STANDARDIZATION OF THE QUICK PROTHROMBIN TEST : WITH REFERENCE TO THE STATISTICAL SIGNIFICANCE OF VARIATIONS IN THE PROTHROMBIN CONCENTRATION WITH USE OF A STABLE THROMBOPLASTIN OF HIGH POTENCY

A large specimen of human brain, obtained at autopsy, was dehydrated and stored in an evacuated desiccator. Portions of the material were removed when needed and the desiccator was re-evacuated. Under these conditions there was no loss of thromboplastic potency over a period of two years.

See PDF for Table

See PDF for Table

The dry oxalate mixture (6 mg. of ammonium oxalate and 4 mg. of potassium oxalate) was used as an anticoagulant in order to obtain plasma for the prothrombin tests.

With the specimen of thromboplastin used in these experiments, normal standards for the Quick prothrombin test were found as shown in table 8. These results do not necessarily mean that the normal range of prothrombin concentrations is from 75 to 140 per cent of the actual amount of prothrombin in the plasma of the average normal subject. They mean only that with the methods here employed, values within this range may be expected to occur.

There was no significant difference between the prothrombin values obtained in a group of normal subjects by methods using diluted or undiluted plasma. In repeated tests on the same subject there appeared to be slightly greater accuracy when the tests were performed on diluted plasma, but there was no significant difference between the results obtained with plasma diluted to 50 or to 25 per cent of its original concentration.

Repeated tests on diluted and undiluted plasma from the same normal subject produced the range of normal variation about the various prothrombin concentrations shown in table 9.

The normal variation that may be expected to occur at any given prothrombin level must be considered in interpreting the variations that appear after administration of vitamin K or dicumarol.

The use of a single specimen of blood from one random normal subject as a normal standard may lead to considerable error, because of unavoidable technical variables and because of differences in the prothrombin concentrations of different normal subjects. To establish reliable normal standards, the blood of at least 5 normal subjects should be individually tested with the same specimen of thromboplastin and the results averaged.