- Email: [email protected]
The inhibitory effects of topical chelating agents and antioxidants on nickel-induced hypersensitivity reactions
The inhibitory effects of topical chelating agents and antioxidants on nickel-induced hypersensitivity reactions Aamir A. Memon, MBBS, DDS, Mohamed M . Molokhia, MBChB, PhD, and Peter S. Friedmann, MD, FRCP Liverpool, United Kingdom Background: Nickel sensitivity is a common problem, often causing significant morbidity from chronic eczematous dermatitis. The main treatment, topical steroids, usually is unable to suppress the dermatitis completely. Objective: Our purpose was to study the effects of "barrier" ointments containing either chelating agents (clioquinol or ethylenediaminetetraacetic acid [EDTAl) or antioxidants (ascorbic acid or a-tocopherol) and 1% hydrocortisone on nickel-induced hypersensitivity reactions. Methods: Nickel-sensitive subjects were challenged with nickel-containing coins coated with the desired barrier ointment and their inhibitory effects observed. Patients with bilateral hand or earring dermatitis explored the efficacy of these agents in clinical use. Results: Clioquinol (3%) completely abolished the allergic reaction in all 29 subjects tested. EDTA (15%), ascorbic acid (20%), and a-tocopherol (10%) were less effective, and 1% hydrocortisone had no significant effect. In clinical use sites treated with a commercially available preparation containing 3% clioquinol and I % hydrocortisone showed marked clinical improvement in all lO subjects. Conclusion: Clioquinol is a potent inhibitor of nickel-induced hypersensitivity reactions and is feasible to use as a barrier ointment to block the allergenic effects of nickel in sensitive patients. (J AM ACAD DERMATOL 1994;30:560-5.)
Sensitivity to nickel is one of the most common causes of allergic contact dermatitis. Nickel sensitivity is more common in women; usual sources of exposure are ear piercings, costume jewelry, wrist watches, and clothing buckles. In men it is often associated with occupational exposure. In the general population the prevalence of nickel sensitivity in women has been estimated at about 10%.1-3 However, Cronin" reported th at the incidence of nickel sensitivity in patch-tested women rose from 12% in 1967to 21 % in 1976, and during the same period the incidence in men rose from 1% to 4%. Lunder" reported a similar rise in women; in the age group 21 to 30 years the incidence increased from 11% ( 1977 to 1981) to 19% (1982 to 1986) . The underlying From the Department of Dermatology, University of Liverpool, P.O. Box 147, Liverpool, L69 3BX, ux. Accepted for publication Sept. 20, 1993. Reprint requests: P. S. Friedmann, MD, FRCP, Department of Dermatology, University of Liverpool, P .O. Box 147,Liverpool L69
3BX, U.K. Copyright @ 1994 by the American Academy of Dermatology, Inc. 0190-9622/94 $3.00 + 0 16/1/51559
560
mechanism of allergic contact dermatitis is cell-mediated (delayed-type) hypersensitivity.f Although the ideal strategy for therapy is avoidance of nickel, this is often impossible and may be incompatible with continued performance of a job. Use of antiinflammatory topical steroids is usually required, but even the most potent steroids are frequently unable to suppress the dermatitis adequately. Another therapeutic approach is the use of chelating agents , given either systemically or topically. Diethyldithiocarbamate, tetraethylthiuram disulfide, and triethylenetetramine have been given systemically with limited success, but these drugs are toxic and abnormal liver function develops in many patients.I: 8 An alternative therapeutic possibility is to use chelating agents topically. Allenby and Goodwin? combined EDTA with 5% nickel sulfate (NiS04) and showed that the number and grade of positive patch test reactions to NiS04 were significantly reduced. Van Ketel and Bruynzeel'" showed that there is a reduction in the number and severity of patch test reactions to 1% NiS04 when the test site was pretreated with 10% EDTA.
Journal of the American Academy of Dermatology Volume 30, Number 4
Fullerton and Hoelgaard 11 compared topical EDTA with other chelating agents, including L-histidine and D-penicillamine, and concluded that EDTA was the most effective chelator and D-penicillamine the least effective. Fischer and Rystedt'? examined the metal chelating properties of 5-chloro-7-iodo-8-hydroxyquinoline (clioquinol, iodochlorhydroxyquin) and found that it can suppress the allergic responses elicited by cobalt chloride and potassium dichromate. Yet another possibility, suggested by Hemingway and Molokhia.P is the use of antioxidants that could reduce the oxygen-dependent dissolution of nickel in sweat. In the present study clioquinol, other chelating agents, and antioxidants were examined for their capacity to block the elicitation of allergic contact hypersensitivity reactions by nickel. The first part of the study examined the effects of concentration, dose, vehicle base, and quantities applied. In the second part pilot studies of efficacy in clinical use were undertaken. MATERIAL AND METHODS The suitability of nickel-containing "silver" coins for elicitation of nickel allergy was compared with that of the standard patch test allergen, 5% NiS04 in yellow soft paraffin. New silver coins (20-pence pieces that contain 15.75% nickel) were cleaned with detergent and ethanol. One coin was applied to the back together with the standard European patch test series on 347 consecutivepatients. All patients gave informed consent. The coin and other allergens were removed after 48 hours, and each site was scored at 48 hours and 96 hours as follows: 0, no reaction ; ± , doubtful reaction; +, weak (nonvesicular) reaction ; ++, strong (edematous or vesicular) reaction; and +++, severe reaction." Barrier ointments were prepared by mixing either 10% clioquinol, 15% EDTA, 20% ascorbic acid, or 10% extocopherol in yellow soft paraffin. The irritancy potential of each barrier ointment was examined. Ten patients were patch tested with each of these ointments in a Finn Chamber. The patches were removed after 48 hours and each site was scored at 48 and 96 hours, as described. To examine the effects of varying the quantity of ointment, known quantities were dispensed directly onto the coin. Fifty, 100, 150, or 200 mg of 10% clioquinol, 15% EDTA, or base were dispensed onto the coin with a syringe. Five nickel-sensitive volunteers took part in the study. The ointment-coated sides of the silver coins were applied to the skin of the back and secured with adhesive tape. Each subject was challenged with 12 coins coated with the desired quantity of ointments. The coins were
M emon et ai. 561 removed after 48 hours, and each site was scored visually at 72 hours, as described. To examine the inhibitory effects of each barrier ointment on nickel-induced allergic reactions, comparisons were made between a range of dilutions of clioquinol in paraffin base (10%, 3%, 1%, 0.3%), 3% clioquinol and 1% hydrocortisone cream (Vioform He cream [ZYMA]), 15% EDTA in base, 20% ascorbic acid in base, 10% atocopherol in base, 1% hydrocortisone cream, and base alone as a control. Silver coins were coated with 150 mg of the ointment or cream to be tested because maximal inhibition was obtained with a quantity greater than 100 mg. Twenty-nine nickel-sensitive volunteers took part in this study. Each subject was tested with a range of ointments and creams including two to four coins coated with base alone as a control. The coated side of the silver coins was applied to the skin of the back for 48 hours, and each site was scored visually at 72 hours as described. To control for the possibility that the base itself could act a s a physical barrier, comparison was made of the allergic reactions induced by nickel-containing coins applied either coated with base or uncoated. Twenty-nine nickel-sensitive volunteers took part in this study. Each subject waschallenged with three coins, ofwhich two were coated with base (150 mg) and one was uncoated . The coated side of the coin was applied to the skin of the back for 48 hours and each site was scored visually at 72 hours . To examine the inhibitory effects of c1ioquinolon nickel salts, comparisons were made between the allergic reactions induced by NiS0 4 alone and NiS04 with added clioquinol. For the purpose of chelation two molecules of clioquinol are required to bind one molecule of nickel. Because the molecular weights of clioquinol and NiS04 are 305 and 58.7 kd, respectively, one part of 10% clioquinol can chelate one part of 0.95% nickel ion. A dilution of 0.625% NiS04 was prepared in base. To one aliquot clioquinol was added to a final concentration of 10%. Twelve nickel-sensitivevolunteers were patch tested with 0.625% NiS0 4 with and without the added clioquinol. Patches were removed a fter 48 hours, and each site was scored visually at 72 hours. . The clinical effects of clioquinol were examined in a small pilot study in which 10 nickel-sensitive volunteers participated. Five had a history of recurrent bilateral earring dermatitis and five had active bilateral hand eczema . The patients with previous earring dermatitis had no existing eczema on the earlobes when tested. They were asked to wear nickel-containing earrings for 2 days , which would usually produce a dermatitis. During this time they applied 3% clioquinol and 1% hydrocortisone Cream to one side and 1% hydrocortisone cream to the control side three times daily. The sides were compared after 3 days, and the degree of eczema was scored as follows: 0, no eczema, 1, mild eczema, 2, moderate eczema, 3, severe eczema.
562
Memon et
Journal of the American Ac adem y of Dermatology April 1994
at. +++
A
-
r::
IV
"'C Q)
29
++
11
~
24
w
a: 0
11
+
U
en
-J
-c
:::>
en >
14
+/ -
x~
-va YSP
100
VitE
HC
25 VitC
EDTA VioHC
29
ell
25
29*
VioHC
CLI
B
CJ N Z
Z
80
~ 0
0
::I:
i=
z
::I: Z
en rn 0
I- W a:: I-
0 W a.. -J a.. 0 a:: ~ a.. 0
60
40
17
20
U
0
11
11
24
VitE
VitC
HC
29 YSP
EDT A
Fig. 1. Inhibitory effects of each ointment on nickel coin-induced allergic hypersensitivity reactions. A, Median reaction scores. B, Proportion of people in whom response was completely abrogated. CLI , Clioquinol 3%; EDTA, 15% EDTA ; HC, I% hydrocortisone, VioHC, 3% ciioquinol/l%hydrocortisone; vue, 20% ascorbic acid; vue, 10%a- tocopherol; YSP, yellow soft paraffin . Asterisk, Number of persons tested. The five nickel-sensitive persons with bilat eral hand dermatitis applied 3% clioquinoij I% hydrocortisone cream to one hand and 1% hydrocortisone cream to the other with rubber gloves or a wooden spatula to minimize cross-contamination . The sides were compared before and after 2 weeks, and the degree of eczema was evaluated by an observer who did not know on which side each treatment had been used. St atistical analysis of results was made with Fisher's exact test.
RESULTS O f 347 subjects patch tested with the European Standard Battery and a metallic coin, 97 showed a positive reaction to 5% NiS0 4, a nd of these 87
(91 %) reacted to the me tallic coin. On ly one patient gave a positive reaction to the coin but not to 5% NiS04. To see whether the base itself acted as a physical barrier, responses were compared between the coins coated wit h base a nd the uncoated coins in 29 subjects, none of whom showed any inhibitory effec ts with the base (F igs. 1 and 2). The irritancy pot ent ial of each barrier ointmen t was examined in 10 volunteers, and none of the barrier ointment elicit ed any irr itan t reaction. To ensure that t he quantity of ointment applied was not critical, coins were loaded with different qu antities of ointments. With 10% clioquinol the inhibitory effect was maximal with 100 mg of oint-
Journal of the Ame rican Academy of Dermatology Volume 30, Number 4
M em on et al. 563
Fig. 2. Responses ind uced by nickel-cont aining coins coated with various ointments. CLI, 3% c1ioquinoIin yellow soft paraffin; COIN, uncoated application ofcoin; EDTA , 15% EDT A in yellow soft paraffin; He, 1% hydr ocortisone; YSP, yellow soft paraffin.
Table I. Clinical scores showing changes in responses by alter ing quantity of ointments 15% EDTA in YSP (mg)
YSP (mg)
Patient No.
1
2 3 4
5
50
I
++ ++ ++ ++ ++
100
++ ++ ++
++ ++
I
150
++ ++ ++ ++ ++
I
200
50
++ ++ ++
++ ++ ++ ++ ++
+
++
I
100
++ ++
+
+ ++
I
150
++ ++
o
+
++
10 % c1ioquinol in YSP (mg)
I
200
50
++ ++
+
o o o
o o o
+
I
100
o o o o o
I
150
o o o o o
I
200
o o o o o
YSP, Yellow soft paraffin.
ment. With 15% EDTA the effect increased with increasing quantity. However , the maximum quantity of EDTA (200 mg) did not give complete inhibition in all subjects . The base alone showed no inhibitory effects, but at the maximum quanti ty of200 mg there was a smaU reduction in the intensity of reaction in one patien t (Table I). When the inhibitory effects of various barrier creams were examined, 10% clioquinol completely abrogat ed the allergic responses in all seven nickelsensitive subjects tested (Fig. 3). Therefore the effect of concentration was examined on 29 nickel-sensitive subje cts. Each was challenged with a series of threefold dilutions (3%, 1%, 0.3%, and 0.1%) of clioquinol. There was a dose-related effect so th at the nickel-induced reactions were abrogated in all 29 patients (100%) tested with 3% clioquinol (Figs. 1 to 3) (X 2 = 58, df= 1, p < 0.000l) and in 20 of 22 subjects (91%) with 1% clioquinol (Fig. 3) whereas
the 0.3% dilution was inhibitory in only 12 of 22 subjects (55%) (Fig. 3). Vioform He cream, wh ich contains 3%clioquinol and 1% hydrocortisone, was tested in 25 subjects, and it also abrogated the allergic responses completely in all 25 subjects (100%) tested (Fig. 1). EDTA ( 15%) was less effective, inhibiting the responses completely in only 6 of 17 subjects (35%) and partially in 7 of 17 subjects (41%) (Figs. 1 and 2) (compared with 3% clioquinol; X2 = 11.7, df= 1, P = 0.0006); 20% ascorbic ' acid inhibited the response completely in only 1 of 11 subjects (9 %) a nd partially in 3 of 11 subjects (2 7%) (Fig. 1); and 10% a -toco pherol inhibited the response in only 1 of 11 (9%) subjects tested (F ig. 1). Hydrocortisone 1% cream used as a control was less effective and inhibited the response completely in only 2 of 24 (8%) and partially in 5 of24 subjects (20%) tested (Figs. 1 and 2) (x 2 = 2.5, df= 1, p > 0.05).
Journal of the American Academy of Dermatology April 1994
564 Memon et al. 100
<.'E ~ Z
29
7
3"
101
22
80
~O
0-
xt:
l/J!E
z
60
22
X
o~ i= w
o0:1W Cl.-1 o
e, 0: ~ e, 0
40
20
U
0.3"
11
CONCENTRATION OF CLioaUINOL
Fig. 3. Inhibition of nickel coin-induced allergic reactions by different concentrations of clioquinol.
Inhibitory effects of clioquinol added to NiS04 were examined in 12 subjects. All subjects gave a positive response to 0.625% NiS04 in base; however, responses to the mixture (0.625% NiS04 and 10% c1ioquinol) showed a marked reduction. Thus in six subjects there was no response and there was a weak response in the other six. In the pilot study of clinical use, all five subjects with bilateral earring dermatitis applied 3% clioquinol and 1% hydrocortisone cream three times daily to one side and 1% hydrocortisone to the other while continuing to wear the provoking earrings. All showed inhibition of allergic responses on the 3% clioquinol/ 1% hydrocortisone-treated side (median score, 1) whereas the 1% hydrocortisone creamtreated side showed no inhibitory effects (median score, 3). In patients with bilateral hand eczema the 3% clioquinol/ 1% hydrocortisone-treated side showed clinical improvement in all five subjects; the median score changed from 3 to 1. The 1% hydrocortisonetreated side showed a small clinical improvement in two subjects, but the median score remained 3. DISCUSSION
In this study we showed that nickel-containing silver coins are useful to elicit nickel sensitivity. It is of particular value as a model challenge system to mimic everyday exposure to metallic nickel. With this challenge system our main finding was that clioquinol completely abrogated allergic reactions induced by metallic nickel or nickel salt. This effect was dose-related, and although 100% of reactions
were abolished by 3% clioquinol, inhibition was evident down to 0.3%. . Clioquinol is an oxine used therapeutically as an antimicrobial agent. Its antimicrobial effects are believedto result from its strong chelating properties that bind to bivalent heavy metals. 15 That the effect was caused by chelating property of clioquinol is supported by the observation that EDTA (15%), another chelating agent, was also able to block the nickel-induced reaction, although to a lesser degree. This was not an effect of the base grease because coating the coins with base alone had no inhibitory effects. Clioquino1 3%/hydrocortisone 1% cream was equally effective,but 1%hydrocortisone cream, ascorbic acid, and a-tocopherol had little or no effect. Allergic responses to nickel salts (NiS0 4) were also suppressed by clioquinol. However, this response was only seen at a low concentration of NiS04 (0.625%). There was no response in 6 of 12 subjects, but small responses were seen in the remaining six subjects. By combining the calculated molar ratios of 0.625% NiS04 and 10% clioquinol we expected a complete inhibition. However, because NiS04 is a water-soluble substance, it is possible that the combination of the two in the vehicle was imperfect and enough NiS04 was available to elicit these small reactions. To see whether clioquinol has direct antiinflammatory properties, 10% c1ioquinol was mixed in equal parts with 5% NiS0 4, which completely saturated the chelating capacity. There was no inhibitory effect on nickel-induced responses in four subjects tested, indicating that clio-
Journal of the American Academy of Dermatology Volume 30, Number 4
quinol has no other effect than chelation of nickel ions. The potential clinical value of clioquinol as a barrier was explored in a pilot study in persons with bilateral nickel dermatitis. Earring dermatitis as a test system has the advantage that cross-contamination can be avoided. There was a good protective effect when volunteers used the 3% clioquinol and 1% hydrocortisone cream even while continuing to wear the provoking earrings. In the nickel-sensitive persons with bilateral hand dermatitis we noted a marked clinical improvement on the 3% clioquinol/ l % hydrocortisone cream-treated side, but in some subjects there was bilateral improvement. Although we encouraged our patients to apply the cream to each hand with either a wooden spatula or by wearing a rubber glove, it is likely that cross-contamination occurred. We presume that clioquinol might act in two ways. First, by chelation of nickel ions at the skin surface, the antigenic effects of nickel-containing objects are blocked. Second, because systemically absorbed nickel may be excreted through sweat.l'' another possibility is that this nickel might be chelated. Our results indicate that clioquinol as a chemical barrier could be a useful part of therapy for nickel dermatitis. However, clioquinol is itself a weak allergen and a low incidence of sensitization has been reported in several large patch test series.17 We are grateful to the Royal Mint for supplying unused silver coins for patch testing. REFERENCES I. Kieffer M. Nickel sensitivity: relationship between history and patch test reaction . Contact Dermatitis 1979;5:398401.
Memon et al. 565 2. Peltonen L. Nickel sensitivity in the general population. Contact Dermatitis 1979;5:27-32. 3. N ielsen NH, Menne T. Allergic contact sensitization in an unselected Danish population. Acta Derm Venereal (Stockh) 1992;72:456-60. 4. Cronin E, ed. Metals . In: Contact dermatitis. Edinburgh: Churchill Livingstone, 1980:344. 5. Lunder M. Variable incidence of nickel dermatitis . 1988;18:287-9. 6. Friedmann PS. Contact hypersensitivity. Curr Opin Immuno! 1989;1:690-3. 7. SpruitD, BongaartsPJM, De Jongh GJ. Dithiocarbamate therapy for nickel dermatitis. Contact Dermatitis 1978; 4:350-8. 8. Kaaber K, Menne T, Tjell JC, et al. Antabuse treatment of nickel dermatitis: chelation-a new principle in the treatmentofnickeldermatitis. Contact Dermatitis 1979;5:221-8. 9. Allenby CF, Goodwin BFJ. Influence of detergent washing powders on minimal eliciting patch test concentrations of nickel and chromium . Contact Dermatitis 1983;9:491-9. 10. Van Ketel WG, Bruynzeel DP. Chelating effect of EDTA on nickel. Contact Dermatitis 1984;11:311-4. 11. Fullerton A, Hoelgaard A. Binding of nickel to human epidermis in vitro. Br J DermatoI1988;119:675-82. 12. Fischer T, Rystedt 1. Influence of topical metal binding substances, vehicles, and corticosteroid creams on the allergic patch test reaction in metal-sensitive patients. Dermato! Clin 1990;8,1:27-31. 13. Hemingway JD , Molokh ia M . The dissolution of metallic nickel in artificial sweat. Contact Dermatitis 1987;16:99-
105. 14. International Contact Dermatitis Research Group. Terminologyofcontact dermatitis.Acta Derm Venereal (Stockh) . 1970;50:287-92. 15. Fischer T. On 8-hydroxyquinoline-zinc oxide incompatibility. Dermatologica 1974; 149:129-35. 16. Christensen OB, Moller H, Andrasko L, et al. Nicke! concentration of blood, urine and sweat after oral administration. Contact Dermatitis 1979;5:312-6. 17. Cronin E, ed. Medicamen ts. In: Contact dermatitis. Edinburgh: Churchill-Livingstone, 1980:218-9.
Sensitivity to nickel is one of the most common causes of allergic contact dermatitis. Nickel sensitivity is more common in women; usual sources of exposure are ear piercings, costume jewelry, wrist watches, and clothing buckles. In men it is often associated with occupational exposure. In the general population the prevalence of nickel sensitivity in women has been estimated at about 10%.1-3 However, Cronin" reported th at the incidence of nickel sensitivity in patch-tested women rose from 12% in 1967to 21 % in 1976, and during the same period the incidence in men rose from 1% to 4%. Lunder" reported a similar rise in women; in the age group 21 to 30 years the incidence increased from 11% ( 1977 to 1981) to 19% (1982 to 1986) . The underlying From the Department of Dermatology, University of Liverpool, P.O. Box 147, Liverpool, L69 3BX, ux. Accepted for publication Sept. 20, 1993. Reprint requests: P. S. Friedmann, MD, FRCP, Department of Dermatology, University of Liverpool, P .O. Box 147,Liverpool L69
3BX, U.K. Copyright @ 1994 by the American Academy of Dermatology, Inc. 0190-9622/94 $3.00 + 0 16/1/51559
560
mechanism of allergic contact dermatitis is cell-mediated (delayed-type) hypersensitivity.f Although the ideal strategy for therapy is avoidance of nickel, this is often impossible and may be incompatible with continued performance of a job. Use of antiinflammatory topical steroids is usually required, but even the most potent steroids are frequently unable to suppress the dermatitis adequately. Another therapeutic approach is the use of chelating agents , given either systemically or topically. Diethyldithiocarbamate, tetraethylthiuram disulfide, and triethylenetetramine have been given systemically with limited success, but these drugs are toxic and abnormal liver function develops in many patients.I: 8 An alternative therapeutic possibility is to use chelating agents topically. Allenby and Goodwin? combined EDTA with 5% nickel sulfate (NiS04) and showed that the number and grade of positive patch test reactions to NiS04 were significantly reduced. Van Ketel and Bruynzeel'" showed that there is a reduction in the number and severity of patch test reactions to 1% NiS04 when the test site was pretreated with 10% EDTA.
Journal of the American Academy of Dermatology Volume 30, Number 4
Fullerton and Hoelgaard 11 compared topical EDTA with other chelating agents, including L-histidine and D-penicillamine, and concluded that EDTA was the most effective chelator and D-penicillamine the least effective. Fischer and Rystedt'? examined the metal chelating properties of 5-chloro-7-iodo-8-hydroxyquinoline (clioquinol, iodochlorhydroxyquin) and found that it can suppress the allergic responses elicited by cobalt chloride and potassium dichromate. Yet another possibility, suggested by Hemingway and Molokhia.P is the use of antioxidants that could reduce the oxygen-dependent dissolution of nickel in sweat. In the present study clioquinol, other chelating agents, and antioxidants were examined for their capacity to block the elicitation of allergic contact hypersensitivity reactions by nickel. The first part of the study examined the effects of concentration, dose, vehicle base, and quantities applied. In the second part pilot studies of efficacy in clinical use were undertaken. MATERIAL AND METHODS The suitability of nickel-containing "silver" coins for elicitation of nickel allergy was compared with that of the standard patch test allergen, 5% NiS04 in yellow soft paraffin. New silver coins (20-pence pieces that contain 15.75% nickel) were cleaned with detergent and ethanol. One coin was applied to the back together with the standard European patch test series on 347 consecutivepatients. All patients gave informed consent. The coin and other allergens were removed after 48 hours, and each site was scored at 48 hours and 96 hours as follows: 0, no reaction ; ± , doubtful reaction; +, weak (nonvesicular) reaction ; ++, strong (edematous or vesicular) reaction; and +++, severe reaction." Barrier ointments were prepared by mixing either 10% clioquinol, 15% EDTA, 20% ascorbic acid, or 10% extocopherol in yellow soft paraffin. The irritancy potential of each barrier ointment was examined. Ten patients were patch tested with each of these ointments in a Finn Chamber. The patches were removed after 48 hours and each site was scored at 48 and 96 hours, as described. To examine the effects of varying the quantity of ointment, known quantities were dispensed directly onto the coin. Fifty, 100, 150, or 200 mg of 10% clioquinol, 15% EDTA, or base were dispensed onto the coin with a syringe. Five nickel-sensitive volunteers took part in the study. The ointment-coated sides of the silver coins were applied to the skin of the back and secured with adhesive tape. Each subject was challenged with 12 coins coated with the desired quantity of ointments. The coins were
M emon et ai. 561 removed after 48 hours, and each site was scored visually at 72 hours, as described. To examine the inhibitory effects of each barrier ointment on nickel-induced allergic reactions, comparisons were made between a range of dilutions of clioquinol in paraffin base (10%, 3%, 1%, 0.3%), 3% clioquinol and 1% hydrocortisone cream (Vioform He cream [ZYMA]), 15% EDTA in base, 20% ascorbic acid in base, 10% atocopherol in base, 1% hydrocortisone cream, and base alone as a control. Silver coins were coated with 150 mg of the ointment or cream to be tested because maximal inhibition was obtained with a quantity greater than 100 mg. Twenty-nine nickel-sensitive volunteers took part in this study. Each subject was tested with a range of ointments and creams including two to four coins coated with base alone as a control. The coated side of the silver coins was applied to the skin of the back for 48 hours, and each site was scored visually at 72 hours as described. To control for the possibility that the base itself could act a s a physical barrier, comparison was made of the allergic reactions induced by nickel-containing coins applied either coated with base or uncoated. Twenty-nine nickel-sensitive volunteers took part in this study. Each subject waschallenged with three coins, ofwhich two were coated with base (150 mg) and one was uncoated . The coated side of the coin was applied to the skin of the back for 48 hours and each site was scored visually at 72 hours . To examine the inhibitory effects of c1ioquinolon nickel salts, comparisons were made between the allergic reactions induced by NiS0 4 alone and NiS04 with added clioquinol. For the purpose of chelation two molecules of clioquinol are required to bind one molecule of nickel. Because the molecular weights of clioquinol and NiS04 are 305 and 58.7 kd, respectively, one part of 10% clioquinol can chelate one part of 0.95% nickel ion. A dilution of 0.625% NiS04 was prepared in base. To one aliquot clioquinol was added to a final concentration of 10%. Twelve nickel-sensitivevolunteers were patch tested with 0.625% NiS0 4 with and without the added clioquinol. Patches were removed a fter 48 hours, and each site was scored visually at 72 hours. . The clinical effects of clioquinol were examined in a small pilot study in which 10 nickel-sensitive volunteers participated. Five had a history of recurrent bilateral earring dermatitis and five had active bilateral hand eczema . The patients with previous earring dermatitis had no existing eczema on the earlobes when tested. They were asked to wear nickel-containing earrings for 2 days , which would usually produce a dermatitis. During this time they applied 3% clioquinol and 1% hydrocortisone Cream to one side and 1% hydrocortisone cream to the control side three times daily. The sides were compared after 3 days, and the degree of eczema was scored as follows: 0, no eczema, 1, mild eczema, 2, moderate eczema, 3, severe eczema.
562
Memon et
Journal of the American Ac adem y of Dermatology April 1994
at. +++
A
-
r::
IV
"'C Q)
29
++
11
~
24
w
a: 0
11
+
U
en
-J
-c
:::>
en >
14
+/ -
x~
-va YSP
100
VitE
HC
25 VitC
EDTA VioHC
29
ell
25
29*
VioHC
CLI
B
CJ N Z
Z
80
~ 0
0
::I:
i=
z
::I: Z
en rn 0
I- W a:: I-
0 W a.. -J a.. 0 a:: ~ a.. 0
60
40
17
20
U
0
11
11
24
VitE
VitC
HC
29 YSP
EDT A
Fig. 1. Inhibitory effects of each ointment on nickel coin-induced allergic hypersensitivity reactions. A, Median reaction scores. B, Proportion of people in whom response was completely abrogated. CLI , Clioquinol 3%; EDTA, 15% EDTA ; HC, I% hydrocortisone, VioHC, 3% ciioquinol/l%hydrocortisone; vue, 20% ascorbic acid; vue, 10%a- tocopherol; YSP, yellow soft paraffin . Asterisk, Number of persons tested. The five nickel-sensitive persons with bilat eral hand dermatitis applied 3% clioquinoij I% hydrocortisone cream to one hand and 1% hydrocortisone cream to the other with rubber gloves or a wooden spatula to minimize cross-contamination . The sides were compared before and after 2 weeks, and the degree of eczema was evaluated by an observer who did not know on which side each treatment had been used. St atistical analysis of results was made with Fisher's exact test.
RESULTS O f 347 subjects patch tested with the European Standard Battery and a metallic coin, 97 showed a positive reaction to 5% NiS0 4, a nd of these 87
(91 %) reacted to the me tallic coin. On ly one patient gave a positive reaction to the coin but not to 5% NiS04. To see whether the base itself acted as a physical barrier, responses were compared between the coins coated wit h base a nd the uncoated coins in 29 subjects, none of whom showed any inhibitory effec ts with the base (F igs. 1 and 2). The irritancy pot ent ial of each barrier ointmen t was examined in 10 volunteers, and none of the barrier ointment elicit ed any irr itan t reaction. To ensure that t he quantity of ointment applied was not critical, coins were loaded with different qu antities of ointments. With 10% clioquinol the inhibitory effect was maximal with 100 mg of oint-
Journal of the Ame rican Academy of Dermatology Volume 30, Number 4
M em on et al. 563
Fig. 2. Responses ind uced by nickel-cont aining coins coated with various ointments. CLI, 3% c1ioquinoIin yellow soft paraffin; COIN, uncoated application ofcoin; EDTA , 15% EDT A in yellow soft paraffin; He, 1% hydr ocortisone; YSP, yellow soft paraffin.
Table I. Clinical scores showing changes in responses by alter ing quantity of ointments 15% EDTA in YSP (mg)
YSP (mg)
Patient No.
1
2 3 4
5
50
I
++ ++ ++ ++ ++
100
++ ++ ++
++ ++
I
150
++ ++ ++ ++ ++
I
200
50
++ ++ ++
++ ++ ++ ++ ++
+
++
I
100
++ ++
+
+ ++
I
150
++ ++
o
+
++
10 % c1ioquinol in YSP (mg)
I
200
50
++ ++
+
o o o
o o o
+
I
100
o o o o o
I
150
o o o o o
I
200
o o o o o
YSP, Yellow soft paraffin.
ment. With 15% EDTA the effect increased with increasing quantity. However , the maximum quantity of EDTA (200 mg) did not give complete inhibition in all subjects . The base alone showed no inhibitory effects, but at the maximum quanti ty of200 mg there was a smaU reduction in the intensity of reaction in one patien t (Table I). When the inhibitory effects of various barrier creams were examined, 10% clioquinol completely abrogat ed the allergic responses in all seven nickelsensitive subjects tested (Fig. 3). Therefore the effect of concentration was examined on 29 nickel-sensitive subje cts. Each was challenged with a series of threefold dilutions (3%, 1%, 0.3%, and 0.1%) of clioquinol. There was a dose-related effect so th at the nickel-induced reactions were abrogated in all 29 patients (100%) tested with 3% clioquinol (Figs. 1 to 3) (X 2 = 58, df= 1, p < 0.000l) and in 20 of 22 subjects (91%) with 1% clioquinol (Fig. 3) whereas
the 0.3% dilution was inhibitory in only 12 of 22 subjects (55%) (Fig. 3). Vioform He cream, wh ich contains 3%clioquinol and 1% hydrocortisone, was tested in 25 subjects, and it also abrogated the allergic responses completely in all 25 subjects (100%) tested (Fig. 1). EDTA ( 15%) was less effective, inhibiting the responses completely in only 6 of 17 subjects (35%) and partially in 7 of 17 subjects (41%) (Figs. 1 and 2) (compared with 3% clioquinol; X2 = 11.7, df= 1, P = 0.0006); 20% ascorbic ' acid inhibited the response completely in only 1 of 11 subjects (9 %) a nd partially in 3 of 11 subjects (2 7%) (Fig. 1); and 10% a -toco pherol inhibited the response in only 1 of 11 (9%) subjects tested (F ig. 1). Hydrocortisone 1% cream used as a control was less effective and inhibited the response completely in only 2 of 24 (8%) and partially in 5 of24 subjects (20%) tested (Figs. 1 and 2) (x 2 = 2.5, df= 1, p > 0.05).
Journal of the American Academy of Dermatology April 1994
564 Memon et al. 100
<.'E ~ Z
29
7
3"
101
22
80
~O
0-
xt:
l/J!E
z
60
22
X
o~ i= w
o0:1W Cl.-1 o
e, 0: ~ e, 0
40
20
U
0.3"
11
CONCENTRATION OF CLioaUINOL
Fig. 3. Inhibition of nickel coin-induced allergic reactions by different concentrations of clioquinol.
Inhibitory effects of clioquinol added to NiS04 were examined in 12 subjects. All subjects gave a positive response to 0.625% NiS04 in base; however, responses to the mixture (0.625% NiS04 and 10% c1ioquinol) showed a marked reduction. Thus in six subjects there was no response and there was a weak response in the other six. In the pilot study of clinical use, all five subjects with bilateral earring dermatitis applied 3% clioquinol and 1% hydrocortisone cream three times daily to one side and 1% hydrocortisone to the other while continuing to wear the provoking earrings. All showed inhibition of allergic responses on the 3% clioquinol/ 1% hydrocortisone-treated side (median score, 1) whereas the 1% hydrocortisone creamtreated side showed no inhibitory effects (median score, 3). In patients with bilateral hand eczema the 3% clioquinol/ 1% hydrocortisone-treated side showed clinical improvement in all five subjects; the median score changed from 3 to 1. The 1% hydrocortisonetreated side showed a small clinical improvement in two subjects, but the median score remained 3. DISCUSSION
In this study we showed that nickel-containing silver coins are useful to elicit nickel sensitivity. It is of particular value as a model challenge system to mimic everyday exposure to metallic nickel. With this challenge system our main finding was that clioquinol completely abrogated allergic reactions induced by metallic nickel or nickel salt. This effect was dose-related, and although 100% of reactions
were abolished by 3% clioquinol, inhibition was evident down to 0.3%. . Clioquinol is an oxine used therapeutically as an antimicrobial agent. Its antimicrobial effects are believedto result from its strong chelating properties that bind to bivalent heavy metals. 15 That the effect was caused by chelating property of clioquinol is supported by the observation that EDTA (15%), another chelating agent, was also able to block the nickel-induced reaction, although to a lesser degree. This was not an effect of the base grease because coating the coins with base alone had no inhibitory effects. Clioquino1 3%/hydrocortisone 1% cream was equally effective,but 1%hydrocortisone cream, ascorbic acid, and a-tocopherol had little or no effect. Allergic responses to nickel salts (NiS0 4) were also suppressed by clioquinol. However, this response was only seen at a low concentration of NiS04 (0.625%). There was no response in 6 of 12 subjects, but small responses were seen in the remaining six subjects. By combining the calculated molar ratios of 0.625% NiS04 and 10% clioquinol we expected a complete inhibition. However, because NiS04 is a water-soluble substance, it is possible that the combination of the two in the vehicle was imperfect and enough NiS04 was available to elicit these small reactions. To see whether clioquinol has direct antiinflammatory properties, 10% c1ioquinol was mixed in equal parts with 5% NiS0 4, which completely saturated the chelating capacity. There was no inhibitory effect on nickel-induced responses in four subjects tested, indicating that clio-
Journal of the American Academy of Dermatology Volume 30, Number 4
quinol has no other effect than chelation of nickel ions. The potential clinical value of clioquinol as a barrier was explored in a pilot study in persons with bilateral nickel dermatitis. Earring dermatitis as a test system has the advantage that cross-contamination can be avoided. There was a good protective effect when volunteers used the 3% clioquinol and 1% hydrocortisone cream even while continuing to wear the provoking earrings. In the nickel-sensitive persons with bilateral hand dermatitis we noted a marked clinical improvement on the 3% clioquinol/ l % hydrocortisone cream-treated side, but in some subjects there was bilateral improvement. Although we encouraged our patients to apply the cream to each hand with either a wooden spatula or by wearing a rubber glove, it is likely that cross-contamination occurred. We presume that clioquinol might act in two ways. First, by chelation of nickel ions at the skin surface, the antigenic effects of nickel-containing objects are blocked. Second, because systemically absorbed nickel may be excreted through sweat.l'' another possibility is that this nickel might be chelated. Our results indicate that clioquinol as a chemical barrier could be a useful part of therapy for nickel dermatitis. However, clioquinol is itself a weak allergen and a low incidence of sensitization has been reported in several large patch test series.17 We are grateful to the Royal Mint for supplying unused silver coins for patch testing. REFERENCES I. Kieffer M. Nickel sensitivity: relationship between history and patch test reaction . Contact Dermatitis 1979;5:398401.
Memon et al. 565 2. Peltonen L. Nickel sensitivity in the general population. Contact Dermatitis 1979;5:27-32. 3. N ielsen NH, Menne T. Allergic contact sensitization in an unselected Danish population. Acta Derm Venereal (Stockh) 1992;72:456-60. 4. Cronin E, ed. Metals . In: Contact dermatitis. Edinburgh: Churchill Livingstone, 1980:344. 5. Lunder M. Variable incidence of nickel dermatitis . 1988;18:287-9. 6. Friedmann PS. Contact hypersensitivity. Curr Opin Immuno! 1989;1:690-3. 7. SpruitD, BongaartsPJM, De Jongh GJ. Dithiocarbamate therapy for nickel dermatitis. Contact Dermatitis 1978; 4:350-8. 8. Kaaber K, Menne T, Tjell JC, et al. Antabuse treatment of nickel dermatitis: chelation-a new principle in the treatmentofnickeldermatitis. Contact Dermatitis 1979;5:221-8. 9. Allenby CF, Goodwin BFJ. Influence of detergent washing powders on minimal eliciting patch test concentrations of nickel and chromium . Contact Dermatitis 1983;9:491-9. 10. Van Ketel WG, Bruynzeel DP. Chelating effect of EDTA on nickel. Contact Dermatitis 1984;11:311-4. 11. Fullerton A, Hoelgaard A. Binding of nickel to human epidermis in vitro. Br J DermatoI1988;119:675-82. 12. Fischer T, Rystedt 1. Influence of topical metal binding substances, vehicles, and corticosteroid creams on the allergic patch test reaction in metal-sensitive patients. Dermato! Clin 1990;8,1:27-31. 13. Hemingway JD , Molokh ia M . The dissolution of metallic nickel in artificial sweat. Contact Dermatitis 1987;16:99-
105. 14. International Contact Dermatitis Research Group. Terminologyofcontact dermatitis.Acta Derm Venereal (Stockh) . 1970;50:287-92. 15. Fischer T. On 8-hydroxyquinoline-zinc oxide incompatibility. Dermatologica 1974; 149:129-35. 16. Christensen OB, Moller H, Andrasko L, et al. Nicke! concentration of blood, urine and sweat after oral administration. Contact Dermatitis 1979;5:312-6. 17. Cronin E, ed. Medicamen ts. In: Contact dermatitis. Edinburgh: Churchill-Livingstone, 1980:218-9.