Clinical case review
Review the complete Medicine case, its five progressive clues, the diagnosis, and the clinical reasoning behind the answer.
Want to test yourself first? Play the original case before reading the clues and explanation below.
WhatsTheDx publishes a new Medicine case every day. Five clues, one diagnosis, and no account needed to play.
Play today's Medicine caseThese are the case findings in their original reveal order, moving from the broader presentation to the most discriminating evidence.
Clue 1
A 74-year-old is brought in after his legs went from under him as he stood up out of an armchair. He did not black out and he did not hurt himself — they simply would not hold him. For two days before that he had felt weak all over, in a way he found hard to describe, and had noticed his heart thumping now and again.
Clue 2
The obvious things are not there. He has no chest pain and never has had. He has no fever and no cough. Neurologically he is symmetrical — no facial droop, no one-sided weakness, no sensory loss — so this is not a stroke, and the weakness is genuinely everywhere rather than anywhere in particular. His blood count is entirely normal.
Clue 3
His medicines are worth reading in order. He has been on ramipril for years without trouble. Three weeks ago spironolactone was added for his heart failure — a sensible prescription, made by someone who knew what they were doing. And for the past week, entirely separately and without telling anyone, he has been buying ibuprofen for a bad back and taking it every day.
Clue 4
The chemistry gives the setting before it gives the answer. Creatinine is 2.1 mg/dL with an estimated filtration rate of 28 mL/min — kidneys working at well under a third of normal, which is the threshold below which the body loses its margin for handling exactly this kind of problem. Sodium, calcium and glucose are all normal, with only a mildly reduced bicarbonate.
Clue 5
The tracing is abnormal in a very specific way. The T waves are tall and sharply peaked across the precordial leads, with no ischaemic ST change to go with them. The PR interval is prolonged. And in several leads the P waves have flattened out and disappeared altogether, with the QRS sitting at the upper limit of normal width. The laboratory phones the result through while the tracing is still being read: potassium 7.2 mEq/L, on a sample with no haemolysis.
Diagnosis
The symptoms are almost useless on their own, and that is the point worth taking away: weakness and palpitations are the typical complaints, but most patients with mild-to-moderate elevation are relatively asymptomatic, so the level is often a surprise. What makes this diagnosable is the drug history read in order. Ramipril has been on board for years without incident. Spironolactone was added three weeks ago — a correct and evidence-based prescription for heart failure. Then, entirely separately and unrecorded, he began buying ibuprofen for his back a week ago. Each of these three is a recognised cause: renin-angiotensin-aldosterone system inhibitors, potassium-sparing diuretics and non-steroidal anti-inflammatory drugs. No individual prescriber made a mistake; the harm is in the combination, and the last component was bought off a shelf. The kidney explains why he had no margin. Hyperkalaemia typically occurs once the filtration rate falls below 30 mL/min, and his is 28. The electrocardiogram is the finding that determines urgency, and its changes track the level in a recognised sequence: tall peaked T waves at 5.5 to 6.5 mEq/L, flattening or loss of P waves at 6.5 to 7.5, QRS widening at 7 to 8, and severe arrhythmias with a sine-wave pattern at 8 to 10. His tracing shows peaked T waves, a prolonged PR interval and P waves already disappearing, with the QRS at the upper limit of normal — which places him around the 7 mark before the laboratory confirms 7.2 mEq/L, and shows that the heart is already affected. That is what dictates the first move: calcium gluconate is generally preferred as the initial agent when there is evidence of cardiac toxicity, because it stabilises the myocardium immediately, before insulin with dextrose is used to shift potassium into cells. Confirming the sample was not haemolysed is not a formality either, since a spuriously raised result is common enough to be worth excluding before treating.
The explanation and decisive case findings were checked against the source below.
Open the clinical referenceFor education and entertainment only. This fictional case is not medical advice and does not replace supervised clinical training, diagnosis, or treatment.
