6
QUANTITATIVE ULTRASONOGRAPHY OF FACIAL MUSCLES GERD FABIAN VOLK, MD, 1 NADJA WYSTUB, 1 MARTIN POHLMANN, MD, 1 MIRA FINKENSIEPER, MD, 1 HEATHER J. CHALMERS, DVM, 2 and ORLANDO GUNTINAS-LICHIUS, MD 1 1 Department of Otorhinolaryngology, Jena University Hospital, Lessingstrasse 2, D-07740 Jena, Germany 2 Department of Clinical Studies, Ontario Veterinary College, University of Guelph, Guelph, Canada Accepted 1 October 2012 ABSTRACT: Introduction: There is no standardized method for examination of facial muscles with ultrasound. The purpose of this study was to identify those facial muscles accessible for reliable identification and to provide reference data. Methods: In healthy subjects all facial muscles were screened for visibility, separation from adjacent muscles, and reliability of landmarks. Bilateral scans of reliable muscles were performed in 40 adult volunteers. Results: Six facial muscles were clearly demarcated with ultrasound. These were: frontalis, orbicularis oculi, orbicu- laris oris, depressor anguli oris, depressor labii inferioris, and mentalis muscles. Cross-sectional area and muscle thickness showed gender differences and were independently related to age for some muscles. A significant left–right side difference was only seen for the orbicularis oculi muscle in women. Conclusions: These data demonstrate the usefulness of ultraso- nography to assess facial muscles and provide reference values that can be applied in the clinical setting. Muscle Nerve 47: 878–883, 2013 Severe facial nerve lesions after Bell palsy, trauma, or tumor surgery lead to atrophy of facial (mi- metic) muscles. So far, decision making for facial nerve reconstruction is based mainly on clinical ob- servation rather than on quantitative measurement. 1 Additionally, needle electromyography (EMG), although useful diagnostically to detect facial mus- cle atrophy, is not particularly useful prognostically. 2 At present there is no quantitative tool that can directly monitor mimetic muscle regeneration after reconstructive surgery and/or directly verify facial muscle size. A tool such as ultrasound, which is pain- less and noninvasive would allow for quantitative se- rial studies which are needed to map out the time course of loss and regeneration of muscle bulk after facial nerve injury and recovery or repair. One study has been published which showed that the degree of facial muscle atrophy can be detected by magnetic resonance imaging (MRI). 3 MRI and computed tomography (CT) may also be used to monitor structural changes of facial muscles after reconstructive surgery. 4 Yet, both CT and MRI have specific drawbacks. When per- formed in a routine manner, the sectional planes acquired in CT or MRI often do not correspond to the axial symmetry of the face due to the technical difficulty in achieving perfectly symmetrical slices, and reconstructions which aim to correct these deficiencies typically have reduced spatial resolu- tion. CT studies involve radiation exposure to the patient. Both CT and MRI might have limited resource accessibility and are costly and immobile. These restrictions may explain why photography is currently the main visualization technique used routinely in the clinical setting, even for cases undergoing reconstructive surgery. In contrast, ultrasonography allows individual optimized cross-sections to be obtained quickly and in a manner that is reproducible by using rou- tinely visible landmarks. Ultrasound can provide functional and structural information at the same time, can be used for sequential follow up evalua- tion that is efficient and free of ionizing radiation, and reveals distinctive patterns in muscles affected by neuromuscular disease. 5,6 The first report on visualization of muscles of facial expression with ultrasonography was published in 1988. 7 Of inter- est, but perhaps due to technical limitations at the time, the possibilities of ultrasonography have not been explored further. In this study, we describe the process for identi- fication of those facial muscles appropriate for reli- able ultrasonographic examinations, and we report reference data of muscle cross-sectional area and thickness. METHODS Subjects. Healthy adult volunteers were recruited. Only subjects without a history of facial or head trauma, facial palsy, or any other neurological dis- order were included. None of the volunteers had a history of a hereditary neuromuscular disorder or any other congenital disorder. Initially, 10 subjects participated in the pilot study to identify the facial muscles which could be visualized clearly by ultra- sonography. Afterward, reference data for the determined facial muscles were estimated in 40 subjects. Three additional volunteers were meas- ured 10 times at different dates to determine the reliability of the method. The study was approved by the local ethics committee, and informed consent was obtained from all participants. Age, gender, body weight, body height, body mass index (BMI), and handedness of all subjects were recorded. Abbreviations: CT, computed tomography; EMG, electromyography; MRI, Magnetic resonance imaging; NA, not applicable Correspondence to: O. Guntinas-Lichius; e-mail: orlando.guntinas@med. uni-jena.de V C 2012 Wiley Periodicals, Inc. Published online 8 October 2012 in Wiley Online Library (wileyonlinelibrary. com). DOI 10.1002/mus.23693 Key words: anatomy, facial muscles, facial nerve, quantitative ultrasound, reconstructive surgery 878 Sonography of the Facial Muscles MUSCLE & NERVE June 2013

Quantitative ultrasonography of facial muscles

  • Upload
    orlando

  • View
    214

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Quantitative ultrasonography of facial muscles

QUANTITATIVE ULTRASONOGRAPHY OF FACIAL MUSCLESGERD FABIAN VOLK, MD,1 NADJA WYSTUB,1 MARTIN POHLMANN, MD,1 MIRA FINKENSIEPER, MD,1

HEATHER J. CHALMERS, DVM,2 and ORLANDO GUNTINAS-LICHIUS, MD1

1 Department of Otorhinolaryngology, Jena University Hospital, Lessingstrasse 2, D-07740 Jena, Germany2 Department of Clinical Studies, Ontario Veterinary College, University of Guelph, Guelph, Canada

Accepted 1 October 2012

ABSTRACT: Introduction: There is no standardized methodfor examination of facial muscles with ultrasound. The purposeof this study was to identify those facial muscles accessible forreliable identification and to provide reference data. Methods: Inhealthy subjects all facial muscles were screened for visibility,separation from adjacent muscles, and reliability of landmarks.Bilateral scans of reliable muscles were performed in 40 adultvolunteers. Results: Six facial muscles were clearly demarcatedwith ultrasound. These were: frontalis, orbicularis oculi, orbicu-laris oris, depressor anguli oris, depressor labii inferioris, andmentalis muscles. Cross-sectional area and muscle thicknessshowed gender differences and were independently related toage for some muscles. A significant left–right side differencewas only seen for the orbicularis oculi muscle in women.Conclusions: These data demonstrate the usefulness of ultraso-nography to assess facial muscles and provide reference valuesthat can be applied in the clinical setting.

Muscle Nerve 47: 878–883, 2013

Severe facial nerve lesions after Bell palsy, trauma,or tumor surgery lead to atrophy of facial (mi-metic) muscles. So far, decision making for facialnerve reconstruction is based mainly on clinical ob-servation rather than on quantitative measurement.1

Additionally, needle electromyography (EMG),although useful diagnostically to detect facial mus-cle atrophy, is not particularly useful prognostically.2

At present there is no quantitative tool that candirectly monitor mimetic muscle regeneration afterreconstructive surgery and/or directly verify facialmuscle size. A tool such as ultrasound, which is pain-less and noninvasive would allow for quantitative se-rial studies which are needed to map out the timecourse of loss and regeneration of muscle bulk afterfacial nerve injury and recovery or repair.

One study has been published which showedthat the degree of facial muscle atrophy can bedetected by magnetic resonance imaging (MRI).3

MRI and computed tomography (CT) may also beused to monitor structural changes of facialmuscles after reconstructive surgery.4 Yet, both CTand MRI have specific drawbacks. When per-formed in a routine manner, the sectional planesacquired in CT or MRI often do not correspond tothe axial symmetry of the face due to the technical

difficulty in achieving perfectly symmetrical slices,and reconstructions which aim to correct thesedeficiencies typically have reduced spatial resolu-tion. CT studies involve radiation exposure to thepatient. Both CT and MRI might have limitedresource accessibility and are costly and immobile.These restrictions may explain why photography iscurrently the main visualization technique usedroutinely in the clinical setting, even for casesundergoing reconstructive surgery.

In contrast, ultrasonography allows individualoptimized cross-sections to be obtained quicklyand in a manner that is reproducible by using rou-tinely visible landmarks. Ultrasound can providefunctional and structural information at the sametime, can be used for sequential follow up evalua-tion that is efficient and free of ionizing radiation,and reveals distinctive patterns in muscles affectedby neuromuscular disease.5,6 The first report onvisualization of muscles of facial expression withultrasonography was published in 1988.7 Of inter-est, but perhaps due to technical limitations at thetime, the possibilities of ultrasonography have notbeen explored further.

In this study, we describe the process for identi-fication of those facial muscles appropriate for reli-able ultrasonographic examinations, and we reportreference data of muscle cross-sectional area andthickness.

METHODS

Subjects. Healthy adult volunteers were recruited.Only subjects without a history of facial or headtrauma, facial palsy, or any other neurological dis-order were included. None of the volunteers had ahistory of a hereditary neuromuscular disorder orany other congenital disorder. Initially, 10 subjectsparticipated in the pilot study to identify the facialmuscles which could be visualized clearly by ultra-sonography. Afterward, reference data for thedetermined facial muscles were estimated in 40subjects. Three additional volunteers were meas-ured 10 times at different dates to determine thereliability of the method. The study was approvedby the local ethics committee, and informedconsent was obtained from all participants. Age,gender, body weight, body height, body mass index(BMI), and handedness of all subjects wererecorded.

Abbreviations: CT, computed tomography; EMG, electromyography;MRI, Magnetic resonance imaging; NA, not applicable

Correspondence to: O. Guntinas-Lichius; e-mail: [email protected]

VC 2012 Wiley Periodicals, Inc.Published online 8 October 2012 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/mus.23693

Key words: anatomy, facial muscles, facial nerve, quantitative ultrasound,reconstructive surgery

878 Sonography of the Facial Muscles MUSCLE & NERVE June 2013

Page 2: Quantitative ultrasonography of facial muscles

Ultrasound Equipment. The ultrasonographyexaminations were performed by 2 of the investiga-tors (N.W., G.F.V.). All measurements were per-formed using a diagnostic ultrasound system(HD11 XE, Philips, The Netherlands). Two differ-ent linear-array transducers were used: a trans-ducer with 3-12-MHz (L12-3, Philips) and anotherwith 7-15-MHz (L15-7io, Philips). After the pilotstudy, the equipment settings were fixed as follows:mechanical index - 1.5 (L12-3) and 1.4 (L15-7io);thermal index - 0.6 (L12-3 and L15-7io); and gain -75 dB using constant time gain compensation. Allequipment settings were kept constant during themeasurements.

Pilot Study: Identification of Visualizable Facial

Muscles. The pilot study was necessary to deter-mine which facial muscles were identifiable in areliable manner, to define those anatomical land-marks which allow muscle identification, and tostandardize the planes for transducer positioningto optimize imaging at each site. All subjects wereexamined in the supine position and were com-pletely relaxed. During the examination, directedactivation of facial muscles was used to confirm thecorrect position of the transducer. The transducerwas always placed perpendicular to the skin sur-face. A muscle was considered to be reliably identi-fied when it was possible to trace the muscle alongits complete anatomic course without interferencefrom other radiating adjacent muscles. Overlapwith other facial muscles with mutual anchoring isa characteristic feature of some facial muscles. Inthese areas of overlap, a distinction of individualmuscles was impossible. This circumstance wouldskew the muscle area measurement. Effort wasmade to identify at least 1 target muscle for eachof the 5 peripheral facial motor nerve branches.Several aqueous stand-off pads were tested but didnot improve muscle visualization. Finally, all meas-urements were performed only with transmissiongel but without any other coupling media. Toavoid systematic error by compressing the muscles,before acquiring a picture, the pressure betweenprobe and skin was minimized.

Quantitative Measurements. All ultrasound pictureswere stored in DICOM format and had a resolu-tion of 1280 � 1040 pixels (8-bit, 256 gray-scale lev-els). Muscle cross-sectional area and muscle thick-ness were estimated using quantification software(QLAB, Philips, The Netherlands). To determinemuscle area, the muscle borders were delimitedmanually on each picture as a region of interest.Maximal muscle thickness was determined perpen-dicular to the muscle fiber orientation.

Statistics. All statistical analyses were performedusing IBM SPSS, version 19.0. Each facial musclewas studied separately for men and women. Thenonparametric Mann-Whitney U-test was used toanalyze gender differences. The nonparametricWilcoxon test was applied to analyze left–right dif-ferences separately for men and women. Pearsoncorrelation was used to analyze the relationshipbetween age, weight, height, and the muscle meas-urements. A multivariate linear regression analysiswas performed including gender, age, and heightas independent variables Finally, to determineintra-rater reliability, the intraclass correlation(ICC) of the muscle measurements of the 3 volun-teers, measured 10 times were calculated. Therater was not blinded from 1 measurement tothe next. The interval between the measurementswas 7 days. Nominal P values of two-tailed testsare reported. The significance level was set atP < 0.05.

RESULTS

Identification of 6 Index Facial Muscles. Six facialmuscles could be delineated clearly from surround-ing connective tissue, bone, and adjacent facialmuscles (Table 1; Fig. 1). In the cranio-caudaldirection, these muscles were as follows: frontalis,orbicularis oculi, orbicularis oris, depressor angulioris, depressor labii inferioris, and mentalismuscles. The delimitation of these 6 muscles wasnot disturbed by radiation of adjacent facial mus-cle fibers. Separation of the zygomatic major mus-cle from the zygomatic minor muscle was not pos-sible in all cases. Distinct delineation of the nasalismuscle and other small facial muscles was not pos-sible. Furthermore, reproducible separation of thenasalis muscle from parts of the levator labii supe-rioris muscle was not achievable. A plane wasdefined to visualize the complete cross-sectionalarea of the depressor anguli oris, depressor labiiinferioris, and mentalis muscles. The landmarksused to locate and verify these muscles, along withthe imaging planes and the concurrent clinicalmaneuvers used to activate the muscles are listedin Table 1. The frontalis, orbicularis oculi, andorbicularis oris muscles were too large to displaythe muscle cross-section in a single image plane.Here, the muscle thickness was estimated perpen-dicular to the muscle fiber course. There were sizeand shape differences between the 2 transducersapplied (L12-3 and L15-7io). As indicated in Table1, one probe was preferred for some muscles andanother for other muscles. For muscles that werelocated on more curved portions of the face, suchas the orbicularis oculi and orbicularis orismuscles, the smaller probe (L15-7io) was

Sonography of the Facial Muscles MUSCLE & NERVE June 2013 879

Page 3: Quantitative ultrasonography of facial muscles

preferable due to its smaller footprint. Otherwise,the larger transducer (L12-3) was used.

Muscle Cross-section Area and Thickness. The 40volunteers consisted of 20 women and 20 men.The mean age was 46 6 19 years (range: 18–95years). The mean height for men was 178.6 6 7.1cm and 167.1 6 5.5 cm for women. The meanweight was 80.2 6 8.7 kg for men and 69.9 6 12.2kg for women. All subjects were of westernEuropean descent. Table 2 gives an overview of thefacial muscle sizes in women and men. The depres-sor labii inferioris muscle had the highest cross-sec-tional area, and the orbicularis oculi muscle wasthe thinnest muscle. The depressor anguli orismuscle on both sides of the face had a significantlylarger area and was thicker in men than in women.The orbicularis oculi muscle was significantly thin-ner in men than in women.

The normal ranges for the difference betweenthe left and right sides are shown in Table 3. Asignificant side-to-side difference was only seenfor the orbicularis oculi muscle in women. Here,the right muscle was thicker than the left muscle(P ¼ 0.042).

Only the depressor anguli oris muscle show acorrelation with age. Higher age was correlatedwith smaller muscle cross-sectional area on theright side (r ¼ �0.317; P ¼ 0.046) and as a trendalso on the left side (r ¼ �0.297; P ¼ 0.063). Theright frontalis muscle was thicker in older patients(r ¼ 0.337; P ¼ 0.0.034). There was a non-signifi-

cant trend on the left side (r ¼ 0.303; P ¼ 0.057).All other muscles did not correlate with age (all P> 0.05). The height of the subjects influencedsome muscles: Larger individuals had a thinnerfrontalis muscle on the right side (r ¼ �0.441; P ¼0.004) and on the left side (r ¼ �0.385; P ¼0.014). Also the left orbicularis oculi muscle wasthicker in larger individuals (r ¼ 0.417; P ¼0.007). On the right side a nonsignificant correla-tion was seen (r ¼ 0.265; P ¼ 0.098). Weight hadno influence on muscle cross-sectional area ormuscle thickness (all P > 0.05).

The multivariable linear regression analysisrevealed a significant influence of age (beta ¼�0.384; P ¼ 0.030) and gender (beta ¼ 0.546; P ¼0.012) on the cross-sectional area of the right de-pressor anguli oris muscle. The same was seen forthe left side (age: beta ¼ �0.356; P ¼ 0.052; gen-der: beta ¼ �0.484; P ¼ 0.030). In accordance, thesame independent influence was estimated forthickness of the right depressor anguli oris muscle(age: beta ¼ �0.391; P ¼ 0.027; gender: beta ¼�0.555; P ¼ 0.010) and also for the left depressoranguli oris muscle (age: beta ¼ �0.373; P ¼ 0.042;gender: beta ¼ �0.479; P ¼ 0.032). Gender was anindependent factor for the thickness of the rightorbicularis oculi muscle (beta ¼ �0.712; P ¼0.001) and the left orbicularis oculi muscle (beta¼ �0.807; P < 0.001).

The intraclass correlation coefficient (ICC)within the repetitive measurement of the 3 volun-teers was 0.93 (n ¼ 10; P < 0.001) for the right

Table 1. Identification of the 6 index mimetic muscles.

Index muscleTransducer*

(MHz) Landmarks Imaging planeManeuver for

muscle identification

Frontalis L12-3 Frontal bone, supraorbital margin Transverse position 1 cmcranial supraorbital margin

Frowning

Orbicularis oculi L15-7io Frontal process of zygomatic bone,lateral wall of orbital cavity

Transverse positionperpendicular on thefrontal process

Blinking

Orbicularis oris L15-7io Columella, philtrum ridges, upper lip Sagittal position within thephiltrum ridges

Pressing thelips together

Depressoranguli oris

L12-3 Infralabial fossa, mandibular premolarteeth, alveolar processes ofmandibular body, inferiorlabial artery, facial artery

Positioning parallel tomandibular body, movementin cranio-caudal directionalong the alveolar processesof mandibular body

Pulling the cornersof the mouthdownwards

Depressor labiiinferioris

L12-3 Infralabial fossa, mandibular premolarteeth, alveolar process of mandibularbody, inferior labial artery, facial artery

Positioning parallel tomandibular body, movementin cranio-caudal direction alongthe alveolar processes ofmandibular body

Pulling the cornersof the mouthdownwards

Mentalis muscle L12-3 Mandible, mental protuberance,contralateral muscle

Transverse position, movementfrom mental protuberancetoward mentolabial sulcusand lower lip.

Pressing thelips together

*Transducer from Philips, The Netherlands.

880 Sonography of the Facial Muscles MUSCLE & NERVE June 2013

Page 4: Quantitative ultrasonography of facial muscles

mentalis muscle and 0.83 (n ¼ 30; P < 0.001) forthe left mentalis muscle. The ICC for the depres-sor anguli oris muscle was 0.68 (P < 0.001) for theright side and 0.66 (P < 0.001) for the left side.The ICC for the depressor labii inferioris musclewas 0.72 (P < 0.001) for the right side and 0.33 (P¼ 0.007) for the left side. For the orbicularis oris

muscle, the ICC for the right side was 0.45 (P <0.001) and 0.54 for the left side (P < 0.001). TheICC for the orbicularis oculi muscle was 0.40 (P ¼0.002) for the right side and 0.59 (P < 0.001) forthe left side. For the frontalis muscle, the ICCfor the right side was 0.44 (P < 0.001) and 0.55for the left side (P < 0.001).

FIGURE 1. Seven facial muscles could be delineated clearly from surrounding connective tissue, bone, and adjacent facial muscles.

1, frontalis muscle; 2, orbicularis oculi muscle; 3, temporalis muscle; 4, orbicularis oris muscle; 5, depressor anguli oris muscle; 6, de-

pressor labii inferioris muscle; 7, mentalis muscle. A, frontal bone; B, lateral wall of orbital cavity; C, incisor; D, alveolar process of

mandibular body; E, mental protuberance; g, upper lip; h, lower lip; i, inferior labial artery. [Color figure can be viewed in the online

issue, which is available at wileyonlinelibrary.com.]

Sonography of the Facial Muscles MUSCLE & NERVE June 2013 881

Page 5: Quantitative ultrasonography of facial muscles

DISCUSSION

Quantitative muscle ultrasonography as a diagnos-tic tool has been implemented recently.8 Further-more, normative data of muscle thickness andecho intensity have been published for somemuscles of the extremities in children andadults.6,9 Such data is very helpful to estimatemuscle atrophy in neuromuscular diseases and tomonitor muscle changes in longitudinal studies.10

The accuracy of ultrasonography in estimatingmuscle size is reported to be comparable to CTand MRI in limb muscles.11,12 The combinationof EMG and ultrasonography is reported to be apowerful tool for evaluation of healthy musclefunction and for quantification of muscle degen-eration due to severe motor neuron lesions.13,14

In contrast to the large skeletal muscles, theelectrodiagnostic options to assess the facialnerve and facial muscles are very limited.15 Thisis related to the limited accessibility of the pe-ripheral facial nerve within the temporal bone.Evaluation of the nerve is only achievable alonga short distance from its exit at the stylomastoid

foramen to its branching in the parotid gland.Furthermore, the facial muscles are very thin,directly anchored to the facial skin, and some ofthem overlap, which complicates the diagnosticapproach.16 In cases of congenital facial palsy orin patients who are candidates for facial nervereconstruction surgery after longer denervationtimes, and to monitor such patients after surgery,it would be very helpful to have ultrasonographyas an additional tool to evaluate the functionalstatus of the facial muscles.

Against this background, it seems to be valua-ble that this study was able to successfully definemeasurements for 6 facial muscles representativeof all 5 peripheral motor end branches of the fa-cial nerve. The study from Balogh et al.7 held outthe possibility of a future prospective study to pro-vide standardized data for facial muscle measure-ments. This goal has been achieved in this study. Arecent study evaluating the results of cleft lip sur-gery using ultrasonography to monitor orbicularisoris muscle thickness does not provide referencedata.17

Table 2. Comparison of muscle thickness and cross-sectional area of facial muscles in men and women.*

Comparison of thickness of facial muscles Comparison of cross-sectional area

Muscle thickness (mm) Muscle cross-sectional area (mm2)

Side Men Women P-value** Men Women P-value**

Frontalis L 2.27 6 0.95 2.86 6 0.58 0.076 NA NA NAR 2.20 6 1.00 2.88 6 0.56 0.050 NA NA NA

Orbicularis oculi L 1.07 6 0.17 0.72 6 0.22 <0.0001 NA NA NAR 1.02 6 0.16 0.76 6 0.26 <0.0001 NA NA NA

Orbicularis oris L 3.54 6 0.90 3.55 6 0.65 0.543 NA NA NAR 3.49 6 0.79 3.40 6 0.74 0.507 NA NA NA

Depressor anguli oris L NA NA NA 27.42 6 7.57 22.49 6 6.39 0.037R NA NA NA 27.38 6 6.89 21.55 6 6.94 0.008

Depressor labii inferioris L NA NA NA 42.98 6 18.51 47.08 6 18.00 0.465R NA NA NA 43.85 6 19.71 46.10 6 20.38 0.705

Mentalis L NA NA NA 28.78 6 11.57 22.55 6 7.78 0.105R NA NA NA 28.96 6 9.47 23.74 6 9.94 0.105

*Data represented as mean 6 standard deviation.**P-values < 0.05 in bold; NA ¼ not applicable.

Table 3. Normal range* of right – left differences in thickness and cross-sectional area of facial muscles in men and women.

Muscle thickness (mm) Muscle cross-sectional area (mm2)

Men P-value** Women P-value** Men P-value** Women P-value**

Frontalis -0.47 to 0.34 0.219 -0.46 to 0.42 0.601 NA NA NA NAOrbicularis oculi 0.73 to 1.41 0.087 0.37 to 1.42 0.286 NA NA NA NAOrbicularis oris -1.20 to 1.06 0.355 -0.99 to 0.77 0.042 NA NA NA NADepressor

anguli orisNA NA NA NA -7.50 to 7.41 1.00 -9.06 to 8.07 0.526

Depressorlabii inferioris

NA NA NA NA -14.49 to 16.23 0.709 -15.38 to 15.28 0.643

Mentalis NA NA NA NA -9.12 to 9.47 0.526 -8.58 to 9.95 0.391

*Data represent -2 SD to þ2 SD from the mean right–left difference. NA, not applicable.**Difference between right and left side; P-values < 0.05 in bold.

882 Sonography of the Facial Muscles MUSCLE & NERVE June 2013

Page 6: Quantitative ultrasonography of facial muscles

According to the methodology of a previousstudy in adult skeletal muscles of the extremities,6

this study provides reference data of adult facialmuscles and a reproducible technique for perform-ing facial ultrasonography. As some of the facialmuscles are small and have complex shapes, it wasuseful in these muscles to measure cross-sectionalarea instead of muscle thickness. In accordancewith the results in larger skeletal muscles, musclethickness and area of facial muscles is also gender-and age-dependent, at least for some facialmuscles. Gender and age have been revealed to beindependent factors for muscle thickness for somefacial muscles in multivariate analysis. Further-more, height but not weight was related to themuscle measurements in the univariate analysis.The gender and age effect were not as different asin larger limb muscles, and there was no differ-ence for some facial muscles. This might beexplained by the function of facial muscles. Mi-metic muscles certainly are exposed to a smallerphysical burden than the functions of larger limbmuscles. This study is limited by its sample size. Tofurther explore the age effect on facial muscles,larger cohorts with age groups would be needed.

The results of the ICC show a moderate or bet-ter agreement in 8 of 12 muscles. This demon-strates the value of the method to detect even thedifferences in healthy probands. It remains to beseen whether these measurements will be of useclinically in differentiating diseased patients fromnormal subjects; this will be the subject of study inthe future.

The addition of routine facial ultrasonographyto the available clinical tests is anticipated to be ofsubstantial value both in diagnosis and monitoringof neuromuscular disorders of the face. The mini-mally invasive nature, easy access, cost effective-ness, and reproducibility of ultrasonography makeit an ideal tool offering many advantages over CTand MRI. Moving forward, other static parameters,including echo intensity and dynamic muscle ultra-sonography could further expand the utility of thisexciting modality.18 A clinical trial in patients with

facial palsy using the measurement parameters pre-sented in this study is already ongoing.

Informed written consent was obtained for publication of thepatient images. The authors issued the whole manuscript. Allauthors have read and approved the final manuscript. The authorsdeclare that they have no competing interests.

REFERENCES

1. Guntinas-Lichius O, Streppel M, Stennert E. Postoperative functionalevaluation of different reanimation techniques for facial nerverepair. Am J Surg 2006;191:61–67.

2. Volk GF, Pantel M, Guntinas-Lichius O. Modern concepts in facialnerve reconstruction. Head Face Med 2011;6:25.

3. Kaylie DM, Wax MK, Weissman JL. Preoperative facial muscle imag-ing predicts final facial function after facial nerve grafting. AJNR AmJ Neuroradiol 2003;24:326–330.

4. Gargiulo P, Klingner CM, Friogeirsson EA, Burmeister HP, Volk GF,Guntinas Lichius O. Side differences im MRI-scans in facial palsy: 3-D modelling, segmentation and grey value analysis. Rome: Septem-ber 12–14, 2011.

5. Pillen S, Scholten RR, Zwarts MJ, Verrips A. Quantitative skeletalmuscle ultrasonography in children with suspected neuromusculardisease. Muscle Nerve 2003;27:699–705.

6. Arts IM, Pillen S, Schelhaas HJ, Overeem S, Zwarts MJ. Normal val-ues for quantitative muscle ultrasonography in adults. Muscle Nerve2010;41:32–41.

7. Balogh B, Fruhwald F, Millesi W, Millesi H, Firbas W. Sonoanatomyof the muscles of facial expression. Surg Radiol Anat 1988;10:101–106.

8. Walker FO, Cartwright MS, Wiesler ER, Caress J. Ultrasound of nerveand muscle. Clin Neurophysiol 2004;115:495–507.

9. Scholten RR, Pillen S, Verrips A, Zwarts MJ. Quantitative ultrasonog-raphy of skeletal muscles in children: normal values. Muscle Nerve2003;27:693–698.

10. Arts IM, Overeem S, Pillen S, Schelhaas HJ, Zwarts MJ. Musclechanges in amyotrophic lateral sclerosis: a longitudinal ultrasonogra-phy study. Clin Neurophysiol 2011;122:623–628.

11. Sipila S, Suominen H. Muscle ultrasonography and computed to-mography in elderly trained and untrained women. Muscle Nerve1993;16:294–300.

12. Walton JM, Roberts N, Whitehouse GH. Measurement of the quadri-ceps femoris muscle using magnetic resonance and ultrasound imag-ing. Br J Sports Med 1997;31:59–64.

13. Bayrak IK, Bayrak AO, Tilki HE, Nural MS, Sunter T. Ultrasonogra-phy in carpal tunnel syndrome: comparison with electrophysiologicalstage and motor unit number estimate. Muscle Nerve 2007;35:344–348.

14. Simoneau EM, Longo S, Seynnes OR, Narici MV. Human muscle fas-cicle behavior in agonist and antagonist isometric contractions. Mus-cle Nerve 2012;45:92–99.

15. Grosheva M, Wittekindt C, Guntinas-Lichius O. Prognostic value ofelectroneurography and electromyography in facial palsy. Laryngo-scope 2008;118:394–397.

16. Wilkinson C. Facial reconstruction–anatomical art or artistic anat-omy? J Anat 2010;216:235–250.

17. Ridgway EB, Estroff JA, Mulliken JB. Thickness of orbicularis orismuscle in unilateral cleft lip: before and after labial adhesion. J Cra-niofacial Surg 2011;22:1822–1826.

18. van Alfen N, Nienhuis M, Zwarts MJ, Pillen S. Detection of fibrilla-tions using muscle ultrasound: diagnostic accuracy and identificationof pitfalls. Muscle Nerve 2011;43:178–182.

Sonography of the Facial Muscles MUSCLE & NERVE June 2013 883