Deriving Segmental Luminal Mass Balances

Why Write the Balance in Terms of Mass?

A segmental gut model needs to specify what happens to dissolved drug in each region. Drug can enter a segment from an upstream region, be added by dissolution, leave through downstream transit, or leave the lumen by absorption. Writing each process as a separate term makes the model interpretable and provides a direct way to check that drug is not created or lost inadvertently.

The state variable here is the amount of dissolved drug mass in a segment, measured, for example, in milligrams. The balance does not itself explain how dissolution or absorption rates are generated. It takes dissolution as an input rate and represents absorption as a mass-proportional loss. This keeps the bookkeeping distinct from the mechanisms that may determine those rates.

Definition: Let \(M_i(t)\) be the dissolved drug mass in segment \(i\) at time \(t\). Let \(k_{\mathrm{tr},i}\) be its first-order transit rate constant, \(k_{\mathrm{abs},i}\) its first-order absorption rate constant, and \(R_{\mathrm{dis},i}(t)\) its dissolution input rate. The transit and absorption constants have units of inverse time; \(M_i\) has units of mass; and \(R_{\mathrm{dis},i}\) has units of mass per time.

The transit-rate convention is that \(k_{\mathrm{tr},i}M_i(t)\) is the mass leaving segment \(i\) by transit per unit time. This first-order form is consistent with representing a segment as well mixed: the amount transferred during a short interval is proportional to the mass present. The relation between a segment’s residence time and its transit rate is covered in the earlier material on transit rates from segment residence times.

Deriving the Segmental Equation

Consider a segment \(i\) over a short interval from \(t\) to \(t+\Delta t\). Denote the incoming transit mass flow rate by \(J_{i,\mathrm{in}}(t)\). During this interval, incoming transit contributes approximately \(J_{i,\mathrm{in}}(t)\Delta t\) mass, and dissolution contributes approximately \(R_{\mathrm{dis},i}(t)\Delta t\). Outgoing transit removes approximately \(k_{\mathrm{tr},i}M_i(t)\Delta t\), while absorption removes approximately \(k_{\mathrm{abs},i}M_i(t)\Delta t\).

The mass at the end of the interval is the starting mass plus these gains minus these losses. Dividing the change by \(\Delta t\) and taking the limit as \(\Delta t\) approaches zero gives the differential equation.

\frac{dM_i}{dt} = J_{i,\mathrm{in}}(t) + R_{\mathrm{dis},i}(t) - k_{\mathrm{tr},i}M_i(t) - k_{\mathrm{abs},i}M_i(t)

For an interior segment in a sequential chain, the upstream segment’s transit is its incoming flow:

J_{i,\mathrm{in}}(t)=k_{\mathrm{tr},i-1}M_{i-1}(t)

Thus the equation for an interior segment becomes:

\frac{dM_i}{dt} = k_{\mathrm{tr},i-1}M_{i-1}(t) + R_{\mathrm{dis},i}(t) - \left(k_{\mathrm{tr},i}+k_{\mathrm{abs},i}\right)M_i(t)

The first segment requires an upstream boundary input \(J_{1,\mathrm{in}}(t)\), rather than an upstream segment term. That input is left unspecified here; its formulation depends on how drug enters the modeled gut. For the final segment, \(k_{\mathrm{tr},i}M_i(t)\) is the mass flow leaving the modeled chain.

Assumptions: Each segment is represented by a single dissolved-mass state; transit and absorption losses are proportional to the mass in that segment; dissolution is supplied as a specified nonnegative rate; and there are no unlisted reactions, degradation processes, or other sources and sinks. Rate constants may be constant or time varying. If they vary with time, the same balance applies with the corresponding time-dependent values.

Worked Balances

Worked Example: One Segment With a Specified Upstream Flow

Suppose one segment contains \(8.0\) mg of dissolved drug. At the time of interest, its incoming transit rate is \(1.0\) mg/h, its dissolution input is \(0.40\) mg/h, its transit rate constant is \(0.30\) h\(^{-1}\), and its absorption rate constant is \(0.20\) h\(^{-1}\). Find the instantaneous rate of change of dissolved mass.

The balance is incoming transit plus dissolution minus outgoing transit and absorption:

\frac{dM}{dt} = J_{\mathrm{in}}+R_{\mathrm{dis}}-(k_{\mathrm{tr}}+k_{\mathrm{abs}})M

Substituting the stated values gives:

\frac{dM}{dt} = 1.0\ \mathrm{mg/h} + 0.40\ \mathrm{mg/h} - (0.30\ \mathrm{h}^{-1}+0.20\ \mathrm{h}^{-1})(8.0\ \mathrm{mg}) = -2.6\ \mathrm{mg/h}

The transit loss is \(0.30\times8.0=2.4\) mg/h, and the absorption loss is \(0.20\times8.0=1.6\) mg/h. The total losses, \(4.0\) mg/h, exceed the total inputs, \(1.4\) mg/h, by \(2.6\) mg/h. Therefore the dissolved mass is decreasing at that instant. The negative derivative does not mean the mass is negative; it describes its local direction of change.

A Two-Segment Model and Its Conservation Check

For two sequential segments, segment 1 loses mass by transit to segment 2 and by absorption. Segment 2 receives the transit outflow from segment 1, then loses mass by its own transit and absorption. If \(J_{1,\mathrm{in}}(t)\) is the input to segment 1, the equations are:

\begin{aligned} \frac{dM_1}{dt} &= J_{1,\mathrm{in}}(t)+R_{\mathrm{dis},1}(t) - (k_{\mathrm{tr},1}+k_{\mathrm{abs},1})M_1(t),\\ \frac{dM_2}{dt} &= k_{\mathrm{tr},1}M_1(t)+R_{\mathrm{dis},2}(t) - (k_{\mathrm{tr},2}+k_{\mathrm{abs},2})M_2(t). \end{aligned}

Worked Example: Calculating Both Segment Derivatives

Consider a hypothetical two-segment model at one instant. Let \(M_1=12.0\) mg and \(M_2=5.0\) mg, with no external input to segment 1 at that instant. Set \(R_{\mathrm{dis},1}=1.2\) mg/h and \(R_{\mathrm{dis},2}=0.30\) mg/h. Let \(k_{\mathrm{tr},1}=0.40\) h\(^{-1}\), \(k_{\mathrm{tr},2}=0.25\) h\(^{-1}\), \(k_{\mathrm{abs},1}=0.10\) h\(^{-1}\), and \(k_{\mathrm{abs},2}=0.20\) h\(^{-1}\). Calculate both instantaneous derivatives.

For segment 1, the outgoing transit and absorption rates are \(0.40(12.0)=4.8\) mg/h and \(0.10(12.0)=1.2\) mg/h. Therefore:

\frac{dM_1}{dt} = 0+1.2-(0.40+0.10)(12.0) = -4.8\ \mathrm{mg/h}

The transit flow entering segment 2 is \(0.40(12.0)=4.8\) mg/h. Segment 2 loses \(0.25(5.0)=1.25\) mg/h by transit and \(0.20(5.0)=1.0\) mg/h by absorption. Hence:

\frac{dM_2}{dt} = 4.8+0.30-(0.25+0.20)(5.0) = 2.85\ \mathrm{mg/h}

Segment 1 is losing dissolved mass while segment 2 is gaining it. Their combined dissolved mass is changing at \(-4.8+2.85=-1.95\) mg/h. The same result follows from the external balance: total dissolution is \(1.2+0.30=1.50\) mg/h; total absorption is \(1.2+1.0=2.20\) mg/h; and transit out of the final segment is \(1.25\) mg/h. Thus \(1.50-2.20-1.25=-1.95\) mg/h. The \(4.8\) mg/h moving from segment 1 to segment 2 is internal transfer, so it cancels from the combined balance.

Why the Conservation Check Works

Adding the two equations shows why internal transit should not be counted as a net gain or loss for the whole modeled chain. More generally, for \(m\) sequential segments, the outflow from every segment except the final one appears once as a loss and once as the next segment’s gain. Those terms cancel in the sum. If \(J_{1,\mathrm{in}}(t)\) is the external input at the first segment, the combined balance is:

\frac{d}{dt}\left(\sum_{i=1}^{m}M_i(t)\right) = J_{1,\mathrm{in}}(t) + \sum_{i=1}^{m}R_{\mathrm{dis},i}(t) - \sum_{i=1}^{m}k_{\mathrm{abs},i}M_i(t) - k_{\mathrm{tr},m}M_m(t)

This is a bookkeeping identity under the stated sequential-transfer structure. It is useful both as a derivation check and as a diagnostic: if internal transit remains as a net loss from the sum of all modeled segments, a transfer term may be missing from the receiving segment. Likewise, if the sum contains neither absorption nor final outflow, the model may be omitting real exits from the modeled luminal states.

Worked Example: Auditing a Two-Segment Balance

Use the two-segment values in the previous example. The initial combined dissolved mass is \(12.0+5.0=17.0\) mg. The combined derivative computed from the segment equations is \(-1.95\) mg/h. Independently, the external balance gives:

\text{dissolution input} - \text{absorption loss} - \text{final transit loss} = 1.50-2.20-1.25 = -1.95\ \mathrm{mg/h}

The two calculations agree, so the segment equations conserve mass across the internal boundary at this instant. The net derivative is not the total amount absorbed: absorption is a loss of \(2.20\) mg/h, while dissolution adds \(1.50\) mg/h and final transit removes another \(1.25\) mg/h. Each quantity has a distinct interpretation and should be reported separately when it is needed.

Assumptions, Diagnostics, and Alternatives

The balance is simple, but the assumptions behind its terms deserve attention. The first-order transit expression assumes that a segment’s outflow is proportional to its current dissolved mass. A single well-mixed state also treats the segment as having no internal spatial gradient in dissolved mass. These are modeling choices, not universal features of gastrointestinal transport. The balance further assumes that all transfers and losses included in the model have been listed; omitting a source or sink makes the resulting equation incomplete even if the algebra is correct.

Check dimensions. Every term in \(dM_i/dt\) must have units of mass per time. A rate constant in h\(^{-1}\) must multiply a mass in mg to produce mg/h.
Check signs and destinations. Transit is negative in the segment it leaves and positive in the receiving segment. Absorption is a loss from the luminal dissolved state.
Check nonnegative states. With nonnegative inputs, nonnegative starting masses, and nonnegative loss rates, the continuous-time balance should not drive a state below zero. A numerical solver that produces negative masses may be using steps that are too large or an unsuitable method.
Check the whole-chain balance. Summing segment equations should cancel internal transfers and leave only external inputs, absorption, and outflow from the final segment.
Check what the state represents. These equations track mass, not concentration. If the intended state is concentration, changes in luminal volume also affect its dynamics and cannot be ignored by simply relabeling \(M_i\).

One alternative is to represent the gut with a single lumped luminal state. That can be simpler, but it cannot distinguish where transit, dissolution, or absorption occurs. A segmented model preserves regional bookkeeping while still making a well-mixed approximation within each segment. More spatially resolved descriptions may be appropriate when within-segment gradients matter, but they require additional states and assumptions. The mass-balance form remains useful in each case: gains and losses must still be assigned consistently.

Common pitfall: It is incorrect to subtract \(k_{\mathrm{tr},i}\) or \(k_{\mathrm{abs},i}\) directly from \(M_i\), or to treat either rate constant as a mass flow. A rate constant has units of inverse time; the corresponding mass flow is \(kM_i\), with units of mass per time. Also, the transit flow leaving one segment must enter the next if that next segment is included in the model.

Check Your Understanding

Use the stated mass-balance assumptions. For each question, compare your reasoning with the short solution below.

A segment contains \(6.0\) mg. Its incoming transit rate is \(0.50\) mg/h, its dissolution input is \(0.20\) mg/h, \(k_{\mathrm{tr}}=0.25\) h\(^{-1}\), and \(k_{\mathrm{abs}}=0.10\) h\(^{-1}\). What is \(dM/dt\)?
In a two-segment chain, what happens to the transit flow from segment 1 when the two segment equations are added?
Why is \(k_{\mathrm{abs}}M_i\), rather than \(k_{\mathrm{abs}}\) alone, the absorption loss rate?
If all dissolution and external input are zero, what terms remain in the whole-chain mass balance?

Solutions. (1) The losses are \((0.25+0.10)(6.0)=2.10\) mg/h, while inputs total \(0.50+0.20=0.70\) mg/h. Thus \(dM/dt=0.70-2.10=-1.40\) mg/h. (2) The flow is a loss from segment 1 and a gain to segment 2, so it cancels in the sum. (3) \(k_{\mathrm{abs}}\) is a rate constant in inverse time; multiplying by the dissolved mass gives a loss rate in mass per time. (4) The combined derivative is the negative of total absorption and transit out of the final segment: \(-\sum_i k_{\mathrm{abs},i}M_i-k_{\mathrm{tr},m}M_m\).