In contrast to coeliac disease where ingested gluten levels lower than 50 mg/d have been shown to be tolerated(Reference Catassi, Fabiani and Iacono1), no safety thresholds of exposure to cereal proteins have been established in allergy to cereals. Therefore, concern has arisen whether potential trace amounts of proteins in products derived from cereal starch could induce allergic symptoms in cereal-allergic subjects.
The prevalence of true food allergy and self-perceived adverse reactions to food show large discrepancies(Reference Young, Stoneham, Petruckevitch, Barton and Rona2–Reference Woods, Stoney, Raven, Walters, Abramson and Thien4). Epidemiological studies indicate that only about one third of perceived food reactions in children and 10 % of those in adults are due to food allergy, i.e. abnormal immunological reactions to food proteins(Reference Teufel, Biedermann, Rapps, Hausteiner, Henningsen, Enck and Zipfel5). Food allergy can manifest as immediate hypersensitivity responses, usually mediated by specific IgE antibodies to food proteins, or delayed type hypersensitivity reactions, mediated by T cells directed against food-derived peptides. In the worst case, food allergic reaction may generalise to several organs, leading to anaphylaxis.
Food allergies are most common among young children, affecting about 8 % of them(Reference Bock6). Many food allergies disappear over time, as tolerance to the offending allergens develops(Reference Wood7). Wheat is among the most common food items causing allergic reactions in early childhood(Reference Allen, Hill and Heine8), and immune cross-reactivity has been shown between wheat, barley and rye(Reference Palosuo, Alenius, Varjonen, Kalkkinen and Reunala9). In general, wheat-allergic children react to lesser amounts of allergen than do adults(Reference Hirschenhuber, Crevel, Jarry, Maki, Moneret-Vautrin, Romano, Troncone and Ward10).
The aim of this study was to find out, using double-blind placebo-controlled food challenge (DBPCFC), which is the gold standard of food-allergy diagnosis, whether glucose syrup produced from barley (barley starch syrup) can cause allergic reactions in individuals with allergy to wheat, barley, rye or oats.
Materials and methods
Patients
Fifteen patients with allergy to at least one of the cereals wheat, rye, barley or oats, confirmed by DBPCFC, were enrolled in DBPCFC testing with barley starch syrup. The patients were recruited from the paediatric outpatient clinics in the catchment area of the Turku University Central Hospital. If allergy to the above cereals was proved by DBPCFC within 3 months before the first study visit, the patient was directly eligible for DBPCFC with barley starch syrup. Otherwise, the presence of allergy to wheat, barley, rye or oats was studied in DBPCFC before the subject underwent DBPCFC with barley starch syrup.
Clinical examination of the patient was carried out at each study visit. The diagnostic criteria of Hanifin & Rajka(Reference Hanifin and Rajka11) were used to determine the possible presence of atopic dermatitis, and scoring atopic dermatitis (SCORAD) assessment(12) to determine its severity.
The permission of the Ethics Committee of the Hospital District of South-western Finland was obtained for the study and informed, written consent was received from the participants and/or parents.
Double-blind placebo-controlled food challenge with wheat, barley, rye and oats
Water-based gruel prepared of a flour mixture containing wheat, barley, rye and oats (four-cereal gruel) in equal amounts was used in DBPCFC to detect the presence of cereal allergy. Water-based placebo gruel containing rice and buckwheat was mixed with the four-cereal gruel (1:1) in order to mask the appearance and taste of the latter. If the patient was known or suspected to be allergic to buckwheat, placebo gruel made of corn was used instead.
DBPCFC with the four-cereal gruel, masked with the placebo, was carried out after the cereals had been eliminated from the diet for at least 1 week. In DBPCFC, gruel containing four cereals and mere placebo gruel were served in randomised order. Increasing doses of 3 g, 15 g and 30 g of cereal in gruel were given at 20 min intervals, provided that no objective allergic symptoms appeared with the lower dose. The doses of 3 g, 15 g and 30 g of the four-cereal gruel contained 0·17 g, 0·85 g and 1·70 g protein, respectively. The subjects were monitored for immediate reactions in the study unit for at least 1 h after the last challenge dose, and if symptoms appeared, for as long as their clinical status required. If no immediate symptoms occurred, administration of gruel made with 30 g flour was continued at home twice daily, for a maximum of 5 d, to monitor the development of delayed type reactions. If any symptoms developed during the home-consumption period, the patient was advised to attend the study unit for a clinical assessment, and, if objective allergic symptoms were seen, the challenge testing was discontinued. If allergic symptoms developed with exposure to four-cereal gruel but not to placebo, the patient was judged as having cereal allergy to wheat, rye, barley or oats.
Double-blind placebo-controlled food challenge with barley starch syrup
DBPCFC with barley starch hydrolysate (isoglucose), Neste 70 FSS (Finnsugar Ltd, Jokioinen, Finland) and saccharose solution, Neste 67 S (Finnsugar Ltd, Kantvik, Finland) used as placebo, was conducted according to the regimens described earlier. Fresh test solutions were provided every 2 weeks, since the shelf life of the solutions was about 3 weeks. The test solutions came from the normal manufacturing line and represented the typical commercially available solutions. The N content of barley starch hydrolysate was regularly analysed by the Kjeldahl method. Ten lots were used for the study.
Three increasing doses of barley starch hydrolysate and saccharose solution, 5 g, 10 g and 15 g of both, were used in the challenges. Unless objective allergic symptoms occurred, administration of test solutions was continued at home, with 15 g twice daily, for a maximum of 5 d.
Skin-prick tests
To study IgE-mediated sensitisation, skin-prick tests (SPT) were carried out with wheat, barley, rye, oats (each as 10 % (w/v) in physiological saline), barley starch hydrolysate as undiluted standard foodstuff (Finnsugar, Kantvik, Finland) and 1 mg/ml gliadin, which is the major allergen of wheat(Reference Palosuo, Varjonen, Kekki, Klemola, Kalkkinen, Alenius and Reunala13). Commercial cereal grains were milled to produce cereal flour for the allergen extract. Histamine dihydrochloride (10 mg/ml; ALK, Copenhagen, Denmark) was used as a positive control and physiological saline as a negative control. The perpendicular diameters of each SPT reaction were recorded after 15 min, and the results were expressed as mean wheal diameter. Mean wheal diameters ≥ 3 mm were considered positive.
Statistics
The data were analysed using SPSS 13.0 statistical software (SPSS Inc., Chicago, USA). In subjects with atopic dermatitis, the Friedman test was used to compare the mean of SCORAD indices during the barley starch syrup and placebo challenges, and the Wilcoxon signed ranks test was used to compare changes in SCORAD indices during the challenges.
Results
The age range of the study subjects allergic to wheat, barley, rye or oats was 0·9–13·8 years (median 2·1 years). Nine of them were boys. Table 1 shows the symptoms of the allergic patients induced by cereals, but not by placebo, in DBPCFC as well as the SPT results to cereals and gliadin. The diluent of the allergen solutions, saccharose and barley starch syrup gave negative results on SPT in all patients. Seven (47 %) of the patients showed immediate allergic reactions (within 1 h after the last dose at the study unit) in DBPCFC with cereals, and eight of them had delayed type reactions (later than 1 h after the last dose at the study unit). One patient experienced an anaphylactic reaction (generalised urticaria, abdominal pain and vomiting) after the first challenge dose of four-cereal gruel. He was treated with intramuscular epinephrine, oral prednisolone and antihistamine and recovered in a few hours.
Table 1 Characteristics of the patients with allergy to cereals and their skin-prick test results with wheat, barley, rye, oats and gliadin*
M, male; F, female; DBPCFC, double-blind placebo-controlled food challenge; AD, atopic dermatitis.
* For details see Materials and methods.
† Delayed-type allergic reaction, i.e. emerging after 1 h after the last challenge dose.
‡ Immediate allergic reaction, i.e. emerging within 1 h after the last challenge dose.
After normal refining process the N content of barley starch hydrolysate was 1–30 mg N in 1 kg fresh-weight barley starch hydrolysate(Reference Kilpi14). In DBPCFC with barley starch hydrolysate, none of the fifteen patients with allergy to cereals developed allergic symptoms. Regarding atopic dermatitis, the mean values of SCORAD indices showed no significant differences between barley starch syrup and placebo challenges. Neither did changes in SCORAD indices during the two challenge periods show any significant differences. The result of the study provided statistical power of 98 % certainty that at least 90 % of cereal allergic patients will not have any allergic reaction when exposed to barley starch syrup.
Discussion
In the present study, none of the cereal-allergic patients showed objectively detectable symptoms when exposed to barley starch syrup in a double-blind placebo-controlled manner. Notably, even the patient showing the most intense IgE-mediated sensitisation to cereals on SPT and having the most severe form of allergic reaction, anaphylaxis, when exposed to cereals, had no symptoms on the barley starch syrup challenge.
Barley starch syrup is a commercial product mainly used as a sweetener in non-alcoholic or alcoholic beverages, confectionery, bakery products and food sauces such as mustard and ketchup. In our study, the daily challenge dose of 30 g barley starch syrup was estimated to correspond to maximal daily consumption(Reference Kilpi14). According to the laboratory assessments, this amount of barley starch syrup contains a maximum of 0·9 mg N, possibly as degradation products of lysophospholipids or derived from some unknown non-protein source(Reference Kilpi14). However, using the generally accepted conversion factor of 6·25 for dietary N(Reference Crim, Munro, Shils, Olson and Shike15), it can be calculated that 0·9 mg N equals 5·6 mg of protein. On this basis, it could be concluded that barley starch syrup is tolerated by coeliac patients, but its safety in cereal-allergic patients could not be determined without a clinical trial. The fact that none of the study patients showed positive SPT results with barley starch hydrolysate further confirmed the suggestion that barley starch hydrolysate does not contain any intact proteins or peptides, and the results of DBPCFC finally confirmed the safety aspect.
To our knowledge, NESTE 70 FSS is the only barley-starch based glucose syrup on the market. Instead, due to the higher starch content e.g. wheat is frequently used as a source of glucose syrup in the areas where climate conditions favour the growth of wheat. Low amounts of residual gluten and peptides (0·3–1·4 mg/kg) have been reported in wheat-starch based glucose syrup(16).
Peanut oil is the most thoroughly studied among the processed food stuffs that may contain residual allergenicity. In the study of Olszewski et al. (Reference Olszewski, Pons, Moutété, Aimone-Gastin, Kanny, Moneret-Vautrin and Guéant17) the residual proteins of peanut oil showed immunologic activity in vitro and positive SPT results in some individuals, but the implication of these proteins in clinical allergic reactions was not completely established in the absence of DBPCFC. In general, it is now concluded on the basis of several other well-defined studies that refined peanut oil can be safely consumed by peanut-allergic individuals, whereas unrefined oil can provoke reactions in some of the same individuals(Reference Crevel, Kerkhoff and Koning18).
In conclusion, this clinical study of ours confirmed the safety of barley starch syrup in at least the vast majority of cereal-allergic patients. Further studies are needed to address the same issue for several other foodstuffs derived from potentially allergenic sources.
Acknowledgements
We thank Eija Hoppendorff-Koskinen, RN, for carrying out the challenge and skin-prick tests, and product specialist Kyllikki Kilpi, Finnsugar Ltd, Kantvik, Finland, for providing the data about the study product. This study was financially supported by Finnsugar Ltd, Finland. The sponsor of the study and the authors had no conflict of interests.
