Speech and Higher Mental Function Examination — Written Clinical Reasoning

SAQ 1 — Acute Wernicke Aphasia: Classification, the Delirium Trap, and the Stroke Pathway (15 marks, 20 minutes)

Prompt: Describe the classification of this patient's speech disorder, the localising interpretation, the critical distinction from an acute confusional state, the urgent investigations, and the initial management. Justify why this is not delirium and why the timing matters. [1]

Model Answer

Classification and localisation (3 marks): [1]

This patient has a fluent aphasia with impaired comprehension and impaired repetition — a Wernicke (receptive) aphasia. The three axes settle it: the speech is fluent (long phrases, normal melody, effortless), comprehension is impaired (cannot follow two-step commands), and repetition is impaired (cannot echo the loaded sentence). The empty content with paraphasias and neologisms, and her lack of insight (anosognosia for the language deficit), are the hallmarks of a posterior aphasia. A fluent aphasia with poor comprehension and poor repetition localises to the left superior temporal gyrus (Wernicke area), supplied by the inferior division of the left middle cerebral artery. The naming difficulty is non-discriminating — anomia is present in every aphasia. The intact hearing excludes a peripheral auditory cause. The absence of limb signs does not exclude the diagnosis, because Wernicke area lies outside the primary motor and sensory cortex. [1]

Why this is not delirium (4 marks): [1]

The emergency registrar's label of "confused" is the classic and dangerous trap. An acute confusional state (delirium) can mimic a Wernicke aphasia because the speech is incoherent and the patient appears not to understand. The distinguishing features that confirm aphasia and exclude delirium are: [1]

1. Alertness and attention are preserved. She is fully awake and attending — she manages one-step commands and orients to the examination. In delirium, attention is the core impairment: the patient is distractible, fluctuates, and cannot sustain a task. I would confirm attention formally with digit span and months backwards; a preserved digit span supports aphasia, an impaired one supports delirium.
2. The deficit is focal and language-specific. The fluency and melody are normal; only the content is disordered. In delirium the speech is globally disorganised — rambling, tangential, off the point — rather than specifically paraphasic.
3. The onset is sudden and discrete. A vascular event produces a focal deficit in seconds to minutes. Delirium typically fluctuates and develops over hours to days with a precipitating cause (infection, drug withdrawal, metabolic derangement).
4. There are neologisms and paraphasias — language errors of a specific type — rather than the jumbled, incoherent thinking of delirium. [1]

The practical danger of the mislabel is that a patient with an acute dominant-hemisphere stroke is sent for a "confusion workup" (urinalysis, cultures, electrolytes) while the thrombolysis clock runs out. The correct action is to recognise the focal language deficit and activate the stroke pathway. [1]

Urgent investigations (4 marks): [1]

1. Urgent non-contrast CT brain to exclude haemorrhage — the first step before any thrombolysis decision. A left MCA-territory ischaemic stroke may not be visible on early CT, which is expected; the role of the initial CT is to exclude a bleed.
2. CT angiography of the intracranial circulation to identify a large-vessel occlusion (a proximal MCA or M2 branch), which, if found within the thrombectomy window, makes the patient a candidate for mechanical thrombectomy.
3. MRI brain with diffusion-weighted imaging if the CT is negative but the clinical picture is strokelike, to confirm an acute left temporoparietal infarct and exclude a mass or encephalitis.
4. Baseline stroke bloods — glucose (already done, to exclude hypoglycaemia which can mimic aphasia), electrolytes, full blood count, coagulation profile, lipids, HbA1c, and an ECG to look for atrial fibrillation. A normal glucose is critical because hypoglycaemia is a reversible stroke mimic. [1]

Initial management (4 marks): [1]

1. Activate the acute stroke pathway immediately. Because she is within the thrombolysis time window (under 4.5 hours from onset) and has no contraindication, she is assessed for intravenous thrombolysis with alteplase (or tenecteplase per local protocol) once haemorrhage is excluded on CT. The focal, sudden-onset language deficit is a disabling stroke symptom and qualifies for thrombolysis.
2. Thrombectomy assessment. If the CT angiogram shows a proximal left MCA occlusion, she is referred to the neurointervention service for mechanical thrombectomy, which can extend the treatment window to 24 hours for selected patients with a favourable imaging profile (per the DAWN and DEFUSE-3 criteria).
3. Blood pressure management per the stroke protocol — permissive hypertension (typically below 185/110 if thrombolysing) to maintain penumbral perfusion.
4. Early referral to speech and language therapy. The Cochrane review of speech and language therapy for aphasia following stroke shows that therapy improves functional communication, receptive and expressive language compared with no therapy, with higher intensity and dose likely more effective [4]. The referral is made early, even in the acute phase, because early assessment informs the communication strategy for the whole admission.
5. Swallow assessment before any oral intake — aphasia often coexists with dysphagia in an MCA stroke, and aspiration pneumonia is preventable.

SAQ 2 — Interpreting a Cognitive Domain Profile: Alzheimer Disease versus Frontotemporal Dementia (10 marks, 15 minutes)

Prompt: A 69-year-old man is referred to the memory clinic with 18 months of progressive cognitive decline. His Addenbrooke's Cognitive Examination III (ACE-III) returns a total of 76 out of 100, with the following domain subscores: attention 16 out of 20, memory 8 out of 20, fluency 16 out of 20, language 18 out of 20, visuospatial 18 out of 20. Outline how you interpret this domain profile, the most likely dementia subtype, the role of the screening tool, and the next steps in the workup. [1]

Model Answer

Interpreting the domain profile (3 marks): [1]

The ACE-III is a 100-point test scored across five domains of 20 each — attention, memory, fluency, language and visuospatial — and the total of 76 is below the 82 cut-off that suggests dementia [3]. The pattern of domain involvement is the discriminating feature, not the total alone. This profile shows a profound and isolated memory deficit (8 out of 20) with relative preservation of language (18), visuospatial (18) and fluency (16), and only mild attentional difficulty. A memory-dominant profile with sparing of language and visuospatial function in the early stage is the classical signature of Alzheimer disease, in which the earliest pathology (entorhinal cortex and hippocampal) produces an episodic memory deficit — impaired learning and delayed recall — while the neocortical language and visuospatial functions are relatively preserved until later. By contrast, frontotemporal dementia typically presents with a fluency-dominant or language-dominant profile (a progressive non-fluent aphasia or semantic dementia) with relative memory sparing early on; vascular cognitive impairment often shows an attention- and processing-speed-dominant profile; and dementia with Lewy bodies shows prominent visuospatial impairment. So this profile points to Alzheimer disease.

The role of the screening tool (2 marks): [1]

The ACE-III was chosen appropriately here because its five-domain breakdown allows a pattern-of-deficit analysis that the single-score MMSE cannot provide [1]. The MMSE is weighted toward language and orientation, is insensitive to mild cognitive impairment, and gives a single number that masks the domain pattern. The MoCA, while more sensitive than the MMSE to mild impairment and better for subcortical and Parkinson-related cognitive profiles, does not give the same five-domain resolution as the ACE-III [2]. So in a memory clinic where the question is "which dementia subtype?", the ACE-III is the correct extended screen. No screen is diagnostic — all are triggers for formal neuropsychological assessment and structural imaging.

Next steps in the workup (5 marks): [1]

1. Collateral history from a reliable informant. The timeline, the first symptoms, the functional impact and the psychiatric features (apathy, depression, behavioural change) are best established from the family. An insidious, progressive episodic-memory deficit with functional decline supports Alzheimer disease.
2. Formal neuropsychological assessment to confirm and quantify the domain profile, characterise the memory deficit (encoding versus retrieval; cueing response — a patient with Alzheimer disease does not benefit from cueing, whereas one with depression or subcortical disease may), and exclude pseudodementia from depression.
3. Structural imaging — MRI brain. Look for hippocampal and medial temporal atrophy (the structural correlate of Alzheimer disease) and exclude a mass, a subdural haematoma, or a vascular burden that would change the diagnosis. A CT is acceptable if MRI is contraindicated.
4. Mood screen and reversible-cause bloods. A Geriatric Depression Scale or PHQ-9 to exclude depression (which can mimic dementia and which itself impairs memory), and a screen for reversible causes — thyroid function, vitamin B12, folate, calcium, syphilis serology if risk factors, and HIV if appropriate.
5. Aetiological management and follow-up. If Alzheimer disease is confirmed, the discussion covers a cholinesterase inhibitor (donepezil, rivastigmine or galantamine) for mild-to-moderate disease, an NMDA antagonist (memantine) for moderate-to-severe disease, advance care planning, carer support, driving assessment, and regular review. The domain profile guides both the diagnosis and the prognosis: a memory-dominant profile has a different trajectory and management than a fluency-dominant one. [1]