Occurrence and removal of frequently prescribed pharmaceuticals and corresponding metabolites in wastewater of a sewage treatment plant.

ABSTRACT

The present study determines removal rates (RR) of 56 pharmaceuticals and metabolites, respectively, in an urban sewage treatment plant using mass flow analysis by comparing influent and effluent loads over a consecutive ten-day monitoring period. Besides well investigated compounds like carbamazepine and metoprolol, less researched targets, such as topiramate, pregabalin, telmisartan, and human metabolites of pharmaceuticals were included. Another aim was to determine the ratio of pharmaceuticals and corresponding metabolites in raw wastewater. Valsartan and gabapentin were detected at the highest average concentrations in influent (c(val) = 29.7 (± 8.1) µg/L, c(gab) = 13.2 (± 3.3) µg/L) and effluent (c(val) = 22.1 (± 5.1) µg/L, c(gab) = 12.1 (± 2.6) µg/L) samples. The comparison of mass loads in influent and effluent showed a significant removal (p<0.1) for 20 compounds but only enalapril, eprosartan, losartan, pregabalin, and quetiapine were removed from the aqueous phase by more than 50%. Another 20 compounds were determined without significant difference and for five compounds (clindamycin, lamotrigine, oxcarbazepine, O-desmethyl venlafaxine, triamterene), a significant higher mass load in the effluent than in the influent was observed. It has to be noticed that metabolites like 10,11-dihydro-10-hydroxy carbamazepine (MHD) are found in higher mass loads than the corresponding parent compound in the sewage samples. Furthermore, metabolites and parent compound behave differently in the sewage treatment process. While MHD (RR = 15.1%) was detected with lower mass load in the effluent than in the influent, oxcarbazepine (RR = -73.2%) showed the contrary pattern. When comparing expected and measured ratios of parent compound and metabolite in raw sewage, citalopram/N-desmethyl citalopram for example, showed good results. However, a major problem exists due to insufficient data regarding metabolism and excretion of many pharmaceuticals. This complicates the prediction of relevant metabolites and further efforts are needed to overcome this problem.