Capabilities of EU labs to determine inorganic arsenic in food have improved

Maximum levels for heavy metals in mushrooms based on wet weight are set by the European Regulation (EC) No. 1881/2006. For common, oyster and shiitake mushroom the maximum levels are: 0.20 mg kg-1 cadmium and 0.30 mg kg-1 for lead. For other species the maximum level for cadmium of 1 mg kg-1 applies.

No maximum levels have been set yet for inorganic arsenic and methylmercury, although they are the most toxic species of arsenic and mercury, respectively. The reason for such discrepancy is that legislators still discuss whether analytical laboratories are able to supply reliable data for the speciation analysis rather than total element analysis. Also, not much information is available in the literature for metal speciation in mushrooms. Both, European Food Safety Authority (2009, 2014) and the Joint FAO/WHO Expert Committee on Food Additives (JECFA) (2011) have recently shown their interest in the content of iAs in food.

Since mushroom consumption has increased in the last years due to their nutritional properties, DG SANCO of the European Commission requested the EURL-HM test analytical capabilities of National Reference Laboratories (NRLs) to determine some  metal and metalloid species in (shiitake) mushrooms.

For this reason, the Institute for Reference Materials and Measurements (IRMM) of the Joint Research Centre (JRC) initiated two proficiency tests as part of their International Measurement Evaluation Program (IMEP): IMEP-116 and IMEP-39.
While participation in IMEP-116 was restricted to NRLs appointed by national authorities in EU member states, IMEP-39 was open to all laboratories. Thirty-seven participants from 25 countries gave results in IMEP-116, and 62 laboratories from 36 countries reported for the IMEP-39 study.

The test item used in both PTs was a blend of mushrooms of the variety shiitake (Lentinula edodes). Five laboratories, with demonstrated measurement capability in the field, provided results to establish the assigned values (Xref). Laboratory results were rated with z- and zeta- scores in accordance with ISO 13528. “The percentage of satisfactory z-scores ranged from 81% (iAs) to 97% (total Cd) in IMEP-116 and from 64% (iAs) to 84% (total Hg) in IMEP-39,” found the study.

The first observation is that the percentage of laboratories reporting realistic uncertainties for all measurands is higher in IMEP-116 than in IMEP-39. Interestingly, participants in IMEP-116 more often overestimated the uncertainty, while laboratories in IMEP-39 tended to underestimate the uncertainties.

Methylmercury was determined during the preparatory phase using the method validated in a collaborated trial (IMEP-115). For methylmercury, an approximate concentration of 0.0042 mg kg-1 was found, which corresponds to about 5% of the total content of Hg in the mushroom sample. However this result can only be considered as approximate because the value is below the limit of quantification of the method.

The screening for iAs indicates that about 50% of the total As mass fraction is present in the form of iAs. Other species present were dimethylarsinic acid (DMA) and monomethylarsonic acid (MMA). The presence of arsenobetaine was not reported, however the methodology used for the speciation analysis was not optimum to determine arsenobetaine.

For total concentrations two clear tendencies were observed only in IMEP-39:
Participants using atomic absorption spectrometry (AAS)-based techniques reported lower values for As than those who used ICP-based techniques (ICP-MS and ICP-AES). It is speculated that low recovery for AAS-based determination is due to the sensitivity of this technique for incomplete digestion especially when using hydride generation-AAS (see our summary on: "Speciation matters even if the interest is in total element concentration").

The influence of the technique used was not so significant for the total Cd, Pb and Hg mass fractions. Unfortunately, A relatively high number of laboratories overestimated the total Pb and Hg mass fractions. Contamination during sample preparation could be a possible explanation.  

Interestingly, the results for iAs where much better than in previous PTs. Sixteen NRLs reported values for this measurand and 81% of these obtained a satisfactory z-score. On the other hand, five out of the seven laboratories that obtained satisfactory z-scores for iAs determination in IMEP-39, used AAS-based techniques. This clearly shows, that  AAS-based methods can be used for speciation analysis and they are cheap and easy-to-use methods which can provide correct results if proper method validation is carried out.