Why the kidneys' job needs a machine to replace it
Healthy kidneys constantly filter nitrogenous waste products left over from protein metabolism — chief among them urea — out of the blood and into urine. When the kidneys fail, that waste has nowhere to go and accumulates in the bloodstream, a state called uremia, which becomes toxic well before it becomes obviously symptomatic. Haemodialysis is an external, artificial substitute for that filtering job: blood is pumped out of the body, cleaned by a machine, and returned, several times a week, for as long as the patient needs it.
The dialyzer: a membrane, not a filter you can see through
The core of the machine is the dialyzer, a cartridge packed with thousands of hollow, semi-permeable fiber membranes. Blood is pumped through the inside of these fibers while a separate fluid, the dialysate, bathes the outside. The membrane's pores are sized to block blood cells and large proteins but let small solutes like urea, creatinine and excess electrolytes cross freely by diffusion, moving from the high concentration in blood to the low concentration in fresh dialysate. A separate pressure gradient across the same membrane drives ultrafiltration, pulling excess water out of the blood at the same time, which is how dialysis also manages fluid overload.
Why countercurrent flow, not parallel flow
Blood and dialysate are pumped through the dialyzer in opposite directions, not the same one. If they flowed the same way, the concentration gap between them would be largest at the inlet and shrink toward the outlet as both fluids approached equilibrium, wasting the membrane area near the outlet where almost no further diffusion could occur. Running them countercurrent keeps blood meeting progressively fresher dialysate along the entire length of the fiber, so a favourable concentration gradient exists from one end of the dialyzer to the other. This single design choice — copied directly from countercurrent heat exchangers — maximises the total urea clearance achievable from a given membrane area and given flow rates.
Single-pool kinetics: one falling exponential
The simplest useful model of what happens to blood urea during a session treats the whole body as one well-mixed compartment of volume V — the urea distribution volume, roughly equal to total body water. Under that assumption, urea concentration falls exponentially over time: C(t) = C₀ · e^(−Kt/V), where K is the dialyzer's urea clearance, essentially the volume of blood fully cleared of urea per minute, set by blood flow rate, dialysate flow rate and the membrane's own permeability. Everything in that formula — how strong the machine's clearance is, how long the session runs, how large the patient's own distribution volume is — folds into one dimensionless number.
C(t) = C0 · e^(-Kt/V) // single-pool urea kinetics Kt/V // dimensionless adequacy target, KDOQI: ≥ 1.2 per session Illustrative example: K (clearance) ≈ 200 mL/min = 0.2 L/min t (session length) = 240 min (4 hours) V (distribution vol.) = 40 L Kt/V = (0.2 x 240) / 40 = 48 / 40 = 1.2 // meets the KDOQI target Fractional reduction: C(t)/C0 = e^(-1.2) ≈ 0.30 >> blood urea falls to about 30% of its pre-dialysis value, a ~70% reduction
Kt/V as the clinical adequacy target — and its honest limits
Because Kt/V directly sets the fractional drop in urea via C(t)/C₀ = e^(−Kt/V), nephrologists use Kt/V itself, rather than any single one of K, t or V, as the standard measure of how much dialysis was actually delivered. National guidelines such as KDOQI target a single-pool Kt/V of at least 1.2 per session, usually reported through the Daugirdas second-generation formula, which corrects the simple single-pool estimate for two things the basic model ignores: ongoing urea generation during treatment, and urea rebound. That rebound exists because the body is not really one compartment — urea inside cells crosses membranes more slowly than urea in the bloodstream, so a gap opens between intracellular and extracellular urea during treatment, and once dialysis stops, urea diffuses back out of cells and blood urea creeps upward again for 30 to 60 minutes before settling. The single-pool model is a deliberate simplification, and clinical Kt/V targets are calibrated with that rebound already in mind.
Frequently asked questions
Why do blood and dialysate flow in opposite directions?
Countercurrent flow keeps fresh, low-solute dialysate meeting blood along the whole length of the fiber, so a favourable concentration gradient exists from one end to the other. With co-current (parallel) flow, blood and dialysate would equalise their solute concentration part-way along the fiber and the gradient would collapse toward the outlet, wasting the remaining membrane area and reducing total clearance for the same flow rates.
What does the Kt/V number actually mean and why is 1.2 the target?
Kt/V is dialyzer clearance (K) multiplied by session time (t), divided by the body's urea distribution volume (V). It is dimensionless and sets the fractional drop in blood urea directly, since C(t)/C0 = e^(-Kt/V). Guidelines such as KDOQI target a single-pool Kt/V of at least 1.2 per session because trials link values below that to worse outcomes, while going much higher adds diminishing benefit for added treatment time.
Why does blood urea rebound upward after dialysis ends?
The body is not one single well-mixed compartment. Blood urea is cleared quickly from the bloodstream (extracellular water), but urea inside cells (intracellular water) crosses membranes more slowly, so a concentration gap opens up during treatment. Once dialysis stops, urea diffuses out of cells to re-equalise the two compartments, and measured blood urea rises for roughly 30 to 60 minutes even though nothing is being added back.
Try it live
Everything above runs in your browser — open Haemodialysis and adjust clearance, session length and distribution volume to watch the blood urea curve and Kt/V respond in real time. Nothing is installed, nothing is uploaded, the whole model lives in one tab.
▶ Open Haemodialysis simulation