Blood constituents such as albumin, glycoproteins, and other proteins can cling to drug molecules, leaving just a fraction free to exert an effect. Yet clinicians adjust dosage mostly according to the total instead of the free concentration: The larger pool of molecules in the total is easier to measure; and laboratory manipulations can upset the proportions of free and bound drug, which exist in equilibrium.

“You want to remove the free fraction without disturbing the bound,” says bioanalytical chemist David Hage of the University of Nebraska in Lincoln.

Hage and his collaborators decided that they had to extract free molecules quickly, before proteins released any captives (Anal. Chem. 2001, 73, 2157–2164). First, they calculated the speed of the equilibrium that they chose to outwit—between albumin and the anticoagulant drug warfarin. Then, assuming a flow rate that is standard for laboratory machines, they figured the length of tubing within which their extraction had to be finished. Finally, they fabricated cylindrical cartridges with a layer of antiwarfarin antibodies that was thinner than the calculated length, while packing in enough antibodies to absorb the free warfarin in the time allotted. From calibrated solutions, the column sponged 95% of the free warfarin in 120 ms.

Endocrinologist Inder Chopra of the University of California, Los Angeles, says that counting the few molecules that stick to the column will rarely be so easy as for warfarin, which fluoresces. The column “doesn’t solve any particular measurement in my mind,” he says. But pharmaceutical chemist George Wilson of the University of Kansas in Lawrence says that the method is progress. “This allows you to do something that would be difficult to do in any other way,” he says.