Clinical and Experimental Ophthalmology (2001) 29, 411415

Original Article
Carotid dissection: a common cause of Horners syndrome
Colin CK Chan MBBS(Hons), Mark Paine FRACP and Justin ODay FRACO
Department of Neuro-Ophthalmology, The Royal Victorian Eye and Ear Hospital, East Melbourne, Victoria, Australia

ABSTRACT
Purpose: To highlight that internal carotid artery dissection
is a common cause of Horners syndrome and that it is
important to diagnose dissection as anticoagulation can
prevent carotid thrombosis and embolism.
Methods: Five case reports are presented of patients with
Horners syndrome secondary to carotid dissection. One
patient had carotid dissection secondary to trauma, two
had spontaneous dissections and two had dissections in the
settings of other illness. A literature search was performed
on carotid dissection as a cause of Horners syndrome and
its diagnosis and management.
Results: The case reports and literature highlight that
dissection is under-recognized as a cause of Horners
syndrome and can be missed.The investigation of choice is
a magnetic resonance imaging and angiography scan of the
head and neck.The treatment advocated is anticoagulation
for 36 months.
Conclusion: Carotid dissection should be suspected in
patients with Horners syndrome, particularly if head or
neck pain is present.
Key words: anticoagulation, carotid dissection, headache,
Horners syndrome, magnetic resonance angiography,
magnetic resonance imaging.

INTRODUCTION
It is traditionally thought that internal carotid artery (ICA)
dissection is a serious but uncommon cause of Horners syndrome, and that dissection is usually flagged by a clear
history of neck trauma with resultant head and neck pain. It
is our experience that many dissections do not present this

way and a dissection may be missed with potentially serious

consequences. Some doctors may be reluctant to use tests
such as digital subtraction angiography and magnetic resonance imaging in patients without a classical presentation.
With our series of patients, we hope to illustrate the varied
ways in which carotid dissection may present, suggest that
is may be much more common than previously thought, and
discuss appropriate investigations and management.

CASE REPORTS
Case 1: Dissection secondary to trauma
A 47-year-old woman with hypertension presented with
pain and paraesthesiae in her right temporal and periorbital
region a few days after a fall down a flight of stairs. She fractured her left navicular bone but did not sustain any other
injuries. In particular, she did not recall having hit her head
or twisting her neck. The headache and numbness were
continuous for 12 h but then became intermittent. She then
noticed a right ptosis and miosis. She did not have any visual
disturbances but did complain of ringing in her right ear.
Investigations included a full blood count, ESR, thyroid
function tests, syphilis serology and a computed tomography (CT) brain scan, all of which were normal.
On examination, there was a moderate right ptosis with
a miotic right pupil, consistent with a right Horners syndrome (Fig. 1). Visual acuity was 6/6 in both eyes and the
visual fields were full. The remainder of the cranial nerve
examination was normal. Neurological examination revealed
absent brachioradialis and biceps jerks on the right;
however, there were no other neurological signs.
An urgent magnetic resonance imaging/magnetic resonance angiography (MRI/MRA) head and neck confirmed a
dissection beginning just above the right internal carotid
artery origin and extending up into the skull base (Figs 2,3a).
She was admitted, heparinized and commenced on warfarin.
A repeat MRI/MRA performed 6 months later showed

Correspondence: Dr Colin Chan, Department of Neuro-Ophthalmology, The Royal Victorian Eye and Ear Hospital, 32 Gisborne Street, East Melbourne, Vic
3002, Australia. Email: chunkeng@yahoo.com

412

Chan et al.
recanalization of the right internal carotid artery. The warfarin was then ceased.

Case 2: Spontaneous dissection

Figure 1.

Partial ptosis and miosis of Horners syndrome (case 1).

A 52-year-old man with a past history of bilateral retinal

tears, benign lid myokymia and migraine presented with a
2-day history of sudden onset, severe throbbing jaw and
facial pain. The pain was different to the headache that had
been ascribed to migraine. There was no history of recent
trauma or illness. On examination, he had a left Horners
syndrome. No other neurological abnormalities were
present. The initial MRI/MRA report identified a thickened
area of the left internal carotid artery at the skull base that
was initially thought not to be a dissection, but an area of
atherosclerosis or focal inflammation. On further review of
the films, the final opinion was that of a dissection. The
patient was anticoagulated with repeat MRI scans showing
recanalization. The Horners syndrome resolved partially as
did the facial pain. After 3 months, warfarin was ceased and
aspirin commenced.

Case 3: Spontaneous dissection

Figure 2. Classic string sign indicating carotid dissection on

reconstructed magnetic resonance angiography images (case1).

A 54-year-old woman with known hypertension presented

to her general practitioner following the sudden onset of left
jaw pain radiating to her forehead and orbit, associated with
jaw and scalp tenderness. She also complained of mild
photophobia. There was no history of trauma. Her general
practitioner suspected giant cell arteritis and commenced
her on high-dose oral prednisolone. Over the next 2 weeks,
the pain persisted and she developed an intermittent left
ptosis and red eye. She also described several brief episodes

Figure 3. (a) Crescent shaped hyperintense lesion representing internal carotid artery dissection on fat suppressed magnetic resonance
(MR) image (case 1). (b) Non-fat suppressed MR image showing similar hyperintensity of dissection/haematoma and high velocity carotid
flow (reproduced with permission from Dr N Trost, Consultant Radiologist, St. Vincents Hospital, Melbourne, Australia).

Carotid dissection and Horners syndrome

of a grey curtain coming down over her vision and the
appearance of flashing lights and bright, wiry lines. On
further questioning, she admitted to having a similar rightsided headache 7 years earlier, associated with a left lower
homonymous field defect.
Examination revealed 6/6 vision in her right eye and
6/5 vision in her left. A slight anisocoria was noted, with the
left pupil smaller than the right but there was no relative
afferent pupillary defect. Her left temporomandibular joint
(TMJ) was moderately tender. No other cranial nerve
defects were found. A brain CT scan was normal. Ocular
migraine with TMJ dysfunction was thought to be the most
likely reason for her symptoms.
She was reviewed 2 weeks later. The headache and facial
pain had improved but she now had a clear left sided
Horners syndrome with a 2-mm ptosis and 1-mm anisocoria.
An urgent MRI/MRA of the head and neck was performed
which revealed a left ICA dissection extending intracranially
with associated carotid occlusion. There were also changes
consistent with deep white matter ischaemia. She was
commenced on warfarin. After 6 months this was changed
to aspirin. She had persistent facial pain 1 year later despite
partial resolution of the Horners syndrome. The pain eventually responded to amitryptylline after failed trials of
non-steroidal anti-inflammatory drugs and carbamazepine.

Case 4: Dissection in the presence of

intercurrent illness
A 49-year-old previously well man initially presented with
recurrent episodes of right-sided maxillary pain and congestion. He also was bothered by bouts of coughing which he
attributed to nasopharyngeal discharge. He was diagnosed
with sinusitis, but because of its recurrent nature, he was
referred to an ear, nose and throat surgeon. At the consultation, he mentioned to the surgeon that he had noticed that
his right eye was red and the pupil was smaller. A CT scan
of the sinuses revealed a large inclusion cyst suspended from
the roof of the maxillary sinus.
However, this did not explain the eye symptoms and he
was referred to an ophthalmologist. On examination, there
was a right 1.5-mm miosis in the light and a 1-mm miosis in
the dark. There was also a slight right ptosis. Visual acuity
was 6/4 in both eyes. Horners syndrome was diagnosed and
an urgent MRI/MRA of the head and neck organized. The
initial MRI failed to show any abnormality, but a repeat scan
revealed a right carotid dissection at the second cervical
vertebra extending to the skull base. He was commenced on
warfarin.

Case 5: Dissection in the presence of

intercurrent illness
A previously well 52-year-old man presented with a 6-week
history of increasing right fronto-orbital and maxillary pain
on a background of recent otitis media. The pain kept him
awake at night despite taking Panadeine forte (SmithKline

413
Beecham, Boronia, Victoria, Australia). He also complained
of altered sensation over the right supraorbital region,
extending back to the scalp and a recent onset right ptosis.
However, his vision was normal.
On examination, he had a right 23 mm ptosis and anisocoria of 12 mm. His visual acuity, fields and fundi were
normal. The rest of the cranial examination was also normal;
notably he did not have any evidence of trigeminal nerve
dysfunction. Both carotid pulses were present and no bruits
were heard. A MRI/MRA of the head and neck showed evidence of a right internal carotid dissection just below the
skull base, with displacement of the carotid artery and a ring
of hyperintense tissue representing haematoma surrounding
the artery posteriorly. He was commenced on warfarin. A
repeat MRI/MRA scan 3 months later showed recanalization
of the artery lumen and the anticoagulation was ceased.
Despite this, the Horners syndrome and facial pain has
persisted. Numerous membrane stabilizers and epileptic
drugs were tried with limited success.

DISCUSSION
Ophthalmologists are often the first specialists to see
patients with Horners syndrome. The aetiology of Horners
syndrome has traditionally been classified anatomically into
central, preganglionic and postganglionic lesions.1,2 Usually
postganglionic lesions can be differentiated from central and
preganglionic defects clinically. Current teaching is that
postganglionic Horners syndrome usually has a benign aetiology.35 However, much of this presumption comes from a
few prospective and retrospective studies performed before
the advent of magnetic resonance (MR). Thus these patients
did not have any screening tests for dissection unless they
had symptoms such as amaurosis fugax or neck pain. It has
been our experience that spontaneous carotid dissection is
more commonly a cause of Horners syndrome than previously recognized. This is also suggested by Miller and
Newman.3 Leira et al. further suggested that carotid dissection should be suspected in cases of transient anisocoria
also.6 There is only one prospective trial, to our knowledge,
that looks at how often Horners syndrome is caused by dissection. Digre et al. performed MRI/MRA of the head and
neck in 20 patients with postganglionic Horners syndrome.7
Three patients (15%) were found to have carotid dissection.
Interestingly, the study concluded that this was too low a
yield to make MRI/MRA useful in the management of postganglionic Horners syndrome.
As illustrated with our case reports, there are three basic
groups of patients that present with carotid dissection: (i)
patients with a clear history of preceding trauma, sometimes
as trivial as a coughing spasm (case 1); (ii) patients with no
history of trauma (cases 2 and 3) and otherwise well, the
spontaneous carotid dissection; and (iii) patients with an
intercurrent illness (cases 4 and 5).8,9 It is thus our experience that dissection may not present with the classic history
of preceding trauma, abrupt onset of neck and head pain
and Horners syndrome. However, having said this, the

414
Table 1. Symptoms and signs associated with carotid dissection
in order of decreasing frequency11,32,33
Cephalic pain/headache
Cerebral ischaemic symptoms (hemiparesis/paraesthesiae/
hemianopia)
Horners syndrome
Monocular visual loss
Photopsia (acute choroidal hypoperfusion)
Amaurosis fugax
Bruit
Dysgeusia (loss of taste)

combination of headache with Horners syndrome should

immediately alert the ophthalmologist to the possibility of a
dissection. The characteristic pain of carotid dissection
starts in the neck and radiates to the face.4,10,11 However, for
some patients, as illustrated in our case studies, the pain only
manifests in the face. Multiple studies show that headache is
a very common feature of dissection (92%). Horners syndrome is also common (up to 82%).5,6,1113 The differential
diagnoses for a painful Horners syndrome also include
cluster headaches, tumours and Raeders paratrigeminal syndrome.3,14 Raeders syndrome is a distinct entity as discussed
below. Cluster headaches with Horners syndrome can
usually be differentiated from dissection by a prior history
and also their episodic nature, although over time the
Horners syndrome may become permanent. Patients with a
cavernous sinus mass usually have features consistent with
compression of other structures within the sinus such as
third, fourth or sixth nerve palsies. The classical features of
dissection in decreasing order of frequency are listed in
Table 1.
It is worth a reminder that, strictly speaking, a painful
postganglionic Horners syndrome is not synomonous
with Raeders syndrome. When Raeder defined his syndrome in 1924, he was specifically referring to a lesion at the
point where the oculosympathetic pathways leave the
internal carotid artery to join the ophthalmic division of
the trigeminal nerve. This lesion typically resulted in multiple parasellar cranial nerve involvement and was associated
with a benign outcome.15,16 This contrasts with carotid
dissection, which can lead to serious complications such
as retinal and cerebral ischaemia.4,17 The headache in
carotid dissection is thought to be due to the dissection
itself and associated haematoma formation with stretching
of sensory nerves, not trigeminal nerve involvement. Kline
et al. suggest that many cases originally diagnosed with
Raeders may in fact been cases of dissection.2 It was
interesting that a number of our patients had a persistent
facial neuralgia. This is not commonly reported in the
literature and its aetiology and management are uncertain.
We found that the pain was usually refractory to opiatebased analgesics but responded to membrane stabilisers such
as amitriptylline and carbamazepine.
Various underlying pathologies have been suggested to
predispose to carotid dissection including fibromuscular

Chan et al.
dysplasia, Marfans syndrome, cystic medial necrosis and
hypertension.1820 However, in the majority of cases no
predisposing factors are found. The mechanism of oculosympathetic palsy is thought to be enlargement of the
artery, associated with haematoma formation and localized
inflammatory response. This results in compression of the
superior cervical ganglion lying in the posterior wall of the
carotid sheath.9,19 The defect is thus postganglionic and
usually incomplete.6
It is now widely accepted that a combination of MRI and
MRA of the head and neck is the best method to diagnose
carotid dissection.9,21 Intra-arterial digital subtraction
angiography (DSA), previously considered the gold standard, is no longer used where MR is available. It is invasive,
more expensive and is less accurate than MR. Diagnosis of
dissection with DSA is by the presence of the classic string
sign which is only present in subintimal dissections where
haemorrhage into the subintimal space causes narrowing of
the vessel (Fig. 2). Some patients have subadventitial dissections that are less likely to produce stenosis and thus may be
missed on DSA.22,23 For the same reason, dissection can be
missed if the radiologist looks only at the reconstructed
images of the MRA as the reconstructed images provide
similar images to DSA. Instead it is the source images that
must be looked at as these will show high-intensity crescent
shaped haematoma in the wall of the artery (Fig. 3a).
Requests for a MRI/MRA should specifically ask for fatsuppressed T1 images. Without fat suppression, both
arterial flow and haematoma will look equally bright and a
dissection can be missed (Fig. 3b). The request should also
specify that both the head and neck images cover the entire
length of the internal carotid artery from bifurcation to skull
base (N Trost, 2001, pers. comm.).12,2430
Carotid Doppler has been proposed by a few studies as a
first line diagnostic test.8,17,31 The premise is that ultrasonography (US) may be more sensitive for detection of
early thrombus than MR. The main problem with US is that
it is unable to reliably detect dissection high in the neck,
near the skull base, where most dissections seem to occur.18
However, it may be a useful tool in monitoring recanalization of the artery and guiding anticoagulation therapy after
initial MR scans to diagnose the lesion.
Treatment of dissection should be with an antiplatelet
drug such as aspirin or anticoagulation with heparin or warfarin, but this is controversial and there is limited supportive
data.8,9,11,17 Admittedly, it would be difficult to justify a
prospective trial looking at this issue but more retrospective
data may clarify this.
In summary, carotid dissection is probably underrecognized as a cause for Horners syndrome. It should
probably be excluded by MRI/MRA in all patients with
painful postganglionic Horners syndrome and maybe
also patients without headache and neck pain because of
the potential complications. The treatment advocated is
36 months of anticoagulation. More studies are necessary
to clarify the management of Horners syndrome and
dissection.

Carotid dissection and Horners syndrome

REFERENCES
1. Monteiro ML, Coppeto JR. Horners syndrome associated
with carotid artery atherosclerosis. Am. J. Ophthalmol. 1988;
105: 934.
2. Kline LB, Vitek JJ, Raymon BC. Painful Horners syndrome due
to spontaneous carotid artery dissection. Ophthalmology 1987;
94: 22630.
3. Miller NR, Newman NJ (eds). Walsh and Hoyts Clinical Neuroophthalmology, 5th edn. Baltimore: Williams & Wilkins, 1998;
96971.
4. Biousse V. Carotid disease and the eye. Curr. Opin. Ophthalmol.
1997; 8: 1626.
5. Ast G, Woimant F, Georges B, Laurian C, Haguenau M.
Spontaneous dissection of the internal carotid artery in 68
patients. Eur. J. Med. 1993; 2: 46672.
6. Leira EC, Bendixen BH, Kardon RH, Adams HP. Brief, transient Horners syndrome can be the hallmark of a carotid
artery dissection. Neurology 1998; 50: 28990.
7. Digre KB, Smoker WR, Johnston P et al. Selective MR imaging
approach for evaluation of patients with Horners syndrome.
AJNR Am. J. Neuroradiol. 1992; 13: 2237.
8. Zetterling M, Carlstrom C, Konrad P. Internal carotid artery
dissection. Acta Neurol. Scand. 2000; 101: 17.
9. Rosebraugh CJ, Griebel DJ, DiPette DJ. A case report of
carotid artery dissection presenting as cluster headache. Am. J.
Med. 1997; 102: 41819.
10. Stapf C, Elkind MS, Mohr JP. Carotid artery dissection. Ann.
Rev. Med. 2000; 51: 32947.
11. Balcer LJ, Galetta SL. Neuro-opthalmology of the pregeniculate afferent visual system. J. Neuroophthalmol. 2000; 20: 17391.
12. Foster RE, Kosmorsky GS, Sweeney PJ et al. Horners syndrome secondary to spontaneous carotid dissection with
normal angiographic findings. Arch. Ophthalmol. 1991; 109:
1499500.
13. Biousse V, DAngelejan-Chatillon J, Massiou H et al. Head pain
in non-traumatic carotid artery dissection. A series of 65
patients. Cephalgia 1994; 14: 336.
14. Hedges TR, Friedman DI, Horton JC, Newman SA, Striph
GC. Neuro-Ophthalmology. Section 5 Basic and Clinical Science Course
20002001. San Francisco: The Foundation of the American
Academy of Ophthalmology, 2000; 10912.
15. Grimson B, Thompson HS. Raeders syndrome: a clinical
review. Surv. Ophthalmol. 1980; 24: 199209.
16. Nolph MB, Dion MW. Raeders syndrome associated with
internal carotid artery dilation and sinusitis. Laryngoscope 1982;
92: 11448.
17. Guillon B, Bousser MG. Internal carotid dissection: an update.
J. Neurol. Sci. 1998; 153: 14658.
18. Bilbao R, Amoros S, Murube J. Horner syndrome as an isolated
manifestation of an intrapetrous internal carotid dissection.
Am. J. Ophthalmol. 1997; 123: 5624.

415
19. West TE, Davies RJ, Kelly RE. Horners syndrome and headache due to carotid artery disease. Br. Med. J. 1976; 1: 81820.
20. Kilpatrick C, Sage MR, Burns R. Spontaneous dissection of the
internal carotid artery. Aust. NZ J. Med. 1983; 13: 16972.
21. Mignon F, Qanadli SD, Sissakian JF et al. Post traumatic
aneurysmal dissection of the extracranial artery: helical computed tomographic and angiographic aspects. J. Radiol. 1999;
80: 5857.
22. Brown J, Danielson R, Donahue SP, Thompson HS. Horners
syndrome in subadventitial carotid artery dissection and the
role of magnetic resonance angiography. Am. J. Ophthalmol.
1995; 119: 81113.
23. Stern MJ, Patel M, Davis J, Friedwald J, Slavin M. Painful
Horner syndrome associated with occlusion of the extracranial
internal carotid artery: MR imaging. J. Comput. Assist. Tomogr.
1989; 13: 91820.
24. Gouny P, Nowak C, Smarrito S, Fadel E, Hocquet-Cheynel C,
Nussaume O. Bilateral thrombosis of the internal carotid arteries after a closed trauma. Advantages of magnetic resonance
imaging and review of the literature. J. Cardiovasc. Surg. 1998;
39: 41724.
25. Fiebach J, Brandt T, Knauth M, Jansen O. [MRI with fat
suppression in the visualisation of wall haematoma in spontaneous dissection of the internal carotid artery.] Rofo Fortschr.
Geb. Rontgenstr. Neuen Bildgeb. Verfahr. 1999; 171: 29093 (in
German).
26. Sturzenegger M. Spontaneous internal carotid dissection.
Early diagnosis and management in 44 patients. J. Neurol. 1995;
242: 2318.
27. Stringaris K, Liberopoulos K, Giaka E et al. Three-dimensional
time-of-flight MR angiography and MR imaging versus conventional angiography in carotid artery dissections. Int. Angiol.
1996; 15: 2025.
28. Oelerich M, Stogbauer F, Kurlemann G, Schul C, Schuierer G.
Craniocervical artery dissection. MR imaging and MR angiographic findings. Eur. Radiol. 1999; 9: 138591.
29. Auer D, Karnath HO, Nagle T, Dichgans J. Noninvasive
investigation of pericarotid syndrome. Headache 1995; 35:
1638.
30. Bakke SJ, Smith HJ, Kerty E, Dahl A. Cervicocranial artery
dissection. Detection by Doppler ultrasound and MR angiography. Acta Radiol. 1996; 37: 52934.
31. Pannone A, Bertoletti GB, Nesi F, Varakin I, Maritati G,
Rabitti G. Carotid artery dissection. Correlation of different
diagnostic techniques. Minerva Cardioangiol. 2000; 48: 1927.
32. Kline LB, Vitek JJ, Raymon BC. Painful Horners syndrome due
to spontaneous carotid artery dissection. Ophthalmology 1987;
94: 22630.
33. Bertram M, Ringleb P, Fiebach J, Orberk E, Brandt T, Hacke W.
[Spectrum of neurological symptoms in dissections of brainsupplying arteries.] Dtsch. Med. Wochenschr. 1999; 124: 2738
(in German).