I have patch tested (true test) an 11 year old girl who develops recurrent erythema, itching and swelling of her face, without exanthema elsewhere. She uses a lot of scented creams, etc, and is sensitive to fragrances, on testing. However, 5 days after the test patches were removed shecalso developed a 2+ reaction to the site where imidazolidinyl urea was tested. What is the significance/reliability of this very delayed reaction? Thanks as always for your help.


Thank you for your inquiry.

There is a significant body of literature which substantiates that in some cases clinically significant patch test reactions do not appear until Day 5. Of course, as with any positive patch test, the clinical significance of a positive test can only be judged by its correlation with the history and the substances which containe the ingredient in question. However, it is possible, based upon the results of delayed patch test readings (three abstracts in this regard are copied for you below) that a reaction appearing in five days could be significant.

Thank you again for your inquiry and we hope this response is helpful to you.

J Am Acad Dermatol. 2008 Aug;59(2):225-33. Epub 2008 May 29.
Delayed patch test reading after 5 days: the Mayo Clinic experience.
Davis MD, Bhate K, Rohlinger AL, Farmer SA, Richardson DM, Weaver AL.
Department of Dermatology, Mayo Clinic, Rochester, Minnesota 55905, USA.
Background: Some patients may have delayed allergic patch test reactions that are not identified with a typical 5-day reading protocol.
Objective: To identify allergens with delayed-positive reactions and to determine whether a late reading (day 7-9) can be substituted for the day-5 reading.
Methods: We retrospectively reviewed medical records of patients who underwent patch testing for suspected allergic contact dermatitis from October 1997 through December 2006 and returned for delayed readings between days 7 and 10 or beyond. (This cohort was predominantly patients with suspected allergies to metals and corticosteroids).
Results: A total of 36,064 individual reactions (471 allergens, 372 patients) were interpreted on day 5 and at least once during days 7 through 21. We analyzed the 170 allergens that were each tested in at least 50 patients. The 4 allergens with the highest percentage of delayed-positive reactions were gold sodium thiosulfate 0.5% (delayed-positive reactions in 22/353 patients), dodecyl gallate 0.25% (6/105), palladium chloride 2% (8/194), and neomycin sulfate 20% (10/253). We observed a low number of delayed-positive readings for p-phenylenediamine 1% (1/251) and for corticosteroids. Reactions to certain preservative and fragrance allergens dissipated after the day-5 reading. Most reactions that dissipated after day 5 were mild, whereas the reactions that became apparent after day 5 were strong.
Limitations: This was a retrospective study. Most patients had suspected allergies to metals and corticosteroids.
Conclusion: Late patch test readings (day 7 or beyond) were useful when interpreting reactions to metals and topical antibiotics, but it was not useful in the diagnosis of reactions to other allergens, including topical corticosteroids. Because reactions to certain fragrance and preservative allergens may dissipate after 5 days, patch test reactions therefore are optimally read at days 3 and 5, but an additional reading on day 7 or beyond is useful if patch tests to metals and topical antibiotics are performed.

Contact Dermatitis. 1999 Mar;40(3):119-26.
Patch test reactions at D4, D5 and D6.
Geier J, Gefeller O, Wiechmann K, Fuchs T.
Department of Dermatology, University of Göttingen, Germany.
In this retrospective study, patch test reactions in 3 groups of patients were analysed, in order to obtain information on the best day for the 2nd patch-test reading after day 2 (D2), on the usefulness of additional readings after D3, and on the dependence of patch-test reactions at D4, D5, or D6 on allergen and/or patient characteristics. In the years 1990 to 1995, patch tests were routinely read at D3 and D4 in 1096 patients, at D3 and D5 in 1243 patients, and at D3 and D6 in 1136 patients. In all of the 3 groups, significantly more positive reactions diminished than appeared de novo from D3 to the later reading. Virtually identical results were observed in subgroups of patients formed by sex, age or atopy. However, men might possibly react more slowly than women on patch testing, showing more increasing than diminishing reactions in the D3/D4- and the D3/D5-comparison. Reactions to individual allergens showed wide differences in this connection. Neomycin sulfate, cobalt salts, and p-phenylenediamine can be characterized as slow allergens, with more reactions increasing than diminishing from D3 to the later readings. With fragrance mix and balsam of Peru, the opposite pattern occurred. In all subgroups of patients, and with most allergens, the gain in positive reactions was biggest when an additional reading was performed at D5. We conclude that for a single 2nd patch test reading after D2, D3 is the best day, and especially better than D4. If a 3rd reading is performed, it should be done at D5 to get the maximum information out of patch testing. However, this extends the test procedure to at least 1 day of the weekend.

Contact Dermatitis. 2000 Jun;42(6):330-5.
The outcome of an additional patch-test reading on days 6 or 7.
Jonker MJ, Bruynzeel DP.
Department of Occupational Dermatology, Free University Hospital, Amsterdam, The Netherlands.
In a retrospective study (period 1997-1998), the usefulness of an additional patch-test reading on day (D) 6 or 7 was investigated. In 62 out of 760 patients (8.2%), 77 late-positive reactions, those manifest after D3, were seen. 4 late reactions of allergens of which a related allergen reacted earlier (at D2 or D3) in the same patient were not incorporated in the study as late-positive. Allergens most involved in producing late-positive reactions were nickel sulfate (20 reactions), neomycin sulfate (7), tixocortol-21-pivalate (5), p.t. butylphenol formaldehyde resin (5) and Cl+ Me isothiazolinone (5). Special attention is paid to ?+ reactions at D6 or D7, because these reactions could add extra information to the outcome of patch-test readings, considering that this might be a group of allergens that has been positive between D3 and D6/7 or become positive later on. When this group is included, the total number of late-positive reactions is 105 in 88 out of 760 patients (11.6%). By doing D3 readings in combination with D6 or D7 readings only, one would miss 24 positive reactions. However, we concluded that it was preferable to have a reading on D2 and D3. An extra reading on D6 or D7 is very useful as it gives additional information in 8.2% of patch-tested patients.

Phil Lieberman, M.D.

Close-up of pine tree branches in Winter Close-up of pine tree branches in Winter