The sad reality is that food insecurity is a major problem in communities around the world, with estimates that over 1 million children die from starvation annually.
The clinical management of acute starvation typically includes ready-to-use therapeutic food (RUTF)--a mixture of peanut paste, milk, and vitamins. Some studies have suggested that supplementation with antibiotics may also improve clinical outcomes by reducing the chance of infections or modulating the gut commensal environment. But there is debate whether this is appropriate in the face of potential adverse events and major organizational difficulties in distributing these drugs to rural communities (not to mention the growing issue of drug resistance).
In 2013, a study in Malawi saw significant improvements (in the 3-5% range) in reducing malnutrition with the addition of antibiotics. They found that on average, antibiotics improved the overall efficacy and rate of improvement compared to a placebo group. So question answered, yes?
Well, no. In a study just published in NEJM, Isanaka at al. repeated the study in Niger using amoxicillin. Here, they found no differences in overall recovery from starvation with antibiotic treatment compared to a placebo arm. Neither the rate of mortality nor the number of bacterial infections were different between groups. That is not to say that amoxicillin didn't do anything--drug treatment promoted faster weight gain (duration of clinical starvation was shortened by 2 days) and reduced incidence of diarrheal disease, but it also increased likelihood for resistance in gut enterobacteria by 35-66%.
So why the difference in studies? The authors here suggest that the Malawi study could have had better clinical outcomes due to significant underlying co-morbidities in the population such as HIV infection. Another difference was in the type of starvation being treated--in Malawi it was predominantly kwashiorkor (a deficiency in protein despite sufficient caloric intake overall), while here it was marasmus (lack of all energy sources).
While the debate will continue (hopefully informed by more clinical studies in different settings), the Niger study has made an important point: indiscriminate population-wide antibiotic treatment is not a panacea and should be approached with care. Who benefits most will be an important question moving forward. And one that we as microbiologists can help tackle.
In light of the work from Gordon and colleagues showing that kwashiorkor can be transferred between mice via the microbiota, it will be interesting to analyze the stool microbiota changes observed upon antibiotic treatment and identify community differences that may help us understand which patients will benefit (or not) from prophylactic antibiotics.