The Use of Photos from Patch Test Reactions on Day 7 in Professional Ice Hockey Players in Sweden

To trace contact allergy, the International Contact Dermatitis Research Group (ICDRG) recommended patch test reading on day 3 or 4 (D3/4) ¹. However, a substantial number of contact allergies is missed unless an additional reading is performed on D7 ^{2, 3}. A second patch test reading on D7 is mandatory at the Malmö department since 1992 ⁴. A test reading on D7 is also mandatory according to the recent recommendations by the European Society of Contact Dermatitis ⁵.

In a study on occupational dermatoses in professional ice hockey players in Sweden, patch testing was performed with a baseline series and a series representing the work environment ⁶. A second patch test reading on D7 was virtually impossible. The aim of this study was therefore to investigate if photos from mobile phones taken on D7 by the subjects themselves or someone helping them could add positive reactions to those noted from traditional test readings on D3 when the images of tested backs were sent to the investigators for evaluation.

The study was approved by the Central Ethical Review Board in Lund, Sweden. The subjects were informed in writing and they gave their written informed consent.

Four subjects out of 107 patch tested were excluded due to too short application time of the tests. Thus, 103 should have sent in photos but photos were obtained from only 100. In 5 players the quality of the photos were too bad to allow evaluation. Thus, photos of 95/103 (92.2%) players were evaluated.

Additional contact allergies were noted to 7 different test preparations in 5 players. 2 of these 5 players did not have any contact allergy diagnosed on D3. One allergic reaction was noted to each one of oxidized linalool, mercapto mix, mercaptobenzothiazole, phenol-formaldehyde resin 2 (PFR-2; a resol resin based on phenol and formaldehyde (9)), methylenedianiline (MDA), imidazolidinyl urea, and paraben mix (Table 1). Contact allergy only detected on the image evaluation was noted to 4 different sensitizer in 2 players, one contact allergy in one player and 3 allergies in another player (Table 1).

The validation evaluation showed agreement between the dermatologist reading and the image evaluation. Doubtful reactions noted on the dermatologist reading on D6 were not registered as positive when viewing the photos without knowledge of the D3 readings. The allergic reactions on the dermatologist reading on D6 were registered as positive at the photo evaluation.

The patch test reading after one week based on mobile photos resulted in the detection of 7 additional contact allergic reactions. The contact allergies concerned 7 different test preparations and most likely 6 sensitizers as mercaptobenzothiazole giving a contact allergic reaction also is present in mercapto mix. The first test reading on D3 by dermatologists resulted in 50 allergic reactions ⁶. The additional reading thus gave 14% more contact allergic reactions diagnosed. Of all contact allergic reactions detected in the players (57 reactions), the allergies detected after one week constitute 12.3% which is a figure similar to the one for routinely patch tested dermatitis patients at our department (unpublished results).

In a way, it is again demonstrated that a reading after one week should be mandatory ⁴. A reading after one week is also recommended in the guidelines of the European Contact Dermatitis Society ⁵. However, do we know that these reactions absent on D3 but appearing on the photos from D7 really represent contact allergic reactions? Actually, we do not know as the ordinary D6 reading of the players in one team did not result in any additional test reactions registered why the performed validation evaluation in this study has a limited value. On the other hand, it is highly likely that the new reactions detected on the images represent patch test reactions. The decisive question is whether they are irritant, doubtful, or allergic. The photos are from the D6/7 reading and not from D2 or D3/4 readings when both doubtful and irritant reactions are more common, particularly if substances/products outside the common test series are tested. All sensitizers detected on the D7 reading in this study are included in our baseline series used for dermatitis patients ⁶. According to our experience, we virtually never see irritant reactions to these preparations appearing first on D7 without any reactions at all on the D3 reading. Thus, irritant reactions can be excluded. It is on the other hand difficult to exclude doubtful reactions for the D7 reactions. For sensitizers with a tendency to appear late, even after D7 without being a sign of active sensitization, such as gold, acrylates, and corticosteroids ^{10, 11, 12}, may not uncommonly present as a negative reaction on D3/4 followed by a doubtful reaction on D7 (10). Except for the phenol-formaldehyde resin (PFR-2) ^{9, 13}, the additional contact sensitizers detected in this study do not belong to those sensitizers having a tendency to appear late but any sensitizer may occasionally appear late, i.e. a doubtful or allergic reaction on D7 following a negative reaction on D3/4. Importantly many of the doubtful reactions appearing first on D7 and with a negative reaction on D3/4 will develop further into an obvious allergic reaction why individuals with such reactions should be read further ¹⁰.

Already today mobile photos are used in the dermatology clinics. Patients may show photos of dermatoses which have changed character or disappeared at the time of consultation. Sometimes we ask the patient to take a photo of a test area and send us in case of any reaction.

A more standardized form of tele-dermatology has been used since many years and is increasingly used in various areas of dermatology ^{14, 15, 16}. It has also been used for evaluation of test reactions. A study from Linköping University Sweden published in 2007 compared the inter-expert evaluation from five experienced dermatologists of photographic images to test readings according to ICDRG ¹⁷. Disagreement on ICDRG criteria was found but there was agreement on simplified tripartite reading for positive, negative (including doubtful) and irritant reactions. Unlike our study, the Linköping study investigated test reactions on D3 when both irritant and doubtful reactions are substantially more common than on D7.

Over the years, mobile cameras and frequent digital imaging have become an everyday technique in the population. Since this study was done, mobile phone cameras and computer screens have improved, as they do continuously, which suggests that evaluation of images can be even better performed nowadays and in the future. In addition to this software to improve image quality can be used.

Are there any limitations of using mobile phones and digital imaging for assessment of patch test reactions? In an ideal situation, we would have performed the D7 reading live but this was not possible. Instead of risk missing 10-15% contact allergy, obviously the technique has a place in situations like ours when most teams of ice hockey players were on tours with frequent games when it was time for the second test reading. We think that the technique should first focus on the reading after one week as there are substantially fewer irritant and doubtful reactions confusing the reading at this time point. Hopefully, the technique will in the future allow the mobile camera to detect whether an erythematous test area also is infiltrated/edematous as this according to the ICDRG classification ¹ discriminates an allergic test reaction from a doubtful one. A larger validation study than ours is also needed when comparing all reactions noted on D7 when a dermatologist is reading live in the ordinary way and when another dermatologist is performing the reading based on digital images from D7.

As mentioned, mobile photos are already used today for patch test reading. We think that the technique can be used for the D7 or later readings, particularly when limited test areas are involved and the photos are taken according to written advice. It is unlikely that a negative reaction on the photo does not represent a negative reaction. The problem arises when there is a reaction and it might be difficult to conclude whether it is a doubtful or allergic reaction. Occasionally, it might not matter as the tested individual is not exposed to the tested substance/product anymore or it is easy to avoid the substance/product. In this situation it might not be necessary for the individual to come to the clinic for evaluation of the test area. On the other hand, if there is a reaction on the photo where the decision on whether it is an allergic reaction or not is very important, we think that the tested person must come to the clinic for evaluation. Such examples are a possible formaldehyde reaction in any individual as chemical analyses may be required and a possible epoxy reaction in a worker exposed to epoxy professionally, as it may affect compensation and rehabilitation matters.

We found 14% (7/50) additional contact allergies on D7 in mobile photos of patch tested backs of the ice hockey players. Our results showed high compliance to send digital photos and benefit from a simplified two partite method as a late test reading, when an ICDRG reading could not be done. Improvements in mobile photo hard ware and image analyzing techniques in a future test reading situation can be of importance. Further investigations are needed.

1. 1.Fregert S. (1981) Manual of Contact Dermatitis. , 2ndEdn., Munksgaard, Copenhagen
   Search at Google Scholar

1. 2.Jonker M J, Bruynzeel D P. (2000) The outcome of an additional patch-test reading on days 6 or 7. Contact Dermatitis. 42, 330-335.
   PubMed·Search at Google Scholar

1. 3.Ahlgren C, Isaksson M, Möller H, Axéll T, Liedholm R et al. (2014) The necessity of a test reading after 1 week to detect late positive patch test reactions in patients with oral lichen lesions. Clin Oral Investig. 18, 1525-1531.
   PubMed·View article·Search at Google Scholar

1. 4.Bruze M. (1990) Den yrkesdermatologiska verksamheten I Malmö efter oktober. In Jubileumsskrift. Yrkes- och miljödermatologisk verksamhet I södra Sverige firar50 år,1960-2010(In Swedish) 16-9.
   Search at Google Scholar

1. 5.Johansen J D, Aalto-Korte K, Agner T, Andersen K E, Bircher A et al. (2015) European Society of Contact Dermatitis guideline for diagnostic patch testing - recommendations on best practice. Contact Dermatitis. 73, 195-221.
   PubMed·View article·Search at Google Scholar

1. 6.TBJ Eriksson, Bruze M, Engfeldt M, Tegner Y, Runeson-Broberg R et al.Contact allergy in Swedish professional ice hockey players. In manuscript
   Search at Google Scholar

1. 7.Mowitz M, Zimerson E, Svedman C, Bruze M. (2012) Stability of fragrance patch test preparations applied in test chambers. , Br J Dermatol 167, 822-827.
   PubMed·View article·Search at Google Scholar

1. 8.Mowitz M, Svedman C, Zimerson E, Bruze M. (2014) Fragrance patch tests prepared in advance may give false-negative reactions. Contact Dermatitis. 71, 289-294.
   PubMed·View article·Search at Google Scholar

1. 9.Isaksson M, Inerot A, Lidén C, Matura M, Stenberg B et al. (2011) Multicentre patch testing with a resol resin based on phenol and formaldehyde. Contact Dermatitis. 65, 34-37.
   PubMed·View article·Search at Google Scholar

1. 10.Bruze M, Hedman H, Björkner B, Möller H. (1995) The development and course of test reactions to gold sodium thiosulfate. Contact Dermatitis. 33, 386-391.
   PubMed·Search at Google Scholar

1. 11.Isaksson M, Lindberg M, Sundberg K, Hallander A, Bruze M. (2005) The development and course of patch-test reactions to 2-hydroxyethyl methacrylate and ethyleneglycol dimethacrylate. Contact Dermatitis. 53, 292-297.
   PubMed·View article·Search at Google Scholar

1. 12.Isaksson M, Bruze M, Björkner B, Hindsén M, Svensson L. (1999) The benefit of patch testing with a corticosteroid at a low concentration. Am J Contact Dermat. 10, 31-33.
   Search at Google Scholar

1. 13.Isaksson M, Ale I, Andersen K, Diepgen T, Elsner P et al. International Contact Dermatitis Research Group. Multicenter Patch (2015) . Testing With a Resol Resin Based on Phenol and Formaldehyde Within the International Contact Dermatitis Research Group. Dermatitis 26, 230-234.
   PubMed·View article·Search at Google Scholar

1. 14.McKoy K, Antoniotti N M, Armstrong A, Bashshur R, Bernard J et al.Practice Guidelines for Teledermatology. Telemed J E Health.2016Sep30.[Epub ahead of print] PubMed PMID: 27690203.
   PubMed·Search at Google Scholar

1. 15.Fogel A L, Teng J, Sarin K Y. (2016) Direct-to-consumer teledermatology services for pediatric patients: Room for improvement. J Am Acad Dermatol. 75, 887-888.
   PubMed·View article·Search at Google Scholar

1. 16.Tensen E, JP van der Heijden, Jaspers M W, Witkamp L. (2016) Two Decades of Teledermatology: Current Status and Integration in National Healthcare Systems. Curr Dermatol Rep. 5, 96-104.
   PubMed·View article·Search at Google Scholar

1. 17.Ivens U, Serup J, O’goshi K. (2007) Allergy patch test reading from photographic images:disagreement on ICDRG grading but agreement on simplified tripartite reading. Skin Res Technol. 13, 110-113.
   PubMed·View article·Search at Google Scholar

1. 1.Vincent Rosie, Ormerod Emma, Gollins Charlotte E, Rai Nabina, Conduit-Turner Karen, et al, 2023, Semi-virtual patch testing: the future?, British Journal of Dermatology, 188(1), 137, 10.1093/bjd/ljac013