Front Neurol Neurosci. Basel, Karger, 2014, vol 33, pp 41-68 (DOI: 10.1159/000351891)

______________________

Risk Scores for Transient Ischemic Attack

M.E. Wolf · V.E. Held · M.G. Hennerici

Department of Neurology, UniversitätsMedizin Mannheim UMM, University of Heidelberg, Mannheim, Germany

______________________

Abstract

The risk of recurrent ischemic stroke after a transient ischemic attack (TIA) has been reported to be 5-10%, and is elevated especially within the first days after the index event. Since TIA primarily has a good outcome without persisting new deficits, interest has been growing to predict stroke recurrence after TIA. This has led to the development of scores, initially for long-term prognosis such as the Stroke Prognosis Instrument (SPI) or the Hankey score, which both have shown a good predictive value at 1 or 2 years after TIA. Risk factors such as age, hypertension or cardiovascular disease were integrated in these systems. Since the early risk prediction for stroke in patients presenting within 24 h after onset of symptoms became clinically more and more relevant in emergency stroke units, the ABCD score (for the predictive factors Age, Blood pressure, Clinical symptoms, Duration of symptoms) was developed. Validation was promising, and hence further scores were developed, which entailed a large number of studies trying to validate these systems or to improve them (e.g. ABCD², ABCD²I, ABCD³, ABCD³I, CIP model, ASPIRE approach, ABCDE+ etc.). The main approaches were to include imaging results (such as DWI positivity) or etiologic considerations (e.g. carotid stenosis or atrial fibrillation). However, these new scores necessitate an extensive diagnostic workup, and therefore can only be used in large stroke centers. Currently, for acute TIA management, the use of ABCD² is recommended in several guidelines.

Copyright © 2014 S. Karger AG, Basel

The risk of recurrent ischemic stroke after a transient ischemic attack (TIA) has been reported to be 5-10% [1]. Especially within the first days after the index event, a high percentage of recurrent neurologic symptoms has been reported [2]. Correspondingly, 15% of patients suffering from ischemic stroke report previous TIA [3], and the prevalence of TIA preceding ischemic stroke (anamnestic details and informa-tion of chronic ischemic lesions on MRI) has been estimated at >20% in some studies [2].

Due to the transient character of neurologic symptoms in TIA, the neurologic outcome of patients seems favorable. However, with the elevated risk of early recurrent ischemic stroke, TIA must be considered a warning sign for an unfavorable course. Therefore, a complete stroke workup after TIA should be performed to detect potential etiologies (e.g. atrial fibrillation (AF) or carotid stenosis), which can be treated immediately to prevent major ischemic stroke in the further course.

Since subsequent stroke might occur within 24 h after TIA, there is a need to early identify patients with a risk to develop ischemic stroke because they need a full workup and/or treatment and therefore necessarily hospitalization. A distinction between TIA patients needing acute diagnostic workup in hospital and TIA patients who can be further evaluated in an outpatient clinic would permit a patient-adapted workflow, sparing some patients an unnecessary long stay at hospital while providing a secure stay at hospital for high-risk patients. Additionally, in a time period when economic constraints become more and more important, the risk- and cost-benefit relation could be optimized in this way.

These notions have led to several attempts to stratify the risk of stroke after a first TIA by creating risk scores and application of these to validate the screening tool. In the closely related specialty of cardiology, the Framingham Score has shown good results and usefulness of such a score, in this particular case for detection of coronary heart disease [4]. Therefore, the idea was to find a similar scoring system to identify TIA patients with a high risk for recurrent stroke.

The idea to build a score needs many variables to consider before developing a useful tool. A good score should include characteristics such as [5] predictive value and consistency of performance (in different studies with different settings), and the score should be easy to calculate for daily use in emergency room situations.

It is important to know that the risk estimation also depends on the clinical specialty of the physician (e.g. a general practitioner or a specialized stroke neurologist) and the setting of evaluation (e.g. emergency room or TIA clinic), which therefore influence the results of validation. Finally, the method of data extraction might bias the results as well.

Several attempts have been made since the 1990s, somehow with different points of view or interests. Scores could be divided into short-time risk of recurrent stroke versus middle- or long-time risk [6]. Another classification could differentiate between scores with only clinical information versus scores including complementary investigations: scores for TIA only, or TIA and minor stroke, or even risk of stroke after TIA versus the prediction of a first stroke or TIA based on clinical information. In this chapter, however, we would like to take a sequential approach, describing consecutive developments of different scores, since this might best reflect the dynamics and growing importance of such a system in stroke neurology.

Chapters will be organized into 3 paragraphs: ‘Background’ will shortly explain the aim of the study or development of the score in its time, ‘Definition of the System’ will follow for the reader, who wants a quick overlook. In the ‘Validation’ paragraph, the development of the score (which might be the basis for the ‘Definition’ paragraph) and further evaluation and application of the score are discussed.

Risk of Recurrent Stroke after Transient Ischemic Attack

The history of risk score developments can be divided into two main aspects: evaluation of long-term risk for stroke, which coincides with the time period in the 1990s, and evaluation of short-time prognosis, which was focused on in the 2000s. In consequence, we divided this chapter into these two parts.

Subsequently, we chose to show the most important results in our opinion to provide a best available comprehensive presentation and tried to generate a systematic overview despite the heterogeneous data due to different designs, populations, and outcome parameters in the performed studies. Table 1 summarizes the scores with their definitions and risk group assessment.

First Long-Term Approaches: Stroke Prognosis Instrument I/II and Hankey Score, etc.

Stroke Prognosis Instrument I (SPI-I) [7]

Background

The SPI was developed for patients with transient ischemia or minor ischemic stroke in the vascular territory of the internal carotid artery. Patients had no previous cerebrovascular event; patients with an artificial heart valve were excluded. The focus on the anterior circulation was due to the special interest of the investigators in carotid artery morphology.

Definition of the System

The initial SPI prognostic system used three parameters to define risk groups: age >65 years (3 points); diabetes mellitus (3 points); arterial hypertension (2 points). Arterial hypertension was defined using measurements on day 1-3 at hospital and categorized as follows: severe hypertension > 180/100 mm Hg (on 2 days); no hypertension < 170/94 mm Hg (not higher at ≥1 day); mild hypertension between ‘severe’ and ‘no’ hypertension. Potential risk groups were defined as follows: risk group 1: 0 points; risk group 2: 1-5 points; risk group 3: 6-8 points.

The initial study to validate these factors led to a modified SPI, defined as the final SPI-I. Coronary heart disease (1 point) and the distinction between stroke and TIA for baseline event (2 points for stroke) were integrated ending up in the following sys-tem: age >65 years (3 points); diabetes (3 points); hypertension (2 points); coronary heart disease (1 point); distinction between stroke and TIA for baseline event (2 points for stroke). Risk groups were divided into risk group 1: 0-2 points; risk group 2: 3-6 points and risk group 3:7-11 points. Table 1 gives an overview of all scores with their definition and risk group assessment.

Table 1. Risk scores for TIA (some for TIA and minor stroke) to predict the recurrence of stroke after TIA (detailed explanations are given in the corresponding paragraphs)

Related posts:

Clinic: A Major Advance in Management of Transient Ischemic Attacks Therapy in Transient Ischemic Attack Patients of Transient Ischemic Attack Definition Ischemic Attack as a Medical Emergency Concept of Acute Cerebrovascular Syndrome of Transient Ischemic Attack

Stay updated, free articles. Join our Telegram channel

Join

Full access? Get Clinical Tree