Horner Syndrome: A Clinical Review - American Chemical Society


Horner Syndrome: A Clinical Review - American Chemical Societypubs.acs.org/doi/pdf/10.1021/acschemneuro.7b00405Nov 30, 2...

3 downloads 125 Views 10MB Size

Review Cite This: ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

pubs.acs.org/chemneuro

Horner Syndrome: A Clinical Review Timothy J. Martin* Department of Ophthalmology, Wake Forest University School of Medicine, Wiston-Salem, North Carolina 27157, United States ABSTRACT: Horner syndrome results from an interruption of the oculosympathetic pathway. Patients with Horner syndrome present with a slightly droopy upper lid and a smaller pupil on the affected side; less commonly, there is a deficiency of sweating over the brow or face on the affected side. This condition does not usually cause vision problems or other significant symptoms, but is important as a warning sign that the oculosympathetic pathway has been interrupted, potentially with serious and even life-threatening processes. The oculosympathetic pathway has a long and circuitous course, beginning in the brain and traveling down the spinal cord to exit in the chest, then up the neck and into the orbit. Therefore, this syndrome with unimpressive clinical findings and insignificant symptoms may be a sign of serious pathology in the head, chest, or neck. This clinical review discusses how to identify the signs, confirm the diagnosis, and evaluate the many causes of Horner syndrome. KEYWORDS: Horner syndrome, ptosis, oculosympathetic pathway, anisocoria



in 1852.2 Therefore, this condition is sometimes called Claude Bernard−Horner syndrome, especially in the French literature.3,4 Signs and Symptoms. Patients with Horner syndrome typically present with a small upper eyelid ptosis (1−2 mm) and an anisocoria with the smaller pupil on the affected side. Ipsilateral anhidrosis of the forehead or face is a far less reliable sign, as it is not clinically evident (or not present at all) most of the time. Similar to the upper eyelid, the lower eyelid can be slightly more closed as well, narrowing the palpebral fissure, which makes the eye look enophthalmic even though true enophthalmos is not present. Transient conjunctival injection and relative hypotony have been reported in the acute setting.5 Miosis. In Horner syndrome, the pupil in the affected eye is smaller than that in the opposite eye due to the loss of sympathetic tone of the pupillary dilator. A difference in the size of one pupil compared to that of the other is called anisocoria. In a normal state, the two pupils are the same size. The size of the pupillary aperture of the iris is controlled by autonomic innervation to two opposing sets of muscles in the iris. The pupillary sphincter is a circular muscle in the iris that frames the pupil; it is innervated by the parasympathetic autonomic system, and activation makes the pupil smaller. The pupillary dilator consists of radial muscle fibers that dilate the pupil when activated and is innervated by the sympathetic autonomic system; the pupils become large with sympathetic “fight or flight” response. Ultimately, the size of the pupil is determined by the balance of these opposing systems. There are many factors that influence sympathetic and parasympathetic

INTRODUCTION Horner syndrome describes the clinical findings (signs and symptoms) that result from an interruption of sympathetic innervation to the eye (oculosympathetic paresis). The classic triad in Horner syndrome is unilateral ptosis, miosis, and anhidrosis, but the ptosis (slight narrowing of the ocular fissure) and miosis (smaller pupil on the affected side) are far more commonly recognized than anhidrosis (lack of perspiration on the forehead or face). Horner syndrome is not likely to cause any functional visual disturbance but is of great importance clinically as a “red flag” warning that the oculosympathetic pathway has been interrupted. This pathway is a three-neuron chain that originates in the hypothalamus in the brainstem, travels down the spinal cord to the lower cervical and upper thoracic levels, then traverses the upper chest cavity and apex of the lung, traveling with the carotid artery into the cavernous sinus, traversing the orbit to innervate the pupillary sphincter; it also branches to innervate accessory muscles for eyelid retraction. This long, circuitous route covers a lot of anatomy, and a patient with a lesion anywhere along its course can present with Horner syndrome. Therefore, this syndrome with minimal symptoms and often subtle findings is of great importance in the diagnosis of potentially life-threatening lesions in the head, neck, and chest!1 Horner syndrome bears the name of Johann Friedrich Horner (1831−1886), a Swiss ophthalmologist who published a case report in 1869 describing a 40 year old woman with unilateral miosis, ptosis, and facial anhidrosis. However, there were case reports of the syndrome that predated Horner’s by many years: a report by Edward Selleck Hare in 1838 and another by Silas Weir Mitchell in 1864. The French ophthalmologist Claude Bernard was the first to identify the triad of findings as the manifestations of an oculosympathetic paresis in animal studies © XXXX American Chemical Society

Received: October 25, 2017 Accepted: November 30, 2017

A

DOI: 10.1021/acschemneuro.7b00405 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Review

ACS Chemical Neuroscience input to the pupil, including the amount of light entering the eyes, the state of accommodative tone (looking at something close versus far away), but also emotional factors, systemic medications, and disease states. In general, the autonomic innervation controlling the pupil is equal in both eyes, so both pupils are normally the same size as each other, even as they change dynamically throughout the day. In the case of Horner syndrome, there is a loss of sympathetic tone to the pupillary dilator in one eye. Therefore, the pupil in the affected eye is smaller than the pupil in the opposite eye, as parasympathetic tone to the pupillary sphincter is relatively unopposed. The degree of anisocoria in Horner syndrome is greater in darkness than in bright light; in fact, anisocoria can be missed altogether if the pupils are only observed in bright light.6 This is because parasympathetic tone (to the pupillary constrictor) is maximized and sympathetic tone is minimized in bright light; thus, the difference in pupil size is not as noticeable. However, in darkness, the pupillary dilator is activated, and the anisocoria becomes greater as the affected side fails to dilate as well as the opposite side. It is important to note that the pupil will dilate in darkness, even in Horner syndrome, as much of the dilation in darkness comes from relaxation of the powerful pupillary sphincter, allowing the mechanical elastic forces of the iris to open the pupil. This passive dilation of the pupil is much slower than the dilation of the pupil when the sympathetic system activates the dilator muscles. This is why the pupil in Horner syndrome is said to have a “dilation lag”: minimal anisocoria in bright light, a larger degree of anisocoria when the lights are turned off for the first 5 s and then less anisocoria after 10−15 s as passive dilation in the affected eye gradually catches up with the size of the pupil in the normal eye. This is best observed by evaluating pupil size first in bright light and then in darkness as the room lights are suddenly turned off (the pupils are observed in relative darkness by illuminating the patient’s eyes with a penlight tangentially from below).7 The dilation lag can also be recorded by taking photographs of the pupils in darkness after five seconds and then again at 15 s.8 Digital pupillography can also be used to plot the pupillary dilation in darkness, producing a characteristic dilation lag pattern.9 The dynamic nature of the dilation lag needs to be understood by the clinician, as it is critical that the pupils be observed as they are in the process of dilating, particularly in the first 5−15 s after the lights are turned off (Figure 1). The dilation lag is very characteristic of Horner syndrome10 but may not always be present.6,11 The anisocoria can be further exaggerated by giving the patient a “scare” a few seconds after the lights are turned off. A loud noise or other stimulus that causes a systemic sympathetic discharge will exaggerate dilation in the unaffected eye, maximizing anisocoria. Pinching the patient (not recommended!) has the same effect.12 These noxious stimuli techniques are more of an intriguing side note than a clinically practical tool. In summary, the pupillary abnormality Horner syndrome is an anisocoria with a smaller pupil on the affected side. This is best identified by comparing the degree of anisocoria in bright light and in darkness with the greatest degree of anisocoria expected 5−7 s after the lights are turned off because of a deficiency of sympathetic tone to the dilator muscle in the affected eye. Note that an anisocoria that is greater in bright light than in darkness implicates that the larger of the two pupils is abnormal, suggesting a parasympathetic deficiency, as occurs with a third cranial nerve palsy or Adie tonic pupil. Ptosis. Increased sympathetic tone to the eyelids causes a slight retraction of the eyelids making the palpebral fissure

Figure 1. A 22 year old with right Horner syndrome. Photograph in room light shows anisocoria with the right pupil smaller than the left and subtle ptosis of the right upper eyelid; also note the lower lid sits higher on the globe (“upside-down ptosis”) (A). In bright light, the anisocoria is minimized (B) and is greatest after ∼5 s in the dark (C) with lessening of the anisocoria at 15 s as the right pupil “catches up” (D). Forty-five minutes after apraclonidine eye drops were instilled in both eyes, the anisocoria is reversed due to supersensitivity of the pupillary dilator in the right eye from Horner syndrome (E). Note that the ptosis is also reversed in this case due to similar supersensitivity of the sympathetically innervated eyelid retractors; however, the ptosis reversal alone is not reliable enough to act as an indicator of a positive test.

wider (as expected because wide open eyes are a part of the “fight or flight” sympathetic response). With a deficiency in oculosympathetic tone in Horner syndrome, there is the opposite effect, a narrowing of the palpebral fissure on the affected side. The primary elevator of the eyelid is the levator palpebrae muscle, which is innervated by the third cranial nerve. HowB

DOI: 10.1021/acschemneuro.7b00405 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Review

ACS Chemical Neuroscience

suggested for identifying this finding clinically, such as sliding a plastic prism bar across the skin of the brow on one side versus the other (it will slide more smoothly where there is no perspiration),17 or applying a powder to the face or body that will change color depending on degree of skin moisture.18 In the acute setting, loss of vasomotor control from sympathetic denervation can cause facial flushing, conjunctival hyperemia, tearing, and even nasal stuffiness with consequent dilation of the vasculature. Later, the skin may be paler than the normal side with vasoconstriction caused by denervation supersensitivity of the vasculature to normally circulating adrenergic elements.7 The Harlequin sign is a striking manifestation of loss of sympathetic vasomotor innervation with hemifacial flushing that respects the vertical midline, typically in infants with an oculosympathetic paresis.19 Neuroanatomy and Clinical Implications. As noted in the Introduction, the oculosympathetic “chain” is a three-neuron pathway: The first-order neurons are in the hypothalamus with axons traveling through the brainstem and spinal cord to synapse in the lower cervical/upper thoracic spinal cord. Second-order axons originate from this spinal nucleus and travel through the upper chest cavity to synapse in the superior cervical ganglion; then, third-order axons originating in the superior cervical ganglion travel along the carotid artery system to reach the orbit and eye. Horner syndrome can result from lesions anywhere along this pathway, which is logically divided into central or first-order, preganglionic (proximal to the superior cervical ganglion) or second-order, and postganglionic or third-order neuron regions (Figure 3). One study demonstrated that 65% of patients presenting with Horner syndrome were found to have an identifiable cause with 13% of these having a central lesion, 44% with a preganglionic lesion, and 43% with a postganglionic lesion.16 First-Order (Central) Horner Syndrome. The cell bodies of first-order neurons reside in the hypothalamus. From there, axons descend through the brainstem and down the spinal cord to synapse in the ciliospinal center of Budge−Waller located at the level of C8 to T2. Therefore, lesions in the brainstem and cervical cord can present with a first-order neuron Horner syndrome often called a “central” Horner syndrome (Table 1).20 This is rarely a cause of isolated Horner syndrome given the proximity of this pathway to important brainstem structures, though there are exceptions.21 Hypothalamic/thalamic lesions (tumor, infarct, hemorrhage) can cause ipsilateral Horner syndrome associated with contralateral hemiparesis and hypesthesia22−24 Lesions of the dorsal midbrain can result in the combination of an ipsilateral Horner syndrome with a contralateral fourth cranial nerve palsy because the fourth cranial nerve fibers cross to the contralateral side as they exit the brainstem.25 Lesions affecting the pons can produce Horner syndrome associated with an ipsilateral or bilateral abducens palsy.26 The most commonly recognized central Horner syndrome occurs in the setting of a lateral medullary plate syndrome from infarction (rarely demyelination), producing a constellation of symptoms called Wallenberg syndrome: Horner syndrome, ipsilateral ataxia, and contralateral hypalgesia; nystagmus, facial weakness, dysphagia, and vertigo may also be present.27 Finally, cervical or upper thoracic spinal cord lesions (trauma, demyelination, tumor, syrinx, or vascular causes) can cause an isolated Horner syndrome28 but most often are associated with long tract signs or spinal cord syndromes such as Brown− Séquard syndrome.7

ever, a small amount of eyelid elevation is produced by Mueller’s muscle, a small sympathetically innervated eyelid retractor that contributes ∼1−2 mm of lift to the upper eyelid.13 Therefore, with an oculosympathetic paresis, there is a 1−2 mm ptosis of the upper eyelid, narrowing the ocular fissure. A ptosis greater than 1−2 mm cannot be fully explained by Horner syndrome. There is a similar rudimentary muscle in the lower lid; thus, the lower lid may be slightly elevated in Horner syndrome (sometimes called “upside-down ptosis”).14 This may only be evident by comparing the position of the lower lid to the limbus between the two eyes. The end result of a loss of sympathetic tone to the eye is a narrowing of the ocular fissure. The narrowed fissure can give the appearance of enophthalmos, though there is no true measurable enophthalmos in Horner syndrome.15 The ptosis in Horner syndrome may be variable and subtle; one study noted that ptosis was absent in 12% of patients.16 Sudomotor and Vasomotor Deficiency. The oculosympathetic system also carries sudomotor fibers (for perspiration) to the face. Interruption of this pathway can cause a deficiency in sweating (anhidrosis) on the affected side. Most of the face is innervated by sudomotor fibers that travel with the common carotid via the external carotid; however, fibers that travel with the internal carotid supply a small area on the forehead and side of the nose. Therefore, lesions involving the superior cervical ganglion or more proximal pathway can result in ipsilateral facial anhidrosis (Figure 2), but more distal lesions along the

Figure 2. A 45 year old man presenting with right Horner syndrome. He had a right paraspinal cervical abscess as a late complication of previous spine surgery producing a right preganglionic Horner syndrome. This clinical photograph was taken at a time when the patient had fever and was diaphoretic, revealing right hemifacial anhidrosis.

third-order neuron will only result in anhidrosis of a small patch of skin on the forehead above the brow. Theoretically, Horner syndrome from lesions in the brainstem or spinal cord could even produce hemibody anhidrosis. However, anhidrosis is almost never evident in the environmentally controlled clinical setting and only rarely recognized by patients. Because of this difference in facial perspiration, women may sometimes note a difference when makeup is applied on one side of the face versus the other. In the past, elaborate methods have been C

DOI: 10.1021/acschemneuro.7b00405 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Review

ACS Chemical Neuroscience

Figure 3. Oculosympathetic pathway. Reproduced with permission from Martin, T.J.; Corbett, J.J. Practical Neuroophthalmology; McGraw Hill, 2013. This illustration was originally redrawn with permission from Weinstein, J.M.; The pupil. In Slamovits, T.L., Burde, R., associate eds.; Neuroophthalmology, vol. 6. In Podos, S.M., Yanoff, M., eds.; Textbook of ophthalmology, St. Louis, 1991, Mosby.

Second-Order (Preganglionic) Horner Syndrome. Second-order neuron lesions are sometimes referred to as “preganglionic” because they are proximal to the superior orbital ganglion. In one study, up to 25% of preganglionic Horner syndrome were the result of malignancy.29 Another study showed that 28% of preganglionic Horner syndrome may have no identifiable etiology.5 Second-order neurons originate in the ciliospinal center of Budge−Waller, a spinal nucleus extending from C8 to T2. They exit the spinal cord as a part of the paraspinal sympathetic plexus. The axons form a sympathetic nerve, traveling under the aorta, draping over the apex of the lung before passing through the stellate ganglion and up the carotid sheath, synapsing in the superior cervical ganglion at the level of the carotid bifurcation. For this reason, lesions in the upper chest cavity can result in Horner syndrome (Figure 4). The classic example is the Pancoast tumor, an apical lung tumor resulting in Horner syndrome and sometimes also producing ipsilateral shoulder pain and signs and symptoms associated with thoracic outlet syndrome.30 Trauma, including injury to the brachial plexus or soft tissue of the neck, or pneumothorax can also cause Horner syndrome. Surgical procedures involving upper chest (including central

lines and anesthetic procedures) or the neck, such as cervical spine surgery or thyroid surgery, can be an iatrogenic cause of Horner syndrome. Third-Order (Postganglionic) Horner Syndrome. Third-order neurons have their cell bodies in the superior cervical ganglion located at the bifurcation of the carotid artery near the angle of the jaw. Axons travel to their final destination not as a single nerve but as a net or plexus (rete) that surrounds the common carotid artery and then the internal carotid artery to reach the eye (or the external carotid artery to supply the face). This plexus of oculosympathetic nerves travels with the carotid artery into the cavernous sinus. Within the cavernous sinus, the plexus coalesces to form a well-defined nerve that travels with the sixth cranial nerve through the middle of the cavernous sinus (unlike cranial nerves III, IV, and V1 that travel relatively protected in the wall of the cavernous sinus). Sympathetic fibers then travel with the first division of the fifth cranial nerve through the superior orbital fissure into the orbit. Sympathetic fibers destined to innervate the pupillary dilator travel through the ciliary ganglion (without synapsing), via the long ciliary nerves that penetrate the sclera near the optic nerve, and then through the suprachoroidal space to segmentally D

DOI: 10.1021/acschemneuro.7b00405 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Review

ACS Chemical Neuroscience Table 1. Anatomic Location and Pathological Processes in Horner Syndrome anatomic location first-order (central)

hypothalamus mesencephalon pons medulla spinal cord

second-order (preganglionic)

thoracic cavity

cervical sympathetic chain neck third-order (postganglionic)

superior cervical ganglion carotid artery cavernous sinus skull base orbit/superior orbital fissure other or unknown

types of lesions

associated symptoms

infarction, hemorrhage, tumor as above, also demyelination as above, more specifically infarction, arterial dissection, cardiac embolism, rarely demyelination trauma, infarction, vascular malformation, demyelination, tumor, inflammatory or infectious myelitis, syrinx, syringomyelia, cervical disc herniation breast and lung cancer, mediastinal masses, chest surgery, thoracic aortic aneurysm, central vascular access, trauma apical lung lesions: non-small-cell lung carcinoma, other tumors, metastatic disease, infection neuroblastoma, schwannoma, neuroectodermal tumor, vagal paraganglioma, mediastinal tumors, cysts

contralateral hemiparesis and/or hypesthesia contralateral trochlear nerve palsy ipsilateral or bilateral abducens palsy Wallenberg syndrome radicular signs, Brown−Séquard syndrome, alternating Horner syndrome

Pancoast syndrome

trauma, abscess, tumor, lymphadenopathy, thyroid neoplasm, thyroidectomy, radical neck surgery, central vascular access, cervical rib trauma, jugular venous ectasia, surgical neck dissection penetrating intraoral injury, intraoral surgery, tonsillectomy traumatic or spontaneous dissection, aneurysm, fibromuscular dysplasia, Ehlers− Danlos syndrome, Marfan syndrome, arteritis cavernous carotid aneurysm, tumor, thrombosis mass lesion, basilar skull fracture

facial pain, stroke, ocular ischemia, cerebral ischemic symptoms Abducens and other ocular motor palsy cranial nerve deficits, trigeminal pain and sensory loss

herpes zoster, nasopharyngeal carcinoma cluster headache trigeminal autonomic cephalgias microvascular ischemia, giant cell arteritis, autonomic neuropathies

severe transient unilateral headache with tearing, conjunctival injection, nasal stuffiness ipsilateral trigeminal neuralgia

Horner Syndrome in Children. Although Horner syndrome identified during the first year of life is most often idiopathic (70% in one study),41 the fact that it can be associated with potentially fatal neuroblastoma means that Horner syndrome in children is a serious matter. It is important to distinguish between congenital and acquired Horner syndrome. More than half of infants with acquired Horner syndrome during the first year of life had an associated underlying potentially fatal disorder in one study.42 Horner syndrome at birth is often idiopathic or associated with birth trauma (often accompanied by brachial plexus injury);42 however, congenital Horner syndrome can also be a sign of intrauterine neuroblastoma.43 Early diagnosis is important as survival declines if the diagnosis is made after the first year of life.44 In addition to ptosis and anisocoria, patients who had Horner syndrome at birth (or acquired during early childhood) can have a lighter-colored iris on the affected side (iris heterochromia) because sympathetic innervation plays an important role in the development of iris melanocytes, which ultimately determine iris color.42 Harlequin sign is a striking presentation of Horner syndrome in infancy with contralateral facial flushing with absolute respect of the vertical midline.19 One study demonstrated that in a group of pediatric patients with Horner syndrome, 40% were congenital, 42% were acquired after a surgical procedure in the thorax, neck, or central nervous system, and 15% were acquired from causes that included neuroblastoma, spinal cord tumors, rhabdomyosarcoma embryonal cell carcinoma, vascular malformations, intrathoracic aneurysm, and trauma.42 Children with Horner syndrome without obvious surgical or traumatic cause require a thorough investigation for a mass lesion including MRI of brain, neck, and chest with and without contrast and urinary catecholamine testing.45

innervate the dilator muscles of the iris. Therefore, pathology in the neck, skull base, and orbit can cause “postganglionic” (thirdorder) Horner syndrome. As noted previously, the sudomotor fibers that supply the majority of the face travel with the external carotid with only a small portion traveling with the internal carotid to supply a patch of skin above the brow (and side of the nose). Therefore, when anhidrosis is identifiable in postganglionic Horner syndrome, the only affected area is a small patch above the brow. On the other hand, pathology affecting the superior cervical ganglion or more proximal may cause hemifacial anhidrosis, though this is rarely recognized clinically (see Figure 2). One of the most common causes of postganglionic Horner syndrome is carotid dissection with Horner syndrome occurring in 20 to 30% of patients with this condition.31,32 Carotid dissection can occur as a result of trauma (including chiropractic manipulation)33,34 but can also occur spontaneously. In carotid dissection, tears in the intimal wall allow blood to enter into the wall of the carotid artery.35 This results in a narrowing of the lumen and occlusion of carotid branches but also an increase in the diameter of the carotid artery, stretching and breaking the plexus of sympathetic nerves. The resulting Horner syndrome is usually accompanied by pain (in the neck, eye, ear, teeth, or head) and other neurologic findings.31,36 Other causes of postganglionic Horner syndrome include cluster headaches (both transient with episodes and chronic after repeated episodes)37 or less commonly tumors of the skull base (often accompanied by facial pain or anesthesia from trigeminal involvement).38 Cavernous sinus lesions, such as cavernous sinus carotid aneurysms, classically cause ipsilateral Horner syndrome associated with a sixth nerve palsy due to the close association of oculosympathetic fibers with the sixth cranial nerve running through the cavernous sinus.39,40 E

DOI: 10.1021/acschemneuro.7b00405 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Review

ACS Chemical Neuroscience

disease can also create a small pupil in the setting of anisocoria. Many of these entities can be identified on the slit lamp examination or are evident in the patient’s history. Horner syndrome is actually an uncommon cause of ptosis; aging changes in the eyelid causing a mechanical drooping (levator dehiscence) is the most common cause of ptosis in older patients. This can also occur in young patients who wear contact lenses, particularly hard contact lenses. Ocular myasthenia gravis and other neuromuscular conditions can also produce ptosis. Clinical Evaluation of Horner Syndrome. The first, and perhaps most important, aspect of evaluating suspected Horner syndrome is a careful history. The history often reveals an obvious cause, such as trauma or neck/chest surgery, or will show that the Horner syndrome had been present for many years and therefore may not require an extensive investigation. The history may also reveal other symptoms that would help localize the lesion causing the Horner syndrome. A careful examination is obviously also critical, as other causes of anisocoria or ptosis may need to be considered rather than Horner syndrome, and concomitant signs (for example, a sixth nerve paresis) may help localize a potential lesion. After this, pharmacologic testing can be helpful to confirm Horner syndrome, as discussed extensively below. Next, all this information needs to be combined to decide if additional investigation is necessary, and if so decide on the best imaging strategy to evaluate the suspected cause of Horner syndrome. Though there is some disagreement on the exact interval, many clinicians will elect not to evaluate Horner syndrome if it has been present for greater than two years. For this reason, evaluating old pictures, particularly those with sufficient resolution to see the pupils clearly, may avert an extensive and expensive investigation. So-called “FAT scan” (family album tomography) can be greatly beneficial and easier to obtain in this digital age. Usually, a driver’s license is not sufficient to see the pupils, though is helpful in showing ptosis. The best pictures are usually graduation or school pictures. Pharmacologic Testing. The diagnosis of Horner syndrome may be convincing based on clinical examination and history alone, but often, the diagnosis is uncertain. Pharmacologic testing can be performed in the clinic by placing eye drops into the eyes to confirm the presence of Horner syndrome. Apraclonidine is now the most commonly used agent, largely supplanting cocaine eye drops as the pharmacologic test of choice for Horner syndrome in adults. Hydroxyamphetamine drops can be used to distinguish third-order neuron lesions from first- and second-order lesions but are used infrequently as the entire oculosympathetic pathway is usually imaged when Horner syndrome is being evaluated. Activation of the oculosympathetic end organs in the eye (the pupillary dilator and Mueller’s muscle in the eyelid) occurs when norepinephrine is released by the terminal axon into the synaptic cleft, which then binds to receptors on the postsynaptic cell membrane to initiate activation of the muscle (pupil dilator or eyelid retractor). The amount of norepinephrine in the synapse is increased by release from the presynaptic bulb in response to neuronal activation. Norepinephrine is also continuously reabsorbed and recycled by the presynaptic cleft. In Horner syndrome, when there is no oculosympathetic activation, there is no release of norepinephrine into the synaptic cleft. The postsynaptic membrane responds to this lack of stimulation by upregulation, actually increasing the postsynaptic receptors for norepinephrine (α-1 receptors). Pharmacologic

Figure 4. This 65 year old patient presented with a left Horner syndrome from a left upper mediastinal mass (arrow, A) shown to be a schwannoma following resection. The Horner syndrome persisted following resection. Note the 2 mm ptosis of the left upper lid and anisocoria with a smaller pupil on the left side (B). Forty-five min after instilling apraclonidine eye drops, the anisocoria reversed with prompt dilation of the left pupil, confirming the presence of oculosympathetic paresis (Horner syndrome) (C). Hydroxyamphetamine testing was not performed, but one would anticipate that both pupils would dilate well because the third-order neuron is intact in this example of preganglionic Horner syndrome.

Differential Diagnosis. There are many causes of anisocoria and many causes of ptosis. Occasionally, these two findings occur together for other reasons (pseudo-Horner syndrome). Therefore, a working knowledge of the differential diagnosis of these two entities is important to arrive at a correct diagnosis. When the pupils are different sizes (anisocoria), the first step is to determine which pupil is the abnormal one: Is the larger pupil too big or is the smaller pupil too small? This is determined by looking at the degree of anisocoria in bright light and in darkness. Obviously, if the large pupil is implicated as the abnormal one (anisocoria greatest in bright light), then Horner syndrome is not a consideration and attention should be turned to disorders of the parasympathetic system, such as Adie tonic pupil or third cranial nerve palsy. Horner syndrome is a consideration when the smaller pupil is identified as abnormal (anisocoria greatest in darkness when compared to light). However, there are other entities that can cause an abnormally small pupil. Essential anisocoria is a small anisocoria (usually less than 0.5 mm) not associated with disease that can be identified in 15−30% of the normal population;46 the anisocoria is usually about the same in darkness and in light, unlike pathologic states.5 Some topical eye drops, contamination of the tear film with systemic medications, and local ocular disease such as uveitis, eye trauma, and diabetic eye F

DOI: 10.1021/acschemneuro.7b00405 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Review

ACS Chemical Neuroscience agents used to tests for Horner syndrome exploit these mechanisms to produce clinical tests that can confirm the presence of an oculosympathetic paresis or localize the lesion within the oculosympathetic chain.1 Apraclonidine. Apraclonidine is an α-2 adrenergic agonist with weak α-1 activity. Clinically, it is an agent used to lower intraocular pressure due to its α-2 agonist properties. Normally, it has a negligible effect on pupil size; however, in patients with Horner syndrome, it reliably dilates the pupil because of supersensitivity of the iris dilator muscle from upregulation of α-1 postsynaptic receptors. This clinically useful property was serendipitously discovered when patients with Horner syndrome were used to evaluate the intraocular pressure-lowering properties of apraclonidine. Patients with Horner syndrome were specifically chosen because any pressure-lowering effect would be entirely from local effects of the drug rather than any secondary systemic absorption and a sympathetic-mediated effect. It was quickly realized that this unanticipated dilating effect in an eye with Horner syndrome could provide a useful pharmacologic diagnostic test,47 and indeed, this has proven to be the case. A number of clinical studies show that apraclonidine is both sensitive and specific for diagnosing Horner syndrome,48−51 largely supplanting cocaine, which has historically been the diagnostic eye drop of choice (see Figure 4). Apraclonidine has many advantages over cocaine eye drops: The end point is obvious as it creates a reversal of the anisocoria (unlike cocaine in which a positive test is when cocaine fails to dilate the affected pupil); apraclonidine is readily commercially available and does not have the stringent storage and use regulations of cocaine. The test is performed by placing 0.5% apraclonidine in both eyes. The test is considered positive for the presence of Horner syndrome if the anisocoria reverses: the smaller pupil in the eye suspected of having the oculosympathetic paresis becomes larger than the opposite eye at the end of 45 min (Box 1). One potential drawback to this test is that it requires supersensitivity to be present, which means that an oculosympathetic paresis has been present for a sufficient time for upregulation of postsynaptic receptors to occur. Therefore, the test is not useful in a suspected acute Horner syndrome but is generally thought to be reliable after 2 weeks from onset of symptoms.52 Because there are a few reported cases in which the test was positive even within a few days,21,53 it is reasonable to try apraclonidine testing in the acute period. A positive test should be considered diagnostic, but a negative test in the acute period has no meaning and cocaine testing should be performed.6 There are safety concerns with regard to the use of apraclonidine eye drops in infants, including drowsiness and even unresponsiveness,54 consistent with the known effects of other adrenergic receptor agonists.55 Though with limited numbers, other reports suggest apraclonidine 0.5% is safe in children younger than 10 years old.22 However, given the current reports in the literature, there is a general consensus that cocaine is the preferred choice for pharmacologic testing of Horner syndrome in infants and young children.45,56 Cocaine. Until the past decade, cocaine eye drops were the agent of choice for diagnosing Horner syndrome.57,58 Cocaine blocks the uptake of norepinephrine by the presynaptic membrane. Therefore, it increases the concentration of norepinephrine in the synaptic cleft and causes pupillary dilation in the normal eye. This is because in the normal eye there is always some baseline sympathetic tone and therefore some release of norepinephrine from the presynaptic cleft, balanced by the

Box 1. Pharmacologic Testing for Horner Syndrome Apraclonidine (Iopidine) 0.5% is readily available commercially and is frequently used in the ophthalmologist office as a pressure lowering agent prior to laser procedures such as YAG capsulotomy. It has emerged as the preferred agent for confirming Horner syndrome but should not be used in infants and young children because of reported respiratory depression. Because it depends on the development of denervation supersensitivity, this test may produce a false negative result if performed within 2 weeks after the onset of Horner syndrome. A positive test is indicated by reversal of the anisocoria, providing an unambiguous end point. Cocaine hydrochloride 10% ophthalmic eye drops were the standard for confirming Horner syndrome prior to apraclonidine and remains the preferred agent in infants and young children. This test should be valid even in the acute phase of Horner syndrome. However, because this agent is a controlled substance, there are many obstacles with compounding and storage. Patients should be notified that cocaine eye drops can produce a positive urine drug screen. A positive cocaine test for Horner syndrome is indicated by the failure of the affected pupil to dilate with anisocoria of at least 1 mm remaining at the end of the test. Hydroxyamphetamine 1% solution is no longer commercially available and thus must be compounded by a pharmacy. After Horner syndrome has been confirmed, this agent can be used to distinguish pre- from postganglionic Horner syndrome, as it reliably dilates any pupil with an intact third-order neuron (even a preganglionic Horner syndrome pupil) but will not dilate a pupil with loss of the third-order neuron. If this test is to be performed after apraclonidine or cocaine testing, one should wait at least 48 h. Therefore, the failure of a pupil with Horner syndrome to dilate with hydroxyamphetamine indicates a postganglionic Horner syndrome. These topical pharmacologic tests are most accurate if they are performed in eyes that have not had instrumentation (such as applanation tonometry) or other eye drops at the time of the evaluation. The tests are performed by placing a single drop of the pharmacologic agent in the inferior cul-desac in each eye with care to instill the same amount in each eye. A second application is performed after several minutes. Systemic absorption can be minimized by having the patient close his/her eyes for several minutes after the drop is instilled. After 45 min, the end point is evaluated. External photography showing the pupils and eyelids is the best way to document the pretest baseline and the test result; serving as a record for interpretation and for medical documentation. reuptake of norepinephrine. Therefore, blocking the reuptake increases the amount of norepinephrine in the synaptic cleft, activating the end organ (dilating the pupil, raising the eyelid). However, in a patient with Horner syndrome with no oculosympathetic outflow, there is no (or far less) norepinephrine present in the synaptic cleft. Therefore, blocking the uptake of norepinephrine with cocaine would have no effect on pupil size. Therefore, this test relies on comparing the dilation response to cocaine eye drops in an affected eye versus the normal eye. Cocaine 10% is placed in both eyes, and the test is considered positive for Horner syndrome if the affected eye does not dilate as well as the normal eye. Clinical tests have shown that a practical threshold for considering a test positive is a remaining G

DOI: 10.1021/acschemneuro.7b00405 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Review

ACS Chemical Neuroscience anisocoria of at least 0.8 mm after 45 min59 or negative if the suspected pupil dilates more than 2 mm.6 Several potential problems are immediately evident. First, a positive end point is not much different from the starting point; for example, mechanical issues that keep a pupil from dilating (such as posterior synechiae or neovascularization of the iris), or eye drops without active cocaine for whatever reason, can produce a false positive test. In addition, the compounding, storage, and use of a controlled substance such as cocaine are difficult from a practical perspective. It also has a short shelf life, and metabolites can remain in the urine for up to 2 days causing potential problems for patients who may have to have a drug screening test.60 Therefore, given the limitations noted above, it is no surprise that apraclonidine has essentially replaced this historically important test in adults (but not infants, where cocaine is deemed safer). Hydroxyamphetamine. Hydroxyamphetamine dilates a normal pupil as it forces presynaptic norepinephrine into the synaptic cleft regardless of sympathetic activation.61 Hydroxyamphetamine will even dilate a patient with Horner syndrome but only if the third-order neuron is intact. Dilation of a pupil affected with an oculosympathetic paresis with hydroxyamphetamine 0.5% demonstrates that the lesion is a first- or secondorder neuron process because the third-order neuron is shown to be intact. Failure to dilate with hydroxyamphetamine demonstrates that the third-order neuron is not intact, as there is no norepinephrine present at the terminal synapse, and is therefore diagnostic of a postganglionic Horner syndrome.62,63 It is important to note that a false-negative test can occur if hydroxyamphetamine testing is performed too soon after the third-order neuron is affected (within 2−3 weeks of symptom onset), before atrophy of the presynaptic bulb has occurred.64 Furthermore, transynaptic degeneration of the third-order neuron can occur even from preganglionic lesions in congenital Horner syndrome (or if acquired in the first year of life), which may falsely implicate a third-order (postganglionic) lesion in infants.65 Hydroxyamphetamine testing should not be performed within 48 h of apraclonidine or cocaine eye drop tests. There is no doubt that the hydroxyamphetamine test is an elegant method for refining the location of a lesion causing Horner syndrome. However, most clinicians insist on imaging the entire oculosympathetic pathway when a workup is indicated in Horner syndrome, so hydroxyamphetamine testing is not routinely performed.66 In addition to hydroxyamphetamine, there is some discussion in the literature that a first-order (central) Horner syndrome will not develop pupillary supersensitivity to the same degree as second- and third-order lesions. These authors advocate the use of very dilute phenylephrine (1%) or epinephrine (2%) eye drops to separate these groups67 but this method has not proven to be clinically reliable.7,68 Pharmacologic Testing Overview. All things considered, using the apraclonidine test to test for Horner syndrome is by far the easiest and most reliable test in adults. As noted above, it does not work in acute Horner syndrome, as it takes several weeks for supersensitivity to develop. If the diagnosis is required on an acute basis, cocaine testing may still have some utility despite the potential drawbacks discussed above. Furthermore, cocaine is preferred over apraclonidine for diagnosing Horner syndrome in infants and young children due to potential side effects of apraclonidine. Once Horner syndrome is confirmed, hydroxyamphetamine eye drops can be used to distinguish between pre- and postganglionic lesions. Hydroxyamphetamine

testing has some academic interest but usually does not change the general indication for evaluating the entire oculosympathetic pathway with imaging; thus, this step is often omitted. Radiologic Investigation of Horner Syndrome. Imaging is the primary investigative tool in Horner syndrome. The first question to be answered by the clinician is whether or not further investigation with imaging is even warranted. Patients with long-standing isolated Horner syndrome (greater than two years), or patients with an obvious cause of Horner syndrome, may not require further investigation.69 When imaging is indicated, the next question is which study or studies to order, particularly given the large anatomic area that could harbor a lesion. Imaging recommendations for evaluating Horner syndrome in the literature range from “one size fits all” to elaborate decision trees that precisely tailor the study. In practice, the imaging method depends on additional localizing signs and symptoms and if the presentation of Horner syndrome is acute or chronic.70−72 Sometimes the patient has additional symptoms or signs that localize the cause of Horner syndrome with a reasonable degree of certainty. For example, a patient with Horner syndrome and neurologic symptoms of Wallenberg syndrome should have an MRI of the head with and without contrast. A patient with acute Horner syndrome and pain in the neck and face should have an immediate CTA or MRA/MRI of the neck and head for suspicion of a carotid dissection. A patient with arm pain and a smoking history should have a CT of the chest or other study directed at a suspected lung apex tumor. If directed studies are negative, then any remaining portion of the oculosympathetic pathway not adequately addressed should be imaged. However, most often patients present with an isolated Horner syndrome without other convincing localizing signs or symptoms. In these cases, most clinicians advocate imaging the entire oculosympathetic pathway (regardless of hydroxyamphetamine localization, if performed). In one study of 88 patients with isolated Horner syndrome, 20% had a causative lesion on neuroimaging with carotid dissection being the most common.32 The imaging choice depends on whether the symptoms are acute or chronic. In the acute setting, an urgent CTA study that includes the circle of Willis and orbits down to the level of the aortic arch is recommended. This study allows excellent imaging of the vascular structures for the possibility of carotid dissection or other vascular disorders but also allows for visualization of the lung apices and soft tissues of the neck and orbit. This protocol has a number of advantages, including covering the entire pathway and minimal imaging time to reduce movement artifact.7 Patients who present with nonacute Horner syndrome, with no pain, no localizing signs, and without a history of trauma, may not require urgent imaging.73 In this setting, a contrastenhanced MRI with or without MRA including the brain and neck from the hypothalamus to the T2 level in the chest has been advocated.66,70 MRI offers better imaging of the brainstem, hypothalamus, cervical cord, and brachial plexus compared to CT imaging. Children with Horner syndrome require MRI to best visualize the sympathetic chain with some experts saying that the MRI scan should also extend into the abdomen and pelvis73 to fully investigate the possibility of neuroblastoma. Additional recommended studies include spot urine catecholamines, VMA, and homovanillic acid.45 Management/Treatment. Horner syndrome usually does not result in loss of function. Having a smaller pupil does not H

DOI: 10.1021/acschemneuro.7b00405 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Review

ACS Chemical Neuroscience

(11) Crippa, S. V., Borruat, F.-X., and Kawasaki, A. (2007) Pupillary dilation lag is intermittently present in patients with a stable oculosympathetic defect (Horner syndrome). Am. J. Ophthalmol. 143, 712−715. (12) Reeves, A. G., and Posner, J. B. (1969) The ciliospinal response in man. Neurology 19, 1145−1152. (13) Beard, C. (1985) Muller’s superior tarsal muscle: anatomy, physiology, and clinical significance. Ann. Plast. Surg. 14, 324−333. (14) Nielsen, P. J. (1983) Upside down ptosis in patients with Horner’s syndrome. Acta Ophthalmol. 61, 952−957. (15) van der Wiel, H. L., and van Gijn, J. (1987) No enophthalmos in Horner’s syndrome. J. Neurol., Neurosurg. Psychiatry 50, 498−499. (16) Maloney, W. F., Younge, B. R., and Moyer, N. J. (1980) Evaluation of the causes and accuracy of pharmacologic localization in Horner’s syndrome. Am. J. Ophthalmol. 90, 394−402. (17) Rosenberg, M. L. (1989) The friction sweat test as a new method for detecting facial anhidrosis in patients with Horner’s syndrome. Am. J. Ophthalmol. 108, 443−447. (18) Wolfe, G. I., Galetta, S. L., Teener, J. W., Katz, J. S., and Bird, S. J. (1995) Site of autonomic dysfunction in a patient with Ross’ syndrome and postganglionic Horner’s syndrome. Neurology 45, 2094−2096. (19) Abe, M., et al. (2006) Harlequin sign (hemifacial flushing and contralateral hypohidrosis) in a. Pediatr. Dermatol. 23, 358−360. (20) Nagy, A. N., Hayman, L. A., Diaz-Marchan, P. J., and Lee, A. G. (1997) Horner’s syndrome due to first-order neuron lesions of the oculosympathetic pathway. AJR, Am. J. Roentgenol. 169, 581−584. (21) de Seze, J., et al. (2006) Unusual ocular motor findings in multiple sclerosis. J. Neurol. Sci. 243, 91−95. (22) Stone, W. M., de Toledo, J., and Romanul, F. C. (1986) Horner’s syndrome due to hypothalamic infarction. Clinical, radiologic, and pathologic correlations. Arch. Neurol. 43, 199−200. (23) Austin, C. P., and Lessell, S. (1991) Horner’s syndrome from hypothalamic infarction. Arch. Neurol. 48, 332−334. (24) Rossetti, A. O., Reichhart, M. D., and Bogousslavsky, J. (2003) Central Horner’s syndrome with contralateral ataxic hemiparesis: a diencephalic alternate syndrome. Neurology 61, 334−338. (25) Guy, J., Day, A. L., Mickle, J. P., and Schatz, N. J. (1989) Contralateral trochlear nerve paresis and ipsilateral Horner’s syndrome. Am. J. Ophthalmol. 107, 73−76. (26) Kellen, R. I., Burde, R. M., Hodges, F. J., 3rd, and Roper-Hall, G. (1988) Central bilateral sixth nerve palsy associated with a unilateral preganglionic Horner’s syndrome. J. Clin. Neuroophthalmol. 8, 179− 184. (27) Sacco, R. L., et al. (1993) Wallenberg’s lateral medullary syndrome. Clinical-magnetic resonance imaging correlations. Arch. Neurol. 50, 609−614. (28) Pomeranz, H. (2002) Isolated Horner syndrome and syrinx of the cervical spinal cord. Am. J. Ophthalmol. 133, 702−704. (29) Thompson, H., Maxner, C., and Corbett, J. (1991) Horner syndrome due to damage to the preganglionic neuron of the oculosympathetic pathway. Eye 5, 36. (30) Herbut, P., and Watson, J. (1946) Tumor of the thoracic inlet producing the Pancoast syndrome: Report of 17 cases and a review of the literature. Arch. Path. 42, 88−103. (31) Baumgartner, R. W., and Bogousslavsky, J. (2005) Clinical manifestations of carotid dissection. Front. Neurol. Neurosci. 20, 70−76. (32) Beebe, J. D., Kardon, R. H., and Thurtell, M. J. (2017) The Yield of Diagnostic Imaging in Patients with Isolated Horner Syndrome. Neurol. Clin. 35, 145−151. (33) Khan, A. M., Ahmad, N., Li, X., Korsten, M. A., and Rosman, A. (2005) Chiropractic sympathectomy: carotid artery dissection with oculosympathetic palsy after chiropractic manipulation of the neck. Mt. Sinai J. Med. N. Y. 72, 207−210. (34) Parwar, B. L., Fawzi, A. A., Arnold, A. C., and Schwartz, S. D. (2001) Horner’s syndrome and dissection of the internal carotid artery after chiropractic manipulation of the neck. Am. J. Ophthalmol. 131, 523−524.

typically create symptoms, though there may be some exceptions; for example, patients with a central cataract might experience greater glare and a decrease in vision when the resting pupil is smaller. By definition, the ptosis in Horner syndrome is only 1−2 mm, and thus it is unlikely to affect vision. Therefore, patients with Horner syndrome are usually asymptomatic and mainly concerned about their appearance rather than complaining of loss of function. It is readily evident with diagnostic testing that apraclonidine not only acts on the pupil in an eye with Horner syndrome but likewise can elevate the eyelid in an affected eye as the postsynaptic receptors and Mueller’s muscle are also upregulated and become supersensitive to apraclonidine. Apraclonidine eye drops therefore may have utility as a therapeutic agent for temporary cosmetic reversal of the ptosis in Horner syndrome.74 Over-thecounter eye drops containing naphazoline (sympathomimetic agent used as a vasoconstrictor for red eyes) can have a similar effect.75,76 Surgical intervention with a levator resection or other method of ptosis repair has a high success rate because the ptosis is reliably small and stable.



CONCLUSIONS Although the ptosis and miosis of Horner syndrome may not result in significant symptoms, recognizing Horner syndrome is critical, as these findings point to a lesion in the oculosympathetic pathway. Though often benign or idiopathic, the cause of Horner syndrome can be very threatening or even lethal, so understanding how to recognize, diagnose, and appropriately evaluate Horner syndrome is important to all clinicians.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Timothy J. Martin: 0000-0001-7194-5560 Notes

The author declares no competing financial interest.



REFERENCES

(1) Martin, T. J., and Corbett, J. J. (2013) The Pupil. In Practical Neuroophthalmology, pp 261−286, McGraw-Hill Education/Medical. (2) Abbas, A. et al. Johann Friedrich Horner and the Repeated Discovery of Oculosympathoparesis: Whose Syndrome Is It? Neurosurgery 2015, 77, 486−491; discussion 491. (3) van der Wiel, H. L. (2002) Johann Friedrich Horner (1831− 1886). J. Neurol. 249, 636−637. (4) Thompson, H. S. (1986) Johann Friedrich Horner (1831−1886). Am. J. Ophthalmol. 102, 792−795. (5) Martin, T. J. (2007) Horner’s syndrome, Pseudo-Horner’s syndrome, and simple anisocoria. Curr. Neurol. Neurosci. Rep. 7, 397− 406. (6) Wilhelm, H., et al. (1992) Horner’s syndrome: a retrospective analysis of 90 cases and recommendations for clinical handling. Ger. J. Ophthalmol. 1, 96−102. (7) Kanagalingam, S., and Miller, N. R. (2015) Horner syndrome: clinical perspectives. Eye Brain 7, 35−46. (8) Czarnecki, J. S., Pilley, S. F., and Thompson, H. S. (1979) The analysis of anisocoria. The use of photography in the clinical evaluation of unequal pupils. Can. J. Ophthalmol. J. Can. Ophtalmol. 14, 297−302. (9) Yoo, Y. J., Yang, H. K., and Hwang, J.-M. (2017) Efficacy of digital pupillometry for diagnosis of Horner syndrome. PLoS One 12, e0178361. (10) Pilley, S. F., and Thompson, H. S. (1975) Pupillary ‘dilatation lag’ in Horner’s syndrome. Br. J. Ophthalmol. 59, 731−735. I

DOI: 10.1021/acschemneuro.7b00405 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Review

ACS Chemical Neuroscience

(56) Liu, G. T. (2011) Pediatric horner syndrome. Arch. Ophthalmol. 1960 (129), 1108−1109. (57) Thompson, H. S. (1977) Diagnosing Horner’s syndrome. Trans. Sect. Ophthalmol. Am. Acad. Ophthalmol. Otolaryngol. 83, 840−842. (58) Van der Wiel, H. L., and Van Gijn, J. (1986) The diagnosis of Horner’s syndrome. Use and limitations of the cocaine test. J. Neurol. Sci. 73, 311−316. (59) Kardon, R. H., Denison, C. E., Brown, C. K., and Thompson, H. S. (1990) Critical evaluation of the cocaine test in the diagnosis of Horner’s syndrome. Arch. Ophthalmol. 1960 (108), 384−387. (60) Jacobson, D. M., Berg, R., Grinstead, G. F., and Kruse, J. R. (2001) Duration of positive urine for cocaine metabolite after ophthalmic administration: implications for testing patients with suspected Horner syndrome using ophthalmic cocaine. Am. J. Ophthalmol. 131, 742−747. (61) Cremer, S. A., Thompson, H. S., Digre, K. B., and Kardon, R. H. (1990) Hydroxyamphetamine mydriasis in normal subjects. Am. J. Ophthalmol. 110, 66−70. (62) Thompson, H. S., and Mensher, J. H. (1971) Adrenergic mydriasis in Horner’s syndrome. Hydroxyamphetamine test for diagnosis of postganglionic defects. Am. J. Ophthalmol. 72, 472−480. (63) Cremer, S. A., Thompson, H. S., Digre, K. B., and Kardon, R. H. (1990) Hydroxyamphetamine mydriasis in Horner’s syndrome. Am. J. Ophthalmol. 110, 71−76. (64) Donahue, S. P., Lavin, P. J., and Digre, K. (1996) False-negative hydroxyamphetamine (Paredrine) test in acute Horner’s syndrome. Am. J. Ophthalmol. 122, 900−901. (65) Weinstein, J. M., Zweifel, T. J., and Thompson, H. S. (1980) Congenital Horner’s syndrome. Arch. Ophthalmol. 1960 (98), 1074− 1078. (66) Gross, J. R., McClelland, C. M., and Lee, M. S. (2016) An approach to anisocoria. Curr. Opin. Ophthalmol. 27, 486−492. (67) Danesh-Meyer, H. V., Savino, P., and Sergott, R. (2004) The correlation of phenylephrine 1% with hydroxyamphetamine 1% in Horner’s syndrome. Br. J. Ophthalmol. 88, 592−593. (68) Smit, D. P. (2010) Pharmacologic testing in Horner’s syndrome - a new paradigm. South Afr. Med. J. Suid-Afr. Tydskr. Vir Geneeskd. 100, 738−740. (69) Al-Moosa, A., and Eggenberger, E. (2011) Neuroimaging yield in isolated Horner syndrome. Curr. Opin. Ophthalmol. 22, 468−471. (70) Chen, Y., Morgan, M. L., Barros Palau, A. E., Yalamanchili, S., and Lee, A. G. (2015) Evaluation and neuroimaging of the Horner syndrome. Can. J. Ophthalmol. 50, 107−111. (71) George, A., Haydar, A. A., and Adams, W. M. (2008) Imaging of Horner’s syndrome. Clin. Radiol. 63, 499−505. (72) Lee, J. H., et al. (2007) Neuroimaging strategies for three types of Horner syndrome with emphasis on anatomic location. AJR, Am. J. Roentgenol. 188, W74−81. (73) Davagnanam, I., Fraser, C. L., Miszkiel, K., Daniel, C. S., and Plant, G. T. (2013) Adult Horner’s syndrome: a combined clinical, pharmacological, and imaging algorithm. Eye 27, 291−298. (74) Garibaldi, D. C., Hindman, H. B., Grant, M. P., Iliff, N. T., and Merbs, S. L. (2006) Effect of 0.5% apraclonidine on ptosis in Horner syndrome. Ophthal. Plast. Reconstr. Surg. 22, 53−55. (75) Pemberton, J. D., MacIntosh, P. W., Zeglam, A., and Fay, A. (2015) Naphazoline as a confounder in the diagnosis of carotid artery dissection. Ophthal. Plast. Reconstr. Surg. 31, e33−35. (76) Tomelleri, G., Vattemi, G., Filosto, M., and Tonin, P. (2007) Eyelid ptosis from sympathetic nerve dysfunction mistaken as myopathy: a simple test to identify this condition. J. Neurol., Neurosurg. Psychiatry 78, 632−634.

(35) Patel, R. R., et al. (2012) Cervical carotid artery dissection: current review of diagnosis and treatment. Cardiol. Rev. 20, 145−152. (36) Yang, S.-T., Huang, Y.-C., Chuang, C.-C., and Hsu, P.-W. (2006) Traumatic internal carotid artery dissection. J. Clin. Neurosci. 13, 123−128. (37) Riley, F. C. J., and Moyer, N. J. (1971) Oculosympathetic paresis associated with cluster headaches. Am. J. Ophthalmol. 72, 763− 768. (38) Jimenez-Caballero, P. E., Marsal-Alonso, C., and AlvarezTejerina, A. (2005) [Horner syndrome as the first symptom of nasopharyngeal cancer. Two case reports]. Rev. Neurol. 40, 541−543. (39) Kurihara, T. (2006) Abducens nerve palsy and ipsilateral incomplete Horner syndrome: a significant sign of locating the lesion in the posterior cavernous sinus. Intern. Med. 45, 993−994. (40) Tsuda, H., et al. (2009) Combination of abducens nerve palsy and ipsilateral postganglionic Horner syndrome as an initial manifestation of uterine cervical cancer. Intern. Med. 48, 1457−1460. (41) George, N. D., Gonzalez, G., and Hoyt, C. S. (1998) Does Horner’s syndrome in infancy require investigation? Br. J. Ophthalmol. 82, 51−54. (42) Jeffery, A. R., Ellis, F. J., Repka, M. X., and Buncic, J. R. (1998) Pediatric Horner syndrome. J. AAPOS Off. Publ. Am. Assoc. Pediatr. Ophthalmol. Strabismus 2, 159−167. (43) Zafeiriou, D. I., et al. (2006) Congenital Horner’s syndrome associated with cervical neuroblastoma. Eur. J. Paediatr. Neurol. EJPN Off. J. Eur. Paediatr. Neurol. Soc. 10, 90−92. (44) Nitschke, R., et al. (1991) Postoperative treatment of nonmetastatic visible residual neuroblastoma: a Pediatric Oncology Group study. J. Clin. Oncol. 9, 1181−1188. (45) Mahoney, N. R., et al. (2006) Pediatric horner syndrome: etiologies and roles of imaging and urine studies to detect neuroblastoma and other responsible mass lesions. Am. J. Ophthalmol. 142, 651−659. (46) Lam, B. L., Thompson, H. S., and Corbett, J. J. (1987) The prevalence of simple anisocoria. Am. J. Ophthalmol. 104, 69−73. (47) Morales, J., Brown, S. M., Abdul-Rahim, A. S., and Crosson, C. E. (2000) Ocular effects of apraclonidine in Horner syndrome. Arch. Ophthalmol. 1960 (118), 951−954. (48) Brown, S. M. The utility of 0.5% apraclonidine in the diagnosis of horner syndrome. Arch. Ophthalmol., 2005, 1960, 123, 578; author reply 578. (49) Freedman, K. A., and Brown, S. M. (2005) Topical apraclonidine in the diagnosis of suspected Horner syndrome. J. Neuro-Ophthalmol. Off. J. North Am. Neuro-Ophthalmol. Soc. 25, 83− 85. (50) Koc, F., Kavuncu, S., Kansu, T., Acaroglu, G., and Firat, E. (2005) The sensitivity and specificity of 0.5% apraclonidine in the diagnosis of oculosympathetic paresis. Br. J. Ophthalmol. 89, 1442− 1444. (51) Chen, P.-L., Chen, J.-T., Lu, D.-W., Chen, Y.-C., and Hsiao, C.H. (2006) Comparing efficacies of 0.5% apraclonidine with 4% cocaine in the diagnosis of horner syndrome in pediatric patients. J. Ocul. Pharmacol. Ther. 22, 182−187. (52) Kardon, R. (2005) Are we ready to replace cocaine with apraclonidine in the pharmacologic diagnosis of Horner syndrome? J. Neuro-Ophthalmol. Off. J. North Am. Neuro-Ophthalmol. Soc. 25, 69− 70. (53) Lebas, M., Seror, J., and Debroucker, T. (2010) Positive apraclonidine test 36 h after acute onset of horner syndrome in dorsolateral pontomedullary stroke. J. Neuro-Ophthalmol. Off. J. North Am. Neuro-Ophthalmol. Soc. 30, 12−17. (54) Watts, P., Satterfield, D., and Lim, M. K. (2007) Adverse effects of apraclonidine used in the diagnosis of Horner syndrome in infants. J. AAPOS Off. Publ. Am. Assoc. Pediatr. Ophthalmol. Strabismus 11, 282−283. (55) Enyedi, L. B., and Freedman, S. F. (2001) Safety and efficacy of brimonidine in children with glaucoma. J. AAPOS Off. Publ. Am. Assoc. Pediatr. Ophthalmol. Strabismus 5, 281−284. J

DOI: 10.1021/acschemneuro.7b00405 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX