The pH of ex vivo plasma, bile and urine was monitored at different times and temperatures of storage, and following different sample processing methods such as ultrafiltration, centrifugation, precipitation and evaporation. The results showed that the pH of ex vivo plasma, bile and urine increased upon storage, and following sample processing and could lead to significant degradation of pH-labile compounds. Several compounds were used to illustrate the impact of pH shifts on drug stability and interpretation of results obtained from in vivo studies. For example, after 1 h of incubation (37 degrees C) in rat plasma (pH 8.3), about 60%, of I was lost. However, in phosphate buffer, losses were about 12% at pH 7.4 and 40% at pH 8.0. Plasma pH also increased during ultrafiltration, centrifugation and extraction. After methanol precipitation of plasma proteins, and evaporation of the supernatant and redissolution of the residue, the resulting solution had a pH of 9.5. Under these conditions, II was degraded by 60% but was stable when phosphate buffer was used to maintain the pH at 7.4. The shift in plasma pH can yield misleading results from in vivo studies if the pH is not controlled. For example, the major circulating metabolite of II was also formed in plasma ex-vivo. This ex vivo degradation was prevented when blood samples were collected into tubes containing 0.1 volume of phosphate buffer (0.3 M, pH 5). The pH of ex vivo plasma can best be stabilized at physiological conditions using 10% CO2 atmosphere in a CO2 incubator. Changes in pH of ex vivo urine and bile samples can have similar adverse effect on pH-labile samples. Thus, processing of plasma samples under a 10% CO2 atmosphere is a method of choice for stability or protein binding studies in plasma, whereas citrate or phosphate buffers are suitable for stabilizing pH in bile and urine and for plasma samples requiring extensive preparations or long term storage.