Idiosyncratic reactions are adverse reactions to foods that occur through unknown mechanisms. As suggested by this definition, idiosyncratic illnesses are poorly understood. In fact, the cause-and-effect relationship implicating the food or food ingredient as the cause of the illness is often quite weak. Many of the adverse reactions associated with food additives including antioxidants (BHA and BHT), artificial colors (tartrazine and sunset yellow), monosodium glutamate (MSG), and others fall into this category but, as noted, the cause-and-effect relationship is often not well established. In some cases, many clinical reports do exist but the vast majority are anecdotal and few have rigorously tested the hypothesis that the specific ingredient is the cause of the illness. With this wide range of substances and the possibility of a number of different mechanisms, a range of symptoms could occur.
Sulfite-Induced Asthma
Sulfites definitely cause idiosyncratic reactions in some consumers through an undefined mechanism associated with the ingestion of sulfites which are common ingredients in foods and medications. The cause-and-effect relationship in this case is quite strong. Asthma is the primary symptom that has been clearly linked to sulfite ingestion in multiple subjects as the result of carefully controlled clinical challenge studies which thus have served to establish the cause-and-effect relationship. Scattered reports exist of other manifestations of sulfite sensitivity including anaphylaxis but the role of sulfites in triggering these other adverse reactions is not well established by controlled clinical challenge trials. Thus, sulfite-induced asthma will be the focus here.
Sources, Properties and Occurrence in Foods
Sulfites are common food additives used in a variety of foods and providing several technical attributes. Several forms of sulfites exist and are allowed for use in foods: sulfur dioxide, sodium metabisulfite, potassium metabisulfite, sodium bisulfite, potassium bisulfite, and sodium sulfite. All of these ingredients serve similar functions in foods because all of them have similar chemistries in foods dependent upon pH. Sulfites can also occur naturally in foods, especially fermented foods, as the result of sulfite formation by yeast. The residual levels of sulfites in foods range from <10 ppm in many food products to >2000 ppm in certain dried fruits. Naturally occurring levels of sulfites are typically quite low, <10 ppm with a few exceptions. Sulfites are added to foods for a variety of purposes including the control of enzymatic and non-enzymatic browning (e.g. potatoes), the prevention of undesirable bacterial growth (e.g. corn wet milling and wine making), the conditioning of doughs (e.g. some frozen dough products), the prevention of oxidation, and the bleaching of selected products (e.g. maraschino cherries and hominy).
The fate of sulfites after their addition to foods is complex. In acidic foods (<pH 4), sulfites can be released from foods into the surrounding atmosphere as SO2 gas. Sulfites are very reactive with a range of food components including carbohydrates, proteins, and others. These reactions can be either reversible or irreversible depending upon the nature of the reaction and existing conditions (pH, etc.). Very little free, unbound sulfite remains in most sulfited foods with a few exceptions such as lettuce.
Prevalence and Severity
Asthma is a relatively common affliction within the population. However, only a small percentage of asthmatics are sulfite-sensitive. Challenge studies indicate that severe asthmatics who rely on steroid-based drugs for the control of their asthma symptoms, are primarily at risk. Severe asthmatics make up only about 20% of the overall asthmatic population. Even among severe asthmatics, the prevalence of sulfite sensitivity is only about 5%. By extrapolating from the challenge study results and knowing the percentage of asthmatics with severe asthma, the number of sulfite-sensitive asthmatics in the U.S. can be estimated at perhaps 150,000.
Sulfites are capable of causing severe reactions in sensitive individuals. The provocation of asthma can be life-threatening in some situations. Deaths have occurred from the ingestion of sulfites by sulfite-sensitive asthmatics, although the number of deaths is rather small.
Management and Thresholds
Sulfite-sensitive asthmatics must avoid the ingestion of sulfites in their diets (and in medications). Fortunately, the presence of sulfites must be declared on the ingredient labels of packaged foods when residual levels exceed 10 ppm (10 mg/kg of food). Also, the use of sulfite is banned from many fresh food products such as lettuce where high residual levels were associated with provocation of particularly severe reactions until this use was prohibited. Sulfite-sensitive asthmatics can tolerate the ingestion of small quantities of sulfites. In controlled clinical challenge trials, the most sensitive individuals react to 5 mg of potassium metabisulfite, equivalent to 3.7 mg of free SO2. Limited challenge studies have been conducted with sulfited foods. But the results of these studies support the hypothesis that sulfite-sensitive asthmatics are more tolerant of sulfites in foods than they are of inorganic sulfites in capsules or other common challenge vehicles. Apparently, the reaction of sulfites with food components serves to lessen the hazard associated with sulfites. FARRP hypothesizes that, if the sulfites are irreversibly bound to various food components, then they are not free to trigger the adverse reaction. The tolerance for sulfited foods appears to vary with the nature of the food suggesting that the form of bound sulfite is likely an important issue. Due to the release of SO2 vapor from acidic beverages, sulfite-sensitive asthmatics may be more sensitive to sulfited beverages than to other forms of sulfite in foods. Sulfite-sensitive asthmatics also appear to be more sensitive to residues of unbound sulfites in foods such as lettuce than they are to sulfited foods that contain bound sulfites such as shrimp and potatoes.
Overall, sulfites pose a considerable risk to sulfite-sensitive asthmatics. While the size of the affected population is comparatively quite small, the reactions can be severe and life-threatening. Sulfite-sensitive asthmatics can tolerate some sulfite in the diets, but the thresholds are low in some individuals. However, the current labeling regulations appear to be sufficient to protect these sensitive consumers. Sulfite-sensitive asthmatics must invoke avoidance diets that exclude most significant sources of sulfite from their diets; they should avoid all packaged foods labeled as containing one of the sulfite ingredients.
Tartrazine (FD&C YELLOW #5)
Tartrazine is an artificial color that has been allowed for use in foods and medications for decades. Tartrazine was first implicated in idiosyncratic reactions in the late 1950's associated with its use in certain medications. Ultimately, the allegations of idiosyncratic reactions occurring to tartrazine prompted the U.S. FDA to require specific labeling of FD&C Yellow #5 even though other artificial colors could be declared under a general heading of "artificial colors".
FARRP would argue that the cause-and-effect relationship between tartrazine and several different idiosyncratic reactions remains unproven. Numerous studies have been conducted on the role of tartrazine in various idiosyncratic reactions but especially in asthma and chronic urticaria. Unfortunately, many of these studies were methodologically flawed making the interpretation of the results quite difficult. First, one must appreciate the difficulties encountered in conducting research on asthma and chronic urticaria (hives). Both asthma and chronic urticaria are chronic illnesses that flare episodically and unpredictably. Medications must be taken continuously in many cases to lessen the frequency and severity of such episodes. In any challenge study conducted on patients with asthma or chronic urticaria, reactions may result from the challenge substance (e.g. tartrazine) or may occur spontaneously simply owing to the unpredictability of the condition. Thus, it is clearly quite important to establish a cause-and-effect relationship through very careful study design. Unfortunately, many investigators were apparently not sufficiently aware of the importance of such design factors.
Tartrazine-Induced Urticaria
Tartrazine has been linked to urticaria (hives). Chronic urticaria is a condition where periodic flares occur often for unexplained reasons. Many such patients must take antihistamines to prevent the frequent occurrence of hives. While many studies seem to confirm the role of tartrazine in chronic urticaria, these studies were badly flawed. Several flaws were identified. In some studies, antihistamine drugs were withheld prior to challenge. Antihistamines are important for the control of chronic urticaria. Without antihistamines, urticarial reactions are likely to flare in these patients regardless of the treatment. On these studies, the placebo was often administered before the tartrazine and thus the study was not truly blinded. Break-through urticaria from the withdrawal of antihistamines is much more likely to occur later in the challenge sequence. Thus, positive reactions were observed with high frequency in these studies, but the reactions are likely to be false positive responses due to the order effect of always administering the placebo before the tartrazine. Several studies involved a double-blind challenge design, but even in these studies there could have been an unreported order effect based upon those patients who received placebo before tartrazine. The percentage of positive responders was lower in the double-blind challenge studies. In all of these double-blind challenge studies, antihistamines were also either withheld or no information was provided on medication status during challenge which again raises the question of break-through urticaria. Several of the studies have other complicating factors. A mixture of colors were used in several studies so that the positive reactions cannot easily be attributed to any one substance. In a single study by Stevenson et al. (1986), a series of patients with chronic urticaria whose history suggested possible tartrazine sensitivity were challenged in blinded fashion. However, in these patients, antihistamines were continued throughout the challenge period. Only 1 of 24 patients developed urticaria on single-blind challenge and this reaction was confirmed by double-blind challenge. A comparatively large challenge dose of 50 mg of tartrazine was used to provoke this response (such doses would rarely be encountered with food ingestion).
The study by Murdoch et al. (18) is somewhat puzzling and deserves further discussion. Murdoch et al. (18) identified 3 patients who reacted to a panel of azo dyes (tartrazine, sunset yellow, amaranth, and carmoisine); 2 of these 3 patients reacted to tartrazine alone in a double-blind, placebo-controlled, in-hospital challenge. Murdoch et al. (18) also documented the release of histamine and its metabolites into plasma and urine during these urticarial flares adding some credibility to a diagnosis of tartrazine-induced urticaria in these two cases. However, medications were withheld from one of these patients during challenge, and no information is presented on the medication status of the other patient. Perhaps, these positive reactions are simply break-through urticaria. Furthermore, one of these patients required rather large provoking doses of tartrazine, either 50 or 150 mg. Murdoch et al. (19) subsequently demonstrated that even normal subjects release histamine into plasma and urine when subjected to such large and unrealistic challenge doses with tartrazine.
Stevenson et al. (1986) concluded that tartrazine can provoke urticaria in an occasional patient. Critics might argue that the continuation of antihistamine therapy during the challenge period would suppress the tartrazine-induced response. Stevenson et al. (1986) posed the relevant question to such critics: "If the reaction to massive exposure to tartrazine is so feeble that regular antihistamines suppress it, of what significance is the regular ingestion of smaller amounts of dietary tartrazine in the pathogenesis of the patient's urticaria?". To the knowledge of FARRP faculty, this single case is the only well documented case of tartrazine-induced urticaria in the entire medical literature.
However, the dispute over the role of tartrazine in chronic urticaria remains. The data presented by Stevenson et al. (1986) seems quite compelling and seems to refute any concerns regarding the role of tartrazine in chronic urticaria.
Tartrazine-Induced Asthma
Studies in the late 1950's also implicated tartrazine as a cause for asthma in some children. The early studies were largely observational and anecdotal. Early studies attempted to link aspirin and tartrazine as cause of asthma but the linkage was never fully documented. While numerous studies have been conducted on tartrazine-induced asthma, methodological flaws existed in many of these studies. Asthmatic patients, especially those with severe asthma, often take a wide variety of medications including theophylline, inhaled and systemic corticosteroids, beta agonists, bronchodilators, cromolyn, and antihistamines. In several studies, medications were withheld leading to concerns that the withdrawal of medications resulted in the asthmatic reactions. This is particularly likely since these studies used single-blind or open challenge designs with the placebo administered before tartrazine.
When some of the medications were continued during the challenge, the frequency of positive responses to tartrazine was almost zero in the few studies that have been conducted using this design. Most of these studies were conducted with double-blind designs. However, aspirin-sensitive asthmatics were used in several of these studies and are known to often have rather unstable lung function. With unstable asthmatics, it can be especially important to intersperse placeboes throughout the challenge period since some unstable asthmatics will develop bronchospasm simply from the exertion of the spirometry tests. This approach was not taken in the studies where a few tartrazine-sensitive asthmatics were apparently identified. In several studies where bronchodilator drugs were continued during challenge and double-blind, placebo-controlled challenges were conducted, only one aspirin-sensitive asthmatic was identified as tartrazine-reactive and the investigators discounted the significance of this single reactor on the basis of highly unstable airways.
Again, a study by Stevenson et al. (1992) is particularly enlightening. Their study involved 43 aspirin-tolerant and 194 aspirin-sensitive asthmatics. Beta agonists, cromolyn, and antihistamines were withheld, but theophylline, bronchodilators, and inhaled and systemic corticosteroids were continued. The studies were conducted with single-blind challenges but all positive responses were confirmed by double-blind challenges. None of their patients reacted to tartrazine at doses up to 50 mg.
The conclusion from these studies is obvious. Tartrazine-induced asthma does not exist. Early studies of tartrazine in asthmatics were flawed and positive reactions were likely the result of unstable airways in chronic, severe asthmatics.
Well-Designed Studies Documenting that Tartrazine Is Highly Unlikely to Cause Asthma or Chronic Uriticaria
- Stevenson DD, Simon RA, Lumry WR, Mathison DA. Adverse reactions to tartrazine. J Allergy Clin Immunol 78:182, 1986.
- Stevenson DD, Simon RA, Lumry WR, Mathison DA. Pulmonary reactions to tartrazine. Pediatr Allergy Immunol 3:222, 1992.
- Vedanathan P, Menson MM, Bell TD, Bergin D. Aspirin and tartrazine oral challenge: incidence of adverse response in chronic childhood asthma. J Allergy Clin Immunol 60:8, 1977.
- Spector SL, Wangaard CH, Farr RS. Aspirin and concomitant idiosyncrasies in adult asthmatic patients. J Allergy Clin Immunol 64:500, 1979.
- Weber RW, Hoffman M, Raine DA, Nelson HS. Incidence of bronchoconstriction due to aspirin, azo dyes, non-azo dyes and preservatives in a population of perennial asthmatics. J Allergy Clin Immunol 64:32, 1979.
- Tarlo SM, Broder I. Tartrazine and benzoate challenge and dietary avoidance in chronic asthma. Clin Allergy 12:303, 1982.
- Morales MC, Basomba A, Pelaez A, Villalmanzo IG, Campos A. Challenge tests with tartrazine in patients with asthma associated with intolerance to analgesics (ASA-Triad). A comparative study with placebo. Clin Allergy 15:55, 1985.
Erythrosine (FD&C #3)
Erythrosine is an artificial food color (red). Erythrosine has been implicated as a causative factor in idiosyncratic reactions on only a few occasions. All of the clinical studies that have occurred on erythrosine have been seriously flawed due to one of more of several factors: use of a mixture of colorants or additives, enrollment of patients with severe chronic conditions such as chronic urticaria or asthma, withdrawal of critical medications from these patients prior to challenges, administration of placeboes first in all challenges, and failure to intersperse placeboes throughout the challenge. Thus, the cause-and-effect relationship has not been documented for erythrosine in idiosyncratic reactions.
Sunset Yellow (FD&C #6)
Sunset yellow is an artificial color allowed for use in foods and medications. Sunset yellow has been implicated as a causative factor in idiosyncratic reactions on only a few occasions. All of the clinical studies that have occurred on sunset yellow have been seriously flawed due to one of more of several factors: use of a mixture of colorants or additives, enrollment of patients with severe chronic conditions such as chronic urticaria or asthma, withdrawal of critical medications from these patients prior to challenges, administration of placeboes first in all challenges, and failure to intersperse placeboes throughout the challenge. Thus, the cause-and-effect relationship has not been documented for sunset yellow in idiosyncratic reactions.