The Deteriorating Ward Patient — Clinical Case

DCE Long Case

Patient brief (provided to trainee)

Patient: Mrs Margaret O'Sullivan, 78 years old, retired teacher. ^[1]

Presenting complaint: Acute deterioration overnight on day 4 of admission for community-acquired pneumonia. The nursing staff called the ward registrar at 02:00 because her NEWS2 had risen from 3 (on admission-day observations) to 9 over the preceding six hours, and she had become confused and short of breath. ^[1]

Past history: Ischaemic heart disease (prior NSTEMI three years ago, treated with PCI and a drug-eluting stent; currently on aspirin, atorvastatin, bisoprolol, ramipril), chronic kidney disease stage 3 (baseline creatinine 140 micromol/L, attributed to hypertensive and ischaemic nephropathy), type 2 diabetes (metformin 1 g twice daily and empagliflozin 10 mg daily, last HbA1c 62 mmol/mol), hypertension, and moderate aortic stenosis (echo two years ago showed a valve area of 1.1 cm2, mean gradient 32 mmHg, considered for surveillance but not yet for valve intervention). She lives independently with her husband of 50 years. She has no formal advance care directive but has told her family she "does not want to be a burden." ^[1]

Examination findings (trainee elicits):

• Airway: patent, speaking in short sentences.
• Breathing: respiratory rate 30, SpO2 88 per cent on room air, 92 per cent on 15 L via reservoir mask. Increased work of breathing, accessory muscle use. Right base: dull to percussion, bronchial breath sounds and coarse crackles.
• Circulation: heart rate 128, irregularly irregular, blood pressure 84/52. Capillary refill 4 seconds. JVP raised at 6 cm. Peripheries cool and mottled to the knees. Urine output 15 mL/hour via urinary catheter (was 50 mL/hour 12 hours ago).
• Disability: GCS 14 (E4 V4 M6), new mild confusion, oriented to person but not to time or place. Pupils equal and reactive. Bedside glucose 8.2 mmol/L.
• Exposure: temperature 38.7. Abdomen soft, non-tender. No rash. Two peripheral cannulae in situ. Right internal jugular central line in situ (placed on admission). Surgical wound: nil (medical admission). ^[1]

Investigations (available results):

• Lactate 4.6 mmol/L (raised).
• Creatinine 210 micromol/L (up from baseline 140; up from 148 on admission).
• Potassium 5.8 mmol/L. Sodium 134. Bicarbonate 18. Chloride 108.
• Haemoglobin 112 g/L (unchanged from admission). White cell count 18.4 (neutrophilia). Platelets 210.
• CRP 210 (up from 140 on admission).
• INR 1.3.
• Troponin: pending.
• Chest X-ray: progressive right lower lobe consolidation compared with the admission film, now involving approximately two-thirds of the right lower lobe. No effusion.
• ECG: atrial fibrillation at a ventricular rate of 128, with peaked T waves across the anterior leads. No acute ischaemic changes. QRS 100 ms.
• Venous blood gas: pH 7.28, PaCO2 4.2 kPa, PaO2 9.0 kPa (on 15 L), bicarbonate 18, base excess minus 8, lactate 4.6. ^[1]

Drug chart on admission (current): aspirin 100 mg daily, atorvastatin 80 mg nocte, bisoprolol 5 mg daily, ramipril 5 mg daily, metformin 1 g twice daily, empagliflozin 10 mg daily, IV benzylpenicillin 1.2 g six-hourly (day 4), oral clarithromycin 500 mg twice daily (day 4), subcutaneous enoxaparin 40 mg daily (venous thromboembolism prophylaxis). ^[1]

Candidate's structured presentation (model)

Opening statement (SASPOP): ^[1]

"Mrs Margaret O'Sullivan is a 78-year-old retired teacher who presents with acute deterioration on day 4 of admission for community-acquired pneumonia, with a NEWS2 that has risen from 3 to 9 over six hours. She has a background of ischaemic heart disease, chronic kidney disease stage 3, type 2 diabetes, hypertension, and moderate aortic stenosis. She lives independently with her husband. ^[1]

Her main problems are:

1. Septic shock from progressive pneumonia — refractory hypotension at 84 over 52, lactate 4.6, mottled peripheries, oliguric.
2. New atrial fibrillation with a rapid ventricular response at 128 — a complication of the sepsis and the electrolyte disturbance.
3. Acute kidney injury with hyperkalaemia — creatinine risen from 140 to 210, potassium 5.8 with peaked T waves, oliguric.
4. Respiratory failure — type 1 with a metabolic acidosis; pH 7.28 on the venous gas.
5. New delirium — a marker of severity.
6. Significant comorbidity and the ceiling-of-care question — aortic stenosis, CKD, ischaemic heart disease, diabetes. ^[1]

My immediate priorities are the ABCDE assessment with high-flow oxygen, cautious intravenous fluid resuscitation in 250 mL aliquots given her aortic stenosis, the Sepsis Six bundle with escalation of antibiotics to piperacillin-tazobactam given progression on first-line therapy, treatment of the hyperkalaemia with calcium gluconate and insulin-dextrose, withholding the nephrotoxic and harmful drugs (metformin, empagliflozin, ramipril, and reviewing the bisoprolol given the hypotension), and an urgent MET call for ICU outreach review. I will also ask the family about her wishes given the comorbidities and the severity of the acute illness." ^[1]

Management plan: ^[1]

1. Call for help. Activate the MET call / ICU outreach now. This is a multi-organ-failure patient with refractory hypotension, hyperkalaemia with ECG changes, and new arrhythmia; a single registrar cannot safely deliver the simultaneous interventions required.
2. ABCDE resuscitation. High-flow oxygen via reservoir mask to target SpO2 92 to 96 per cent. Two large-bore cannulae (in addition to the central line). 250 mL bolus of balanced crystalloid (Hartmann's) over 15 minutes, reassess for responsiveness and overload, repeat cautiously. Continuous cardiac monitoring. Prepare for noradrenaline via the central line if the blood pressure does not respond to the first bolus.
3. Sepsis Six within one hour. Blood cultures (peripheral and from the central line), broad-spectrum IV antibiotics (piperacillin-tazobactam 4.5 g IV eight-hourly given progression on benzylpenicillin and clarithromycin, and the possibility of a hospital-acquired or resistant organism), 30 mL/kg crystalloid in aliquots as above, lactate already measured, hourly urine output, oxygen. Send sputum and consider a urinary antigen panel.
4. Treat the hyperkalaemia. Calcium gluconate 10 mL of 10 per cent IV over 5 to 10 minutes for the ECG changes (peaked T waves). Insulin-dextrose (10 units Actrapid in 50 mL of 50 per cent dextrose over 15 minutes). Salbutamol nebuliser 5 mg. Recheck potassium and ECG in 30 minutes. Monitor blood glucose for the next 2 hours (hypoglycaemia risk). If refractory or recurs, she will need renal replacement therapy.
5. Manage the atrial fibrillation. Treat the sepsis and the hyperkalaemia first; the AF is likely secondary. Hold the bisoprolol in the hypotensive state. If the AF is itself causing haemodynamic compromise that does not respond to the resuscitation, amiodarone 300 mg IV over one hour or, if peri-arrest, synchronised DC cardioversion. Plan anticoagulation once the acute phase resolves.
6. Drug review. Withhold metformin (lactic acidosis risk in AKI and shock), empagliflozin (euglycaemic DKA risk, nephrotoxic in volume depletion), ramipril (hypotension and AKI), and the NSAIDs if any. Convert her diabetes management to a sliding-scale insulin infusion. Continue aspirin, atorvastatin, and enoxaparin. Hold the bisoprolol temporarily and reintroduce once the blood pressure is stabilised.
7. Escalation and the ceiling of care. ICU admission for vasopressor support, treatment of the hyperkalaemia (likely including renal replacement therapy given the oliguric AKI), and possibly non-invasive ventilation. Discuss with the family early — ask about an advance care directive and her expressed wishes; she is a previously independent 78-year-old with reversible sepsis, which in most institutions supports a time-limited trial of ICU care with a defined review point at 48 to 72 hours. Keep the family informed at every step. ^[1]

Communication and shared decision-making: Explain to the family that Mrs O'Sullivan has become much more unwell overnight with worsening of the pneumonia and that the body is struggling to compensate. Explain that the immediate priorities are to support the blood pressure, treat the infection, and correct the potassium. Explain that the next two to three days will tell us whether the treatment is working. Ask, gently, what she would want — surface the ceiling-of-care conversation early, not in a crisis. Document the discussion and the plan. ^[1]

Examiner discussion questions

Q1: "Her NEWS2 rose from 3 to 9 over six hours. At what points should the escalation have occurred, and what does the trajectory tell you about the afferent limb of the rapid response system?" ^[1]

"The trajectory tells me she was deteriorating throughout, and the escalation should have occurred at the NEWS2 of 5 to 6 threshold — that is the point at which urgent ward-registrar review and hourly monitoring are mandated, and at which the response should have been escalated if the score continued to rise. By time score reached 7, the emergency critical care assessment was overdue. The lesson is about the afferent limb of the rapid response system: the recognition. A single 'acceptable' value in a rising trend is not reassuring; the trend is the signal, and the observation set must be measured, recorded, scored, and acted upon at every point. The failure mode here is the delayed response to the intermediate score — the patient whose NEWS2 reaches 5 or 6 and is rechecked in an hour rather than reviewed by a registrar. The system only works if the intermediate thresholds trigger action, not just the emergency threshold." ^[1]

Q2: "How would you adjust the standard 30 mL/kg sepsis fluid resuscitation for her aortic stenosis?" ^[1]

"The standard Surviving Sepsis Campaign 2021 recommendation of 30 mL per kilogram crystalloid in the first three hours gives the total amount; her aortic stenosis dictates that I deliver it in smaller aliquots with reassessment between each ^[3]. My approach is 250 mL of balanced crystalloid over 15 minutes, then reassess for fluid responsiveness — a rise in blood pressure, a fall in heart rate, an improvement in capillary refill or mottling, a rise in urine output — and for signs of overload — a rise in JVP, new basal crackles, a fall in SpO2, increasing work of breathing. If she is fluid-responsive and not overloading, I repeat the bolus, working towards the 30 mL per kilogram total over the three hours. If she is not fluid-responsive, or if she shows signs of overload, I start an early vasopressor — noradrenaline via the central line — to restore the mean arterial pressure to above 65 mmHg, and I use a focused bedside echocardiogram or a passive leg raise to guide the fluid if available. The principle: the comorbidity does not override the guideline, but it modifies its execution — smaller aliquots, closer reassessment, and a lower threshold for vasopressor support."

Q3: "She is oliguric with a rising creatinine and hyperkalaemia. What is your approach to the acute kidney injury?" ^[1]

"The AKI is multifactorial — pre-renal from the septic shock and hypovolaemia, nephrotoxic from the empagliflozin and the ramipril in the volume-depleted state, and potentially from the pneumonia itself (infection-related glomerulonephritis is less likely, but intrinsic AKI from sepsis is common). My management is supportive: restore the intravascular volume cautiously as above, withhold all nephrotoxins (the metformin, the empagliflozin, the ramipril, and any NSAID), monitor the urine output hourly via the catheter, and treat the hyperkalaemia as described. I send a urinary sodium and a urinary osmolality early to help discriminate pre-renal from intrinsic AKI (a fractional excretion of sodium below 1 per cent suggests pre-renal; above 2 per cent suggests intrinsic), though in practice the distinction is often blurred by the diuretics and the sepsis. I review the medications that need dose adjustment for the renal failure — the antibiotics, the enoxaparin (which accumulates in renal failure and may need to be switched to unfractionated heparin or dose-reduced), the bisoprolol. The indications for renal replacement therapy are the standard ones — refractory hyperkalaemia, severe metabolic acidosis, fluid overload that does not respond to diuretics, and uraemic complications — and she is likely to meet at least one of these given the hyperkalaemia and the oliguria, which is another reason for the ICU admission." ^[1]

Q4: "What is the role of the bedside echocardiogram in this patient?" ^[1]

"The bedside focused echocardiogram is invaluable in the deteriorating patient with shock because it discriminates the shock type and guides the fluid and vasopressor strategy. In this patient, I would expect to see a small, under-filled left ventricle with hyperdynamic contraction (consistent with hypovolaemic and septic shock with low preload), and the assessment of the inferior vena cava collapsibility gives a dynamic measure of fluid responsiveness. The echo also reassesses the aortic stenosis (the valve gradient may change with the altered haemodynamics) and screens for a new wall motion abnormality (the troponin is pending and the ECG is in AF, so a regional wall motion abnormality would suggest an acute coronary syndrome contributing to the shock). The echo does not replace the clinical assessment, but it sharpens it — it tells me whether the heart is under-filled and hyperdynamic (give more fluid cautiously), or dilated and poorly contracting (cardiogenic component, use an inotrope rather than fluid). In the hands of a trained operator, it is one of the highest-yield investigations at the bedside of the deteriorating patient." ^[1]

Q5: "She has told her family she does not want to be a burden. How does this inform the ceiling-of-care conversation?" ^[1]

"This is a statement of values, not a specific treatment refusal, and it requires careful exploration. I would sit with the family — with her husband if he is present, and with her children — and ask what she meant. Did she mean she would not want invasive life support if the outcome were likely to be dependence on others? Did she mean she would not want to be kept alive artificially if she could not return home? Did she have a specific scenario in mind? The conversation is to translate a general value statement into a specific treatment plan. My framing is honest and hopeful: that the sepsis is treatable, that she is in the best place for the supportive care she needs, that the next 48 to 72 hours will tell us whether the treatment is working, and that if the treatment is working and she is recovering towards her baseline, we will continue; if it is not, and she is heading towards a level of dependence she would not accept, we will refocus on comfort. I would offer a time-limited trial of ICU care as the reasonable middle path — maximal support for a defined period, with a clear review point, and a commitment to revisit the plan with the family at every stage. I document the conversation, the values expressed, the plan, and the review point. And I keep the family informed — the conversation does not end at the decision; it continues through the course." ^[1]

Q6: "What is the single most important lesson from this case for a registrar managing the deteriorating ward patient?" ^[1]

"The single most important lesson is that the deterioration is not sudden — it has been building for hours, and the system that measures, scores, and escalates is the system that catches it. The NEWS2 trajectory from 3 to 9 over six hours is a patient who has been deteriorating throughout, and the response at each intermediate threshold is what determines whether the deterioration is intercepted early or whether it culminates in an arrest. The registrar's job is to measure the full vital sign set at every observation, to calculate the NEWS2 and watch the trend, to act at the intermediate thresholds rather than waiting for the emergency threshold, and to call for help early. The second lesson is the integration: the ABCDE assessment, the Sepsis Six, the hyperkalaemia protocol, the drug review, and the ceiling-of-care decision are not separate tasks but a single coordinated response, delivered by a team that has been called early. The registrar who tries to do all of this alone, without calling for help, is the registrar who is most likely to fail the patient." ^[1]