Abstract
Cerebral vasospasm is a phenomenon that can occur after subarachnoid hemorrhage, most notably after aneurysm rupture. It represents a significant cause of morbidity and mortality in an already devastating disease. Much research has been conducted to understand the pathophysiology and to use this information to develop targeted therapies to both prevent and treat this disease. Despite this, there are very few effective options available. The rapid onset and significant sequelae mean that the treating physician must have a thorough understanding of the current state of therapeutic options to be able to act quickly and mitigate the effects of vasospasm-induced ischemia.
Key words
vasospasm – delayed cerebral ischemia – subarachnoid hemorrhage – aneurysm – angioplasty – triple-H therapy6 Cerebral Vasospasm
6.1 Goals
Define cerebral vasospasm and differentiate it from other potential causes of neurological decline in the setting of aneurysmal subarachnoid hemorrhage (SAH).
Review the literature that has been paramount to developing our current clinical management strategy of cerebral vasospasm.
Discuss recent therapeutic interventions.
6.2 Case Example
6.2.1 History of Present Illness
A 45-year-old male experienced a sudden onset of severe headache with no loss of consciousness but elected to present to the hospital 3 days later with persistent headache but neurologi-cally intact. Computed tomography (CT) of the head demonstrated diffuse SAH throughout the anterior basal cisterns, and CT angiography revealed a broad-based blister aneurysm of the right supraclinoid internal carotid artery (ICA) (Fig. 6.1a). The patient was started on nimodipine, as well as aspirin and clopidogrel prior to undergoing uncomplicated placement of a flow-diverting stent from the right ICA into the middle cerebral artery (MCA) (Fig. 6.1b). Postoperatively, he developed hypoxia requiring an intensive care unit admission. Imaging studies revealed significant bilateral pulmonary edema. On SAH day 7, the patient’s level of consciousness suddenly deteriorated.
Past medical history: Noncontributory.
Family history: Noncontributory.
Social history: Occasional social alcohol. No smoking history.
Review of systems: Tachypnea and increased work of breathing. S3 noted on auscultation.
Neurologicalexamination: Glasgow Coma Scale (GCS)— E3V4M6; drowsy and disoriented; weak left arm.
Imaging studies:CT and digitalsubtraction angiography (DSA) (see Fig. 6.1 and Fig. 6.2).
6.2.2 Treatment Plan
The patient initially did well after the flow-diverting stent was placed across the aneurysm. Intravenous crystalloid was infused to maintain euvolemia and normotension. Shortly after the treatment, he developed neurogenic pulmonary edema and required supplemental oxygen as well as diuretic therapy. The hypoxia was short lived and he was successfully weaned off of oxygen. Daily transcranial Doppler (TCD) studies were performed beginning on the date of admission, which showed progressively increasing velocities in the proximal right MCA. On post-SAH day 8, the patient developed a decreased level of consciousness and left arm weakness consistent with symptoms of cerebral vasospasm. Plain CT and CT angiography were performed that confirmed significant vasospasm of the right MCA, but no obvious infarction. He was immediately started on a norepinephrine infusion. Despite reaching a systolic blood pressure of 200 mmHg, his symptoms persisted. He was then brought to the angiography suite where an angiogram redemonstrated the severe right proximal MCA stenosis, distal to the flow-diverting stent (Fig. 6.2a). Balloon angioplasty was performed by partial inflation of a 4-mm compliant balloon system, followed by the selective administration of 3 mg of verapamil into the MCA. The procedure resulted in clinical improvement, but not angio graphic resolution, of the vasospasm (Fig. 6.2b). His symptoms subsequently resolved and his level of consciousness returned to baseline while being maintained on a norepinephrine infusion. The patient was gradually weaned off the vasopressor infusion and was discharged home on post-SAH day 18.


6.2.3 Follow-up
The patient was discharged home in a good condition with slight residual left arm weakness. He was seen in the clinic for a follow-up 6 weeks later with a normal neurological examination. DSA at 6 months demonstrated complete parent vessel remodeling with no filling of the blister aneurysm.
6.3 Case Summary
What is the definition of cerebral vasospasm and is it synonymous with delayed cerebral ischemia (DCI)?
Cerebral vasospasm is defined as the prolonged radiographic narrowing of cerebral arteries and can be seen in several pathologies, most notably aneurysmal SAH. The peak incidence in this scenario is between 4 and 14 days postrupture, and radiographic vasospasm is estimated to have an overall incidence of approximately 43%. 1 However, radiographic vasospasm does not imply symptomatology, and thus, it should be differentiated from symptomatic vasospasm, which is estimated to occur in about 33%. The pathogenesis of vasospasm appears to be multifactorial and remains elusive. Several mechanistic processes have been described, including endothelial injury, vascular smooth muscle contraction, and inflammatory cascade activation. 2 Endothelial injury is thought to largely occur in an outside-in fashion, beginning with oxyhemoglobin found in the subarachnoid blood-inducing hydroxyl radical and lipid peroxide formation. 3 , 4 These are thought to diffuse through the arterial wall and disrupt both the vascular smooth muscle cells and the endothelium, by upregulating endothelin-1 (vasoconstrictor) and downregulating nitric oxide production (vasodilator), respectively. 5 , 6 Acute arterial smooth muscle cell contraction is dependent on adenosine triphosphate, while the chronic vasospasm condition appears not to be. Instead, research has shown that after short-term smooth muscle contraction occurs, various contractile proteins are activated (including rho-kinase, protein kinase C, and protein tyrosine kinase), leading to sustained muscle shortening independent of intracellular calcium stores. 7 Finally, various inflammatory proteins are upregulated postaneurysmal SAH in both serum and cerebrospinal fluid and are suspected to play a role in inducing vasospasm. 2 , 8 , 9
DCI, on the other hand, is a more encompassing term for neurological decline and includes other etiologies, including microthrombi, cortical spreading depression, distal microvascular constriction, etc. The literature is fraught with inconsistent terminology, and in an effort to standardize research efforts, a group recently proposed a unifying definition of DCI: “The occurrence of focal neurological impairment, or a decrease of at least 2 points on the Glasgow Coma Scale. This should last for at least 1 hour, is not apparent immediately after aneurysm occlusion, and cannot be attributed to other causes by means of clinical assessment, CTor MRI scanning of the brain, and appropriate laboratory studies.” 10 DCI is, therefore, a diagnosis of exclusion and should be reserved for situations where other potential causes for neurological decline, such as hyponatremia, hydrocephalus, infection, and others, have been ruled out. In the example case, the patient was suffering from DCI secondary to vasospasm of the proximal right MCA as other potential causes were ruled out and an angiogram revealed severe stenosis of a culprit artery.
Are there any predictive factors for the onset of cerebral vasospasm?
The original Fisher grading scale aimed to predict the onset of symptomatic vasospasm after aneurysmal SAH, allowing practitioners to implement protective measures in anticipation. 11 Using CT, the authors found that thick SAH and intraventricular blood were associated with symptomatic vasospasm. Later, Frontera et al published a modification of the Fisher grading scale which better predicted the onset (▶ Table 6.1 ). 12 Our patient had a modified Fisher grade 3 SAH and thus would be quoted a 33% chance of developing symptomatic vasospasm.
Other predictors of symptomatic vasospasm have more recently been identified, which include prolonged clearance of cisternal blood, smoking history, hyperglycemia, systemic inflammatory response syndrome, and hydrocephalus. 13 , 14 Furthermore, there is some evidence that female gender, hypertension, initial loss of consciousness, and perhaps others may portend a higher risk. 13
Are there any prophylactic measures that can be taken? To date, nimodipine is the only therapeutic intervention aimed at preventing cerebral vasospasm that has been shown to improve outcomes. It is a dihydropyridine calcium channel blocker which primarily acts upon the cerebral vasculature and has minimal effects on the cardiac conduction system. Administered 60 mg orally every 4 hours (or 30 mg every 2 h if hypotension is encountered) has been shown in a metaanalysis to improve the vasospasm-induced deficit and mortality rates (odds ratio [OR]: 0.46-0.58). 15 However, no reduction in incidence of cerebral vasospasm has been found.
It is important to also implement general medical measures including maintaining euvolemia; avoiding hyponatremia, hypotension, anemia, and high intracranial pressure; optimizing ventilation and oxygenation; and avoiding surgical clipping during the period of peak vasospasm incidence. 2 , 16 , 17 Syndrome of inappropriate antidiuretic hormone (SIADH) secretion and cerebral salt wasting are potential complications that must be managed aggressively without compromising the aforementioned physiological parameters. For instance, a patient with SIADH should not be fluid restricted, as is the standard treatment. Instead, patients should receive hypertonic saline as needed. During vasospasm, oral fludrocortisone 0.3 mg per day may be used as well. 18 In the example case, our patient received nimodipine 60 mg orally every 4 hours and tolerated the dosing without an issue. Furthermore, neurogenic pulmonary edema occurred, requiring prompt treatment. Many experimental therapies have been trialed in humans with limited success as some have reduced the incidence of angiographic vasospasm but have had no significant effect on clinical outcome. These include strategies to improve subarachnoid blood clearance (intracisternal thrombolytics), prophylactic balloon angioplasty, endothelin-lA receptor antagonists (CONSCIOUS-1, 2, and 3 trials), intrathecal vasodilators, magnesium infusion (MASH and MASH-2 trials), statins (STASH trial), and many others. 14 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26
How would you monitor for the development of vasospasm? It is important to remember the distinction between symptomatic vasospasm and radiographic vasospasm. Of course, symptomatic vasospasm requires clinical deterioration or alteration and is one etiology of DCI. Radiographic cerebral vasospasm may be diagnosed using several different modalities. TCD measures blood flow velocity, which increases during vasospasm, and is a feasible method of noninvasive bedside monitoring. The transtemporal window is most frequently used but limits the assessment to the proximal intracerebral arteries. While blood velocity trends in all arteries may be useful, parameters have been established for predicting MCA and basilar artery (BA) vasospasm.
Blood velocities greater than 200 cm/s for the MCA and 85 cm/s for the BA are highly predictive of vasospasm. 27 However, these increased velocities may be indicative of a hyperemic state, so the Lindegaard ratio was developed to minimize false positives by normalizing velocity by dividing by the proximal arterial velocity (ICA for MCA velocity, and extracranial vertebral artery for BA velocity). A Lindegaard ratio greater than 3 is highly suggestive of vasospasm.
While TCD may be used to monitor for vasospasm, other imaging modalities are used for diagnosis and include CT angiography, CT perfusion, magnetic resonance angiography (MRA), MR perfusion, positron emission tomography (PET), and single-photon emission computed tomography (SPECT). DSA, although the most invasive, remains the gold standard for diagnosing cerebral vasospasm. Two-dimensional DSA exhibits a nearly 100% sensitivity and specificity. Mills et al recently published a thorough review comparing the advantages and disadvantages of these various modalities. 28 More invasive monitoring techniques are also available, which include microdialysis and brain tissue oxygenation.
Microdialysis can only provide a highly specific regional alteration in metabolism, with increased lactate and gluta-mate occurring approximately 24 hours prior to clinical ischemia. 29 Similarly, brain tissue oxygenation is only measured in one region but can detect tissue hypoxia with an oxygen tension below 15 mmHg.
Finally, electroencephalography can be used to predict vasospasm. Studies have shown that the alpha wave to delta wave ratio and percent alpha wave variability can predict vasospasm and often precedes the onset by approximately 2 days. 30
Our patient developed gradually increasing velocities in the right MCA territory. As the patient remained asymptomatic initially, we elected to observe carefully. At our center, we do not routinely rely on perfusion imaging at the onset of symptoms. Instead, we generally confirm TCD results with CT angiography, then perform DSA with an intention to treat, after other potential causes of neurological decline have been ruled out.
What factors would you consider when deciding whether or not to treat cerebral vasospasm in this patient?
As mentioned previously, a distinction must be made between radiographic and symptomatic vasospasm. Neurological symptoms appear to occur in approximately half of the patients demonstrating angiographic large artery narrowing. 31 Treatment of cerebral vasospasm, whether through medical or interventional means, should not be performed if the patient is asymptomatic. However, it may be prudent to consider treatment in patients who are comatose and demonstrate severe angiographic vasospasm if a reliable neurological examination is not available. In this case, our patient was not comatose but demonstrated a decreased level of consciousness and focal neurological deficits attributable to the vasospastic artery. With good neurological function prior to his decline, and minimal lag time after symptom onset, we elected to treat swiftly and aggressively.
6. What options are available for treatment?
Once the diagnosis of symptomatic cerebral vasospasm has been made, there are several treatment options available. Prior to endovascular therapy, triple-H therapy was the mainstay treatment, theorized to improve rheology to optimize cerebral blood flow and oxygen transport. This regimen consists of hemodilution, hypervolemia, and hypertension. Expansion of the effective circulating volume using an infusion of isotonic crystalloid effectively addresses the former two; however, there is evidence that cerebral blood flow does not change between a hypervolemic state and a euvole-mic state. 17 A hypervolemic state is associated with potential complications in those with cardiorespiratory comorbidities and thus euvolemia should be targeted. Furthermore, excessive hemodilution may be detrimental to blood’s oxygen-carrying capacity. Perhaps a more important strategy is to achieve relative hypertension only, as it may be more beneficial than hemodilution and hypervolemia. 32 , 33 Hypertension should be aggressively achieved using inotropic infusions, but care should be taken in the setting of an unsecured aneurysm.
In the setting of symptomatic vasospasm with correlating angiographic findings, angioplasty may be employed as early as possible if inducing hypertension and maintaining euvolemia fail to resolve symptoms. The prophylactic use of angioplasty is not indicated. 19 Angioplasty is generally performed on affected proximal arteries only with a carefully sized balloon for the artery of interest. Distal arterial balloon angioplasty (beyond Ml or Al segments) has traditionally been thought to incur a high risk of rupture given the thinner wall. However, this notion has been challenged. 34 For proximal artery vasospasm, success rates have improved since the technique’s inception, with a recent study demonstrating approximately a 97% success rate in relieving radiographic vasospasm with a complication rate of 1%. 35 The rate of clinical improvement appears to be slightly lower. Local intraarterial infusions of vasodilators, including nicardipine (dihy-dropyridine calcium channel blocker), verapamil (non-dihy-dropyridine calcium channel blocker), and milrinone (phos-phodiesterase-III inhibitor), have been studied for treating more distal arterial vasospasm with mixed results. 36 , 37 , 38 , 39 , 40 , 41 Intravenous milrinone infusions have also been investigated due to the drug’s inotropic and vaso-dilational effects. 42 Although many centers employ intraarterial infusion of vasodilators, including our own, at this time there is insufficient evidence to recommend the routine use of these therapies.
Several other investigational therapies for symptomatic vasospasm are undergoing animal and human trials, including endothelin-lA receptor antagonists, magnesium, nitric oxide, statins, and others. 2 , 43 Thus far, no other successful therapeutic strategies have been borne out of these trials. We rapidly initiated vasopressor therapy upon symptom onset and did not have target blood pressure limitations due to cardiorespiratory comorbidities. Despite achieving a systolic blood pressure of 200 mmHg, our patient did not demonstrate symptom resolution; thus, we brought him to the angiography suite for immediate assessment of the vasculature and treatment of vasospasm with balloon angioplasty and selective intra-arterial verapamil administration.
Grade | Description | Symptomatic vasospasm rate (%) |
0 | No SAH or IVH | 0 |
1 | Focal or diffuse thin SAH, no IVH | 24 |
2 | Focal or diffuse thin SAH, IVH present | 33 |
3 | Thick SAH, no IVH | 33 |
4 | Thick SAH, IVH present | 40 |
Abbreviations: IVH, intraventricular hemorrhage; SAH, subarachnoid hemorrhage. Note: Thin SAH < 1mm< thick SAH. |

Stay updated, free articles. Join our Telegram channel

Full access? Get Clinical Tree

