1 Nickel

Approximately 20% of patch-tested patients in the United States are allergic to nickel [1].

Many double-blind, placebo-controlled studies have shown that oral challenge with nickel leads to clinical symptoms more frequently in those with positive patch test reactions to nickel than in those with negative patch test reactions to nickel. This chapter will briefly review the evidence that oral nickel causes clinical symptoms but will focus on describing what is known about the pathophysiology of these reactions and their treatment.

In this chapter, the term systemic contact dermatitis to nickel (SCDN) will refer to skin-limited symptoms and findings induced by nickel consumption, while the term systemic nickel allergy syndrome (SNAS) will refer to the combination of skin symptoms and findings with associated systemic symptoms, primarily of the gastrointestinal (GI) tract.

2 Evidence that Nickel Consumption Leads to Clinical Symptoms

There is a substantial body of evidence demonstrating that dietary nickel causes clinical symptoms, so substantial in fact, that there can no longer be a question of whether this phenomenon exists. Briefly, though, there are two obvious types of evidence that dietary nickel consumption leads to clinical symptoms: (1) development of clinical symptoms with oral nickel challenge, especially when the challenge is double-blind, placebo controlled, and (2) resolution of clinical symptoms with dietary nickel restriction.

2.1 Development of Clinical Symptoms with Oral Nickel Challenge

Numerous studies have demonstrated that oral challenge with nickel, typically as nickel sulfate, leads to flares of clinical symptoms in patients who have positive patch tests to nickel, with the percentage of patients having cutaneous symptoms with challenge increasing with the dose [2,3,4,5,6,7,8,9,10,11,12]. A dose-response relationship has also been shown for systemic symptoms resulting from oral challenges [13]. A meta-analysis of nickel challenge studies found a clear dose-response relationship, with it being estimated that roughly 10% of nickel-allergic patients will have a reaction to an oral challenge with 1 mg of elemental nickel [5].

The likelihood of reacting to oral nickel challenge does not appear to correlate with how sensitive the patient is to nickel patch testing, as when serial dilutions of nickel were used for patch testing and serial doses of oral nickel were used for oral challenge, there was no correlation between the minimum concentration that caused a patch test reaction and the likelihood of reacting to oral challenge or the oral dose needed to cause a reaction [14]. Another study did find a correlation between the intensity of patch test reactions and the likelihood of responding to an oral challenge [6].

In those with a positive oral challenge, the likelihood of improving on a low-nickel diet is higher in those with 1+ or 2+ patch test reaction than in those with 3+ reactions [15].

Most studies use challenges with nickel sulfate and can be criticized because the dose of nickel used is higher than could be consumed in a single meal. There are fewer studies that look at challenge with a diet naturally high in nickel. In one such study, 12 nickel-allergic patients were challenged in a blinded manner with a high-nickel diet for 4 days (containing almost 500 ug/d of nickel). Urinary nickel excretion quadrupled while on the diet. At day 4, 50% flared based on both patient and investigator assessment, and at day 11 (7 days after the dietary nickel challenge had ended), 100% had flared based on both patient and investigator assessment [16].

2.2 Improvement of Symptoms with Reduction in Dietary Nickel Intake

Numerous studies have demonstrated that a substantial percentage, usually in the range of 40%, of patients with a positive patch test to nickel and widespread dermatitis or hand dermatitis improve when placed on a low-nickel diet [11, 15, 17,18,19]. If the group put on a low-nickel diet is limited to those who also have a positive oral challenge to nickel, the likelihood of improving on the diet rises to the 60–80% range [12, 20]. Studies showing improvement on oral disulfiram do not truly offer evidence that dietary nickel is the causative factor, as while there is evidence that disulfiram chelates nickel out of the body, it has not been proven that there isn’t some other effect of disulfiram that leads to the improvement.

3 Pathophysiology of Systemic Contact Dermatitis to Nickel

There are two primary aspects of understanding SCDN and SNAS: (1) understanding the normal physiology of nickel in the body – its absorption in the GI tract, transport in the body, and excretion – and (2) understanding the immunologic reaction to ingested nickel in nickel-sensitive individuals who react to dietary nickel, both the specific effects on the immune system and the reason(s) why only some nickel-sensitive patients react.

3.1 Nickel Physiology

Nickel is a component of numerous foods, and it is estimated that the human body contains 10 mg of nickel [21, 22]. Estimates of daily intake of nickel vary from country to country, with estimates of up to 4 mg per day in some Swedish diets (although with an estimated average consumption of 0.75 mg) [23]. In the United States (USA), a reasonable average daily dietary nickel intake is in the range of 0.5 mg.

Some foods are generally high in nickel, but the nickel content of any given food can vary widely based on the nickel content of the soil where the food was produced [22]. In general, whole grains, legumes, and cocoa beans are very high in nickel. Urinary nickel levels increase directly in proportion with ingestion of nickel-rich foods [24]. Between 10 and 40% of the nickel consumed is absorbed, and nickel absorption may be greater in patients over age 30 [23,24,25]. Taking dietary supplements and drinking water that has been stagnant in nickel-containing pipes increases systemic nickel exposure as well [24].

The foods that are co-ingested with nickel also affect absorption, although comprehensive studies of which foods and drinks affect nickel absorption and how they affect it are lacking – for example, milk, orange juice, tea, and coffee reduced nickel absorption compared to ingestion with water, while ingestion with Coca-Cola did not [23]. Ingestion with phytic acid did not affect nickel absorption, ingestion with ascorbic acid (vitamin C) reduced nickel absorption, and ingestion with disodium ethylenediaminetetraacetic acid (EDTA), a common food preservative, led to a dramatic drop in nickel absorption and even a drop in serum nickel levels [23].

When nickel is given orally on an empty stomach, the nickel blood level peaks between 1.5 and 4 h and remains elevated for up to 96 hours [23, 25]. There is a 20% variation from individual to individual in the rise in serum nickel in response to a given dose of oral nickel [23]. With increased nickel ingestion over long periods of time, the proportion of nickel that is absorbed declines [26].

Absorbed nickel is excreted primarily in the urine, with between 50% and 80% of absorbed nickel being excreted in the urine [25]. Urinary nickel rises in proportion to the serum nickel [27]. Renal excretion of nickel increases with increases in nickel ingestion [26].

Nickel is also excreted in the sweat, with nickel concentration in sweat being substantially higher in women than in men, but nickel concentration in sweat does not correlate with nickel concentration in blood or urine [27, 28].

While there is conflicting data, nickel sensitization does not seem to affect nickel absorption in the GI tract nor renal excretion, as when allergic and nonallergic women were challenged with oral nickel, there were no differences between the groups in the resultant increases in serum and urinary nickel [29, 30]. In addition, urinary nickel levels were the same in sensitized and non-sensitized groups [24]. In one study, individuals with positive nickel patch tests had higher serum nickel levels at baseline [31].

Atopic dermatitis does appear to affect nickel metabolism, as serum nickel levels increased more in atopic patients in response to an oral nickel challenge than in non-atopics [32]. Specifically, patients with intrinsic atopic dermatitis (IAD), defined as meeting the Hanifin and Rajka criteria but having a normal IgE level and not being sensitized to dust mite, had a serum nickel concentration that was double that in extrinsic atopic dermatitis (EAD) patients and seven times that of healthy controls [33]. In another study, IAD patients were nearly three times as likely to be nickel sensitized as EAD patients, and the IAD patients had over three times higher nickel concentrations in their sweat compared to EAD patients [34]. These data may be interpreted in two ways: (1) that some proportion of patients diagnosed with IAD really have SCDN or (2) that patients with IAD truly differ from those with EAD in terms of nickel metabolism.

3.2 Immunologic Response to Ingested Nickel

When the cutaneous reactions to ingested nickel are biopsied, the histology is similar to regular allergic contact dermatitis (ACD) to nickel, with infiltrates consisting primarily of CD4+ T-cells and smaller numbers of CD8+ T-cells [35, 36]. In patients who react to an oral nickel challenge, there is a drop in blood monocytes at approximately 4 h after the challenge and then a marked decrease in blood B-cell and T-cell counts at 24 h [30, 37]. More specifically, patients who have a reaction to oral nickel show decreases in CD3 + CLA+ (but not CD3 + CLA-), CD4 + CD45RO+, CD4 + CD45RO-, CD8 + CLA+ (but not CD3 + CLA-), CD19+, and CD5-CD19+ lymphocytes [7, 30, 37].

When cytokine levels are followed, there is a statistically significant increase in IL-5 in the nickel-allergic patients who react to oral nickel compared to those who do not react to oral nickel [7]. IL-6 and IL-10 also increased in these patients but did not reach statistical significance [7]. IL-2, IL-4, TNF-alpha, and IFN-gamma did not increase in the nickel reactors [7].

When duodenal biopsies were performed 2 days after nickel challenge, those who reacted to oral nickel challenge showed dramatic inflammation compared to non-nickel-sensitive patients and nickel-sensitive patients who didn’t react to the oral challenge. The infiltrating cells were mainly CD45RO+ [37].

Finally, when nickel patch test positive patients with urticarial, respiratory, oral, and/or skin symptoms that were thought to be related to dietary nickel were studied with prick testing, 70% of patients had positive prick tests to 10 mg/ml nickel suspension compared to 30% of controls, with the urticarial patients being the most likely to have a positive prick test [17]. When lymphocyte transformation testing was performed with nickel as the stimulating agent, patient lymphocytes showed elevated production of IL-4, IL-5, IL-10, and IFN-gamma compared to controls [17].

In summary, it appears that in patients who react to an oral challenge, circulating lymphocytes are activated in response to nickel in the blood, migrating into the gut mucosa and skin and producing IL-4, IL-5, and IL-10.

4 Treatment of SCDN and SNAS

There are four potential treatment approaches: (1) reduce the amount of nickel ingested, (2) reduce the amount of ingested nickel that is absorbed in the GI tract, (3) remove nickel from the body, and (4) desensitize the immune system. Each approach has been shown to have efficacy, although there are not well done head-to-head trials.

4.1 Reduction of the Amount of Nickel Ingested

As previously noted, around 40% of patients with a positive patch test to nickel and suspected reactions to dietary nickel will improve when placed on a low-nickel diet [11, 15, 17,18,19]. Limiting it to those who respond to an oral nickel challenge increases the likelihood of improving to the 60–80% range [12, 20]. For patients in the United States, the most recently published low-nickel diet based on a point system is recommended (Table 17.1) [22]. Because the mineral content of foods varies based on the geographic area where the food was produced, it is uncertain how accurate this diet would be for those living in other parts of the world.

Table 17.1 Instructions for low-nickel diet (Reproduced with permission from [22])
Full size table

Two fundamental questions are (1) why don’t all nickel-allergic patients react to a nickel challenge and (2) why don’t all patients who react to an oral challenge improve on a low-nickel diet. Evidence does not exist to definitively answer these two questions, but there are hypotheses that can be drawn based on interesting findings. First, in one of the oral challenge studies, the patients who reacted to the oral challenge had a greater rise in urinary nickel than those who didn’t [13]. The most likely explanation was that those who reacted either absorbed more of the nickel from the challenge or ingested additional nickel via their regular diet in the days preceding and following the challenge. In another study, urinary nickel levels dropped more consistently and to a greater degree in those who improved on the diet compared to those who did not, although this did not reach statistical significance (likely due to small sample size) [9]. In a patient who was carefully followed, serum and urine nickel levels dropped by half while on the low-nickel diet, and timing of the improvement in symptoms correlated with the timing of the drop in nickel levels [38]. Overall, the literature suggests that (1) variation in the amount of nickel actually absorbed into the blood, either because of differences in what proportion of nickel is absorbed or because of differences in nickel ingestion related to diet, is the primary determinant of who responds to a nickel challenge; and (2) inability to adequately lower systemic nickel exposure, either due to poor compliance with the diet, living in a region that has high nickel content in food and water due to soil concentration, or having an intestinal system that absorbs a higher proportion of ingested nickel than normal, or some other factor, is the primary reason for failure of patients to improve on a low-nickel diet [12].

4.2 Reduction of the Amount of Ingested Nickel that Is Absorbed

This has not been attempted as a sole therapeutic approach. There are two approaches, though, that have good data to support their efficacy in reducing nickel absorption. The first is the co-administration of vitamin C (ascorbic acid) with meals. Co-administration of 1 g of ascorbic acid with a dose of nickel sulfate hexahydrate substantially reduced the rise in plasma nickel concentration compared to when the nickel was given by itself [23]. Vitamin C is cheap, widely available, and safe, so taking a chewable vitamin C tablet with every meal is a simple recommendation for clinicians to make and for patients to implement.

Disodium cromoglycate (DSCG) also showed significant efficacy in reducing nickel absorption in a study of 24 patients who were nickel allergic and reacted to a challenge with 10 mg NiSO4 (2.09 mg of elemental nickel) [18]. Patients were randomized to either a low-nickel diet group or a regular diet supplemented with DSCG. Both groups improved clinically and had reductions in urinary nickel, but the DSCG group improved significantly more and had a greater reduction in urinary nickel. On intestinal permeability testing, the DSCG group showed a reduction in osmosis through aqueous pores of enterocytes, with this being the presumed cause of the reduction in urinary nickel and clinical improvement.

More interesting, and possibly more effective than either of the above, is the ingestion of calcium disodium EDTA with meals. When 40 mg of iron sodium or disodium EDTA were ingested with a dose of 5 mg of elemental nickel (22.4 mg of nickel sulfate), both not only completely blocked the expected rise in serum nickel but actually led to subsequent decreases in serum nickel, suggesting that they bound nickel so tightly that they prevented its absorption and also actually drew nickel from the blood into the intestines where it could be excreted [23]. Calcium disodium EDTA is available at a very low cost, making this a reasonable treatment alternative. However, it is uncertain if it is possible for deficiencies to develop due to other minerals it binds, such as iron or magnesium, so care should be taken if this treatment is recommended, and monitoring of iron, magnesium, and other minerals should be considered.

4.3 Removal of Nickel from the Body

There have been several reports of the effectiveness of the oral chelating agent disulfiram, both in conjunction with a low-nickel diet and as a standalone treatment. The combination is more effective and has fewer side effects, although disulfiram used alone is also effective in the significant majority of patients [10, 39,40,41]. Unlike calcium disodium EDTA, which is very poorly absorbed from the GI tract (approximately 5% absorbed) and is thought to have its effect by binding nickel in the GI tract and preventing absorption, disulfiram is absorbed into the blood and is thought to bind circulating nickel and remove it from the body.

In one of the earlier reports in the literature, it was demonstrated that in alcoholics being treated chronically with disulfiram to prevent alcohol consumption, their serum, blood, and urinary nickel levels all increased. The median increases compared to pretreatment levels were 17 times in serum, 15 times in whole blood, and 39 times in urine [42]. Similar increases were seen in other studies with disulfiram, with the nickel levels remaining elevated in serum and urine as long as the patient stayed on disulfiram [39, 40, 42]. This supports the concept that the disulfiram is binding nickel in the blood and the disulfiram/nickel complex is then excreted in the urine.

It is important to understand that the assays are measuring the total amount of nickel in the serum and urine, not the amount of free nickel. So, while the increase in serum levels could be concerning, this is unlikely because the nickel in the serum is not immunologically active due to binding to diethyldithiocarbamate (active metabolite of disulfiram) [40]. That being said, a substantial proportion of patients will flare, with symptoms ranging from exacerbation of existing dermatitis to significant systemic symptoms and even leukocytoclastic vasculitis, when starting disulfiram, presumably due to a transient increase in the serum concentration of immunologically active free nickel [39, 40]. Whether disulfiram is started at the full dose or started at a lower dose and titrated upward, the majority of patients will still have a flare, with up to 80% of patients being affected [40, 41]. Initiating a low-nickel diet for several weeks prior to initiation of the disulfiram may reduce the likelihood of a flare reaction and reduce the severity of the flare if it does occur [10].

4.4 Desensitization to Nickel

Desensitization to oral nickel using progressively increasing doses of nickel has been shown to be effective in a number of studies, being effective in up to 85% of patients [43]. There are two proposed mechanisms, each with experimental support: (1) immunologic desensitization and (2) intestinal alterations that result in reduced nickel absorption.

One study used a stepwise increase in oral nickel dosage: 0.627 mg elemental nickel (3 mg NiSO4) from day 1 to 20, 1.254 mg elemental nickel (6 mg NiSO4) from 21 to 40, and 2.090 mg elemental nickel (10 mg NiSO4) from day 41 onward (between 49 and 152 days). Nickel-sensitive patients had a higher serum nickel at baseline than nonsensitive individuals, suggesting they absorbed nickel more efficiently. However, nickel serum levels did not increase over the duration of the study and no patients flared, suggesting that as the amount of nickel ingested increased, the proportion that was absorbed decreased [31].

In another similar study, 22 patients with positive challenges to 2.090 elemental nickel (10 mg NiSO4) were treated with 0.627 mg elemental nickel (3 mg NiSO4)/day × 1 month, 1.254 mg elemental nickel (6 mg NiSO4)/day x 1 month, and 2.090 mg elemental nickel (10 mg NiSO4)/day × 1 month. Three had to stop due to flares in the first month (days 4, 8, and 15) but no others flared, even when taking the same dose daily for a month that had previously caused a flare when taken once during the initial oral challenge. Very importantly, there was no effect on patch test reactions or reactions to earrings after completing the 3 months, suggesting strongly that the mechanism of tolerance is not immunologic hyposensitization but instead reduced absorption [2].

Another randomized study showed the importance of dose and frequency of administration in oral hyposensitization and suggested an immunologic mechanism. Patients were treated with either 0.5 mg per day or 5.0 mg per week of nickel for 6 weeks. A single patient in the small daily dose group flared, and there were no changes in patch test reactivity to nickel after 6 weeks. In the high weekly dose, over half had flares and there was a significant reduction in patch test reactivity to nickel after the 6 week treatment course [44].

Very small doses of nickel, 0.5 ng per day, combined with a low-nickel diet, were shown potentially useful in a study completed by 30 out of 50 initial participants. Fewer than 20% had flares, and those who completed the regimen showed a decrease in reactivity to oral challenge, but less than half showed any change in patch test reactivity [45].

In another very low dose protocol, SNAS patients were randomized to either a low-nickel diet only (12 patients) or to a low-nickel diet combined with desensitization using NiOH starting at 0.3 ng per week and titrating up to 1.5 ug per week, which was then continued for 12 months. Patients in the desensitization group had greater clinical improvement and increased tolerance to nickel-rich food after completing the protocol, as well as reduced in vitro peripheral blood mononuclear cell release of IL-13, IL-5, and IFN-gamma in response to nickel stimulation [46]. In a different trial looking at the minimum doses necessary, patients were randomized to 1.5 ug per week, 0.3 ug per week, and 0.03 ug per week for 1 year. The 1.5 ug per week group had improvements in GI symptoms and reduced sensitivity to oral challenge while the other groups did not show similar changes [19].

A higher dose protocol showed more impressive results from desensitization. Twenty-six patients who completed a 3-month protocol taking 50 ug per day showed a significant decrease in patch test reactivity to nickel, improved clinical symptoms, and reduced in vitro T lymphocyte reactivity to nickel. However, this was an open label study, so placebo effects cannot be excluded as the cause of the changes [47].

In the highest quality study reported to date, 141 patients with a positive nickel patch test, symptoms consistent with SNAS, improvement on a low-nickel diet, and a positive response to an oral nickel challenge were randomized to treatment with different doses of nickel in a double-blind, placebo-controlled manner. Dose levels were 1.5 ug per week, 0.3 ug per week, or 30 ng per week (in all cases the total weekly dose was divided into three weekly doses). A low-nickel diet was implemented 1 month prior to beginning oral nickel supplementation, and high-nickel foods were intentionally reintroduced in a gradual manner after 3 months of nickel supplementation. Those in the highest dose group showed improvements in clinical symptoms overall and reductions in patch test reactivity and responsiveness to oral challenge, but the changes did not become apparent until between 7 and 12 months of oral treatment [19].

5 Summary: Nickel Allergy and Dietary Nickel

A substantial number of patients with nickel allergy will manifest cutaneous and systemic symptoms related to dietary nickel ingestion in a dose-dependent fashion. It is unclear what the distinguishing factor is between patients who do and do not react to dietary nickel – it may be an immunologic difference or a difference in GI absorption of nickel. In patients who react to oral nickel challenges, there is an increase in IL-4, IL-5, and IL-10 released from T lymphocytes. This happens in the gut mucosa for certain but may be happening in multiple other tissue sites as well.

In the ideal setting, the combination of (1) a diffuse, pruritic dermatitis and (2) a positive nickel patch test would trigger additional testing to confirm the diagnosis of SCDN or SNAS, but such testing is not generally feasible at present. Thus, when the diagnosis is suspected based on criteria (1) and (2) above, a therapeutic trial is appropriate.

Therapeutic options include a low-nickel diet alone or in conjunction with one or more adjunctive therapies, including oral disulfiram, oral disodium cromoglycate, oral calcium disodium EDTA, or oral nickel desensitization. There are no comparative trials to help in determining which therapy is most effective or has the best cost-effectiveness ratio. Unfortunately, disodium cromoglycate is generally not available in appropriate oral doses in the United States to allow its use to be considered, nor are the materials for oral nickel desensitization, so these approaches will not be discussed further.

The author’s current approach is to initiate therapy with a low-nickel diet and a 1-month course of calcium disodium EDTA at 600 mg taken three times daily with meals, supplemented with a non-nickel-containing, iron-rich multivitamin taken immediately before bed. Calcium disodium EDTA is available via multiple online sources at a low cost (less than $20 for enough to complete 1 month of therapy). No laboratory monitoring is necessary, and there is essentially no risk of side effects or of a flare of dermatitis. If there is clear clinical benefit, the low-nickel diet is continued and calcium disodium EDTA is repeated if and when symptoms relapse.

Disulfiram therapy is reserved for patients who only respond partially to combined low-nickel diet and calcium disodium EDTA and who relapse quickly when chelation therapy is discontinued. It is typically used at 250 mg/day for 8 weeks while the low-nickel diet is continued. Liver function tests are drawn prior to initiation and are repeated at the 1-month mark. Patients are warned that a flare of cutaneous and/or systemic symptoms is likely and that absolute avoidance of alcohol consumption is necessary. Chelation therapy is continued for 1 month out of every three in an attempt to avoid reaccumulation of nickel and thus to avoid the need for repeated courses of disulfiram.

6 Chromium

Chromium is present in most dietary items, and the amount varies substantially based on the soil content where the food is produced [48]. There is substantially less hard data on chromium content of foods compared to nickel, and there is substantially less experimental evidence linking dietary ingestion of chromium to dermatitis, but clinical experience strongly suggests that at least some patients who are patch test positive to chromium will experience clinical exacerbation of dermatitis, vesicular hand dermatitis in particular, with a high-chromium diet and improvement with avoidance of dietary chromium [49,50,51,52].

The published information on the chromium content of foods is less comprehensive than for nickel. However, the two sources that seem to be most reliable in recent publication include an analysis by Thor et al. that reviewed the published data on chromium content in food and a study by Anderson et al. that directly measured the chromium content of various foods [48, 53].

7 Cobalt

Similar to the situation with chromium, there is much less data for cobalt than for nickel regarding the frequency, mechanism, or management of systemic contact dermatitis from its ingestion in food. Once again, clinical experience and published studies with oral challenges strongly suggest that at least some individuals who are patch test positive to cobalt will have clinical exacerbation of their dermatitis, especially vesicular hand dermatitis, with a high-cobalt diet and improvement with dietary cobalt restriction [51, 54, 55].

Fortunately, unlike chromium and like nickel, there is a recent, practical, useful article by Stuckert and Nedorost with specific instructions for following a low-cobalt diet (Table 17.2) [56]. The instructions presented in this article are quite useful and user-friendly for patients in whom a low-cobalt diet is recommended.

Table 17.2 Instructions for low-cobalt diet (Reproduced with permission from Stuckert et al. [56])
Full size table

8 Conclusion

There is an enormous body of published work and clinical experience supporting the role of dietary nickel, chromium, and cobalt in cases of systemic contact dermatitis, especially vesicular hand eczema. Reliable data exist for the content of these elements in various foods, along with evidence that avoidance of foods high in the relevant metal leads to improvement in many patients. It is incumbent on those managing patients with vesicular hand dermatitis to consider the potential role of dietary metal as a causative factor, patch test these patients, and implement dietary avoidance in those with positive patch tests.