Abstract
Background
Reactive granulomatous dermatitis (RGD) is an umbrella term used to describe interstitial granulomatous dermatitis (IGD), palisaded neutrophilic and granulomatous dermatitis (PNGD), and interstitial granulomatous drug eruption (IGDR).
Objective
The aim of this study was to describe systemic associations of RGD, explore possible associations between histopathologic findings and systemic RGD associations and determine clinical relevance of RGD subtypes.
Methods
We retrospectively studied clinical and histopathologic characteristics of patients with RGD from 1990 through 2020.
Results
Of 65 patients with RGD (41 women, 24 men; median age at diagnosis, 62 years), 37 had IGD, 26 had PNGD, and 2 had IGDR. Fifty patients (76.9%) had an associated systemic condition; rheumatologic conditions were identified for 34 (52.3%) patients. The associated systemic condition occurred before RGD in approximately 75% of patients. Statistical analyses did not show significant associations between specific subtypes of RGD and systemic diseases or treatment response, and specific histopathologic findings were not predictive of an associated systemic disease.
Conclusions
Although most patients with RGD had an associated systemic condition, subtypes of RGD did not correlate with systemic associations, lending support to the use of the umbrella term RGD.
Introduction
Granulomatous inflammatory skin conditions occur with varying degrees and patterns of microscopic inflammation. The umbrella term reactive granulomatous dermatitis (RGD) has been proposed to describe the subtypes interstitial granulomatous dermatitis (IGD), palisaded neutrophilic and granulomatous dermatitis (PNGD) and interstitial granulomatous drug reaction (IGDR).1 These conditions have been studied clinically and histopathologically since the early 1990s, and their systemic associations have been described in detail2,3; however, most reports have been case reports and smaller case series.4–6 Few clinical and microscopic descriptions exist for larger patient populations. Therefore, we conducted a clinicopathologic review to describe the systemic associations of RGD, to explore possible associations between histopathologic characteristics and systemic associations of RGD, and to determine the clinical relevance of RGD subtypes.
Materials and Methods
After obtaining approval from the Mayo Clinic Institutional Review Board, we retrospectively reviewed electronic health records of patients who had a diagnosis of IGD, PNGD or IGDR at Mayo Clinic in Rochester, Minnesota, from 1 January 1990 to 15 September 2020. These patients were followed up through 15 November 2020. Patients were considered to have RGD based on the proposed classification criteria of Rosenbach and English (see Table 6 of the referenced manuscript for specific criteria).1 These criteria stratify RGD into the following subtypes based on clinical and histopathologic characteristics: IGD, PNGD and IGDR. All patients in our study had clinical and histopathologic features of RGD1 and a final diagnosis of RGD (or 1 of its subtypes)2 rendered by the treating dermatologist based upon clinicopathologic correlation.
For this study, skin biopsy slides for all patients were retrospectively reviewed in a blinded fashion by a board-certified dermatopathologist (J.S.L.). Based on the predominant histopathologic inflammatory pattern observed, patients were then classified as having either the IGD or PNGD subtype of RGD. Histopathologic features of each skin biopsy specimen were categorized according to the following dermal parameters: degree of cellularity; inflammatory cell type (i.e. neutrophils, histiocytes, lymphocytes and eosinophils); inflammatory pattern (i.e. interstitial, palisading and perivascular); and presence of mucin, basophilic collagen and vascular damage. These features were integrated holistically by the reviewing dermatopathologist as fitting best into either the IGD or PNGD category. Characteristics favouring the IGD subtype were interstitial histiocytic inflammation with variable degrees of degenerated collagen, whereas characteristics favouring PNGD were the presence of neutrophils, karyorrhexis or leukocytoclastic vasculitis in addition to dermal granulomatous inflammation. Additional differentiating histopathologic features were noted.
Patients who had a probable temporal association of their granulomatous skin eruption with a medication (≤1 year), as deemed by the treating dermatologist at the time of clinical care (without the use of objective drug causality scoring criteria), were classified as having IGDR. Patients with IGDR can potentially have vacuolar interface dermatitis and eosinophilia on histopathology.
We excluded patients who were younger than 18 years, who had an alternate clinical or histopathologic diagnosis (e.g. infection, sarcoidosis, necrobiosis lipoidica, granuloma annulare and cutaneous lymphoma), whose skin biopsy slides were not available for review, and who did not consent to be included in retrospective research studies.
Data collection
We abstracted the following data: age at diagnosis, sex, skin type, date of diagnosis, date of the last follow-up, parts of the body involved, medications used, relevant laboratory data, history of autoimmune diseases or cancer, family history, symptoms, clinical examination findings, treatment and clinical course with response to treatment.
We retrospectively classified clinical outcomes of treatment as follows: (i) complete response (CR), i.e. total resolution of lesions; (ii) partial response (PR), i.e. persistence of some active lesions, with PR graded as >50% or ≤50%; or (iii) no response (NR), i.e. persistence, worsening, or increase in lesions.
Statistical analysis
Continuous data, such as age, were summarized as median (range), and categorical data were summarized as frequency (percentage). We used logistic regression models to explore the association between having (i) a systemic disease association and any subtype of RGD; (ii) an association between the presence of a systemic condition and any particular histopathologic feature; and (iii) an association between treatment response and subtype of RGD. A P value <0.05 was considered significant.
Results
Clinical characteristics
Sixty-five patients met our inclusion criteria: 41 women (63.1%) and 24 men (36.9%). The median (range) patient age at RGD diagnosis was 62 (15–81) years. Of the patients, 37 had the IGD subtype, 26 had the PNGD subtype, and 2 had the IGDR subtype (culprit drugs in the latter group were ganciclovir vs. temozolomide and trimethoprim-sulfamethoxazole). The median patient age was 63 years at IGD diagnosis and 49.5 years at PNGD diagnosis. Most patients (64.9%) did not have symptomatic skin involvement. For patients with skin symptoms, pruritus was most common (31.6%). The most common laboratory abnormality was a positive antinuclear antibody [ANA titre ≥1:80; 18 of 51 tested patients (35.3%)]. Additional clinical characteristics are summarized in Table 1 and depicted in Figs 1 and 2.
Table 1.
Clinical characteristics and laboratory abnormalities of 65 patients with reactive granulomatous dermatitis, classified by clinicopathologic subtype
| No. (%)† |
||||
|---|---|---|---|---|
| RGD | IGD | PNGD | IGDR | |
| Patients | 65/65 (100) | 37/65 (56.9) | 26/65 (40.0) | 2/65 (3.1) |
| Sex | ||||
| Female | 41 (63.1) | 23 (62.2) | 16 (62.0) | 2 (100.0) |
| Male | 24 (36.9) | 14 (37.8) | 10 (38.0) | 0 (0.0) |
| Median (range) age at diagnosis, years | 62 (15–81) | 63 (17–81) | 49 (15–81) | 69 (62–76) |
| Clinical symptoms | n = 57 | n = 34 | n = 21 | n = 2 |
| Asymptomatic | 37 (64.9) | 24 (70.6) | 12 (57.1) | 1 (50.0) |
| Pruritus | 18 (31.6) | 10 (29.4) | 7 (33.3) | 1 (50.0) |
| Tenderness | 3 (5.3) | 2‡ (5.9) | 1 (4.8) | 0 (0.0) |
| Other | 2/57 (3.5) | 0 (0.0) | 1§ (4.8) | 1¶ (50.0) |
| Laboratory abnormalities (No. abnormal/No. tested) | ||||
| ANA | 18/51 (35.3) | 8/29 (27.6) | 9/21 (42.9) | 1/1 (100.0) |
| Anti-dsDNA | 5/36 (13.9) | 3/21 (14.3) | 2/15 (13.3) | NA |
| SS-A | 4/34 (11.8) | 2/20 (10.0) | 2/14 (14.3) | NA |
| SS-B | 0/35 (0.0) | 0/20 (0.0) | 0/15 (0.0) | NA |
| RNP | 3/32 (9.4) | 0/20 (0.0) | 3/12 (25.0) | NA |
| Anticentromere | 2/35 (5.7) | 1/20 (5.0) | 1/15 (6.7) | NA |
| Anti-Jo1 | 1/35 (2.9) | 1/20 (5.0) | 0/15 (0.0) | NA |
| Rheumatoid factor | 4/36 (11.1) | 2/21 (9.5) | 2/15 (13.3) | NA |
| ANCA | 6/36 (16.7) | 2/19 (10.5) | 4/13 (30.8) | NA |
Abbreviations: ANA, antinuclear antibody; ANCA, antineutrophil cytoplasmic antibodies; anti-dsDNA, anti-double-stranded DNA; IGD, interstitial granulomatous dermatitis; IGDR, interstitial granulomatous drug reaction; NA, not applicable; PNGD, palisaded and neutrophilic granulomatous dermatitis; RGD, reactive granulomatous dermatitis; RNP, ribonucleoprotein; SS, Sjögren syndrome.
Data are displayed as No. (%) unless indicated otherwise.
Two of 3 patients who reported tenderness also reported pruritus.
One patient reported ulcerations.
One patient reported swelling.
Figure 1.
Erythematous-to-violaceous, circinate plaques of the anterior thighs in reactive granulomatous dermatitis-palisaded neutrophilic and granulomatous dermatitis subtype.
Figure 2.
Erythematous papules, patches and thin plaques involving the back in reactive granulomatous dermatitis-interstitial granulomatous dermatitis subtype.
Associated systemic diseases
Of all patients with RGD, 50 of 65 (76.9%) had an associated systemic condition; 34 patients had rheumatologic conditions (52.3%). Other associated conditions included cancer [6 patients (9.2%)], gastrointestinal disease [5 patients (7.7%)] and other conditions [5 patients (7.7%)] (Table 2). For patients with available data (n = 50), most associated systemic conditions developed before RGD [39 (78.0%)] (Table 2). A specific systemic condition was not significantly associated with a particular subtype of RGD (P = 0.99 for both IGD and PNGD).
Table 2.
Categories and timing of associated conditions of 65 patients with reactive granulomatous dermatitis, classified by clinicopathologic subtype
| No. (%)† |
||||
|---|---|---|---|---|
| RGD | IGD | PNGD | IGDR | |
| Associated conditions | 50 (76.9) | 27 (73.0) | 21 (80.8) | 2 (100.0) |
| Rheumatologic conditions | 34 (52.3) | 19 (51.4) | 15 (57.7) | 0 (0.0) |
| Rheumatoid arthritis | 7 (10.8) | 5 (13.5) | 2 (7.7) | NA |
| SLE | 5 (7.7) | 2 (5.4) | 3 (11.5) | NA |
| Seronegative inflammatory arthritis | 6 (9.2) | 4 (10.8) | 2 (7.7) | NA |
| Undifferentiated CTD | 5 (7.7) | 0 (0.0) | 5 (19.2) | NA |
| CREST syndrome | 2 (3.1) | 2 (5.4) | 0 (0.0) | NA |
| Other‡ | 9 (13.8) | 6 (16.2) | 3 (11.5) | NA |
| Gastrointestinal conditions | 5 (7.7) | 1 (2.7) | 4 (15.4) | 0 (0.0) |
| Crohn disease | 4 (6.2) | 0 (0.0) | 4 (15.4) | NA |
| Ulcerative colitis | 1 (1.5) | 1 (2.7) | 0 (0.0) | NA |
| Cancer§ | 6 (9.2) | 3 (8.1) | 1 (3.8) | 2 (100.0) |
| Other¶ | 5 (7.7) | 4 (10.8) | 1 (3.8) | NA |
| Timing of associated conditions | n = 50 | n = 27 | n = 21 | n = 2 |
| Before | 39 (78.0) | 20 (74.0) | 17 (81.0) | 2 (100.0) |
| Synchronous†† | 2 (4.0) | 2 (7.4) | 0 (0.0) | NA |
| After‡‡ | 9 (18.0) | 5 (18.5) | 4 (19.0) | NA |
Abbreviations: AML, acute myeloblastic leukaemia; CTD, connective tissue disease; IGD, interstitial granulomatous dermatitis; IGDR, interstitial granulomatous drug reaction; MAC, mycobacterium avium complex; MALT, mucosa-associated lymphoid tissue; NA, not applicable; PNGD, palisaded and neutrophilic granulomatous dermatitis; RGD, reactive granulomatous dermatitis; SLE, systemic lupus erythematosus.
Data are displayed as No. (%) unless indicated otherwise.
Morphea, giant cell arteritis, dermatomyositis, juvenile rheumatoid arthritis, gout and polymyositis were reported for a single patient with a diagnosis of IGD. Polymyalgia rheumatica, psoriasis, and Behҫet disease were reported for a single patient with a diagnosis of PNGD.
Malignancy reported included 2 patients with acute myelocytic leukaemia and 1 with multiple myeloma with IGD, 1 patient with Hodgkin lymphoma and PNGD, and 1 patient each with glioblastoma multiforme and MALT lymphoma and IGDR.
Hypothyroidism was reported for 2 patients with IGD and 1 patient with PNGD. Graves’ disease and MAC infection were reported in single patients with IGD.
One patient with IGD synchronously developed SLE, and 1 patient with IGD synchronously developed a MAC infection.
Rheumatologic conditions that developed after RGD included RA (2 patients, 1 with IGD and 1 with PNGD), undifferentiated CTD and PNGD (1 patient), seronegative inflammatory arthritis (2 patients with IGD and 1 patient with PNGD), and SLE and PNGD (1 patient). Two patients with IGD had a malignancy after RGD (AML). These conditions developed a median of 12 months (range: 3 weeks–10 years) after the diagnosis of RGD for the 9 patients.
Histopathologic findings
We classified subtypes of RGD on the basis of histopathologic criteria and assessed skin biopsies of each patient for specific microscopic characteristics, as described in Methods. An associated systemic condition and the following histopathologic findings were not significantly associated: inflammatory cell type; degree of dermal cellularity; inflammatory pattern; and presence of mucin, necrobiosis, basophilic collagen or vascular damage (P > 0.90 for rheumatologic condition, gastrointestinal disease and malignancy, respectively). A summary of specific histopathologic features present for each RGD subtype is presented in Table 3 and depicted in Figs 3 and 4.
Table 3.
Histopathologic features of 65 patients with reactive granulomatous dermatitis, classified by clinicopathologic subtype
| IGD† (n = 39) | PNGD (n = 26) | Total (N = 65) | |
|---|---|---|---|
| Dermal inflammation (location): Neutrophils, No. (%) | |||
| Interstitial | 2 (5.1) | 6 (23.1) | 8 (12.3) |
| Perivascular | 2 (5.1) | 2 (7.7) | 4 (6.2) |
| Associated with granulomas | 0 (0.0) | 6 (23.1) | 6 (9.2) |
| Interstitial and perivascular | 0 (0.0) | 3 (11.5) | 3 (4.6) |
| Interstitial and with granulomas | 0 (0.0) | 9 (34.6) | 9 (13.8) |
| Dermal inflammation (density): Neutrophils, No. (%) | |||
| Absent/mild | 38 (97.4) | 10 (38.5) | 48 (73.8) |
| Moderate/abundant | 1 (2.6) | 16 (61.5) | 17 (26.2) |
| Dermal inflammation (location): Histiocytes, No. (%) | |||
| Interstitial | 6 (15.4) | 3 (11.5) | 9 (13.8) |
| Perivascular | 3 (7.7) | 1 (3.8) | 4 (6.2) |
| Palisaded | 3 (7.7) | 0 (0.0) | 3 (4.6) |
| Interstitial, perivascular | 7 (17.9) | 4 (15.4) | 11 (16.9) |
| Interstitial, palisaded | 16 (41.0) | 16 (61.5) | 32 (49.2) |
| Perivascular, palisaded | 1 (2.6) | 1 (3.8) | 2 (3.1) |
| Interstitial, perivascular, palisaded | 2 (5.1) | 1 (3.8) | 3 (4.6) |
| Dermal inflammation (density): Histiocytes, No. (%) | |||
| Absent/mild | 8 (20.5) | 2 (7.7) | 10 (15.4) |
| Moderate/abundant | 31 (79.5) | 24 (92.3) | 55 (84.6) |
| Dermal inflammation (location): Lymphocytes, No. (%) | |||
| Interstitial | 2 (5.1) | 0 (0.0) | 2 (3.1) |
| Perivascular | 27 (69.2) | 18 (69.2) | 45 (69.2) |
| With granuloma | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| Interstitial, perivascular | 0 (0.0) | 3 (11.5) | 3 (4.6) |
| Interstitial, with granuloma | 1 (2.6) | 0 (0.0) | 1 (1.5) |
| Perivascular, with granuloma | 6 (15.4) | 1 (3.8) | 7 (10.8) |
| Dermal inflammation (density): Lymphocytes, No. (%) | |||
| Absent/mild | 22 (56.4) | 15 (57.7) | 37 (56.9) |
| Moderate/abundant | 17 (43.6) | 10 (38.5) | 27 (41.5) |
| Dermal inflammation (location): Eosinophils, No. (%) | |||
| Interstitial | 2 (5.1) | 2 (7.7) | 4 (6.2) |
| Perivascular | 2 (5.1) | 3 (11.5) | 5 (7.7) |
| With granuloma | 1 (2.6) | 2 (7.7) | 3 (4.6) |
| Interstitial, with granuloma | 1 (2.6) | 0 (0.0) | 1 (1.5) |
| Interstitial, perivascular | 0 (0.0) | 1 (3.8) | 1 (1.5) |
| Dermal inflammation (density): Eosinophils, No. (%) | |||
| Absent/mild | 36 (92.3) | 24 (92.3) | 60 (92.3) |
| Moderate/abundant | 3 (7.7) | 2 (7.7) | 5 (7.7) |
| Dermal necrobiosis (location), No. (%) | |||
| Interstitial | 10 (25.6) | 4 (15.4) | 14 (21.5) |
| Perivascular | 0 (0.0) | 1 (3.8) | 1 (1.5) |
| With granuloma | 15 (38.5) | 11 (42.3) | 26 (40.0) |
| Interstitial, perivascular, with granulomas | 0 (0.0) | 1 (3.8) | 1 (1.5) |
| Interstitial, with granulomas | 9 (23.1) | 9 (34.6) | 18 (27.7) |
| Dermal necrobiosis, No. (%) | |||
| Present | 35 (89.7) | 26 (100.0) | 61 (93.8) |
| Absent | 4 (10.3) | 0 (0.0) | 4 (6.2) |
| Dermal mucin (location), No. (%) | |||
| Interstitial | 4 (10.3) | 5 (19.2) | 9 (13.8) |
| Perivascular | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| With granuloma | 10 (25.6) | 2 (7.7) | 12 (18.5) |
| Interstitial, with granulomas | 6 (15.4) | 5 (19.2) | 11 (16.9) |
| Dermal mucin, No. (%) | |||
| Absent | 19 (48.7) | 14 (53.8) | 33 (50.8) |
| Present | 20 (51.3) | 12 (46.2) | 32 (49.2) |
| Basophilic collagen, No. (%) | |||
| Absent | 39 (100.0) | 20 (76.9) | 59 (90.8) |
| Present | 0 (0.0) | 6 (23.1) | 6 (9.2) |
| Vascular damage, No. (%) | |||
| Absent | 39 (100.0) | 21 (80.8) | 60 (92.3) |
| Present | 0 (0.0) | 5 (19.2) | 5 (7.7) |
Abbreviations: IGD, interstitial granulomatous dermatitis; IGDR, interstitial granulomatous drug reaction; PNGD, palisaded and neutrophilic granulomatous dermatitis.
Two cases of IGDR had similar histopathologic findings as cases of IGD (n = 37) and were thus included in this subgroup (total of 39 patients in this subgroup). Additionally, both cases of IGDR had focal vacuolar changes and focal floating collagen bundles.
Figure 3.
Palisading granulomas with foci of neutrophils in the mid-dermis in reactive granulomatous dermatitis-palisaded neutrophilic and granulomatous dermatitis subtype (haematoxylin and eosin; original magnification × 300).
Figure 4.
Interstitial lymphohistiocytic infiltrate in the upper and mid-dermis in reactive granulomatous dermatitis-interstitial granulomatous dermatitis subtype (haematoxylin and eosin; original magnification × 200).
Treatment
Topical steroids were used to treat RGD for 32 patients (49.2%), and systemic corticosteroids were used for 13 patients (20.0%). Additional treatments are described in Table 4.
Table 4.
Treatment of 65 patients with reactive granulomatous dermatitis, classified by clinicopathologic subtype
| No. (%)† |
||||
|---|---|---|---|---|
| RGD | IGD | PNGD | IGDR | |
| Treatment‡ | ||||
| Topical steroids | 32 (49.2) | 17 (45.9) | 14 (53.8) | 1 (50.0) |
| Systemic steroids | 13 (20.0) | 3 (8.1) | 9 (34.6) | 1 (50.0) |
| Hydroxychloroquine | 5 (7.7) | 3 (8.1) | 2 (7.7) | 0 (0.0) |
| Tetracycline antibiotics | 4 (6.2) | 1 (2.7) | 3 (11.5) | 0 (0.0) |
| Observation | 4 (6.2) | 4 (10.8) | 0 (0.0) | 0 (0.0) |
| Intralesional steroids | 3 (4.6) | 1 (2.7) | 2 (7.7) | 0 (0.0) |
| Topical tacrolimus | 3 (4.6) | 2 (5.4) | 1 (3.8) | 0 (0.0) |
| Methotrexate | 3 (4.6) | 1 (2.7) | 2 (7.7) | 0 (0.0) |
| Other§ | 4 (6.2) | 1 (2.7) | 3 (11.5) | 0 (0.0) |
| Unknown | 10 (15.4) | 8 (21.6) | 2 (7.7) | 0 (0.0) |
| Treatment response | ||||
| CR | 18 | 10 | 8 | NA |
| PR >50 | 10 | 5 | 5 | NA |
| PR <50 | 4 | 2 | 2 | NA |
| No data | 33 | 20 | 11 | 2 |
Abbreviations: CR, complete response; IGD, interstitial granulomatous dermatitis; IGDR, interstitial granulomatous drug reaction; NA, not applicable; PNGD, palisaded and neutrophilic granulomatous dermatitis; PR, partial response; RGD, reactive granulomatous dermatitis.
Data are displayed as No. (%) unless indicated otherwise.
Multiple modalities were used for some patients.
Etanercept, clindamycin, loratadine and azathioprine were each used once.
At the last follow-up, 18 patients with RGD (27.7%) had a CR, and 10 patients (15.4%) had a PR >50. Treatment response was not available for 33 patients (50.8%). Among patients with available data for treatment response, 15 of 17 (88.2%) with IGD had a CR or a PR >50, and 13 of 15 (86.7%) patients with PNGD had a CR or a PR >50.
Discussion
We described clinical and histopathologic characteristics of RGD in a large case series of 65 patients and correlated these findings with systemic associations to determine whether subclassification of RGD as IGD or PNGD subtypes could offer additional prognostic information. The most common subtype of RGD was IGD, a finding similar to that in a recently published study.7 Our patients were predominantly women; this result is similar to results of other studies, which have shown that RGD was at least three times as common for women.3,4,8–10 Our patients were of all age groups, and patients with the PNGD subtype were younger than those with the IGD subtype, which has also been described by others.7,11
Associated systemic diseases
More than three-fourths of our patients (76.9%) had an associated condition, which was most commonly rheumatologic (52.3%). Of those tested, 35.3% had a positive ANA. Prior studies showed a positivity rate for ANA as high as 59.6%.7 Associated systemic disease developed before RGD in 78.0% of patients, which may have implications for determining which patients should be more thoroughly evaluated for an underlying systemic disease at the time of RGD diagnosis. We did not find a significant association between a patient’s having an associated systemic disease and a specific subtype of RGD. The most common associations described in the literature for IGD are inflammatory arthritis, connective tissue diseases and cancer,3,10,12 whereas the most commonly described associations with PNGD include connective tissue diseases (particularly systemic lupus erythematosus), inflammatory arthritis, hematologic conditions and, rarely, infections.5,13,14 Although an association with RGD and antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis has been described,1 there were no cases of ANCA-associated vasculitis in our cohort. Our data support evaluating patients with the IGD or PNGD subtype for underlying rheumatologic diseases, including inflammatory arthritis and autoimmune connective tissue disease.
Given the small number of patients with inflammatory bowel disease or hematologic malignancy associated with RGD in our retrospective study, we cannot make definitive screening recommendations for these conditions. For patients with IGDR, the most commonly implicated drugs are calcium channel blockers, β-blockers, angiotensin-converting enzyme inhibitors, statins, tumour necrosis factor (TNF)-β inhibitors and furosemide.1,15–18 The initial descriptions of IGDR note long latency times of ≥1 year after medication exposure in some patients.15 Although only two of our 65 patients had a drug association with RGD, we recommend thoroughly reviewing the medication list of patients with skin findings suggestive of RGD.
Histopathologic findings
We performed a detailed histopathologic assessment of skin biopsies for all 65 patients with RGD in our study, and we did not find any significant associations between a patient’s having an associated systemic condition and a specific histopathologic finding. Thus, our data do not support predicting whether or not a patient is more likely to have an associated systemic condition based on microscopic features of RGD. Because of the association of both IGD and PNGD subtypes of RGD with underlying systemic disease, we recommend that dermatopathologists consider using the term RGD within their dermatopathology reports (in addition to providing a descriptive readout, which may include information regarding the predominant type of inflammatory cell observed), in order to alert the treating dermatologist to perform a review of systems and targeted workup to assess for the presence of an associated systemic disorder.
Treatment
For about 20% of patients with PNGD, skin findings have been reported to resolve, even as early as 1 week.4 Spontaneous remission has also been described for IGD.3,19 A prior study reported that lesions completely resolved for 52 (76%) patients.7 After drugs are eliminated as a trigger, most patients are treated with topical corticosteroids, although the benefit is controversial.1,3,20 Treatment of the underlying trigger often helps control PNGD and IGD.1 Other treatment options described for RGD include systemic corticosteroids, non-steroidal anti-inflammatory drugs, hydroxychloroquine, colchicine, methotrexate, cyclosporine, cyclophosphamide, TNF-α inhibitors, intralesional corticosteroids and dapsone.1,3,6,21,22 Because our treatment response data were incomplete, we could not perform analyses to assess for significant associations between a particular subtype of RGD and response to treatment. However, the percentages of patients with available data were similar for patients with a CR or a PR >50 in the IGD (88.2%) and PNGD (86.7%) subtypes.
How does our study add to the existing literature about RGD?
Our study findings are similar to those of a recently published experience7 and add further clinical and histopathologic data to support the use of the term RGD in clinical practice. Table 5 summarizes pertinent findings of our study and compares them to findings of other selected RGD case series.
Table 5.
Comparison of the current study of reactive granulomatous dermatitis with selected previous case series
| Study authors; study type | Demographic characteristics | Associated diseases | Histopathologic and other key findings |
|---|---|---|---|
| Kumar et al., 2022 (present study); single-centre retrospective study | 65 patients (63% women); median age, 62 years; IGD, 37; PNGD, 26; IGDR, 2 | 50/65 patients (76.9%) had an associated systemic condition (e.g. rheumatologic, 34; cancer, 6; gastrointestinal, 5); associated systemic condition occurred before RGD in 78% of patients; specific systemic conditions were not significantly associated with a particular subtype of RGD (P = 0.99) | Specific histopathologic findings were not significantly associated with the presence of an associated systemic condition (P > 0.90) |
| Rodriguez-Garijo et al., 20217; multicentre retrospective study | 52 patients (73% women); median age, 56.5 years; IGD, 46; PNGD, 6 | 39/52 patients (75%) had an associated systemic condition [autoimmune, 28 patients (53.8%), including 12 with SLE; 6 (11.5%), cancer; and 4 (7.7%) infection] | Karyorrhexis and neutrophils were observed in patients with a limited course (resolving within a few weeks); patients with a deeper and predominantly lymphocytic inflammatory infiltrate had a limited recurrent course (cutaneous flare-ups resolving within 1–2 years) |
| Peroni et al., 20123; single-centre retrospective study | 12 patients with IGD (75% women); mean age, 58.5 years | 6 patients had an associated rheumatologic condition (rheumatoid arthritis, spondyloarthritis, arthralgias); 3 had cancer | No vasculitis or significant mucin deposition was found in skin biopsies of IGD; overall clinical disease duration was months to years |
| Akinshemoyin Vaughn et al., 202011; multicentre retrospective study | 7 paediatric patients (5 female); mean age, 12.7 years; ethnic background Hispanic, 5 patients; Asian, 2 | All 7 patients had associated SLE (3 patients, prior diagnosis; 2, concomitant; and 2, subsequent) | Neutrophils or karyorrhexic debris observed in 4 of 7 skin biopsies, and eosinophils were present in 2; 3 patients had questionable drug exposures (etanercept, enalapril and valproic acid); systemic treatment of SLE improved RGD in all 7 patients |
Abbreviations: IGD, interstitial granulomatous dermatitis; IGDR, interstitial granulomatous drug reaction; PNGD, palisaded neutrophilic and granulomatous dermatitis; RGD, reactive granulomatous dermatitis; SLE, systemic lupus erythematosus.
What are the novel findings of our study?
As described in Table 5, very limited case series of RGD exist to guide dermatologists in improving patient outcomes. Similar to a recent multicentre review describing 52 patients with RGD from 3 dermatology departments in Spain,7 our study sought to explore the clinical relevance at an academic practice of the term RGD as coined by Rosenbach and English in 2015.1 Our study is unique given the size of the patient cohort (65 patients) and given that patients came from a single-centre academic dermatology practice (in comparison with the analysis7 noted above). In our study, we used a novel methodology of statistical analyses to assess whether specific histopathologic features of RGD could predict the presence of an associated systemic disorder: the lack of an association in our study suggests that an underlying systemic condition should be considered by clinicians whenever a patient has received a final diagnosis of RGD through clinicopathologic correlation, regardless of the specific histopathologic features that are present in an individual patient. Another novel finding of our study is the detailed description of the timing of RGD (before, synchronous, after) in relation to the development of associated systemic conditions (Tables 2 and 5), which may have clinical monitoring implications for the treating dermatologist.
Limitations
This was a retrospective study. In addition, we could not compare clinical morphologic patterns of RGD for our patient cohort with previously published results in the literature because details were inconsistent in physical examination documentation, and clinical photographs were often unavailable. Our study was not designed for optimal assessment of treatment responses, and therefore, we could not derive general treatment guidelines for RGD from our data. We also were unable to uniformly determine the level of disease activity of the associated systemic diseases at the time RGD developed.
Conclusion
In this study of 65 patients with RGD, no significant associations were shown between specific RGD subtypes and associated systemic diseases and treatment response. The study findings also did not support using specific histopathologic findings to predict the presence of an associated systemic disease. Thus, our data suggest that the term RGD is relevant in clinical practice and may be more helpful to clinicians caring for patients with suspected RGD (particularly in guiding the initial evaluation for associated diseases) than previously described terms such as IGD, PNGD and IGDR. Future RGD studies should aim to assess whether RGD is a good or poor prognostic factor for underlying associated disease, whether RGD and the underlying disease follow a parallel clinical course, whether there are any clinical morphologic differences between IGD and PNGD, and whether certain skin-directed treatments (such as dapsone) have differing effectiveness depending upon the presence of specific histopathologic features (such as a predominance of neutrophils, histiocytes or eosinophils).
Acknowledgements
The authors would like to thank Austin Todd, MS and Paul Novotny, MS, for assistance with the statistical analyses performed in the study. Marianne Mallia, ELS, MWC, senior scientific/medical editor, Mayo Clinic, substantively edited the manuscript. The Scientific Publications staff at Mayo Clinic provided proofreading, administrative and clerical support. This publication was supported by Grant Number UL1 TR002377 from the National Center for Advancing Translational Sciences (NCATS). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. The patients in this manuscript have given written informed consent to the publication of their case details.
Funding sources
This article has no funding source.
Footnotes
Conflict of interest
The authors have no conflict of interest to declare.
This study was approved by the Mayo Clinic Institutional Review Board.
Data availability statement
Will be provided if requested by senior author
References
- 1.Rosenbach M, English JC 3rd. Reactive granulomatous dermatitis: a review of palisaded neutrophilic and granulomatous dermatitis, interstitial granulomatous dermatitis, interstitial granulomatous drug reaction, and a proposed reclassification. Dermatol Clin 2015; 33: 373–387. [DOI] [PubMed] [Google Scholar]
- 2.Wilmoth GJ, Perniciaro C. Cutaneous extravascular necrotizing granuloma (Winkelmann granuloma): confirmation of the association with systemic disease. J Am Acad Dermatol 1996; 34(5 Pt 1): 753–759. [DOI] [PubMed] [Google Scholar]
- 3.Peroni A, Colato C, Schena D, Gisondi P, Girolomoni G. Interstitial granulomatous dermatitis: a distinct entity with characteristic histological and clinical pattern. Br J Dermatol 2012; 166: 775–783. [DOI] [PubMed] [Google Scholar]
- 4.Hantash BM, Chiang D, Kohler S, Fiorentino D. Palisaded neutrophilic and granulomatous dermatitis associated with limited systemic sclerosis. J Am Acad Dermatol 2008; 58: 661–664. [DOI] [PubMed] [Google Scholar]
- 5.Germanas JP, Mehrabi D, Carder KR. Palisaded neutrophilic granulomatous dermatitis in a 12-year-old girl with systemic lupus erythematosus. J Am Acad Dermatol 2006; 55(2 Suppl): S60–S62. [DOI] [PubMed] [Google Scholar]
- 6.Busquets-Perez N, Narvaez J, Valverde-Garcia J. Interstitial granulomatous dermatitis with arthritis (Ackerman syndrome). J Rheumatol 2006; 33: 1207–1209. [PubMed] [Google Scholar]
- 7.Rodriguez-Garijo N, Bielsa I, Mascaro JM Jr et al. Reactive granulomatous dermatitis as a histological pattern including manifestations of interstitial granulomatous dermatitis and palisaded neutrophilic and granulomatous dermatitis: a study of 52 patients. J Eur Acad Dermatol Venereol 2021; 35: 988–994. [DOI] [PubMed] [Google Scholar]
- 8.Moon HR, Lee JH, Won CH et al. A child with interstitial granulomatous dermatitis and juvenile idiopathic arthritis. Pediatr Dermatol 2013; 30: e272–e273. [DOI] [PubMed] [Google Scholar]
- 9.Lehman JS, Sokumbi O, Peters MS et al. Histopathologic features of non-infectious granulomatous disorders involving the skin. Hum Pathol 2020; 103: 127–145. [DOI] [PubMed] [Google Scholar]
- 10.Dubey S, Merry P. Clinical vignette. Interstitial granulomatous dermatitis (Ackerman’s syndrome) in SLE presenting with the ‘rope sign. Rheumatology (Oxford) 2007; 46: 80. [DOI] [PubMed] [Google Scholar]
- 11.Akinshemoyin Vaughn OL, Siegel DH, Chiu YE et al. Clinical and histologic presentation of pediatric reactive granulomatous dermatitis. Pediatr Dermatol 2020; 37: 498–503. [DOI] [PubMed] [Google Scholar]
- 12.Felcht M, Faulhaber J, Gottmann U, Koenen W, Goerdt S, Goebeler M. Interstitial granulomatous dermatitis (Ackerman’s syndrome). Eur J Dermatol 2010; 20: 661–662. [DOI] [PubMed] [Google Scholar]
- 13.Finan MC, Winkelmann RK. The cutaneous extravascular necrotizing granuloma (Churg-Strauss granuloma) and systemic disease: a review of 27 cases. Medicine (Baltimore) 1983; 62: 142–158. [DOI] [PubMed] [Google Scholar]
- 14.Singh M, Comfere N. Images in clinical medicine. Palisaded neutrophilic and granulomatous dermatitis. N Engl J Med 2012; 366: e33. [DOI] [PubMed] [Google Scholar]
- 15.Magro CM, Crowson AN, Schapiro BL. The interstitial granulomatous drug reaction: a distinctive clinical and pathological entity. J Cutan Pathol 1998; 25: 72–78. [DOI] [PubMed] [Google Scholar]
- 16.Siami K, Wilkerson M, Clark SH, Crowson AN. Pathologic quiz case: an indurated plaque on the ankle of a 74-year-old woman. Interstitial granulomatous drug reaction. Arch Pathol Lab Med 2004; 128: e129–e130. [DOI] [PubMed] [Google Scholar]
- 17.Deng A, Harvey V, Sina B et al. Interstitial granulomatous dermatitis associated with the use of tumor necrosis factor alpha inhibitors. Arch Dermatol 2006; 142: 198–202. [DOI] [PubMed] [Google Scholar]
- 18.Perrin C, Lacour JP, Castanet J, Michiels JF. Interstitial granulomatous drug reaction with a histological pattern of interstitial granulomatous dermatitis. Am J Dermatopathol 2001; 23: 295–298. [DOI] [PubMed] [Google Scholar]
- 19.Aloi F, Tomasini C, Pippione M. Interstitial granulomatous dermatitis with plaques. Am J Dermatopathol 1999; 21: 320–323. [DOI] [PubMed] [Google Scholar]
- 20.Warycha MA, Fangman W, Kamino H, Schaffer JV. Interstitial granulomatous dermatitis in a child with chronic uveitis. J Am Acad Dermatol 2008; 58(5 Suppl 1): S100–S102. [DOI] [PubMed] [Google Scholar]
- 21.He Y, Maverakis E, Ramirez-Maverakis D, Fitzmaurice S. Combination therapy with intralesional triamcinolone and oral dapsone for management of palisaded neutrophilic and granulomatous dermatitis. Dermatol Online J 2013; 19: 17. [PubMed] [Google Scholar]
- 22.Pastar Z, Rados J, Pavic I et al. Palisaded neutrophilic and granulomatous dermatitis in association with subcutaneous nodular and systemic sarcoidosis. Acta Dermatovenerol Croat 2013; 21: 245–249. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Will be provided if requested by senior author