Editor's note: This story was updated Jul 1 with comments from Sally Miller, PhD.
A new study by researchers with a group that advises farmers on plant diseases suggests antibiotics may be used more frequently on crops, and on a wider variety of crops, than previously thought.
The findings are based on data from Plantwise, an agricultural development program that trains extension workers to provide assistance and advice to farmers in countries in Africa, Asia, and the Americas. An analysis of more than 436,000 records from the Plantwise database containing recommendations provided to farmers for management of crop problems found that antibiotics, including some medically important antibiotics, are being recommended for use on more than 100 crops, most commonly rice.
According to the records, antibiotics are being recommended not only for specific bacterial diseases but also for fungal problems. They were also recommended for controlling insects and mites and viral infections, neither of which antibiotics have any activity against.
Overall, the proportion of records containing a recommendation for an antibiotic was less than 1%, and the records do not indicate whether the recommendation was followed. In addition, the quantities of antibiotics used, where the recommendations were followed, are far smaller than those used in human and veterinary medicine.
But the findings are noteworthy because there is very little documentation or monitoring of antibiotic use in crops. A recent joint investigation by the Food and Agriculture Organization, the World Health Organization (WHO), and the World Organization for Animal Health found that only 3% of 158 countries regularly assess the types and amounts of antibiotics used on crops. Antibiotic use in human and veterinary medicine, seen as the primary avenues for the development of antibiotic resistance, get far more attention.
Lead study author Phil Taylor of CABI, an international nonprofit that focuses on problems in agriculture and the environment, specifically plant diseases and pests, said that although it's difficult to say how much of a human health threat antibiotic use in crops presents, it provides another potential route for the spread of antibiotic resistance.
"Antibiotic use on crops is certainly another way bacterial communities are exposed to low levels of antibiotics, which will of course amplify resistance genes," Taylor said. "Coating the phylloplane [leaf surface] in antibiotics is certainly going to expose a huge number of bacteria to antibiotics they will never have encountered previously."
The study was published in CABI Agriculture and Bioscience, the official journal of CABI.
Concerns about medically important antibiotics
In their analysis of data from Plantwise, which is operated by CABI, Taylor and his CABI colleague and coauthor Robert Reeder found a total of 436,674 records from 32 countries, but the proportion containing an antibiotic recommendation accounted for just 0.38%, and antibiotic recommendations were recorded in only 17 of the countries analyzed. The final analysis included records from 11 countries representing four WHO regions—South East Asia, Eastern Mediterranean, Western Pacific, and the Americas.
The authors note that the data were grouped into regions rather than being presented on a per country basis because of potential sensitivities surrounding the recommendation of antibiotics for use on crops. In addition, the dataset does not contain records from China, which does not disclose data to CABI. And no records from Africa were found to contain recommendations for antibiotics, for reasons that are unclear, Taylor said. He doesn't think it's due to cost or to differences in crops or the types of pathogens that affect crops.
"The crops are similar across the globe and so are the diseases," he said. "I could imagine that the supply chains have not yet reached Africa and if nothing changes it will only be a matter of time before antibiotics are used [there]."
The analysis found that a total of 11 antibiotics belonging to eight classes were recommended for use on more than 100 different crops. The types of antibiotics recommended varied by region, as did the amounts recommended. While six of the antibiotics are used primarily to control plant disease, the other six—amoxycillin, tetracycline, oxytetracycline, streptomycin, gentamicin, and cefadroxil—are considered medically important by the WHO.
Streptomycin was the most frequently recommended antibiotic, followed by tetracycline and kasugamycin, an antibiotic developed specifically for the control of bacterial diseases in rice that is used on a variety of other crops.
Rice was the dominant crop in which antibiotics were recommended. In South East Asia, 7.4% of the recommendations for rice contained an antibiotic. Taylor and Reeder estimated that if the recommended rate of Plantomycin (a blend of streptomycin and tetracycline that was the most widely recommended antibiotic by trade name in the dataset) was applied to 7.4% of the rice-growing area in South East Asia, a single application would represent 63 tons of streptomycin and 7 tons of tetracycline.
Other crops for which antibiotics were recommended included tomatoes, citrus, potatoes, and cabbage.
Taylor, who was a farmer for 16 years before joining CABI, said that many people around the world feel that antibiotics are a reasonable means of controlling plant disease and that antibiotics used strictly on crops may not pose a problem for human health. Still, the use of medically important antibiotics on crops concerns him.
"I do think it is dreadful that medically important antibiotics are being used in this way," he said. "The use of antibiotics in crop production further strengthens the case for plant health to be included under the 'one health' banner."
Taylor and Reeder also noted that antibiotics used on crops are routinely mixed with other agrochemicals, which has led to concerns over interactions that might promote cross-resistance or co-selection for antibiotic resistance.
"There is some alarming evidence that the blending of antibiotics with other agrochemical products induces resistance much faster than exposure to antibiotic in isolation, and this is a concerning issue, but it is not something we have researched," Taylor said.
Sally Miller, PhD, distinguished professor of food, agriculture and environmental sciences in plant pathology at The Ohio State University, said that even though the findings are based on recommendations and not actual use, and are only representative of smallholder farmers who sought advice on crops from plant clinics, the data are useful because they point to countries and regions where antibiotics are available for use on crops and are being recommended. She's also concerned about the number of records in which antibiotics were incorrectly recommended .
"The antibiotics listed are not effective against insects/mites, viruses, etc.," said Miller, who was not involved in the study. "If the incorrect recommendations are followed, the farmers are likely to experience economic losses due to lack of disease control plus antibiotic application costs, and antibiotics are needlessly placed in the environment."
Antibiotic use on US fruit trees
Although North America not was among the regions covered in the dataset, to date the most widely documented use of antibiotics on crops is in the United States, where streptomycin and oxytetracycline have long been used to combat fire blight, a bacterial disease that strikes apple and pear trees. Streptomycin is also applied to tomato seedlings in greenhouses before the plants are transplanted to fields.
More recently, the US Environmental Protection Agency has allowed for expanded use of oxytetracycline on roughly 700,000 acres of citrus farms in Florida and California to prevent citrus greening disease and is reviewing a request for expanded use of streptomycin.
Public health and environmental advocacy groups, as well as the Centers for Disease Control and Prevention and the Food and Drug Administration, have expressed concern that spraying massive amounts of these antibiotics on citrus trees could pose a threat to human health by selecting for antibiotic-resistant bacteria in the soil, which could then share resistance genes with pathogens that cause disease in humans and animals.
Taylor and Reeder also say there is evidence to suggest that crops, especially if eaten raw, could be a potential vehicle for resistant bacteria to enter the human gut. But they note that those who advocate for antibiotics to be used against plant diseases point out there is no evidence of resistant bacteria spreading from plant pathogenic bacteria to human or animal pathogens, despite more than 50 years of continual use.
Miller pointed out, however, that resistance to streptomycin is common in xanthomonas, a plant pathogen that causes bacterial spot on tomatoes.
"So it is a concern that resistance to streptomycin may be widespread in bacterial pathogens of tomatoes and other crops where antibiotic use is common, and this resistance may be transferred to human or animal pathogens," she said.
Taylor said Plantwise will continue to collect data from trained agriculture extension workers around the world, and he and his colleagues hope to be able over time to catalog the various types of antibiotics used on crops and to monitor their use.
