Clinical manifestations

Presentation and course

Before the development of sustained closure of the eyelids, about a third of the patients reported an increased frequency of blinking, suggesting that blepharospasm may be due to impairment of mechanisms associated with normal blinking. One study showed that blinking in normal individuals is influenced by various behavioral tasks; the blink rate decreases by 74% during reading and increases by 100% during conversation (17). Although healthy individuals blink more frequently during conversation than at rest, in patients with blepharospasm, the pattern is reversed, suggesting that conversation may act as geste antagonistique (18). Although the reasons for spontaneous eyeblinks every few seconds, other than for ocular lubrication, are not well understood, there is some evidence that eyeblinks are actively involved in the release of attention (167). This “mental break” during an eyeblink, lasting only a second or so, is apparently sufficient for attention to be fully restored.

The blink rates are increased in patients with blepharospasm in both conditions, at rest (median 55 blinks per minute) and during conversation (median 31.5), as compared to healthy controls (17 vs. 26, respectively) (18). This is in contrast to a more variable blink rate in patients with Parkinson disease “off” medications (1.1 to 31.6 per minute, mean 11.5 vs. 1.7 to 17.2; mean 8.7 in healthy controls) (130). The increased blinking that precedes blepharospasm is commonly associated with a feeling of irritation in the eyes and photophobia (better termed “photosensitivity” or “photodynia,” or “photo-oculodynia”) (15; 51). Increased frequency of blinking usually progresses to clonic and later tonic (sustained) contractions of the orbicularis oculi, leading to forceful closure of eyelids, often associated with involvement of the corrugator and procerus muscles and compensatory contractions of the frontalis muscles. Although increased blinking often precedes the onset of blepharospasm, high blink rates found in some individuals apparently have a different mechanism than the more sustained orbicularis oculi contractions seen in patients with blepharospasm (31). The increased R2 response is an excellent biomarker of blepharospasm.

Spasms associated with the contraction of the orbital portion of the orbicularis oculi induce lowering of the eyebrow beneath the superior orbital margin (“Charcot sign”). Up to 20% of patients have unilateral involvement at the onset, but the opposite eye becomes involved later in all patients. In contrast to hemifacial spasm, which remains unilateral, the eyebrow in patients with blepharospasm is usually lowered during the spasms (“Charcot sign”), whereas in patients with hemifacial spasm, the ipsilateral eyebrow is often elevated (“Other Babinski sign”) (199; 222). Apraxia of eyelid opening is often confused with blepharospasm, although both conditions can coexist, particularly in the setting of parkinsonism, such as Parkinson disease, progressive supranuclear palsy, multiple system atrophy, or other parkinsonian or neurologic disorders, including deep brain stimulation (11). The condition can be defined as difficulty in voluntarily opening the eyes in the absence of visible contraction of the orbicularis oculi muscles but is often accompanied by voluntary contractions of the frontalis muscles in an attempt to open the eyes. The mechanism of apraxia of eyelid opening has not been well understood and has been attributed to levator palpebrae inhibition, contractions of pretarsal orbicularis oculi, parkinsonian freezing, and other mechanisms.

Blepharospasm is seldom an isolated condition. This form of dystonia is often associated with dystonia in other facial cervical perioral and mandibular muscles (oromandibular dystonia) (190). In addition, patients with blepharospasm may have dystonia in the limbs, trunk, and vocal cords (spasmodic dysphonia). Unlike idiopathic blepharospasm, which is most prominent when the patient is active, secondary blepharospasm often persists during rest. This distinction, however, is not reliable enough to differentiate primary from secondary blepharospasm.

Blepharospasm may vary from only a slightly annoying condition to a disabling disorder that interferes with daily activities such as reading, watching television, and driving. In our original study of botulinum toxin in patients with cranial dystonia, we used a rating scale, referred to as the Jankovic Rating Scale (JRS) (187; 109), to assess the severity and frequency of involuntary eyelid contractions (116). The self-rating response scale Blepharospasm Disability Index (BDI) has been found to correlate well with the Jankovic Rating Scale, traditionally used to quantify the symptoms of blepharospasm (109). Other clinical rating scales include quality-of-life scales (140). Health-related quality-of-life instruments have found that blepharospasm can markedly impair the functioning of the affected individuals (186; 81). With the emphasis on quality-of-life outcome measures, there is a need to develop instruments that measure this domain. In this regard, the craniocervical dystonia questionnaire (CDQ-24) will help evaluate the effects of botulinum toxin treatment on blepharospasm and cervical dystonia (166). A scale with good sensitivity to change has been developed to assess the severity of blepharospasm (41). This rating scale includes items such as apraxia of eyelid opening, orbicularis oculi spasms while writing, timing of duration of spasm (shorter than 3 to longer than 5 seconds), and counting of number of blinks and eyelid spasms. Besides the Jankovic Rating Scale, other rating scales have been proposed, including the Blepharospasm Severity Rating Scale (BSRS), but they seem more complicated and do not offer any meaningful advantage (38).

Up to two thirds of patients are rendered functionally blind by their blepharospasm. Blepharospasm is usually exacerbated by bright light; as a result, many patients wear sunglasses both outside and inside. The spasms may be transiently alleviated by pulling on an upper eyelid or an eyebrow, pinching the neck, talking, humming, yawning, singing, sleeping, relaxing, reading, concentrating, looking down, and performing other maneuvers or sensory tricks (geste antagonistique) or alleviating maneuvers (77; 76; 112; 177). Although adult-onset focal dystonias tend to remain focal, among the focal dystonias, the risk of spread was highest in patients with blepharospasm (31% past the head) compared to those starting in the upper extremities (16%), larynx (12%), or the neck (9%) (221). In most cases of blepharospasm, the spread occurs in the first 5 years after onset, as shown in several studies (221; 01). Another study, involving 124 patients presenting with blepharospasm (73 with cervical dystonia and 24 with focal hand dystonia; all with 10 years or more of symptom duration), showed that age at dystonia onset, age at initial spread, and the risk of initial spread were higher, and the time from onset to initial spread was shorter for the patients with blepharospasm compared to other focal dystonias (01). Similar findings were seen in a group of 132 patients followed for a mean of 7.5 years (203). In a review of 10,324 of patients with blepharospasm, 61% experienced the spread of dystonia to other regions, most commonly the oromandibular region and neck (195). One study identified several risk factors for spread of blepharospasm: previous head or face trauma with loss of consciousness, young age at onset of blepharospasm, and female gender (35). In another study, patients with blepharospasm were found to have a 2-fold higher rate of spread than those presenting with cervical dystonia, but spread developed at a similar age, about 50 years (149). In an international cohort study of 487 patients with dystonia, spread was observed in 50% of blepharospasm patients, particularly to the oromandibular region (42.2%) and neck (22.4%) (20). In one study, cluster analysis suggested the existence of three groups of patients who differed by type of spasms, motor severity, and co-existence of psychiatric symptoms (37). Further studies are needed to confirm these subtypes of blepharospasm.

Psychiatric symptoms, such as anxiety, depression, and psychosis, may be present even before or at onset of blepharospasm and were identified in 18% of 264 patients (78; 37; 195). The prevalence of obsessive-compulsive symptoms, often attributed to basal ganglia dysfunction, in patients with blepharospasm was significantly higher than in those with hemifacial spasm, despite the clinical similarities (23). This coexistence with mild psychiatric symptoms may explain the tendency to label blepharospasm as a psychogenic problem. However, psychogenic forms of blepharospasm are extraordinarily rare, and there is usually little or no evidence of any psychopathology in patients with blepharospasm (192). Nevertheless, psychogenic forms of blepharospasm are rarely encountered. One sign that was found to be characteristic of psychogenic blepharospasm was the presence of contraction of corrugator and procerus muscles without spasms of the orbicularis oculi (73). In one survey, the diagnosis of blepharospasm was delayed 4 to 10 years in more than half of the patients (115). Although the latency between the onset of symptoms and the diagnosis is shortening, largely as a result of intensive education of physicians and the public, the delays in making the diagnosis are still unacceptably long. In a study of 240 patients with blepharospasm, there was 2.8:1 female preponderance, 50% had pure blepharospasm, 31% had other forms of facial dystonia, and 4% had a combination of blepharospasm and eyelid opening apraxia (178).

In order to provide clinical guidelines for the clinical diagnosis of blepharospasm, the following 7 of 19 clinical items were analyzed by neurologists and ophthalmologists for interobserver correlation: “involuntary eyelid narrowing/closure due to orbicularis oculi spasms,” “bilateral spasms,” “synchronous spasms,” “stereotyped spasm pattern,” “sensory trick,” “inability to voluntarily suppress the spasms,” and “blink count at rest.” The following combination yielded 93% sensitivity and 90% specificity: “stereotyped, bilateral, and synchronous orbicularis oculi spasms inducing eyelid narrowing/closure,” followed by recognition of “sensory trick” or, alternatively, “increased blinking” (39). A sensory trick, the item that reached the greatest specificity, was defined as “any kind of manual maneuver performed by the patient that led to a transient reduction of the severity of dystonic posturing or movements in the period of time immediately after its execution.” The combination involuntary eyelid spasms and sensory trick (alleviating maneuver) yielded greater than 80% diagnostic sensitivity and specificity (42).

Prognosis and complications

Blepharospasm is a lifelong disorder in most patients. In most series, less than 3% of all patients experienced prolonged spontaneous remission (115; 78; 152), but the remission rate may be as high as 10%, mostly within the first 5 years (25). In a series of 238 patients who responded to a questionnaire, 11.3% were found to be symptom-free after less than 5 years of blepharospasm (25). The duration of remission averaged 6.3 years. It is important, however, to note that this was a retrospective study, and the only way to obtain reliable data on remission is through a longitudinal study of a prospectively followed cohort of patients. In one study, complete remission occurred in 15.4% of patients with cervical dystonia and in 5.8% of patients with blepharospasm (145). Remission occurred on average 4.5 years after onset of dystonic symptoms, but the majority of patients (63.8%) relapsed. Patients with remission were significantly younger at symptom onset than patients without remission.

Generally, patients have progressive worsening of their symptoms during the first 5 years after onset, following which the symptoms stabilize. Up to 15% become legally blind. Complications of chronic untreated blepharospasm include “dry eyes” and dermatochalasis (abnormal looseness of the eyelid skin due to constant pulling on the eyelids in an effort to keep the eyes open). In well over 80% of patients with blepharospasm, other facial oromandibular, pharyngeal, laryngeal, and cervical muscles become involved, and the focal dystonia gradually evolves into segmental (cranial-cervical) dystonia. In addition to the obvious physical disability, the patients often experience an uncomfortable “pulling” sensation behind their eyes. The partial blindness, discomfort, and social embarrassment caused by the blepharospasm often lead to anxiety and depression, although psychopathology is remarkably rare in patients with blepharospasm (192).

Clinical vignette

A 63-year-old woman first noted increased blinking and eye irritation at age 60 while walking outside on a sunny but windy day. She initially consulted an ophthalmologist and was diagnosed with “dry eyes.” Eye drops and lubricants, however, provided no benefit, and she began to notice involuntary closure of her eyes that interfered with her ability to drive, read, and watch television. Within 2 years after the onset of her eye symptoms, she began to notice that in addition to her involuntary eyelid spasms, she had involuntary contractions of her paranasal muscles, causing an embarrassing facial grimacing. The involuntary facial movements progressed to spasms of the jaws with marked trismus and bruxism, resulting in severe jaw pain; this was diagnosed as temporomandibular joint syndrome by her dentist. She was subsequently evaluated by a neurologist who confirmed the diagnosis of cranial dystonia, manifested chiefly by blepharospasm and jaw-closing oromandibular dystonia. Diazepam, trihexyphenidyl, and baclofen provided only minimal benefit, but botulinum toxin injections into her eyelids, eyebrows, paranasal muscles, and masseter muscles markedly improved her symptoms. She had occasional unilateral ptosis after the treatment, but overall, she was pleased with the results of the botulinum toxin injections.

Biological basis

Etiology and pathogenesis

That blepharospasm represents a forme fruste of idiopathic (primary) torsion dystonia is now well accepted (112). In addition to the frequent coexistence of blepharospasm and dystonia in other body segments, the relatively frequent occurrence of a family history of dystonia, essential-type tremor, or both supports the hypothesis that blepharospasm and other forms of dystonia may be genetically related. In our experience, one third of all patients with cranial-cervical dystonia have an action hand tremor similar to essential tremor or dystonia, and one third of patients have a first-degree relative with tremor or dystonia (104). In the series of Grandas and colleagues, a family history of movement disorders was present in 20% of their patients (78). Family history of blepharospasm has been reported to range between 9.5% and 27% among first-degree relatives of patients with blepharospasm, but only a single large family with blepharospasm has been reported (36). In a study of 56 families that included 436 first-degree relatives of probands, 233 of whom were examined, 27% of the index cases had at least one first-degree relative with blepharospasm, with estimated 20% penetrance if autosomal dominant transmission is assumed (47). Another study by the Italian group suggested that about 20% of patients with blepharospasm have a family member with dystonia (34). A Colombian family with childhood onset of rapid blinking was studied, and a tentative linkage was found on chromosome 3p (Singleton, personal communication). A study of idiopathic focal dystonias in the United Kingdom confirmed that these movement disorders were genetically transmitted. Several genetically determined dystonias (THAP1, TOR1A, SGCE, ATCAY, CIZ1, GNAL, ANO3, PRRT2, PRKRA, GCH1, TAF1, PNKD, ATP1A3, SLC2A1, TH, SPR, TIMM8A, DRD5, CP, PANK2, FTL, FBX07, and DJ1) have been associated with blepharospasm and cranial dystonia (128; 168). In one study of 132 patients with blepharospasm, all variants detected in GNAL, CIZ1, and TOR2A were thought to be benign, but the sequencing of REEP4 revealed two nonsynonymous variants that were considered possibly pathogenic; the authors concluded, however, that “functional studies will need to be performed to further elucidate the association between REEP4 and the disease” (87).

In two Italian families with adult-onset blepharospasm inherited as an autosomal dominant trait with reduced penetrance, the DYT1 gene mutation was excluded as linkage to the known dystonia loci (DYT1, DYT6, DYT7, and DYT13) (36). Another rare form of blepharospasm is the Mohr-Tranebjaerg syndrome due to a mutation in the deafness/dystonia peptide (DDP1) gene (127). Studies of genetic disorders that affect brainstem development, such as Duane syndrome and congenital ptosis, may provide important clues to the genetics of blepharospasm (57).

The study of blepharospasm due to some specific, identifiable cause (secondary blepharospasm) can provide insights into the pathogenesis of the primary, idiopathic disorder, sometimes referred to as “benign, essential blepharospasm.” Although the most common cause of blepharospasm is adult-onset idiopathic torsion dystonia, there are many other less common causes (Table 1). Even though most patients initially consult ophthalmologists, ocular disorders probably only rarely cause blepharospasm (189). In a study investigating symptoms related to ocular problems, a questionnaire was administered to 165 patients with blepharospasm and 180 control subjects with hemifacial spasm (150). They found that the questionnaire was highly predictive of ocular disease, particularly symptoms of dry eyes, and that patients with such symptoms had a significantly greater risk of blepharospasm with an adjusted odds ratio of 7.3 for 40- to 59-year age group and 4.2 for patients with ocular symptoms after the age of 59 years. It is possible that some of the patients with blepharospasm after ocular insult had peripherally induced blepharospasm (108; 150).

Table 1. Etiology of Blepharospasm

Tardive dystonia is probably the most common cause of secondary dystonia, including blepharospasm (106; 151; 71; 110). Tardive dystonia consists of a persistent dystonic movement involving chiefly the face, jaws, neck, trunk, and arms (216). Blepharospasm may be the initial presentation of tardive dystonia (188). In addition to dopamine-receptor-blocking drugs (neuroleptics), other drugs that can cause blepharospasm are lithium (157), lamotrigine (213), and others (151).

A variety of CNS lesions involving the rostral brainstem, thalamus, and basal ganglia (eg, stroke, multiple sclerosis, thalamotomy, subthalamic stimulation, hydrocephalus) have been reported in association with blepharospasm and other forms of cranial dystonia (117; 102; 158; 212; 79; 220; 126). In our initial report, we described six patients with clinical and radiographic evidence of rostral brainstem lesion and bilateral blepharospasm. Four patients suffered rostral brainstem strokes, and two had multiple sclerosis (117). This and subsequent reports of lesions producing blepharospasm, oromandibular dystonia, or both support the notion that in addition to the basal ganglia, other structures, including the thalamus, brainstem, cerebellum, and even cortex play an important role in the pathophysiology of cranial dystonia. Based on a review of medical records of 1114 patients with blepharospasm seen over the past 10 years at Emory University, only 18 had evidence of focal lesions on imaging studies, and together with cases published in the literature, information was available for 48 cases (126). Lesions thought to play a role in the pathogenesis of blepharospasm were found in the thalamus (n=12), lower brainstem (n=11), basal ganglia (n=9), cerebellum (n=9), midbrain (n=7), and cortex (n=1).

An animal model of blepharospasm has been proposed. In a rat rendered mildly deficient of dopamine by unilateral injection of 6-hydroxydopamine (6-OHDA), sectioning of the zygomatic branch of the facial nerve contralateral to the 6-OHDA lesion results in blepharospasm (193; 63). This is probably due to a reduction in the tonic inhibition of the trigeminal reflex blink circuits as a result of a mild striatal dopamine depletion coupled with an adaptive increase in the drive on the trigeminal sensory-motor blink circuit in response to the facial nerve lesion. They demonstrated the presence of pause neurons that terminate the blink generated by burst neurons; the pause but not the burst neurons are involved in lid adaptation (194). The dorsal accessory olive projects to the burst neurons, and medial inferior olive projects to the pause neurons, which may be involved in blepharospasm.

By definition, the cause of essential (primary, idiopathic) blepharospasm is unknown, but several lines of evidence suggest that genetic factors are important in the pathogenesis of this form of focal dystonia. Several gene loci have been identified for autosomal dominant dystonia, but only one gene mutation has been discovered; a three base-pair deletion in a gene coding for a novel ATP-binding protein in the 9q34 locus, termed torsinA, resulting in a loss of a pair of glutamic acid residues (174). Although the most common form of genetic dystonia, DYT1 dystonia, is almost never associated with blepharospasm, many patients with familial forms of blepharospasm have been reported (104). No gene mutation, however, has been identified. An overexpression of allele 2 with microsatellite repeat in the D5 dopamine receptor gene on chromosome 4 was found in patients with blepharospasm, but it is unlikely that this polymorphism in the DRD5 is functionally related to the pathogenesis of blepharospasm (159; 29). The authors suggest that a certain haplotype associated with allele 2 of the D5 receptor microsatellite confers susceptibility to developing blepharospasm.

High-resolution T1-weighted MRI imaging in 16 patients with primary blepharospasm showed evidence of bilateral grey-matter increase in the putamen, but this did not correlate with the duration of the blepharospasm (62). Other volumetric MRI studies showed increased volume of putamen and globus pallidus in patients with blepharospasm (53). Further studies using large numbers of subjects and paying special attention to technical details, using for example voxel based morphometry of MRI, are needed to resolve the discrepancies in the various studies reporting either increased or decreased grey matter volumes in patients with blepharospasm and other forms of dystonia. Functional MRI studies have shown activation of S1 and SMA and underactivity of M1 (54). Using T1 and diffusion-weighted MRI scans in 14 female pateitns with blepharospasm and 14 healthy matched controls, grey matter changes within the primary sensorimotor and the anterior cingulate cortices were detected in patients with blepharospasm compared to controls, but these changes did not correlate with disease duration (97).

Measuring metabolism by using F-18 fluorodeoxyglucose positron emission tomography, Esmaeli-Gutstein and colleagues showed significantly increased activity in the striatum and thalamus (59), and Hutchinson and colleagues found increased metabolic activity in the pons and cerebellum (99). In another study using positron emission tomography to measure cerebral blood flow before and after vibration applied to the lower face, Feiwell and colleagues showed that the normal activation of the primary sensorimotor area significantly decreased following vibration in patients with blepharospasm (67). There was also decreased activation of the primary sensorimotor area and supplementary motor area in response hand vibration. This suggests that patients with blepharospasm have abnormal sensorimotor processing, a phenomenon also described in patients with other forms of dystonia. Lack of cortical inhibition rather than abnormal excitation is thought to best explain the excessive movement in patients with dystonia (82; 83). This concept, however, has been challenged as neurophysiological markers of cortical and spinal disinhibition have been identified in patients with organic as well as psychogenic dystonia, raising the possibilities that these neurophysiological abnormalities may be secondary to the dystonic movement or posture rather than the primary pathophysiological mechanism (60). Using positron emission tomography to measure the in vivo binding of the dopaminergic radioligand (18F)spiperone in putamen in 21 patients, including those with blepharospasm, and comparing the findings with those from 13 normals, a decrease of dopamine D2-like binding was found in the putamen (179). These findings were confirmed later in eight patients with blepharospasm using another D2 receptor ligand, (11C)raclopride, showing 11.7% reduction in the uptake in the caudate and 11.6% in the anterior putamen compared to controls (96).

To understand the pathoanatomic and pathophysiologic mechanisms of blepharospasm, it is important to understand the brainstem pathways controlling lid closure. The trigeminal sensory neurons provide sensory input from the cornea and eyelashes to the trigeminal sensory nucleus, which extends from the pons to the upper cervical cord. From the caudal trigeminal nucleus, there are projections to the ipsilateral oculomotor nucleus; from the rostral trigeminal nucleus, there are bilateral projections to both oculomotor nuclei. The midbrain pretectal olivary nucleus controls the size of the pupil in response to the brightness of the light entering the eye, and this nucleus may also project to the oculomotor nucleus and cause contraction of the orbicularis oculi. By injecting fluorescent retrograde tracers into selected muscles of the upper and lower face and then injecting anterograde tracers into corresponding regions of the motor cortex of rhesus monkeys, Morecraft and colleagues defined the musculotopic organization of the facial nucleus (163). They found that the orbicularis oculi region was innervated chiefly by the rostral cingulate region, called M3, but the other four motor areas of cingulate cortex also innervate the facial motor nucleus. Three of the five motor areas project to the facial nucleus that innervates the lower part of the face, but the caudal and medial cingulate area, particularly M3 and M4, projects to the amygdala and to the part of the facial nucleus that bilaterally supplies the upper face and may be involved in emotional facial expression (164). The cortical areas involved in eyelid function have been defined by various techniques, including transcranial magnetic stimulation mapping and functional magnetic resonance imaging (88). Findings show a major input to M3 from the amygdala, which presumably plays a role in behaviors such as emotional facial expressions, but further research is needed to define other inputs.

Amygdala projection terminates in the face representation of M3. Stimulation of the mesial frontal lobe and anterior cingulate gyrus results in eye closure (198). Furthermore, the anterior and medial cingulate cortex receives rich dopaminergic input, and dopaminergic deficiency in Parkinson disease may be responsible for parkinsonian hypomimia and hypokinesia. Because most of the cingulate cortex is not affected in typical middle artery stroke, the upper face is usually spared in this type of stroke.

The blink reflex may be elicited by electrical stimulation of the supraorbital nerve and consists of an early first response (R1) (ipsilateral to the stimulated side with a latency of about 10 msec) and a late second (R2) (bilateral, longer duration and latency of about 30 msec) response; increased duration of the corneal reflex in patients with blepharospasm and oromandibular dystonia has been confirmed by several groups (19; 80). But these abnormalities are also seen in patients with dystonia without blepharospasm. Blink reflex excitability, as measured by R2 recovery cycle, appears to be increased in patients with blepharospasm but not in patients with increased blinking, suggesting different mechanisms for these two phenomena (31). Lew and colleagues showed that 87% of patients with cranial-cervical dystonia display increased latency and reduced amplitude of the acoustic reflex in cranial-cervical dystonia (139). In a study of 17 patients with dystonic blepharospasm and 11 age-matched controls, Gomez-Wong and colleagues found that in patients with blepharospasm there was normal prepulse inhibition occurring at 60 to 100 millisecond intervals, but the prepulse inhibition for the R2 response was abnormally reduced in 11 (64.7%) patients, including nine who did not use sensory tricks (geste antagonistique) (77). Experimental studies in cats and monkeys have found that serotonin (5-HT) facilitates the excitability of facial motoneurons, partly through 5-HT2 receptors that are densely concentrated in the facial nucleus, and serotonin agonists produce blepharospasm whereas serotonin antagonists reduce blink frequency (136). Furthermore, the normal prepulse inhibition of the trigeminal reflex is abnormal in a percentage of patients with blepharospasm. It is postulated that in these patients the normal sensory gating on trigeminal afferents is disturbed, the normal contact-induced reduction in the gain of trigeminofacial reflexes is lost, and the usual sensory tricks are no longer effective (76).

A conditioning protocol using high-frequency trains of electrical stimuli repeatedly given over the right supraorbital nerve, and timed to coincide with the R2 response elicited by a preceding supraorbital stimulus, was used to induce long-term potentiation-like plasticity in trigeminal, wide dynamic range neurons of the blink reflex circuit (181). In patients with blepharospasm, the facilitation of the bilateral R2 response was markedly increased compared to healthy controls if the high frequency stimulation coincided with the reflex blink but was suppressed if the stimulation preceded the reflex blink (146). These findings provide evidence that long-term potentiation-like plasticity is increased in the trigeminal reflex circuit of patients with blepharospasm. The investigators have proposed that an abnormal corneal input induced by excessive blinking exacerbates increased long-term potentiation-like plasticity in blepharospasm, which is removed with botulinum toxin treatment, thus, restoring plasticity toward normal values. Associative plasticity has been found to be increased in the hand cortical area in patients with blepharospasm, but not in patients with hemifacial spasm, suggesting that this finding is relatively specific for blepharospasm (Berardelli et al 2006, personal communication). Also, tactile discrimination threshold is significantly increased in patients with blepharospasm, but this is also present in patients with hand dystonia (Berardelli et al 2006, personal communication). Besides electrical stimulation, subthreshold repetitive transcranial magnetic stimulation (rTMS) has been used to probe mechanisms of synaptic plasticity in the sensorimotor circuits of patients with focal dystonia (133). These and other studies support the hypothesis that at least some cases of blepharospasm result from reduced inhibition or hyperexcitability of brainstem interneurons, possibly as a result of dysfunction of descending basal ganglia pathways (102; 84).

Various in vitro and in vivo studies suggest that sensitization of the trigeminal system plays a role in the pathophysiology of photophobia and blepharospasm (150). For example, an in vitro model of orbicularis oculi control suggests that blepharospasm arises from a change in the trigeminal input to the facial nucleus. The mechanism of photophobia is not well understood but it probably involves the melanopsin photopigment identified in the retinal ganglion cells, iris, and the trigeminal ganglion nerve fibers that innervate the cornea. It is likely that photophobia involves the convergence of signals from the optic nerve and trigeminal nerves at the level of the thalamus or the pretectum (51). In addition, specific light-sensitive cells, different from rods and cones, called intrinsically photosensitive retinal ganglion cells (IPRG), which mediate pupillary constriction and are involved in modulation of circadian rhythm, may be involved in oculodynia (66; 137).

Peripheral trauma is increasingly recognized as a cause of dystonia, and peripheral trauma may trigger dystonia in carriers of the idiopathic torsion dystonia gene (105; 108). Up to 12% of patients with blepharospasm report the occurrence of ocular trauma prior to the onset of their movement disorder (78). About 11% of otherwise healthy individuals have dry eyes, but most patients do not have dry-eye syndrome as seen in Sjögren disease, although they may have dry-eye symptoms. Many patients experience symptoms of “dry eyes” and other ocular symptoms shortly before the onset of blepharospasm, suggesting that disorders of the anterior segment of the eye may actually trigger blepharospasm (150). Whether dry eye syndrome is the cause or result of blepharospasm is not known. People with dry eyes blink more frequently than those without dry eyes, presumably to reduce the chance of holes developing in the tear film and blink oscillations help thicken the lipid layer of the tear film (64). The production of proteins, particularly lacritin, normally secreted by the lacrimal glands is markedly reduced in some patients with blephritis, which reduces the lipid layer (129). Although blephritis often precedes blepharospasm, it is not clear that it is a risk factor for blepharospasm as blephritis is a frequent symptom in otherwise normal individuals. It is possible that some form of injury to the anterior segment of the eye causes trigeminal sensitization leading to photophobia, increased blinking, and blepharospasm. The mechanisms of photophobia and dry eyes are complex and may be due to an ocular problem as well as a central effect (Henriquez and 63). Furthermore, dryness may be more related to the make-up of the tear film and its dynamics than the actual amount of tears produced (07; 70; 162; 72).

Schicatano and colleagues proposed a 2-factor model based on the observation that dopamine depletion reduces the tonic inhibition of trigeminal blink circuit, thus, creating “a permissive environment within the trigeminal blink circuits” that, along with an external ophthalmic insult (second factor), precipitates blepharospasm (193). They suggest that this 2-factor model may also be applicable to the genesis of other dystonias. Another animal model may be severe unilateral keratoconjunctivitis sicca and blepharospasm, reported in 16 juvenile Yorkshire Terriers (92).

Only few patients with cranial dystonia, including blepharospasm, have been studied at autopsy (74; 85; 84; 86). In a study of four brains from patients with generalized dystonia, genetically confirmed DYT1, McNaught and colleagues found perinuclear inclusion bodies in the midbrain reticular formation and periaqueductal gray of the pedunculopontine nucleus, cuneiform nucleus, and griseum centrale mesencephali (155). These inclusions stained positive for ubiquitin, torsinA, and the nuclear envelope protein lamin A/C. In addition, tau/ubiquitin-immunoreactive aggregates were found in the substantia nigra pars compacta and locus coeruleus. These findings support the notion that DYT1 dystonia is associated with impaired protein handling, particularly in the brainstem nuclei. No specific pathological abnormalities have been identified in the autopsied brains of patients with blepharospasm. We reported a 68-year-old woman with 7-year history of progressive blepharospasm, spasmodic dysphonia, and cervical dystonia who died in the hospital shortly after the diagnosis of poorly differentiated metastatic adenocarcinoma (120). Her brain was collected within 30 minutes and examined histologically as well as biochemically. No abnormalities were noted on histological examination, but the norepinephrine levels were markedly increased in the brainstem (199.6% in red nucleus and 415.2% in substantia nigra). Two patients with atypical cranial dystonia were found to have a mosaic neuronal cell loss and gliosis in the striatum. Neuronal cell loss in the substantia nigra and other brainstem nuclei was reported in three patients, two of whom had associated Lewy bodies (148).

Epidemiology

Although considered an unusual or rare disorder, cranial-cervical dystonia is the most common form of dystonia at Baylor College of Medicine (78% of patients with dystonia). The prevalence of blepharospasm is estimated to be 5 per 100,000 (78; 171; 121). In two large series of patients with blepharospasm, women outnumbered men at a ratio of about two to one, and in two thirds of the patients, the movement disorder began after 50 years of age (115; 78). Pooling data from eight European countries, the cases diagnosed as blepharospasm by adult neurologists with specialist movement disorder (and botulinum toxin) clinics resulted in a prevalence rate for blepharospasm of 36 (95%, confidence interval 31 to 41) per million (58). The authors suggest that because of ascertainment bias, the true prevalence is probably considerably higher. In a study of a region of Southern Italy, Defazio and colleagues found a prevalence of 133 per million, and apraxia of eyelid opening was found to coexist in one third of cases (45). In their review of published epidemiologic data, they concluded that the crude estimated prevalence ranges from 16 to 133 per million; age at onset and female gender are putative risk factors; and prior head and face trauma may increase the risk of spread of dystonia to adjacent body regions (43). In another report, they estimated the crude prevalence rate of apraxia of eyelid opening to be 59 per million (95%, confidence interval 24 to 173); it coexisted with adult-onset blepharospasm in 75% of cases and with atypical parkinsonism in 25% of cases (135). We examined the glabellar reflex in 100 subjects, including patients with Parkinson disease (n=41), progressive supranuclear palsy (n=12), multiple system atrophy (n=7), and healthy, age-matched controls (n=40) and found that this reflex is a relatively sensitive sign of parkinsonian disorders, but it lacks specificity as it does not differentiate between the three most common parkinsonian disorders (22). In addition to Parkinson disease and progressive supranuclear palsy, blepharospasm may be one of the most disabling features in patients with multiple system atrophy (21; 123). In a study of 659 patients, of which 357 had parkinsonism (276 idiopathic Parkinson disease), 81 atypical parkinsonism (57 progressive supranuclear palsy, 11 multiple system atrophy, 13 corticobasal degeneration), 274 essential tremor, 22 cervical dystonia, and 6 spinocerebellar ataxia; blepharospasm was present in 7.41% of those with atypical parkinsonism and 3.3% of those with Parkinson disease but in none of the other patients (184).

There appears to be a protective effect of smoking in patients with blepharospasm compared to those with hemifacial spasm, similar to Parkinson disease (46). The inverse relationship between blepharospasm and coffee consumption was confirmed by another prospective family-based study, which also showed that prior eye disease and coffee drinking increase the risk of blepharospasm (34). In one study, patients with blepharospasm were much more prone to anxiety and depression as compared to those with hemifacial spasm (81).

Prevention

Because the majority of cases of blepharospasm are thought to be genetically determined, there is little hope that the disorder will be preventable in the near future. Certainly, avoiding dopamine-receptor-blocking drugs will prevent the development of tardive blepharospasm (71; 110).

Differential diagnosis

Idiopathic torsion dystonia accounts for the majority of cases of blepharospasm. Tardive dystonia, caused by exposure to dopamine-receptor-blocking drugs, is probably the second most common cause of blepharospasm. A variety of lesions in the upper midbrain and the basal ganglia, eg, stroke, multiple sclerosis, thalamotomy, and hydrocephalus, have been reported in association with blepharospasm (104; 85). A lesion in the dorsomedial caudal pontine tegmentum has also been associated with blepharospasm (08). Cranial-cervical dystonia can also occur in association with diseases of the central nervous system, such as Tourette syndrome, Wilson disease, Parkinson disease, progressive supranuclear palsy, multiple system atrophy, postencephalitic parkinsonism, and X-linked dystonia-parkinsonism syndrome. Levodopa replacement in Parkinson disease can also cause blepharospasm. A few reports have suggested that there may be an association between blepharospasm and autoimmune diseases, especially systemic lupus erythematosus, Sjögren syndrome, rheumatoid arthritis (118), and myasthenia gravis (134; 170). Increased blinking can also be a sign of a seizure (16).

In addition to dystonia, other conditions can lead to closure of the eyelids. Ptosis may result from weakness or paralysis of the levator palpebrae muscle or the smooth muscle of Müller. Some patients cannot open their eyes because they cannot "activate" the levator palpebrae muscles. This is analogous to the motor blocks or the freezing phenomenon experienced by some, and the terms “apraxia of eyelid opening” and "eyelid freezing" are used to describe this disorder.

The inability to open eyes has been attributed to absence of contraction or even inhibition of the levator palpebrae (despite compensatory frontalis contraction) (75; 132). Based on clinical and electrophysiological observations in six patients, Elston argued that this sign was caused by isolated contraction of the pretarsal orbicularis oculi (56). In some cases, electromyographic recording from the levator palpebrae and orbicularis oculi muscles is required to differentiate this persistent pretarsal orbicularis oculi contraction from the levator inhibition (09). EMG has been used to detect an abnormal persistence of orbicularis oculi activity in patients with apraxia of eyelid opening (207). Progressive supranuclear palsy is the most common cause of eyelid freezing seen in the Parkinson clinic, but other parkinsonian syndromes, Huntington disease, hemispheric cerebral vascular disease, and neuroacanthocytosis are occasionally associated with this phenomenon. Apraxia of eyelid opening, however, can occur in isolation without any other motor deficits, and it may improve with levodopa (50) and recur when levodopa is reduced, as may be the case after deep brain stimulation (211).

Involuntary contractions of the orbicularis oculi muscle can also be caused by ophthalmologic disorders, possibly mediated by the trigeminal-palpebral reflex (eg, blepharitis, conjunctivitis, "dry eye syndrome," keratitis, iritis, uveitis) or the opticopalpebral reflex (eg, albinism, achromatopsia, maculopathies) (104). Reflex blepharospasm is also seen in premature infants, patients with various parkinsonian syndromes, and patients with lesions in the nondominant temporoparietal lobe (Fisher sign) and in response to loud noise (cochleopalpebral reflex), sudden free fall (vestibulopalpebral reflex), and gag (palatopalpebral reflex). Medial frontal hypometabolism has been demonstrated with (18F) fluorodeoxyglucose positron emission tomography in four patients with lid opening apraxia who had no pyramidal or extrapyramidal dysfunction (197).

Contraction of orbicularis oculi with eyelid closure can also result from hyperactivity of the peripheral nervous system. Hemifacial spasm, a form of segmental myoclonus, is characterized by involuntary, paroxysmal tonic, or clonic contractions of the muscles innervated by the 7th cranial nerve (219). In 1688 patients, mostly women, the cause was unknown in 163, and a vascular abnormality, identified in 509 patients, was the most frequent cause (52). In one study, magnetic resonance tomographic angiography showed that 64.9% of patients with hemifacial spasm had ipsilateral vascular compression (03). The presumed pathophysiologic mechanism of hemifacial spasm involves the generation of ortho- and antidromic impulses by a damaged area of the facial nerve. The constant antidromic stimulation may result in "kindling," causing neuronal discharge in the facial motor nucleus, leading to hemifacial spasm. At onset, the patients typically experience occasional twitches in the eyelids, but with progression, the spasms and twitches become more constant and involve the lower facial musculature. The clonic and tonic contractions are triggered by action (smiling, talking, eating, blinking). Hemifacial spasm is easily distinguished from blepharospasm caused by dystonia because it is virtually always unilateral, although there are rare exceptions (205). Furthermore, in contrast to blepharospasm, patients with hemifacial spasm often exhibit paradoxical raising of the eyebrow as the eye closes (the “other” Babinski sign) (49). The term "tic convulsif" describes the rare coexistence of trigeminal neuralgia and hemifacial spasm. The phenomenology of aberrant facial regeneration or facial synkinesis is similar to hemifacial spasm, but the onset usually follows facial palsy. Studies in macaque monkeys show that following facial nerve injury, the orbicularis oculi motoneurons innervate the perioral muscles, causing co-contraction (synkinesia) of eyelid and perioral muscles (12). In a series of 164 patients with hemifacial spasm, nine (5.5%) were thought to have coexistent blepharospasm (204). Blepharospasm also has been reported after Bell palsy (28; 13; 26), but prospective studies could not demonstrate an association between Bell palsy and subsequent blepharospasm (47; personal communication). Hemimasticatory spasm is a rare disorder whose underlying mechanism is similar to hemifacial spasm, but the trigeminal rather than the facial nerve is involved (10). The spasms of the masticatory muscles may or may not be associated with hemifacial atrophy. Facial myokymia, a rapid undulation, and flickering of the facial muscles from the frontalis to the platysma, is thought to be due to an intramedullary lesion close to the facial motor nucleus. Multiple sclerosis is probably the most common cause, but intra-axial tumors and Guillain-Barré syndrome have also been described as associated with this movement disorder. Tetanus is caused by tetanus toxin, a product of Clostridium tetani, and it is characterized by hyperactivity of motor neurons, which causes forceful closure of the eyelids. Although rare in the United States, it still remains a major public health problem in underdeveloped areas. Amyloidosis V and Schwartz-Jampel syndrome (autosomal recessive disorder manifested by a combination of blepharospasm, blepharophimosis, dwarfism, muscular hypertrophy, generalized muscular stiffness, and myotonia) represent additional causes of blepharospasm.

Blepharoclonus refers to rhythmic contractions of the orbicularis oculi closely resembling tremor, present during gentle closure of the eyelids (147). Although no apparent cause can be identified in many cases, blepharoclonus is occasionally associated with multiple sclerosis, obstructive hydrocephalus, and Arnold-Chiari malformation (100; 101). Blinking, the most common motor tic present in 70% of patients with Tourette syndrome, is characterized by bursts of rapid, nonsustained contractions of the orbicularis oculi. On the other hand, dystonic tics of the eyelids, found in 15% of patients, can cause diagnostic difficulties because they are transiently sustained and may resemble blepharospasm.

An important cause of facial movements is Whipple disease. In addition to gastrointestinal symptoms, patients with Whipple disease typically exhibit supranuclear ophthalmoparesis and rhythmic contractions of the eyelids, face, and mouth in synchrony with convergent eye oscillations. This oculomasticatory myorhythmia is usually associated with contractions of the neck as well as the pharyngeal and proximal and distal musculature.

The phenomenology of blepharospasm is usually the same regardless of its cause. The presence of associated findings, however, may suggest a specific etiology. The recognition of stereotypies (repetitive, patterned, seemingly purposeful but purposeless movements), for example, suggests the diagnosis of tardive dystonia, whereas corneal Kayser-Fleischer ring and evidence of hepatic failure indicate Wilson disease (202).

Diagnostic workup

Except for a careful ophthalmologic evaluation, there is usually no need for any diagnostic studies. Neuroimaging studies are helpful in the evaluation of patients suspected of having secondary blepharospasm related to acute stroke, multiple sclerosis, or other etiologies. Rarely are tests for collagen-vascular and autoimmune diseases indicated.

Management

There are few controlled therapeutic trials in blepharospasm (94; 110). Furthermore, as noted above, only a few studies have employed objective assessments, validated clinical rating scales, and other instruments to measure the severity of blepharospasm and the impact of the therapeutic intervention on functional capabilities, activities of daily living, and quality of life (116; 140; 186; 81; 187). The most frequently used scales are the Jankovic Rating Scale and the Blepharospasm Disability Index (109). Glasses fitted with wire loops to press against the brow (Lundie loops) or to lift the upper eyelid (eyelid crutch) have been reported to be helpful in some patients with blepharospasm and apraxia of eyelid opening (95; 182). Information is available at http://www.eyeglassrepair.net. The Lundie loops may serve as a sensory trick or alleviating maneuver. Indeed, “sensory trick frames” were found to be effective in reducing blepharospasm-related disability in a small study involving nine patients (141). In another study, 39 of 58 patients with blepharospasm reported benefit from a spectacle-mounted device (“Pressop”), which provides continuous individually localized focal pressure on the temple to mimic the effect of finger pressure used as a sensory trick (65). Occasionally, biofeedback and other muscle relaxation techniques and stress management may be helpful, particularly for those patients in whom stress exacerbates the symptoms (206). Because light often precipitates eyelid spasms, filtering the light using various special lenses has been proposed as a potential treatment for blepharospasm.

In a study of 87 blepharospasm, migraine, and normal control subjects, both gray and FL-41-tinted lenses (eg, cocoons) significantly improved light sensitivity thresholds in all groups (02). In another study designed to quantify light sensitivity in blepharospasm, 24 patients and 10 controls were evaluated with seven different chromatic lenses (93); 71% of patients with blepharospasm reported the greatest relief of photophobia with the FL-41 lens, which filters well under 400 nm, and moderately between 400 and 550 nm. Other studies concluded that sensory photophobia might be related as much to the wavelength as to the intensity of the light exposure (200).

Many patients with blepharospasm complain of eye pain and photodynia, which may, in part, be due to local inflammation and, therefore, local anti-inflammatory drops, such as nepafenac (Nevanac) eye drops, may be helpful. Also, treatment with lubricant eye drops (eg, transiently preserved or non-Bion tears, NanoTears, or Lacrilube) is highly soothing to patients who experience eye irritation as a result of associated dry eyes. Some patients require cytokine-blocking agents such as cyclosporin, oral pilocarpine, cevimeline, topical androgens, lacrimal pump, punctual occlusion, or other procedures to relieve the dry-eye syndrome. Some patients with blephritis in the setting of rosacea may require a course of antibiotics. Although some patients with blepharospasm obtain satisfactory relief from medications such as clonazepam (1 to 8 mg/day), trihexyphenidyl (2 to 12 mg/day), levodopa, tetrabenazine, or baclofen, in the majority of patients, medications do not effectively control blepharospasm. Capsaicin, the hot ingredient of chilies, has been described to dramatically improve blepharospasm in one woman when applied to her forehead, although it did not affect normal blinking (138). Zolpidem, an imidazopyridine agonist with a high affinity on benzodiazepine subtype receptor BZ1 (ω1), was found to improve blepharospasm and other forms of dystonia at doses of 5 to 20 mg/day (161).

Although there is a paucity of well-designed, double-blind, controlled studies, botulinum toxin injections into the eyelids and eyebrows are now considered by many as the treatment of choice (122; 32; 109; 30; 124; 103; 110; 90; 05). Indeed, in a review of 2,026 subjects enrolled in the Dystonia Coalition Study, 49% of patients with blepharospasm were treated with botulinum toxin, the highest percentage of all focal dystonias (180). Nevertheless, in the American Academy of Neurology Practice Guideline, botulinum toxin was assigned only level B evidence in treating blepharospasm (196). This conclusion is based on a relative lack of double-blind, placebo-controlled studies, even though such studies would now have been considered unethical. Botulinum toxin injections provide moderate or marked improvement in over 90% of patients. Pretarsal injections, particularly into the Riolan’s part of the pretarsal orbicularis oculi, seem to provide the most benefit (107; 131; 144; 24). Some investigators suggested that combined orbital and pretarsal injections may be needed for those patients who do not obtain a satisfactory response from pretarsal injection alone (61). The average latency from the time of the injection to the onset of improvement is 2 to 5 days, and the average duration is 3 to 4 months. Following botulinum toxin treatments, as a result of reduced eyelid and eyebrow spasms, most patients can function normally, and they have fewer difficulties driving, watching television, or reading. In addition to the observed functional improvement, there is usually a meaningful amelioration of discomfort, and because of less embarrassment, the patients' self-esteem also frequently improves. Although about 10% to 15% of all treatment sessions are followed by some side effects (ptosis, blurring of vision or diplopia, tearing, and local hematoma), the complications only rarely affect a patient's functioning and usually resolve spontaneously in less than 2 weeks. There is no apparent decline in benefit, and the frequency of complications actually decreases after repeat botulinum toxin treatments (119).

Botulinum toxin type B (BTX-B) is as effective in the treatment of blepharospasm as BTX-A but is associated with more discomfort and possibly shorter duration of effect; BTX-B but may be particularly useful in patients with immunoresistance to BTX-A (Wan and Jankovic 2005). Botulinum toxin F (BTX-F) may also be a useful alternative for those rare patients who develop blocking antibodies to BTX-A, but its usefulness is limited by a substantially shorter benefit duration (156). In a double-blind study of 212 consecutive patients with essential blepharospasm who received one injection of BOTOX® and one injection of Dysport® in two separate treatment sessions, using the empirical ratio BOTOX: Dysport of 1:4 (IU), the average dose of BOTOX per treatment was 45.4 IU ± 13.3 (range 25 to 85 IU) and of Dysport 182.1 IU ± 55.1 (range 100 to 340 IU). There was no difference in the duration of the effect (about 8 weeks), but side effects, particularly ptosis, were more frequent with Dysport as compared to Botox (24.1% vs. 17.0%) (p < 0.05). Botulinum toxin treatment improves the frequency and intensity of blepharospasm and markedly improves the quality of life (210).

The efficacy and safety of another type of BTX-A, NT-201 (or IncobotulinumtoxinA, Xeomin), in the treatment of blepharospasm was also confirmed by a prospective, double-blind, placebo-controlled, randomized, multicenter study involving 109 patients (mean total dose of Xeomin per treatment visit was 64.8 U), the Jankovic Rating Scale severity subscore was significantly reduced compared to placebo (p< 0.001). The most commonly reported adverse effects related to Xeomin versus placebo were eyelid ptosis (18.9 vs. 8.8%), dry eye (16.2 vs. 11.8%), and dry mouth (16.2 vs. 2.9%) (109). In a randomized, placebo-controlled, double-blind trial of efficacy and safety, 109 patients with blepharospasm were randomized in a 2:1 ratio into treatment with incobotulinumtoxinA (up to 50 U per eye) or placebo (111). After a follow-up of 20 weeks, a significant difference was observed in the Jankovic Rating Scale severity subscore, rated by an independent rater 6 weeks following treatment, favoring incobotulinumtoxinA by (P < .001). Functional impairment, as measured by the Blepharospasm Disability Index also significantly improved (P = .002) compared with placebo. Adverse events were reported in 70.3% of incobotulinumtoxinA patients and 58.8% of placebo patients; eyelid ptosis (18.9% vs. 5.9%), dry eye (18.9% vs. 11.8%), and dry mouth (14.9% vs. 2.9%) occurred most frequently. In a multicenter, clinical, fixed-dose trial, Dysport (40, 80, and 120 units/eye) or placebo (in a 3:1 randomization ratio) were administered to 119 patients with blepharospasm, 85 of whom completed the 16-week follow-up; several parameters showed robust improvement in Dysport arms compared to placebo (208). IncobotulinumtoxinA (Xeomin(®), NT 201) injections were studied in subjects with blepharospasm who completed a ≤ 20 weeks, double-blind, placebo-controlled study and then entered a ≤ 69 weeks open-label extension period during which they received five or fewer additional incobotulinumtoxinA treatments at flexible doses (≤ 50 U per eye) and flexible injection intervals (minimum of 6 weeks) (209). All 102 subjects who completed the main period entered the open-label trial, and of those, 82 completed the study. At 6 weeks, all scores, including the investigator-rated Jankovic Rating Scale and patient-rated Blepharospasm Disability Index, were significantly improved (p ≤ 0.001l). The most frequently reported adverse events were eyelid ptosis (31.4%) and dry eye symptoms (17.6%). Apraclonidine, an alpha-2 adrenergic receptor agonist with weak alpha 1 agonist effects, has been found to be transiently effective in the treatment of BTX-induced ptosis and is being evaluated in the treatment of blepharospasm (215). In another study of incobotulinumtoxinA in treating blepharospasm, the latency from injection to perceived effect was about 5 days, and the benefits lasted up to 20 weeks (160).

Cochrane database analysis concluded that BTX resulted in a moderate to large improvement in blepharospasm-specific severity, with a reduction of 0.93 points on the Jankovic Rating Scale (JRS) severity subscale, and resulted in a moderate to large improvement in blepharospasm-specific disability and blepharospasm-specific involuntary movements at 4 to 6 weeks after injection compared to placebo (55).

Several strategies have been investigated to alleviate the most common side effect of BTX treatment, namely ptosis. Apraclonidine, an alpha-2 adrenergic receptor agonist with weak alpha 1 agonist effects, has been found to be effective in the treatment of BTX-induced ptosis, and it may also improve blepharospasm, as it recurs when the effects of BTX start wearing off (215). Another medication with a similar mechanism of action and potential benefits in treating ptosis and blepharospasm is oxymetazoline hydrochloride ophthalmic solution (201).

Several long-term studies have demonstrated that BTX continues to be effective over a period of decades, and no serious complications have been reported (33; 183).

In addition to the four different BTX-A preparations available on the market in one or more countries, Botox® (Allergan Inc., Irvine, California, United States), Dysport (Ipsen Ltd, Slough, United Kingdom), Xeomin (Merz Pharmaceuticals, Frankfurt am Main, Germany), and Prosigne (Lanzhou Biological Products, China), Meditoxin® (Medy-Tox, Seoul, Korea, another name: Neuronox®) was introduced for the treatment of blepharospasm and is available in Korea (223). All these types were evaluated in blepharospasm and were found to be comparable with respect to clinical efficacy and adverse effects.

Apraxia of eyelid opening is more difficult to treat than blepharospasm, but some patients with apraxia of eyelid opening improve with botulinum toxin injections, particularly if it is triggered by blepharospasm (69). In 16 patients suffering from blepharospasm and dry eyes, botulinum toxin injections were effective in relieving blepharospasm but were unsuccessful in treating dry-eye syndrome (98). A formulation of BTX-A that is free of complexing proteins, NT 201 (Xeomin®), has been reported to be equivalent in terms of efficacy and safety to BOTOX in patients with blepharospasm (187). The most important determinant of a successful outcome is the customization of dose and site of injection according to the individual patient (172).

Patients who fail to obtain satisfactory control of their blepharospasm with botulinum toxin may be candidates for surgical treatment. Facial nerve lysis and orbicularis oculi myectomy, once used extensively in the treatment of blepharospasm, have been essentially abolished because botulinum toxin treatment is usually effective, and postoperative complications, such as ectropion, exposure keratitis, facial droop, and postoperative swelling and scarring are common (06; 27; 176). Patients with severe apraxia of eyelid opening may require frontalis suspension combined with blepharoplasty (48). Besides botulinum toxin and surgery, chemomyectomy with muscle necrotizing drugs, such as doxorubicin, has been tried in some patients with blepharospasm and hemifacial spasm. Severe local irritation limits the usefulness of this therapy. However, a modification of the procedure using a combination of bupivacaine/hyaluronidase and Doxil® (doxorubicin HCl liposome injection), a liposome-encapsulated form of doxorubicin, may be more effective and safer (154). In one study, the immunotoxin ricin-mAb35 was injected in one eyelid of adult rabbits, and within 1 week after injection, the total number of orbicularis oculi myofibers was significantly decreased in the ricin-mAb35-treated eyelids. This myofiber loss was maintained 6 months after the initial injection (89). Daily application of topical acetyl hexapeptide-8 (AH8), a competitive SNAP25 inhibitor, has been investigated in a double-blind, placebo-controlled, randomized trial in 24 patients with blepharospasm (142). There was a trend for longer time until return to baseline Jankovic Rating Scale after injection, and there were no adverse effects. Further studies are needed before this treatment can be recommended as a standard therapy for blepharospasm.

Various neurophysiologic techniques have been used to investigate the pathophysiology and treatment of blepharospasm. In a randomized, sham-controlled, observer-blinded prospective study of 12 patients with blepharospasm, a 15-minute session of low-frequency (0.2 Hz) rTMS over the anterior cingulate cortex was associated with a significant clinical improvement, which was detectable even 1 hour after stimulation, suggesting that rTMS could be a useful therapeutic strategy in blepharospasm (133).

Finally, because photophobia may be caused by sympathetically maintained pain, a blockade of the superior sympathetic ganglion with local anesthetic has been reported to improve light-induced eyelid spasm and, as such, may be useful as a therapeutic modality in patients with blepharospasm (153). One report suggested that administering a combination of linoleic acid and alpha-linoleic acid, which presumably modifies the composition and function of neuronal membranes, has improved blepharospasm in a catecholamine-depleted rat model (165). This polyunsaturated free fatty acid, however, has not been tested in patients with blepharospasm. It is not clear whether bipolar stimulation of the nerve to the levator palpebrae, found to be potentially useful in opening eyes in experimental rabbits (Scott, personal communication), will be applicable in the treatment of blepharospasm.

Surgery may be required when medical therapy does not provide adequate relief. In addition to orbicularis oculi myectomy (176), other procedures reported to be effective include levator aponeurosis advancement, frontalis suspension, and browlift using polypropylene suture. The latter procedure may be necessary to manage eyelid apraxia associated with blepharospasm (169). Deep brain stimulation is effective in the treatment of generalized dystonia, and it has been applied in the treatment of patients with disabling cranial dystonia, including blepharospasm, who have become refractory to botulinum toxin injections (68). In a report of six patients with cranial-cervical dystonia, bilateral GPi deep brain stimulation showed a 72% mean improvement in the Burke-Fahn-Marsden dystonia rating scale (BFMDRS) total movement score at 6 months (173). There was also a trend to improvement in the mean BFMDRS disability score. It is important to note, however, that despite improvement in cranial-cervical dystonia, motor function was mildly worse in previously nondystonic body regions in four patients. Given the clear utility of deep brain stimulation for dystonia, further studies for blepharospasm will certainly be undertaken. In one of the largest long-term outcome studies involving 12 patients (six women, six men) with cranial dystonia followed for up to 6.5 years after bilateral GPi deep brain stimulation, the dystonia severity as assessed by the BFMDRS showed a mean improvement of 45% at short-term follow-up (4.4 ± 1.5 months; P < 0.001), and of 53% at long-term follow-up (38.8 ± 21.7 months; P < 0.001); the blepharospasm scales also significantly improved (185). One subphenotype, manifested by blepharospasm, apraxia of eyelid opening, and anterocollis, is therapeutically particularly challenging, and some of these patients may require deep brain stimulation in combination with botulinum toxin injections (217).