The colour of the global crisis of antibiotic resistance is red (if te Gram stain is your reference). In rich countries we have ESBL-producing Enterobacterales (mainly E. coli), but the real problem are carbapenemase-producing strains (Klebsiella, Pseudomonas and Acinetobacter) that are already endemic in lower and middle-income countries. The unanswered question is “where did these resistant bacteria come from”? Animals or bathrooms?

I previously blogged on this topic, and now again, because of 2 new publications. The first one, a month ago in Science, addressed the occurrence of antibiotic resistant bacteria in lower and middle income countries. It actually was a literature review. The authors identified 901 point-prevalence  surveys reporting resistance data in animals and pooled the data. The first analysis was simple: count the number of publications in time and demonstrate that this increases (almost exponentially). Then the findings in these studies were categorized as high prevalence if “the proportion of antimicrobial compounds with resistance was higher than 50%”. If the percentage was lower, it was not called “low”, but “emerging”. And in time the proportion of publications per year with high prevalence increased for chickens and pigs. Not very obvious, but recognizable “if the trend is your friend”. The conclusion is that reported antibiotic resistance in animals in lower and middle income countries is increasing (although we will never hear about studies with low resistance prevalence that don’t get published). Yet, these studies do not provide evidence that the animals contribute anything to the emergence of carbapenamase-producing bacteria among humans. In fact, they were hardly detected in any of these studies, not even in the continents and countries where these bacteria cause infections in humans. Apparently, these bacteria are transmitted between humans, rather than between animals and humans.

The second study, in Lancet ID, is from the UK. 913 ESBL-producing E. coli isolated in defined time periods and regions “from human faeces (360 isolates), sewage (n=65), farm slurry (n=24), animals (n=83) and retail foodstuffs (n=111)” and 293 human bloodstream isolates were sequenced and sequence types were determined. The conclusion: “non-human reservoirs made little contribution to invasive human disease”. Thus, other humans are the most important reservoir for ESBL-producing E. coli.

These findings are similar to those from previous (and similarly designed) studies in the Netherlands, reaching similar conclusions. I add them here and here, as you can’t find them among the references (!).

So, what now? If these bacteria didn’t come from animals, where then from? And if the animals were not the amplified of the epidemiology, how did ESBL-producing E. coli manage to colonize 5% of the Dutch population, without leaving a trace that points to a risk factor? Bob Weinstein did send me this 25 year old scene from Seinfeld, that offers an explanation.

For carbapenemase-producing bacteria failing infection control in healthcare settings is the big smoking gun. In some countries (such as Greece and Italy) this seems related to economic crises that destroyed the minimal infrastructure required for infection control. But why did these bacteria emerge in Asia and Africa in recent years ? Same problem? Bad luck? Or did we miss something? Nice question for an ambitious PhD student (with skills).