文摘
We assessed the ecotoxicological hazard potential of 42pharmaceuticals from 22 therapeutic groups, includingmetabolites formed in humans. We treated each parent drugand its metabolites as a mixture of similarly actingcompounds. If physicochemical or effect literature datawere missing, we estimated these with quantitative structure-activity relationships (QSAR). Additionally, we estimatedmicropollutant removal efficiency of urine source separationusing pharmaceutical information. On average, 50% of aparent drug was metabolized, and 70% was excreted withurine, albeit with large variations among drugs. Metabolismreduced the toxic potential of all but eight drugs. Thesubsequently modeled risk quotient was mostly below thethreshold of one. However, ibuprofen and its metabolitesin a mixture could pose an ecotoxicologal risk; and possiblyalso acetylsalicylic acid, bezafibrate, carbamazepine,diclofenac, fenofibrate, and paracetamol. Lipophilicity andsale quantities of parent drugs alone were insufficientto estimate their ecotoxicological risk. Urine separationcould decrease the ecotoxicological risk of some, but notall drugs. The estimated risk quotients were equal inurine and feces, again with large variations amongcompounds. Because of scientific limitations of the modeland inconsistent literature data the results are somewhatuncertain. However, this new approach allows first tierscreening of single drugs, thus supporting decision-making.