Orbicularis Oculi Muscle Immunohistochemical, Metabolic, and Morphometric Differences in Affected and Nonaffected Sides in Hemifacial Spasm vs Healthy Subjects

Background: Subtle morphological alterations have been reported even in the nonaffected side of the orbicularis oculi muscle in patients with hemifacial spasm. However, no previous study assessed immunohistochemical, metabolic, and morphometric alterations in orbicularis oculi muscle fibers in affected and nonaffected sides in patients with this condition, compared with samples obtained from healthy subjects. The purpose of this study is to objectively assess and compare orbicularis oculi muscle (OOM) samples of hemifacial spasm affected and nonaffected sides and healthy subjects. Methods: Orbicularis oculi samples from 8 patients with hemifacial spasm who had not been previously treated and 6 healthy subjects were prepared using hematoxylin and eosin, nicotinamide adenine dinucleotide tetrazolium reductase, cytochrome oxidase, succinate dehydrogenase, Gomori staining, and monoclonal antibodies against myosin slow and myosin fast. A digital image analysis software was used for objective analysis. Results: OOM fiber area was significantly greater in both affected ( P = 0.0379) and nonaffected sides ( P = 0.0012) of HFS samples when compared with control subjects' fibers. A significantly greater number of oxidative fibers were observed in both affected and nonaffected sides of patients with HFS when compared with control subjects ( P < 0.001 for both). A significantly greater percentage of slow fibers was observed in the affected side of HFS patients ( P = 0.0012) compared with control subjects. Conclusions: This study's findings suggest that repeated contractions might lead to OOM fiber hypertrophy, increased mitochondrial metabolism, and possible conversion of fast-twitch orbicularis oculi muscle fibers into slow-twitch fibers in patients with HFS. Alterations were observed in affected and nonaffected sides, confirming initial findings that the nonaffected side is not normal in this unique condition.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Orbicularis Oculi Muscle Immunohistochemical, Metabolic, and Morphometric Differences in Affected and Nonaffected Sides in Hemifacial Spasm vs Healthy Subjects Tammy H. Osaki, MD, PhD, Gustavo R. Gameiro, MD, Midori H. Osaki, MD, PhD, Teissy Osaki, MD, PhD, Eliene D. Campos, PhD, Rubens Belfort Jr, MD, PhD, Suely K. N. Marie, MD, PhD Background: Subtle morphological alterations have been reported even in the nonaffected side of the orbicularis oculi muscle in patients with hemifacial spasm. However, no previous study assessed immunohistochemical, metabolic, and morphometric alterations in orbicularis oculi muscle fibers in affected and nonaffected sides in patients with this condition, compared with samples obtained from healthy subjects. The purpose of this study is to objectively assess and compare orbicularis oculi muscle (OOM) samples of hemifacial spasm affected and nonaffected sides and healthy subjects. Methods: Orbicularis oculi samples from 8 patients with hemifacial spasm who had not been previously treated and 6 healthy subjects were prepared using hematoxylin and eosin, nicotinamide adenine dinucleotide tetrazolium reductase, cytochrome oxidase, succinate dehydrogenase, Gomori staining, and monoclonal antibodies against myosin slow and myosin fast. A digital image analysis software was used for objective analysis. Results: OOM fiber area was significantly greater in both affected (P = 0.0379) and nonaffected sides (P = 0.0012) of HFS samples when compared with control subjects’ fibers. A significantly greater number of oxidative fibers were observed in both affected and nonaffected sides of patients with HFS when compared with control subjects (P , 0.001 for both). A significantly greater percentage of slow fibers Department of Ophthalmology and Visual Sciences (THO, GRG, MHO, TO, RB), Paulista School of Medicine/EPM, Federal University of São Paulo/UNIFESP, São Paulo, Brazil; and Laboratory of Molecular and Cellular Biology (GRG, EDC, SKNM), Department of Neurology, University of São Paulo/USP, São Paulo, Brazil. Supported in part by CAPES Foundation (GRG). The authors report no conflicts of interest. T. H. Osaki and G. R. Gameiro contributed equally to the manuscript. Address correspondence to Tammy H. Osaki, MD, PhD, Department of Ophthalmology and Visual Sciences, Federal University of São Paulo, Eye & Eyelid Experts, 661, João Moura Street, St 84, São Paulo, SP 05412-001, Brazil; E-mail: [e-mail] 410 was observed in the affected side of HFS patients (P = 0.0012) compared with control subjects. Conclusions: This study’s findings suggest that repeated contractions might lead to OOM fiber hypertrophy, increased mitochondrial metabolism, and possible conversion of fasttwitch orbicularis oculi muscle fibers into slow-twitch fibers in patients with HFS. Alterations were observed in affected and nonaffected sides, confirming initial findings that the nonaffected side is not normal in this unique condition. Journal of Neuro-Ophthalmology 2023;43:410–416 doi: 10.1097/WNO.0000000000001770 © 2022 by North American Neuro-Ophthalmology Society H emifacial spasm (HFS) is a movement disorder composed of involuntary, intermittent, clonic, and tonic spasmodic contractions of the muscles innervated by the ipsilateral facial nerve (1,2). In most cases, it is associated with the compression of the facial nerve at the exit zone by the anterior inferior cerebellar artery (3,4). Spasms usually initiate in the orbicularis oculi muscle before affecting the midface and lower third of the face over months to years. Although bilateral cases have been described, they are rare, and most cases are unilateral (1). HFS is a peculiar movement disorder, as one hemiface is phenotypically normal without abnormal contractions. Firstline treatment consists of optimized doses of botulinum toxin-A injections (BTX-A) on the affected muscles. Muscles that are typically injected are orbicularis oculi, major zygomatic, minor zygomatic, risorius, and depressor anguli oris muscles. Little is known regarding alterations in the orbicularis oculi muscle (OOM) fiber in affected and nonaffected sides in this condition. A previous study demonstrated morphological alterations in the muscle fiber on both the affected and nonaffected sides; endomysial and perimysial connective tissue Osaki et al: J Neuro-Ophthalmol 2023; 43: 410-416 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution areas were significantly increased on the affected side when compared with normal controls; cell density was observed to be significantly reduced on the affected side and also on the nonaffected side when compared with normal controls (5). These morphological alterations may be related to some degree of nonclinical evident muscular contraction or some type of muscle homeostatic disturbance even in the nonaffected side (5). Nevertheless, to the best of our knowledge, no previous studies have assessed metabolic, immunohistochemical, and morphometric changes in OOM fibers in patients with hemifacial spasm. The purpose of this study was to perform a deeper analysis of OOM samples obtained from patients with HFS and healthy subjects. Samples were obtained during cosmetic blepharoplasty to investigate differences in OOM fiber type distribution, mitochondrial metabolism, and morphometry on the affected and nonaffected sides in patients with HFS, in comparison with healthy subjects. METHODS This study was approved by the Federal University of S. Paulo Institutional Review Board (IRB). Written informed consents were obtained from all the participants. Twentyfive samples of OOM from 14 subjects who underwent cosmetic upper lid blepharoplasty were evaluated. All subjects were treated in accordance with the tenets of the Declaration of Helsinki and expressed the desire to undergo upper blepharoplasty for cosmetic purposes. Eight patients with moderate and moderate-to-severe HFS, according to the Jankovic Rating Scale (6), who had not been previously treated with botulinum toxin injections, and 6 healthy subjects were included in the study. All patients who underwent the surgical procedure underwent preoperative standard examinations to identify any potential surgical risk. Exclusion criteria included (1) previous treatment with BTX-A injections in the affected muscles, (2) patients with bilateral HFS, (3) prior eyelid surgery, (4) no evident dermatochalasis, (5) lagophthalmos, (6) neuromuscular junction and peripheral neuropathic diseases, (7) local hemostasis disorders, (8) high systemic surgical risk, and (9) patients who did not wish to undergo upper eyelid blepharoplasty before receiving BTX-A applications on the affected hemiface. All patients underwent cosmetically associated upper blepharoplasty, which was associated with the excision of a thin strip of preseptal orbicularis oculi muscle. All patients with HFS underwent surgery before receiving botulinum toxin injections. These patients were scheduled to undergo BTX-A applications in the affected muscles 30 days after upper blepharoplasty. Surgical procedures were performed in the operating room by a single surgeon, following standard aseptic Osaki et al: J Neuro-Ophthalmol 2023; 43: 410-416 conditions and local anesthesia using 1.0 mL of 2% lidocaine with epinephrine (1: 100,000). An upper eyelid incision was made using a #15 blade, and meticulous dissection of the orbicularis from the overlying skin was performed. After skin-only excision, a thin strip of preseptal orbicularis oculi muscle obtained from all cases was snap frozen in liquid nitrogen and stored at 280°C until further analysis. The skin was closed with 6-0 interrupted nylon sutures. Sutures were removed on the seventh postoperative day, and all patients were re-evaluated 1 month after the procedure. Serial transverse sections of 6 mm thickness were performed in all samples in a cryostat at 225°C and subsequently stained with mitochondrial stains: nicotinamide adenine dinucleotide tetrazolium reductase (NADH-TR), cytochrome oxidase (COX), and succinate dehydrogenase (SDH), beyond modified Gomori (GOM) and hematoxylin & eosin (HE) stainings. Immunohistochemistry (IHC) was also performed using monoclonal antibodies against Myosin Slow and Myosin Fast (Novocastra, Leica, Newcastle, United Kingdom). The dilution of antibodies used was 1:100 (1 mL of antibody in 99 mL of diluent). Morphological characterization was performed through subjective observation in optic microscopy followed by objective analysis (digital high-definition photographs and Image J software). Subjective analysis-wise, digital images of the most representative findings were obtained. Objective analysis-wise, photographs of the histological slides under the ·400 power field were taken by a single operator. The area containing the highest number of cells was standardized for analysis. Digital photographs were taken using the same Nikon image retrieval system. Open-source ImageJ version 1.53e, developed by the National Institute of Health (NIH), is a world-renowned powerful and effective software for image analysis. Endomysial and perimysial connective tissue areas, obtained using the “color threshold” tool in Image J, were recorded for the affected and nonaffected sides in the HFS and control groups in the Gomori slides by a blinded researcher. After manually setting the threshold that best fit all the connective tissue of each slide, the “measure” tool was used, and the percentage of the area was acquired (Fig. 1). COX, NADH, and SDH stainings were used to assess mitochondrial metabolism. NADH and SDH analysis was qualitative, performed by 2 investigators. High-definition digital photographs for COX were converted to 8-bit type (256 shades of gray between the white and black). All the images were photographed on the same day under the same light conditions by the same operator to compare the staining of different images. Moreover, the COX staining was from the same batch. The closer the value to 0, the darker the image; likewise, the closer the value gets to 255, 411 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 1. Representative objective analysis of the total muscle fiber area and connective tissue area. Left column. Orbicularis oculi muscle sample slide (top: hematoxylin & eosin; bottom: modified Trichrome Gomori staining). Center column. Correspondent total muscle fiber area after manually setting the threshold using Image J. Right column. Correspondent connective tissue area after Image J analysis. Magnification: ·400. the whiter the image is. The fiber area was manually selected using the “threshold” tool, and after that, the “average” pixel value was obtained and computed for posterior comparison. The Mann–Whitney U test was used to compare (1) the control group vs. the affected side of the HFS group and (2) the control group vs the nonaffected side. The Wilcoxonmatched pairs signed-rank test was used to compare (3) nonaffected vs. affected sides in patients with HFS. Statistical analyses were performed with GraphPad Prism 9.0 (GraphPad Software, Inc, San Diego, CA). P-values less than 0.05 were considered statistically significant. RESULTS The mean age was 60.25 ± 7.59 years (47–70 years) in the HFS group and 68.6 ± 6.11 years (64–79 years) in the control group. Three patients from the hemifacial spasm group were men; all other subjects included were women. The mean disease duration was 7.5 ± 4.57 years. All patients reported satisfaction with the cosmetic results 1 month postoperatively. During the follow-up period, neither group observed complications such as lagophthalmos, blepharoptosis, infection, or wound dehiscence. Morphological and Morphometric Analysis Regarding morphometric analysis, the OOM fiber area was significantly increased in both affected (12.64 ± 3.10 mm2; P = 0.0379) and nonaffected sides (13.46 ± 1.59 mm2; P = 0.012) HFS samples, when compared with 412 control subjects fibers (9.03 ± 1.76 mm2). There was no statistically significant difference between the mean OOM fiber area in affected vs unaffected sides of patients with HFS (P = 0.578). A significant reduction in cell number was observed in both affected (42.57 ± 18.74; P = 0.0286) and nonaffected (41.71 ± 7.56; P = 0.023) sides compared with healthy controls (67.14 ± 15.26). There was no statistically significant difference between muscle fiber count in affected vs unaffected sides of patients with HFS (P = 0.984). Endomysial and perimysial connective tissue area percentage was assessed with modified Trichrome Gomori staining, and a significant increase in the affected side (28.76 ± 8.11) when compared with healthy subjects samples (14.12 ± 5.38; P = 0.0041) was observed. A statistically significant difference in the connective tissue area in the affected vs. nonaffected (20.07 ± 9.04; P = 0.04) sides of patients with HFS was observed. Figures 2 and 3A and B show differences in OOM fiber number, area, and connective tissue area in affected and nonaffected sides in patients with HFS compared with healthy subjects. Mitochondrial Metabolism Analysis The mean histogram gray pixel value, assessed with COX staining, was significantly lower (darker shades) in HFS samples, as displayed in Figure 3C. That means that there are more oxidative fibers in the OOM of patients with HFS on both affected and nonaffected sides (P , 0.001** for both). NADH and SDH stainings analysis also showed more oxidative fibers in HFS samples. Osaki et al: J Neuro-Ophthalmol 2023; 43: 410-416 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 2. Morphometrical aspects of orbicularis oculi muscle samples of normal control subjects and patients with hemifacial spasm (affected and nonaffected sides). Left column. Orbicularis oculi muscle samples from healthy subjects: preserved morphology. Central column. Orbicularis oculi muscle samples from the nonaffected side of patients with hemifacial spasm; lower cell density and larger fiber area can be observed when compared with the normal control samples. Right column. Orbicularis oculi muscle samples from the affected side of patients with hemifacial spasm; a reduction in the muscle fiber density and an increase in both the cell area and the endomysial and perimysial connective tissue portions can be observed when compared with control samples (Top: hematoxylin and eosin; bottom: modified Trichrome Gomori staining; ·400). Figure 4 presents a macro panel of those findings. Decoupling of respiratory chain reactions with COX, NADH, and SDH stainings was also observed in the affected and nonaffected sides of patients with HFS. Immunohistochemical Analysis A significant increase in the percentage of slow fibers (22.62 ± 2.21) was observed in the affected side of patients with HFS when compared with control subjects (13.85 ± 2.31; P = 0.0012) and the nonaffected side (18.42 ± 4.24; P = 0.0411). Figure 3D shows differences in OOM slow fiber percentage in affected and nonaffected sides in patients with HFS compared with healthy subjects. Table 1 summarizes all studied parameters and comparisons between groups. DISCUSSION Our study comprehensively assessed immunohistochemical, metabolic, and morphometric differences in both affected and nonaffected sides of orbicularis oculi muscle in patients with HFS and in healthy subjects so that we could better understand muscular fiber changes associated with this unique condition. OOM fiber size, distribution of endomysial and perimysial connective tissue, study of mitochondrial metabolism, and distribution of slow fibers, assessed with Osaki et al: J Neuro-Ophthalmol 2023; 43: 410-416 immunohistochemistry, were studied in patients with HFS (affected and nonaffected sides) and healthy subjects. As observed in a previous research from our group, the findings of this study corroborate that the OOM fiber of the nonaffected side in patients with HFS is not normal, although there were no signs of increased muscular contraction on the nonaffected side in any of the patients with HFS included in this study. Regarding OOM muscle fiber density and area, a significant reduction in cell number and hypertrophy of the muscle fiber were observed in both affected and nonaffected sides compared with healthy controls. We can hypothesize that the nonaffected side is also subjected to chronic stress, perhaps presenting some degree of nonclinically evident muscular contraction or some type of muscle homeostasis disturbance, leading to cell hypertrophy. However, this degree of contraction would not be substantial enough to lead to muscle degeneration, as observed on the affected side, in which a significantly increased connective tissue area was observed compared with healthy subjects. Immunohistochemical analysis showed that a significantly greater percentage of slow fibers was observed on the affected side when compared with the nonaffected side and with control samples. In addition, the metabolic analysis showed that OOM fibers from both affected and nonaffected sides of patients with HFS were more oxidative 413 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 3. A. Differences in orbicularis oculi muscle connective tissue area in healthy subjects, affected and nonaffected sides of patients with hemifacial spasm. *P , 0.05; **P , 0.01. B. Differences in orbicularis oculi muscle fiber area in healthy subjects, affected and nonaffected sides of patients with hemifacial spasm. *P , 0.05; **P , 0.01. C. Differences in the amount of oxidative fibers, assessed with COX (cytochrome oxidase) using the histogram gray pixel value, in healthy subjects, affected and nonaffected sides of patients with hemifacial spasm. *P , 0.05; **P , 0.01. D. Differences in orbicularis oculi muscle slow fibers percentage in healthy subjects, affected and nonaffected sides of patients with hemifacial spasm. AFT, affected side; H&E, hematoxylin and eosin; NAFT, nonaffected side. *P , 0.05; **P , 0.01. than the control group. It is well established that Type I muscle fibers are usually slow, smaller, and more oxidative. On the other hand, Type II fibers are larger and faster and can be subdivided into 2 subtypes: IIa and IIb. Type IIb muscle fibers are the largest, with high power of contraction, low endurance, hence glycolytic, and less oxidative. Type IIa fibers are between those extremes (7,8). Our findings may suggest that the conversion of some of the glycolytic muscle fibers (IIb) to oxidative fibers (IIa) or even fibers Type I, increasing the oxidative stain, may occur in OOM fibers in patients with HFS after years of chronic contractions. This conversion could explain the higher percentage of slow and more oxidative fibers observed in the HFS groups. Although fiber type transformation is controversial, some studies support the idea of that conversion in runners and space exercises investigations (8,9). Furthermore, previous studies have evaluated eyelid kinematics in patients with HFS (10,11), and it has been shown that eyelid closing velocity is lower on the affected side of patients with HFS (10). Conversion of fast-twitch orbicularis oculi muscle fibers into slow-twitch fibers fiber in these patients could explain this reduced velocity on the affected side. 414 Cytochrome C oxidase or complex IV is located in the inner mitochondrial membrane and is the last enzyme of the electron transport chain. Reduced NADH is oxidized by Complex I in the mitochondria electron transport chain to produce at the end 2.5 adenosine triphosphate. Finally, SDH is responsible for oxidizing succinate to fumarate in the citric acid cycle (Krebs cycle). Hence, those stains show fibers with the preferred oxidative phosphorylation pathway in the mitochondria (9). The current study shows that the OOM fibers from controls present fewer oxidative fibers than the hemifacial spasm groups. Notwithstanding, one cannot say that it means that those oxidative fibers are more efficient as it was observed uncoupling of the respiratory chain enzymes even on the nonaffected side of HFS. Further studies assessing gene expression profiles of those muscle samples might provide clues concerning signaling pathways involved in the initial muscular tissue modification. Digital image analysis software (Image J) has already been described as valuable in demonstrating significant differences between normal controls and affected and nonaffected sides of patients with HFS and between affected and nonaffected sides (5). In this study, ImageJ Osaki et al: J Neuro-Ophthalmol 2023; 43: 410-416 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 4. Representative metabolic panel (NADH, SDH, and COX) of orbicularis oculi muscle samples from normal subjects, affected and nonaffected sides of patients with hemifacial spasm. Both the affected and nonaffected sides of patients with HFS have more oxidative fibers (observed as darker shades) than healthy subjects. COX, cytochrome oxidase; HFS, hemifacial spasm; NADH, nicotinamide adenine dinucleotide; SDH, succinate dehydrogenase. Magnification: ·400. also permitted to objectively assess the mean pixel value in COX staining and the area of OOM fibers. Limitations of this study include the size sample. Nevertheless, due to the low prevalence of disease-related literature reports (7.4–14.5/100,000 individuals) (1) and also to the inclusion criterion in which only nonpreviously treated patients were assessed, our comparative study sample, including 8 patients, can be considered relevant and resulted in statistically significant results. Although the mean ages varied between hemifacial spasm and control groups, it has been demonstrated that no agerelated histopathological changes were observed in orbicularis oculi samples (12), and similar percentages of COX-negative and “ragged-red” fibers were observed over 40 years old (13). CONCLUSIONS In conclusion, this study adds relevant findings regarding alterations in OOM fibers in patients with hemifacial Osaki et al: J Neuro-Ophthalmol 2023; 43: 410-416 spasm. Our results suggest that repeated orbicularis oculi contractions might lead to fiber hypertrophy, increased mitochondrial metabolism, and possible conversion of fast-twitch muscle fibers into slow-twitch fibers in patients with HFS. Alterations were observed in both affected and nonaffected sides, confirming initial findings that the nonaffected side is not normal in this unique condition. The alterations may be related to some degree of nonclinical evident muscular contraction or some type of muscle homeostatic disturbance even in the nonaffected side. One may also hypothesize that a supranuclear neurological alteration is present in this facial movement disorder, not only a peripheric alteration. Perhaps future studies assessing evoked potential may help to elucidate this hypothesis. The main clinical implication of our study is that due to the findings showing that the unaffected side is not normal, in future clinical studies assessing patients with HFS, the nonaffected side may not be considered as 415 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. 416 Slow Fibers (%) Mean ± SD 13.85 ± 2.31 18.42 ± 4.24 22.62 ± 2.21 P = 0.0012* P = 0.0734* P = 0.0411† Mean Histogram Value Mean ± SD 56.40 ± 5.00 42.69 ± 9.52 43.39 ± 12.48 P , 0.001* P , 0.001* P = 0.5614† control. Further quantification of the mitochondrial DNA and novel techniques such as real-time sequencing technology may enhance our comprehension of this disease. STATEMENT OF AUTHORSHIP Conception and design: T. H. Osaki, M. H. Osaki, S. K. N. Marie; Acquisition of data: T. H. Osaki, G. Rosa Gameiro, M. H. Osaki, T. Osaki, E. D. Campos; Analysis and interpretation of data: T. H. Osaki, G. Rosa Gameiro, M. H. Osaki, T. Osaki, R. Belfort Jr, S. K. N. Marie. Drafting the manuscript: T. H. Osaki, G. Gameiro; Revising the manuscript for intellectual content: M. H. Osaki, T. Osaki, E. D. Campos, R. Belfort Jr, S. K. N. Marie. Final approval of the completed manuscript: T. H. Osaki, G. Rosa Gameiro, M. H. Osaki, T. Osaki, E. D. Campos, R. Belfort Jr, S. K. N. Marie. 9.03 ± 1.76 13.46 ± 1.59 12.64 ± 3.10 P = 0.0379* P = 0.0012* P = 0.5781† P-values , 0.05 are highlighted in bold. *Statistic comparison between groups: Mann–Whitney. † Statistic comparison between groups: Wilcoxon-matched pairs. COX, cytochrome oxidase; HFS, hemifacial spasm. 67.14 ± 15.26 41.71 ± 7.56 42.57 ± 18.74 P = 0.0286* P = 0.0023* P = 0.9844† 14.12 ± 5.38 20.07 ± 9.04 28.76 ± 8.11 P = 0.0041* P = 0.1649* P = 0.0469† Controls Nonaffected HFS side Affected HFS side Affected HFS side Controls vs Nonaffected HFS side HFS affected vs nonaffected side Connective Tissue Area (%) Mean ± SD Fiber Number Mean ± SD Fiber Area (mm2) Mean ± SD REFERENCES Case TABLE 1. Connective tissue area, fiber number and area, mean histogram value (COX), and percentage of slow fibers in each group Original Contribution 1. Ross AH, Elston JS, Marion MH, Malhotra R. Review and update of involuntary facial movement disorders presenting in the ophthalmological setting. Surv Ophthalmol. 2011;56, 54–67. 2. Chaudhry N, Srivastava A, Joshi L. Hemifacial spasm: the past, present and future. J Neurol Sci. 2015;356, 27–31. 3. Girard N, Poncet M, Cases F, Tallon Y, Chays A, MartinBouyer P, Magnan J, Raybaud C. Three-dimensional MR of hemifacial spasm with surgical correlation. Neuroradiol. 1997;39:46–51. 4. Tan EK, Chan LL, Lim SH, Lim WE, Khoo JB, Tan KP. Role of magnetic resonance imaging and magnetic resonance angiography in patients with hemifacial spasm. Ann Acad Med Singap. 1999;28:169–173. 5. Osaki MH, Osaki TH, Osaki T, Gameiro GR, Belfort R Jr, Marie SKN. Orbicularis oculi morphological alterations in affected and nonaffected sides in hemifacial spasm. J Neuroophthalmol. 2020;40:193–197. 6. Jankovic J, Kenney C, Grafe S, Goertelmeyer R, Comes G. Relationship between various clinical outcome assessments in patients with blepharospasm. Mov Disord. 2009;24, 407–413. 7. Brooke MH, Kaiser KK. Muscle fiber types: how many and what kind? Arch Neurol. 1970;23:369–379. 8. Scott W, Stevens J, Binder–Macleod SA, Human skeletal muscle fiber type classifications. Phys Ther. 2001;81:1810– 1816. 9. Lammens M, ter Laak H. Contribution of histopathological examination to the diagnosis of OXPHOS disorders. In: Smeitink JAM, Sengers RCA, Trijbels JMF, Eds Oxidative Phosphorylation in Health and Disease. Boston, MA: Medical Intelligence Unit. Springer, 2004. doi: 10.1007/0-387-26992-4_4. 10. Manning KA, Evinger C, Sibony PA. Eyelid movements before and after botulinum therapy in patients with lid spasm. Ann Neurol. 1990;28:653–660. 11. Osaki MH, Osaki TH, Garcia DM, Osaki T, Gameiro GR, Belfort R Jr, Cruz AAV. Analysis of blink activity and anomalous eyelid movements in patients with hemifacial spasm. Graefes Arch Clin Exp Ophthalmol. 2020;258:669–674. 12. Pottier F, El-Shazly NZ, El-Shazly AE. Aging of orbicularis oculi anatomophysiologic consideration in upper blepharoplasty. Arch Facial Plast Surg. 2008;10:346–349. 13. Mckelvie P, Satchi K, McNab AA, Kennedy P. Orbicularis oculi: morphological changes mimicking mitochondrial cytopathy in a series of control normal muscles. Clin Exp Ophthalmol. 2012;40:497–502. Osaki et al: J Neuro-Ophthalmol 2023; 43: 410-416 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.