Aphasia is an acquired disorder of language resulting from brain damage. Understanding the history, controversies, and neuropsychology of aphasia are requirements for every neurologist in training, and there are several excellent reviews available that discuss these topics in greater detail.1–3 For bedside purposes, however, a simpler, clinically focused approach consisting of the following three steps is necessary:
1. Determine if the problem is aphasia or a mimic.
2. Classify the type of aphasia based on bedside examination.
3. Determine the etiology and attempt to treat it if possible.
Mimics of aphasia
The first step in evaluating a patient with a possible acute language disturbance is to determine whether they are actually aphasic. The two problems that are most often misidentified as aphasia are dysarthria and confusion.
Dysarthria is an abnormality in the mechanical production of speech (Chapter 8). It is most easily distinguished from aphasia by the absence of word-finding or comprehension difficulties. In most cases, dysarthria is secondary to intoxication with drugs (both prescription and illicit) or alcohol, or to a metabolic disturbance such as hyponatremia or hypoglycemia. In rare instances, acute-onset isolateddysarthria may be secondary to stroke.4
Differentiating confusion from aphasia is often challenging, but doing so is very important because the metabolic derangements that produce confusion are quite distinct from the vascular lesions that are most commonly responsible for aphasia. Although confusion is principally a disorder of attention, any cognitive domain, including language, may be affected (Chapter 1). Language examination in the confused patient will often show normal fluency, poor comprehension, and normal repetition, potentially leading to the erroneous impression that the patient has a transcortical sensory aphasia. Confused patients are most reliably distinguished from aphasic ones by the presence of widespread behavioral abnormalities outside the language domain.
Bedside examination of the aphasic patient
Because the most common acute cause of aphasia is ischemic stroke, it is important to evaluate the patient rapidly in order to determine their eligibility for intravenous thrombolysis. Once the patient is stabilized and any possible treatment for acute stroke is administered, you can assess language in greater detail. More than an academic exercise, the classification of aphasia helps to define the patient’s language capabilities and to predict the recovery and evolution of their language. Correct classification of aphasia also allows succinct communication with other health-care providers including speech pathologists and rehabilitation experts. For a neurologist’s purposes, assessment of aphasia requires evaluation of:
1. Spontaneous speech
2. Comprehension ability
4. Confrontation naming
5. Reading, both aloud and for comprehension
Careful listening to spontaneous speech while taking the history often provides most of the essential details about a patient’s language dysfunction.
Fluent speech is characterized by a normal or increased rate of word production with normal phrase lengths, while nonfluent speech is characterized by a paucity of verbal output and short phrase lengths. Fluent aphasics use excessive numbers of “filler” words such as prepositions, conjunctions, and adjectives. Despite the excessive number of words, content is lacking. Nonfluent aphasics generally use a preponderance of content-rich words such as nouns and verbs. While their utterances are short and agrammatic, they often convey a great deal of meaning. As a general rule, fluent aphasias are caused by lesions posterior to the central sulcus, while nonfluent aphasias are caused by lesions anterior to the central sulcus.
Paraphasic errors are word substitutions, and may be classified broadly into semantic and phonemic errors. Semantic paraphasic errors are those in which the word produced is related in meaning to the target word. Examples of common semantic paraphasic errors include simplifications (e.g. finger for thumb), substitutions of one item for another of the same class (e.g. toe for thumb), and substitutions of the whole for the part (e.g. hand for thumb). Phonemic paraphasic errors are those in which individual phonemes (segments of sound) are substituted (e.g. tadle for table), omitted (e.g. tale for table), or added (e.g. tadable for table) incorrectly.
Articulatory errors are apparent word substitutions produced by patients with dysarthria. These are not technically paraphasias, but in some cases may resemble phonemic errors.
Neologisms are new words that are formed from appropriate phonemes but which do not resemble an identifiable target word. These are particularly characteristic of posterior aphasias such as Wernicke’s aphasia.
Circumlocution is a circling in on the target word in which the patient uses descriptors of the word rather than the word itself (e.g. thing you eat with for fork). It is a characteristic feature of conduction aphasia.
While it is generally true that nonfluent aphasics comprehend spoken language better than fluent aphasics do, all aphasic patients have some degree of comprehension impairment. Conversely, the comprehension abilities of fluent aphasics (in whom comprehension is traditionally described as being poor) are often preserved to some degree. When assessing comprehension, it is important to establish a floor and a ceiling of performance. In sequence, ask the patient to do the following:
• Follow commands that involve the midline of the body, such as opening and closing the eyes and sticking out the tongue. The ability to follow these commands is preserved in patients with all but the most severe comprehension deficits.
• Answer simple yes/no questions that require only head nodding. Be sure to alternate questions that elicit both yes and no responses, as many patients may continue to nod “yes” without actually understanding the questions.
• Perform simple limb movements such as raising the left or right hand or pointing to objects around the room. Keep in mind that right arm weakness and apraxia (Chapter 4) may accompany aphasia, thus limiting the ability to perform and interpret this type of test.
• Follow sequential commands such as “point to the door, then the light, then the window.”
• Follow out-of-sequence commands such as “after pointing to the door, but before pointing to the light, point to the window.” These are more difficult, especially for nonfluent aphasics in whom understanding complex syntactic structures is impaired.
• Answer listening comprehension questions such as, “The wolf was chased and eaten by the sheep. Who died?” Questions such as these in which both the traditional subject–verb–object syntax and the expected logic of the sentence are altered are often helpful in detecting subtle deficits in nonfluent aphasic patients with apparently preserved comprehension.
While the ability to repeat is seldom applied to everyday life situations, testing repetition is useful in distinguishing among the different aphasia syndromes. The general rule is that repetition is poor in aphasias derived from lesions adjacent to the Sylvian fissure, whereas repetition is relatively preserved in extrasylvian aphasias. To test repetition, start with common, single-syllable words such as “cat” and “dog.” The ability to repeat these words is preserved in all but those with the most severe difficulties. Next, test the ability to repeat simple subject–verb–object sentences such as, “The boy threw the ball.” Finally, ask the patient to repeat complex sequences such as, “After coming home from work, they ate breakfast in the living room.” Note specific problems with repetition in each case. Patients with nonfluent aphasias tend to omit prepositions and conjunctions but repeat content-rich words correctly. Inattentive patients might be able to repeat only the first few (or last few) words of a complex phrase correctly. All repetition impairments should be judged in terms of their relative severity compared with other language deficits. For example, patients with transcortical motor aphasia may make mild errors when repeating, but compared with a near absence of spontaneous speech, any deficits in repetition are relatively minor.
Some patients with aphasia have few deficits beyond problems with confrontation naming. A commonly used bedside method to test confrontation naming is to ask the patient to name your hand, finger, thumb, knuckle, and cuticle in sequence. This is a rough screen for anomia for high-, medium-, and low-frequency items. Standardized materials such as the Boston Naming Test allow better analysis of naming deficits. Be cautious not to assign too much weight to mild anomia, as it is common in people who are older or less educated.
Reading deficits generally mirror spoken language deficits. Test for the ability to read individual words, short sentences, and brief passages of 100–200 words aloud. Look for both the ability to pronounce the words correctly and for an understanding of content.
Like reading, writing performance usually mirrors spoken language. Patients with nonfluent aphasias write short, agrammatic sentences full of content words, while those with fluent aphasias write long, often incomprehensible sentences with many paraphasic errors.
While the following descriptions of the classic aphasia syndromes are clinically useful, they have several important limitations. First, they are largely a cluster of language deficits associated with specific ischemic strokes. Impairments from aphasia secondary to trauma, tumor, or hemorrhage do not conform to these patterns. Secondly, there are numerous exceptions to even the most basic rules such as left-sided lesions cause aphasia in right-handed people, posterior lesions produce fluent aphasias, and extrasylvian aphasias do not affect repetition. Finally, aphasia may undergo a
Figure 3.1 Left lateral view of the brain demonstrating the rough localizations of the classical cortical aphasias.
semiological transformation over time, resembling global aphasia at initial presentation and changing to Wernicke’s aphasia several days later, for example.
Broca’s aphasia is characterized by nonfluent speech, relatively preserved comprehension, and impaired repetition. In its classical form, spontaneous output is limited to content-rich phrases of one or two words, which are produced with great effort and poor melodic intonation. While patients understand simple sentences and can follow commands, longer, syntactically complex sentences reveal important comprehension limitations. Patients may be able to repeat single words but longer phrases prove more challenging. Accompanying deficits include contralateral arm and face weakness. Broca’s aphasia is traditionally due to infarction of the frontal operculum and adjacent subcortical white matter (Figure 3.1), although many patients with lesions isolated to these areas have fractional syndromes rather than full-blown Broca’s aphasia.5
Patients with Wernicke’s aphasia have fluent speech, often with an increase in the rate and quantity of verbal output to the point where it can be labeled as pressured. While the quantity of output is great, the content is sparse, consisting of excessive semantic and phonemic paraphasic errors, adverbs, adjectives, and grammatical connectors. The prosody, or rhythm and inflection of verbal output, is normal. Comprehension of both spoken and written language is poor. Patients with Wernicke’s aphasia are unable to repeat in a reliable manner, and in many cases are not even able to understand the request to repeat. Patients with Wernicke’s aphasia may be completely unaware of or unconcerned by their deficits. Additional neurological deficits include contralateral hemibody sensory abnormalities and right-sided visual field cuts. The lesion associated with Wernicke’s aphasia is in the left superior temporal gyrus (Figure 3.1) and is often produced by an embolus to the inferior division of the left middle cerebral artery.
Global aphasia is most often caused by a large left middle cerebral artery infarction. Output is nonfluent and agrammatical, and in some cases the patient may be mute. Both comprehension and repetition are poor. Global aphasia is usually accompanied by a severe right hemiparesis and often by forced leftward eye deviation.
Transcortical motor aphasia
The most obvious feature of transcortical motor aphasia is poor initiation of speech, with phrases that may be only one or two words in length. Comprehension is relatively preserved. The difference between spontaneous speech and repetition is quite striking. Some patients with transcortical motor aphasia may be able to repeat long sentences almost verbatim. Others have very little spontaneous speech, but are able to repeat the examiner’s questions (echolalia). The most common lesion location is in the frontal subcortical white matter anterolateral to the left lateral ventricle.6 Transcortical motor aphasia may be due to left anterior cerebral artery infarction, in which case it is accompanied by right foot weakness.
The most obvious feature of spontaneous speech in conduction aphasia is the excessive number of paraphasic errors, most often phonemic ones. Circumlocution is also quite striking and may take the form of conduit d’approche in which the patient
makes successive paraphasic errors that more closely approximate the desired word before finally arriving at the target. Fluency is generally preserved, although it may be slightly reduced as the patient attempts to correct their paraphasic errors or search for the appropriate word. Comprehension is relatively normal, but repetition is very poor. Conduction aphasia may be associated with contralateral homonymous hemianopsia or inferior quadrantanopsia. Conduction aphasia is usually due to lesions of the left supramarginal gyrus, insula, or the underlying white matter (Figure 3.1).
Transcortical sensory aphasia and mixed transcortical aphasia
These aphasias are uncommon in the acute setting. Transcortical sensory aphasia is characterized by fluent output of nonspecific words such as “that” or “things,” poor comprehension, and relatively spared repetition. It is most common in Alzheimer’s disease and semantic dementia (Chapter 4). Patients with mixed transcortical aphasia show poor fluency and comprehension, with relatively preserved repetition. It is uncommon, and is most often due to large frontal or anterior thalamic lesions.1
Subcortical lesions may result in a wide variety of aphasias, which I will not discuss in great detail. Deficits may resemble one of the classical aphasias or may be somewhat nonspecific, or may be restricted to anomia. The most commonly described locations for subcortical aphasia are the thalamus and the striatum.7
Essentially any left hemispheric lesion may produce anomia without other language abnormalities. Anomia may be the only deficit in acute aphasia, or it may be the long-term remnant of a resolving aphasia.
Aphemia is characterized by severe articulatory planning deficits that may mimic nonfluent aphasia.8 In most cases, patients are mute at presentation, and fluency improves over several days. Unlike patients with nonfluent aphasias, comprehension and written language are preserved in aphemia. The capacity to write lengthy, well-constructed sentences that contrast markedly with the sparse spontaneous verbal output is often astounding. Aphemia is caused by any number of left hemispheric lesions, including those involving Broca’s area, the premotor cortex, the motor strip, and the insula. It is usually accompanied by right hemiparesis. As acute mutism resolves, speech is initially slow, effortful, and poorly articulated. Complete recovery may occur over several days to a few weeks. It is important to recognize aphemia, as it usually has a better prognosis than the nonfluent aphasia syndromes with which it is confused.
Determining the cause and treatment of aphasia
Acute aphasia is most commonly due to ischemic or hemorrhage stroke. Other sources of acute aphasia include head trauma, intracranial masses, seizures, and the postictal state. In the acute setting, a comprehensive evaluation of language must assume a secondary priority, as the pathologies that produce acute aphasia are potentially devastating and sometimes reversible. Until proven otherwise, assume that all patients who present with acute-onset aphasia have an ischemic stroke that requires immediate evaluation and treatment (Chapter 21).
Recovery and rehabilitation of acute aphasia
Most recovery from aphasia takes place within the first 3–6 months after symptom onset.9 The traditional notion that recovery beyond 1 year is rare may not necessarily be true.10 Predictors of better prognosis include smaller lesions, younger patient age, and left-handedness. Patients with traumatic lesions tend to have better outcomes than those with ischemic or hemorrhagic ones. Recovery is mediated by the cortex adjacent to the lesion, subcortical structures, and the right hemisphere.11 Both formal (with a speech therapist) and informal (reintegrating the patient back into everyday communication with family, friends, and coworkers) rehabilitation programs are beneficial and should be initiated as soon as possible.12
1. Alexander MP. Aphasia: clinical and anatomical aspects. In: Feinberg TE, Farah MJ, eds. Behavioral Neurology and Neuropsychology. New York: McGraw Hill; 1997; 133–150.
2. Hillis AE. Aphasia: progress in the last quarter of a century. Neurology 2007;69:200–213.
3. Benson DF, Ardila A. Aphasia: a Clinical Perspective. Oxford: Oxford University Press; 1996.
4. Kim JS. Pure dysarthria, isolated facial paresis, or dysarthria-facial paresis syndrome. Stroke 1999;25:1994–1998.
5. Mohr J, Pessin M, Finkelstein S, et al. Broca aphasia: pathological and clinical. Neurology 1978;28:311–324.
6. Freedman M, Alexander MP, Naeser MA. Anatomic basis of transcortical motor aphasia. Neurology 1984;34:409–417.
7. Nadeau SE, Crosson B. Subcortical aphasia. Brain Lang 1997;58:355–402.
8. Ottomeyer C, Reuter B, Jager T, et al. Aphemia: an isolated disorder of speech associated with an ischemic lesion of the left precentral gyrus. J Neurol 2009;256:1166–1168.
9. Kertesz A, McCabe P. Recovery patterns and prognosis in aphasia. Brain 1977;100:1–18.
10. Moss A, Nicholas M. Language rehabilitation in chronic aphasia and time postonset: a review of single-subject data. Stroke 2006;37:3043–3051.
11. Saur D, Lange R, Baumgaertner A, et al. Dynamics of language reorganization after stroke. Brain 2006;129:1371–1384.
12. Robey RR. The efficacy of treatment for aphasic persons: a meta-analysis. Brain Lang 1994;47: 582–608.