The File That Stopped Me
The case — as the file told it
She was 22 years old — intubated, sedated, connected to every monitor in the room. I stopped. I looked deeper into that file than I had looked at anything all morning. The kind of deep where time slows down and your brain starts sending quiet little panic signals it doesn't want you to notice.
She was 22. I was a first-year resident. And somehow, I was the one holding her file.
Turning the pages, each one had something new to say — a new finding, a new direction, a new decision that had determined where she was today.
Fifteen days ago, she walked in with fever. Icterus. Purpuric patches scattered across her skin. The kind of presentation that makes you pause before you even write the first differential.
Investigations came back with a leukocytosis, a platelet count in freefall, and every hemoparasite panel — malaria, dengue, scrub typhus, the full board — negative. She was labelled Pyrexia of Unknown Origin and managed symptomatically while the team looked harder for answers.
But she didn't wait for answers.
Her sensorium began to slip. GCS dropping. Drowsy, unresponsive to commands. By the time she reached our centre, she needed a tube. Post-intubation ABGs told the next part of the story — metabolic acidosis, raised lactate, and electrolyte chaos.
And somewhere in that chaos — the potassium was wrong.
Deep in the differentials, two names kept surfacing — TAFRO syndrome and HLH (Haemophagocytic Lymphohistiocytosis). Two rare, aggressive conditions where the immune system turns on itself — both capable of explaining the fever, the cytopenias, the collapsing sensorium, the entire picture.
Meanwhile, her potassium was low. A KCl infusion was started. Reasonable, necessary, textbook.
Her evening potassium came back at 4.7 mEq/L. Perfectly normal. No flag, no alarm, no reason to pause. The infusion continued.
Later that night, on continuous ECG monitoring, something shifted.
I ran an ABG. Potassium: 5.2 mEq/L.
Still within the upper limit of normal. Still a number that — on paper — shouldn't cause concern. But the ECG had already told us the truth before the lab could print it.
We stopped the infusion immediately. By morning, her potassium had settled back to 4.7. The T waves had normalised. She was stable.
The number never crossed the upper limit. But her heart reacted anyway.
It isn't just about the value — it's about the direction, the speed, and the patient in front of you.
And that is exactly what this blog is about.
The Five Questions
Our roadmap through hyperkalemia
Before we go further, let me lay down our roadmap. Every topic on this blog follows the same five questions — because good clinical thinking is always structured.
Every question answered. In order.
When Does It Get Overwhelming?
Classification and the alarm threshold
Potassium has a normal range of 3.5 to 5.0 mEq/L. Anything above 5.0 is hyperkalemia — but not all hyperkalemia is created equal.
Often asymptomatic. Subtle ECG changes may be present.
Clinical manifestations begin to appear.
High risk of life-threatening arrhythmias.
Textbooks will tell you symptoms begin beyond 6.0 mEq/L. And they are not wrong.
But here is what our patient taught us at 2 AM — her potassium was 5.2. Mild by classification. Within the upper limit by most standards. And yet her ECG had already changed.
Because hyperkalemia does not read textbooks.
The number matters — but so does the rate of rise, the underlying condition, and the heart that is receiving it. A patient with chronic kidney disease may tolerate a potassium of 5.8 for weeks. But an acutely ill patient, on a KCl infusion, in a compromised system — their heart may react at 5.2.
The specification is not just the number.
It is the number in context of the patient.
How Does the Potassium Rise?
Drivers and causes
Rather than a long explanation, let this visual do the talking —
Before you scroll past it, two things in this image matter most clinically:
First — Pseudohyperkalemia. Before you treat a high potassium, always ask: is this real? A haemolysed sample, a tight tourniquet, excessive fist pumping — and your lab will hand you a number that was never in the bloodstream. Repeat the sample before you panic.
Second — notice where our patient sits in this image. Sepsis. Metabolic acidosis. IV potassium supplementation. She had three simultaneous drivers pushing her potassium up — which is exactly why her heart reacted before the number crossed the upper limit.
One cause is manageable. Three at once, in a critically ill 22-year-old, is a different conversation entirely.
Why Does It Become Dangerous?
The pathophysiology
To understand why potassium becomes dangerous, you need to understand where it lives and what it does.
98% of the body's potassium sits inside the cell. Only 2% circulates in the blood — and that 2% is what we measure. That 2% is what the heart listens to.
Potassium controls the electrical charge across every cell membrane — the resting membrane potential. Think of it as a battery, always slightly charged, always ready to fire. When the potassium outside the cell rises, that charge narrows. The battery weakens. And a weakened battery in the heart means unpredictable firing — too easy at first, then not at all.
The heart does not wait for a critical lab value. It responds to the electrical environment around it. And that environment changes faster than any blood test can capture.
What Does It Look Like?
Clinical history and examination
Hyperkalemia is a quiet condition until it isn't.
In mild to moderate cases, the symptoms are non-specific enough to be dismissed — generalised weakness, fatigue, palpitations, an occasional syncopal episode. The kind of symptoms that could belong to a dozen other diagnoses on a busy ward.
On examination, look for hypertension and oedema in the context of renal disease. Signs of hypoperfusion may be present. Muscle tenderness should raise the suspicion of rhabdomyolysis — a significant intracellular source of potassium release. Jaundice points toward haemolytic conditions where cell destruction drives potassium into the bloodstream.
The neuromuscular examination tells the most — generalised muscle weakness, flaccid paralysis, depressed deep tendon reflexes. These are the body's warning that the electrical imbalance has moved beyond the heart and into every excitable cell it owns.
Our patient had jaundice. She had weakness. She had the full picture of a system under siege — and potassium was one of the many things going wrong inside it.
How Do We Confirm It?
Evaluation and investigations
The first investigation in any suspected hyperkalemia is not a blood test.
It is the ECG.
Because the arrhythmias that hyperkalemia produces can cause sudden death — and the ECG tells you in real time what the potassium is doing to the heart, sometimes before the lab result is even back.
| Potassium Level | ECG Change |
|---|---|
| 5.5 – 6.5 mEq/L | Tall, peaked T waves |
| 6.5 – 7.5 mEq/L | Flattening or loss of P waves |
| 7.0 – 8.0 mEq/L | Widening of QRS complex |
| 8.0 – 10.0 mEq/L | Sine wave pattern, severe arrhythmias, risk of asystole |
The rate of rise has far greater influence than the absolute potassium level.
A fast rise at 5.2 can be more dangerous than a slow climb to 6.0.
The Night It Happened
Back to 2 AM — the moment of recognition
The evening was quiet. ICU quiet — which is not really quiet at all, just a specific kind of stillness made of ventilator hums and monitor beeps and the soft shuffle of night staff moving between beds. Our eyes moved from monitor to monitor in a slow, almost hypnotic loop. The way they always do on a long night shift.
And then one monitor stopped us.
The T waves were tall. Peaked. Different from an hour ago.
A moment of confusion settled over us — not panic, not immediately — just that particular stillness when your brain registers something wrong before it can name what. The kind of moment where time slows just enough for instinct to take over.
My hands went straight to the KCl infusion. Off.
My senior reached for calcium gluconate without a word exchanged between us.
No discussion. No deliberation. Just two people who had been trained for exactly this moment, moving at the same time, in the same direction.
GRBS checked. Normal. That mattered — because insulin was likely coming next, and you never reach for insulin without knowing where the glucose stands.
A 12-lead ECG was taken. The tall, peaked T waves were there in black and white now — no longer just a monitor finding, but a documented rhythm telling us something had shifted.
Then the ABG. Potassium: 5.2 mEq/L.
5.2. Technically within range. Not a number that should, on paper, produce ECG changes of this kind. But the ECG didn't care about the number. It cared about what had happened to the potassium — how fast it had moved, in which direction, and in what kind of patient.
Rapid correction. That was the answer.
It wasn't where the potassium had landed. It was the speed at which it got there. A slow rise gives the body time to adapt. A fast rise — even into the upper range of normal — gives it no such warning.
How Do We Treat It?
Management — step by step
Now we come to the part that matters most when you are standing at that bedside at midnight — what do you actually do?
The answer depends on four things: how fast the potassium rose, how high it climbed, what symptoms the patient is showing, and what caused it. Not every hyperkalemia is treated the same way.
Who needs aggressive intervention — right now, no waiting:
Our patient had ECG changes at 5.2. She qualified.
Stop whatever is driving the potassium up. In our case, the KCl infusion. In others, it may be a medication, a diet, or an ongoing process. Treat the reversible cause simultaneously.
Calcium therapy comes first — always — when there are ECG changes or arrhythmias. It does not lower the potassium. It protects the cardiac membrane while you bring the number down.
Calcium gluconate over calcium chloride — gluconate is safe through a peripheral line. Calcium chloride carries three times the elemental calcium, caustic enough to cause severe tissue necrosis if it extravasates peripherally. Gluconate for the peripheral line. Chloride for the central line or the arrest.
Insulin drives potassium intracellularly — one of our most reliable and fastest shifts. In euglycaemic patients, always co-administer dextrose to prevent hypoglycaemia. Check the GRBS first — which is exactly why my senior checked it before anything else that night.
Beta-2 agonists such as albuterol work through a similar mechanism. The doses required for potassium shift are significantly higher than those used for bronchodilation. Do not assume a standard nebuliser dose will do the job.
Sodium bicarbonate — beneficial specifically in the setting of metabolic acidosis. Not a standalone treatment, but a useful adjunct.
A note on epinephrine — it is not recommended for hyperkalemia management. The risk of inducing angina outweighs any theoretical benefit. Do not reach for it.
Shifting potassium into cells buys time. Eliminating it from the body is what actually resolves the problem.
Diuretics — loop and thiazide diuretics enhance renal potassium excretion. Useful in non-oliguric, volume overloaded patients, never as monotherapy.
· Hypervolaemic: Furosemide 40mg every 12 hours or continuous infusion
· Euvolaemic/hypovolaemic with preserved renal function: isotonic saline first, then Furosemide 40mg
Cation exchangers — Sodium Polystyrene Sulfonate has fallen out of favour. If used, never co-administer with sorbitol. Newer agents — Patiromer and Sodium Zirconium Cyclosilicate — are increasingly preferred.
For patients with end-stage renal disease or severe renal impairment, haemodialysis remains the definitive answer. When the kidneys cannot excrete potassium, you need a machine to do it for them.
Treat the potassium. Watch the patient. Always both, never just one.
What Happens When We Miss It?
Complications and a final word on diet
The progression of untreated hyperkalemia is predictable, and it is unforgiving.
We caught ours at the T wave stage. Tall, peaked, and early enough to act on. Not every team will be that fortunate. This is why hyperkalemia is not a condition you manage retroactively — it is one you anticipate, monitor for, and move on fast.
A final word on diet: Generally, dietary potassium restriction is unnecessary for most people. The healthy kidney is remarkably efficient at maintaining potassium balance regardless of what you eat. The exception — patients with chronic kidney disease or severe, recurrent hyperkalemia. For them, limiting high potassium foods — bananas, oranges, potatoes, tomatoes, nuts — becomes a meaningful part of long-term management.
For everyone else, the potassium in your diet is not your enemy. Your kidneys have it handled.
This blog began at 2 AM, in a quiet ICU, watching a monitor that refused to stay quiet.
A 22-year-old with a diagnosis we were still chasing, a potassium that never technically crossed the upper limit, and an ECG that told us the truth before the lab could.
Hyperkalemia taught me that night — not from a textbook, but from a patient.
And that is the only way medicine ever really teaches you anything.