Investigating the environmental transport of human pharmaceuticals to streams in the United Kingdom

The occurrence of 12 selected pharmaceutical compounds and pharmaceutical compound metabolites in sewage treatment works (STW) effluents and surface waters was investigated. The substances selected for the monitoring programme were identified by a risk ranking procedure to identify those substances with the greatest potential to pose a risk to the aquatic environment. STW final effluent and surface water samples were collected from Corby, Great Billing, East Hyde, Harpenden and Ryemeads STWs. Ten of the 12 pharmaceutical compounds were detected in the STW effluent samples: propranolol (100%, median = 76 ng/l), diclofenac (86%, median = 424 ng/l), ibuprofen (84%, median = 3086 ng/l), mefenamic acid (81%, median = 133 ng/l), dextropropoxyphene (74%, median= 195 ng/l), trimethoprim (65%, 70 ng/l), erythromycin (44%, <10 ng/l), acetyl-sulfamethoxazole (33%, median = <50 ng/l), sulfamethoxazole (9%, median = <50 ng/l), tamoxifen (4%, median = <10 ng/l). In the corresponding receiving streams, fewer compounds and lower concentrations were found: propranolol (87%, median = 29 ng/l), ibuprofen (69%, median = 826 ng/l), mefenamic acid (60%, median = 62 ng/l), dextropropoxyphene (53%, median = 58 ng/l), diclofenac (47%, median = <20 ng/l), erythromycin (38%, median = <10 ng/l), trimethoprim, (38%, median = <10 ng/l), acetyl sulfamethoxazole (38%, median = <50 ng/l). Four human pharmaceutical compounds were detected in samples upstream of the STWs sampled: ibuprofen (57%, median = 181 ng/l), trimethoprim (36%, median <10 ng/l), erythromycin (17%, median = <10 ng/l), propranolol (14%, median = <10 ng/l), suggesting that longer range stream transport of some compounds is possible. The particular STW that was sampled and the month that it was sampled significantly influenced the measured concentrations of several, but not all, substances. There was no significant relationship between usage data and the overall frequency with which different substances were detected. There was however, some evidence to suggest that usage data are positively associated with concentrations of pharmaceuticals in effluent and, particularly, with concentrations measured in surface waters below STWs. These results suggest that most sewage treatment works in England and Wales are likely to be routinely discharging small quantities of pharmaceuticals into UK rivers. None of the pharmaceuticals were found at concentrations that were high enough to cause acute toxic impacts to aquatic organisms. However, insufficient data were available to be able to comment on whether the concentrations measured have the potential to result in more subtle long-term effects on aquatic organisms (e.g. effects on growth, ability to reproduce). (C) 2004 Elsevier B.V. All rights reserved.