Clinical Analysis: 72M with Pontine + Lentiform Nucleus Stroke, Recurrent Vomiting, Drooling, Hiccups, Na 116, TLC 24,000

1. Localizing the Symptoms to the Stroke

Pons

Lentiform Nucleus (Putamen + Globus Pallidus)

• Typically causes contralateral hemiparesis, hemibody sensory loss, or dysarthria
• The lentiform nucleus lesion likely explains the vascular risk factor pattern (hypertension commonly causes lacunar/small vessel disease here) and may contribute to dysarthria/dysphagia via corticobulbar involvement
• Combined pons + basal ganglia involvement suggests small vessel disease / lacunar strokes from hypertension and diabetes

2. The Hyponatremia (Na = 116 mEq/L)

Severity Classification

Critical Clinical Dilemma in This Patient

1. Na 116 - falls in the < 120 range where ODS risk is very high with rapid correction
2. Pre-existing pontine lesion - the pons is the most commonly affected site in ODS. This patient's pons is already damaged - rapid correction could cause catastrophic additional injury to an already vulnerable brainstem
3. Diabetes - an independent risk factor for ODS
4. Age 72 - likely has some degree of malnutrition/metabolic vulnerability

"ODS is a life-threatening complication of overly rapid correction of severe chronic hyponatremia (< 120 mEq/L for 2 or more days). The pons is most commonly affected... Develops 2-6 days after correction."

• Symptom to Diagnosis, 4th Ed.

"50% of rapidly corrected patients develop osmotic demyelination if their serum sodium levels at baseline were lower than 105 mmol/L."

• Bradley and Daroff's Neurology in Clinical Practice

Correction Targets

Hyponatremia Likely Etiology

• SIADH secondary to CNS pathology (stroke) is the most likely cause - strokes, especially involving the hypothalamus, pons, and brainstem, are classic triggers of SIADH via inappropriate ADH secretion
• Vomiting for 4 days could also be contributing via volume depletion, but the neurological picture favors SIADH
• Risk factors for SIADH in this patient: stroke, diabetes medications (e.g., sulfonylureas), advanced age

3. Leukocytosis (TLC 24,000, Neutrophil Dominant)

A. Aspiration Pneumonia (Most Likely)

• The patient has dysphagia (drooling, impaired swallowing from pontine/bulbar involvement), recurrent vomiting, and a recent stroke
• This is the classic setup for aspiration pneumonia
• Chest X-ray / CT chest is mandatory
• Right lower lobe / right upper lobe predominance typical
• Management: Broad-spectrum antibiotics (e.g., piperacillin-tazobactam or ampicillin-sulbactam to cover anaerobes + gram negatives), airway protection, NPO with NG tube feeding

B. Stress Leukocytosis

• Severe stroke and severe metabolic derangement (Na 116) can cause physiological stress leukocytosis even without infection
• However, with 24,000 and neutrophil dominance in the context of dysphagia, infection should be the presumed cause until proven otherwise

C. Urinary Tract Infection

• Common in elderly, diabetic, post-stroke (catheterized) patients
• Urine culture and urinalysis should be sent

4. Problem Summary and Priority List

5. Key Red Flag: ODS in a Pre-damaged Pons

• ODS will preferentially demyelinate the pons (the most vulnerable white matter site)
• This will add new demyelinating injury on top of an ischemic scar
• Clinically: worsening quadriparesis, locked-in syndrome, dysarthria, dysphagia worsening in 2-6 days after correction

6. Investigations to Order

• Urgent: Serum sodium every 2h, serum potassium (hypokalemia worsens ODS risk), serum osmolality, urine sodium + osmolality (to confirm SIADH)
• Imaging: MRI brain with DWI to confirm stroke sites + baseline pontine status; consider repeat if neurological deterioration
• Infection workup: CXR, blood cultures x2, urine R/E + culture, sputum culture
• Metabolic: RBS, HbA1c, LFTs, RFTs, TSH (hypothyroidism is a cause of SIADH and hyponatremia)
• ABG: for ventilatory status given aspiration risk

• Symptom to Diagnosis: An Evidence Based Guide, 4th Ed., Table 24-2/24-3
• Bradley and Daroff's Neurology in Clinical Practice
• National Kidney Foundation Primer on Kidney Diseases, 8th Ed.

Clinical Update Review: Day 2+ Progress and Prescription Concerns

What Has Improved

• Na: 116 → 126 mEq/L (+10 mEq/L) - this is right at the upper safety limit for 24h
• TLC: 24,000 → 19,000 - partial response to antibiotics, consistent with evolving aspiration pneumonia
• No hypercapnia on ABG, renal function normal - these are reassuring

1. The Antibiotic Regimen - Significant Concern

2. CRITICAL Drug Interaction: Azithromycin + Amiodarone

• Amiodarone blocks multiple ion channels (Na, K, Ca) - class III antiarrhythmic with well-known QT prolongation
• Azithromycin blocks hERG (IKr) potassium channels - causes QT prolongation even as a "safe" macrolide

• Age 72, male
• Active AF (already structurally abnormal conduction)
• Possible hypokalemia/hypomagnesemia (recent severe hyponatremia + correction)
• Receiving amiodarone infusion (high loading doses)

1. Check QTc on ECG immediately - if QTc > 500 ms, azithromycin should be stopped
2. Replace azithromycin with a non-QT-prolonging agent (e.g., doxycycline for atypical coverage, or simply rely on meropenem if atypical coverage is not strongly indicated)
3. Monitor serum K+ and Mg2+ - replace aggressively (K+ > 4.0 mEq/L, Mg2+ > 0.8 mmol/L)
4. Daily ECG monitoring while on amiodarone

3. Polypharmacy Concern: Bisoprolol + Ivabradine + Amiodarone

• Ivabradine is not indicated in AF. Ivabradine works by blocking the If (funny) current in the SA node - it only slows sinus rhythm. In AF, there is no sinus node activity driving the ventricular rate; the rate is determined by AV nodal conduction. Ivabradine has no proven rate-controlling benefit in AF and its use here is pharmacologically inappropriate.
• The combination of amiodarone + bisoprolol already carries a risk of profound bradycardia and AV block (additive AV node depression). Adding ivabradine on top - even if ineffective in AF - adds unnecessary polypharmacy and monitoring burden.
• Ranolazine 500 mg BD: While ranolazine has some antiarrhythmic properties and modest QT effects at standard doses, combining it with amiodarone adds further QT risk. Ranolazine also inhibits CYP3A4 to a degree, potentially raising amiodarone levels.

• Stop ivabradine - not indicated in AF, no mechanism of action applies
• Bisoprolol + amiodarone is the appropriate combination for AF rate/rhythm control
• Re-evaluate need for ranolazine - if it was added for angina/CAD, note its interaction profile with amiodarone and monitor QTc closely
• Ensure continuous cardiac monitoring (telemetry) given triple QT-prolonging potential

4. Persistent Disorientation Despite Sodium Correction

On the Addition of Modafinil 100 mg OD

• In a delirious patient (disoriented, not simply somnolent), modafinil can worsen agitation and delirium
• Modafinil can also increase blood pressure - relevant in a hypertensive patient
• It is a CYP3A4 inducer - may reduce amiodarone levels unpredictably
• Better first step: Identify and treat the cause of disorientation before adding a stimulant

5. Priority Action List

Summary

1. Ivabradine for Sinus Tachycardia - Partially Valid, But Caveat

• Severe bradycardia - triple nodal suppression
• Symptomatic pauses, especially dangerous in an elderly post-stroke patient with a compromised brainstem

• Active infection (WBC 19,000, consolidation) - most likely cause
• Pain/discomfort
• Fever
• Hypovolemia from vomiting

2. Modafinil for Drowsiness - Rationale Understood, But Timing Is Wrong

• Active sepsis (aspiration pneumonia, WBC 19,000) - septic encephalopathy causes drowsiness
• Na corrected from 116 to 126 - brain still re-equilibrating
• Multiple sedating drugs on board (baclofen, amiodarone has CNS effects)
• ODS not yet ruled out

3. Clindamycin for Anaerobes + Staphylococcus - Clinically Valid, But Redundant Here

• If no MRSA risk: stop clindamycin (meropenem covers anaerobes + MSSA adequately)
• If MRSA risk present: replace clindamycin with vancomycin or linezolid
• Obtain sputum culture and sensitivity to guide de-escalation

Overarching Safety Concern That Remains Unchanged

1. Stopping Clindamycin + Azithromycin ✓ Correct

2. Adding Doxycycline ✓ Appropriate

3. Adding Vancomycin - Conditional

• Check renal function before dosing - vancomycin is nephrotoxic, and while creatinine is currently normal, this patient is elderly (72), diabetic, and septic. Monitor renal function daily and trough levels (target AUC-guided dosing 400-600 mg·h/L, or trough 15-20 mg/L if AUC monitoring not available)
• Does he actually have MRSA risk factors? Healthcare exposure, prior hospitalization, known MRSA, cavitation, or failure to respond to beta-lactam therapy? If none of these apply, meropenem + doxycycline alone may be sufficient. Add vancomycin if cultures return MRSA or if there is clinical deterioration
• If concerned about nephrotoxicity with meropenem + vancomycin together, linezolid 600 mg BD is an alternative with better lung penetration (and avoids vancomycin monitoring complexity)

4. LMWH 0.6 mg SC OD - Important Concerns

1. VTE prophylaxis (immobile post-stroke patient) - appropriate, use 40 mg OD
2. Bridge anticoagulation for AF - if the plan is to anticoagulate for stroke prevention in AF, LMWH is only a bridge. He will need long-term oral anticoagulation (DOAC or warfarin) - this should be planned now
3. Acute AF with hemodynamic instability - if AF is causing hemodynamic compromise, the primary treatment is rate/rhythm control (already underway with amiodarone), not heparin alone

• CHA₂DS₂-VASc score: Age ≥75 (1) + Hypertension (1) + Diabetes (1) + Stroke (2) = minimum score of 5 - very high thromboembolic risk, anticoagulation is strongly indicated
• However, the timing after ischemic stroke matters. Conventional guidance: delay full anticoagulation for 2-4 weeks post-ischemic stroke (depending on infarct size) to reduce hemorrhagic transformation risk. His stroke was 2 months ago, so full anticoagulation is now appropriate from a timing standpoint
• Transition to a DOAC (apixaban or rivaroxaban) once clinically stable - superior to warfarin in AF with fewer intracranial bleeds

5. Stopping Baclofen Since Hiccups Are Controlled - Yes, Taper, Don't Abruptly Stop

• Seizures
• Hallucinations, confusion, agitation
• Hyperthermia
• In severe cases: rhabdomyolysis, multiorgan failure

• Taper over 1-2 weeks: reduce to 5 mg OD for 5-7 days, then 2.5 mg OD for 3-5 days, then stop
• Monitor for return of hiccups during taper - if they recur, hold at the current dose and reassess
• If hiccups remain controlled throughout taper, safe to discontinue

Summary of Decisions

Enoxaparin for AF Stroke Prevention - Is This the Right Approach?

1. Correct Dose for This Patient

• Weight 65 kg → 65 mg (0.65 mL) SC BD
• Not OD at 0.6 mg - that is neither a standard prophylactic nor treatment dose

2. The Fundamental Problem: LMWH Is Not the Right Long-Term Strategy for AF

3. Critical Warfarin-Amiodarone Interaction (If Warfarin Is Considered)

• INR must be checked weekly when both are co-administered
• Warfarin dose typically needs to be reduced by 30-50%
• This interaction alone makes a DOAC strongly preferable in this patient

4. Timing - Is It Safe to Anticoagulate Now?

5. CHA₂DS₂-VASc Score Confirmation

Recommended Plan

1. Start apixaban 5 mg BD as the definitive long-term anticoagulant - once tolerating oral/NG feeds reliably
2. Use enoxaparin 65 mg SC BD as a short-term bridge until apixaban is established and the patient is haemodynamically stable (usually 48-72 hours overlap is not needed with DOACs - can start apixaban directly)
3. Do not use warfarin given the amiodarone co-administration and monitoring complexity
4. Confirm no haemorrhagic transformation on imaging before initiating
5. Monitor renal function every 48-72h - vancomycin + enoxaparin together in an elderly patient is a nephrotoxicity risk combination

Apixaban via Ryle's Tube - Yes, But With Conditions

1. Can Apixaban Be Crushed and Given via NG Tube?

• Crush the tablet, suspend in 60 mL of water, and administer immediately via NG tube
• Flush with another 60 mL water after administration
• Bioavailability is preserved - pharmacokinetic studies confirm absorption is not significantly altered by crushing
• Do not suspend in enteral feeds - flush tube, give drug in water, then flush again before resuming feeds

2. Should You Start It Straight Away - Right Now?

A. Confirm No Haemorrhagic Transformation on Imaging

• Get a CT brain (non-contrast) or MRI today before first dose
• If clean - start apixaban
• If haemorrhagic transformation present - delay and discuss with neurology

B. Confirm NG Tube Position

3. Stop Enoxaparin When Starting Apixaban

• Give the last enoxaparin dose, then start apixaban at the time the next enoxaparin dose would have been due
• No gap, no overlap - this avoids double anticoagulation and bleeding risk

4. Dose Confirmation for This Patient

5. Watch Points After Starting

Bottom Line

MRI Result: Chronic Infarcts Only - What This Means

• The lesions are in the subacute-to-chronic phase - no active ischaemia, no haemorrhagic transformation, no fresh DWI restriction
• The blood-brain barrier has largely restored at the infarct sites
• No contraindication to anticoagulation from a haemorrhagic standpoint

Decision: Start Apixaban Now ✓

Administration:

1. Pause enteral feed
2. Flush NG tube with 30 mL water
3. Crush apixaban 5 mg tablet, suspend in 60 mL water, administer immediately
4. Flush with another 30-60 mL water
5. Resume feeds after 15-20 minutes

Timing with enoxaparin:

• Give last enoxaparin dose now if it is due
• Start apixaban at the time the next enoxaparin dose would have been due (i.e., 12 hours later if BD dosing)
• No overlap

What the MRI Also Tells You

Revised Problem Summary at This Point

Situation: Prolonged QT + HR 50-60 + Frequent PVCs

Immediate Actions (Next 30 Minutes)

Step 1 - Quantify the QTc

• QTc < 500 ms: High vigilance, electrolyte correction, drug review
• QTc 500-550 ms: Stop all QT-prolonging drugs, correct electrolytes urgently
• QTc > 550 ms: Emergency. Stop all QT-prolonging agents immediately, IV magnesium now, consider temporary pacing

Step 2 - Stop All QT-Prolonging Drugs Immediately

Step 3 - IV Magnesium Sulphate (First-Line Treatment)

• MgSO₄ 2g IV over 5-10 minutes (for active TdP or as prophylaxis when QTc is critically prolonged)
• Follow with MgSO₄ infusion 3-4g over 4 hours if PVCs persist
• Magnesium suppresses the triggered activity that initiates TdP - it works even when serum Mg is normal

Step 4 - Correct Electrolytes Urgently

Step 5 - The Bradycardia Problem

1. Bradycardia itself prolongs the QT interval - the slower the heart, the more time for abnormal repolarization
2. PVCs at slow rates are more likely to fall on the T-wave (R-on-T) → triggers TdP

Step 6 - If TdP Occurs

Step 7 - Monitoring

• Continuous telemetry - watch for R-on-T PVCs, runs of TdP
• 12-lead ECG every 2-4 hours until QTc trending down
• QTc should begin to fall once QT-prolonging drugs are stopped and electrolytes corrected
• Target QTc < 500 ms before considering any further antiarrhythmic

What About AF Recurrence?

• AF may recur - accept this risk for now; the VT/TdP risk outweighs AF
• If AF recurs and needs rate control: bisoprolol alone at low dose (2.5 mg) is safer than amiodarone
• For rhythm control once stable: flecainide or sotalol are contraindicated (both prolong QT); consider dronedarone in the longer term only after QT normalises and cardiac function is assessed
• Apixaban is already on board - AF recurrence is covered from a thromboembolic standpoint

Summary - Priority Sequence Right Now

What Is Causing the Hypotension?

Noradrenaline - Is It the Right Vasopressor Here?

• HR 50-60 (already poor cardiac output)
• Frequent PVCs (myocardial irritability)
• Prolonged QT

Immediate Priorities Alongside Noradrenaline

1. Address Bradycardia - This Is Now Urgent

• Atropine 0.5-1 mg IV - first line for symptomatic bradycardia
• If no response: Temporary transvenous pacing - this simultaneously fixes the bradycardia AND shortens the QT (overdrive pacing at 80-90 bpm is therapeutic for TdP risk)
• This is now the most important intervention - fixing the heart rate will do more for the BP than vasopressors alone

2. Stop Amiodarone Infusion Now

3. Fluid Resuscitation

• Give a 250-500 mL crystalloid bolus (normal saline or Plasmalyte) to assess fluid responsiveness
• In septic shock with no haemorrhagic transformation on MRI, cautious fluid challenge is safe
• Raises preload, improves cardiac output

4. Check Lactate Now

• Serum lactate tells you whether hypotension is causing tissue hypoperfusion
• Lactate > 2 mmol/L = tissue dysoxia = true shock requiring urgent escalation
• Lactate > 4 mmol/L = severe shock, consider ICU escalation immediately

The Full Shock Picture Right Now

What to Have Ready

• Crash cart at bedside - QT prolonged + bradycardia + PVCs + hypotension is a VF arrest waiting to happen
• Temporary pacing setup - ideally transvenous; transcutaneous pacing as emergency bridge
• Defibrillator charged - if TdP degenerates to VF, immediate defibrillation is the only intervention

One Question to Answer Right Now

• If yes, and lactate is < 2: septic shock is the dominant mechanism, continue noradrenaline, treat infection aggressively
• If no, or lactate > 4: cardiogenic component is dominant - temporary pacing is the priority, consider adding dobutamine (inotrope) alongside noradrenaline

What Just Happened - Root Cause Analysis

• Amiodarone + Azithromycin (now stopped, but amiodarone half-life 40-55 days - still active in the body)
• Ranolazine (QT prolongation)
• Bradycardia not treated early enough
• Likely hypokalaemia/hypomagnesaemia from vomiting + diuretic effect

Immediate Post-Cardioversion Priorities

1. IV Magnesium Sulphate - Give Now If Not Already Done

• MgSO₄ 2g IV over 10 minutes immediately post-cardioversion
• Follow with MgSO₄ 4-8g over next 24h as infusion
• This is the most important pharmacological step to prevent TdP recurrence
• Works regardless of serum magnesium level

2. Electrolytes - Correct Aggressively

3. Increase Heart Rate - Critical

Drug Review Post-VT - What Stays, What Goes

Longer-Term Antiarrhythmic Strategy

• Digoxin - rate control in AF, does not prolong QT, safe in bradycardia-prone patients; caution with renal function
• Diltiazem (calcium channel blocker) - rate control in AF if no significant LV dysfunction

• Requires formal cardiology/electrophysiology review
• No antiarrhythmic is safe to restart until QTc is consistently < 450 ms and electrolytes are stable

Monitoring Requirements Right Now

The Bigger Picture

What the Document Contains

Why Rate Control Is Failing

• Digoxin controls rate primarily via vagal tone enhancement - it works best at rest and is significantly less effective during sympathetic activation (which this patient has in abundance - active sepsis, pneumonia, fever, noradrenaline infusion)
• At 0.125 mg, this is a low/maintenance dose - may not even be at therapeutic level
• Digoxin level should be checked (therapeutic 0.8-2.0 ng/mL; toxicity risk especially in elderly)
• Digoxin alone or with bisoprolol is often insufficient in high-sympathetic states

• You held bisoprolol during the QT/bradycardia event - is it back on board now?
• If recently restarted, it may not yet be at full effect
• Even if running, 5 mg may be insufficient with the sympathetic drive from sepsis

Distinguishing AF vs Atrial Flutter - Clinically Important

• Flutter is harder to rate-control with drugs than AF
• Flutter responds extremely well to cardioversion (50-100J synchronised)
• Flutter can be cured with catheter ablation of the cavotricuspid isthmus (CTI ablation) - highly effective, >95% success

Management Options

Option 1 - Optimise Current Drugs First

• Can uptitrate to 7.5 mg or 10 mg OD if haemodynamically tolerated (MAP > 65, no worsening of BP)
• Check if today's dose was actually given - easy to miss in a complex ICU regimen

• If subtherapeutic (< 0.8 ng/mL): can give a small incremental dose (0.125 mg extra)
• Do NOT push digoxin aggressively in age 73, elderly, post-VT - narrow therapeutic index
• If already therapeutic: digoxin is doing its job and the sympathetic drive is overriding it

Option 2 - Add a Rate-Controlling Agent

• Excellent AV nodal rate control in both AF and flutter
• IV diltiazem 0.25 mg/kg bolus over 2 minutes, then infusion 5-15 mg/h
• Caution: Negative inotrope - check LV function; if LV dysfunction present, avoid
• Safe from QT standpoint - does not prolong QT
• Can be combined with digoxin; use carefully with bisoprolol (risk of AV block and hypotension)

Option 3 - Electrical Cardioversion (If Flutter Confirmed)

• Patient is already anticoagulated (apixaban)
• Haemodynamically compromised (rate 140 + sepsis + vasopressor)
• Synchronised DC cardioversion 50-100J (biphasic) is highly effective and appropriate
• Do not wait for drug optimisation if the patient is haemodynamically deteriorating

• Apixaban on board? If yes - no need to wait 3-4 weeks (he is already anticoagulated)
• QTc normalised - confirmed ✓
• Sedation: IV midazolam 1-2 mg or propofol for comfort before shock
• Synchronise mode - confirm R-wave synchronisation on monitor before discharge

Option 4 - Rate Control Target Reassessment

• Strict rate control (< 80 bpm) is less important than maintaining adequate cardiac output
• A rate of 110-120 may be acceptable temporarily while sepsis is being treated
• Chasing rate < 100 aggressively with multiple negative inotropes/chronotropes in a vasopressor-dependent patient can drop BP further
• Lenient rate control target (< 110 bpm) is acceptable in acute illness - AHA/ACC guidelines support lenient rate control (< 110) as non-inferior to strict control (< 80) in stable AF (RACE II trial)

Practical Step-by-Step Recommendation

One Important Note on Anticoagulation

ECG Interpretation – 06/06/2026 01:15

Measurements (from the machine + visual confirmation)

Primary Rhythm: Atrial Flutter with 2:1 AV Block

• Atrial rate ~300 bpm with 2:1 conduction giving ventricular rate ~136-150 bpm
• Sawtooth flutter waves are visible - best seen in leads II, III, aVF as the classic negative sawtooth pattern, and as positive deflections in V1
• The ventricular rhythm is regular (not irregularly irregular like AF) - each QRS is preceded by a consistent flutter wave relationship
• Narrow QRS - confirms this is supraventricular, not aberrant conduction

Secondary Findings

1. Paired Ventricular Premature Complexes (VPCs)

• Machine confirms: "Paired ventricular premature complexes - sequence of 2 V complexes"
• You can see the wide, aberrant beats interspersed among the narrow flutter-conducted beats
• These are ventricular couplets - two consecutive VPCs in a row
• In the context of the recent VT event, these couplets are a warning - couplets → non-sustained VT → sustained VT is a recognised escalation pathway
• The QTc of 489 ms means the pro-arrhythmic substrate is not fully resolved yet

2. Old Inferior Infarct

• Machine reports: "Inferior infarct, old"
• Q waves in II, III, aVF consistent with old inferior MI
• This is a new finding relative to your documented history - this patient has had a prior inferior MI that was not part of the known history
• This is clinically significant: old inferior MI + hypertension + diabetes = structural heart disease with likely LV dysfunction or diastolic dysfunction
• This makes diltiazem less safe and further supports the case for cardioversion over drug escalation

3. QTc 489 ms - Still Elevated

• Despite QT "normalising" clinically, the machine-calculated QTc is 489 ms
• Normal is < 450 ms in males
• 489 ms is still in the borderline prolonged range
• Amiodarone tissue levels are still active (t½ 40-55 days)
• This means adding any further QT-prolonging drug is still risky
• Do not restart amiodarone, sotalol, or class IA agents

Management Decision: Cardioversion Is the Right Call

1. Confirmed atrial flutter (not AF) - responds excellently to cardioversion
2. Haemodynamic compromise - rate 136 on vasopressor with sepsis
3. Old inferior MI - structural heart disease - drug escalation carries more risk
4. VPC couplets - residual arrhythmic instability, rate 136 driving them
5. Patient already anticoagulated on apixaban - no delay needed
6. QTc 489 - cannot safely add QT-prolonging antiarrhythmics
7. Digoxin + bisoprolol already failing to control rate

Cardioversion Protocol for This Patient

Post-Cardioversion - Maintenance of Sinus Rhythm

Summary

What Has Improved

Persistent Disorientation - This Now Needs a Definitive Answer

Cause 1: Post-Stroke Cognitive Impairment / Vascular Dementia

• Stroke involving pons + lentiform nucleus (basal ganglia)
• Lentiform nucleus (putamen/globus pallidus) connections to frontal cortex via thalamocortical loops - damage here causes frontal-subcortical cognitive dysfunction: inattention, disorientation, slowed processing
• This may be the baseline after the stroke 2 months ago
• Key question: Was he cognitively normal before this admission? Get collateral history from family urgently

Cause 2: Non-Convulsive Status Epilepticus (NCSE)

• This has not been excluded yet - EEG has not been mentioned as done
• NCSE is the single most commonly missed diagnosis in a disoriented post-stroke ICU patient
• Presents as: fluctuating consciousness, confusion, staring, subtle eye deviation, no obvious motor convulsions
• Post-stroke patients have significantly elevated seizure risk
• If EEG has not been done - this is the most urgent pending investigation

Cause 3: Osmotic Demyelination Syndrome (ODS)

• Na corrected from 116 to 126 (+10 mEq/L) in a patient with a pre-existing pontine infarct
• ODS classically appears on MRI 48-96 hours after correction as FLAIR hyperintensity in the pons
• The original MRI was done acutely - it was too early to show ODS changes
• A repeat MRI FLAIR now (day 4-5 post correction) will show ODS if it has occurred
• ODS on top of existing pontine infarct would explain persistent disorientation without new motor deficits
• This investigation is overdue

Cause 4: Septic Encephalopathy (Resolving)

• TLC falling from 24,000 to 12,000 - infection improving
• Septic encephalopathy can persist for days to weeks after infection control
• Spontaneous eye opening is a good sign - GCS trending up
• If this is the cause, expect gradual improvement over next 48-72 hours as WBC normalises

Cause 5: Drug-Induced Encephalopathy

• Baclofen - even at 5 mg BD, elderly patients can develop baclofen encephalopathy with features of sedation and confusion. Is the taper underway?
• Digoxin toxicity - digoxin 0.125 mg in a 73-year-old with any renal fluctuation can cause CNS effects (confusion, visual disturbance, delirium). Digoxin level should be checked
• Modafinil - if still running, can cause paradoxical agitation/disorientation in delirium

Cause 6: AF with Rapid Ventricular Rate → Cerebral Hypoperfusion

• Rate 136 on ECG with known old inferior MI and likely reduced cardiac reserve
• Sustained tachycardia → reduced diastolic filling → reduced cerebral perfusion
• Controlling the rate (ideally cardioversion to sinus) may improve sensorium directly
• This is a treatable contributor - controlling rate/rhythm may improve cognition

Atrial Fibrillation Continuing

• If not done yet - still indicated, especially given ongoing haemodynamic stress and cerebral hypoperfusion contribution to delirium
• If cardioversion was done and AF recurred - accept rate control for now

• Digoxin 0.125 mg + bisoprolol 5 mg
• If ventricular rate is still 130-140: increase bisoprolol to 7.5-10 mg cautiously (MAP permitting, noradrenaline weaning?)
• Check if noradrenaline is still running - if it is, the sympathetic drive will continue to counteract rate-control efforts

Action Plan - Priority Order

Encouraging Signs to Note

• Spontaneous eye opening = GCS Eye component 4 - this is meaningful improvement
• TLC 12,000 = approaching normal, infection responding
• No hypercapnia, no renal failure, no new focal neurology reported

What This Tells Us

• Post-stroke cognitive impairment is NOT the explanation for current disorientation
• His neurological baseline was intact despite the 2-month-old pontine + lentiform stroke
• The disorientation is therefore entirely acquired during this admission
• Something has happened to his brain during this hospitalisation that was not there before

The Differential Reranked

1. Osmotic Demyelination Syndrome (ODS) - NOW TOP PRIORITY

• Confusion, disorientation
• Dysarthria (already present from pontine stroke - may be worsening)
• Dysphagia (already present - may be worsening)
• In severe cases: spastic quadriparesis, locked-in syndrome

2. Non-Convulsive Status Epilepticus (NCSE) - Still Not Excluded

• Still on the list until EEG is done
• Post-stroke + metabolic disturbance + sepsis = high seizure risk
• A coherent patient who becomes persistently disoriented without explanation = NCSE until proven otherwise
• EEG today - this cannot wait further

3. Septic Encephalopathy - Less Likely Now

• TLC falling (12,000), infection responding
• Septic encephalopathy as the sole explanation is less convincing when:
  • Infection is clearly improving
  • Patient was cognitively intact before admission with the same background comorbidities
  • Disorientation is not improving proportionally with the infection

4. Cerebral Hypoperfusion from AF/Rate 136

• Sustained AF with ventricular rate 136 in a patient with old inferior MI and likely reduced cardiac reserve
• Reduced diastolic filling time → reduced cerebral perfusion pressure
• In an elderly patient with pre-existing cerebrovascular disease, this matters more than in a young patient
• Controlling the ventricular rate or cardioverting to sinus may produce visible improvement in sensorium
• This is a treatable, reversible contributor

5. Digoxin Toxicity

• Digoxin causes CNS toxicity (confusion, disorientation, visual changes) even at "therapeutic" levels in the elderly
• Age 73 + any renal fluctuation during sepsis/vasopressor use = reduced digoxin clearance
• Level must be checked today
• Digoxin toxicity ECG signs: ST scooping ("digoxin effect"), prolonged PR - review the ECG for these

What Needs to Happen Today - No Further Delay

MRI Brain with FLAIR Sequence

• Specifically request: DWI + FLAIR + T2
• ODS appears as symmetric pontine FLAIR hyperintensity - the classic "trident" or "bat-wing" pattern
• Can also involve extrapontine sites: basal ganglia (his lentiform nucleus is already damaged - ODS here may be harder to distinguish from old infarct), thalamus, cerebellum
• Timing is now optimal - day 4-5 post correction is when ODS becomes MRI-visible
• If MRI is not immediately available, CT will miss early ODS - push for MRI

EEG

• Request continuous EEG monitoring if possible, not just a spot EEG
• NCSE can be intermittent - a single 20-minute EEG may miss it
• If continuous EEG not available: standard 30-minute EEG with activation procedures
• If NCSE found: IV levetiracetam 1000 mg loading dose (safe, non-sedating, no hepatic metabolism issues, minimal drug interactions)

If ODS Is Confirmed on MRI

One Immediate Action You Can Take Right Now

• This is contributing to ODS risk ongoing
• Target should have been capped at 126-130 for this high-risk patient
• If Na is now > 132-134: consider desmopressin 2 mcg IV to slow further correction
• Each additional mEq/L rise in a patient who may already have ODS is additional myelin damage

Summary

Why Are They Recurrent? - Find the Cause First

Step 1: Quantify the Loss - Urine Electrolytes

Most Likely Causes in This Patient

1. Meropenem / Antibiotic-Driven Renal Tubular Loss

2. Vancomycin Nephrotoxicity - Renal Tubular Dysfunction

• Vancomycin causes proximal and distal tubular injury
• Tubular damage → impaired K⁺ and Mg²⁺ reabsorption → renal wasting
• This is the most likely culprit in this patient
• Check: rising creatinine? Casts on urinalysis? These would confirm tubular injury
• If vancomycin is causing this: switch to linezolid 600 mg BD via NG - equally effective for MRSA/Staphylococcal pneumonia, no nephrotoxicity, no electrolyte wasting

3. Noradrenaline / Catecholamine Effect

• Catecholamines (endogenous from sepsis + exogenous noradrenaline) drive K⁺ into cells via beta-2 receptor stimulation
• This causes redistributive hypokalemia - total body K is not low, but serum K falls
• As noradrenaline is weaned, K⁺ will shift back out of cells
• Implication: Do not over-replace K⁺ during high-dose vasopressor - when vasopressor weaned, K⁺ will rebound and you risk hyperkalemia

4. NG Tube Feeds - Inadequate Electrolyte Content

• Standard enteral feeds often contain insufficient K⁺ and Mg²⁺ for a critically ill patient with ongoing losses
• Check the electrolyte content of the enteral formula being used
• May need supplemented formulas or additional electrolytes added to feeds

5. Ongoing GI Losses

• Vomiting (even if reduced, may still be occurring via NG aspirates)
• Gastric secretions are rich in H⁺ and Cl⁻ → metabolic alkalosis → renal K⁺ wasting (kidney excretes K⁺ to retain H⁺ in alkalosis)
• Check: is there a metabolic alkalosis on ABG? If pH > 7.45 with low Cl⁻ = contraction alkalosis from vomiting = ongoing K⁺ loss

6. Refractory Hypokalemia from Unresolved Hypomagnesemia

• This is the most important physiological link and is frequently missed
• Magnesium is required to maintain intracellular potassium via Na-K-ATPase pump activity
• You cannot correct hypokalemia if hypomagnesemia persists - the K⁺ will keep leaking out of cells regardless of how much you give
• Always replace Mg²⁺ first or simultaneously - hypokalemia will be refractory until Mg²⁺ is ≥ 0.8 mmol/L
• Are you giving enough magnesium? (see doses below)

7. Hyperaldosteronism / Secondary Aldosteronism

• Sepsis + vasopressor use + volume depletion activates the RAAS
• High aldosterone → increased renal K⁺ excretion
• This is likely contributing to the renal wasting pattern

8. Diabetes - Osmotic Diuresis

• If blood glucose is poorly controlled, glycosuria causes osmotic diuresis
• Osmotic diuresis washes out K⁺ and Mg²⁺
• Check: is glucose well controlled? HbA1c? Current CBG readings?

Replacement - Are You Giving Enough?

Potassium Replacement

• Maximum rate via central line: 20 mEq/hour with continuous cardiac monitoring
• Never give > 10 mEq/hour via peripheral line
• Add KCl to NG feeds if enteral route functioning - 20-40 mEq BD via NG tube

Magnesium Replacement

• Renal function: Reduce dose if CrCl < 30 - but currently renal function normal
• Oral magnesium (magnesium glycinate or oxide) via NG is an option for maintenance once acute phase resolves - better sustained levels than intermittent IV

The Vicious Cycle to Break

Hypomagnesemia
      ↓
Na-K-ATPase pump fails → K⁺ leaks out of cells
      ↓
Hypokalemia (refractory to K replacement alone)
      ↓
QTc prolongation + VPC/TdP risk
      ↓
Replace K but not Mg → K keeps falling
      ↓
Cycle repeats

Practical Protocol for This Patient

Why This Matters Beyond Just Electrolyte Numbers

• QTc is already 489 ms - every drop in K⁺ and Mg²⁺ pushes QTc higher → TdP risk returns
• AF with rapid ventricular response is harder to rate-control with low K⁺ and Mg²⁺
• VPC couplets seen on ECG - these will worsen with electrolyte depletion
• You had one VT arrest already - recurrent hypokalemia is the most direct path back to another one

Two Separate Problems to Address

Problem 1: Multiple Mouth Blisters / Ulcers

Differential Diagnosis

How to Differentiate - Examine the Lesions Carefully

Red Flag - Rule Out SJS/TEN First

• Skin involvement - blistering, peeling, target lesions elsewhere on body
• Eye involvement - conjunctival injection, discharge
• Genital mucosal involvement
• Systemic features - high fever, elevated LFTs, eosinophilia

Most Likely Diagnosis: Oral Candidiasis

• Diabetes (impaired mucosal immunity)
• Prolonged meropenem (destroys normal oral flora)
• Prolonged ICU stay + NG tube
• No oral hygiene possible (intubated/sedated phase)

• Fluconazole 200 mg loading dose, then 100-200 mg OD via NG tube - first line for oropharyngeal candidiasis
• Duration: 7-14 days
• Alternative: nystatin suspension 5 mL QDS (swish and swallow) if systemic azole not preferred
• Monitor for fluconazole interactions: apixaban levels can rise with fluconazole (CYP3A4 inhibition) - watch for bleeding signs; may need dose reduction of apixaban to 2.5 mg BD

• Acyclovir 400 mg TDS via NG tube for 7-10 days
• Or IV acyclovir 5 mg/kg TDS if unable to absorb orally

Problem 2: Inability to Swallow - Neurogenic Dysphagia

Why Is He Unable to Swallow?

• The swallowing centre (nucleus tractus solitarius + nucleus ambiguus) is located in the dorsal medulla/lower pons
• His pontine infarct directly damages the central pattern generator for swallowing
• Dysphagia from isolated pontine infarction is well documented - affects both oral and pharyngeal phases
• The lentiform nucleus (basal ganglia) stroke contributes via corticobulbar disconnection - poor swallow triggering and upper oesophageal sphincter control

• Mouth ulcers causing pain on attempted swallowing - pain itself inhibits swallowing
• Muscle deconditioning from prolonged NG feeding - "use it or lose it" applies to swallowing musculature
• Drooling (from day 1) confirms ongoing bulbar/pseudobulbar involvement

Management of Neurogenic Dysphagia

If Dysphagia Persists Beyond 4-6 Weeks

• More comfortable than long-term NG tube
• Better nutrition delivery
• Reduces risk of NG tube-related sinusitis and mucosal injury
• ESO guidelines recommend considering PEG if dysphagia persists > 2-4 weeks with no prospect of early recovery

Priority Action List

Denosumab - The Unifying Diagnosis for Three Persistent Problems

Mechanism 1: Electrolyte Disruption

Denosumab → Osteoclasts suppressed → Calcium stays in bone
      ↓
Serum calcium falls → PTH surges (secondary hyperparathyroidism)
      ↓
PTH increases renal phosphate excretion → Hypophosphatemia
PTH increases renal magnesium excretion → Hypomagnesemia
      ↓
Hypomagnesemia → Refractory hypokalemia (Na-K-ATPase failure)
      ↓
All three electrolytes fall and resist replacement

• Hypocalcaemia (may be severe - FDA issued a boxed warning January 2024 for severe hypocalcaemia with denosumab)
• Hypophosphataemia
• Elevated PTH
• Low Vitamin D (likely pre-existing in a diabetic elderly patient - exacerbates everything)

Mechanism 2: Mouth Ulcers / MRONJ

• Inhibiting osteoclast-mediated bone remodelling → jaw bone cannot repair microdamage or heal mucosal defects
• Impairing angiogenesis → reduced vascular supply to jaw mucosa → ischaemic environment
• Soft tissue toxicity → mucosal healing is impaired → ulcers do not close

Critical Additional Concern: Hypocalcaemia + Recent VT

• Prolongs the QT interval independently
• Was likely a contributing factor to the QT prolongation and TdP throughout this admission alongside the drug-induced QT effects
• The refractory electrolyte abnormalities were fuelling the arrhythmic instability

Management Plan

Immediate - Electrolytes

• Calcium + Vitamin D supplementation → reduces PTH drive → reduces renal Mg/K wasting
• Oral calcium carbonate 1-1.5 g elemental calcium/day via NG alongside vitamin D

MRONJ Management

Vitamin D - Urgent

• Vitamin D deficiency is almost certain
• Denosumab-induced hypocalcaemia is dramatically worsened by Vitamin D deficiency
• Start cholecalciferol 60,000 IU weekly via NG immediately - do not wait for the level
• Check level; if < 10 ng/mL (severely deficient): loading dose 1,00,000-2,00,000 IU over 2 weeks

Revised Understanding of the Whole Admission

Priority Actions Right Now