4.1 Introduction

Acute ischemic stroke (AIS) is at present a high-ranking health problem on account of the high percentage of the population it affects and the sequels it can determine.

During the last few years the approach to this disease has undergone considerable modification, mainly on account of the publication of numerous clinical trials ^{1 ,​ 2} demonstrating excellent therapeutic results.

Clearly, the main therapeutic objectives are to restore blood flow in the affected area and to decrease the effect of ischemia on tissue. ³

The main change has taken place at the hyperacute stage, where the introduction of different protocols of patient selection and their corresponding treatment guidelines have had an impact on the prognosis regarding survival and functional abilities. ⁴

While therapy progressed, different imaging techniques provided substantial information as a basis for therapeutic protocols and also for a better understanding of this pathology. ⁵

Imaging has opened up, consequently, a series of possibilities: accurate diagnosis of ischemia or ischemia-like diseases (differential diagnoses), identification of potentially salvageable tissue (penumbra or tissue at risk for ischemia), localization of vascular occlusion and analysis of same, and finally selection of patients to be treated during the hyperacute phase and choice of the corresponding imaging method. ¹

AIS is, beyond doubt, a medical emergency. Therefore, the chosen imaging method must be quickly and accurately performed and promptly interpreted, so as not to delay treatment.

A subject of extensive discussion in the last few years is which would be the best method and how to select patients for it.

Regarding the first stages, the choice between computed tomography (CT) and magnetic resonance (MR) has been the subject of much discussion and analysis. ⁶ It has been clearly demonstrated that CT is excellent at detecting hemorrhage while MR is the most sensitive detector of ischemia at the hyperacute stage. ⁷

As we shall see later on, MR is beyond doubt the most complete method for the study of AIS patients at the hyperacute stage, but its use in an emergency setting poses such logistic problems that many centers keep it as a resource for specific situations. ⁷ Notwithstanding, the last years have seen a trend toward the increasingly frequent use of MR as a basis for the choice of therapy, on account of its unique contributions.

We will review the present state of the different available imaging techniques for the study of AIS patients at the hyperacute stage and the impact of those techniques on choice and therapeutic management of patients.

4.2 Objectives of Imaging Methods

Neuroimaging plays an important role in the study of a stroke patient and affects therapy in a significant way. That is why most of the algorithms proposed in clinical trials include criteria or information derived from imaging.

The main objective of imaging methods, CT above all, is to establish a diagnosis of AIS in a patient and discard possible differential diagnoses or mimics. Notwithstanding, in the last few years different therapeutic options have provided evidence that has considerably modified the objectives of imaging.

The assessment of parenchyma, the absence of hemorrhage (main contraindication of the use of fibrinolytics) and the confirmation of ischemia are all important, but they are not the only factors. It is also essential to assess the state of brain circulation regarding the vascular tree itself and the main hemodynamic parameters. ⁸

There is no consensus on the imaging method of choice for the selection of AIS patients. In spite of that, it is clear that the chosen method must provide information quickly in order to screen patients adequately for early treatment.

The AHA recommends both CT and MR as methods of choice in patients clinically suspected for hyperacute ischemia. Therefore they should be considered as methods providing complementary information. A guideline should not determine which method is the most adequate one.

After ruling out hemorrhage, it is basic to define whether confirmed ischemia is present and its extension, that is to say the size of the infarct core. As we will see later on, this is a key parameter for the first therapeutic decisions.

Another basic point is the detection of occlusion in a main cerebral artery (a proximal occlusion). The best parameters for patient prognosis are precisely the detection of occlusion and the neurological findings. ⁹

At the present moment it is also necessary to evaluate potentially recoverable tissue, what is called the penumbra area, which is the main target of reperfusion therapies.

Although PET has been established as the gold standard for the definition of ischemic area and penumbra area, it cannot enter daily clinical practice. That is why we have reviewed the contribution of both CT and multimodal MR to the definition and evaluation of those parameters.

4.3 Analysis of Brain Parenchyma: Infarct Core

When a patient presents with a probable AIS at the hyperacute stage the first and basic step is to ascertain the state of the cerebral parenchyma.

Once hemorrhage has been excluded, the existence or absence of ischemia must be ascertained quickly; if ischemia is present, its extension must be defined, because it will constitute the infarct core.

The presence and extension of irreversibly damaged tissue is one of the factors that determines prognosis.

Non-contrast CT is clearly established as the method of first choice and must be performed immediately in AIS is suspected (Class I, Level of Evidence A). ¹⁰ The main objectives remain the elimination of differential diagnoses and mimics of brain ischemia, and the exclusion of hemorrhage, which is the main contraindication for revascularization treatments.

Contrast CT is highly sensitive for the diagnosis of parenchymal and subarachnoid hemorrhage. It is therefore sufficient to exclude patients with these pathologies from reperfusion treatments. ¹

Non-contrast CT has long been accepted as being sufficient in itself as the single method to decide the use of intravenous thrombolytics in AIS patients (Class I, Level of evidence A). ^{11 ,​ 12}

Although this is a firmly established principle, CT is a low-performance procedure for the diagnosis of ischemia in the first hours after its onset. CT performance depends likewise from the location of the infarcted areas. ^{13 ,​ 14}

Within the first 3 to 6 hours of established ischemia, which is the critical period for the selection of patients, CT signs are very subtle; the scan is usually normal.

As the hours go by, ischemic areas show up on CT. Some 36–48 hours after the clinical onset, the compromised areas are visible as well-defined hypodense zones in 100% of cases. ¹⁵

At the hyperacute stage it is of the essence to be able to detect the so-called early ischemic signs on CT. These signs must be quantified.

Early ischemia signs include: effacement of lentiform nucleus, loss of gray-white differentiation at the insula (insular ribbon sign), localized loss of differentiation between white and gray matter and localized effacement of sulci. (Fig. 4‑1 )

Another high-ranking sign is hyperdensity of the middle cerebral artery. This sign does not indicate the presence of ischemia, but it locates the vascular occlusion responsible for the affected area and usually correlates with a proximal thrombus (at the M1 segment of the above mentioned artery). (Fig. 4‑2)

Although this sign has been described for other arteries in the cerebral circulation, it does not perform or correlate so well in those cases.

It is important to remember that this is a low-sensitivity sign and that false positives do appear, in patients with high hematocrit or in cases with severe atheromatosis with calcified vessel walls, for example.

It has been pointed out that undoubtedly the perception of early AIS signs is highly variable, from one observer to another and even for the same observer.

Global sensitivity for these signs during the early period (3 hours) amounts to 30–60%, specificity being around 85%; the negative predictive value is very low, less than 30%.

When the area is clearly identifiable as ischemic, it will present to the eye of the observer as a hypodense zone with loss of gray-white differentiation and a well-defined margin. It is also clearly part of an arterial vascular territory.

Strategies have been devised to quantify the areas of ischemic tissue, which would be a means of patient selection. The ASPECTS (Alberta Stroke Program Early CT score) score is one of them; it allows the quantification of irreversibly damaged tissue in a simple way for patients with middle cerebral artery compromise. ¹⁶

The use of ASPECTS for the selection of patients soon prevailed. Although the extension of ischemic areas does not exclude the patient from treatment, it does exert a significant influence on prognosis. Consequently, low-score patients with the corresponding zones of confirmed ischemia are excluded from early reperfusion therapy.

It has been demonstrated that the use of ASPECTS has some limitations, mainly regarding the borders of affected zones and the concept of volume (singly considered) correlating poorly with prognosis.

Notwithstanding, the ASPECTS concept is widely used at present, even if MR is the preferred method.

MR is a method of high sensitivity and specificity for the diagnosis of ischemia in the early hours after onset, especially since the introduction of diffusion–weighted imaging and apparent diffusion coefficient mapping (DWI/ADC). ^{9 ,​ 17}

On this ground among others, many centers have decided to use MR as a method for the selection of patients.

The DWI/ADC sequence can evaluate the movements of water molecules, which are limited or restricted on account of neuronal membrane alterations produced by brain ischemia. ¹⁷

The sensitivity and specificity of the DWI/ADC sequence for the detection of ischemia in the first 6 hours is over 95% (Level of Evidence A). This diagnostic yield bears no relation to the volume or the location of the infarct. (Fig. 4‑3, Fig. 4‑4)

According to available evidence this imaging modality can detect ischemia in less than 15 minutes after onset.

In clinical practice the tissue areas with restricted diffusion are considered unrecoverable. In spite of that, there have been reports of reversion of altered signals in such areas after reperfusion therapy.

Classically CT was spoken of as a method superior to MR for the detection of early-stage hemorrhage. Nevertheless, MR is similar in sensitivity and specificity to CT in the diagnosis of acute cerebral hemorrhage (Level of Evidence B).

In order to optimize diagnostic yield in this aspect it is basic to add susceptibility sequences (GRE T2 or SWI/SWAN) to the study protocol. These sequences also play a key role in the detection of asymptomatic micro-bleeds. ¹⁸

These focal findings, which were considered of fundamental importance a few years ago, are now viewed differently: the presence of less than 5 micro-bleeds does not significantly increase the risk of symptomatic bleeding in patients treated with thrombolytics (Class IIa, Level of evidence B).

As mentioned above, once the ischemic areas have been identified, their extension must be quantified. (Fig. 4‑5)

Volume quantification of core or infarct nucleus (non-salvageable tissue) is of crucial prognostic importance: it is one of the highest-ranking independent predictive signs.

At present, it is firmly established that core volume, quantified by means of DWI/ADC, is one of the most influential factors for patient prognosis. ¹⁹

A volume of 70–100ml of ischemia in the territory of the middle cerebral artery is highly specific for poor outcome, independently from the existence of penumbra (potentially salvageable tissue) or a successful vascular recanalization. ^{20 ,​ 21}

Furthermore, the risk of significant hemorrhage during treatment increases in proportion to the size of infarcted tissue, especially if the area is bigger than 100 ml. ²² That is why many authors do not recommend reperfusion therapy in this type of patient.

In spite of that fact, no precise cutoff point has been established that would enable to determine which patients should be excluded from reperfusion therapy, because there have been reports on favorable outcome for considerable ischemic areas after treatment.