Granuloma annulare (GA) is a noninfectious granulomatous skin condition that can present with a variety of cutaneous morphologies. It is characterized by collagen degeneration, mucin deposition, and palisaded or interstitial histiocytes. Although the mechanism underlying development of GA is unknown, studies point to a cell-mediated hypersensitivity reaction to an as-yet undetermined antigen. Systemic associations with diabetes, thyroid disorders, lipid abnormalities, malignancy, and infection are described in atypical GA. Treatment is divided into localized skin-directed therapies and systemic immunomodulatory or immunosuppressive therapies. The selected treatment modality should be based on disease severity, comorbid conditions, consideration of potential side effects, and patient preference.
Key points
- •
Granuloma annulare (GA) presents with multiple morphologies, including localized, generalized, macular/patch, subcutaneous, perforating, and atypical variants (palmar, mucosal, and photosensitive).
- •
GA is a cell-mediated hypersensitivity reaction to an unknown antigen resulting in elevated levels of interleukin (IL)-12, interferon (IFN)-γ, and tumor necrosis factor α (TNF-α).
- •
Many recent studies have failed to find an association between GA and diabetes mellitus (DM).
- •
Treatment can broadly be divided into localized/skin-directed therapy and systemic immunomodulating therapy.
- •
Selection of modality should take into consideration disease severity, comorbid conditions, potential side effects, and patient preference.
Introduction
First described by Colcott and Fox in 1895 and later formally named in 1902 by Radcliffe-Crocker, GA is classified as a noninfectious granulomatous skin condition. The condition is characterized by multiple morphologies, including localized, generalized, subcutaneous, macular or patch, and various atypical morphologies. Histologically, GA shows collagen degradation, mucin deposition, and either a palisaded or interstitial histiocytic infiltrate. Although the exact cause of the condition remains unknown, a cell-mediated hypersensitivity reaction is favored as the mechanism underlying the development of GA. There are multiple reports of systemic disease associations with GA, including DM, malignancy, thyroid dysfunction, lipid abnormalities, and infection. Many of these associations are seen with the atypical clinical variants and are discussed in this article. In regard to therapy, a skin-directed or systemic approach can be taken. There are no large randomized controlled trials, however, evaluating the therapies used to treat GA. Consideration for disease severity, comorbid conditions, side effects, and patient preference is imperative.
Introduction
First described by Colcott and Fox in 1895 and later formally named in 1902 by Radcliffe-Crocker, GA is classified as a noninfectious granulomatous skin condition. The condition is characterized by multiple morphologies, including localized, generalized, subcutaneous, macular or patch, and various atypical morphologies. Histologically, GA shows collagen degradation, mucin deposition, and either a palisaded or interstitial histiocytic infiltrate. Although the exact cause of the condition remains unknown, a cell-mediated hypersensitivity reaction is favored as the mechanism underlying the development of GA. There are multiple reports of systemic disease associations with GA, including DM, malignancy, thyroid dysfunction, lipid abnormalities, and infection. Many of these associations are seen with the atypical clinical variants and are discussed in this article. In regard to therapy, a skin-directed or systemic approach can be taken. There are no large randomized controlled trials, however, evaluating the therapies used to treat GA. Consideration for disease severity, comorbid conditions, side effects, and patient preference is imperative.
Etiopathogenesis
The histologic findings of GA are well known, with focal areas of collagen degeneration surrounded by an inflammatory infiltrate composed of palisaded histiocytes and lymphocytes. The etiology and pathogenesis, however, remain poorly understood. There have been reports of GA appearing after localized subcutaneous trauma for desensitization, bacille Calmette-Guérin vaccination, tetanus and diphtheria vaccination, hepatitis B vaccination, ultraviolet light exposure, mesotherapy, and drug exposure. Studies have pointed to an immunologic mechanism with resultant necrobiosis underlying the development of GA. Various studies have suggested a role of immunoglobulin-mediated vasculitis, lysozyme production, cell-mediated hypersensitivity, and and elastic fibers as targets in GA.
Dahl and colleagues examined the vasculature of patients with GA. Immunoglobulin M (IgM) and C3 were found in the vessel walls and IgM, C3, and fibrinogen were all observed along the dermal epidermal junction. Additionally, they found vessel wall necrosis, fibrinoid change, and thickening or occlusion in many specimens. These results suggested that an immunoglobulin-mediated vasculitis may be involved in the pathogenesis of GA. In a study by Umbert and Winkelmann, however, these changes were thought more in response to tissue and vessel changes related to the mononuclear infiltrate characteristic of GA.
Beuchner and colleagues identified the presence of T-cell subsets in lesional skin. Using monoclonal antibodies targeting peripheral T cells, the predominant cell population was identified as activated helper T cells of the T h 1 class. These T-cell populations were found not only within the granulomas but also surrounding the vasculature. Furthermore, the histiocytes in this study demonstrated diffuse activity. The results suggested a cell-mediated response to an unidentified antigen as the dominant pathogenic event in GA.
With the thought that GA represents a delayed-type hypersensitivity reaction resulting in matrix degradation, Fayyazi and colleagues sought to identify the cytokine profile. Specifically, the roles of IFN-γ, matrix metalloproteinases, and TNF-α were investigated. In their study, large numbers of lymphocytes expressed IFN-γ, which subsequently activated macrophages. These activated macrophages in turn produced IL-12, the major cytokine driving differentiation of naïve T cells to the T h 1 subset. Furthermore, a vast majority of both the lymphocytes and the macrophages contained TNF-α. In addition, immunohistochemistry revealed that the macrophages producing the TNF-α additionally coexpress matrix metalloproteinases-2 and -9. These results taken together suggest that in GA, IFN-γ secreting T h 1 cells cause a delayed type of hypersensitivity reaction whereby macrophages are differentiated to aggressive effector cells. The end result of this process is the degradation of the connective tissue matrix characteristic of GA. A recent study by Mempel and colleagues confirmed the presence of T h 1 cells in the granulomas and found high levels of IL-2 within these granulomas, ultimately recruiting more T cells to the sites of activity. Taken together, these 3 studies illustrate that GA is a cell-mediated response to an unknown antigen.
Lastly, the role of genetic predisposition and GA has been examined. Generalized GA was found in association with HLA Bw35. More recently, there has been a case report of generalized GA occurring in monozygotic twins carrying the HLA AH8.1. Further studies are warranted to determine the role of genetics in the development of GA.
Clinical presentations
The presentation of GA is variable. Localized GA is the classic variant described in the literature. In this form, flesh-colored to erythematous papules in annular configurations are appreciated ( Fig. 1 ). The lesions are commonly found on the dorsal hands or feet. The localized form is the most common presentation of GA, is found in women more often than in men, and tends to appear in patients under the age of 30. Generalized annular GA, by definition, is the presence of 10 or more skin lesions or by widespread annular plaques. This form is notable for a later age of onset, a more chronic and relapsing course, rare spontaneous resolution, and, in general, a poorer response to therapy than its localized counterpart.
Subcutaneous GA is a rare presentation that is seen predominantly in children and young adults ( Fig. 2 ). Lesions appear as firm, asymptomatic, subcutaneous nodules with a tendency to appear on the anterior lower legs, hands, head, and buttocks. In 25% of cases, the subcutaneous lesions appear in association with intradermal lesions. These lesions are often biopsied to rule out clinical mimickers, such as rheumatoid nodule, necrobiosis, and epithelioid sarcoma, and sarcoidosis. Macular or patch GA is a rare presentation. In this form, asymptomatic erythematous to brown patches favoring the proximal extremities in women are seen ( Fig. 3 ). The differential diagnosis of macular GA includes morphea, parapsoriasis, and cutaneous T-cell lymphoma (CTCL); however, biopsy is diagnostic and in most cases is notable for an interstitial pattern of histiocytes.
Atypical variants of GA have also been reported. Perforating GA is a rare presentation of disease and represents an atypical variant. This form is notable for a central umbilication with a keratotic core. Within this umbilication, histologically, elimination of degenerated collagen is seen. These lesions may be pruritic or painful, and, unlike their localized, macular, and generalized counterparts, scar on healing. Additional atypical forms include photosensitive GA ( Fig. 4 ), palmar GA, disseminated popular, and oral mucosal GA. Generalized GA, disseminated GA, and other atypical GA, as discussed previously, are important to identify because they have been associated with systemic disease.
Systemic associations
The relationship between DM and GA has been controversial. Early studies have suggested a link between the 2 conditions. More recent studies, however, have called this association into question. A retrospective study of patients conducted by Studer and colleagues did not find a statistically significant association between GA and DM. Patients with DM and GA tended, however, to have more chronic and relapsing disease than did their nondiabetic counterparts. Gannon and Lynch found an absence of carbohydrate intolerance in patients with GA. Most recently, Nabesio and colleagues conducted a case-control study between patients with GA and matched controls with psoriasis. This study was unable to find a statistically significant relationship between GA and type 2 DM.
GA, both generalized and localized, has been reported to occur in the setting of thyroid disease. Case reports associating localized GA with autoimmune thyroiditis are found in the literature. A small, case-controlled study conducted by Vazquez-Lopez and colleagues sought to establish the frequency and type of thyroid disease in adult women with localized GA. In their study, 24 women with localized GA and 100 age-matched female patients with non-GA cutaneous disease were evaluated for underlying thyroid disorders. In the patients with localized GA, 12% demonstrated autoimmune thyroiditis whereas only 1% of the controls were noted to have autoimmune thyroiditis. These statistically significant results ( P = .022) suggest that a subset of female patients with localized GA may have autoimmune thyroid disease. Therefore, in female patients with localized GA, screening for thyroid disease with thyroid-stimulating hormone and antibodies may be considered.
GA has also been reported to occur in the setting of malignancy. Hematologic malignancies are the most frequently reported. Both Hodgkin and non-Hodgkin lymphoma and leukemias have been reported in patients with GA. Solid tumors have also been reported, however, in association with GA. These include cancers of the lung, breast, cervix, colon, prostate, testicles, and thyroid. Li and colleagues conducted a review of the literature examining the association between GA and malignancy. Their review of 14 case reports and 2 correlation studies found that most of the patients with both GA and malignancy were older (mean age of 54 y) and had atypical clinical presentations of their cutaneous disease and that more than half of the malignancies reported were lymphomas. There was no definitive relationship between GA and malignancy. Although no causative relationship has been established between GA and malignancy, older patients, patients with atypical clinical presentations, and patients with recalcitrant disease should have routine age-appropriate cancer screening for both solid organ and hematologic malignancies.
There have been multiple reports of GA occurring in a setting of HIV infection. Presenting morphologies are variable, but often the presentation is atypical. These presentations include photosensitive, molluscum contagiousum–like, disseminated papular, generalized, oral mucosal, and perforating. Toro and colleagues conducted a study that illustrated that the most common morphology seen in HIV infection was that of generalized GA and that although GA can occur at all stages of infection, it was slightly more common in patients with AIDS. HIV-infected patients can develop a multitude of mucocutaneous conditions. These have been categorized into 3 groups. Group 1 includes conditions that are nearly always associated with HIV infections in appropriate clinical scenarios. Group 2 conditions are those conditions that occur with increased prevalence in HIV-infected individuals or whose detection in a previously seronegative patient should prompt consideration for testing. Group 3 mucocutaneous conditions occur in patients with HIV but that are not related to the viral infection. GA has been categorized as a group 2 mucocutaneous condition. As a result, in patients with atypical or generalized presentations of GA, HIV screening should be considered, in particular in a setting of appropriate risk factors. In addition, there have been multiple reports of GA resolving with treatment with highly active antiretroviral therapy.
Infections with hepatitis B and C have also been reported to occur in patients with GA. A recent study examining 10 biopsy specimens found no molecular or culture-based evidence of bacterial, mycobacterial, or fungal infection. With these studies and reports in mind, assessing for risk factors for hepatitis B, hepatitis C, and HIV should be performed in patients with atypical presentations. Follow-up testing with serologies should be performed in the appropriate settings.
The association between lipid abnormalities and GA has been proposed. A study conducted by Dabski and Winkelmann evaluating 100 patients with generalized GA found hypercholesterolemia in 19.6% and hypertriglyceridemia in 23.3% of patients whose levels were tested. More recently, Wu and colleagues conducted a case-control study examining the prevalence of dyslipidemia in patients with GA. The study included 140 patients with GA and 420 matched controls. In patients with GA, dyslipidemia was found in 79.3% of patients compared with 51.9% of the matched controls ( P = .001). In addition, there were statistically significant differences between the quantitative values for total cholesterol, triglycerides, and low-density lipoprotein cholesterol between the GA group and the control group. Generalized GA had a higher prevalence of dyslipidemia and the annular morphology was associated with dyslipidemia, notably hypercholesterolemia. They found that the odds of finding dyslipidemia among patients with GA were approximately 4 times (odds ratio 4.04; 95% CI, 2.53–6.46) than those seen in matched controls without GA. In a comment to that study, it was recommended that adults and children over age 11 with GA be screened for lipid abnormalities.
In addition to these systemic associations, cases of drug-induced GA have recently been reported in the literature. Allopurinol, paroxetine, thalidomide, pegylated IFN-α, topiramate, TNF-α inhibitors (infliximab, adalimumab, and etanercept), and intranasal calcitonin have all been reported to induce lesions clinically and histologically consistent with GA. As a result, for patients with new-onset GA, a detailed drug history should be performed. A trial off of the medication, if possible, can be performed to assess for resolution of cutaneous disease.
In general, a majority of patients with GA are healthy. Dahl conducted a retrospective chart review of 32 patients with GA diagnosed between 1950 and 1970 with a mean follow-up of 35 years and found that most patients had common conditions, such as hypertension, hyperlipidemia, degenerative joint disease, and atherosclerosis. Five patients were noted to have thyroid disorders (not specified). In summary, although a vast majority of patients with GA are healthy, screening for these systemic conditions should be considered in patients with atypical lesions or generalized lesions and in some patients with localized GA.
Evaluation and management
Patients with classic features of localized or generalized annular GA are often diagnosed based on clinical findings alone. But in patients with clinically atypical lesions, or clinical lesions that include other nongranulomatous conditions in the differential diagnosis, biopsy should be performed to confirm the diagnosis. Regardless of clinical presentation, histologically, there is focal degeneration of collagen surrounded by an inflammatory infiltrate composed of lymphocytes and histiocytes. Although this is the classic description of GA, there are 3 patterns of histiocytes that have been described. The most common variant is that of an infiltrative or interstitial pattern. In this form, scattered histiocytes are seen between and surrounding collagen bundles in the upper and the mid-dermis. In a study by Umbert and Winkelmann, this form was found in 71% of reviewed cases. The second pattern is that of the classic palisading granulomas, found in 26% of cases. These palisaded granulomas are scattered within the dermis and show the classic palisaded histiocyties, central degeneration of the connective tissue, and abundant mucin characteristic of GA. The final pattern described in GA is the epithelioid nodule. This was rare and seen in only 3% of studied cases. In the epithelioid form, the histiocytes are aggregated with a nodular appearance with some giant cells. This can be difficult to differentiate from other granulomatous conditions, in particular sarcoidosis; however, the interstitial pattern can often be seen surrounding the epithelioid granuloma, aiding in the diagnosis of GA.
The presence of mucin aids in the diagnosis of GA. The finding of prominent mucin is helpful in the differentiation of GA from other granulomatous conditions, such as necrobiosis lipoidica, sarcoidosis, rheumatoid nodule, and cutaneous Crohn disease. Recently, clonality in sarcoidosis, GA, and granulomatous mycosis fungoides (CTCL) was evaluated. Granulomatous findings in primary cutaneous lymphomas are rare, but their presence histologically can be confused with GA. The presence of a polyclonal T-cell population of cells points to the diagnosis of GA, whereas a monoclonal population of T cells raises the concern for a primary cutaneous lymphoma.
Once a diagnosis of GA has been established, the next challenge facing the clinician and patient is the selection of a treatment modality. Broadly, treatment can be categorized into local, skin-directed therapy and systemic immunomodulatory or immunosuppressive therapy. The treatment selected is often based on severity of disease as well as patient preference.
For localized disease, topical corticosteroids are generally considered first-line therapy. Depending on site, high potency corticosteroids with or without occlusion can be used. Intralesional corticosteroids can be used for localized lesions not responding to topical corticosteroids. Patients must be counseled on risks of corticosteroid use (both topical and intralesional), in particular the risk of atrophy, dyschromia, and striae. Topical tacrolimus has also been reported in the successful treatment of perforating, periorbital, and disseminated GA. These results may be related to the recent report of glioma-associated oncogene ( gli-1 ) highly expressed in granulomatous skin disorders, including GA, and may offer future therapeutic options for patients with disseminated and recalcitrant disease.
Various destructive measures have been reported as effective in the treatment of GA. The role of biopsy in resolution of lesions remains controversial. Levin and colleagues reported a case of rapid resolution of biopsied lesions on 2 separate occasions in the same patient with an interstitial pattern seen histologically. Although the mechanism is unclear, it is thought that the induction of wound healing changes the inflammatory milieu with the replacement, rather than the destruction, of extracellular matrix proteins. Contact cryosurgery has also been reported to clear the lesions of GA. A prospective trial of 31 patients with localized GA found resolution in all patients treated with cryosurgery. The patients were treated with one freeze-thaw cycle ranging in duration from 10 to 60 seconds. Blistering occurred in all patients and the therapy was well tolerated. Intralesional therapy with low-dose recombinant human IFN-γ was successfully used to treat 3 patients with GA. In all 3 patients, treated lesions resolved and remained clear at 12 months post-therapy.
Within the past several years, the role of laser therapy in treatment of individual lesions in both localized and generalized GA has been examined; 15 treatments with excimer laser at 300 mJ/cm 2 with 5 doses per treatment session resulted in resolution of GA lesions on the hands with no recurrence at 6-month follow-up. Pulsed dye laser (PDL) therapy has also been examined as a modality to treat isolated lesions in GA. Sliger and colleagues reported near complete clearance using a 595-nm PDL with maintenance of results at 36 weeks. Sniezek and colleagues performed 3 treatments over the course of 13 months with a 585-nm PDL and reported decreased erythema, thickness, and diameter of treated lesions. These results were maintained at 3 years of follow-up. Most recently, fractionated thermolysis has been reported to be helpful in patients with GA. Improvement in lesions treated with the 1550-nm erbium-doped YAG (Er:YAG) laser as well as the 1440-nm Nd:YAG laser have been reported.
Although localized GA or localized lesions of generalized GA may be amenable to these destructive and topical methods, treatment of widespread GA often requires systemic therapy. There are no large studies examining these treatment modalities. Many of the treatments reported include case reports or small case series. These range from ultraviolet light therapy and immunomodulating antiinflammatory drugs to immunosuppressive medications. The modality selected should focus on disease severity, potential side effects, comorbidities, and patient preference.
There have been various reports of skin-directed ultraviolet therapy resulting in improvement and clearance of lesions of GA. Inui and colleagues reported a case of disseminated GA improving after 24 treatments with narrow-band UV-B (NB-UVB) therapy with sustained response 6 months after cessation of therapy. Improvement in 5 patients treated with 16 weeks of thrice-weekly therapy with broadband UV-B was demonstrated by Do and colleagues. The use of various forms of UV-A has also been efficacious. Small studies using UV-A1 have shown improvement in lesions; however, relapses were common after discontinuation of therapy. Psoralen–UV-A (PUVA) has been effective in clearing the lesions of GA in case reports and small case series. As in the case of UV-A1 therapy, although many patients showed improvement, recurrence was common and maintenance therapy was required. One challenge with any form of phototherapy is the need for frequent and consistent office visits. PUVA carries the additional risk of cutaneous malignancy.
Successful treatment of GA lesions with photodynamic therapy (PDT) has been reported. Piaserico and colleagues reported 3 patients successfully treated with methyl aminolevulinate (MAL) and red light PDT. In another study, 7 patients were treated with aminolevulinic acid (ALA) and a red light source with marked improvement in 2 patients and complete clearance in 2 patients. These patients were treated with 2 to 3 sessions at 2- to 4-week intervals. Clazavara-Pinton and colleagues conducted a retrospective analysis of patients treated with MAL and red light PDT. In the 13 patients with GA, there was a marked to moderate response in 9 of the treated patients.
Oral antibiotics, in particular those in the tetracycline family, have been used in the treatment of generalized GA. The tetracyclines are broad-spectrum antibiotics that are additionally antiinflammatory. Duarte reported a patient with generalized GA who cleared with 10 weeks of doxycycline (100 mg daily). This patient remained clear at 1 year of follow-up. Treatment with monthly rifampicin, ofloxacin, and minocycline has been reported effective in the treatment of GA. Marcus and colleagues treated 6 patients with recalcitrant biopsy-proved GA with 3 months of monthly rifampicin (600 mg), ofloxacin (400 mg), and minocycline hydrochloride (100 mg) combination therapy. Complete clearance of lesions was noted in all patients 3 to 5 months after therapy. Side effects were mild and some patients developed postinflammatory hyperpigmentation as the lesions resolved. Dapsone therapy has also been reported to result in clearance of generalized and localized GA lesions. Reports of dapsone (100–200 mg/d) have shown clearance of lesions within 6 weeks of initiating therapy.
Treatment with fumaric acid esters (FAEs) has been reported effective in the treatment of widespread disease. Weber and colleagues reported a series patients treated with low-dose FAEs for 1 to 18 months. One patient showed complete clearance and 4 were reported to have marked improvement. One patient failed to show any improvement. Kreuter and colleagues reported a female patient with a 25-year history of recalcitrant disseminated disease successfully treated with a combination of low- and high-dose fumaric acids. Therapy was discontinued after 3 months and she remained clear at 6 months. Acharya treated 2 patients with longstanding disseminated GA with FAEs. One patient had clearance after 6 months and was maintained on low-dose FAEs, whereas a second patient had suboptimal clearance of lesions and discontinued therapy after 12 months. Eberlein-Konig and colleagues treated 8 patients with FAEs and noted remission in 3, partial remission in 4, and no change in 1 patient. Breuer and colleagues treated 32 patients with various noninfectious granulomas with FAEs. Of the 13 patients with GA, 8 were noted to have improved. The main side effects associated with FAEs include gastrointestinal, flushing, reversible elevation of transaminases, lymphocytopenia, and eosinophilia.
Antimalarials are often used in the treatment of generalized GA. Chloroquine (250 mg daily) was successfully used in a patient with photodistributed GA. Hydroxychloroquine has been used in the treatment of generalized GA. Cannistraci and colleagues treated 9 patients with generalized GA with 4 months of hydroxychloroquine (9 mg/kg/d for 2 months; 6 mg/kg/d for month 3; and 2 mg/kg/d for month 4). In this series of patients, complete remission was seen in all patients within 6 to 7 weeks of initiation of therapy. The 1 child in the study was treated with half of the dosing scheme.
The systemic retinoids isotretinoin and etretinate have been reported useful in the treatment of generalized GA. Sahin and colleagues reported a case of generalized GA in a diabetic patient successfully treated with isotretinoin (50 mg/d). Their patient had 90% clinical improvement in their lesions. Adams and Hogan also reported a case of longstanding generalized GA showing 90% improvement with oral isotretinoin therapy (40 to 80 mg/d). Additional reports and small case series have shown improvement in generalized GA with the treatment of oral isotretinoin at doses of 0.5 to 1 mg/kg/d.
In addition to these treatment modalities, there are reports of additional nonimmunosuppressive systemic treatments of generalized GA. These have included calcitriol (0.025 μg daily), niacinamide (1500 mg/d), pentoxifylline (400 mg 3 times daily), potassium iodide, allopurinol (300 mg twice daily), and zileuton (2400 mg daily) combined with vitamin E (400 IU daily).
There are a few case reports and small case series evaluating the role for systemic immunosuppression in the treatment of GA. Cyclosporine was used to treat 4 patients with disseminated GA at a starting dose of 4 mg/kg/d for 4 weeks with gradual taper. Resolution was noted within 3 weeks and no relapse was noted either during taper or in the 12 weeks after discontinuation of cyclosporine. There is an additional case report of cyclosporine (5 mg/kg/d) clearing GA in a patient with leukemia. Methotrexate was successfully used to treat 1 patient with disseminated GA. Methotrexate (15 mg weekly) and daily folic acid supplementation resulted in clearance of a majority of lesions in a female patient. With cessation of therapy, lesions recurred and again resolved with reinstitution of therapy. Administration of the alkylating agent chlorambucil in low doses has been reported to result in improvement of GA lesions. For any systemic therapy, consideration of both short- and long-term side effects is necessary and close monitoring is required.
Lastly, the role of biologic therapy, specifically anti–TNF-α, has been investigated as a treatment option for patients with recalcitrant generalized or disseminated disease. As discussed previously, TNF-α is important in the formation and maintenance of granulomas and as such may offer a target for therapy. In particular, adalimumab and infliximab have been reported in the treatment of generalized GA. Knoell reported a rapid response of disseminated GA in identical twins with the AH8.1 haplotype. Werchau and colleagues reported a case of generalized GA that resolved after treatment with adalimumab (80 mg initially followed by 40 mg every 2 weeks). Torres and colleagues treated a patient with disseminated GA with adalimumab (80 mg at week 0 followed by 40 mg [starting at week 1] every other week). This patient had complete clearance by week 8. At 6 months, due to sustained response without recurrence, adalimumab was discontinued and at 9 months the patient remained clear. Infliximab has also been used to treat generalized and disseminated GA. Hertl and colleagues reported a case of a patient treated with infliximab (5 mg/kg) at weeks 0, 2, and 6, followed by monthly infusions for 4 months. Improvement was noted within 4 to 6 weeks and no new lesions had appeared at follow-up through 16 months. Murdaca and colleagues treated a patient with disseminated GA with infliximab (5 mg/kg) at weeks 0, 2, and 6 and monthly for 10 months. They also noted improvement within 8 weeks with gradual resolution of lesions over the course of treatment. Most recently, Amy de la Breteque and colleagues reported treatment of recalcitrant GA with infliximab. The dosing in this patient was similar with infliximab (5 mg/kg) administered at 0, 2, and 6 weeks and then every 8 weeks thereafter. After 8 treatments, the lesions had cleared and therapy was discontinued. The patient had mild recurrence of lesions 18 months later and was treated with topical corticosteroids. The results with etanercept are mixed. Although Shupack and Siu reported resolution of generalized GA in 1 patient after 12 weeks of etanercept weekly therapy, Kreuter and colleagues found no improvement in 4 patients with GA treated with etanercept. These recent reports suggest that TNF-α inhibition with adalimumab and infliximab are viable options for patients with severe, recalcitrant disease, or comorbidities contraindicating other therapies previously listed. Caution must be observed, however, because cases of GA induced by TNF-α inhibitors have been reported.
Summary
GA is a common noninfectious granulomatous condition that can present with a variety of morphologies. Histologically, it is characterized by central collagen degradation with prominent mucin deposition and a peripheral histiocytic infiltrate in either a palisaded or interstitial pattern. GA has been associated with several systemic conditions. Recent studies have called into question the association with DM. In patients with chronic, widespread, or relapsing disease, however, screening with hemoglobin A 1C is recommended. In female patients with localized disease in particular, screening for thyroid disorders is recommended. The recent association between generalized GA and hyperlipidemia should prompt a lipid panel. Lastly, in patients with acute onset of GA who are over the age of 60 years or in patients with atypical presentations of GA, age-appropriate malignancy work-up, hematologic evaluation, and consideration of screening for HIV, hepatitis B, and hepatitis C is recommended ( Box 1 ).