J Neurol (2009) 256:1146–1151

DOI 10.1007/s00415-009-5104-8

ORIGINAL COMMUNICATION

Ischemic stroke of the cortical ‘‘hand knob’’ area:

stroke mechanisms and prognosis
Nils Peters Æ Stefanie Müller-Schunk Æ
Tobias Freilinger Æ Marco Düring Æ
Thomas Pfefferkorn Æ Martin Dichgans

Received: 15 October 2008 / Revised: 26 January 2009 / Accepted: 11 March 2009 / Published online: 8 April 2009
Ó Springer-Verlag 2009

Abstract Cortical ischemic stroke affecting the precentral
‘‘hand knob’’ area is a rare but well known stroke entity. To
date, little is known about the underlying stroke mechanisms
and the prognosis. Twenty-nine patients admitted to our
service between 2003 and 2007 were included in the study on
the basis of an acute ischemic infarct of the cortical ‘‘hand
knob’’ area confirmed by diffusion-weighted magnetic
resonance imaging with contralateral hand paresis. For all
patients clinical, epidemiological as well as imaging data at
the time point of admission were analysed retrospectively
and follow-up data on all patients was obtained. The majority
(n = 21/72%) had an isolated infarct of the cortical ‘‘hand
knob’’ area. In 23 (79%) patients it was a first ever stroke. Ten
patients (34%) had ipsilateral extracranial stenosis of the
internal carotid artery (ICA), whereas potential cardiac
embolic sources were less frequent (n = 4/14%). No patient
exhibited ipsilateral MCA stenosis. All but two patients
(93%) had marked atherosclerotic alterations of the ICA.
Hypertension was the most prevalent vascular risk factor
(n = 23/79%). At follow-up (mean 25.0 months, range 0.4–
47.4 months) no patient had died and only one (3%) expe-
rienced a recurrent stroke. The majority of patients (79%)
reported improvement of hand paresis, 17 (59%) were
asymptomatic (modified Rankin score = 0). Only one
patient was significantly disabled due to a recurrent stroke. In
N. Peters (&)
T. Freilinger
M. Düring
T. Pfefferkorn

M. Dichgans
Department of Neurology, Klinikum Grosshadern,
Ludwig-Maximilians-University, Marchioninistr. 15,
81377 Munich, Germany
e-mail: Nils.Peters@med.uni-muenchen.de
S. Müller-Schunk
Department of Neuroradiology, Klinikum Grosshadern,
Ludwig-Maximilians-University, Munich, Germany
conclusion, ischemic infarcts affecting the cortical ‘‘hand
knob’’ area are frequently associated with atherosclerotic
changes of the carotid artery, suggesting an arterio-arterial
thrombembolic stroke mechanism. It mostly reflects first
ever ischemic stroke, and follow-up data suggest a rather
benign course.
Keywords Cortical
Hand knob
Stroke
Prognosis

Precentral gyrus
Introduction
The omega or epsilon shaped region within the precentral
gyrus referred to as the cortical ‘‘hand-knob’’ is the site of
hand motor function [1, 2]. Interestingly, the cortical ‘‘hand
knob’’ area is known to be the isolated site of acute ischemic
strokes leading to contralateral hand or distal arm paresis
[3–6], which often is functionally very disabling in the acute
phase. To date, little is known about the underlying stroke
mechanisms and the clinical course of this rather rare, but
well known stroke entity. A recent short term study sug-
gested embolic mechanisms as the main cause and reported a
benign short term course within the first few days after stroke
[3]. However, no long term data exist on the clinical course.
Aims of the present study were to study: (1) the underlying
stroke mechanisms; and (2) the prognosis regarding func-
tional outcome as well as risk of stroke recurrence in patients
with cortical ischemic stroke of the ‘‘hand knob’’ area.
Methods
Twenty-nine patients admitted to our service between 2003
and 2007 were identified by systematic review of our

123
J Neurol (2009) 256:1146–1151 1147

in-patient stroke-unit database and an electronic search of Table 1 Baseline characteristics at time of admission
medical records as well as MR-imaging reports with sub- Subjects (n = 29)
sequent review of MR-images and included on the basis of:
(i) an acute ischemic infarct of the cortical ‘‘hand knob’’ Demographic characteristics
area confirmed by diffusion-weighted (DWI) magnetic Age (years, mean ± SD) 70.8 ± 9.3
resonance imaging (MRI) with (ii) contralateral hand Gender (male/female) 20/9
paresis. The patients were recruited from a total of 3499 Handedness (right/left) 22/7
patients who were treated on our stroke unit during the Clinical parameters
5 year period. Side of stroke (right/left) 16/13
Cranial MRI was performed on a high field system (1.5 Hand paresis 29 (100%)
Tesla, Magnetom Vision or Magnetom Symphony, Dominant hand affected 16 (55%)
Siemens, Erlangen, Germany/3 Tesla Signa Excite HDX, Sensory disturbance of affected hand 10 (34%)
General Electric, Milwaukee, USA) using a standardized Potential stroke mechanisms
stroke-protocol including the following sequences: DWI Atherosclerosis of carotid artery 27 (93%)
(slice 5 mm, FOV 210, TR 4200 ms, TE 139 ms/slice Ipsilateral ICA-Stenosis C 50% 10 (34%)
5 mm, FOV 2400, TR 6000, TE 80.8), PD/T2 weighted Cardiac embolic source 4 (14%)
imaging (slice 5 mm, FOV 210, TR 2750 ms, TE 13/ MCA-Stenosis 0
91 ms) or FLAIR imaging (slice 5 mm, FOV 220, TR Leukoaraiosis 7 (24%)
7500 ms, TE 74 ms/slice 5 mm, FOV 220, TR 8502, TE Vascular risk factors
122,4) and time of flight MRA (96 slices, slice 0.8 mm, Hypertension 23 (79%)
FOV 202, TR 26 ms, TE 7.15 ms/2 9 56 slices, slice 0.8 Hypercholesterolemia 12 (41%)
FOV 200 9 162, TR 25, TE 3.2). Diabetes mellitus 6 (21%)
All patients received MRI within five days after onset Smoking 12 (41%)
of symptoms (mean 2.3 days). The majority (n = 21; Concomitant vascular disease
72%) had an isolated infarct restricted to the cortical Previous stroke (yes/no) 6/23
‘‘hand knob’’ area. The remaining patients had a cortical Coronary artery disease 8 (28%)
ischemia slightly extending to subjacent cortical or sub-
Peripheral artery disease 3 (10%)
cortical areas or a leading cortical ‘‘hand knob’’ lesion
Treatmenta
accompanied by a maximum of three smaller ischemic
Antithromboticsb 25 (86%)
DWI-lesions affecting areas in the vicinity of the pre-
Anticoagulants 4 (14%)
central cortical ‘‘hand knob’’. Patients with acute terri-
Statins 25 (86%)
torial infarcts or patients with multiple diffusely
Antihypertensives 22 (76%)
distributed DWI lesions also affecting the ‘‘hand knob’’
Antidiabetics 6 (21%)
area were not included. The approximate volume of the
Interventional treatment 8 (28%)
ischemic ‘‘hand knob’’ lesion was calculated from dif-
a
fusion weighted images and T2/Flair images where Following stroke
b
possible (volume = (X 9 Y 9 Z)/2). Aspirin, clopidogrel or aspirin/dipyridamol
The following baseline data were obtained retrospec-
tively on all patients (Table 1): clinical and epidemiolog- Results
ical data including age, gender, handedness, stroke
localization, medical history, vascular risk factors, con- Baseline
comitant vascular disease, and medication; Doppler-/
duplexsonography-findings of extra- and intracranial ves- By definition, all 29 patients had a hand paresis contra-
sels, including assessment of mean flow velocities in the lateral to the site of acute cortical ischemia detected by
middle cerebral artery obtained by transcranial Doppler- DWI. This cohort represented less than 1% of all 3499
sonography; cardiac diagnostics (including transesophageal patients treated on our stroke unit during the five year
echocardiography and continuous Holter-ECG). period.
For all patients, follow-up data were obtained via a Detailed results are shown in Table 1. Sixteen patients
standardized interview, especially addressing the following (55%) had an acute ischemic infarct of the right, and
issues: clinical course regarding hand motor function and thirteen (45%) of the left cortical ‘‘hand knob’’ area. Mean
related disability using the modified Rankin Scale [7]; lesion volume of the acute cortical ischemic ‘‘hand knob’’
stroke recurrence; occurrence of other (vascular) disease; lesion was 0.58 ml (± 0.87). In 14 patients (48%) the
change of stroke related medication. ischemic lesion was restricted to the lateral and in 8 (28%)

123
1148 J Neurol (2009) 256:1146–1151

to the medial ‘‘hand knob’’ area. In 7 (24%) the ischemic
lesion comprised the complete ‘‘hand knob’’ area (Fig. 1).
No relevant pathological MRA-findings (e.g. stenosis or
low flow signal) of the middle cerebral artery (MCA) of the
affected side were observed.
In sixteen patients (55%) the dominant hand was
affected. For 23 (79%) patients it was a first ever stroke.
Upon clinical examination, isolated hand paresis was the
only pathological finding in 22 patients (76%). The most
frequent concomitant finding was ipsilateral facial weak-
ness (n = 6), while transient dysarthria was present in only
one patient. Ten patients reported sensory disturbance of
the affected hand.
Ten (34%) patients had ipsilateral extracranial internal
carotid artery (ICA) stenosis (C 50%) and four (14%) had
a potential cardiac embolic source (three patients with
atrial fibrillation, one patient with patent foramen ovale and
atrial septum aneurysm). No patient exhibited MCA ste-
nosis. All but two patients (93%) had marked atheroscle-
rotic alterations of the carotid artery (atherosclerotic
plaques and IMT [ 1 mm). There was no significant dif-
ference of average mean flow velocities in the MCA of the
affected side versus the non-affected side as assessed by
transcranial Dopplersonography. Relevant ischemic white
matter lesions were observed in only seven (24%) patients.
Hypertension was the most prevalent vascular risk factor
(n = 23; 79%). Medical treatment for secondary stroke
prevention was initiated in all patients (Table 1). Eight
patients underwent stenting of ipsilateral ICA stenosis.
These patients received medical treatment according to
standard procedures, i.e. initial dual antithrombotic treat-
ment followed by monotherapy.
Follow-up
The mean follow-up period was 25.0 months (range 0.4–
47.4 months). No patient had died and only one (3%)
experienced a second stroke. Twenty-three (79%) patients
reported improvement of hand paresis. Of the remaining
six patients, five were the ones with the most recent strokes

Fig. 1 Ischemic stroke of the

precentral cortical ‘‘hand knob’’
area. Examples of
corresponding cranial MRI
scans (1a–c: 3.0 Tesla, 2 and
3a–c: 1.5 Tesla) of three
patients including FLAIR/T2
weighted images (a), diffusion
weighted images (b) and ADC
map (c). All patients show an
acute ischemic lesion of the
‘‘hand knob’’ (b, c) with
additional chronic post ischemic
lesions (a). Acute ischemic
lesions affect the medial part
(1), the lateral part (2) or the
complete cortical ‘‘hand knob’’
(3)

123
J Neurol (2009) 256:1146–1151
with a mean follow-up time of just 6.9 months (range 0.4–
13.6).
Seventeen (59%) were asymptomatic at follow-up
(modified Rankin score = 0). Nine (31%) patients had
mild disability (modified Rankin Score = 1), and only one
patient was significantly disabled due to a recurrent stroke.
There was no significant age difference between the
asymptomatic subjects and those with disability.
Two patients had suffered myocardial infarction; for
three patients, antihypertensive treatment had been inten-
sified; one patient reported newly diagnosed diabetes
mellitus.
Excluding the patients that had undergone interventional
treatment and for whom antithrombotic treatment was
changed from dual to monotherapy, seven patients (24%)
had changed their antithrombotic medication during the
follow-up period: three patients reported change of anti-
thrombotic medication (i.e. dosage change or change from
one antithrombotic to another), two patients had been
switched to anticoagulation and two patients had stopped
antithrombotic medication (Table 2).
Discussion
In the present study, we analysed the underlying vascular
stroke mechanisms and the clinical course related to
ischemic infarcts affecting the precentral cortical ‘‘hand
knob’’ area in a cohort of twenty-nine patients. Given the
rather rare occurrence of isolated cortical ‘‘hand knob’’
stroke compared to the overall number of patients treated in
our stroke unit during the 5 year period, our cohort of
twenty-nine patients represents a rather large cohort,
which—to our knowledge—constitutes the largest cohort
to date.
Cardiovascular disease was present in the vast majority
of patients, with arterial hypertension being the most pre-
valent risk factor. These findings indicate that ischemic
stroke of the cortical ‘‘hand knob’’ area is related to
Table 2 Follow-up data
N (%)
Recurrent stroke 1 (3%)
Deaths 0
Clinical improvement 23 (79%)
Asymptomatic (mRS = 0) 17 (59%)
Mild disability (mRS = 1) 9 (31%)
Change of medicationa 7 (24%)
Mean follow-up 25.0 months (range 0.4–47.4 months)
mRS modified Rankin Scale
a
Secondary prevention
1149
common vascular disease. Overall, in approximately half
of the patients an embolic source could be identified (ten
with ipsilateral ICA stenosis, four with a potential cardiac
embolic source). Of note, nearly all patients exhibited
marked atherosclerotic changes of the carotid artery, indi-
cating a possible arterio-arterial embolic mechanism for the
majority of patients. These findings support previous
observations that arterial embolic stroke is associated with
smaller and more superficial infarcts compared to cardio-
genic embolism [8]. In contrast to the high frequency of
large vessel disease, relevant ischemic white matter chan-
ges were observed in only seven patients, arguing against a
microangiopathic pathomechanism. Our finding of a most
likely embolic stroke mechanism is in line with observa-
tions from a previous study, in which a possible source of
embolism could be detected in 11 of 14 patients with
cortical ‘‘hand knob’’ stroke [3]. Of note, none of our
patients had higher degree ipsilateral MCA stenosis. This
finding is in line with previous observations of MCA ste-
nosis being more frequently associated with deep perfora-
tor and internal borderzone infarcts [9].
In our own experience, isolated cortical ischemic strokes
affecting the precentral ‘‘hand knob’’ area are clinically
observed more frequently than strokes of other cortical
areas. Clearly, this observation may be biased: cortical
ischemic strokes affecting less eloquent areas may be
clinically underdiagnosed. Alternatively, yet hypotheti-
cally, cerebral hemodynamic factors may make this func-
tionally important area more susceptible than other cortical
regions within the MCA territory. MCA stroke has been
shown to be associated with thrombo-embolism and
extracranial large carotid artery disease in Caucasians and
the MCA territory is most frequently affected by thrombo-
embolisms compared other cerebral vascular territories [10,
11]. Yet, isolated infarcts in other functionally important
cortical areas within the MCA territory, e.g. leading to
isolated dysarthria, occur infrequently [12]. Although
individual variability of cerebral arterial territories is
known to exist [13–15], it has been shown that emboli may
go to the same arterial branches [16]. Despite variability of
arterial territories, there may be possible inter-individual
similarity of the vascular supply of functionally important
regions such as the cortical ‘‘hand knob’’. The MCA blood
supply from the superior and intermediate trunk of the
primary motor cortex has been shown to be mostly domi-
nated by a central group of vessels (72.5%), in fewer cases
by a precentral group (10%) or both (17.5%), but in no case
by the postcentral group, explaining few postcentral lesions
in combination with embolic lesions of the ‘‘hand knob’’
[17]. Yet, overall, we are aware that such susceptibility of
the cortical ‘‘hand knob’’ area remains hypothetical and
that the apparently more frequent clinical occurrence of
isolated cortical ‘‘hand knob’’ stroke compared to infarcts
123
1150
of other cortical areas may be biased, with the latter fre-
quently being clinically silent infarcts.
In the present study, we could not identify obvious
functional or hemodynamic parameters possibly explaining
stroke localization within the ‘‘hand knob’’ area in our
cohort: no side-differences of flow velocity of the MCA
assessed by Dopplersonography were observed. This was
also true for the subgroup of patients without upstream
ICA-stenosis. Also, no obvious abnormalities of the ipsi-
lateral MCA or its branches were detected via MRA. As
expected, we did not observe any association of stroke
localization with handedness. This was also true if the ten
patients with ipsilateral, stroke-defining ICA-stenosis were
excluded.
Most patients presented with isolated hand paresis. Some
patients, however, had concomitant symptoms, especially
sensory disturbance of the affected hand. In the majority of
these patients, sensory symptoms occurred despite the
absence of a DWI lesion of the major sensory pathways
including the postcentral gyrus. Sensory disturbance was
mostly transient, possibly suggesting transient ischemia with
absence of an ischemic DWI lesion. On the other hand,
sensory fibers are known to also project to the precentral
gyrus (Brodman area 4), explaining possible sensory
symptoms from isolated motor cortex infarcts. Finally, in
many cases sensory disturbance was reported by patients
without clear findings upon clinical examination (such as
tactile hypaesthesia). Thus, these symptoms may reflect
subjective ‘‘sensory’’ disturbance due to motor dysfunction.
Overall, the follow-up data of our study indicate a rather
good functional outcome. Approximately three quarters of
patients reported improvement of hand paresis, with about
two-thirds being asymptomatic at follow-up (modified
Rankin score = 0). Also, patients who still reported dis-
ability were mostly those with rather recent strokes and
thus shorter follow-up periods. This suggests possible
further long-term clinical improvement also in these
patients, which may for example be due to recruitment of
other cortical areas [18]. So far, only data on the short-term
prognosis of cortical ‘‘hand knob’’ ischemia regarding
functional outcome in a clinical setting have been reported,
with a follow-up period of days [3]. Our findings of a good
functional outcome without disability in activities of daily
living over a mean period of approximately two years
extend these results.
We further found a low stroke recurrence rate. In fact,
there was only one patient with multiple extra- and intra-
cranial arterial stenoses who experienced a new stroke,
leading to increased disability. The low recurrence rate
may reflect the natural course, the effects of medical
treatment, or—most likely—a combination of both. Over-
all, patients exhibited good medical compliance regarding
secondary prevention. In the presence of common vascular
123
J Neurol (2009) 256:1146–1151
risk factors—especially arterial hypertension—and a high
rate of atherosclerosis in our cohort, thorough medical
treatment in these patients is warranted.
A limitation to our study is the retrospective data
assessment. However, given the standardized diagnostic
work-up of stroke patients and detailed electronic docu-
mentation of clinical and epidemiological data, we believe
that this is no major concern and that relevant aspects could
be addressed. Also, follow-up data could be obtained for all
patients.
In conclusion, our data illustrate that ischemic infarcts
affecting the cortical ‘‘hand knob’’ area are frequently
associated with arterial hypertension and atherosclerosis,
suggesting an arterio-arterial thrombembolic stroke mech-
anism in these patients. Our follow-up data suggest a rather
benign course and may be of value for patient counseling.
References
1. Yousry TA, Schmid UD, Alkadhi H, Schmidt D, Peraud A,
Buettner A, Winkler P (1997) Localization of the motor hand area
to a knob on the precentral gyrus. A new landmark. Brain
120:141–157
2. Caulo M, Briganti C, Mattei PA, Perfetti B, Ferretti A, Romani
GL, Tartaro A, Colosimo C (2007) New morphologic variants of
the hand motor cortex as seen with MR imaging in a large study
population. AJNR Am J Neuroradiol 28:1480–1485
3. Gass A, Szabo K, Behrens S, Rossmanith C, Hennerici M (2001)
A diffusion-weighted MRI study of acute ischemic distal arm
paresis. Neurology 57:1589–1594
4. Tei H (1999) Monoparesis of the right hand following a localised
infarct in the left ‘‘precentral knob’’. Neuroradiology 41:269–270
5. Phan TG, Evans BA, Huston J (2000) Pseudoulnar palsy from a
small infarct of the precentral knob. Neurology 54:2185
6. Takahashi N, Kawamura M, Araki S (2002) Isolated hand palsy
due to cortical infarction: localization of the motor hand area.
Neurology 58:1412–1414
7. Newcommon NJ, Green TL, Haley E, Cooke T, Hill MD (2003)
Improving the assessment of outcomes in stroke: use of a struc-
tured interview to assign grades on the modified Rankin Scale.
Stroke 34:377–378
8. Timsit SG, Sacco RL, Mohr JP, Foulkes MA, Tatemichi TK,
Wolf PA, Price TR, Hier DB (1993) Brain infarction severity
differs according to cardiac or arterial embolic source. Neurology
43:728–733
9. Lee PH, Oh SH, Bang OY, Joo SY, Joo IS, Huh K (2004) Infarct
patterns in atherosclerotic middle cerebral artery versus internal
carotid artery disease. Neurology 62:1291–1296
10. Hedberg M, Boivie P, Edström C, Engström KG (2005) Cere-
brovascular accidents after cardiac surgery: an analysis of CT
scans in relation to clinical symptoms. Scand Cardiovasc
J 39:299–305
11. Yoo KM, Shin HK, Chang HM, Caplan LR (1998) Middle
cerebral artery occlusive disease: the New England medical
center stroke registry. J Stroke Cerebrovasc Dis 7:344–351
12. Kumral E, Celebisoy M, Celebisoy N, Canbaz DH, Calli C (2007)
Dysarthria due to supratentorial and infratentorial ischemic
stroke: a diffusion-weighted imaging study. Cerebrovasc Dis
23:331–338
J Neurol (2009) 256:1146–1151
13. van der Zwan A, Hillen B (1991) Review of the variability of the
territories of the major cerebral arteries. Stroke 22:1078–1084
14. van der Zwan A, Hillen B, Tulleken CA, Dujovny M (1993) A
quantitative investigation of the variability of the major cerebral
arterial territories. Stroke 24:1951–1959
15. Hennerici M, Daffertshofer M, Jakobs L (1998) Failure to iden-
tify cerebral infarct mechanisms from topography of vascular
territory lesions. AJNR Am J Neuroradiol 19:1067–1074
16. Whisnant JP (1982) Multiple particles injected may all go to the
same cerebral artery branch. Stroke 13:720
1151
17. Ugur HC, Kahilogullari G, Coscarella E, Unlu A, Tekdemir I,
Morcos JJ, Elhan A, Baskaya MK (2005) Arterial vascularization
of primary motor cortex (precentral gyrus). Surg Neurol 64 Suppl
2:48–52
18. Butefisch CM, Kleiser R, Korber B, Muller K, Wittsack HJ,
Homberg V, Seitz RJ (2005) Recruitment of contralesional motor
cortex in stroke patients with recovery of hand function.
Neurology 64:1067–1069
123