The "Harlequin" Signand Congenital Horner's Syndrome

62666ournal ofNeurology, Neurosurgery, and Psychiatry 1997;62:626-628 SHORT REPORT The "harlequin" sign and congenital Horner's syndrome D A Morrison, K Bibby, G Woodruff Abstract ganglion) may be made by observing the dis- When trying to establish the likely tribution of anhidrosis and facial flushing.56 anatomical site (preganglionic or post- The site of the lesion may be determined phar- ganglionic) of a lesion causing congenital macologically by testing the response of the Horner's syndrome, the distribution of pupil to pholedrine drops.7 Pholedrine is facial flushing (the "harlequin" sign), chemically similar to hydroxyamphetamine, may be seen. In babies and young chil- and works by releasing catecholamines from dren, facial flushing is a relatively simple intact nerve endings. clinical sign to demonstrate, compared It has been suggested that pharmacological with facial sweating. In unilateral facial methods of localising sympathetic lesions may flushing the areas that do not flush are be less reliable in congenital Homer's syn- almost always identical to the anhidrotic drome than acquired Homer's syndrome.5 areas. However, neither facial flushing Such atypical responses are not widely nor testing the pupil reactions with acknowledged and may have implications in pholedrine or hydroxyamphetamine can the management of congenital Homer's syn- be relied on to predict the probable site of drome. any lesion causing congenital Horner's We describe two patients with congenital syndrome. Two patients with congenital Homer's syndrome in whom clinical examina- Horner's syndrome are presented which tion and pharmacological testing gave conflict- demonstrated the "harlequin" sign and in ing evidence for localisation of the causative whom clinical examination and pharma- lesion. cological testing gave conflicting evidence for localisation of the site of the causative lesion. The presentation of congenital Case reports Horner's syndrome should be investi- CASE 1 gated and include MRI or CT to exclude a A 4 month old boy was referred to the eye serious underlying cause. clinic after his mother noticed that the left eye "looked smaller" than the right. He had been delivered quickly with the cord around his (7 Neurol Neurosurg Psychiatry 1997;62:626-628) neck but his birth was otherwise normal. Ocular examination showed anisocoria: the Keywords: congenital Homer's syndrome; pholedrine right pupil measured 4 mm in diameter and the left 2 mm. There was narrowing of the left palpebral fissure and left ptosis (fig 1A), but The clinical features of congenital Homer's no iris hypochromia. On attempting to instil syndrome include upper and lower lid ptosis, diagnostic eye drops the child became dis- Department of Ophthalmology, pupil miosis, iris hypochromia, and anhidrosis tressed and flushed all over with the exception University of of the face and upper limbs on the affected of the left side of the face (fig 1B) (facial Leicester, Robert side. sweating was not measured). The absence Kilpatrick Clinical cause of Sciences Building, Classically, the congenital Homer's of flushing suggested a left preganglionic Leicester Royal syndrome has been birth trauma resulting in Homer's syndrome. The diagnosis of left Infirmary, LE2 7LX, brachial plexus injury (Klumpke's palsy). Homer's syndrome was confirmed with 4% UK More sinister such as neuroblas- D A Morrison aetiologies, cocaine drops. K Bibby toma and carotid artery thrombosis, are occa- General examination was normal with no G Woodruff sionally seen. 1-3 However, in many cases no limb weakness. Pholedrine drops were instilled Correspondence to: cause is found, despite extensive investiga- into both eyes and after 40 minutes the right Mr D A Morrison, South East of Scotland Clinical tion.3 pupil measured 7 mm and the left 4 mm, this Genetics Service, Western The diagnosis of Homer's syndrome is con- increase in anisocoria being typical in a post- General Hospital, Crewe Road, Edinburgh EH4 2XU, firmed pharmacologically by the use of 4% ganglionic Homer's syndrome. Magnetic reso- UK. cocaine drops, which cause an increase in nance imaging of the head, neck and thorax Received 13 June 1996 anisocoria (difference in pupil size).4 was normal. One year later this boy was well, and in revised form 2 January 1997 Clinically, localisation of the probable site of but still showed asymmetric facial flushing Accepted 23 January 1997 the lesion (presuperior or postsuperior cervical when distressed or in hot weather. The "harlequin" sign and congenital Homer's syndrome 627 Figure 1 (A) Case 1: left congenital Homer's syndrome. (B) Case 1: left i1 .-i congenital Homer's syndrome showing asymmetric facialflushing. 4-.- ::,J.li -s- A 4. W. A CASE 2 A 7 year old boy presented to the eye clinic with the right pupil smaller than the left since birth. His mother stated that after exerc:ise, in hot weather, or when upset, the left side of his face and upper chest would go red whereas the right side of his face did not. Examirnation showed the right pupil to be 3 mm in diaLmeter and the left 5 mm, and there was 1-5 nnm of right ptosis. There was no iris hypochrromia. Visual acuity was 6/6 right and left. The right B pupil did not change in size after the addition of 4% cocaine drops whereas the left dilalted to 8 mm, confirming the diagnosis of a right of the brain, neck, and thorax did not show Homer's syndrome. After running indoors any paravertebal mass or abnormal signal. and climbing stairs for a few minutes, there was a pronounced asymmetric flushing of the Discussion face with the left side red and sweaty com- Congenital Homer's syndrome may be recog- pared with the right, implying a pregang lesion on the right side (fig 2). Some 'weoni nised clinically by the features of blepharopto- later, pupil testing with pholedrine drops sis, miosis, facial anhidrosis, and iris caused the left pupil to increase in size firom 5 hypochromia. The diagnosis can be confirmed mm to 6 mm. The right pupil remainecd at 3 with cocaine drops.4 mm, suggesting a postganglionic lesion. Hydroxyamphetamine drops (1%) are General examination was normal and MRI highly effective in distinguishing preganglionic and postganglionic lesions.8 With a postganglionic lesion the pupil fails to dilate; with a Figure 2 Case 2: right preganglionic lesion the affected pupil dilates congenital Homer's the same as or more than the normal side. syndrome with asymmetric facialflushing (after 4% Hydroxyamphetamine drops are no longer cocaine drops). available in the United Kingdom, but .1I pholedrine (Ankerpharm GmBH, Ankerwerk ,..ii,;ii,K Rudolstadt, Germany) is structurally similar to hydroxyamphetamine and has been shown to .i be comparable in its effects.7 The distribution of sweating in Homer's syndrome is valuable in localising the site of the lesion clinically.6 The anatomical pathways concerned with thermoregulatory facial sweat- c . I ing are well established.9 When the lesion is .4.. proximal to the bifurcation of the common ;.1Y carotid artery there is hemifacial anhidrosis on a. the affected side, and when the lesion is distal the loss of sweating is confined to the medial aspect of the forehead. In patients not showing anhidrosis or with only mild asymmetry there is no localisation value, and many of these will have central lesions. In unilateral facial flushing the areas that do not flush are virtually identical to the anhidrotic areas.5 910 Both thermoregulatory facial flushing and emotional vasodilatation are thought to be mediated by the cervical sympathetic pathways,9 travelling for the most part with branches of the external carotid artery. 628 Momson, Bibby, Woodruff Thus, in congenital Homer's syndrome, the site of a lesion causing congenital Homer's affected side may show reduced facial flushing syndrome. The use of 4% cocaine drops to and a decrease in skin temperature due to confirm a clinical diagnosis would seem to be impaired sympathetic vasodilatation. The justified. However, the value of pharmacologi- effect is exaggerated by superimposed active cal testing of the pupils with pholedrine or vasoconstriction due to circulating cate- hydroxyamphetamine in congenital Homer's cholamines (enhanced by denervation hyper- syndrome must be questioned, particularly as sensitivity5). this may not be easy or practical in small chil- In the two patients presented we have not dren. The results have no influence on subse- been able to determine the exact anatomical quent investigation or management. site of the lesion. The distribution of facial The interesting but largely unacknowledged flushing, suggesting preganglionic congenital "harlequin" sign in association with congenital Homer's syndrome, is at odds with the results Homer's syndrome has been illustrated; asym- of pharmacological testing. The response of metric facial flushing should alert the clinician the pupils to pholedrine, implying a postgan- to the possibility of Homer's syndrome. In a glionic lesion, and the simultaneous presence possible case of congenital Homer's syn- of a clinically preganglionic lesion in congenital drome, the "harlequin" sign should be sought. Homer's syndrome, could be explained in var- It is easier to show than facial anhydrosis, par- ious ways. ticularly in small babies, although its value as a (1) There may be a diffuse lesion or more localising sign has yet to be defined. than one lesion involving the preganglionic Therefore, to what extent and when should and postganglionic neuron, or a lesion affect- the presentation of congenital Homer's

Load more