Ecological risk assessment of the presence of pharmaceutical residues in a French national water survey.

In this study, we focused on the list of 33 chemicals that was established through a French national prioritisation strategy. Assessing the potential risks to the environment was a step-wise procedure: (i) we determined the Predicted Environmental Concentration (PEC) of all molecules measured in the national survey based on the highest recommended dose used, (ii) we used the Measured Environmental Concentration (MEC) and the Predicted No-Effect Concentration (PNEC) to establish the Risk Quotient (RQ) based on either a PEC/PNEC (estimated risk) or MEC/PNEC (real risk) ratio. The risk assessment was performed using a binary ecological classification suggesting that appreciable risk is likely (RQ⩾1). Of the 15 molecules quantified in the survey, 12 had a PEC higher than the action limit value of 0.01µg/L. According to the EU Guideline, environmental risk was estimated as likely for the following five compounds: acetaminophen (RQ=1.6), ibuprofen (RQ=600), diclofenac (RQ=15), oxazepam (RQ=2.1) and carbamazepine (RQ=3.2). Only ibuprofen was identified as posing real environmental risk based on its MEC (RQ=1.9).

Assessment of the health risks related to the presence of drug residues in water for human consumption: application to carbamazepine.

Pharmaceutical residues have been detected at low (usually ng/L) concentrations in drinking water sources. The detection of drugs in water intended for human consumption (WIHC) has raised questions of safety. In the absence of regulatory or other official guidance, water utilities are faced with a problem of which pharmaceutical residues should be monitored and the toxicological limits that should be required. In this essay, we define an approach for the assessment of health risks related to chemicals found in drinking water. We use the examples of carbamazepine and its main metabolite 10,11-epoxycarbamazepine to demonstrate our approach, which involves application of the following algorithm: (1) when there is human or animal toxicity data, a toxicity reference value (TRV) can be calculated; (2) when this is not applicable, an attempt should be made to derive the TRV using known information about the minimum therapeutic dose (MTD); and (3) when no applicable data is available, at all, a threshold of toxicological concern (TTC) should be estimated. In the case of carbamazepine, where relevant toxicological data exists, we derived a TRV, based on the known minimum therapeutic dose (MTD). For carbamazepine's metabolite 10,11-epoxycarbamazepine, there is no toxicological data, so we applied the TTC approach. Using this approach, and combining our estimates with what is known about these chemicals' margin of exposure (MOE), suggests that there is likely to be no appreciable risk to human health exposure to carbamazepine or its major metabolite, even given the inevitable uncertainties in exposure scenarios.