Treatment of “Other” Stroke Etiologies

Arterial Dissection ^,

Dissection of the internal carotid artery and vertebral artery may occur as a result of significant head and neck trauma but may also occur spontaneously or after a trivial injury. A number of underlying connective tissue disorders appear to be risk factors for spontaneous dissection, including fibromuscular dysplasia (FMD), Marfan syndrome, Ehlers-Danlos syndrome (type IV), osteogenesis imperfecta, and other genetic conditions in which collagen is abnormally formed. At present, none of these underlying conditions are amenable to specific treatment, although several of these conditions may be associated with systemic abnormalities warranting monitoring and in some cases intervention. Further, identification of conditions such as Ehlers–Danlos is relevant as endovascular procedures are extremely high risk in this population and should generally be avoided.

Ischemic stroke may result from either extracranial or intracranial dissection. However, intracranial dissection may also produce subarachnoid hemorrhage (SAH) and will be discussed separately below. The clinical manifestations of extracranial carotid and vertebral artery dissection are distinct, but their treatment appears to be identical.

In the first 4.5 hours after the onset of acute ischemic stroke, thrombolytic therapy should be considered regardless of whether dissection is suggested by the patient’s history or examination. In the setting of dissection, there is a theoretical risk of causing increased hemorrhage into the vessel wall. However, there have been no reports to substantiate this theoretical risk. To the contrary, multiple case series have demonstrated no dissection-specific complications related to intravenous tissue plasminogen activator (tPA) and have found outcomes with tPA comparable to patients with more typical stroke etiologies. ^, In patients with large vessel occlusion eligible for mechanical thrombectomy, outcomes are similar in those with extracranial cervical artery dissection treated with thrombectomy compared to those with other stroke mechanisms. ^, Therefore the presence of dissection should not modify the decision to proceed with thrombectomy in otherwise eligible patients. In the setting of acute tandem occlusions in which the cervical carotid artery is occluded from dissection, thrombectomy alone may achieve successful recanalization of the carotid; whether additional emergent angioplasty and stenting should be performed in such cases is uncertain. ^,

Historically, there has been considerable controversy about the optimal strategy for prevention of recurrent stroke in patients with dissection. Mechanistically, stroke related to dissection may be a result of thromboembolism or hemodynamic compromise, but thromboembolism seems to be the dominant mechanism, with microembolic signals detectable on transcranial Doppler monitoring in about half of patients. For many years, early anticoagulation with heparin or low-molecular-weight heparin (LMWH) was recommended at the time of diagnosis, particularly because the risk of stroke appears to be greatest in the first few days after the initial vascular injury. As with thrombolysis, it has been proposed that immediate anticoagulation of an acute dissection could cause worsening hemorrhage into the vessel wall, but this has never been proven to occur. A Cochrane database systematic review and pooled analysis of 1285 patients with carotid dissection in 36 studies (all of which were observational studies) reported a nonsignificant trend towards greater risk of death or disability with antiplatelet compared to anticoagulant therapy (odds ratio [OR], 1.77; 95% confidence interval [CI], 0.98–3.22, P = .06). Recurrent stroke was not significantly different between anticoagulant and antiplatelet treated groups (1.87% with anticoagulation vs. 2% with antiplatelet therapy). These data are severely limited given the nonrandomized nature of the included studies and the extreme variability in stroke rates reported across the larger observational studies, ranging from 0.3% to 10%. ^, More recently, the Cervical Artery Dissection in Stroke Study (CADISS) ^, was completed and represents the only randomized controlled trial data addressing prevention of recurrent stroke in patients with dissection. CADISS randomized 250 subjects to either antiplatelet or anticoagulant therapy within 7 days of symptom onset. Type of antiplatelet was at the discretion of the enrolling investigator; single antiplatelet agent was used in 56% of patients, and dual in 44%. Mean time to randomization was 3.7 days after symptom onset. Study treatment was continued for 3 months, after which choice of antithrombotic therapy was left to the discretion of the treating physician. At 3-month follow-up, ipsilateral ischemic stroke occurred in 3 of 126 subjects in the antiplatelet group compared to 1 of 124 subjects in the anticoagulant group, but there was 1 SAH with hydrocephalus in the anticoagulant group, yielding no net difference between groups. At 1-year follow-up, there were four ipsilateral strokes in the antiplatelet and two in the anticoagulant group. There was no difference in residual narrowing or persistent vessel occlusion between groups. In both this trial and in earlier observational studies, most recurrent strokes occurred early after symptom onset, such that the choice of antithrombotic therapy may be influenced by the time from initial symptoms to presentation. Given the evidence from CADISS, most patients with dissection should be treated with antiplatelet therapy; dual antiplatelet therapy may be appropriate in those at higher risk (shorter interval between onset and presentation, ischemic symptoms at onset).

Dissections usually heal over time, and patients are commonly maintained on antithrombotic therapy for at least 3 months. The 3 month duration of therapy is arbitrary, and some authors suggest that imaging studies be repeated to confirm recanalization of the dissected vessel prior to a change in therapy. ^, Patients with complete vascular healing and those with complete and persistent occlusions on follow-up imaging appear to be at low risk and prolonged aggressive antithrombotic therapy (such as dual antiplatelet therapy), if employed initially, is not necessary. Whether prolonged aggressive antithrombotic therapy is warranted for dissections with residual luminal irregularities and stenosis is unknown, but given the low risk of recurrent events even in these patients, changing to a single antiplatelet agent is generally favored. There are no reliable data to indicate whether prolonged or indefinite antiplatelet therapy should be used beyond the first few months; the perceived risk of recurrent dissection may be taken into consideration when making this decision. A dissecting aneurysm (often erroneously referred to as “pseudoaneurysm”) may occur in 5%–40% of patients with dissection; these were seen in 16% of the subjects in CADISS with available imaging. ^{, ,} These have been thought to represent a potential source of thromboembolism and to pose a risk of arterial rupture. Consequently, aggressive treatments such as ligation of the parent artery, bypass procedures, and stenting have been advocated. However, long-term observational data have demonstrated a very low risk of complications related to these aneurysms, arguing against aggressive intervention. Three case series have been reported including a total of 89 patients with 109 aneurysms, with average follow-up ranging from 3 to 6.5 years. No cases of aneurysm rupture were identified, and there were only three recurrent cerebral ischemic events, none of which was clearly related to the aneurysm. ^{, ,} Most of these patients were treated with either early anticoagulation for a few months followed by long-term antiplatelet therapy or antiplatelet therapy alone.

While most ischemic strokes due to dissection are a result of early thromboembolic phenomena, a minority appear to be due to hemodynamic compromise. ^, The prognosis may be worse in these cases, and revascularization procedures such as stenting or surgery have occasionally been proposed in this setting, although prospective studies do not currently exist. ^{, ,} Routine treatment of clinically asymptomatic persistent stenosis following dissection is probably not warranted given that procedural risk seems to outweigh the long-term risk of stroke with medical management.

Intracranial dissection may result in either ischemic stroke or SAH. In the setting of SAH, anticoagulation is contraindicated and patients should be medically managed as with aneurysmal SAH. In the setting of ischemic stroke, the same treatment principles likely apply as for extracranial dissection, but this has not been studied. Caution and vigilance is certainly warranted since hemorrhage may be catastrophic. It has been proposed that intracranial dissections that cause ischemia could be treated with surgical extracranial-to-intracranial arterial bypass to avoid anticoagulation, but again data regarding this approach are completely lacking. At least one report of acute extracranial-to-intracranial arterial bypass surgery following intra-arterial thrombolysis in a patient with intracranial carotid dissection has been reported.

Some dissections are believed to occur without producing any symptoms and therefore are presumed to remain completely unrecognized. Therefore it is possible that some dissections have a benign prognosis and do not require therapy. Unfortunately, there is no reliable method to identify these low-risk patients at present and observation without any therapy cannot be recommended.

Patients who have had cervicocerebral arterial dissections should probably avoid activities that may cause sudden rotation or extension of the neck. However, no reliable data exist to define the limits of activity for these patients. There is no apparent reason to manage their physical therapy differently during rehabilitation after stroke because of the dissection.

Fibromuscular Dysplasia

FMD is a noninflammatory arteriopathy that predominantly affects the extracranial cephalic, renal, splanchnic, and iliac arteries. About one-third of patients with FMD also harbor intracranial berry aneurysms. FMD is frequently an incidental finding on conventional angiography but is occasionally associated with stroke. When FMD is implicated as a cause of ischemic stroke, it is usually by way of a poorly characterized predisposition toward arterial dissection, although local atherosclerosis and in-situ thrombosis have been described. Hypertension is common in patients with FMD and must be aggressively controlled to prevent the development of cardiovascular and cerebrovascular atherosclerotic disease. When hypertension is present in a patient with FMD, evaluation of the renal arteries is indicated to identify renal artery stenosis from FMD causing secondary hypertension, which may be effectively treated with renal artery angioplasty in many cases.

Acute treatment of stroke attributed to FMD is similar to that of patients in general. There is no current evidence to suggest a differential response to thrombolytic therapy or mechanical thrombectomy in patients with FMD.

Optimal treatment of FMD to prevent stroke is uncertain since there is a paucity of information about its natural history, though it is often believed to be benign. In a series of 79 patients with FMD, only one elderly patient had a stroke in the territory of the affected artery, and it occurred 18 years after the initial diagnosis. Two other elderly patients had strokes in regions that were unrelated to the affected vessel, about 4 and 11 years after diagnosis. All three of these patients were on no treatment, while no strokes or transient ischemic attacks (TIAs) occurred among patients treated with antithrombotic medications. Thus, even without therapy, the risk of stroke related to FMD appears to be relatively low and may be particularly minimal in younger patients in whom FMD is an often incidental finding. Nevertheless, antithrombotic strategies should be considered since they appear to further ameliorate the risk, most importantly among patients with FMD and stroke without an alternative etiology. In most cases, antiplatelet therapy is preferable to anticoagulation since the annual risk of stroke (approximately 1%–2%) appears to be less than the bleeding risk attributable to anticoagulation. Unless there is concomitant evidence of significant atherosclerotic disease or marked hyperlipidemia, there is no indication for statin therapy.

Surgical and endovascular treatment has been advocated for patients with FMD causing symptomatic focal carotid artery stenosis. Since FMD is often associated with atherosclerosis, patients with accessible symptomatic carotid artery stenosis that appears atherosclerotic should probably be managed as if the FMD were an incidental anomaly. In the absence of concomitant atherosclerosis, surgical intraluminal dilatation of the FMD-related stenosis has been attempted with mixed results. ^{, ,} Perioperative morbidity is estimated at 3%–6% and may negate any potential benefit compared to medical therapy alone or no therapy at all. Angioplasty and/or stenting of FMD has also been successfully performed in a number of cases and is increasingly used, although the efficacy of this approach in stroke prevention has not been formally assessed. Surgical considerations for FMD-associated intracranial aneurysms are likely to be similar to those for all intracranial aneurysms.

One specific subtype of FMD involves the carotid bulb giving a radiographic appearance of a carotid web. While data are limited to small case series, the risk of recurrent stroke in this specific scenario is reported to be substantial with medical therapy, and surgical removal of the affected region may be of benefit. In one series of 25 Afro-Caribbean patients, 6 of 20 (30%) treated with antiplatelet therapy had recurrent stroke compared to none of 7 patients who underwent surgical excision. In a systematic literature review, reports of 157 patients were identified, of whom 25 of 45 treated with medical therapy had a recurrent stroke (with similar rates in those treated with antiplatelet and anticoagulant medication) compared to none of 42 treated with carotid surgery or stenting. These data are obviously limited by selective reporting and publication bias.

Vasospasm After Subarachnoid Hemorrhage

Among survivors of aneurysmal SAH, symptomatic vasospasm is a leading cause of morbidity and mortality. Although vasospasm resembles a dynamic constrictive process, there is evidence that it is largely a proliferative arteriopathy. Compromise of the vascular lumen may lead to impaired cerebral autoregulation and ultimately to ischemia. This nonatherosclerotic vasculopathy and its treatment are discussed extensively elsewhere in the text (see Chapter 29 ) and will not be reiterated here.

Reversible Cerebral Vasoconstriction Syndromes

Reversible cerebral vasoconstriction syndromes (RCVS) encompass several interrelated disorders associated with dysregulation of cerebral vascular tone, leading to vasoconstriction of medium and large arteries. Patients often present with thunderclap headache, and SAH, intracerebral hemorrhage, or ischemic stroke may occur. A number of vasoactive drugs have been identified as possible precipitants of RCVS ( Box 58.4 ). A detailed history with attention to these agents should be taken and their use, if present, permanently discontinued. Rarely, RCVS occur secondary to underlying conditions, including pheochromocytoma, carcinoid tumors, hypercalcemia, porphyria, or following vascular or neurosurgical procedures. Such conditions should be identified and specific treatment implemented as appropriate.