Carmoisine

A randomised, double-blind, placebo-controlled trial in 153 three-year-olds and 144 eight-to-nine-year-olds found that two mixtures of artificial colours (including E122 in both mixtures) plus sodium benzoate were associated with increased hyperactivity scores compared with placebo. Effects were inconsistent across age groups and mixtures.

EFSA's AFC Panel concluded the Southampton study provided limited evidence of a small effect on activity and attention in some children, but inconsistencies across age groups and mixtures meant the ADIs for the colours could not be revised. No biological mechanism was identified.

The mandatory EU and UK warning label 'may have an adverse effect on activity and attention in children' applies to all food and drink containing E122, under Annex V of assimilated Regulation (EC) No 1333/2008, in force from July 2010.

A 2010 EFSA scientific opinion on food azo colours concluded that no data on sensitivity to carmoisine (E122) are available and no well-documented cases of intolerance reactions after oral exposure have been reported, though the panel acknowledged this could partly reflect underreporting.

EFSA's 2009 re-evaluation noted that azo dyes as a class are associated with pseudo-allergic reactions in aspirin-intolerant individuals, but did not identify specific documented cases for carmoisine in food use.

Oral administration of carmoisine to young male rats over 30 days increased liver enzymes (ALT, AST, ALP) and kidney markers (urea, creatinine) and reduced antioxidant capacity, with more pronounced effects at higher doses.

Mice given carmoisine at 200 or 400 mg/kg body weight per day for 120 days showed significant liver and kidney damage, anaemia, and upregulation of apoptosis-blocking genes (Bcl-x, PARP) with downregulation of tumour-suppressor p53 in liver tissue. The low-dose group (4 mg/kg, equivalent to the ADI) showed no adverse effects.

EFSA's 2009 re-evaluation found all genotoxicity tests for carmoisine were negative, no structural alert was present, and carcinogenicity studies in rats and mice showed no increase in tumour incidence. The ADI of 4 mg/kg body weight per day was not revised.

In vitro tests on human peripheral lymphocytes found that carmoisine induced chromosome aberrations, sister chromatid exchanges and increased DNA tail length (comet assay) at the highest tested concentration (300 micrograms per mL), but not at lower concentrations. The researchers called for further in vivo studies.

EFSA's 2009 re-evaluation found that at the 95th percentile of exposure, using either maximum permitted levels or reported industry use levels, intake estimates for 1- to 10-year-old children could be above the ADI of 4mg/kg body weight per day.

A refined 2015 EFSA exposure assessment using actual industry use data and EU food consumption data found the ADI was not exceeded for any population group at mean or high exposure levels. However, under maximum permitted level scenarios, exposure estimates still exceeded the ADI for toddlers and children at the high level.