Medication Dyes as a Source of Drug Allergy

All patients had complete blood count with differential blood testing and comprehensive metabolic panels, which showed no consistent or relevant abnormalities. Only 2 patients demonstrated any evidence of peripheral eosinophilia during their disease course. Seven patients underwent biopsy by hematoxylin and eosin (H&E) staining, and 3 for direct immunofluorescence. H&E staining demonstrated nonspecific findings with perivascular infiltrates with eosinophils. Six patients reported a history of sulfonamide allergy, and 3 reported aspirin allergy. Six of the patients reported multiple drug allergies, including 2 patients who reported 8 and 10 drug allergies, respectively. All patients experienced suboptimal to no responses to topical corticosteroids or generic oral antipruritics, including antihistamines (both sedating and nonsedating) and antileukotrienes. When dye allergy was suspected, identification of dye-free alternatives was accomplished using the National Institutes of Health–based DailyMed database (http://dailymed.nlm.nih.gov/dailymed/about.cfm). Three patients appeared to be sensitive to more than one dye based on reactions observed upon inadvertent rechallenge.

We have identified a small cohort of patients whose history is strongly suggestive of medication dyes as a trigger for itching and dermatitis. This has been an area of interest for a number of years,¹ although little has been written recently in the dermatology literature.² Most of what has been reported is in the form of single case reports or small case series.^3-5 FD&C Yellow No. 5 (tartrazine) has been the most frequently implicated in adverse reactions, including asthma, urticaria, worsening of atopic eczema, eosinophilia, and angioedema. Tartrazine reactions have been purported to be most relevant in aspirin-intolerant individuals.^4,6 The proposed mechanism for these reactions is dose-related histamine release from mast cells.

There are multiple single case reports focusing on what appeared to be reactions to dyes other than tartrazine, including FD&C Yellow No. 6, Yellow No. 10, and Red No. 40,⁷ which was extensively reviewed more than 10 years ago.⁸ Intravenous or local injections of other dyes used for diagnostic purposes are well-known to sporadically induce anaphylaxis in patients.^9,10 Similarly, use of FD&C Blue No. 2 has been reported to cause reactions as severe as anaphylaxis during cystoscopy in patients with sulfa allergies.^9,10 The reaction of the index case (case 1) to the very distinctive blue/black tablets prompted us to consider whether this might be a much more common phenomenon than initially appreciated.

While there are few reports implicating blue dyes, we had at least 3 patients whose pruritus and dermatitis appeared to be aggravated by blue dyes. Blue No. 1 and Blue No. 2 are 2 of 7 certified color additives that can be used in food, drug, and cosmetic products in both the United States and the European Union. FD&C Blue No. 1 is used in beverages, dairy products, powders, jellies, confections, condiments, icing, syrups, and extracts, while FD&C Blue No. 2 is used in baked goods, cereals, snack foods, ice cream, confections, and cherries. In the pharmaceutical industry, FD&C Blue No. 1 is used as a bluing agent in white tablets to confer their brightness, while FD&C Blue No. 2 is used as a dye for drug capsules, and in urologic, gastrointestinal, gynecologic, and obstetrical procedures, including amniotic fluid leaks.

Literature linking FD&C Blue No. 1 and Blue No. 2 to allergic reactions is essentially anecdotal, with a very limited number of reports. In a study of 43 children with a history of angioedema/urticaria who responded to an additive-free diet, 19 were challenged with FD&C Blue No. 2 (indigo carmine), and 3 “reacted.” Interestingly, these 3 patients did not have positive skin tests to common allergens, including dogs, cats, and eggs.¹¹

One study reported that adverse reactions to food additives, including dyes, were very rare in the general population (0.01%-0.23%), but higher in atopic individuals (2%-7%).¹² Given the difficulty in undertaking these studies, we must accept that these numbers are at best crude estimates. However, given the ubiquitous nature of exposure in medications, supplements, and foods, even a small percentage may mean that substantial numbers of patients could be affected. Using the lower bound of these estimates (0.01%) would suggest that up to 30,000 people could be affected in the United States alone. One has to wonder how many patients who report multiple drug allergies may simply be allergic to a color additive or other drug excipients.

Unfortunately, the real incidence of reactions to any additives is essentially unknown because of the lack robust and readily deployable tools required to accomplish vigorously controlled studies. First, there is an inherent problem in challenging patients with chronic urticaria or other potentially serious adverse reactions. ¹³ Even though our cohort of patients did not report airway issues, there are still substantial logistical problems associated with these evaluations. One cannot assume that challenge will not be accompanied by anaphylaxis or other severe reactions. Most of the studies looking at reactions to medication excipients and food additives examined either airway reactivity or objective changes to skin observed within 24 hours of challenge. Given the major complaint in our cohort was pruritus, often experienced without objective findings, this represents a particular challenge to consistently measure. While double-blind challenges are held as the