Studies of the sensitivity and reproducibility of pharmacopoeial pyrogen testing.

Rabbits generally numbering 27 or 36 in each experiment were injected in 1970 with doses of the pyrogenic preparation Pyrago (HUMAN) and in 1971-72 and 76 with doses of the International Pyrogen Standard (I.P.St.). The maximum temperature rises were registered, randomized and interpreted according to the requirements of the BP, USP, P. Hung. and P. Nord. 0.29 degrees C mean temperature rises proved to be non-pyrogenic according to BP and P. Hung; whereas according to the USP and P. Nord. in some percents 'to be rejected' qualifications were also established. At more increasing mean temperature rises 0.49 degrees C and 0.58 degrees C, increasing pyrogenic and decreasing 'passable' qualifications were obtained. Beginning with 0.70 degrees C and at 0.86 degrees C and 1.08 degrees C, 'passable' groups disappeared. Exclusion of the rabbits showing low sensitivity (before randomization) barely influenced the results with 3.5 ng kg-1 I.P.St. in the experiment in which the mean temperature rise was 0.49 degrees C. If, however, the mean temperature rise was higher (0.57 degrees C or 0.69 degrees C), such a selection practically resulted in the disappearance of 'passable' qualifications in the triplet groups and a great predominance of 'to be rejected' qualifications in the large groups.