Why Separate Adipose Size From Its Perfusion?
Obesity can substantially expand adipose tissue, but the size of that tissue and the blood flow it receives are different physiological inputs. A larger adipose compartment does not, by itself, establish how much additional blood reaches it. Nor does an increase in total blood flow necessarily mean that each unit of adipose tissue is as well perfused as before. PBPK parameterization should therefore represent tissue amount and perfusion separately.
This distinction matters because blood flow contributes to drug delivery and removal in a perfusion-limited tissue model. A model that increases adipose volume while leaving total adipose flow unchanged implies a different flow per unit tissue than one that increases flow in proportion to volume. Those assumptions can produce different concentration-time behavior even when they use the same expanded tissue volume.
The earlier material on selecting body-size descriptors provides context for choosing a size measure in an obese adult. Here the focus is narrower: once adipose amount has been specified, how should adipose volume and blood flow be encoded without conflating them?
Definitions and a Scaling Model
Total flow \(Q_{\mathrm{ad}}\) has units such as L/min. Volume-specific perfusion \(q_{\mathrm{ad}}\) has units min\(^{-1}\), because L/min divided by L gives min\(^{-1}\). These quantities should not be given the same name or substituted for each other. PBPK tissue blood flow is often supplied as a total flow, whereas the flow relative to tissue volume helps describe how perfusion changes as the compartment expands.
A simple way to encode the relationship between expanded tissue volume and total flow is a power law. Let subscript 0 denote a reference physiology, and define \(r=V_{\mathrm{ad}}/V_{\mathrm{ad},0}\). Then:
The exponent \(\beta\) is a modeling choice to be supported by physiological data or an explicit modeling rationale; it is not a universal biological constant. When \(\beta=1\), total flow rises in proportion to volume and volume-specific perfusion is unchanged. When \(\beta<1\), total flow rises less than volume, so specific perfusion falls. When \(\beta>1\), total flow rises more than volume, so specific perfusion rises. A value of \(\beta=0\) represents fixed total flow as volume changes.
A mass-based version can be used if mass is the available size measure. If adipose density is held constant across the reference and target physiology, then \(V_{\mathrm{ad}}/V_{\mathrm{ad},0}=M_{\mathrm{ad}}/M_{\mathrm{ad},0}\), so the same exponent can be applied to the mass ratio. If density is not assumed constant or volumes are available directly, use the volume ratio rather than silently treating mass and volume as interchangeable.
Why Specific Perfusion Changes: A Short Derivation
Substitute the flow-scaling equation and \(V_{\mathrm{ad}}=rV_{\mathrm{ad},0}\) into the definition of volume-specific perfusion. The reference perfusion is \(q_{\mathrm{ad},0}=Q_{\mathrm{ad},0}/V_{\mathrm{ad},0}\), so:
Thus the change in specific perfusion is determined by the difference between the flow-scaling exponent and the volume-scaling exponent, which is one in this parameterization. If volume doubles and \(\beta=0.5\), then \(q_{\mathrm{ad}}/q_{\mathrm{ad},0}=2^{-0.5}\), approximately 0.707. Total flow has increased, but perfusion per unit tissue volume has declined by about 29.3%. This result follows from the parameterization; it is not evidence that a particular value of \(\beta\) is correct for every population.
If density is constant and the model is expressed in terms of mass, the same derivation gives \(f_{\mathrm{ad}}/f_{\mathrm{ad},0}=r^{\beta-1}\). The equality of the relative changes in \(f_{\mathrm{ad}}\) and \(q_{\mathrm{ad}}\) depends on the constant-density assumption.
Worked Example: Doubling Adipose Volume With Subproportional Flow
Consider a hypothetical reference adipose volume of 25.0 L receiving 0.250 L/min of blood. In a target physiology, let volume be 50.0 L and suppose the chosen flow-scaling exponent is \(\beta=0.500\). Calculate target total flow and volume-specific perfusion, using the unrounded flow in the perfusion calculation.
The volume ratio is \(r=50.0/25.0=2.00\). The parameterized total flow is:
The reference specific perfusion is \(q_{\mathrm{ad},0}=0.250/25.0=0.0100\ \mathrm{min}^{-1}\). At the target volume:
The ratio is \(0.0070711/0.0100=0.7071\), consistent with \(r^{\beta-1}=2.00^{-0.500}=0.7071\). Total flow has increased by a factor of \(\sqrt{2}\), while volume has doubled; therefore, flow per unit volume has decreased. Rounding the flow to 0.354 L/min before dividing would give 0.00708 min\(^{-1}\), a small rounding difference. Carrying the unrounded value avoids that avoidable discrepancy.
Choosing and Checking the Parameterization
The scaling relationship is useful because it makes a key assumption explicit, but it does not select the assumption for the modeler. Consider the alternatives carefully: holding total flow fixed assumes no increase in adipose flow despite expansion; proportional scaling assumes unchanged flow per unit volume; and a subproportional or superproportional exponent represents a decrease or increase in specific perfusion, respectively. When suitable physiological measurements are available, directly measured target flow and volume can be preferable to imposing a generic exponent.
| Approach | Assumption | Implication for specific perfusion |
|---|---|---|
| Fixed total flow | \(\beta=0\); total flow does not change with volume | Falls as volume increases |
| Proportional scaling | \(\beta=1\); total flow changes in direct proportion to volume | Remains constant |
| Power-law scaling | \(\beta\) is specified or estimated for the intended application | Falls for \(\beta<1\), is constant for \(\beta=1\), and rises for \(\beta>1\) |
| Direct target measurements | Target volume and flow are available and considered applicable | Calculated as measured flow divided by measured volume |
The power law assumes positive reference and target volumes, positive reference flow, and comparable definitions of adipose tissue and blood flow across the two physiologies. It also assumes that one exponent is an adequate description over the volume range being modeled. If the available data concern only a narrow range, extrapolating the same exponent to extreme expansion may be unjustified. If adipose composition or density changes, a mass ratio may not accurately represent the volume ratio.
These assumptions should be checked alongside the arithmetic. Confirm that the reference flow and volume refer to the same tissue definition; use consistent units; verify that target volume and flow are positive; and calculate \(q_{\mathrm{ad}}\) from the final \(Q_{\mathrm{ad}}\) and \(V_{\mathrm{ad}}\). A flow value reported in L/min cannot be divided by a volume reported in mL without converting units. Likewise, a total flow should not be entered into a model field that expects flow per unit tissue volume.
Worked Example: Comparing Flow Assumptions for the Same Expansion
For a hypothetical reference adipose volume of 30.0 L and total flow of 0.300 L/min, consider expansion to 45.0 L. The volume ratio is \(r=1.50\), and reference perfusion is \(0.300/30.0=0.0100\ \mathrm{min}^{-1}\). Compare fixed total flow, proportional flow, and \(\beta=0.500\) scaling.
With fixed total flow, \(\beta=0\), so \(Q_{\mathrm{ad}}=0.300(1.50)^0=0.300\ \mathrm{L/min}\). The target perfusion is \(0.300/45.0=0.006667\ \mathrm{min}^{-1}\).
With proportional flow, \(\beta=1\), so \(Q_{\mathrm{ad}}=0.300(1.50)^1=0.450\ \mathrm{L/min}\). The target perfusion is \(0.450/45.0=0.0100\ \mathrm{min}^{-1}\), unchanged from reference.
With \(\beta=0.500\), \(Q_{\mathrm{ad}}=0.300\sqrt{1.50}=0.367423\ \mathrm{L/min}\), and \(q_{\mathrm{ad}}=0.367423/45.0=0.00816496\ \mathrm{min}^{-1}\). This also follows from \(q_{\mathrm{ad}}/q_{\mathrm{ad},0}=1.50^{-0.500}=0.816497\). The same expanded volume therefore gives three different perfusion values, because each case encodes a different assumption about how total blood flow responds to tissue expansion.
Worked Example: Inferring a Scaling Exponent From Hypothetical Measurements
Suppose a hypothetical reference measurement gives \(V_{\mathrm{ad},0}=30.0\) L and \(Q_{\mathrm{ad},0}=0.300\) L/min. A comparable target measurement gives \(V_{\mathrm{ad}}=45.0\) L and \(Q_{\mathrm{ad}}=0.360\) L/min. Under the power-law model, calculate the exponent that matches these values, then calculate target perfusion.
Rearrange \(Q_{\mathrm{ad}}/Q_{\mathrm{ad},0}=r^\beta\) by taking natural logarithms:
The target specific perfusion is \(q_{\mathrm{ad}}=0.360/45.0=0.00800\ \mathrm{min}^{-1}\), compared with reference perfusion \(0.300/30.0=0.0100\ \mathrm{min}^{-1}\). It is lower because measured total flow rose by a factor of 1.20 while volume rose by a factor of 1.50. The exponent summarizes these two hypothetical measurements under the assumed power law; it should not be treated as a population-wide physiological constant without supporting data.
Diagnostics, Alternatives, and Limitations
A useful diagnostic is to compare the direction of change in total flow with the direction of change in specific perfusion. For \(r>1\), \(\beta<1\) must yield lower specific perfusion, \(\beta=1\) must leave it unchanged, and \(\beta>1\) must yield higher specific perfusion. If calculations violate these implications, check the exponent, volume ratio, units, and division used to calculate \(q_{\mathrm{ad}}\). Also check that the target tissue volume was not increased while the calculation accidentally retained the reference volume.
Another failure mode is to treat a plausible numerical result as proof that its assumptions are sound. A positive flow and a reasonable-looking \(q_{\mathrm{ad}}\) do not establish that the exponent is physiologically appropriate. Compare the inputs with the intended population and available measurements, and examine whether the implied total flow is compatible with the broader physiological specification. Total cardiac-output allocation is a separate modeling issue; this parameterization alone does not determine or justify changes to cardiac output or to flows in other tissues.
Direct target measurements can avoid choosing a scaling exponent, but they may be unavailable, noisy, or based on a tissue definition that does not match the PBPK compartment. A fixed-flow or proportional-flow assumption is transparent and easy to implement, but either can misrepresent a population if its relationship between tissue size and flow differs from that assumption. A fitted exponent depends on the quality and range of the data and may not extrapolate reliably. In all cases, uncertainty in volume, flow, density, and the scaling rule can affect predicted tissue exposure.
The main practical safeguard is to report the assumed or measured adipose volume, the total adipose flow, and the resulting specific perfusion as separate quantities with units. That makes it possible to distinguish an expanded tissue compartment from a change in the perfusion available to each unit of that compartment.
Check Your Understanding
Answer each question, then compare your reasoning with the solutions.
- A target adipose volume is three times its reference volume. Under proportional flow scaling, what happens to total flow and volume-specific perfusion?
- Reference values are \(V_{\mathrm{ad},0}=20.0\) L and \(Q_{\mathrm{ad},0}=0.200\) L/min. If volume doubles and \(\beta=0.500\), calculate target total flow and specific perfusion.
- Why can a larger total adipose blood flow occur at the same time as lower perfusion per unit tissue?
- What assumption allows an adipose mass ratio to be used in place of a volume ratio?
Solutions. 1. With \(\beta=1\), total flow triples and specific perfusion is unchanged, since \(q/q_0=3^{1-1}=1\). 2. \(Q_{\mathrm{ad}}=0.200(2.00)^{0.500}=0.282843\ \mathrm{L/min}\). Target volume is 40.0 L, so \(q_{\mathrm{ad}}=0.200\sqrt{2.00}/40.0=0.0070711\ \mathrm{min}^{-1}\). 3. If total flow increases by a smaller factor than tissue volume, dividing total flow by the larger volume gives lower specific perfusion; mathematically, for expansion \(r>1\) and \(\beta<1\), \(q/q_0=r^{\beta-1}<1\). 4. The substitution requires adipose density to be treated as constant between reference and target, so mass and volume have the same expansion ratio.