International Journal of Pharmaceutics 315 (2006) 24–29 Arsenic release from glass containers by action of intravenous nutrition formulation constituents Denise Bohrer ∗ , Emilene Becker, Paulo C´ ıcero Nascimento, Vanessa M¨ orschb¨ acher, Leandro Machado de Carvalho, Marieli da Silva Marques Chemistry Department, Universidade Federal de Santa Maria, 97105-900 Santa Maria, Brazil Received 1 June 2005; received in revised form 26 January 2006; accepted 27 January 2006 Available online 18 April 2006 Abstract Pharmacopoeias prescribe tests to determine the levels of arsenic in raw materials and glass containers. In this study, glass ampoules for injectables containing individually the main components of intravenous nutrition formulations were submitted to the hydrolytic resistance test by heating at 121 ◦ C for 30 min. As(V) and As(III) levels in these solutions after heating were determined by hydride generation atomic absorption spectrometry. The arsenic content of substances used in these formulations was previously determined, as well as the arsenic content of the glass containers. The results showed that raw substances as well as glass containers contain arsenic. Moreover, arsenic is released during the heating (hydrolytic resistance test). However, the amount released and the arsenic species present in solution depend on the solution composition. While As(V) was the predominant specie in glass, solutions containing reducing substances such as glucose and vitamins had As(III) in higher concentration. Therefore, arsenic is released from glass containers during the heating for sterilization, and reacts with formulation constituents depending on their reducing properties. © 2006 Elsevier B.V. All rights reserved. Keywords: Arsenic species; Intravenous nutrition; Glass; Contaminants 1. Introduction Glass containers for parenterals must meet specifications con- cerning stability (USP 26, BP, IP). One such specification is related to the extractability of arsenic during the heating cycle for product sterilization. The specification, however, deals neither with arsenic speciation nor considers the action of formula- tion constituents on the glass surface since the test addresses water attack only. Glass can contain arsenic as a constituent because arsenic oxide(III) may be added to glass melt as a fin- ing agent to improve its transparency (Scholze, 1988), a feature specially important for solutions for intravenous administration that must be subjected to visual inspection of the content before use (Bacon, 1986). Extraction of arsenic from glass containers could be increased or reduced by action of formulation constituents. Moreover, depending on the nature of the substance, it could increase ∗ Corresponding author. Tel.: +55 55 3220 8870; fax: +55 55 3220 8870. E-mail address: ndenise@quimica.ufsm.br (D. Bohrer). the extraction rate of arsenic from glass and also change the arsenic species in solution by converting As(III) into As(V) or vice-versa. In a previous work we developed a procedure to investigate the presence of As(III) and As(V) in commercial solutions for parenteral nutrition (Bohrer et al., 2005). While in water for injection and in solutions of salts, As(V) predom- inated over As(III), in solutions of some amino acids, glucose and vitamins, As(III) was the most abundant species. As the later substances possess reducing properties, the presence of As(III) in higher concentration was attributed to the reduction of As(V) in solution by these substances. Parenteral nutrition (PN) is the administration of nutrients intravenously to patients that cannot be fed via the gastrointesti- nal tract. Products for PN are commercialized in form of sterile solutions, and include electrolytes, amino acids, carbohydrates, albumin, vitamin and lipids emulsions. Previous studies have shown that PN solutions can be contaminated by toxic metals (Buchman et al., 2001; Leung, 1995). Such contamination can lead to organ deposition of the metals with subsequent dele- terious effects. The presence of arsenic in pharmaceuticals is a matter of great consideration due to its high toxicity. Stud- 0378-5173/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.ijpharm.2006.02.008