Interpreting Ka, Tmax, and Peak Concentration

What the Peak Tells Us

After an oral dose, drug enters the systemic circulation from an absorption depot while drug is being removed from the central compartment. The concentration therefore typically rises, reaches a maximum, and then falls. Two model-derived quantities summarize that peak: $T_{\max}$, the time at which the concentration is greatest, and $C_{\max}$, the concentration at that time. Both depend on the absorption and disposition rates; neither is simply the reciprocal of one rate.

The derivation of the first-order absorption profile gives the one-compartment concentration function below. This tutorial uses that function to find and interpret its peak. Here $k_a$ is the first-order absorption rate constant and $k$ is the first-order disposition rate constant for this one-compartment model. The parameter $k$ is a model rate constant, not automatically a physiologic rate constant in a more complex compartment model.

Model and assumptions: A single oral dose $D$ is placed in an absorption depot at time zero. A constant fraction $F$ is systemically available; absorption from the depot is first order with rate constant $k_a>0$; and disposition from a one-compartment central space of apparent volume $V>0$ is first order with rate constant $k>0$. There is no lag time, and the model is linear with respect to dose. For $k_a\ne k$, the concentration at time $t\geq 0$ is $$ C(t)=\frac{FDk_a}{V(k_a-k)}\left(e^{-kt}-e^{-k_at}\right). $$

The quantities $k_a$ and $k$ have units of inverse time, such as $\mathrm{h}^{-1}$. Time $t$ and $T_{\max}$ use the corresponding time unit. If $D$ is in mg and $V$ is in L, concentration is in mg/L. The formula assumes positive rates and dose, $F>0$, and a positive volume. Under those assumptions, $C(0)=0$, $C(t)>0$ for $t>0$, and the peak concentration is positive. The concentration is not strictly positive at every time in its domain because it is zero at the instant of dosing.

Deriving the Time of the Peak

For $k_a\ne k$, differentiate the concentration with respect to time. The peak occurs at a stationary point where this derivative is zero.

\frac{dC(t)}{dt} = \frac{FDk_a}{V(k_a-k)} \left(-ke^{-kt}+k_ae^{-k_at}\right).

Setting the derivative equal to zero gives $k_ae^{-k_aT_{\max}}=ke^{-kT_{\max}}$. Taking logarithms and collecting the terms in $T_{\max}$ yields

\ln\left(\frac{k_a}{k}\right) =(k_a-k)T_{\max}, \qquad T_{\max}= \frac{\ln(k_a/k)}{k_a-k}.

This result has units of time: the logarithm is dimensionless, and the denominator has units of inverse time. It is positive for either ordering of the rates. If $k_a>k$, both the logarithm and denominator are positive; if $k_a<k$, both are negative. Thus $T_{\max}>0$ in either case.

The stationary point is the unique maximum. In fact, the derivative is zero only when $k_ae^{-k_at}=ke^{-kt}$, or equivalently when $e^{-(k_a-k)t}=k/k_a$. For $k_a\ne k$, the left side is strictly monotone in $t$, so this equation has at most one solution; the formula above gives a positive solution. The concentration starts at zero, initially rises (the right-hand derivative at zero is $FDk_a/V>0$), remains positive for $t>0$, and approaches zero as $t$ tends to infinity. Consequently, that stationary point is the peak.

Worked Example: Calculating Peak Time and Concentration

Consider a hypothetical single oral dose of $D=100$ mg, with $F=0.80$, $V=40$ L, $k_a=1.2\ \mathrm{h}^{-1}$, and $k=0.20\ \mathrm{h}^{-1}$. These values satisfy the model assumptions and have consistent time units.

First calculate the dose scale $FD/V$: $$ \frac{FD}{V}=\frac{0.80(100\ \mathrm{mg})}{40\ \mathrm{L}} =2.00\ \mathrm{mg/L}. $$ Then calculate the peak time: $$ T_{\max} =\frac{\ln(1.2/0.20)}{1.2-0.20}\ \mathrm{h} =\frac{\ln(6)}{1.00\ \mathrm{h}^{-1}} =1.792\ \mathrm{h}. $$

At the peak, the derivative condition gives $e^{-k_aT_{\max}}=(k/k_a)e^{-kT_{\max}}$. Substituting this identity into the concentration profile simplifies the peak concentration to $$ C_{\max}=\frac{FD}{V}e^{-kT_{\max}}. $$ Therefore, $$ C_{\max} =(2.00\ \mathrm{mg/L})e^{-(0.20\ \mathrm{h}^{-1})(1.792\ \mathrm{h})} =2.00(0.6988)\ \mathrm{mg/L} =1.398\ \mathrm{mg/L}. $$

The model predicts a maximum of about $1.398$ mg/L at $1.792$ h after dosing. This is the continuous-time model peak; a measured concentration at a scheduled sample time need not equal it.

Interpreting the Two Rate Constants

With $F,D,V,$ and $k$ fixed, increasing $k_a$ moves the predicted peak earlier and raises its height. The time result can be seen by writing $r=k_a/k$, so that $T_{\max}=\ln(r)/[k(r-1)]$. The function $\ln(r)/(r-1)$ decreases as positive $r$ increases: its derivative has numerator $(r-1)/r-\ln(r)$, which is negative for $r\ne1$. Hence $T_{\max}$ decreases as $k_a$ increases, with a continuous value at $r=1$.

The peak-height formula gives the companion interpretation. For $r\ne1$, $$ C_{\max}=\frac{FD}{V} \exp\left[-\frac{\ln(r)}{r-1}\right]. $$ As $r$ increases, $\ln(r)/(r-1)$ decreases, so the exponent becomes less negative and $C_{\max}$ increases. The dose scale $FD/V$ also shows that, with rates and volume fixed, increasing the systemically available dose increases the predicted peak proportionally. These are conditional model statements: changing several parameters together need not preserve the same comparison.

Worked Example: Comparing Faster and Slower Absorption

Use the same hypothetical dose scale $FD/V=2.00$ mg/L and disposition rate $k=0.20\ \mathrm{h}^{-1}$, but compare $k_a=1.2\ \mathrm{h}^{-1}$ with $k_a=0.10\ \mathrm{h}^{-1}$. Both rate constants are positive; all other model assumptions and parameters are held fixed.

For the slower absorption rate, $$ T_{\max} =\frac{\ln(0.10/0.20)}{0.10-0.20} =\frac{\ln(0.5)}{-0.10\ \mathrm{h}^{-1}} =6.931\ \mathrm{h}. $$ Its peak concentration is $$ C_{\max} =(2.00\ \mathrm{mg/L})e^{-(0.20\ \mathrm{h}^{-1})(6.931\ \mathrm{h})} =2.00(0.2500)\ \mathrm{mg/L} =0.5000\ \mathrm{mg/L}. $$

For $k_a=1.2\ \mathrm{h}^{-1}$, the first worked example gave $T_{\max}=1.792$ h and $C_{\max}=1.398$ mg/L. Under these fixed model conditions, the faster absorption rate produces an earlier, higher peak. This comparison concerns the model-predicted peak, not a claim that every observed sample from one profile must exceed every sample from the other.

When the Rates Are Equal

The formula containing $k_a-k$ cannot be directly evaluated when $k_a=k$. The concentration itself, however, has a finite limit. Taking $k_a$ toward $k$ in the first-order profile gives $C(t)=(FD/V)kte^{-kt}$. Differentiating this limiting profile gives a peak at $T_{\max}=1/k$, and substituting that time gives $C_{\max}=FD/(Ve)$.

Worked Example: Equal Absorption and Disposition Rates

Suppose a hypothetical dose has $FD/V=1.50$ mg/L and the equal rates are $k_a=k=0.40\ \mathrm{h}^{-1}$. The applicable limiting profile is $C(t)=(FD/V)kte^{-kt}$, not the expression with a zero denominator.

The peak time and height are $$ T_{\max}=\frac{1}{0.40\ \mathrm{h}^{-1}}=2.50\ \mathrm{h}, \qquad C_{\max}=\frac{1.50\ \mathrm{mg/L}}{e} =0.5518\ \mathrm{mg/L}. $$ The units are hours for the peak time and mg/L for the concentration. Both are positive, as required for a peak occurring after dosing under the stated positive-parameter assumptions.

Calculation Checks and Common Pitfalls

A practical calculation should be checked against the model's domain, units, and qualitative shape. For positive $F,D,V,k_a,$ and $k$, the concentration is nonnegative for $t\geq0$, equals zero at $t=0$, and is strictly positive for $t>0$. The calculated $T_{\max}$ and $C_{\max}$ should both be positive. If either rate or time is entered in inconsistent units—for example, $k$ in $\mathrm{h}^{-1}$ and $T_{\max}$ in minutes without conversion—the exponent and result will be wrong.

A common mistake is to assume that a negative denominator when $k_a<k$ makes the concentration negative. In that case $e^{-kt}-e^{-k_at}$ is also negative for $t>0$, so their ratio in the profile is positive. Another mistake is to substitute $k_a=k$ into the unequal-rate formula and report a division by zero. The correct approach is to use the limiting profile and its peak formulas. Near equality, direct computation of the difference of two nearly equal exponentials can also lose numerical precision; a stable implementation should use a limiting expression or numerically stable exponential-difference calculation.

Finally, the largest sampled concentration is not necessarily $C_{\max}$. If sampling times fall on either side of $T_{\max}$ but not at it, the largest recorded value can be lower than the continuous-time model peak. A grid-based search over sample times is useful for describing the observed data, but it does not replace the derivative-based model peak. A numerical maximization of the fitted concentration curve is an alternative computational check; it should agree with the analytic result when the model and units are implemented correctly.

Scope and Interpretation

These formulas apply to the specified single-dose, one-compartment, first-order absorption and disposition model without a lag time. They do not by themselves establish that the model describes a particular concentration-time data set well, nor do they account for multiple disposition phases or other absorption processes. $T_{\max}$ and $C_{\max}$ are functions of model parameters, so uncertainty in those parameters also implies uncertainty in the peak quantities. The formulas provide model-based interpretations conditional on the assumptions, rather than guarantees about every measured profile.

When using the results, first check that the selected model and parameter units are appropriate, then calculate $T_{\max}$, and finally evaluate the profile at that time or use the simplified expression for $C_{\max}$. The equal-rate limit and the positivity checks are particularly useful safeguards against algebraic and software errors.

Check Your Understanding

Answer each question, then compare your reasoning with the solutions below.

For $k_a=0.50\ \mathrm{h}^{-1}$ and $k=0.25\ \mathrm{h}^{-1}$, calculate $T_{\max}$. Is the result positive?
For $FD/V=3.00$ mg/L and $T_{\max}=2.00$ h, with $k=0.30\ \mathrm{h}^{-1}$, calculate $C_{\max}$.
What peak-time and peak-concentration formulas apply when $k_a=k=0.50\ \mathrm{h}^{-1}$ and $FD/V=2.00$ mg/L?
A learner says that the concentration must be negative when $k_a<k$, because $k_a-k$ is negative. Identify the error.
Why might the largest observed concentration be smaller than the model-predicted $C_{\max}$?

Solutions. (1) $T_{\max}=\ln(0.50/0.25)/(0.50-0.25)=\ln(2)/(0.25\ \mathrm{h}^{-1})=2.773$ h, which is positive. (2) $C_{\max}=(3.00\ \mathrm{mg/L})e^{-(0.30\ \mathrm{h}^{-1})(2.00\ \mathrm{h})}=3.00e^{-0.60}=1.646$ mg/L. (3) Use the equal-rate limit: $T_{\max}=1/(0.50\ \mathrm{h}^{-1})=2.00$ h and $C_{\max}=(2.00\ \mathrm{mg/L})/e=0.7358$ mg/L. (4) The learner considers only the denominator. For $t>0$ and $k_a<k$, the exponential difference $e^{-kt}-e^{-k_at}$ is also negative, so the profile is positive. (5) Samples may not be taken exactly at the continuous-time peak; concentrations at sampled times on either side can both be below it.