Last week Jon ended his reflection with a grade A recommendation to close the toilet lid before flushing, as the best way to minimize the potential impact of “toilet flushing plumes”. Even better: do not flush at all. This week we take it from there, in a discovery of what happens subsequently. Let’s take the loo with the highest likelihood of being soiled with antibiotic-resistant bacteria (ARB) and antimicrobial resistance genes (ARGs): the hospital loo. Elena Buelow, former Phd student in our group and now post-doc in Limoges, France, quantified how hospital sewage contributes to the quantity and diversity of ARGs in the general sewerage system. The work was published on-line today.

This work started with “old-fashioned” field work: Sampling of material at our hospital (UMC Utrecht) wastewater pipe and at 2 wastewater treatment plants (WWTPs). One plant („urban WWTP‟) treats wastewater of the 290,000 inhabitants of Utrecht, including the UMCU and 2 other hospitals. The other plant („suburban WWTP‟) treats wastewater of a suburban community of 14,000 inhabitants and no hospital. Elena took influent and effluent samples at the WWTPs and distal from there in surface water (in which people like to swim in summer, but not in March/April when sampling occurred).

Then the fancy stuff came: 16S rRNA gene sequencing and nanolitre-scale quantitative PCR testing to compare microbiota composition and levels of ARGs at the different sampling sites.

First, the hospital sewage microbiome appeared markedly different than the microbiome at any of the other sites. In all 67 different ARGs were detected in the different samples, conferring resistance to 13 classes of antimicrobials. Not unexpectedly, hospital sewage was richer in ARGs than the other samples for 12 out of 14 classes of ARGs. In particular so for aminoglycoside, β-lactam and vancomycin resistance genes; being – on average – 12, 15 and 175 fold higher in hospital sewage. The quinolone resistance gene qnrA, the erythromycin resistance gene ermC, the vancomycin resistance gene vanB, the AmpC-type β-lactamases blaDHA-1 and blaCMY-2 and the carbapenemase blaNDM were only detected in hospital sewage. Yet, the carbapenemase blaIMP was detected in effluent and river water samples, but not in hospital sewage or WWTP influent.

And now the surprise: The relative abundance of ARGs in the „urban WWTP‟ influent, receiving sewage from UMCU and two additional hospitals, and the „suburban WWTP‟ influent was comparable for any of the detected ARG families.

Also reassuring; for 9 classes of antibiotics (aminoglycosides, β-lactams, chloramphenicols, macrolides, polymyxins, puromycins, trimpethoprim, quinolones, and tetracyclines) the levels of ARGs in the urban WWTP were lower in the effluent than in the influent samples. Actually, levels of ARGs in WWTPs effluent were comparable to levels in effluent-influenced river water.

So, this study demonstrated that hospital sewage is loaded with ARGs, but that considerable dilution occurs before the WWTP is reached. Imaginable, as hospital sewage contributed to an estimated 0.8% of all sewage. Furthermore, considerable cleaning seems to occur in the WWTP, where sewage is treated with nitrification and denitrification in activated sludge systems and phosphorus removal.

Now think of a country where a considerable part of the population does not suffer from “fear of fear” (as in the Netherlands), where not all shit becomes sewage through a flushable toilet with a lid, and where river bathing is a daily habit. What would, there, be the effects of operational WWTPs in reducing the release of ARGs from human subjects into nature?