In the framework of Articles 23 and 33 of Regulation (EC) No 178/2002 the European Food Safety Authority (EFSA) has received from the European Commission (EC) the mandate to collect on a continuous basis all available data on the occurrence of chemical contaminants in food and feed and to periodically produce topical reports on the occurrence of specific contaminants (mandate M-2010-0374). The present report includes an overview of the occurrence of 3-monochloropropane-1,2-diol (3-MCPD) in food. The report addresses data collected in European Member States from 2009 to 2011 and submitted to EFSA.
3-MCPD is a food processing contaminant formed by heat as a reaction product of triacylglycerols, phospholipids or glycerol and hydrochloric acid in fat-based or fat-containing foods. Depending on the type of food it may occur as a free substance, in the form of an ester with fatty acids or in both forms (Svejkovská et al., 2004). The main target organ for 3-MCPD toxicity is the kidney, with chronic oral exposure resulting in nephropathy and tubular hyperplasia and adenomas (as reviewed by the Joint FAO/WHO Expert Committee on Food Additives, JECFA (JECFA, 2002; in particular Sunahara et al., 1993). The Scientific Committee on Food (SCF) established a tolerable daily intake (TDI) for 3‑MCPD of 2 µg/kg body weight (b.w.) (SCF, 2001). The International Agency for Research on Cancer (IARC) has classified 3-MCPD as a “possible human carcinogen (group 2B)” (IARC, 2012). The Joint FAO/WHO Expert Committee on Food Additives (JECFA) established for the free compound a provisional maximum tolerable daily intake (PMTDI) of 2 µg/kg b.w. (JECFA, 2002, 2007). Commission Regulation (EC) No 1881/2006 established for hydrolyzed vegetable protein (HVP) and soy sauce a maximum level of 20 μg/kg for liquid products containing 40 % dry matter, corresponding to a maximum of 50 µg/kg in the dry matter.
Studies have shown that the highest occurrence of 3-MCPD in food is in food categories where it is ester-linked with fatty acids in the form of mono- or di-esters (Svejkovská et al., 2004; Weisshaar, 2011; Crews et al., 2013). High levels of 3-MCPD esters are found in edible refined plant oils and fats, and composite food containing them. By the action of lipases, 3-MCPD can be released from the esters in vivo and a recent study (Abraham et al., 2013) supports the assumption of equivalent oral bioavailability of 3-MCPD in the free form and in the ester form (BfR, 2007; EFSA, 2008).
As for free 3-MCPD, official methods are not established for the analytical determination of 3-MCPD esters. Scientific activity has taken place in recent years to develop and validate efficient and reliable methods for the analysis of 3-MCPD in its forms (reviewed in Wenzl et al., 2007; Baer et al., 2010; Crews et al., 2013). Depending on the analytical method, a broad range of limits of detection (LOD) and limits of quantification (LOQ) is encountered. Parallel with analytical method development the occurrence of 3-MCPD in different foodstuffs has been investigated.
The data for the present report have been extracted from the EFSA chemical occurrence database, and include data from sampling years 2009 to 2011. The reasons for choosing a limited time frame relate to the improvement in data collection introduced by the Standard Sample Description, SSD (EFSA, 2010a), and the improvement of the analytical methods, to reduce the possible overestimation of occurrence experienced in older methods (Weisshaar and Perz, 2010; Weisshaar et al., 2011).
After extraction and data cleaning, the 3-MCPD dataset included 1 235 analytical results submitted by 14 European countries relating to 11 food groups. Considering also the sub-groups of the 11 groups, a total of 45 food groups were considered in the present assessment.
For all analytical results, it was assumed that reported values represented the total amount of 3-MCPD in the respective sample. No distinction was made in the data analysis between 3-MCPD (free) and 3-MCPD (from ester bounds) based on the assumption of equivalent oral bioavailability (Abraham et al., 2013).
A broad range of the left-censoring limit (LC limit) (i.e. limit of detection - LOD and limit of quantification - LOQ) was observed among food groups and in some cases even within the same food group; the LC limits ranged between 1 and 1000 µg/kg. The 727 left-censored results (LC results) (analytical results reported either as < LOD or < LOQ), corresponding to 59 % of all results, were treated using the substitution method (WHO, 2009; EFSA, 2010b). Three scenarios were considered: lower bound (LB) where the LC results were replaced with 0, upper bound (UB, where they were replaced with the LC limits and middle bound (MB) where they were replaced with half of the LC limit. The mean occurrence of 3-MCPD in different food groups with LC results was calculated as a MB estimate and range [MB (LB – UB)].
High mean [MB (LB – UB)] occurrence of 3-MCPD was observed in the group ‘Animal and vegetable fats and oils’ [1 020 (960 – 1 090) µg/kg] and in its sub-groups: ‘Margarine and similar products’ [1 500 (1480 – 1 530) µg/kg], ‘Fats of mixed origin’ [1 230 (1170 – 1 290) µg/kg] and ‘Vegetable fats and oils’ [820 (740 - 890) µg/kg]. Inside the food group ‘Vegetable fats and oils’, mean occurrence of 3-MCPD was separately evaluated in ‘Walnut oil’ [4 750 (4 740 – 4 750) µg/kg] and in ‘Vegetable fats and oils (excluding walnut oil)’ [430 (350 - 520) µg/kg]. The latter group includes ‘Vegetable fats and oils refined’ [530 (460 - 600) µg/kg], ‘Vegetable fats and oils unrefined (excluding walnut oil)’ [250 (200 - 310) µg/kg] and ‘Vegetable fats and oils unspecified refining (excluding walnut oil)’ [450 (350 - 540) µg/kg].
Mean occurrence of 3-MCPD in other food groups was as follows: ‘Grains and grain-based products’ [33 (26 - 39) µg/kg], ‘Fish meat (smoked)’ [37 (36 - 37) µg/kg], ‘Preserved meat (smoked)’ [47 µg/kg, no LC results] and ‘Herbs, spices and condiments’ [9.2 (6 - 12) µg/kg]. In the group ‘Herbs, spices and condiments’ values higher than the average of the group were found in the sub groups ‘Dressings’ [120 (110 - 140) µg/kg] and ‘Mixed spices’ [25 (23 - 26) µg/kg]. High single values were reported for ‘Infant formulae, powder’ (300 µg/kg, N = 1), ‘Follow-on formulae, powder’ (800 and 1200 µg/kg, N = 2) and ‘Potatoes and potato products’ (390 µg/kg, N = 1).
A preliminary estimate of the mean and 95th percentile (total population) chronic dietary exposure to 3-MCPD of the different age classes across European surveys (population groups) was based on the mean MB occurrence in 15 food groups. Food groups with too few analytical results (1 or 2 results; N = 10) were not considered in the exposure assessment. For the majority of population groups, mean dietary exposure to 3-MCPD was below 1 µg/kg b.w. per day (60 out of 64 population groups). In two dietary surveys on ‘Toddlers’ and two on ‘Other children’ mean dietary exposure was between 1 and 1.5 µg/kg b.w. per day. In the case of high consumers, P95 exposure was below 1 µg/kg b.w. per day in approximately half of the population groups (30 groups out of 64 corresponding to 47 %), while it was between 1 and 2 µg/kg b.w. per day in 26 population groups. In eight population groups (three on ‘Other children’, three on ‘Toddlers’, one on ‘Elderly’ and one on ‘Very elderly’), P95 exposure was between 2 and 3 µg/kg b.w. per day.
The contribution of the food groups to the overall mean exposure in the different population groups was also evaluated: ‘Margarine and similar products’ appears among the top three contributors to 3-MCPD exposure in 47 population groups out of 64 (73 %), with a contribution to the total exposure in the range 13-83 %; ‘Bread and rolls’ in 55 population groups (86 %), with a contribution to the total exposure in the range 6-26 %; ‘Vegetable fats and oils (excluding walnut oil)’ in 34 population groups (53 %), with a contribution to the total exposure in the range 5-74 %; ‘Fine bakery wares’ in 30 population groups (47 %) with a contribution to the total exposure in the range 4-29 %; ‘Preserved meat (smoked)’ in 17 population groups (27 %) with a contribution to the total exposure in the range 3-18 %. In particular, ‘Margarine and similar products’ is the main contributor to the exposure in 45 population groups (70 %) and ‘Vegetable fats and oils (excluding walnut oil)’ is the main contributor to the exposure in 18 population groups (28 %).