When the Number Is the Same and the Blood Is Not

What they found

Forty three-dimensional simulations of a pediatric roller pump, sweeping speed, occlusion and hematocrit. Ten of the conditions delivered the same oxygen, 9 to 12 mL per minute, while peak shear stress ran from about 25 to 70 pascals. Same number, threefold difference in what the blood went through.

What it does not mean

The whole experiment runs at roughly a quarter of clinical flow. Their oxygen delivery tops out near 25 mL per minute. A five kilogram infant at a DO2i of 400 needs about 110. That does not make the finding wrong, it makes it a map of a pump rather than a recipe for a case.

Three more fences. There is no oxygenator, no patient, no temperature, and the blood is modeled as a simple fluid, so the shear they report in a cold case understates what you would get. Peak stress is measured per revolution and never added up across the revolutions, so twice the passes at a lighter load reads as an improvement when it may not be.

And the line that will travel furthest, that lower hematocrit permitted gentler settings, comes from filtering their runs at a minimum delivery rather than comparing matched pairs. Do not read it as permission to dilute a child.

What to do today

Make occlusion a decision instead of a habit. Set it, know what you set, and put the number where the next person can find it. A leakier raceway moves less blood per revolution and concentrates the squeeze; a tighter one buys flow and spends the cells. Oxygen delivery will not tell you which of those you chose.

And when someone says the case made its DO2i number, the answer is: yes, and here is the setting that got us there.