Current pesticide dietary risk assessment in light of comparable animal study NOAELs after chronic and short-termed exposure durations

Abstract

Dietary risk assessment (DRA) of pesticides includes the estimation of chronic and acute exposures from crop residues, but assesses acute exposures only for pesticides with an acute reference dose (ARfD). Acute estimation uses high percentiles of food consumption surveys which are considerably higher than per capita lifetime averaged food consumption values which are used for chronic estimations. Assessing acute risks only for pesticides with an ARfD tacitly assumes that chronic risk assessment covers also intermittent occurring exposures which could significantly exceed chronic estimates. The present investigation conducted on 2200 rat studies from 436 pesticides provides evidence demonstrating that pesticides with and without ARfD have no-observed-adverse-effect levels (NOAELs) which remain statistically unchanged in developmental, subacute, subchronic, reproductive and chronic toxicity studies covering exposure durations between 2 and 104 weeks. DRA of pesticides without ARfD needs reconsideration in light of equally high toxic dose levels after short- and long-term exposures, suggesting that intermittent exposures could be toxic, if they repeatedly exceed the acceptable chronic daily intake (ADI; conceptually the human counterpart of chronic animal NOAEL). As such risks are currently not assessed for pesticides without ARfD, the current DRA concept, which automatically presumes the use of low chronic exposure estimates entirely covers the risks of not acutely toxic pesticides, needs reconsideration. Furthermore, risks to intermittent occurring high exposures are probably also insufficiently assessed for pesticides where the ARfD is significantly higher than the ADI. As an example, the maximum residue limit for bifenazate in peaches is discussed.

Keywords: Dietary risk assessment; Intermittent exposure; NOAEL; Pesticides; Time extrapolation.

Fig. 1

Cumulative distributions of lowest NOAELs for pesticides. The distributions of segregated subsets are presented: pesticides with an ARfD; pesticides with an ARfD but excluding organophosphates (OP) and carbamates (carb); pesticides without an ARfD

Fig. 2

Cumulative distributions of ratios of shorter-term study NOAEL (or LOAEL) to chronic study NOAEL (or LOAEL). Only studies with a dose spacing ≤8 were used to minimize the influence of dose spacing on the NOAEL (or LOAEL) ratio distributions. The types of studies labeled on the graphs were those used to calculate the NOAEL (or LOAEL) ratios. For example, the curve labeled “reproductive” is the cumulative distribution of ratios of reproductive study NOAELs (or LOAELs) to chronic study NOAELs (or LOAELs) for all pesticides. Except for distributions involving developmental studies, all ratio distributions were calculated using the applied concentration of the pesticide in the feed (i.e., ppm). In distributions of ratios of developmental NOAELs (or LOAELs) to chronic study NOAELs (or LOAELs), the NOAELs (or LOAELs) were given as doses (i.e., mg/kg bw per day), because the dose in developmental toxicity studies is usually applied by gavage as a constant dose. In the curves labeled “developmental; adj.”, the chronic NOAELs (or LOAELs) were additionally adjusted to the dose that the animals received at the beginning of the study to account for the dose decrement in feeding studies. For further explanation of this step, see “Study design factors affecting NOAEL ratio distribution”

Fig. 3

Cumulative distributions of ARfD/ADI ratio (295 compounds) and AOEL/ADI ratio (419 compounds), when AOELs are applied as reported (i.e., if appropriate, being corrected for oral absorption, which applies to 20% of the compounds). The graph of ARfD/ADI ratios starts with approximately 15% at a ratio of 1 because, by definition, an ARfD has to be greater than or equal to the ADI

Fig. 4

Cumulative distribution of ratios of acute to chronic exposure for 163 food commodities. The chronic (WHO cluster diet E, representing a considerable part of Europe) and acute (highest reported exposure) food consumption data for exposure estimates are from the “PRIMo rev2” exposure model of EFSA (EFSA 2016)