Equilibrium Partitioning Versus Finite-Rate Exchange

Two Different Questions: Where and How Fast?

A tissue model may specify that a drug has a preferred equilibrium concentration in one compartment relative to another. That relationship answers where the concentrations tend in the long run; it does not say how quickly the drug crosses the membrane. A finite membrane permeability answers the second question. Keeping these roles separate is essential when interpreting tissue concentration profiles: the same eventual partitioning can accompany very different transient behavior.

Consider an extracellular compartment with volume \(V_e\) and concentration \(C_e\), and an intracellular compartment with volume \(V_i\) and concentration \(C_i\). The compartments exchange drug across a membrane. For the derivation and examples below, assume fixed positive volumes, a closed system with no input or elimination, and a constant, positive equilibrium partition ratio \(K\). Assume also that exchange is linear in the concentration difference expressed relative to that ratio. These assumptions isolate membrane exchange; they are not a claim that every tissue has constant volumes or no other processes.

For this two-compartment model, define the net flux from extracellular to intracellular space as \(J=PS(C_e-C_i/K)\). A positive \(J\) means drug is moving into the intracellular compartment. The factor \(1/K\) makes zero flux correspond to the specified equilibrium relationship \(C_i=K C_e\). With this sign convention, a concentration difference from equilibrium can drive flux in either direction.

The model assumes that the concentration-based flux law applies throughout the interval being considered, and that \(PS\) and \(K\) do not change during that interval. It also treats each compartment as well mixed, so a single concentration represents each compartment. If binding, active transport, changing membrane properties, or spatial gradients are important, this simple model may not adequately describe the observed profiles.

Mass Conservation and the Equilibrium Concentrations

The amount of drug in the two compartments is \(A=V_e C_e+V_i C_i\). In this closed system, the exchange flux leaves one compartment at exactly the rate it enters the other:

Thus, \(A\) remains constant. At equilibrium, \(J=0\), so \(C_i=K C_e\). Combining this relationship with the conserved amount gives \(A=(V_e+K V_i)C_e\). The equilibrium concentrations are therefore:

These expressions show what sets the long-time concentrations under the stated closed-system assumptions: the total amount, the compartment volumes, and the equilibrium ratio. The exchange capacity \(PS\) does not appear. Changing \(PS\) alone cannot change these equilibrium concentrations; it changes how long the model takes to approach them.

Deriving the Approach to Equilibrium

To describe the transient, define \(q=C_e-C_i/K\), the concentration difference that drives the modeled flux. At equilibrium \(q=0\). Differentiating \(q\) and substituting the two compartment equations gives a single differential equation:

For constant positive \(PS\), \(V_e\), \(V_i\), and \(K\), this equation has the exponential solution below. The coefficient \(\lambda\) has units of inverse time.

Because \(A\) is conserved, the deviations of \(C_e\) and \(C_i\) from their equilibrium concentrations must offset one another in amount. Combining that constraint with \(q=C_e-C_i/K\) yields the individual concentrations:

For example, if \(q(0)>0\), then \(C_e\) begins above its equilibrium value and \(C_i\) below its equilibrium value. As \(q(t)\) decays toward zero, the extracellular concentration falls and the intracellular concentration rises. If \(q(0)<0\), the directions reverse. This is a relaxation toward the same equilibrium partition relationship, not a changing value of \(K\).

The exponential form follows by separating variables when \(q\ne 0\): \(dq/q=-\lambda\,dt\). Integrating from 0 to \(t\) gives \(\ln|q(t)|-\ln|q(0)|=-\lambda t\), and exponentiating gives \(q(t)=q(0)e^{-\lambda t}\). If \(q(0)=0\), the system is already at equilibrium, \(J=0\), and the same solution gives \(q(t)=0\) throughout.

Fast Exchange: Equilibrium Is Approached Quickly

When \(PS\) is large relative to the compartment volumes and observation period, \(\lambda\) is large and the exponential term decays rapidly. After a short transient, concentrations are close to their equilibrium values. An instantaneous-equilibrium approximation sets \(C_i=K C_e\) at all modeled times; it can be useful when exchange is sufficiently fast for the question being asked. It is not the same as saying that \(K\) itself is large.

Slow Exchange: The Long-Time Ratio May Be Far Away

If \(PS\) is very small, little exchange occurs over a fixed observation interval. The equilibrium ratio still defines the eventual state for every positive \(PS\) in this closed model, but it may be a poor description of concentrations measured before that state is approached. At exactly \(PS=0\), there is no exchange: each compartment retains its initial concentration, and the system does not move toward the partition relationship unless it already satisfies it.

Interpreting the Limits and Checking the Model

The limiting cases clarify the distinction between equilibrium and rate. For any positive finite \(PS\), the model gives \(q(t)\to0\) as \(t\to\infty\), so the concentrations tend toward \(C_i=K C_e\). As \(PS\to\infty\), the decay is arbitrarily rapid; in the idealized limit, an initial mismatch disappears immediately after time zero. As \(PS\to0\), the rate of change tends to zero. At \(PS=0\), an initially unequal partition remains unequal indefinitely. Thus, a zero-permeability model and a very-low-permeability model can look similar over a short interval but differ in their long-time behavior.

A useful diagnostic is to check whether a proposed trajectory respects the assumptions and mathematical constraints. In the closed model, total amount \(V_e C_e+V_i C_i\) must remain constant. The sign of \(J\) must agree with \(q\): if \(C_e>C_i/K\), flux is inward, so \(C_e\) decreases and \(C_i\) increases. For nonnegative initial concentrations and positive parameters, a computed trajectory should not produce negative concentrations. A violation can indicate a sign error, inconsistent volumes, a unit mismatch, or use of the formula outside its assumptions.

Units provide another check. Since \(J\) is amount per time and \(C_e-C_i/K\) is amount per volume, \(PS\) must have units of volume per time. Then \(PS(1/V_e+1/(K V_i))\) has units of inverse time, as required for the exponent \(-\lambda t\) to be dimensionless. The partition ratio \(K\) is dimensionless when both concentrations use the same units.

Perfusion-limited tissue models use a different simplification: they treat tissue and blood as equilibrating sufficiently rapidly for the relevant model purpose. A permeability-limited model retains a finite membrane flux when that assumption is not suitable. Neither approach makes the partition ratio a rate parameter. The choice between them depends on whether finite exchange delay matters for the question and data at hand.

Common Mistake

A frequent error is to treat \(K\) as a measure of how quickly drug enters the cell. For example, reasoning that a larger \(K\) must mean faster intracellular accumulation confuses a concentration ratio at zero net flux with a kinetic property. In this model, changing \(K\) changes the equilibrium condition and also appears in the mathematical expression for \(\lambda\), but the rate still depends on \(PS\), volumes, and the specified exchange law. A large partition ratio does not by itself establish a large permeability or rapid equilibration.

A related error is to impose \(C_i=K C_e\) while also claiming to model a substantial transient intracellular lag. The algebraic equality removes that lag by assumption. To represent delayed equilibration, retain the differential equations and finite flux; to use instantaneous equilibrium, state that the approximation is intended and check that the unresolved transient is unimportant for the application.