A Method for Measuring Plasma Levels of Digitalis

FOR both practical and theoretical reasons there has long been a need for measuring blood levels of cardiac glycosides. From the practical standpoint the clinician is constantly confronted with patients in cardiac failure in whom it is difficult to determine whether there is too much or too little digitalis, and the wrong analysis may have serious or even fatal consequences. The problem is made more difficult by the fact that digitalis has one of the lowest toxic-to-therapeutic ratios of any commonly used drugs.' The most sensitive nonradioactive assay available has been the duck embryo bioassay of Friedman,2' 3 capable of detecting 0.025 /g. of digitoxin per milliliter of fluid. This proved inadequate for measuring normal blood levels of the glycoside and it was necessary to do laborious urine extractions. These results are difficult to interpret, because cardiac glycosides are excreted fecally as well as renally, and active metabolic products as well as the originally administered drug appear in the urine. Okita studied blood and tissue levels with C14 labeled digitoxin and was able to measure as little as 0.02 /.tg.4' 5 He could detect blood levels for about 3 days after an intravenous dose of 0.5 to 1.5 mg. of C14 digitoxin. This method also is not sufficiently sensitive to detect ordinary blood levels of orally administered glycosides, and of course not all patients can be given radioactive digitalis. In 1953 Schatzmann6 showed that extremely small concentrations of cardiac glycosides inhibit radioactive potassium uptake of human red cells. Glynn7 further demonstrated that both influx and efflux of both potassium