Skip to main content
JAMA Network logoLink to JAMA Network
. 2019 Nov 27;156(1):85–91. doi: 10.1001/jamadermatol.2019.3844

Evaluation of Patch Test Findings in Patients With Anogenital Dermatitis

Erin M Warshaw 1,2, Rebecca S Kimyon 1,3,, Jonathan I Silverberg 4, Donald V Belsito 5, Joel G DeKoven 6, Howard I Maibach 7, Kathryn A Zug 8, Amber R Atwater 9, Toby Mathias 10, Denis Sasseville 11, Joseph F Fowler Jr 12, James G Marks Jr 13, Margo J Reeder 14, Vincent A DeLeo 15, Melanie D Pratt 16, Matthew J Zirwas 17, James S Taylor 18, Anthony F Fransway 19
PMCID: PMC6902099  PMID: 31774454

Key Points

Question

For patients with anogenital dermatitis referred for patch testing, what are common relevant allergens?

Findings

In this analysis of cross-sectional data from the North American Contact Dermatitis Group from 2005 to 2016, most individuals with anogenital involvement only referred for patch testing had relevant allergic reactions. Compared with individuals without anogenital involvement, those with anogenital dermatitis had significantly more reactions to balsam of Peru, methylchloroisothiazolinone and methylisothiazolinone, dibucaine, benzocaine, triamcinolone acetonide, budesonide, ethylenediamine dihydrochloride, lidocaine, and desoximetasone.

Meaning

It is important that individuals with anogenital skin disease undergo patch testing, especially for reactions to preservatives, fragrances, topical anesthetics, and topical corticosteroids.

Abstract

Importance

Contact dermatitis in the anogenital area is associated with sleep disturbance and dyspareunia and can profoundly affect quality of life. The literature on anogenital contact dermatitis and culprit allergens is limited. The last large-scale study on common, relevant allergens in patients with anogenital dermatitis was published in 2008.

Objectives

To characterize patients with anogenital dermatitis referred for patch testing by the North American Contact Dermatitis Group, to identify common allergens, and to explore sex-associated differences between anogenital dermatitis and allergens.

Design, Setting, and Participants

A retrospective, cross-sectional analysis was conducted of the North American Contact Dermatitis Group database among 28 481 patients who underwent patch testing from January 1, 2005, to December 31, 2016, at outpatient referral clinics in the United States and Canada.

Exposure

Patch testing for allergens.

Main Outcomes and Measures

Currently relevant allergic patch test reactions in patients with anogenital dermatitis.

Results

Of 28 481 patients tested during the study period, 832 patients (336 men and 496 women; mean [SD] age, 50.1 [26.5] years) had anogenital involvement and 449 patients (177 men and 272 women; mean [SD] age, 49.6 [17.4] years) had anogenital dermatitis only. Compared with those without anogenital involvement, there were significantly more male patients in the group with anogenital dermatitis (177 [39.4%] vs 8857 of 27 649 [32.0%]; relative risk, 1.37; 95% CI, 1.14-1.66; P < .001). In the group with anogenital involvement, female patients were significantly less likely than male patients to have allergic contact dermatitis as a final diagnosis (130 [47.8%] vs 107 [60.5%]; relative risk, 0.78; 95% CI, 0.64-0.94; P = .01), whereas a final diagnosis of other dermatoses (eg, lichen planus, lichen sclerosus, or lichen simplex chronicus) was more frequent for female patients than for male patients (67 [24.6%] vs 28 [15.8%]; relative risk, 1.54; 95% CI, 1.02-2.31; P = .03). Of the 449 patients in the group with anogenital involvement only, 227 (50.6%) had 1 or more relevant reaction with patch testing. Allergens that were statistically significantly more common in patients with anogenital involvement compared with those without anogenital involvement included medicaments such as dibucaine (10 of 250 patients tested [4.0%] vs 32 of 17 494 patients tested [0.2%]; relative risk, 22.74; 95% CI, 11.05-46.78; P < .001) and preservatives such as methylchloroisothiazolinone and methylisothiazolinone (30 of 449 patients tested [6.7%] vs 1143 of 27 599 patients tested [4.1%]; relative risk, 1.61; 95% CI, 1.14-2.41; P = .008). A total of 152 patients met the definition for anogenital allergic contact dermatitis, which is defined as anogenital involvement only, allergic contact dermatitis as the only diagnosis, and 1 or more positive reaction of current clinical relevance.

Conclusions and Relevance

For patients with anogenital involvement only who were referred for patch testing, male patients were more likely to have allergic contact dermatitis, whereas female patients were more likely to have other dermatoses. Common allergens or sources consisted of those likely to contact the anogenital area. For individuals with anogenital involvement suspected of having allergic contact dermatitis, reactions to preservatives, fragrances, medications (particularly topical anesthetics), and topical corticosteroids should be tested.


This cross-sectional analysis characterizes patients with anogenital dermatitis referred for patch testing by the North American Contact Dermatitis Group, examines common allergens, and explores sex-associated differences between anogenital dermatitis and allergens.

Introduction

Contact dermatitis in the anogenital area is associated with sleep disturbance and dyspareunia and can profoundly affect quality of life. It is likely that irritant contact dermatitis and/or other anogenital skin conditions precede the development of allergic contact dermatitis (ACD).1 A total of 13% to 63% of patients with anal and/or genital dermatitis who undergo patch testing have a reaction to 1 or more allergens.2,3,4,5,6,7,8,9 A prior analysis of North American Contact Dermatitis Group (NACDG) patients with anogenital dermatitis spanned from 1994 to 2004.9 Since that study, several studies have identified allergens in patients with anal and/or genital dermatitis (Table 1),2,3,4,5,6,7,8,9 but those studies are limited by small sample size or lack of information on the clinical relevance of the allergens. Because contact allergen trends change with consumer and industrial practices, we sought to update the previous NACDG study with recent data.

Table 1. Studies Published Since 2008 of Patch Test Results for Patients With Anal and/or Genital Involvement.

Source Patch-Tested Patients, No. (% Female) Study Population Age, Mean (range), y Patients Main Findings
With ≥1 Positive Allergic Reaction, No. (%) With ≥1 Currently Relevant Reaction, No. (% of Patients Tested)
Bauer et al,8 2011 (2004-2008; Europe) 1374 (41) Patients with anogenital, genital, or anal dermatosis NR (75% of patients >40 y) 632 (46) NR Bufexamac significantly more common with anogenital symptoms than without; nickel, BOP, and fragrance mix are most common positive allergens in anogenital involvement; patients with anal dermatoses more likely to develop sensitization than those with genital dermatoses
Warshaw et al,9 2008 (1994-2004; North America) 347 (61) Patients with anogenital dermatitis 47.2 (0-92) NR 220 (63) Cinnamal and topical medications significantly more common allergens
Abu-Asi et al,2 2016 (2005-2015; United Kingdom) 150 (47) Patients with perianal dermatoses and/or pruritus ani 45.2 (5-88) 60 (40) 32 (21) MI-MCI was the most common relevant allergen
Vermaat et al,6 2008 (2002-2006; the Netherlands) 53 (100) Patients with chronic anogenital, vulvar, or perianal symptoms 35.3 (16-64) 35 (66) 7 (13) Topical fragrances and ingested spices were the most common relevant allergens
Haverhoek et al,4 2008 (Australia) 43 (100) Patients with vulvar pruritus 42.0 (20-80) 35 (81) 19 (44) Patch testing to a wide variety of allergens is worthwhile in those with vulvar pruritus even without a diagnosis of ACD; nickel and Vagisil (benzocaine and resorcinol) were the most common relevant allergens
González-Pérez et al,3 2014 (2001-2012; Spain) 37 (41) Patients with perianal eczema 56.6 (37-82) 16 (43) 10 (27) MI-MCI was the most common relevant allergen
Bhate et al,5 2010 (1990-2006; United States) 37 (49) Patients with genital dermatitis 43.2 (24-77) 15 (41) 11 (30) Fragrances were the most common relevant allergens
Ljubojević et al,7 2009 (dates NR; Croatia) 33 (67) Patients with persistent or recurrent genital erythema, pruritus, and burning 38.0 (NR) 13 (39) 7 (21) Nickel, BOP, formaldehyde, and neomycin were the most common positive allergens

Abbreviations: ACD, allergic contact dermatitis; BOP, balsam of Peru; MI-MCI, methylisothiazolinone-methylchloroisothiazolinone; NR, not reported.

Methods

The methods used in this study were similar to those in the previous NACDG study.9 In brief, 28 481 patients underwent patch testing from January 1, 2005, to December 31, 2016, for the NACDG screening series (65-70 allergens); data were collected using standard NACDG methods.9 Several subgroups were identified a priori: (1) patients referred for patch testing with anogenital involvement only (AG), (2) patients referred for patch testing without anogenital involvement (noAG), (3) patients with anogenital ACD (AG-ACD: AG plus ACD as the only diagnosis and ≥1 currently relevant positive reaction), and (4) patients with AG plus dermatitis at 1 or more other body site (AG+). This cross-sectional analysis of deidentified data was approved by the Minneapolis Veterans Affairs Medical Center Subcommittee on Human Studies, waiving patient informed consent. Statistical analyses were performed using SAS, version 9.2 (SAS Institute Inc) with 2-sided χ2 tests; P < .05 was considered statistically significant.

Results

Patients With or Without Anogenital Involvement

Of 28 481 patients tested during the study period, 832 patients (336 men and 496 women; mean [SD] age, 50.1 [26.5] years) had anogenital involvement and 449 patients (177 men and 272 women; mean [SD] age, 49.6 [17.4] years) had anogenital dermatitis only. A comparison of the AG (n = 449) and noAG (n = 27 649) groups found that male patients were significantly more likely than female patients to present with anogenital involvement (177 male patients [39.4%] in the AG group vs 8857 male patients [32.0%] in the noAG group; relative risk, 1.37; 95% CI, 1.14-1.66; P < .001). Individuals in the AG group were significantly less likely than individuals in the noAG group to have an atopic predisposition or an occupationally related skin problem (eTable in the Supplement).

Patients With Anogenital Involvement Only

For the 449 patients in the AG group, the most common final diagnoses included ACD (237 [52.8%]), other dermatitis (114 [25.4%]), other dermatoses (95 [21.2%]), and irritant contact dermatitis (66 [14.7%]) (eTable in the Supplement). Stratifying by sex, we found that female patients were significantly more likely than male patients to have a final diagnosis of other dermatoses (67 of 272 female patients [24.6%] vs 28 of 177 male patients [15.8%]; relative risk, 1.54; 95% CI, 1.02-2.31; P = .03) and were significantly less likely than male patients to have a final diagnosis of ACD (130 of 272 female patients [47.8%] vs 107 of 177 male patients [60.5%]; relative risk, 0.78; 95% CI, 0.64-0.94; P = .01). There were no statistically significant associations of sex with a final diagnosis of other dermatitis or irritant contact dermatitis.

Of the 449 individuals in the AG group, 227 (50.6%) had 1 or more positive, currently relevant allergic reactions (Table 2). Allergens that were statistically significantly more common in the AG group compared with those in the noAG group included balsam of Peru, methylchloroisothiazolinone (MI)–methylisothiazolinone (MCI), dibucaine, benzocaine, triamcinolone acetonide, budesonide, ethylenediamine dihydrochloride, lidocaine, and desoximetasone. Dibucaine and triamcinolone acetonide had the highest relative effect size (dibucaine, 22.74 [95% CI, 11.05-46.78]; triamcinolone acetonide, 7.87 [95% CI, 1.86-33.31]), followed by benzocaine (7.50 [95% CI, 4.43-12.71]) and desoximetasone (6.55 [95% CI, 1.99-21.54]). Statistically significantly less frequent allergens included nickel, formaldehyde (1% concentration), paraphenylenediamine, bacitracin, cobalt, thiuram mix, and 1,3-dimethylol-5, 5-dimethyl hydantoin.

Table 2. Most Common Currently Clinically Relevant Allergens Among Patients in the AG Group vs Patients in the noAG Group.

Allergen AG noAG AG vs noAG
No. of Patients Tested Clinically Relevant Reaction, No. (%) Rank Order No. of Patients Tested Clinically Relevant Reaction, No. (%) Rank Order P Value RR (95% CI)
Significantly higher in AG group vs noAG group
Dibucaine 2.5% pet 250 10 (4.0) 9 17 494 32 (0.2) 62 <.001 22.74 (11.05-46.78)
Triamcinolone acetonide 1% pet 117 2 (1.7) 25 9292 20 (0.2) 62 .04 7.87 (1.86-33.31)
Benzocaine 5% pet 449 15 (3.3) 12 27 613 123 (0.5) 57 <.001 7.50 (4.43-12.71)
Desoximetasone 1% pet 410 3 (0.7) 48 23 258 26 (0.1) 64 .01 6.55 (1.99-21.54)
Ethylenediamine dihydrochloride 1% pet 448 6 (1.3) 28 27 623 112 (0.4) 59 .002 3.30 (1.46-7.47)
Lidocaine hydrochloride 15% pet 449 5 (1.1) 31 27 596 120 (0.4) 58 .03 2.56 (1.05-6.23)
Budesonide 0.01% or 0.1% pet 449 7 (1.6) 26 27 619 192 (0.7) 50 .03 2.24 (1.06-4.74)
Methylchloroisothiazolinone and methylisothiazolinone 0.01% aq (100 ppm) 449 30 (6.7) 4 27 599 1143 (4.1) 9 .008 1.61 (1.14-2.41)
Balsam of Peru 25% pet 449 45 (10.0) 2 27 637 2095 (7.6) 4 .04 1.38 (1.01-1.88)
Significantly lower in AG group vs noAG group
Thiuram mix 1% pet 449 4 (0.9) 39 27 620 727 (2.6) 19 .02 0.034 (0.13-0.90)
1,3-Dimethylol-5, 5-dimethyl hydantoin 1% pet 449 1 (0.2) 57 27 618 423 (0.5) 30 .02 0.15 (0.02-1.03)
Cobalt chloride 1% pet 449 5 (1.1) 31 27 608 964 (3.5) 12 .006 0.32 (0.13-0.76)
Bacitracin 20% pet 449 6 (1.3) 29 27 610 927 (3.4) 13 .02 0.40 (0.18-0.88)
Paraphenylenediamine 1% aq 449 8 (1.8) 22 27 576 1069 (3.9) 10 .02 0.46 (0.23-0.92)
Nickel sulfate 2.5% pet 448 26 (5.8) 5 27 538 2938 (10.7) 2 <.001 0.54 (0.37-0.79)
Formaldehyde 1% aq 449 15 (3.3) 12 27 626 1668 (6.0) 6 .02 0.55 (0.34-0.91)
No significant difference between AG group vs noAG group
Dithiomorpholine 1% pet 39 1 (2.6) 19 4309 41 (1.0) 40 .32 2.69 (0.38-19.10)
Lavender oil 2% pet 332 2 (0.6) 53 18 319 58 (0.3) 60 .29 1.90 (0.47-7.76)
Mercapto mix 1% pet 192 2 (1.0) 35 13 472 76 (0.6) 55 .30 1.85 (0.46-7.46)
Benzyl alcohol 1% pet 410 8 (2.0) 21 23 277 248 (1.1) 37 .09 1.83 (0.91-3.68)
Carvone 5% pet 332 3 (0.9) 37 18 318 104 (0.6) 54 .44 1.59 (0.51-4.99)
Neomycin sulfate 20% pet 449 17 (3.8) 10 27 594 664 (2.4) 20 .06 1.57 (0.98-2.52)
Clobetasol propionate 1% pet 449 2 (0.5) 55 27 609 79 (0.3) 61 .37 1.56 (0.38-6.31)
Majantole 5% pet 133 2 (1.5) 27 8224 85 (1.0) 39 .40 1.45 (0.37-5.77)
Iodopropynyl butylcarbamate 0.1% or 0.5% pet 449 18 (4.0) 7 27 609 820 (3.0) 17 .46 1.35 (0.85-2.13)
Paraben mix 15% pet 449 5 (1.1) 31 27 617 237 (0.9) 44 .56 1.30 (0.54-3.13)
Di-α-tocopherol 100% 449 4 (0.9) 39 27 608 202 (0.7) 48 .58 1.22 (0.45-3.26)
Disperse blue 106 or 104/126 mix 1% pet 449 4 (0.9) 39 27 597 205 (0.7) 47 .58 1.20 (0.45-3.21)
Lanolin alcohol 30% or 50% pet 449 16 (3.6) 11 27 629 831 (3.0) 16 .49 1.18 (0.73-1.93)
Methylisothiazolinone 0.2% aq (2000 ppm) 199 26 (13.1) 1 10 091 1113 (11.0) 1 .36 1.18 (0.82-1.70)
Bronopol 0.5% pet 449 8 (1.8) 22 27 613 425 (1.5) 29 .68 1.16 (0.58-2.32)
Mercaptobenzothiazole 1% pet 449 3 (0.7) 50 27 628 161 (0.6) 53 .75 1.15 (0.37-3.58)
Cinnamic aldehyde 1% pet 449 15 (3.3) 12 27 610 806 (2.9) 18 .60 1.14 (0.69-1.89)
Fragrance mix
I 8% pet 449 42 (9.4) 3 27 616 2610 (9.5) 3 .94 0.99 (0.74-1.32)
II 14% pet 410 18 (4.3) 6 23 275 1036 (4.5) 8 .95 0.99 (0.63-1.56)
Jasmine absolute 2% pet 293 2 (0.7) 49 17 234 119 (0.7) 51 >.99 0.99 (0.25-3.98)
Propylene glycol 100% aq 199 6 (3.0) 16 10 095 319 (3.2) 14 .91 0.95 (0.42-2.16)
Methyldibromo glutaronitrile and phenoxyethanol 2% pet 449 13 (2.9) 18 27 607 871 (3.2) 15 .75 0.92 (0.54-1.57)
Amidoamine 0.1% aq 449 5 (1.1) 31 27 613 330 (1.2) 35 .87 0.93 (0.39-2.24)
Shellac 20% alc 133 1 (0.8) 47 8224 67 (0.8) 46 >.99 0.92 (0.13-6.60)
Oleamidopropyl dimethylamine 0.1% aq 332 7 (2.1) 20 18 318 439 (2.4) 22 .73 0.88 (0.42-1.84)
Carba mix 3% pet 449 14 (3.1) 15 27 624 1015 (3.7) 11 .53 0.85 (0.51-1.43)
Tixocortol-21-pivalate 1% pet 449 6 (1.3) 29 27 613 474 (1.7) 27 .54 0.78 (0.35-1.74)
Propylene glycol 10% or 30% aq 449 8 (1.8) 22 27 619 662 (2.4) 21 .40 0.74 (0.37-1.48)
Quaternium 15 2% pet 449 18 (4.0) 7 27 622 1607 (5.8) 7 .10 0.69 (0.44-1.09)
Imidazolidinyl urea 2% aq 117 1 (0.9) 45 9303 118 (1.3) 34 .69 0.67 (0.09-4.78)
Cocamidopropyl betaine 1% aq 449 4 (0.9) 39 27 614 395 (1.4) 32 .42 0.62 (0.23-1.66)
Propolis 10% pet 410 4 (1.0) 36 23 261 390 (1.7) 28 .43 0.58 (0.22-1.55)
Compositae mix 6% pet 449 3 (0.7) 50 27 608 329 (1.2) 36 .50 0.56 (0.18-1.74)
Colophony 20% pet 449 3 (0.7) 50 27 629 358 (1.3) 33 .39 0.52 (0.17-1.60)
Dimethylaminopropylamine 1% aq 332 3 (0.9) 37 18 326 321 (1.8) 26 .39 0.52 (0.17-1.60)
Imidazolidinyl urea 2% pet 449 4 (0.9) 39 27 620 493 (1.8) 25 .20 0.50 (0.19-1.33)
Formaldehyde 2% aq 199 6 (3.0) 16 10 093 642 (6.4) 5 .05 0.47 (0.21-1.05)
Diazolidinyl urea 1% aq 117 1 (0.9) 45 9288 171 (1.8) 24 .73 0.46 (0.07-3.29)
Black rubber mix 0.6% pet 449 1 (0.2) 57 27 628 146 (0.5) 56 .73 0.42 (0.06-3.00)
Diazolidinyl urea 1% pet 449 4 (0.9) 39 27 619 582 (2.1) 23 .09 0.42 (0.16-1.13)
4-Tert-butylphenol formaldehyde resin 1% pet 449 1 (0.2) 57 27 625 178 (0.6) 52 .54 0.35 (0.05-2.46)
2-Hydroxyethyl methacrylate 2% pet 410 2 (0.5) 54 23 256 339 (1.5) 31 .14 0.33 (0.08-1.34)
Methyl methacrylate 2% pet 449 1 (0.2) 57 27 608 198 (0.7) 49 .38 0.31 (0.04-2.21)
Bisphenol A epoxy resin 1% pet 410 1 (0.2) 56 23 276 207 (0.9) 42 .28 0.27 (0.04-1.95)
Tea tree oil, oxidized, 5% pet 449 1 (0.2) 57 27 604 230 (0.8) 45 .28 0.27 (0.04-1.90)
Cocamide DEA 0.5% pet 449 1 (0.2) 57 27 614 242 (0.9) 43 .19 0.25 (0.04-1.81)
Glyceryl thioglycolate 1% pet 449 1 (0.2) 57 27 577 253 (0.9) 41 .20 0.24 (0.03-1.73)
Ylang-ylang oil 2% pet 449 1 (0.2) 57 27 608 291 (1.1) 38 .10 0.21 (0.03-1.50)

Abbreviations: AG, anogenital involvement only; alc, alcohol; aq, aqueous; DEA, diethanolamine; mix, mixture; noAG, no anogenital involvement; pet, petrolatum; RR, relative risk.

Patients With AG-ACD

Of 227 patients with 1 or more clinically relevant allergic reactions, 211 (93.0%) had a final diagnosis of ACD. Of these 211 patients, 59 (28.0%) had additional diagnoses, most frequently other dermatoses (n = 21), other dermatitis (n = 18), and irritant contact dermatitis (n = 18). A total of 152 patients met our strict definition of AG-ACD (anogenital involvement only, ACD as the only diagnosis, and ≥1 positive allergic reaction of current clinical relevance). The most common allergens in this group were preservatives, fragrances, medications, emulsifiers, corticosteroids, nickel sulfate, and surfactants. The most common sources were personal care products, medications, food products, and clothing; Table 3 summarizes the sources of these allergens.

Table 3. Sources of Currently Clinically Relevant NACDG Allergens Among Patients With Anogenital ACD.

Source No. (%) (N = 366)
Personal care products 208 (56.8)
Unspecified 131 (35.8)
Wipes 29 (7.9)
Soaps and cleansers 22 (6.0)
Moisturizers 8 (2.2)
Miscellaneous health devicesa 6 (1.6)
Hair dyes 4 (1.1)
Shampoos 3 (0.8)
Artificial nails 2 (0.5)
Perfumes and fragrances 2 (0.5)
Deodorants 1 (0.3)
Medications 69 (18.9)
Corticosteroids 27 (7.4)
Analgesics and antipruritics 22 (6.0)
Antibiotics 9 (2.5)
Unspecified 6 (1.6)
Antifungals 5 (1.4)
Food products 31 (8.5)
Clothing and jewelry 29 (7.9)
Undergarments and swimwear 11 (3.0)
Jewelry 6 (1.6)
Gloves 5 (1.4)
Unspecified 4 (1.1)
Textiles 2 (0.5)
Socks and stockings 1 (0.3)
Unknown and otherb 29 (7.9)

Abbreviations: ACD, allergic contact dermatitis; NACDG, North American Contact Dermatitis Group.

a

Includes condoms and diaphragms.

b

Nonskin soaps and disinfectants, paper products, miscellaneous consumer items, adhesives and glues, and material handling machinery.

Patients With AG+

Of 28 481 patients, 383 (1.3%) had AG+. This group included 165 male patients and 218 female patients; common additional body sites included the trunk (91 [23.8%]), hand (76 [19.8%]), and leg (67 [17.5%]). Trunk involvement was 2.09 times (95% CI, 1.66-2.64; P < .001) as frequent and leg involvement was 1.69 times (95% CI, 1.30-2.20; P < .001) as frequent in patients with AG+ compared with those in the noAG group.

Discussion

Of 28 481 patients in the 2005-2016 NACDG database, 832 (2.9%) had anogenital involvement and 449 (1.6%) had anogenital involvement only. To our knowledge, the NACDG database is the largest database that has both characterized AG patients and identified the clinical relevance of allergic reactions.

Of 449 patients in the AG group, 237 (52.8%) had a final diagnosis of ACD, which is higher than in studies by Bhate et al5 (30.0%) and González-Pérez et al3 (32.4%); those studies had small sample sizes (n = 37 in each study). In our study, patients in the AG group had a significantly higher frequency of other dermatoses (eg, lichen planus, lichen sclerosus, balanitis, or vulvodynia) and other dermatitis (eg, lichen simplex chronicus) compared with those in the noAG group, likely owing to the high prevalence of other genital dermatoses among those suspected to have ACD.1

Both the previous 1994-2004 NACDG study9 and present study found higher proportions of male patients in the AG group compared with the noAG group (39.5% vs 35.3% in the previous NACDG study9 and 39.4% vs 32.0% in the present study), consistent with studies by González-Pérez et al3 (59.5% vs 36.2%) and Bhate et al5 (51.0% vs 29.0%). We also found that, within the AG group, both the previous and current NACDG studies found a higher frequency of male patients than female patients with a final diagnosis of ACD (49.6% vs 40.5% in the previous NACDG study9 and 60.5% vs 47.8% in the present study). The explanation for male predominance is unclear but could be due to biostructural differences, sexual behaviors, product choices, and/or hygiene practices. It could also be due to female patients seeking care from obstetricians or gynecologists rather than dermatologists for anogenital symptoms. However, no studies, to our knowledge, have investigated these factors.

Currently Clinically Relevant Allergens and Sources

Of 449 patients in the AG group, 227 (50.6%) had 1 or more positive allergic reactions of current clinical relevance, similar to other studies (Table 1).2,3,4,5,6,7,8,9 This finding is lower than that in the prior NACDG study (63%)9; the significance of this difference is unclear. Compared with the prior NACDG study of patients with anogenital dermatitis, statistically significantly more common allergens in the AG vs noAG groups were similar. One exception was preservatives. Compared with individuals in the noAG group, those in the AG group had a significantly higher frequency of relevant reactions to MI-MCI and ethylenediamine, which was not seen in the prior NACDG study.9 This finding is not surprising given the worldwide epidemic of MI allergy.10 Two other studies found that MI-MCI was the most common relevant allergen among patients with AG3,5. Although MI allergy was not significantly more frequent among patients with AG in the present study, the NACDG began testing for this allergen in 2013; thus, the numbers may be too small to detect significance. Historically, MI and MI-MCI were common relevant allergens in wet wipes, although wipe manufacturers may be decreasing the use of MI and MCI.11 Ethylenediamine is an emulsifier and stabilizer in generic nystatin and triamcinolone cream.12

Compared with individuals in the noAG group, those in the AG group had a significantly higher frequency of relevant reactions to balsam of Peru. This ingredient is present in a popular anogenital product in the United States, Boudreaux’s Butt Paste. Two studies5,6 found that fragrance and flavoring allergens were the most common allergens among those with anogenital involvement. A study of 45 patients with fragrance ACD by Salam and Fowler13 found that 47% improved with a low-balsam diet; 31% of patients had anogenital involvement. These patients were suspected to have a systemic contact dermatitis from ingestion of balsam of Peru; however, to our knowledge, no formal studies have evaluated this conjecture.

Compared with individuals in the noAG group, those in the AG group had a significantly higher frequency of relevant reactions to 3 topical anesthetics (benzocaine, dibucaine, and lidocaine) and 3 corticosteroids (triamcinolone, budesonide, and desoximetasone). The frequency of reactions to tixocortol pivalate were not significantly different; this finding may be due to the common use of hydrocortisone for treatment of skin conditions at all body sites.

Although allergy to nickel was significantly less common among patients in the AG group compared with those in the noAG group, it was the fifth most common allergen in the most restrictive group (AG-ACD). The most common reported sources of nickel were food products and jewelry. The role of dietary nickel in anogenital dermatitis is poorly understood. In a study by Vermaat et al,6 dietary nickel was not thought to be clearly relevant to patients’ anogenital dermatitis. However, other reports showed an association between the ingestion of nickel and anogenital dermatitis.14

Not all allergens are captured by the NACDG screening series. The most common sources of relevant non-NACDG allergens were personal care products and medications (eg, pramoxine15 and clotrimazole16), none of which were on the NACDG screening series. This finding underscores the need to test beyond a screening series to include patients’ personal care products.

Limitations

The limitations of this study include the following: (1) the use of cross-sectional data, (2) a tertiary referral population, (3) a strict definition of AG-ACD to confidently identify allergens related to the anogenital area, (4) the exclusion of past and unknown relevance, (5) the lack of long-term follow-up data, and (6) no data on prior patch testing.

Conclusions

It is important for patients with anogenital involvement who are suspected of having ACD to undergo patch testing because more than half will likely have 1 or more currently relevant allergic reactions. Male patients were more likely to have anogenital ACD as a final diagnosis, whereas a final diagnosis of other dermatoses was more frequent among female patients. For patients with anogenital ACD, allergens and sources were consistent with items likely to contact the anogenital region, including preservatives, fragrances, medications (particularly topical anesthetics), and topical corticosteroids.

Supplement.

eTable. Demographics: Patients With Anogenital Involvement Only (AG) Compared With Patients Without Anogenital Involvement (noAG)

References

  • 1.Yale K, Awosika O, Rengifo-Pardo M, Ehrlich A. Genital allergic contact dermatitis. Dermatitis. 2018;29(3):112-119. doi: 10.1097/DER.0000000000000371 [DOI] [PubMed] [Google Scholar]
  • 2.Abu-Asi MJ, White IR, McFadden JP, White JM. Patch testing is clinically important for patients with peri-anal dermatoses and pruritus ani. Contact Dermatitis. 2016;74(5):298-300. doi: 10.1111/cod.12514 [DOI] [PubMed] [Google Scholar]
  • 3.González-Pérez R, Sánchez-Martínez L, Piqueres Zubiaurrre T, Urtaran Ibarzábal A, Soloeta Arechavala R. Patch testing in patients with perianal eczema. Actas Dermosifiliogr. 2014;105(7):694-698. English Edition. doi: 10.1016/j.adengl.2014.07.013 [DOI] [PubMed] [Google Scholar]
  • 4.Haverhoek E, Reid C, Gordon L, Marshman G, Wood J, Selva-Nayagam P. Prospective study of patch testing in patients with vulval pruritus. Australas J Dermatol. 2008;49(2):80-85. doi: 10.1111/j.1440-0960.2008.00442.x [DOI] [PubMed] [Google Scholar]
  • 5.Bhate K, Landeck L, Gonzalez E, Neumann K, Schalock PC. Genital contact dermatitis: a retrospective analysis. Dermatitis. 2010;21(6):317-320. doi: 10.2310/6620.2010.10048 [DOI] [PubMed] [Google Scholar]
  • 6.Vermaat H, Smienk F, Rustemeyer T, Bruynzeel DP, Kirtschig G. Anogenital allergic contact dermatitis, the role of spices and flavour allergy. Contact Dermatitis. 2008;59(4):233-237. doi: 10.1111/j.1600-0536.2008.01417.x [DOI] [PubMed] [Google Scholar]
  • 7.Ljubojević S, Lipozencić J, Celić D, Turcić P. Genital contact allergy. Acta Dermatovenerol Croat. 2009;17(4):285-288. [PubMed] [Google Scholar]
  • 8.Bauer A, Oehme S, Geier J. Contact sensitization in the anal and genital area. Curr Probl Dermatol. 2011;40:133-141. doi: 10.1159/000321065 [DOI] [PubMed] [Google Scholar]
  • 9.Warshaw EM, Furda LM, Maibach HI, et al. Anogenital dermatitis in patients referred for patch testing: retrospective analysis of cross-sectional data from the North American Contact Dermatitis Group, 1994-2004. Arch Dermatol. 2008;144(6):749-755. doi: 10.1001/archderm.144.6.749 [DOI] [PubMed] [Google Scholar]
  • 10.DeKoven JG, Warshaw EM, Belsito DV, et al. North American Contact Dermatitis Group PATCH TEST results 2013-2014. Dermatitis. 2017;28(1):33-46. doi: 10.1097/DER.0000000000000225 [DOI] [PubMed] [Google Scholar]
  • 11.Hamann CR, Sahni S, Zug KA. Methylisothiazolinone: still on leave-on products, but no longer on baby wipes. Dermatitis. 2019;30(2):173-174. doi: 10.1097/DER.0000000000000438 [DOI] [PubMed] [Google Scholar]
  • 12.White MI, Douglas WS, Main RA. Contact dermatitis attributed to ethylenediamine. BMJ. 1978;1(6110):415-416. doi: 10.1136/bmj.1.6110.415 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Salam TN, Fowler JF Jr. Balsam-related systemic contact dermatitis. J Am Acad Dermatol. 2001;45(3):377-381. doi: 10.1067/mjd.2001.114738 [DOI] [PubMed] [Google Scholar]
  • 14.Silvestri DL, Barmettler S. Pruritus ani as a manifestation of systemic contact dermatitis: resolution with dietary nickel restriction. Dermatitis. 2011;22(1):50-55. doi: 10.2310/6620.2011.10062 [DOI] [PubMed] [Google Scholar]
  • 15.Schlarbaum JP, Kimyon RS, Liou YL, Becker O’Neill L, Warshaw EM. Genital dermatitis in a transgender patient returning from Thailand: a diagnostic challenge. Travel Med Infect Dis. 2019;27:134-135. doi: 10.1016/j.tmaid.2019.01.007 [DOI] [PubMed] [Google Scholar]
  • 16.Pullen SK, Warshaw EM. Vulvar allergic contact dermatitis from clotrimazole. Dermatitis. 2010;21(1):59-60. doi: 10.2310/6620.2009.09008 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement.

eTable. Demographics: Patients With Anogenital Involvement Only (AG) Compared With Patients Without Anogenital Involvement (noAG)


Articles from JAMA Dermatology are provided here courtesy of American Medical Association

RESOURCES