REPLY
28 April 2003
Dear Editor,
We are surprised at the tenor of the letter from Professor D Roberts, the Scientific and Technical Director of the Australian Food and Grocery Council (AFGC) in that the chief executive of the organization welcomed the study and encouraged more research be performed1, something we also recommended in the manuscript.
We were extremely conservative in our interpretation of the results, concluding that irritability, restlessness, inattention and sleep disturbance ‘in some children may be caused by a preservative in healthy foods consumed daily’. That's all. We leave it for the readers to decide whether the association we found warrants further attention. However, if the AFGC shares our concerns perhaps they might consider funding additional research.
We agree with Professor Marshall and Mr Bulsara's comment2 on the paucity of human literature on the effects of low doses of propionic acid; however, the association between the very high levels of propionic acid seen in some metabolic diseases and severe neurological problems are well recognized in paediatric medicine. In addition, a number of studies in rats suggest that early administration of propionic acid alters normal development and induces long-lasting behavioural deficits, and that administration of ascorbic acid can prevent the behavioural alterations provoked by propionic acid3−8. While the propionic acid doses used were around four times higher than might be expected from bread alone in the average diet for a child in Australia, the effects seen were marked (http://members.ozemail.com.au/~sdengate/propionateresearch.htm). It seems extraordinary that this widely used substance has not been tested more extensively in humans, especially as neurological presentation is not always associated with metabolic crises9.
We share Professor Marshall and Mr Bulsara's concerns regarding washout periods. This is why we allowed a full 4-day washout as opposed to the usual 24 h in other studies of this type10,11.
It is important that randomized controlled trials, even small ones with equivocal or negative results, be reported. Perhaps other clinical trials have been performed showing the safety of this additive but with an extensive literature search we were unable to find them. Conversely, our results are consistent with a published study in which the behavioural effects of calcium propionate were observed in 38% of 42 hyperactive children during a larger investigation of food intolerance12,13.
Despite the small study numbers, the raw data takes several pages and requires the appropriate scales for interpretation. It is not common practice in manuscripts of this length to include such detail. However, we have made this information available for those who might be interested.
Parents were found to be reliable observers and raters of their children's behaviour by Rowe and Rowe11. There is no evidence for the common assumption that teacher ratings are more objective than parent ratings. As described in the manuscript, teacher ratings for this study were abandoned when found to be incomplete due to teacher absence, ratings on incorrect days or weeks, incorrect reading of forms, and transfer of children to other classes.
We do not accept the supposition that because a compound is ubiquitous it is safe, especially in higher than environmental doses. Vitamins A, D and iron are just three examples of ubiquitous compounds that can be highly toxic in excess. Furthermore, with the introduction of the Food Standards Code 200214, Australia has the highest permitted maximum level of propionate preservatives in bread in the world (4 g/kg) and maximum levels are frequently used. The addition of propionic acid and its salts in Australian foods such as fruit and vegetable juices, fruit and vegetable products and sauces and toppings are limited only by Good Manufacturing Practice (GMP), so the actual weekly intake by Australian children of propionic acid and its salts is unknown.
Cognitive and behavioural effects of propionic acid and its salts were not tested during the World Health Organization's toxicological investigations mentioned by Professor Roberts15. Where are the studies to show that this preservative is safe for the developing brain? Mr Bulsara has described the extremely high and increasing use of pyschostimulant medication in Australia16. Exploration of different causes for the epidemic of behavioural problems in Australia, as well as non-pharmaceutical interventions, therefore seem appropriate.
