Different compositions of pharmaceuticals in Dutch and Belgian rivers explained by consumption patterns and treatment efficiency

In the current study, 43 pharmaceuticals and 18 transformation products were studied in the river Meuse at the Belgian-Dutch border and four tributaries of the river Meuse in the southern part of the Netherlands. The tributaries originate from Belgian, Dutch and mixed Dutch and Belgian catchments. In total, 23 pharmaceuticals and 13 transformation products were observed in samples of river water collected from these rivers. Observed summed concentrations of pharmaceuticals and transformation products in river water ranged from 3.5 to 37.8 µg/L. Metformin and its transformation product guanylurea contributed with 53 to 80 % to this concentration, illustrating its importance on a mass basis. Data on the flow rate of different rivers and demographics of the catchments enabled us to calculate daily per capita loads of pharmaceuticals and transformation products. These loads were linked to sales data of pharmaceuticals in the catchment. Simple mass balance modelling accounting for human excretion and removal by sewage treatment plants revealed that sales could predict actual loads within a factor of 3 for most pharmaceuticals. Rivers that originated from Belgian and mixed Dutch and Belgian catchments revealed significantly higher per capita loads of pharmaceuticals (16.0¿±¿2.3 and 15.7¿±¿2.1 mg/inhabitant/day, respectively) than the Dutch catchment (8.7¿±¿1.8 mg/inhabitant/day). Furthermore, the guanylurea/metformin ratio was significantly lower in waters originating from Belgium (and France) than in those from the Netherlands, illustrating that sewage treatment in the Belgian catchment is less efficient in transforming metformin into guanylurea. In summary, the current study shows that consumption-based modelling is suitable to predict environmental loads and concentrations. Furthermore, different consumption patterns and wastewater treatment efficiency are clearly reflected in the occurrence and loads of pharmaceuticals in regional rivers