Critical evaluation of the standard hydrolytic resistance test for glasses used for containers for blood and parenteral formulations.

As prescribed by pharmacopoeias, containers should meet certain condition of stability to be used for pharmaceutical products. Glass containers are classified according to their resistance to chemical attack, a test executed by heating the glass in contact with water for 30 min at 121 degrees C. The USP powdered glass test for glass containers was applied to different kinds of glasses used as containers for parenteral formulations. In this experiment not only the released alkalinity was measured but also the release of glass constituents: silicate, borate, sodium, and aluminum, and also the release of some impurities as copper and lead. The USP powdered glass test was also carried out with glass ampoules, clear and amber, in the presence of solution of some inorganic salts, NaCl, KCl, CaCl2, MgCl2, NaHCO3, NaH2PO4, KH2PO4, and sodium gluconate, citric acid and glucose. The results showed that even when releasing very low alkalinity, glasses also released their constituents, in concentration ranges from 8.8 to 33 mg/l for silicate, 0.9 to 6.9 mg/l for borate, 3 to 37 for mg/l for sodium and 0.5 to 2.4 mg/l for aluminum. More expressive results were found, however, for the tests done with solutions instead of pure water. The tests showed that, for most of the solutions, while the measured alkalinity was very low, high levels of the other constituents were found. Basic solutions of bicarbonate and gluconate presented the higher levels of all investigated constituents, confirming the ability of basic solutions to attack and dissolve the glass network. Glucose and citric acid interacted with the glass surface, selectively extracting aluminum, copper, and lead. Whereas silicate, borate and sodium found in these solutions were at levels similar to those found with pure water, the aluminum level was almost 20 times higher. This specific action of citrate and glucose could be related to their metal-complexing ability. The results indicate that even so-called "chemical-resistant glasses," as measured by the hydrolytic resistance test, react with many substances when packaged in contact with them. The hydrolytic resistance test, when used as the sole measure of potential drug-container compatibility, is not reliable.