A Clinical Bio-Assay for Chorionic Gonadotropin

THE value of the bio-assay of chorionic gonadotropin in clinical endocrinology is well recognized. However, in spite of the numerous methods employed for the quantitative determination of chorionic gonadotropin in experimental work, there does not exist a satisfactory procedure for routine use in clinical laboratories, especially when the number of determinations is large. A bio-assay suitable for routine use in a clinical laboratory should have the following attributes: 1) it should be sufficiently simple to be performed by average laboratory technicians; 2) it should not require too much time for completion; 3) it should be expressed in terms of the International Standard; 4) the error of the assay should be known, and this should be of an order that would allow determinations accurate enough for clinical use. The purpose of this paper is to describe the procedure that has been developed and in use in our laboratory for the past four years (Albert (1)). The method involves no new principles; rather, well-known reactions and methods were adapted to the function of a routine clinical laboratory. In contrast to older procedures which require concentration of the hormone by chemical or physical methods, serial dilution of the concentrate and multiple injections into animals, the method herein described utilizes a single injection of unconcentrated blood serum or urine. Older methods require days to weeks for completion, whereas this procedure is completed in four hours. Previous methods utilize animal units of one kind or another, a procedure which has only local significance and cannot be duplicated satisfactorily in other laboratories. The results of the method to be described are stated in terms of the international unit (i. u.), which is a standard of general reference and a unit of measurement universally understood, thus allowing a ready exchange of results between different workers and laboratories. Finally, in most methods for assay of this hormone the error of the method is unknown. The method to be described has been found to have an error of ±31 per cent, which is sufficiently accurate for clinical purposes.