Dissection of the internal carotid artery is an increasingly recognised cause of acute ischaemic stroke in young adults and children. It may arise spontaneously or secondary to blunt or penetrating trauma. It has been reported after seemingly trivial incidents, such as reversing a car, washing hair, or holding a mobile telephone by flexing the neck against the shoulder. The incidence of carotid dissection is about 2.5–3/100 000/year—similar to aneurysmal subarachnoid haemorrhage.1 The most common presenting features are ipsilateral temporal, retroorbital, or hemicranial pain, Horner’s syndrome, and local cranial nerve palsies, plus potentially devastating cerebral ischaemic events. Although no trial data exist to support the use of anticoagulants, most cerebrovascular specialists advocate initial intravenous heparin then three to six months of warfarin treatment if there have been ischaemic episodes. There is often an interval between onset of symptoms and cerebral infarction enabling the diagnosis to be made and treatment to be instituted.2 Given the potentially fatal or disabling consequences of carotid dissection this window of opportunity is not to be missed. About 80% of ischaemic strokes arise within the first seven days although they can occur four to five weeks after the onset of symptoms.2 3 We report a patient who developed a disabling stroke five months after ipsilateral carotid dissection. This has implications for instituting treatment in a patient seen weeks or months after the incident, and for the duration of subsequent anticoagulation.

A 24 year old man was involved in a road traffic accident. Emergency fire services were required to free him from the vehicle and he remains amnesic for the event. He sustained soft tissue injuries to the face, chest, and arms, but no fractures. No surgery was required. No neurological sequelae were noted although he mentioned mild visual blurring in retrospect. A non-contrast CT was performed the day after the accident which showed a small right peripheral parietooccipital infarct (but this had been reported as a contusion). The patient was discharged and made a full recovery from his injuries.

About four months later he suddenly developed a left facial weakness. There was associated right retro-orbital and temporal headache. Over three to four days the headache and facial weakness cleared. Medical attention was not sought.

Just over five months (158 days) after the traffic accident he was admitted to our unit. That morning he had suddenly developed a dense left hemiparesis affecting the arm and face more than the leg. The right sided headache had returned. A repeat CT showed maturation of the previous occipitoparietal infarct and a new infarct in the corona radiata on the right, extending into the internal capsule. A colour flow Doppler/duplex scan of the cervical carotid arteries showed normal appearances on the left side, a normal right external carotid artery, but reduced diastolic flow in the right common carotid artery. The origin of the right internal carotid artery was patent but the Doppler waveform was severely damped with the typical “bidirectional” signal associated with distal occlusive disease and commonly seen in dissections. The diagnosis of dissection of the right internal carotid artery was confirmed by T1 weighted axial MRI of the neck. These showed characteristic high signal in the vessel wall due to haemorrhage with associated reduction in diameter of the true lumen. The patient was treated immediately with intravenous heparin. Seven days later warfarin was started for six months. His hemiparesis gradually improved.

The right internal carotid artery probably dissected at the time of the road traffic accident and the infarct seen on the initial CT had the characteristics of a peripheral embolic lesion in the territory of posterior superficial middle cerebral artery branches. A second (presumably) embolic insult causing transient facial weakness arose four months later. Another five weeks passed before the disabling left hemiparesis. During the five months since the traffic accident neither the patient nor his family could recall any occasions when he had complained of a new neck or head pain relating to further neck trauma. It seems likely that he had had only one insult to the right internal carotid artery but that over five months the vessel had failed to heal spontaneously and the damaged intima continued to act as a focus for thrombus formation.

Recurrent dissection of the same extracranial vessel is extremely unusual, and occurred in only one of 81 patients followed up for a median of almost three years by Bassetti et al.4 Recurrent dissection involving other extracranial vessels is slightly more common, with an incidence of about 1% a year.5 Our patient had no clinical evidence of the heritable connective tissue disorders sometimes associated with multiple or recurrent dissections (Marfan’s syndrome, Ehlers Danlos type IV, etc). He clearly had three ischaemic episodes over five months attributable to the right internal carotid artery and we think that persisting damage, rather than true recurrence, was the likeliest cause of his stroke.

It is not uncommon to be asked to see young patients with ischaemic stroke several weeks after the event. Even when arterial dissection is suspected or proved, if the patient has had no further episodes for a month or more it is tempting to prescribe antiplatelet agents rather than submitting the patient for formal anticoagulation. This case shows that disabling stroke can occur as long as five months after the initial dissection; thus anticoagulation should probably be considered even when there is considerable delay in referral. Although empirical, a minimum period of six months on warfarin would seem appropriate. If repeat duplex ultrasound remains abnormal at that time, extension to 12 months may be necessary.