Bonvallot and colleagues described an interesting exploratory metabolomics study of 83 pregnant women in an agricultural region in France [1]. These comprised a subset of subjects from a larger study, including all those who had a live birth and provided a urine sample in 2004 during the 11th week of their pregnancy. The women were a priori divided into three groups depending on the percentage of land surface dedicated to cereal crops in their town of residence (Group 0: 0–17%, Group 1: between 17 and 25% and Group 2: above 25%).

The authors found an increased odds ratio for some urinary metabolite changes (i.e. glycine, threonine, lactate and glycerophosphocholine, which were all increased, and for citrate which was lower) for Groups 1 and 2 compared to Group 0, after adjusting for maternal age, parity, body mass index and smoking habits. They conclude that their study, “suggests that an exposure to complex pesticide mixtures induces modifications of metabolic fingerprints”.

We acknowledge the importance of this study as an exploration of metabolic changes in communities surrounded by agricultural land and the difficulties in assessing exposure to pesticides, many of which have a very short biological half-life in the body. However, we consider that the authors have extended the interpretation of their data beyond that which can be scientifically sustained. In particular, there is no strong evidence that the surrogate measure of exposure used in the study has validity in distinguishing actual differences in either acute or chronic exposure to agricultural pesticides.

There are many different chemical compounds used in modern pesticides and individuals may be exposed to these agents from a diverse range of sources, e.g. in the food consumed, from occupational or para-occupational exposure, from use of home or garden preparations and from proximity to agricultural land where some of the applied pesticide may inadvertently contaminate their home or garden. So far as we are aware there are no published studies to conclusively link living close to agricultural land as a proxy for exposure, although we are currently undertaking a study of this type [2].

Bonvallot et al cite three studies in support of an association between proximity to agricultural land and pesticide exposure. However, these studies were undertaken in California and Chile where agricultural practice and pesticide usage and regulation are different from Europe [3, 4, 5]. Also, the studies only provide weak evidence of an association between the proportion of agricultural land in a community and pesticide exposure. For example, Gunier et al [5] developed two complex metrics of exposure to pesticides based on the density of pesticide use in the community, which they compared to pesticide residues measured in carpets of the homes of their subjects (as a surrogate for human exposure). Even with these more specific exposure metrics there was only about 10 – 15% of the carpet residue concentrations explained by the pesticide usage.

The authors refer to unpublished data from their larger study that shows urinary metabolites of fungicides applied to cereal crops is associated with the percentage of these crops in the town of residence. However, it would have been more convincing if the authors had used the corresponding analyses of the samples from the 86 women described in this paper, and directly investigated the association between the pesticide metabolites and the other metabolic markers.

We strongly advocate that metabolomic studies of environmental exposure to pesticides should focus on co-measurement of primary metabolic biomarkers of the specific pesticides to which the subjects have been exposed.

Sincerely,

John W Cherrie 1
Karen Galea 1
Kate Jones 2
Martie van Tongeren 1
John Cocker 2

1. Institute of Occupational Medicine, Centre for Human Exposure Science, Research Avenue North, Edinburgh EH14 4AP, UK
2. Health and Safety Laboratory, Harpur Hill, Buxton, Derbyshire SK17 9JN, UK


References

1. Bonvallot N, Tremblay-Franco M, Chevrier C, Canlet C, Warembourg C, et al. (2013) Metabolomics Tools for Describing Complex Pesticide Exposure in Pregnant Women in Brittany (France). PLoS ONE 8: e64433. doi:10.1371/journal.pone.0064433.t003.

2. Galea KS, MacCalman L, Jones K, Cocker J, Teedon P, et al. (2011) Biological monitoring of pesticide exposures in residents living near agricultural land. BMC Public Health 11: 856. doi:10.1186/1471-2458-11-856.

3. Bradman A, Castorina R, Boyd Barr D, Chevrier J, Harnly ME, et al. (2011) Determinants of Organophosphorus Pesticide Urinary Metabolite Levels in Young Children Living in an Agricultural Community. Int. J. Environ. Res. Public Health 8: 1061–1083. doi:10.3390/ijerph8041061.

4. Muñoz-Quezada MT, Iglesias V, Lucero B, Steenland K, Barr DB, et al. (2012) Environment International 47: 28–36. doi:10.1016/j.envint.2012.06.002.

5. Gunier RB, Ward MH, Airola M, Bell EM, Colt J, et al. (2011) Determinants of Agricultural Pesticide Concentrations in Carpet Dust. Environ Health Perspect 119: 970–976. doi:10.1289/ehp.1002532.