While many laboratory experiments have shown that antibiotic resistance imposes a fitness cost on resistant bacteria, it’s far less clear if this is the case in natural populations. In Europe, the phasing out of a vancomycin analogue, avoparicin, resulted in a dramatic decrease in vancomycin resistance in enterococci bacteria, from roughly seven percent to about three percent. However, the drop doesn’t appear to have continued further (although the economic and health burdens of treating vancomycin resistant enterococci make this decrease a good thing).
One of the problems with most studies of antibiotic usage is that they are local. Perhaps a hospital or even a county is involved, but there is always the issue that resistant strains can ‘immigrate in’ and obscure the pattern. But a recent study in Israel provides some really good evidence that, in the face of decreased antibiotic use, the frequency of resistance drops.
Israel has a nationalized healthcare system, and consequently can track not only infections, but also the amount of antibiotic prescriptions*. This allowed the authors to compare monthly frequencies of resistance in Streptococcus pneumoniae that ear infections in children to the amount of antibiotics administered. Among Israeli Jews (I’ll get to this in a bit), there was seasonality in prescriptions that matched spikes in resistance: in the cold season (such as it is by Boston standards), antibiotic usage increased along with frequency of resistance, whether the researchers looked at penicillin resistance, erythromycin resistance, or multidrug resistance (resistance to three or more classes of resistance). Basically, a monthly increase of 1 prescription/100 children increased the odds of resistance by five percent.
One interesting thing is that resistance increased (and decreased) in all of the major serotypes of S. pneumoniae. This isn’t a case of epidemic spread of a successful bacterial clone, but a species-wide effect**. The other interesting thing is that this pattern was not observed in Bedouin children. Here’s the neat thing: Bedouin children don’t show seasonal variation in the use of antibiotics. It’s a built-in control experiment.
At least in S. pneumoniae, resistance appears to confer a fitness cost, which means that intelligent and appropriate use of antibiotics can limit the antibiotic resistance problem.
*One more reason to have a national healthcare system: we’ll have the data that will allow us to figure out what all of this crap actually does to us.
**Each of the three major serotypes essentially functions as a replicate experiment.
Cited article: Dagan, Ron, Galia Barkai, Noga Givon-Lavi, Amir Z. Sharf, Daniel Vardy, Ted Cohen, Marc Lipsitch, and David Greenberg. 2008. Seasonality of Antibiotic-Resistant Streptococcus pneumoniae That Causes Acute Otitis Media: A Clue for an Antibiotic-Restriction Policy? The Journal of Infectious Diseases 197:1094-1102. DOI: 10.1086/528995