.114.007953.
11. Sreekrishnan A, Dearborn JL, Greer DM, Shi FD, Hwang DY, Leasure
AC, et al. Intracerebral hemorrhage location and functional outcomes
of patients: a systematic literature review and meta-analysis. Neurocrit
Care. 2016;25:384–391. doi: 10.1007/s12028-016-0276-4.
12. Glasser MF, Coalson TS, Robinson EC, Hacker CD, Harwell J, Yacoub
E, et al. A multi-modal parcellation of human cerebral cortex. Nature.
2016;536:171–178. doi: 10.1038/nature18933.
13. Berger K, Weltermann B, Kolominsky-Rabas P, Meves S, Heuschmann
P, Böhner J, et al. The reliability of stroke scales. The German version of
NIHSS, ESS and Rankin scales. Fortschr Neurol Psychiatr. 1999;67:81–
93. doi: 10.1055/s-2007-993985.
14. Teasdale G, Jennett B. Assessment of coma and impaired consciousness.
A practical scale. Lancet. 1974;2:81–84.
15. Falcone GJ, Biffi A, Brouwers HB, Anderson CD, Battey TW, Ayres
AM, et al. Predictors of hematoma volume in deep and lobar supraten-
torial intracerebral hemorrhage. JAMA Neurol. 2013;70:988–994. doi:
10.1001/jamaneurol.2013.98.
16. Allen JS, Damasio H, Grabowski TJ, Bruss J, Zhang W. Sexual dimor-
phism and asymmetries in the gray-white composition of the human
cerebrum. Neuroimage. 2003;18:880–894.
17. Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR,
Zuccarello M, et al. The ABCs of measuring intracerebral hemorrhage
volumes. Stroke. 1996;27:1304–1305.
18. Burgess RE, Warach S, Schaewe TJ, Copenhaver BR, Alger JR, Vespa
P, et al. Development and validation of a simple conversion model for
comparison of intracerebral hemorrhage volumes measured on CT and
gradient recalled echo MRI. Stroke. 2008;39:2017–2020. doi: 10.1161/
STROKEAHA.107.505719.
19. Graeb DA, Robertson WD, Lapointe JS, Nugent RA, Harrison PB.
Computed tomographic diagnosis of intraventricular hemorrhage.
Etiology and prognosis. Radiology. 1982;143:91–96. doi: 10.1148/
radiology.143.1.6977795.
20. Kuramatsu JB, Gerner ST, Schellinger PD, Glahn J, Endres M, Sobesky
J, et al. Anticoagulant reversal, blood pressure levels, and anticoagulant
resumption in patients with anticoagulation-related intracerebral hemor-
rhage. JAMA. 2015;313:824–836. doi: 10.1001/jama.2015.0846.
21. Banks JL, Marotta CA. Outcomes validity and reliability of the modi-
fied Rankin scale: implications for stroke clinical trials: a literature
review and synthesis. Stroke. 2007;38:1091–1096. doi: 10.1161/01.
STR.0000258355.23810.c6.
22. Broderick JP, Brott TG, Duldner JE, Tomsick T, Huster G. Volume of
intracerebral hemorrhage. A powerful and easy-to-use predictor of
30-day mortality. Stroke. 1993;24:987–993.
23. Kaps M, Seidel G, Bauer T, Behrmann B. Imaging of the intracra-
nial vertebrobasilar system using color-coded ultrasound. Stroke.
1992;23:1577–1582.
 Gerner et al   Outcome of Specific Lobar-Located ICH    595
24. Martin PJ, Evans DH, Naylor AR. Transcranial color-coded sonography
of the basal cerebral circulation. Reference data from 115 volunteers.
Stroke. 1994;25:390–396.
25. Martin PJ, Evans DH, Naylor AR. Measurement of blood flow velocity
in the basal cerebral circulation: advantages of transcranial color-coded
sonography over conventional transcranial Doppler. J Clin Ultrasound.
1995;23:21–26.
26. Schöning M, Walter J. Evaluation of the vertebrobasilar-posterior
system by transcranial color duplex sonography in adults. Stroke.
1992;23:1280–1286.
27. Brodmann K. Vergleichende Lokalisationslehre der Grosshirnrinde in ihren
Prinzipien Dargestellt auf grund des Zellenbaues. Leipzig: JA Barth; 1909.
28. Kelly-Hayes M, Robertson JT, Broderick JP, Duncan PW, Hershey
LA, Roth EJ, et al. The American Heart Association Stroke Outcome
Classification. Stroke. 1998;29:1274–1280.
Downloaded from http://ahajournals.org by on November 7, 2023
29. Weimar C, Kurth T, Kraywinkel K, Wagner M, Busse O, Haberl RL, et
al; German Stroke Data Bank Collaborators. Assessment of functioning
and disability after ischemic stroke. Stroke. 2002;33:2053–2059.
30. Patel AT, Duncan PW, Lai SM, Studenski S. The relation between impair-
ments and functional outcomes poststroke. Arch Phys Med Rehabil.
2000;81:1357–1363. doi: 10.1053/apmr.2000.9397.
31. de Haan R, Limburg M, Bossuyt P, van der Meulen J, Aaronson N.
The clinical meaning of Rankin ‘handicap’ grades after stroke. Stroke.
1995;26:2027–2030.
32. Huth AG, de Heer WA, Griffiths TL, Theunissen FE, Gallant JL. Natural
speech reveals the semantic maps that tile human cerebral cortex. Nature.
2016;532:453–458. doi: 10.1038/nature17637.
33. Izumihara A, Ishihara T, Iwamoto N, Yamashita K, Ito H. Postoperative
outcome of 37 patients with lobar intracerebral hemorrhage related to
cerebral amyloid angiopathy. Stroke. 1999;30:29–33.