Human health risk assessment of carbamazepine in surface waters of North America and Europe.

A human health risk assessment was carried out for environmental exposures to carbamazepine (CBZ) and its major human metabolites, carbamazepine diol (CBZ-DiOH) and carbamazepine N-glucuronide (CBZ-N-Glu). Carbamazepine is an active pharmaceutical ingredient (API) used worldwide as a medicine for treating epileptic seizures and trigeminal neuralgia. Carbamazepine tends to be detected in surface water more frequently, and at relatively higher concentrations, than most other APIs. Predicted no effect levels (PNECs) for CBZ and its major human metabolites were developed for surface waters to be protective of human health from environmental exposures from drinking water and fish consumption. These PNECs were compared to both measured (MEC) and predicted (PEC) environmental concentrations for North America and Europe. PECs were calculated using the geo-referenced models PhATE for North America and GREAT-ER for Europe. The combined PNEC for drinking water and fish consumption for CBZ is 226,000ng/L. Ninetieth percentile MECs ranged from 150 to 220ng/L, while 90th percentile PECs ranged from 333 to 658ng/L. Calculated margins of safety (MOS) therefore range from 340 to 1500. MOS for the major metabolites are significantly higher. This assessment indicates that CBZ and its major metabolites have high MOS (>>1) and thus should have no appreciable risk to human health through environmental exposures based on available human data.

Exposure assessment of 17alpha-ethinylestradiol in surface waters of the United States and Europe.

An evaluation of measured and predicted concentrations of 17-ethinylestradiol in surface waters of the United States and Europe was conducted to develop expected long-term exposure concentrations for this compound. Measured environmental concentrations (MECs) in surface waters were identified from the literature. Predicted environmental concentrations (PECs) were generated for European and U.S. watersheds using the GREAT-ER and PhATE models, respectively. The majority of MECs are nondetect and generally consistent with model PECs and conservative mass balance calculations. However, the highest MECs are not consistent with concentrations derived from conservative (worst-case) mass balance estimates or model PECs. A review of analytical methods suggests that tandem or high-resolution mass spectrometry methods with extract cleanup result in lower detection limits and lower reported concentrations consistent with model predictions and bounding estimates. Based on model results using PhATE and GREAT-ER, the 90th-percentile low-flow PECs in surface water are approximately 0.2 and 0.3 ng/L for the United States and Europe, respectively. These levels represent conservative estimates of long-term exposure that can be used for risk assessment purposes. Our analysis also indicates that average concentrations are one to two orders of magnitude lower than these 90th-percentile estimates. Higher reported concentrations (e.g., greater than the 99th-percentile PEC of approximately 1 ng/L) could result from methodological problems or unusual environmental circumstances; however, such concentrations are not representative of levels generally found in the environment, warrant special scrutiny, and are not appropriate for use in risk assessments of long-term exposures.

Use of acute and chronic ecotoxicity data in environmental risk assessment of pharmaceuticals.

For many older pharmaceuticals, chronic aquatic toxicity data are limited. To assess risk during development, scale-up, and manufacturing processes, acute data and physicochemical properties need to be leveraged to reduce potential long-term impacts to the environment. Aquatic toxicity data were pooled from daphnid, fish, and algae studies for 102 active pharmaceutical ingredients (APIs) to evaluate the relationship between predicted no-effect concentrations (PNECs) derived from acute and chronic tests. The relationships between acute and chronic aquatic toxicity and the n-octanol/water distribution coefficient (D(OW)) were also characterized. Statistically significant but weak correlations were observed between toxicity and log D(OW), indicating that D(OW) is not the only contributor to toxicity. Both acute and chronic PNEC values could be calculated for 60 of the 102 APIs. For most compounds, PNECs derived from acute data were lower than PNECs derived from chronic data, with the exception of steroid estrogens. Seven percent of the PNECs derived from acute data were below the European Union action limit of 0.01 µg/L and all were anti-infectives affecting algal species. Eight percent of available PNECs derived from chronic data were below the European Union action limit, and fish were the most sensitive species for all but 1 API. These analyses suggest that the use of acute data may be acceptable if chronic data are unavailable, unless specific mode of action concerns suggest otherwise.

Predicting concentrations of trace organic compounds in municipal wastewater treatment plant sludge and biosolids using the PhATE™ model.

This article presents the capability expansion of the PhATE™ (pharmaceutical assessment and transport evaluation) model to predict concentrations of trace organics in sludges and biosolids from municipal wastewater treatment plants (WWTPs). PhATE was originally developed as an empirical model to estimate potential concentrations of active pharmaceutical ingredients (APIs) in US surface and drinking waters that could result from patient use of medicines. However, many compounds, including pharmaceuticals, are not completely transformed in WWTPs and remain in biosolids that may be applied to land as a soil amendment. This practice leads to concerns about potential exposures of people who may come into contact with amended soils and also about potential effects to plants and animals living in or contacting such soils. The model estimates the mass of API in WWTP influent based on the population served, the API per capita use, and the potential loss of the compound associated with human use (e.g., metabolism). The mass of API on the treated biosolids is then estimated based on partitioning to primary and secondary solids, potential loss due to biodegradation in secondary treatment (e.g., activated sludge), and potential loss during sludge treatment (e.g., aerobic digestion, anaerobic digestion, composting). Simulations using 2 surrogate compounds show that predicted environmental concentrations (PECs) generated by PhATE are in very good agreement with measured concentrations, i.e., well within 1 order of magnitude. Model simulations were then carried out for 18 APIs representing a broad range of chemical and use characteristics. These simulations yielded 4 categories of results: 1) PECs are in good agreement with measured data for 9 compounds with high analytical detection frequencies, 2) PECs are greater than measured data for 3 compounds with high analytical detection frequencies, possibly as a result of as yet unidentified depletion mechanisms, 3) PECs are less than analytical reporting limits for 5 compounds with low analytical detection frequencies, and 4) the PEC is greater than the analytical method reporting limit for 1 compound with a low analytical detection frequency, possibly again as a result of insufficient depletion data. Overall, these results demonstrate that PhATE has the potential to be a very useful tool in the evaluation of APIs in biosolids. Possible applications include: prioritizing APIs for assessment even in the absence of analytical methods; evaluating sludge processing scenarios to explore potential mitigation approaches; using in risk assessments; and developing realistic nationwide concentrations, because PECs can be represented as a cumulative probability distribution. Finally, comparison of PECs to measured concentrations can also be used to identify the need for fate studies of compounds of interest in biosolids.

Human pharmaceuticals in US surface waters: a human health risk assessment.

The detection of low levels of pharmaceuticals in rivers and streams, drinking water, and groundwater has raised questions as to whether these levels may affect human health. This report presents human health risk assessments for 26 active pharmaceutical ingredients (APIs) and/or their metabolites, representing 14 different drug classes, for which environmental monitoring data are available for the United States. Acceptable daily intakes (ADIs) are derived using the considerable data that are available for APIs. The resulting ADIs are designed to protect potentially exposed populations, including sensitive sub-populations. The ADIs are then used to estimate predicted no effect concentrations (PNECs) for two sources of potential human exposure: drinking water and fish ingestion. The PNECs are compared to measured environmental concentrations (MECs) from the published literature and to maximum predicted environmental concentrations (PECs) generated using the PhATE model. The PhATE model predictions are made under conservative assumptions of low river flow and no depletion (i.e., no metabolism, no removal during wastewater or drinking water treatment, and no instream depletion). Ratios of MECs to PNECs are typically very low and consistent with PEC to PNEC ratios. For all 26 compounds, these low ratios indicate that no appreciable human health risk exists from the presence of trace concentrations of these APIs in surface water and drinking water.

Screening analysis of human pharmaceutical compounds in U.S. surface waters.

The PhATE (Pharmaceutical Assessment and Transport Evaluation) model presented in this paper was developed as a tool to estimate concentrations of active pharmaceutical ingredients (APIs) in U.S. surface waters that result from patient use (or consumption) of medicines. PhATE uses a mass balance approach to model predicted environmental concentrations (PECs) in 11 watersheds selected to be representative of most hydrologic regions of the United States. The model divides rivers into discrete segments. It estimates the mass of API that enters a segment from upstream or from publicly owned treatment works (POTW) and is subsequently lost from the segment via in-stream loss mechanisms or flow diversions (i.e., man-made withdrawals). POTW discharge loads are estimated based on the population served, the API use per capita, the potential loss of the compound associated with human use (e.g., metabolism), and the portion of the API mass removed in the POTW. Simulations using three surrogate compounds showthat PECs generated by PhATE are generally within an order of magnitude of measured concentrations and that the cumulative probability distribution of PECs for all watersheds included in PhATE is consistent with the nationwide distribution of measured concentrations of the surrogate compounds. Model simulations for 11 APIs yielded four categories of results. (1) PECs fit measured data for two compounds. (2) PECs are below analytical method detection limits and thus are consistent with measured data for three compounds. (3) PECs are higher than (i.e., not consistent with) measured data for three compounds. However, this may be the consequence of as yet unidentified depletion mechanisms. (4) PECs are several orders of magnitude below some measured data but consistentwith most measured data forthree compounds. For the fourth category, closer examination of sampling locations suggests that the field-measured concentrations for these compounds do not accurately reflect human use. Overall, these results demonstrate that PhATE may be used to predict screening-level concentrations of APIs and related compounds in the environment as well as to evaluate the suitability of existing fate information for an API.