Deriving the Integrated Michaelis-Menten Concentration-Time Equation

From an Elimination Rate to a Concentration-Time Relationship

The Michaelis-Menten elimination rate law describes how elimination rate depends on concentration at an instant. A concentration-time profile asks a different question: how does that rate accumulate over time as concentration changes? Integrating the one-compartment mass balance gives an exact implicit relationship for a bolus dose. It is implicit because concentration appears both outside and inside a logarithm, so finding concentration at a specified time generally requires numerical solution.

The derivation below assumes the Michaelis-Menten elimination rate law established in the earlier material on Michaelis-Menten elimination. The parameters \(V_{\max}\) and \(K_m\) are treated as fixed for the individual whose concentration trajectory is being described.

Model and assumptions: After an instantaneous intravenous bolus of dose \(D>0\), drug is in one well-mixed compartment of constant volume \(V>0\). Concentration is \(C(t)\), with \(C(0)=C_0=D/V\). Elimination is the only process changing the amount in the compartment, and its rate is \(V_{\max}C/(K_m+C)\), where \(V_{\max}>0\) is the maximum elimination rate and \(K_m>0\) is the concentration at which the rate is half of \(V_{\max}\). There is no input after the bolus.

With amount \(A(t)=VC(t)\), the mass balance is \(dA/dt=-V_{\max}C/(K_m+C)\). Since \(V\) is constant, \(dA/dt=V\,dC/dt\). Consequently, concentration decreases according to

\frac{dC}{dt}=-\frac{V_{\max}}{V}\frac{C}{K_m+C}.

This differential equation describes the instantaneous rate of change. To obtain elapsed time between \(C_0\) and a later concentration \(C\), separate variables and integrate. The logarithm must contain a dimensionless ratio, here \(C/C_0\).

\begin{aligned} \frac{K_m+C}{C}\,dC&=-\frac{V_{\max}}{V}\,dt,\\ \int_{C_0}^{C}\left(\frac{K_m}{u}+1\right)\,du &=-\frac{V_{\max}}{V}\int_0^t ds,\\ K_m\ln\left(\frac{C}{C_0}\right)+(C-C_0) &=-\frac{V_{\max}}{V}t. \end{aligned}

Rearranging gives the integrated concentration-time relationship. The integration variable \(u\) is just a dummy concentration; \(C\) is the concentration at elapsed time \(t\).

Integrated bolus relationship: Under the stated assumptions, for \(t\geq 0\) and \(0<C\leq C_0\),

t=\frac{V}{V_{\max}}\left[(C_0-C)+K_m\ln\left(\frac{C_0}{C}\right)\right], \qquad C_0=\frac{D}{V}.

The expression gives time as a function of concentration. To find concentration from time, solve this equation for \(C\).

Why the Concentration Solution Is Unique

For a specified \(t>0\), define a residual function by moving all terms to one side. A concentration solving the integrated relationship is exactly a root of this function.

H(C)=C_0-C+K_m\ln\left(\frac{C_0}{C}\right)-\frac{V_{\max}t}{V}.

For positive concentration, its derivative is \(H'(C)=-1-K_m/C<0\). Thus \(H\) is strictly decreasing. At \(C=C_0\), \(H(C_0)=-V_{\max}t/V<0\), while \(H(C)\) tends to positive infinity as \(C\) approaches zero from above, because \(\ln(C_0/C)\) grows without bound. Continuity and strict decrease therefore guarantee exactly one root in \((0,C_0)\). At \(t=0\), the solution is \(C=C_0\).

This result justifies a bracketed numerical method such as bisection: start with positive concentrations on either side of the root and repeatedly narrow the bracket. Newton's method can converge faster, using

C_{\text{new}}=C-\frac{H(C)}{-1-K_m/C},

but an unconstrained Newton step can leave the physically meaningful interval \((0,C_0)\). In practice, retain a bracket or reject steps outside it. In either method, check the final concentration by substituting it into the original integrated equation, not merely by trusting the solver's reported convergence.

Worked Calculations

Worked Example: Time to Fall from 4 to 2 mg/L

Consider a hypothetical one-compartment bolus model with \(V=10\) L, \(V_{\max}=20\) mg/h, \(K_m=2\) mg/L, and initial concentration \(C_0=4\) mg/L. Find the time at which concentration reaches \(2\) mg/L. The dose consistent with the initial concentration is \(D=VC_0=(10\text{ L})(4\text{ mg/L})=40\) mg.

Substitute the values into the integrated relationship:

\begin{aligned} t&=\frac{10\text{ L}}{20\text{ mg/h}} \left[(4-2)\text{ mg/L}+(2\text{ mg/L})\ln(4/2)\right]\\ &=0.5\text{ L h/mg}\left[2+2(0.693147)\right]\text{ mg/L}\\ &=0.5(3.386294)\text{ h}\\ &\approx 1.693\text{ h}. \end{aligned}

The concentration has fallen to half its initial value after about 1.693 h under this hypothetical model. In particular, \(\ln(4/2)=\ln 2\approx0.6931\); the factor \(K_m=2\) multiplies that logarithm, giving \(1.3863\) mg/L inside the bracket.

Worked Example: Concentration After Two Hours

Use a hypothetical model with \(V=10\) L, \(V_{\max}=20\) mg/h, \(K_m=2\) mg/L, and \(C_0=8\) mg/L. Find concentration at \(t=2\) h. Here \(D=VC_0=80\) mg, and the parameters meet the positivity assumptions.

The equation to solve is

H(C)=8-C+2\ln(8/C)-\frac{(20)(2)}{10} =8-C+2\ln(8/C)-4=0,

where concentrations are in mg/L and the time term \(V_{\max}t/V\) is \(4\) mg/L. For example, evaluating the residual at two trial concentrations gives

\begin{aligned} H(4.95)&=8-4.95+2\ln(8/4.95)-4\approx 0.0101,\\ H(4.96)&=8-4.96+2\ln(8/4.96)-4\approx -0.00393. \end{aligned}

Because the residual changes sign and has only one root, the solution lies between 4.95 and 4.96 mg/L. Bisection or a safeguarded Newton solve gives \(C\approx4.9572\) mg/L. Substitution checks the result:

8-4.9572+2\ln(8/4.9572)\approx 4.0000\text{ mg/L} =\frac{V_{\max}t}{V}.

Thus the predicted concentration after two hours is approximately \(4.957\) mg/L. Reporting more digits would imply greater precision than is useful for these hypothetical inputs.

Worked Example: Solving at a Different Specified Time

For the hypothetical parameters \(V=10\) L, \(V_{\max}=20\) mg/h, \(K_m=2\) mg/L, and \(C_0=4\) mg/L, find the concentration at \(t=1\) h. As before, \(D=40\) mg, and the assumptions of positive parameters, constant volume, and no post-bolus input hold.

Here \(V_{\max}t/V=2\) mg/L. The root equation is

H(C)=4-C+2\ln(4/C)-2=0.

At \(C=2.75\) mg/L, \(H(2.75)=2-2.75+2\ln(4/2.75)\approx-0.000613\). Since \(H'(C)=-1-2/C\), one Newton update is

C_{\text{new}}=2.75-\frac{-0.000613}{-1-2/2.75} \approx 2.7496\text{ mg/L}.

The root is therefore about \(2.750\) mg/L to four significant figures. It lies below \(C_0=4\) mg/L, as expected for elimination without further input; substitution into \(H\) gives a residual close to zero.

Checks, Alternatives, and Limitations

Several quick checks can catch algebra, unit, or implementation errors. At \(t=0\), the equation must return \(C_0\). For \(t>0\), the unique solution must be positive and less than \(C_0\); the right-hand side for time is positive for \(0<C<C_0\). Units also have to balance: the bracket has concentration units, and \(V/V_{\max}\) has units of time per concentration, so their product is time. The logarithm's argument \(C_0/C\) has no units.

The integrated equation is exact for this model and avoids approximating the trajectory by small time steps. A numerical ordinary differential equation solver is an alternative: it repeatedly advances the differential equation in time and can be useful when the model has additional processes or does not admit this simple integral. For the present model, the implicit equation plus a one-dimensional root solver is direct and provides a monotone root with a clear bracket. Neither computational approach repairs a misspecified structural model.

The equation assumes instantaneous mixing into one compartment, constant volume, fixed \(V_{\max}\) and \(K_m\), and elimination as the only process after the bolus. Absorption, distribution between compartments, time-varying parameters, or other elimination pathways would change the governing equation. Numerical warning signs include a negative concentration, a value above \(C_0\) during elimination-only decay, a non-small equation residual, or a solver that was not constrained to the valid root interval. These are reasons to check units, the parameter mapping, the logarithm, and the model assumptions.

Common pitfall: Treating the time equation as if \(C\) could be isolated by rearranging only the linear term omits its appearance inside \(\ln(C_0/C)\). The result is not an explicit concentration formula. Instead, define \(H(C)\), solve \(H(C)=0\) within \(0<C\leq C_0\), and verify the original equation and units.

Set the initial condition.
For a bolus, calculate \(C_0=D/V\) using dose and volume in consistent units.

Form the root equation.
At the desired time \(t\), set \(H(C)=C_0-C+K_m\ln(C_0/C)-V_{\max}t/V\).

Find and verify the root.
Bracket it in \((0,C_0)\), solve numerically, then check its range and substitution residual.

Check Your Understanding

Try each question, then compare with the solutions below.

For \(D=30\) mg and \(V=6\) L, what is the initial concentration?
Why can the root equation have at most one positive solution less than \(C_0\) at a specified positive time?
For \(V=8\) L, \(V_{\max}=16\) mg/h, \(K_m=1\) mg/L, and \(C_0=3\) mg/L, write the equation for the time at which \(C=1.5\) mg/L.
Name two checks that should be applied to a computed concentration at a positive time.

Solutions. (1) \(C_0=D/V=30/6=5\) mg/L. (2) The derivative of the residual is \(H'(C)=-1-K_m/C<0\), so it is strictly decreasing and cannot have two roots. (3) Substitute into the integrated equation: \(t=(8/16)[(3-1.5)+\ln(3/1.5)]\) h \(=0.5[1.5+0.693147]\) h \(\approx1.097\) h. (4) Check that \(0<C<C_0\), and substitute the result into the original relationship to confirm a small residual; also check that the units and logarithm argument are valid.

Deriving the Integrated Michaelis-Menten Concentration-Time Equation

What You'll Learn