Presentation Abstract

Pharmaceuticals and Personal Care Products as Environmental Contaminants (1): Preliminary Environmental Risk Calculations and Method Development for Analysis in Environmental Media via GC/MS
Kevin Bisceglia, Jim Yu, Ed Bouwer, Mehmet Coelhan, and A. Lynn Roberts
Johns Hopkins University, Baltimore, MD

The purpose of this research is to: (1) conduct a preliminary environmental risk assessment (ERA) by compiling data on pharmaceutical usage, occurrence, and potential (eco)toxic risk; and (2) refine analytical methods for quantifying selected pharmaceuticals in sewage and drinking water samples.

A detailed usage, occurrence, and ecotoxicity database was developed for the highest volume (“top 200”) pharmaceuticals in the brand name, generic, over-the-counter, and hospital categories for the years 1999-2002. Usage data were used to compute expected introductory concentrations (EICs) according to U.S. Food and Drug Administration (FDA) guidelines; results were compared to measured concentrations whenever possible. As very few pharmaceuticals appear in the U.S. Environmental Protection Agency’s (EPA) ECOTOX database, emphasis was placed on toxicity estimates obtained from EPA’s Ecological Structure Activity Relationships (ECOSAR) program. The results were used to select an analyte suite of environmentally relevant compounds potentially amenable to analysis via gas chromatography/mass spectrometry. In developing analytical methods, choice of derivatization agent, solvent, and reaction conditions were systematically optimized, as were necessary sample clean-up procedures. Solid phase extraction (SPE) was explored as a preconcentration step, and the selection of sorbent media, sample pH, and elution solvent identity was investigated systematically as well.

The preliminary ERA yielded no correlation between EIC and pharmaceutical sales ranking, indicating that many more pharmaceuticals could be of potential concern than were identified in this study. The majority (greater than 80%) of pharmaceuticals that are potentially present at measurable concentrations (i.e., > 10 ng/L) have not yet been sought by environmental researchers. EICs appear useful in providing order-of-magnitude estimates of concentrations encountered in sewage treatment plant influents. Preliminary calculations (using experimental and ECOSAR-derived toxicity data) indicate that as many as 10 percent of the pharmaceuticals considered in this study are likely to be of environmental concern. The results suggest that the current FDA “trigger” of 1 μg/L for performing risk assessments on new pharmaceuticals may be insufficiently conservative, as some compounds exert toxicity below that threshold.

This project has developed two new multiresidue methods for a total of 52 acidic, neutral, and basic pharmaceuticals and personal care products employing either pentafluorobenzyl bromide or a combination of N, N-bis(trimethylsilyl)trifluoroacetamide and chlorotrimethylsilane. The method for the acidic compounds is particularly robust and performs best in water:organic solvent mixtures, thus eliminating the need for drying prior to derivatization or after SPE. The SPE method has proved to be reproducible and exhibits minimal interference when extracting target compounds from waters high in natural organic matter. These methods have been demonstrated to perform well in highly complex matrices, such as raw and finished wastewater, as indicated by high recoveries of isotopically labeled surrogates or laboratory fortified field matrices.

The preliminary ERA performed in this study is the most detailed that has been performed in the United States to date. It demonstrates that EICs can be useful tools in predicting pharmaceutical loading into sewage treatment plants, and as such should be useful to environmental scientists in prioritizing future occurrence and ecotoxicity investigations. The analytical methods developed as part of this study employ techniques that can be adopted easily by other laboratories. Together, the methods are capable of analyzing one of the largest and most diverse suite of biologically active chemicals employed in the United States, and thus should prove useful in identifying the nature and scope of pharmaceutical contamination in U.S. waters.

Top of page