The Undifferentiated Altered Mental Status — Written Clinical Reasoning

SAQ 1 — Integrated Management of the Acutely Confused Elderly Patient with Multifactorial Delirium (20 marks, 30 minutes)

Prompt: Outline your integrated management of Mr Chen's acute confusional state, addressing: (a) the ABCDE resuscitation and the immediately reversible causes (the GHOST check); (b) the application of the CAM diagnostic algorithm and the multifactorial model of delirium; (c) the management of the sepsis (pneumonia and UTI) with the appropriate antibiotic regimen in the context of his CKD; (d) the controlled correction of the severe hyponatraemia and the danger of over-rapid correction; (e) the management of the chronic subdural haematoma in a patient on apixaban; (f) the drug review and the medications you would withhold and why; and (g) the common exam trap in this patient. [1]

Model Answer

(a) ABCDE resuscitation and the GHOST check (3 marks): [1]

The approach begins with the ABCDE, treating life-threats in the order they are found. Airway: patent — he is talking, though incoherently. No intervention needed beyond positioning. Breathing: respiratory rate 22, SpO2 94 per cent on room air — give 2 L of oxygen via nasal speculum to target 94 to 98 per cent (he is not a known CO2 retainer). Circulation: heart rate 96 in atrial fibrillation, blood pressure 104/64, capillary refill 3 seconds — he is mildly hypotensive and peripherally shut down; two large-bore cannulae, send the bloods (including the cultures), and a cautious fluid bolus (250 mL of balanced crystalloid, reassess — the hypotension may be septic or pre-renal, and the CKD and the hyponatraemia caution against over-resuscitation). Disability: GCS 13, pupils equal and reactive, no focal neurology; finger-prick glucose 6.2 (normal — no hypoglycaemia, but the check is mandatory). Exposure: temperature 38.3 (fever), and a full examination for the source of the sepsis (the chest and the urine are already identified as positive). [1]

The GHOST reversible-cause check: Glucose 6.2 (normal, checked). Oxygen — give it to target 94 to 98 per cent. Sodium 124 — severe, needs controlled correction (addressed below). Thiamine — give 500 mg IV empirically (he is elderly, possibly nutritionally depleted, and the downside is negligible). Hypertension/hypotension — the blood pressure of 104/64 is low for a baseline hypertensive; address with the cautious fluid and treat the sepsis. [1]

(b) The CAM diagnostic algorithm and the multifactorial model (3 marks): [1]

Mr Chen meets the CAM criteria for delirium: Feature 1 (acute onset and fluctuating course — acute over four days, from a previously independent baseline), Feature 2 (inattention — he cannot do the months of the year backwards), Feature 3 (disorganised thinking — implied by the disorientation), and Feature 4 (altered level of consciousness — drowsy, GCS 13) [1]. The CAM is positive.

The multifactorial model, as described by Inouye, frames delirium as the interaction between a vulnerable patient (predisposing factors — his age, the baseline vascular dementia, the CKD, the polypharmacy) and multiple acute insults (precipitating factors — the pneumonia, the UTI, the hyponatraemia, the AKI, and the subdural) [2]. The implication for management is that there is rarely a single cause; there are several, and the workup must address all of them. The dementia is the predisposing vulnerability; the acute change is the emergency; the precipitants are the targets of the treatment.

(c) Sepsis management — pneumonia and UTI with the antibiotic regimen in CKD (3 marks): [1]

He has sepsis from two sources: a community-acquired pneumonia (right lower lobe consolidation, fever, CRP 96, WCC 16.2) and a urinary tract infection (positive nitrites and leucocytes). The management is the Sepsis Six bundle — oxygen (given), blood cultures (sent), intravenous antibiotics within one hour, intravenous fluids (cautiously, given the CKD and the hyponatraemia), lactate (measured), and hourly urine output (catheterise if not already). [1]

The antibiotic regimen, guided by the local protocol and dose-adjusted for the CKD (creatinine 195, eGFR approximately 30): ceftriaxone 1 g IV daily (no adjustment needed — it is renally excreted but the once-daily dosing is safe at this eGFR) plus azithromycin 500 mg orally daily for three days (for the atypical cover of the community-acquired pneumonia). Add a gram-negative urinary cover if the local guideline requires it (e.g., trimethoprim or nitrofurantoin for the UTI once the culture is available — but these may be contraindicated or dose-limited by the CKD, so I would discuss with microbiology and use the ceftriaxone as the empiric cover for both sources in the acute phase). Paracetamol 1 g IV or orally for the fever. I reassess at 24 to 48 hours with the clinical response, the inflammatory markers, and the culture results. [1]

(d) The controlled correction of the severe hyponatraemia (3 marks): [1]

The sodium of 124 is severe and is contributing to the confusion and the drowsiness. The correction must be controlled — no more than 8 to 10 mmol per litre in 24 hours — because over-rapid correction risks osmotic demyelination syndrome, the devastating pontine and extrapontine injury first described by Sterns in 1986 [5]. The risk factors for osmotic demyelination include the severity of the hyponatraemia, the chronicity, the alcohol use, the malnutrition, and the hypokalaemia — and this patient is elderly and possibly nutritionally depleted.

The approach is the treatment of the underlying cause (the sepsis, the dehydration, the SIADH-like state that can accompany the pneumonia) with cautious normal saline (250 to 500 mL over 4 to 6 hours, reassess) and the serial monitoring of the sodium every 4 to 6 hours. I do not give hypertonic saline unless he seizes or his conscious level falls precipitously — the risk of overcorrection is highest in the patient who is rapidly fluid-resuscitated and whose sodium rises faster than the 24-hour limit. If the sodium rises faster than 10 mmol in 24 hours, I stop the normal saline, give desmopressin and free water to re-lower it, and consult nephrology. [1]

(e) The management of the chronic subdural on apixaban (3 marks): [1]

The CT head shows a small chronic subdural haematoma on the left (hypodense, no mass effect) — this may be contributing to the confusion, though the sepsis and the hyponatraemia are more likely the primary drivers. The management has four steps: [1]

1. Reverse the apixaban. Discuss with haematology — the options are andexanet alfa (a recombinant modified factor Xa that decoys the apixaban) or four-factor prothrombin complex concentrate (PCC). The choice depends on the severity, the local protocol, and the availability. The reversal reduces the risk of further expansion.
2. Neurosurgical consultation. A small chronic subdural with no mass effect may be managed conservatively with serial imaging, but the decision is the neurosurgeon's, and the threshold for surgical evacuation (burr-hole drainage) is lower if the GCS falls or a focal deficit emerges.
3. Hourly GCS monitoring and serial CT. If the GCS drops by 2 points or a new focal deficit appears, repeat the CT urgently and escalate.
4. The fall assessment. How did he fall? Was the fall the cause of the subdural, or was the confusion and the unsteadiness the cause of the fall? Address the underlying contributors (the sepsis, the hyponatraemia, the polypharmacy) and assess for future fall prevention. [1]

(f) The drug review — what to withhold and why (3 marks): [1]

The drug chart is the highest-yield document in the unexplained delirium, and in this patient several drugs must be withheld: [1]

• Gliclazide — withhold immediately. He has an acute kidney injury (creatinine risen from 150 to 195) and the sulfonylurea persists for hours; the risk of prolonged, refractory hypoglycaemia is high. Manage the glucose with a variable-rate intravenous insulin infusion, checking the glucose hourly, and transition back to subcutaneous insulin once the AKI resolves and the glucose is stable.
• Apixaban — withhold immediately, for the subdural haematoma (addressed above).
• Antihypertensives — withhold. His blood pressure is 104/64, which is low for a baseline hypertensive; the antihypertensives will worsen the hypotension, the cerebral hypoperfusion, and the confusion. Restart once the blood pressure is stable and the acute illness is resolving.
• Insulin glargine — continue but adjust the dose for the AKI and the reduced intake. The long-acting basal insulin may need to be reduced; monitor the glucose closely.
• Any other nephrotoxic or sedating drugs — review and withhold as indicated. [1]

(g) The common exam trap (2 marks): [1]

The trap in this patient is attributing the acute confusion to the underlying vascular dementia without diagnosing and treating the delirium. The dementia is the predisposing factor (the reduced cognitive reserve that makes the elderly patient more vulnerable), not the explanation for the acute change. The CAM is positive, the precipitants are multiple and identifiable (the sepsis, the hyponatraemia, the AKI, the subdural), and the management is the treatment of the precipitants alongside the non-pharmacological delirium bundle. The registrar who documents "confusion — due to dementia" and does not work the precipitants has failed the patient. [1]

The second trap is correcting the hyponatraemia too rapidly. The sodium of 124 is severe and tempting to correct fast, but the rate must be controlled at 8 to 10 mmol per 24 hours. The over-rapid correction is irreversible and catastrophic — the osmotic demyelination syndrome produces a quadriparetic, dysarthric, often comatose patient who does not recover. [1]

SAQ 2 — The DIMTOP Approach and the Investigation Strategy (10 marks)

Prompt: A junior doctor asks you to explain: (a) the DIMTOP mnemonic and how it organises the differential diagnosis of the acutely confused patient; (b) when a CT head is indicated (and when it is not) in the workup of the confused patient; and (c) when to request an EEG in the confused patient, and what it is looking for. [1]

Model Answer

(a) The DIMTOP mnemonic and the differential organisation (4 marks): [1]

DIMTOP is a framework that ensures no category of cause is missed in the workup of the acutely confused patient. Each letter corresponds to a category, and within each category there are high-yield specific causes: [1]

• D — Drugs. The single most common iatrogenic cause of acute confusion. The high-yield culprits are anticholinergics (oxybutynin, tricyclics), opioids (respiratory depression and sedation), benzodiazepines (GABA potentiation), anticonvulsants (dose-dependent cerebellar and cortical effects), corticosteroids (steroid-induced psychosis), digoxin (toxicity in renal failure), lithium (narrow therapeutic index), dopaminergics (hallucinations in Parkinson), and drug withdrawal (alcohol, benzodiazepine, opioid). The drug chart is the highest-yield document; review every drug, every dose, and every recent change.
• I — Infection. The commonest single cause of acute confusion in the elderly hospitalised patient. The sources, in order of frequency: urinary tract, chest, skin and soft tissue, intra-abdominal, central nervous system (meningitis, encephalitis), and line-related. The infection screen is blood cultures, urine MC&S, chest X-ray, and CRP. The antibiotic is given within one hour for suspected sepsis.
• M — Metabolic. The most easily missed and most easily treated category. The high-yield derangements: hypoglycaemia (finger-prick glucose), hyponatraemia (U&E), hypercalcaemia (calcium), uraemia (creatinine), hepatic encephalopathy (LFTs, asterixis), thyroid disease (TSH), and acid-base disturbance (blood gas). The bedside panel (glucose, VBG, U&E, calcium, TSH) catches the majority.
• T — Trauma. Dominated by subdural haematoma — the diagnosis most likely to be missed in the elderly patient on anticoagulation with a forgotten or unwitnessed fall. Traumatic brain injury itself causes confusion (the concussed patient with post-traumatic amnesia). The CT head is indicated for focal neurology, trauma, anticoagulation, deteriorating GCS, or a falling GCS trend.
• O — Oxygen (and carbon dioxide). Hypoxia (any cause of low oxygen saturation) and hypercapnia (CO2 narcosis — especially the COPD patient on excessive oxygen). The blood gas catches both. The pulse oximeter is the first measurement.
• P — Psychiatric. Primary psychiatric disorders (depression, psychosis, mania) are diagnoses of exclusion in the acutely confused patient. The cardinal rule: a first episode of acute psychosis or mania in a patient over 40 has a medical cause until proven otherwise. [1]

The framework is applied systematically — the registrar works through each category rather than anchoring on the first plausible cause. In Mr Chen, the categories that are active are I (pneumonia and UTI), M (hyponatraemia and AKI), T (the chronic subdural), and D (the gliclazide and the apixaban). The multifactorial model demands that all active categories are addressed. [1]

(b) When the CT head is indicated (and when it is not) (3 marks): [1]

The CT head is not a blanket reflex in every confused patient. It is indicated for: [1]

• Focal neurology on examination — an asymmetry of power, tone, reflexes, or the plantar response.
• Head trauma — any history of a fall or a head injury, especially in the elderly and the anticoagulated.
• Anticoagulation — warfarin, a DOAC, or antiplatelet therapy, because the threshold for an intracranial bleed is lower.
• Deep or fluctuating coma (GCS less than 12), progressive deterioration of the conscious level, or a failing GCS trend.
• Seizures — a first seizure or a new-onset status epilepticus.
• Suspected raised intracranial pressure — papilloedema, vomiting, headache, or the Cushing response.
• Before a lumbar puncture — to exclude a mass lesion that could precipitate herniation. [1]

The CT is not indicated for the alert, non-focal, non-traumatised patient with a clear metabolic or infectious cause for their confusion (the hypoglycaemic diabetic, the hyponatraemic patient, the septic patient with a clear source) — the metabolic derangement or the infection is the cause, and the CT adds nothing. The registrar who scans every confused patient has not understood the differential and is substituting imaging for clinical reasoning. In Mr Chen, the CT is indicated by the anticoagulation (apixaban) and the new confusion in the elderly, and it has identified the subdural — an appropriate and informative investigation. [1]

(c) The EEG — when to request it and what it is looking for (3 marks): [1]

The EEG is indicated for the suspected non-convulsive status epilepticus (NCSE) and for the support of a metabolic encephalopathy diagnosis when the standard workup is unrevealing. The clinical scenarios: [1]

• The patient who fails to recover from a convulsive seizure — the post-ictal state that does not resolve within 30 to 60 minutes raises the possibility of ongoing electrographic seizure activity without motor manifestations.
• The patient with a known seizure disorder who presents confused — the confusion may be the non-convulsive seizure rather than a post-ictal state or a drug effect.
• The patient with unexplained persistent confusion despite a negative metabolic, infectious, toxic, and structural workup — the EEG may reveal NCSE or a triphasic-wave pattern (the classic but non-specific EEG of metabolic encephalopathy, especially hepatic and uraemic). [1]

The EEG is looking for epileptiform activity (spikes, sharp waves, rhythmic activity that meets the diagnostic criteria for NCSE) or for the diffuse slowing and the triphasic waves that support a metabolic encephalopathy. The limitation: the standard 30-minute EEG may miss intermittent NCSE, and a continuous EEG (cEEG) over 24 to 48 hours is preferred in the ICU and the high-suspicion patient. The teaching point: the EEG is a selective test, not a routine one — request it for the clinical indications, not as a blanket screen. In Mr Chen, the EEG is not currently indicated — his confusion is explained by the multifactorial delirium (the sepsis, the hyponatraemia, the AKI, the subdural), and the EEG would add nothing at this stage. If the confusion persists or fluctuates despite the correction of these causes, the EEG becomes indicated. [1]