Determination Of Solu

Determination of Solubility of Poorly Soluble Drugs in Gastrointestinal Fluids using a Chromatographic Method

Abstract Purpose. Primarily, to determine the solubilities of sixteen structurally different drugs in gastrointestinal (GI) fluids, using an HPLC-based, small-scale method. Secondary, to examine differences in solubility between buffer, simulated GI fluids and human GI fluids. By comparing these results, it might be possible to propose changes in contents of the simulated intestinal fluids that would improve their ability to predict intestinal solubility. Background. Since the oral route of administration remains the most popular it is of great importance to understand and predict oral absorption of drugs. To fully understand the absorption of an orally administrated drug both GI solubility and permeability must be investigated. The solubility in the GI tract can be difficult to predict due to the complexity of the GI fluids. The physicochemical properties of the drug as well as the physiological conditions in the GI tract will determine the in vivo solubility. By increasing knowledge of the in vivo conditions, progress could be made in developing a general in vitro method predicting the in vivo solubility of poorly water-soluble drugs. Studies have been made on the enhancement of drug solubility due to solubilization by various conjugated and nonconjugated bile salts. An improved characterization of the intestinal contents has been suggested for better predictions of drug solubilization (Wiedmann et al., 2002). While several studies have been made on artificial GI fluids and buffers imitating the physiological conditions, only a few have been made in human GI fluids (Pedersen Lomstein, Brönsted et al., 2000; Pedersen Lomstein, Müllertz et al., 2000). By studying solubility of a number of drugs in buffers, simulated and human GI fluids and comparing these results, an improvement of today’s artificial GI fluids might be possible to suggest. Also, due to the high cost and

limited availability of human GI fluids, it is of primary importance to scale down the used volume of fluid. The possibility of automatization should also be considered. Methods. All solubility determinations were made in standard vials for liquid chromatography, using 0,4-1,5 mL of fluid. The samples were incubated, centrifuged and analysed at 37 oC. A number of gradient and isocratic methods were used for HPLC analysis. Results The solubilities of sixteen drug substances were determined in gastric fluid, intestinal fluid, simulated intestinal fluid, 0,01 M HCl and phosphate buffer pH 6,5, at 37oC. It was found that for most of the studied substances, increased solubility could be seen in the simulated intestinal fluid compared to solubility in phosphate buffer. In most cases solubility increased further in intestinal fluid. All of the neutral substances appeared to have higher solubility in simulated intestinal fluid than in buffer. Two of the neutral compounds appeared to be much more soluble in simulated intestinal fluid than in human intestinal fluid. Among the basic drugs, two of three substances were more soluble in human than in simulated intestinal fluid. Solubilities for the acidic substances were higher in the human intestinal fluid than in the simulated. All of the acidic NSAID substances with comparable data had higher solubilities in the simulated intestinal fluid than in buffer. The solubilities in gastric fluid, which were studied for five acidic drugs, were in all cases higher than the corresponding values in HCl. Conclusions Generally, solubility is higher in both simulated intestinal fluid and human intestinal fluid than in phosphate buffer of the same pH. This is likely to be due to solubilization by bile salts and lecithin. It appears as if the simulated intestinal fluid should be further improved regarding its bile acid and lecithin content, before adopting it as a general method for predictions of intestinal solubility. The largest differences in solubilities between the simulated media and human intestinal fluid were found for neutral, poorly soluble substances. The increase in solubility in gastric fluid compared to HCl is somewhat

surprising, since in the fasted state gastric fluid is normally not expected to contain solubilizing contents. Further studies should be performed on different batches of GI fluids because of the large variation in the composition of these biological fluids. Perhaps a better understanding could also be obtained by adding more substances of varying structures to future studies. Any relationship between the concentration of solubilizing components in the GI fluids and the extent of drug solubilization need to be further investigated.