Spatiotemporal profiling of antimicrobial agents during wastewater treatment – a national study

• 55 antimicrobials tested at 10 WWTPs spatially distributed in the UK in a one-year longitudinal study • Significant temporal and spatial variability observed with winter having higher daily AA loads • WWTP treatment efficiency highly variable, AA group dependent, with no preference for any technology observed • Lincosamides, macrolides, nitrofurans with removal >75%; B-lactams, glycopeptides with <75%; quinolones with low/no removal. Effective wastewater treatment is critical for public health; however, the persistence of antimicrobial chemicals within these systems presents challenges and potential risks to microbial ecology, treatment efficiencies, the spread of antimicrobial resistance (AMR), and the potential environmental impacts of partially treated effluents. To understand the inputs and removals of antimicrobials in wastewater treatment, we measured, in the most comprehensive study to date, an extended list of 55 antimicrobial agents (AA) and their metabolites at ten wastewater treatment works (WwTWs) across England and Wales over six sampling campaigns in a one-year longitudinal study. A total of 204 wastewater and 135 sludge samples were collected resulting in 79,555 AA data points. Our results revealed significant temporal and spatial variability among the different AA groups, reflecting their variable usage patterns. Most AAs peaked in daily loads during the January or March sampling campaigns (except for lincosamides, quinolones, tuberculosis drugs, azoles, and antiretrovirals), whereas all AA groups experienced a significant reduction in their daily loads during the July sampling campaign. Clear spatial and temporal variabilities in per-capita AA usage were also observed likely due to seasonal usage patterns. Removal of AAs was group and individual AA dependent and highly variable, with >75% removal of lincosamides, some macrolides, nitrofurans, and <75% removal of β-lactams, glycopeptides, with some macrolides and sulfonamides being highly variable. Quinolones had very low removals with mean and median removal rates of –31.7 and –3.7%, respectively. Furthermore, comparison between trickling filter and activated sludge systems, which are the two most common biological treatment processes employed in the UK, suggests that trickling filter systems had comparable or even higher removal rates than the activated sludge systems for most AAs, except for azoles. Furthermore, due to the high variability in AA removal, no significant differences were observed in the overall removal efficiency among the tested sampling sites or across the same technologies. This indicates substantial heterogeneity in AA removal and highlights potential challenges in optimizing wastewater treatment performance for improved AA removal.