Você está na página 1de 8

Acta Neurochir (2016) 158:181188

DOI 10.1007/s00701-015-2628-9


Does arteriosclerosis contribute to hemifacial spasm?

Miki Ohta 1 & Masahito Kobayashi 1 & Naruhiko Terano 1 & Kenji Wakiya 1 &
Kenji Suzuki 2 & Takamitsu Fujimaki 1

Received: 4 August 2015 / Accepted: 28 October 2015 / Published online: 7 November 2015
# Springer-Verlag Wien 2015

Abstract Conclusions It is suggested that arteriosclerotic changes are

Background Hemifacial spasm (HFS) is caused by pulsative not involved in the pathogenesis of HFS, and that vascular
vascular compression of the root exit zone (REZ) of the facial compression syndromes are attributable to anatomical features
nerve. However, the mechanism that causes the offending of the intracranial arteries and facial nerves formed during the
vessels to compress the REZ has not been clarified. prenatal stage.
Elongation of intracranial arteries due to arteriosclerosis is
one possibility, but such arteriosclerotic changes are not ob- Keywords Arteriosclerosis . Cardio-ankle vascular index .
served very frequently among patients with HFS. The aim of Hemifacial spasm . White matter lesions
the present study was to investigate whether arteriosclerotic
changes would contribute to the pathogenesis of HFS.
Methods This study included 111 HFS patients, all of whom Introduction
were Japanese. The prevalence rates of hypertension, hyper-
lipidemia, and diabetes mellitus were examined as risk factors It is widely accepted that hemifacial spasm (HFS) is caused by
of atherosclerosis, and the cardio-ankle vascular index (CAVI) pulsative vascular compression of the root exit zone (REZ) of
was measured as an indicator of arteriosclerotic change. The the facial nerve [3, 5]. However, the pathogenesis responsible
severity of white matter lesions (WMLs) in HFS patients was for compression of the REZ by intracranial vessels remains
measured by magnetic resonance imaging. These data were unclear. A correlation between this condition and elongation
compared with data from healthy Japanese controls. of the intracranial arteries due to arteriosclerosis has been
Results The prevalence rates of the risk factors for atheroscle- speculated because the symptoms usually begin in midlife or
rosis in the HFS patients were not higher than those in the later [10, 11]. Furthermore, previous reports have discussed
general Japanese population. The CAVI scores for the HFS hypertension as a risk factor for HFS [6, 7, 23, 30, 31].
patients were similar to, or lower than those in the healthy Arteriosclerotic changes in the compressing arteries have been
controls for all age groups except 60 to 69-year-old men. demonstrated intraoperatively and in angiograms of HFS pa-
The severity of WMLs in the HFS patients was not signifi- tients [15]. However, such changes were not always obvious
cantly worse than that in the controls. in our previous operative series of more than 500 HFS pa-
tients, most of whom showed apparently normal compressing
vessels without arteriosclerosis, as revealed using an operating
microscope. In addition, the prevalence of risk factors for
* Miki Ohta atherosclerosis, such as hypertension, hyperlipidemia or dia-
m_johta@saitama-med.ac.jp betes mellitus, was not particularly high in comparison with
that among stroke patients.
Department of Neurosurgery, Saitama Medical University, 38 Recently, the cardio-ankle vascular index (CAVI) has been
Morohongo, Moroyama-machi, Iruma, Saitama 350-0495, Japan established for evaluation of arteriosclerotic changes. The
Japan Health Promotion Foundation, 1-244 Ebisu, Shibuya-ku, CAVI is an index of the overall stiffness of an artery from its
Tokyo, Japan origin on the aorta to the ankle, indicating the degree of
182 Acta Neurochir (2016) 158:181188

arteriosclerosis. The CAVI value increases with age in both Health and Nutrition Examination Survey of Japan, 2012 [2].
men and women [26], and it has been reported that there is a A total of 12,488 Japanese people (5,026 men and 7,462
convincing relationship between the CAVI value and diseases women) aged 20-74 years for whom valid information was
resulting from atherosclerosis [9, 16, 20, 28]. available were extracted and used as the general Japanese
In the present study, we investigated the prevalence of hy- population for comparing the prevalence rates of the diseases.
pertension, hyperlipidemia and diabetes mellitus in a cohort of The criteria for these diseases were a history of the following
patients to determine whether arteriosclerosis is a risk factor conditions, or medication for them: hypertension: systolic
for HFS. In addition, the CAVI was measured to evaluate blood pressure 140 mmHg and/or diastolic blood pressure
arteriosclerotic changes in the HFS patients in comparison 90 mmHg; hyperlipidemia: serum LDL level 140 mg/dl;
with normal controls, and the white matter lesions (WMLs) diabetes mellitus: HbA1c 6.5 %.
in the preoperative magnetic resonance (MR) images were
investigated to estimate consecutive, possible ischemic chang- CAVI
es in the brain parenchyma of patients with HFS.
In addition to the risk factors for atherosclerosis, CAVI scores,
which reflect arteriosclerotic changes, were also determined in
Methods and materials all patients. The CAVI and blood pressure were measured
within a week before MVD surgery. During CAVI measure-
Clinical materials ment, patients were instructed to relax in a supine position
with cuffs applied to the bilateral upper arms and ankles.
This study included 111 consecutive patients who underwent Blood pressure was measured at the bilateral upper brachial
microvascular decompression (MVD) for HFS at our institu- arteries and electrocardiograms were monitored. CAVI was
tion, Saitama Medical University Hospital, between measured with a VaSera CAVI instrument (VS-1500ATN;
May 2012 and February 2014. All of the patients were native Fukuda Denshi, Tokyo, Japan) using the methods described
Japanese and resident in Japan. During this period, 113 pa- previously [27, 29]. Briefly, CAVI was calculated using the
tients with HFS had consulted us about MVD and two of them following formula: CAVI=a{(2/P)ln(Ps/Pd)PWV2}+b,
were excluded as candidates for surgery because of severe where Ps is the systolic blood pressure, Pd the diastolic blood
atrial fibrillation and myelodysplasia syndrome. All of the pressure, PWV pulse wave velocity, P=Ps - Pd, is the
subjects provided written informed consent before participat- blood density, a and b are constants. The PWV was calculated
ing in the study, which had been approved by the institutional from the vascular length and latency of pulse waves propagat-
review board of Saitama Medical University. The mean pa- ing between the aortic valve and the ankle [27, 33]. The CAVI
tient age was 50.7 years (median,50 years; range, 1976 years; values were measured bilaterally in each subject, and the mean
SD,11.9), and 87 (78.4 %) of the patients were women. The values were used for analysis.
mean age at onset was 43.7 years (median,43 years; range, The original control data for the VaSera CAVI instrument
1867 years; SD,11.9). Sixty-one patients (55.0 %) had left- included 69,967 individuals (34,870 men and 35,097 women)
sided HFS. On admission to our institution, five men and who underwent an annual health check organized by the Japan
seven women were receiving medication for hypertension, Health Promotion Foundation at urban workplaces in Japan
two men and eight women for hyperlipidemia, and one man between January 2006 and March 2013. In this study, to ana-
and one woman for diabetes mellitus. The other 92 patients lyze strictly and elucidate the risk factors for HFS, data for
were not receiving these medications. individuals with possible risk factors for atherosclerosis were
In 25 patients (22.5 %), the vertebral artery (VA) was in- excluded from the normal control group when determining the
volved as the offending vessel, compressing the REZ of the CAVI. The exclusion criteria were: (1) hypertension with
facial nerve. In the remaining 86 patients (77.5 %), the poste- medication, or a systolic blood pressure of 140 mmHg, or a
rior inferior cerebellar artery or/and anterior inferior cerebellar diastolic blood pressure of 90 mmHg; (2) hyperlipidemia
artery compressed the REZ without involvement of the VA or with medication, or a serum low-density lipoprotein (LDL)
the basilar artery. level of 140, or a serum high-density lipoprotein (HDL) level
<40, or a serum triglyceride level of 150; (3) diabetes
Risk factors for atherosclerosis mellitus with medication, or HbA1c 6.5 %, or a casual blood
glucose level of 200 mg/dl; (4) history of a cardiovascular
To examine if the risk factors for atherosclerosis are also risk event, or cerebrovascular event. Finally, the normal control
factors for HFS, the prevalence rates of hypertension, hyper- group from which CAVI data were obtained comprised 28,
lipidemia [high low-density lipoprotein (LDL)], and diabetes 350 healthy individuals (10,999 men and 17,351 women)
mellitus among HFS patients were compared with those in the without hypertension, hyperlipidemia, diabetes or a history
general Japanese population, based on data from the National of cardiovascular or cerebrovascular diseases, and the normal
Acta Neurochir (2016) 158:181188 183

CAVI values were determined on this basis. The mean age of Results
the normal control CAVI individuals was 43.4 years (median,
41 years; range, 1691 years; SD,12.6). Risk factors for atherosclerosis

White matter lesions (WMLs) The prevalence rates of hypertension and hyperlipidemia in
the female patients with HFS were significantly lower than in
MR imaging was performed 710 days before MVD surgery, the Japanese controls in the age groups of 5069 and 5059
using a 1.5-tesla imaging unit (MAGNETOM Sonata; years. In the other age groups of female HFS patients and all
Siemens, Erlagen, Germany) with a quadrature head coil. age groups of male patients, there were no significant differ-
The standardized MR imaging protocol included axial T2- ences in the prevalence rates of hypertension, hyperlipidemia,
weighted (repetition time/echo time [TR/TE]=4,000/88 ms) or diabetes mellitus compared with those among the general
and T1-weighted (TR/TE=660/14 ms) spin echo, and fluid- Japanese population (Fig. 1). Thus, the prevalence rates of
attenuated inversion recovery (FLAIR; TR/TE = 10,000/ these diseases were not high among the HFS patients, but
99 ms). The severity of WML was classified according to were rather lower in middle-aged female patients.
the rating scale of the Atherosclerosis Risk in Communities
study [32], as follows: grade 0=no white matter signal abnor- CAVI
malities; grade 1=a discontinuous periventricular rim or min-
imal Bdots^ of subcortical white matter; grade 2=a thin con- In the HFS group, the average CAVI score in men and women
tinuous periventricular rim or a few patches of subcortical was 7.671.47 and 7.341.08, respectively. In the normal
WMLs; grade 3=a thicker continuous periventricular rim with control group, the average CAVI score in men and women
scattered patches of subcortical WMLs; grade 4=a thicker, was 7.200.97 and 7.170.89, respectively. CAVI tended to
shaggier periventricular rim with mild subcortical WMLs, increase with age in both groups, and in both men and women
possibly with minimal confluent periventricular lesions; grade (Fig. 2). The CAVI score of 84.7 % for the HFS patients was
5=mild periventricular confluence surrounding the frontal lower than the mean+1SD lines, and the CAVI score of
and occipital horns; grade 6=moderate periventricular conflu- 60.4 % for the patients was below the mean of the normal
ence surrounding the frontal and occipital horns; grade 7= control group without risk factors for arteriosclerosis. The
periventricular confluence with moderate involvement of the patients with VA-involved vascular compression and non-
centrum semiovale; grade 8=periventricular confluence in- VA-involved vascular compression are presented separately
volving most of the centrum semiovale; grade 9=all of the in this figure because large vessels, such as the VA, would
white matter involved. One neurosurgeon and two radiolo- be strongly influenced by systemic atherosclerotic changes.
gists, who were all blind to the patients data, evaluated and Figure 3 shows the mean CAVI values for the HFS patients
classified the severity of WMLs in the preoperative MR and the normal control group at different ages. For men,
images. ANOVA revealed a significant effect of Bage^ on the CAVI
score [F(5, 10,918)=13.689, p<0.00001] and also a signifi-
Statistical analysis cant interaction between Bage^ and Bgroups^ [F(5, 10,918)=
5.429, p=0.00005], but without any group effect [F(1, 10,
All analyses were performed separately in the groups of men 918)=0.740, p=0.390]. Post hoc tests demonstrated that in
and women. the controls the CAVI score increased significantly with age:
The prevalence rates of hypertension, hyperlipidemia and in the HFS patients the CAVI score for patients in their 60s
diabetes mellitus were classified according to age. Fishers was significantly higher than those for patients in their 30s and
exact probability test was used to compare these prevalence 40s, and the CAVI score for those in their 50s was significant-
rates between the HFS patients and the general Japanese ly higher than that for patients in their 40s. In addition, the
population. CAVI scores for male HFS patients aged 6069 years were
The CAVI data for HFS patients and the normal controls significantly higher than those for controls (Fig. 3, left). As for
were also classified by age group and subjected to two-way women, ANOVA revealed a significant effect of Bage^ on the
analysis of variance (ANOVA) followed by post hoc changes in CAVI score [F(5, 17,401)=37.53, p<0.00001] and
Bonferroni/Dunn correction. All results were reported as an interaction between Bage^ and Bgroups^ [F(5, 17,401)=
meansSDs and grouped according to age. The HFS pa- 2.255, p= 0.046], but without any group effect [F(1, 17,
tients were divided into two groups according to the 401)=0.081, p=0.776]. Post hoc tests showed that in the con-
offending vessels, i.e. VA involvement or not, and the trols the CAVI score for each age group was significantly
CAVI values were compared between these two groups higher than for younger age groups: in the HFS patients the
using ANOVA. All results were reported as meansSDs CAVI score for those in their 60s was significantly higher than
and grouped according to age. that for patients in their 30s and 40s, and the score for patients
184 Acta Neurochir (2016) 158:181188

Fig. 1 Prevalence rates of hypertension, hyperlipidemia, and diabetes asterisksnumbers above columns show the number of people with risk
mellitus. Black columns HFS patients; gray columns data from the factors/total number of people; indicate significant differences
National Health and Nutrition Examination Survey of Japan, 2012;

in their 70s was significantly higher than for patients in their White matter lesions
20s, 30s, 40s and 50s. In addition, the CAVI scores for HFS
patients aged 40-49 years were significantly lower than those Table 1 shows the prevalence of each grade of WMLs by age
for the controls. These results indicate that CAVI scores in- group. Most of the HFS patients preoperative MR images
creased gradually with age, and that the scores for the HFS showed no or only slight WMLs. In total, WMLs were found
patients were quite similar to those for the controls when com- in 27 of the 111 HSF patients (24.3 %) and only two (1.8 %)
pared in each age group, except for male patients aged 6069 patients had moderate grade (grades 4 and 5) WMLs. The
years. There was no significant difference in the CAVI values prevalence rates among the HFS patients appeared to be al-
between patients with non-VA-involved and VA-involved most similar to, and not significantly higher than, those in the
HFS (data not shown). general Japanese population (see also BDiscussion^).
Acta Neurochir (2016) 158:181188 185

Fig. 2 CAVI values for the normal control group and HFS patients. Gray line and gray broken lines indicate the means and standard deviations of the
normal control group. Small and large dots represent non-VA-involved HFS patients and VA-involved HFS patients, respectively

Discussion Japan. These findings indirectly but clearly indicate that these
diseases are not risk factors for HFS, and that arteriosclerotic
Our study analyzed arteriosclerotic changes in patients with changes in HFS patients are similar to, or even less marked
HFS by investigating the prevalence rates of hypertension, than those in the general Japanese population.
high LDL cholesterol levels, and diabetes mellitus as risk The CAVI is an index of the overall stiffness of an artery
factors for atherosclerosis, and we also assessed CAVI scores from its origin on the aorta to the ankle. When the vascular
and the prevalence rates of WMLs as an index of brain arte- walls of the aorta and elastic arteries are quite stiff, they do not
riosclerosis. All of the study results indicated that arterioscle- cushion the pulsatile flow generated by cardiac beats as effec-
rotic changes do not contribute to the pathogenesis of HFS, tively [21]. Without this buffering capacity, such pulsations
especially in individuals younger than 60 years of age. will extend into the microvasculature, and preferentially into
Among our HFS patients, who were all native Japanese, the organs with high perfusion such as the brain and kidneys [21,
prevalence rates of the risk factors for atherosclerosis were all 22]. In individuals with arteriosclerosis, high pulsation pres-
similar to, or even lower than, those indicated by data for the s u r e a n d f l o w w i t h o u t e n e rg y l o s s w i l l i n d u c e
general Japanese population obtained in the national survey of microhemorrhages and infarcts in these organs [21]. Indeed,

Fig. 3 CAVI values of the normal control group and HFS patients in each HFS patients; light-gray columns normal control. Data are presented as
age group. Left: CAVI values for male controls and HFS patients; right: meansSD. Error bar for male HFS patients in their 20s is not shown
those of female controls and HFS patients. Significant differences are because there was only one patient
indicated by plus, asterisk and sharp symbols (see text). Black columns
186 Acta Neurochir (2016) 158:181188

Table 1 White matter lesions

(WMLs) in each age group Age (years) Total (n) Grade of WMLs (n)a Overall prevalence rates (%)

0 1 2 3 4 5 This study Park et al. [25]

<40 18 18 0 0 0 0 0 0.0 3.6

40-49 34 31 3 1 0 0 0 11.8 12.5
50-59 30 20 6 3 0 1 0 33.3 29.5
60-69 21 13 2 4 2 0 0 38.1 49.2
70 7 2 0 4 0 0 1 71.4 81.2
No patient showed WMLs with grade >5

the CAVI score is significantly greater in patients with ische- appear in middle-aged or elderly individuals, and the
mic cerebrovascular disease, such as WMLs, large artery ath- offending arteries are likely to be elongated and arteriosclerot-
erosclerosis, and small vessel occlusion [4, 28]. Our CAVI ic. In contrast, different mechanisms such as the involvement
data for HFS patients, however, revealed no significant differ- of veins as the offending vessels [19], a thickened arachnoid
ences between HFS patients and normal controls without risk surrounding the offending vessels [14], and anatomical varia-
factors for atherosclerosis, demonstrating directly that arterio- tions at the REZ [12] have been suggested for HFS in the
sclerotic changes do not form the basis of HFS pathogenesis. young, based on observations in case reports. Our study of
WMLs, sometimes described as leukoaraiosis, are fre- the HFS patients has shown that the prevalence of risk factors
quently detected in MR images of elderly individuals, but for atherosclerosis, the CAVI data and the incidences of
are especially common in patients with known vascular risk WMLs were similar to or lower than those in the control
factors and symptomatic cerebrovascular disease [24]. Since group, except for CAVI in 60 to 69-year-old men. These
the pathophysiological mechanism of WMLs is similar to that observations imply that arteriosclerotic changes in elderly
of stroke, WMLs have been considered an intermediate sur- men with HFS would be slightly more marked than those in
rogate of stroke [8]. Park et al. [25] analyzed the clinical data normal controls, possibly contributing to the pathogenesis of
and MR images of 1,030 healthy Japanese individuals aged vascular compression. However, the number of elderly male
2878 years (mean age, 52.7 years) and found that the prev- patients in our study was not sufficiently large, and therefore
alence rate of WMLs was 28.8 % (Table 1). The prevalence of further study will be required to elucidate the contribution of
WMLs in our HFS patients did not appear to be higher than, atherosclerosis to the pathogenesis of HFS in elderly male
but rather similar to those in their Japanese population in all patients. In contrast, patients showing onset of HFS in youth
age groups. These findings, together with data for CAVI and or middle age would have minimal arteriosclerotic changes,
the prevalence rates of hypertension, hyperlipidemia and DM, and the early onset of symptoms in such patients may be
demonstrate that arteriosclerotic changes cannot explain the attributable to anatomical features that developed in the pre-
pathogenesis of HFS. natal stage, resulting in compression of the REZ of the facial
It has been suggested that HFS patients with VA-involved nerves.
compression would show atherosclerotic changes [17, 18]. A possible limitation of this study is that the HFS patient
However, in our present series, there was no significant dif- population may have included only those who were healthy
ference in CAVI values between patients with VA-involved enough to be candidates for neurosurgical intervention. In our
compression and those with non-VA-involved compression. series of 113 patients, however, only two were excluded as
While the CAVI values for HFS patients increased with age, surgical candidates because of cardiological or hematological
the changes were quite similar to those observed in normal problems. Furthermore, as a control for the CAVI data, we
controls, and indeed almost all the data for the HFS patients analyzed normal, healthy individuals without risk factors for
were within 1 SD, i.e. normal limits. Thus, it appears unlikely atherosclerosis, thus successfully avoiding any bias for HFS
that arteriosclerotic changes contribute essentially to the path- patients.
ogenesis of vascular compression, even in elderly HFS
While it is widely recognized that the cause of HFS is Conclusion
vascular compression at the REZ of the facial nerve [1], the
pathogenesis that causes small intracranial arteries to com- The results of this study support the contention that arterio-
press the REZ has not yet been clarified. A correlation be- sclerotic changes do not contribute to the pathogenesis of
tween vascular compression and arteriosclerosis has been HFS. Presumably, the anatomical features of the compressing
speculated [10, 13, 31] because the symptoms of HFS often arteries and the REZ of facial nerves would contribute to the
Acta Neurochir (2016) 158:181188 187

manifestation of vascular compression syndromes, and any neuralgia, hemifacial spasm, and glossopharyngeal neuralgia: 11-
year experience and review. Neurosurgery 43:13511361
arteriosclerotic changes that are present in some elderly male
14. Kobata H, Kondo A, Kinuta Y, Iwasaki K, Nishioka T, Hasegawa K
patients may contribute to HFS to some extent. (1995) Hemifacial spasm in childhood and adolescence.
Neurosurgery 36:710714
Acknowledgments We would like to thank Drs. Mamoru Niitsu and 15. Kondo A, Ishikawa J, Konishi T (1981) The pathogenesis of
Kaiji Inoue for diagnosis and helpful comments on the MR images of the hemifacial spasm: characteristic changes of vasculatures in
patients. vertebro-basilar artery system. In: Sammi M, Jannetta P (eds) The
For this work, support was received from a Grant-in-Aid for Young cranial nerves. Springer, New York, pp 494501
Physicians from Saitama Medical University Hospital Grant Number 24- 16. Kubozono T, Miyata M, Ueyama K, Nagaki A, Hamasaki S,
C-1 (Miki Ohta) and the Japan Society for the Promotion of Science Kusano K, Kubozono O, Tei C (2009) Association between arterial
KAKENHI Grant Number 23592105 (Masahito Kobayashi). stiffness and estimated glomerular filtration rate in the Japanese
Compliance with ethical standards general population. J Atheroscler Thromb 16:840845
17. Kurokawa Y, Maeda Y, Toyooka T, Inaba K (2004) Microvascular
Conflict of interest None. decompression for hemifacial spasm caused by the vertebral artery:
a simple and effective transposition method using surgical glue.
Surg Neurol 61:398403
18. Kwon HM, Lee YS (2011) Dolichoectasia of the intracranial arter-
ies. Curr Treat Options Cardiovasc Med 13:261267
19. Levy EI, Resnick DK, Jannetta PJ, Lovely T, Bissonette DJ (1997)
References Pediatric hemifacial spasm: the efficacy of microvascular decom-
pression. Pediatr Neurosurg 27:238241
20. Nakamura K, Tomaru T, Yamamura S, Miyashita Y, Shirai K,
1. Barker FG 2nd, Jannetta PJ, Bissonette DJ, Shields PT, Larkins Noike H (2008) Cardio-ankle vascular index is a candidate predic-
MV, Jho HD (1995) Microvascular decompression for hemifacial tor of coronary atherosclerosis. Cir J 72:598604
spasm. J Neurosurg 82:201210 21. ORourke MF (2007) Arterial aging: pathophysiological principles.
2. Bureau CCaHPDHS (2011) The National Health and Nutrition Vasc Med 12:329341
Survey in Japan, 2011. In: Ministry of Health LaW, Japan (ed) 22. ORourke MF, Safar ME (2005) Relationship between aortic stiff-
Japan ening and microvascular disease in brain and kidney: cause and
3. Campos-Benitez M, Kaufmann AM (2008) Neurovascular com- logic of therapy. Hypertension 46:200204
pression findings in hemifacial spasm. J Neurosurg 109:416420 23. Oliveira LD, Cardoso F, Vargas AP (1999) Hemifacial spasm and
4. Choi SY, Park HE, Seo H, Kim M, Cho SH, Oh BH (2013) Arterial arterial hypertension. Mov Disord 14:832835
stiffness using cardio-ankle vascular index reflects cerebral small 24. Park JH, Kwon HM, Lee J, Kim DS, Ovbiagele B (2015)
vessel disease in healthy young and middle aged subjects. J Association of intracranial atherosclerotic stenosis with severity
Atheroscler Thromb 20:178185 of white matter hyperintensities. Eur J Neurol 22(44-52):e42e43
5. Cohen-Gadol AA (2011) Microvascular decompression surgery for 25. Park K, Yasuda N, Toyonaga S, Yamada SM, Nakabayashi H,
trigeminal neuralgia and hemifacial spasm: naunces of the tech- Nakasato M, Nakagomi T, Tsubosaki E, Shimizu K (2007)
nique based on experiences with 100 patients and review of the Significant association between leukoaraiosis and metabolic syn-
literature. Clin Neurol Neurosurg 113:844853 drome in healthy subjects. Neurology 69:974978
6. Defazio G, Berardelli A, Abbruzzese G, Coviello V, De Salvia R, 26. Shirai K, Hiruta N, Song M, Kurosu T, Suzuki J, Tomaru T,
Federico F, Marchese R, Vacca L, Assennato G, Livrea P (2000) Miyashita Y, Saiki A, Takahashi M, Suzuki K, Takata M (2011)
Primary hemifacial spasm and arterial hypertension: a multicenter Cardio-ankle vascular index (CAVI) as a novel indicator of arterial
case-control study. Neurology 54:11981200 stiffness: theory, evidence and perspectives. J Atheroscler Thromb
7. Defazio G, Martino D, Aniello MS, Masi G, Logroscino G, 18:924938
Manobianca G, La Stilla M, Livrea P (2003) Influence of age on 27. Shirai K, Utino J, Otsuka K, Takata M (2006) A novel blood
the association between primary hemifacial spasm and arterial hy- pressure-independent arterial wall stiffness parameter; cardio-
pertension. J Neurol Neurosurg Psychiatry 74:979981 ankle vascular index (CAVI). J Atheroscler Thromb 13:101107
28. Suzuki J, Sakakibara R, Tomaru T, Tateno F, Kishi M, Ogawa E,
8. Inzitari D (2003) Leukoaraiosis: an independent risk factor for
Kurosu T, Shirai K (2013) Stroke and cardio-ankle vascular stiff-
stroke? Stroke 34:20672071
ness index. J Stroke Cerebrovasc Dis 22:171175
9. Izuhara M, Shioji K, Kadota S, Baba O, Takeuchi Y, Uegaito T, 29. Takaki A, Ogawa H, Wakeyama T, Iwami T, Kimura M, Hadano Y,
Mutsuo S, Matsuda M (2008) Relationship of cardio-ankle vascular Matsuda S, Miyazaki Y, Matsuda T, Hiratsuka A, Matsuzaki M
index (CAVI) to carotid and coronary arteriosclerosis. Circ J 72: (2007) Cardio-ankle vascular index is a new noninvasive parameter
17621767 of arterial stiffness. Circ J 71:17101714
10. Jannetta PJ (1977) Observations on the etiology of trigeminal neu- 30. Tan EK, Chan LL, Lum SY, Koh P, Han SY, Fook-Chong SM, Lo
ralgia, hemifacial spasm, acoustic nerve dysfunction and YL, Pavanni R, Wong MC, Lim SH (2003) Is hypertension associ-
glossopharyngeal neuralgia. Definitive microsurgical treatment ated with hemifacial spasm? Neurology 60:343344
and results in 117 patients. Neurochirurgia 20:145154 31. Tan NC, Chan LL, Tan EK (2002) Hemifacial spasm and involun-
11. Jannetta PJ, Resnick DK (1996) Cranial rhizopathies. In: Youmans tary facial movements. QJM 95:493500
J (ed) Neurological surgery, vol 5. W.B. Saunders, Philadelphia, pp 32. Wong TY, Klein R, Sharrett AR, Couper DJ, Klein BE, Liao DP,
35633574 Hubbard LD, Mosley TH (2002) Cerebral white matter lesions,
12. Jho HD, Jannetta PJ (1987) Hemifacial spasm in young people retinopathy, and incident clinical stroke. JAMA 288:6774
treated with microvascular decompression of the facial nerve. 33. Yambe T, Yoshizawa M, Saijo Y, Yamaguchi T, Shibata M, Konno
Neurosurgery 20:767770 S, Nitta S, Kuwayama T (2004) Brachio-ankle pulse wave velocity
13. Kobata H, Kondo A, Iwasaki K, Nishioka T (1998) Combined and cardio-ankle vascular index (CAVI). Biomed Pharmacother
hyperactive dysfunction syndrome of the cranial nerves: trigeminal 58(Suppl 1):S95S98
188 Acta Neurochir (2016) 158:181188

Comment hemifacial spasm, it does lay the groundwork for further study. Further
investigations of the association between arteriosclerosis and hemifacial
spasm in the elderly population and study of embryological contributions
Hemifacial spasm due to vascular compression of the facial nerve repre- to the genesis of hemifacial spasm are warranted.
sents an important neurologic issue that can be effectively treated by Ryan Hofler
microvascular decompression. At the time of surgery many surgeons Christopher M. Loftus
describe thickened arteries, suggestive of atherosclerotic disease, IL, USA
compressing the root exit zone. This description has persisted in the References
literature for decades with little solid research into this aspect of the Hosemann W (1983) Intimal fibrosis of cerebellar arteriesa contri-
disease. bution to the pathogenesis of neurovascular syndromes. Laryngol Rhinol
The authors compared the risk factors hypertension, hyperlipidemia, Otol (Stuttg) 62:522-525
and diabetes mellitus as well as cardio-ankle vascular indices (CAVI), a Izuhara M, Shioji K, Kadota S, Baba O, Takeuchi Y, Uegaito T,
measure of global atherosclerosis (Izuhara et al. 2008), and cerebral white Mutsuo S, Matsuda M (2008) Relationship of cardio-ankle vascular index
matter changes, as a measure of local atherosclerosis (Park et al. 2015), in (CAVI) to carotid and coronary arteriosclerosis. Circ J 72:1762-1767
a series of 111 patients with hemifacial spasm who had undergone mi- Jannetta PJ (1977) Observations on the etiology of trigeminal neural-
crovascular decompression, and compared these factors to the healthy gia, hemifacial spasm, acoustic nerve dysfunction and glossopharyngeal
population. They found no significant difference in rates of risk factors neuralgia. Definitive microsurgical treatment and results in 117 patients.
or degrees of local or global atherosclerosis between the hemifacial spasm Neurochirurgia 20:145-154
population and the healthy population. Arteriosclerosis is often cited as an Kondo A, Ishikawa J, Konishi T (1981) The pathogeneshis of
etiology for hemifacial spasm and this article provides a fresh perspective hemifacial spasm: characteristic changes of vasculatures in vertebro-
on this assumption (Jannetta 1977; Kondo et al. 1981). The authors con- basilar artery system. In: Sammi M, Jannetta P (eds) The cranial nerves.
firm the findings of the cadaver studies performed by Hosemann (1983). Springer, New York, pp 494-501
The authors are to be commended on the fine work in evaluating this Park JH, Kwon HM, Lee J, Kim DS, Ovbiagele B (2015) Association
important issue. While their work does not elucidate a definite pathophys- of intracranial atherosclerotic stenosis with severity of white matter
iologic etiology or risk factor for microvascular compression causing hyperintensities. Eur J Neurol 22:44-52, e42-43