Examinations were made on the possibility of an in vitro test for bioinequivalence of chloramphenicol tablets, previously found by the measurement of their level in human serum, using the dissolution tests as a marker. Dissolution tests were made by the Japanese Pharmacopoeia VIII disintegration test, USP XVIII dissolution test, Levy's beaker method and Erweka's method. Erweka's method, using water as the solvent, was found to give a result that correlated well with the area under a serum level-time curve. Serum level of 10 products selected from 24 kinds of commercial chloramphenicol tablets was measured in a total of 135 healthy women in the first and 129 healthy women in the second experiment. Dissolution rate of these 10 products was also measured. In addition, serum level and dissolution rate of the original powder for these chloramphenicol tablets were measured and the difference between tablets and their original powder of chloramphenicol was confirmed. Correlation between the human serum level and dissolution rate of commercial chloramphenicol tablets manufactured in Japan was examined and a regression formula for them was calculated. It was found from these experiments that chloramphenicol tablets with 50% dissolution time of over 64 min have a high probability to be bioinequivalent, and 3 out of 24 products of commercially available chloramphenicol tablets on Japanese market in 1974 were found to be bioinequivalent.
