These are the physiological effects of infusing three ampoules of concentrated potassium chloride into a patient.

What am I giving?

These 6 ampoules of potassium chloride contains 60mmol of potassium and 60 mmol of chloride.

The infusion of 60mmol of concentrated potasium chloride

The fate of the electrolytes and the fluids

Boring forgotten chloride ends up trapped in the extracellular fluid. This much is clear. The fate of potassium is more interesting.

Why doesn't the potassium rise by over 1mmol/L?

One must again turn to the ancient origins of physiology, in the men-wore-hats days of 1938. Winkler and Smith anaesthetised several dogs and infused them with potassium, then measured their jugular venous potassium to establish its apparent volume of distribution (and thereby to determine whether any of it has become intracellular).

Turns out the volume of distribution for potassium (at least among dogs marinaded in morphine) is about 55% to 75% of body mass, which means it distributed roughly equally into all body fluid compartments. And it did this quickly, because blood samples drawn even 15 minutes after the infusion already give these numbers. And yes, Winkler and Smith did complain about the difficulty of measuring potassium in context of its rapid excretion. They concluded that potassium must distribute equally into the total body water and that the apparent volume of distribution (which exceeds the total dog-body water volume) must be accounted for by potassium loss into the urine.

So... if 60 mmol went into 43 litres, a 1.4 mmol rise per litre should be experienced, right? It would make sense, but it certainly never seems to do that in practice. Why is that, you ask?

What actually happens to the infused potassium?

Some of it becomes intracellular, some of it becomes extracellular, and then some of the extracellular potassium is excreted in the urine.
We are all aware that it is not confined to the extracellular fluid. If it was, its concentration kinetics would look like this:

behaviour of potassium concentration after infusion if it were confined to the extracellular fluid

However, from experiments on anuric patients (in whom the confusing renal excretion is not an issue) we know it is more like this:

behaviour of potassium concentration after infusion in absence of renal excretion
rate of potassium distribution out of the ECF

The initial curvature of the concentration over time relationship demonstrates that as it is being infused, some of the potassium is leaving the extracellular fluid (otherwise, the curve would be linear). After the inflection point, the curve does become linear, which suggests that the rate of potassium distribution out of the ECF has plateaued.

So; after the end of the infusion, there has been some loss of potassium out of the ECF and into the intracellular fluid.

After that, the loss continues, and at about 2 hours after the end of the infusion a steady state has been reached, at which the potassium is no longer moving between compartments (this is the concentration plateau).

change in potasium concentration depending on preinfusion levels

According to the abovementioned study with the anuric people, at this stage the amount of potassium remaining in the extracellular fluid ranged between 7.6% and 36.3% of the initial infused amount. This range was among patients with a different baseline potassium - those with an already high potassium had less uptake into their cells. The hypokalemic patients had more uptake, and consequently ended up with a lower serum potassium after the infusion, almost as if they were replacing some sort of intracellular deficit.

In summary, and working from rounded figures and rough estimates:

If your hypokalemic patient has no kidneys, about 10% of the potassium you infuse will end up in the extracellular fluid. The administered potassium will be internalised by the cells in order to address the intracellular potassium shortage. No better models for this exists than the DKA patients, whose potassium deficit may be around 5mmol/Kg, and who are notoriously potassium-thirsty.

If your hyperkalemic patient has no kidneys, about 30% will end up in the extracellular fluid

The more extracellular potassium there is, the more equally infused potassium will be distributed into total body water; i.e. the volume of distribution will resemble the total body water volume, just like in the early Winkler and Smith experiments.

To further complicate issues, although hyperkalemia decreases cellular potassium uptake in these people, chronic hyperkalemia in general results in the expression of more NA+/K+ATPase pumps on the skeletal muscle, conferring to it a greater capacity to act as a potassium sink. Experiments on rats fed a high potassium diet and then challenged with a lethal potassium load have shown that this is the case (i.e. they lived).

What happens to the infused potassium if my patient DOES have working kidneys?

The renal handling of potassium is a topic all by itself, and rests comfortably far from the scope of this puny applied physiology page. All we can say about it is that during the 3 hour long infusion of 60mmol of KCl, the renal potassium excretion will increase, and it will be influenced by various things, such as the flow through (and sodium delivery to) the distal nephron, the aldosterone levels, the plasma potassium levels, and dietary potassium intake.

The acute change (i.e. within these first hours) is predominantly due to the increased plasma potassium, because this increases the potassium content of the tubular fluid (it being freely filtered). It takes the aldosterone system a while to activate fully, so it may not be a factor in the short term. Also, seeing as you are neither administering a sodium load nor a fluid challenge, the distal nephron does not experience a change of flow or sodium delivery.

But how much, exactly, is lost in the urine during the infusion? We can try to estimate it, but it relies on so many factors that there is no way to accurately guess what the potassium excretion is going to be like in any given patient. All we can say is that under normal conditions, obligate daily urinary potassium loss is about 20-40mmol. This is roughly 1-2 mmol/hr. Of course, the conditions are never normal. A patient on vigorous frusemide therapy can lose as much as 200-300mmol per day, which means 10-15mmol/hr.

Changes in initial conditions: distribution of the major electrolytes

Changes in compartment volumes and osmolality in response to the infusion of 60mmol of concentrated potassium chloride

Let us assume 10% of the potassium will remain extracellular, with a low-ish pre infusion potassium level of 3.5  and unimpaired renal excretion. Lets assume the kidneys are doing their job, and conserving potassium in this scenario.

So, of the 60mmol infused, 57mmol remain after 3 mmol is excreted renally, and thus 5.7mmol remain in the extracellular fluid.

This means the serum potassium will increase by 0.4 mmol/L. up to 3.9mmol/L.

Lets peg the extracellular chloride at a flat 100mmol/L.  With the addition of 60mmol of chloride to the extracellular fluid, this concentration would rise to 104mmol/L. That’s even more than normal saline; remember that KCl infusion mixture is highly concentrated, and practically no water is administered along with it.

Change to the initial conditions: distribution of volume

I know I promised not to talk about the laughably small fluid shifts, but I feel compelled to. The osmolality of the extracellular fluid increases out of proportion to that of the intracellular fluid. When the water distributes between these compartments, it will distribute itself in such a way as to achieve isotonicity, at around 282.4 mmol/L. Because extracellular ions are staying extracellular, in order to achieve this isotonicity an additional 29 ml of water must move into the intravascular space, 86ml into the extravascular space, which leaves the intracellular compartment  115ml dryer.

Osmoreceptor response

There is no appreciable osmoreceptor response; the osmolality of the compartments only increases by 2.4mmol, less than the required 1%.

Baroreceptor response

The baroreceptors never even flinch at the miniscule fluid movement which occurs.

References, as always;

Winklet A.W, Smith P.K. The apparent volume of distribution of potassium injected intravenously.J. Biol. Chem. 1938 124: 589-598.

GINSBURG JM, WILDE WS: Distribution kinetics of intravenous radiopotassium. Am J Physiol 179:63-75, 1954

Richard H Sterns1, Peter U Feig1, Martin Pring1, Joseph Guzzo1 and Irwin Singer1 Disposition of intravenous potassium in anuric man: A kinetic analysis Kidney International (1979) 15, 651–660; doi:10.1038/ki.1979.85

Gerhard Giebisch Renal potassium transport: mechanisms and regulation Am J Physiol Renal Physiol 274:F817-F833, 1998.

Blachley JD, Crider BP, Johnson JH. Extrarenal potassium adaptation: role of skeletal muscle. Am J Physiol. 1986 Aug;251(2 Pt 2):F313-8.

Giebisch G, Wang W.. Potassium transport: from clearance to channels and pumps. Kidney Int. 1996 Jun;49(6):1624-31

Badoe EO, Appeadu-Mensah W, Hesse A, Maddy SO.The daily water, sodium and potassium excretion in urine of Ghanaian children aged 5 to 12 years. West Afr J Med. 2005 Jul-Sep;24(3):231-3. - This article was exactly what I wanted; its surprising how hard it was to find experimental urinary potassium excretion data in Caucasian adults. We do know that there are racial differences in electrolyte excretion after all.

Abraham Aviv, Norman K. Hollenberg, Alan Weder Urinary Potassium Excretion and Sodium Sensitivity in Blacks Hypertension. 2004; 43: 707-713

And if you have a couple of spare hours, reading Dileep N. Lobo's thesis on fluid physiology will be an ideal way to spend them. It contains beautiful digressions. To wit, when discussing the effects of starvation and injury on fluid balance, Lobo muses "Life began in the sea and the intracellular environment of early life forms was isotonic with the external environment, as these unicellular organisms had no means of regulating the internal osmotic pressure"... and so on.