Advances in Anatomic Pathology:
Systemic Lupus Erythematosus–associated Neutrophilic Dermatosis: A Review and Update
Larson, Allison R. MD; Granter, Scott R. MD
Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston MA
The authors have no funding or conflicts of interest to disclose.
Reprints: Scott R. Granter, MD, Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, 75 Francis St. J-820, Jimmy Fund Building, Boston, MA 02115 (e-mail: email@example.com).
Neutrophilic dermatoses are a rare manifestation of systemic lupus erythematosus (SLE). In recent years, a growing body of literature describes a pathologic spectrum of neutrophilic infiltrates that may be seen in lupus patients. It is particularly important to recognize that neutrophilic dermatoses can be the initial manifestation of SLE in a third of patients. We were able to identify 47 patients with SLE associated with neutrophilic tissue reactions. In this review, we describe the histologic and clinical features of these cases in the hope that increased awareness of this unusual manifestation of SLE will generate prompt diagnosis and improved patient care.
Neutrophilic infiltrates of skin in the setting of lupus patients with bullous systemic lupus erythematosus (SLE) and leukocytoclastic vasculitis are well documented.1 However, most dermatopathology texts do not consider cutaneous involvement by SLE in the histologic differential diagnosis of neutrophilic inflammatory dermatoses in nonbullous or nonvasculitic lesions. Nevertheless, there is a growing body of literature documenting cell-rich Sweet-like neutrophilic reactions as well as a small number of patients with skin lesions characterized by paucicellular neutrophilic dermal infiltrates. We and others have encountered some patients with moderately cellular neutrophilic infiltrates in the setting of SLE that bridge the gap between these extremes, providing evidence that there is a spectrum of neutrophilic dermatoses that may be seen in SLE patients.
We identified 47 lupus patients with neutrophilic dermatoses in the literature (Table 1).2–16 Forty-one patients (87%) were female, ranging in age from newborn infants to 67 years old (mean age, 36.96 y). Six patients (13%) were male. A neutrophilic dermatosis was a presenting symptom in 15 patients (32%). In most patients the lesions were violaceous, pink, or erythematous in color, and were most commonly described as papules and plaques. The extremities were involved in 83% of patients, trunk in 60%, and the face or head and neck region in 23% of patients. In addition to patients with SLE, patients with hydralazine-induced lupus and neonatal lupus may also develop or present with neutrophilic dermatoses.17–20 We were able to identify 4 patients with hydralazine-induced lupus who developed Sweet-like tissue reactions as a presenting symptom (Table 2). The lesions, like those associated with SLE, were described as papules or plaques and involved the face or head and neck area in all 4 patients, the extremities in 2 patients, and the trunk in 1 patient. Of note, 3 of the 4 patients were male and only 1 was female.
Three cases of neonatal lupus were identified and included 2 females and 1 male (Table 3).21,22 The lesions were a presenting clinical feature in all 3 patients. In keeping with SLE-associated neutrophilic tissue reactions, the lesions were described as erythematous and macules, papules, or plaques.
The lesions show a spectrum range from paucicellular neutrophilic infiltrates in 12 patients (26%) often limited to the papillary dermis, to cell-rich Sweet-like neutrophilic tissue reactions in 24 (52%) patients. Ten patients (22%) fell somewhere in the middle of this spectrum with moderately cellular infiltrates (Table 1 and Fig. 1).
In patients for whom detailed pathologic descriptions were available, interface changes were seen in 22 patients and dermal mucin was seen in 7. Basement membrane thickening was seen in only 1 patient. Follicular plugging was documented in 1 patient. One case in our series showed a minute discrete focus resembling neutrophilic and granulomatous dermatitis in a background of an otherwise typical neutrophilic dermatosis.
Immunofluorescence testing was performed in 14 patients, and immunoreactants at the dermal epidermal junction were seen in the biopsies from 7 patients (50%).
In 1972, Fryha et al23 reported the case of a 37-year-old woman with glomerular renal disease, butterfly rash, and a cutaneous neutrophilic infiltrate consistent with Sweet syndrome. Because the patient did not have antinuclear antibodies at the time, she did not meet strict criteria for a diagnosis of SLE at that time. In addition, the authors also argue that the biopsy showing Sweet syndrome did not support a diagnosis of SLE—in fact, they posit the histologic findings argued against it. Of course, if the patient did indeed have SLE, the authors would have no way of knowing of an association with neutrophilic tissue reactions as this would be the first case documented. We believe in retrospect that this patient likely represents the first documented case of SLE associated with a Sweet-like neutrophilic dermatosis. Although we strongly suspect that the patient had SLE, we have not included her in our meta-analysis of the literature as she did not meet strict criteria for SLE at the time of biopsy. Goette2 is generally credited with documenting the first case of lupus associated with Sweet syndrome. In 2009, Hou et al4 reported the first case of Sweet syndrome as an initial presentation of SLE. The patient, a 38-year-old woman, presented with joint symptoms, fever, and a rash typical of Sweet syndrome. The constellation of findings hardly points to SLE for either the physician or the pathologist if they are unaware of this association. Patients presenting in this way may easily be suspected clinically of having an infection. In our review of the literature, approximately one third of patients did not have a prior history of SLE at presentation and the presentation of the patient described by Goette is quite typical.
We (S.R.G.) first learned of the association between SLE and neutrophilic dermatitis in textbooks by Ackerman.24,25 In his books, Ackerman describes the presence of “neutrophils and, sometimes, nuclear dust of neutrophils sprinkled immediately beneath the epidermis along the dermoepidermal junction.” Ackerman’s description emphasized the paucicellular end of the histologic spectrum. In our first report, we documented cases similar to Ackerman’s paucicellular example and more moderately cellular infiltrates that still seem to fall short of the exuberant neutrophilic infiltrates that characterize Sweet syndrome.6 Over the last several years, we encountered 13 additional patients.16 Our experience and the literature suggests the moderately to highly cellular neutrophilic tissue reactions showing Sweet-like features are more common than paucicellular lesions.
We identified 4 cases of hydralazine-induced SLE associated with neutrophilic dermal tissue reactions.17–20 As expected, the demographics of this small sample are different from the idiopathic cases described above. Three of the 4 patients were men and the average age was 56 years (range, 44 to 68 y), likely reflecting characteristics of the population that receives hydralazine for hypertension. In all 4 patients, histologic analysis showed a cell-rich Sweet-like pattern of inflammation. The relationship between hydralazine and development of lupus and concomitant Sweet syndrome raises an interesting question. Is the Sweet syndrome a hydralazine-induced drug eruption or it is a manifestation of lupus? A review of the literature fails to identify cases of hydralazine-associated Sweet-like dermatosis in the absence of evident features of SLE. Although sparse, these facts seem to suggest the presence of lupus is required for the development of hydralazine-associated Sweet-like cutaneous lesions.
Another distinctive subset of lupus patients that may develop neutrophilic dermal tissue reactions are neonates. To date, we have identified 3 cases.21,22 Lesions were present at birth in 2 patients, and 1 infant developed lesions at 4 months of age. These cases also show overlapping histologic features from paucicellular lesions21 to frank a Sweet-like tissue reaction.22
The histologic differential diagnosis of the lesions described in SLE patients is broad and to some extent depends on the cellularity of the inflammatory infiltrate. The paucicellular examples are identical to bullous SLE and can be distinguished only by the clinical absence of blisters (Fig. 1A). The histologic features can also be indistinguishable from dermatitis herpetiformis and linear IgA disease (Fig. 1B). Distinction should be made easily on clinical grounds and with immunofluorescence testing when necessary. Rarely, interface dermatitis may be present, which helps with the diagnosis (Fig. 1E). The presence of a mixed pattern of inflammation that includes lymphocytes and eosinophils helps distinguish the paucicellular end of the spectrum from drug reactions, such as a fixed-drug eruption. Still’s disease can show an eruption with features indistinguishable from the paucicellular or moderately cellular variant of SLE-associated neutrophilic dermatoses and clinical correlation is needed to make the correct diagnosis. Similarly, the eruption of rheumatic fever—erythema marginatum rheumaticum—has been described to have similar features and it likely requires clinical correlation to make this diagnosis.
The differential diagnosis of the cellular or Sweet-like end of the spectrum (Figs. 1C, D) includes Behçet disease. The presence of suppurative folliculitis or vasculitis, which may be present in some cases, and clinical features should allow for distinction. The presence of ulcers and pathergy should separate pyoderma gangrenosum from Sweet-like cutaneous SLE.26 Rheumatoid arthritis may uncommonly be associated with a Sweet-like eruption that can be indistinguishable from the lesions we have encountered in the setting of SLE.27 In fact, some might simply regard rheumatoid neutrophilic dermatitis as Sweet syndrome in the setting of rheumatoid arthritis. It is should be pointed out that Sweet-like tissue reactions have been documented in the setting of other connective tissue diseases, such as dermatomyositis.28,29 The designation SLE-associated neutrophilic dermatosis should only be applied to patients who fit criteria for a diagnosis of SLE. Although we would only consider palisaded neutrophilic and granulomatous dermatitis in the broadest of differential diagnoses, as the presence of a granuloma-annulare like histiocytic component with necrobiosis should allow for easy distinction, we have encountered a single case of SLE-associated neutrophilic dermatosis associated with a small focus that showed typical features of palisaded neutrophilic and granulomatous dermatitis.16 Finally, infection should be considered in the histologic differential diagnosis of neutrophilic tissue reactions and stains for bacteria and fungus should be used when appropriate. The pathologist should have a low threshold for recommending culture if concern for infection exists.
It would be convenient if all SLE-associated neutrophilic dermatoses had histologic clues to the underlying disease, such as interface change, dermal mucin, and basement membrane thickening. Unfortunately, this is often not the case. To make accurate diagnosis even more challenging, in our review of the literature we could identify histologic features suggestive of background SLE in one of the 15 patients who had undiagnosed SLE at the time of presentation—precisely the patients who present a diagnostic challenge. Background changes of SLE were much more common in patients with established SLE.
In summary, a spectrum of SLE-associated neutrophilic dermatoses exist and are important to recognize as this is a presenting manifestation of SLE in a third of patients. We anticipate the clinicopathologic spectrum of neutrophilic dermatoses in the setting of SLE to expand as more pathologists are made aware of this association. We would not be surprised if additional cases of neutrophilic dermatoses associated with other types of connective tissue disease are reported in the future. We hope that increased awareness of the association of neutrophilic dermatoses and SLE, particularly in young women, assists in timely diagnosis and improved patient care.
1. Yell JA, Mbuagbaw J, Burge SM .Cutaneous manifestations of systemic lupus erythematosus.Br J Dermatol. 1996; 135:355–362.
2. Goette DK .Sweet’s syndrome in subacute cutaneous lupus erythematosus.Arch Dermatol. 1985; 121:789–791.
3. Choi JW, Chung KY .Sweet’s syndrome with systemic lupus erythematosus and herpes zoster.Br J Dermatol. 1999; 140:1174–1175.
4. Hou TY, Chang DM, Gao HW, et al .Sweet’s syndrome as an initial presentation in systemic lupus erythematosus: a case report and review of the literature.Lupus. 2005; 14:399–402.
5. Bunham JM, Cron RQ .Sweet syndrome as an initial presentation in a child with lupus erythematosus.Lupus. 2005; 14:974–975.
6. Gleason BC, Zembowicz A, Granter SR .Non-bullous neutrophilic dermatosis: an uncommon dermatologic manifestation in patients with lupus erythematosus.J Cutan Pathol. 2006; 33:721–725.
7. Brinster NK, Nunley J, Pariser R, et al .Nonbullous neutrophilic lupus erythematosus: a newly recognized variant of cutaneous lupus erythematosus.J Am Acad Dermatol. 2012; 66:92–97.
8. Kieffer C, Cribier B, Lipsker D .Neutrophilic urticarial dermatosis; a variant of neutrophilic urticaria strongly associated with systemic disease. Report of 9 new cases and review of the literature.Medicine. 2009; 88:23–31.
9. Camarillo D, McCalmont TH, Frieden IJ, et al .Two pediatric cases of nonbullous histiocytoid neutrophilic dermatitis presenting as a cutaneous manifestation of lupus erythematosus.Arch Dermatol. 2008; 144:1495–1498.
10. Fernandes NF, Castello-Soccio L, Kim EJ, et al .Sweet syndrome associated with new-onset systemic lupus erythematosus in a 25-year-old man.Arch Dermatol. 2009; 145:608–609.
11. Tabache F, El Kartouti A, Abdelilah T, et al .Systemic lupus erythematosus revealed by sweet syndrome.Joint Bone Spine. 2011; 78:420–421.
12. Pavlidakey P, Mills O, Bradley S, et al .Neutrophilic dermatosis revisited: an initial presentation of lupus? J Am Acad Dermatol. 2012; 67:e29–e35.
13. Barton JL, Pincus L, Yasdany J, et al .Association of Sweet’s syndrome and systemic lupus erythematosus.Case Rep Rheumatol. 2011; 2011:242681
14. Saeb-Lima M, Charli-Joseph Y, Rodriquez-Acosta ED, et al .Autoimmunity-related neutrophilic dermatosis: a newly described entity that is not exclusive of systemic erythematosus.Am J Dermatopathol. 2013; 35:655–660.
15. Tsuji H, Youshifuji H, Nakashima R, et al .Sweet’s syndrome associated with systemic lupus erythematosus: a case report and review of the literature.J Dermatol. 2013; 40:641–648.
16. Larson AR, Granter SR .Systemic lupus erythematosus-associated neutrophilic dermatosis—an under-recognized neutrophilic dermatosis in patients with systemic lupus erythematosus.Hum Pathol. 2014; 45:598–605.
17. Sequeira W, Polisky RB, Alrenga DP .Neutrophilic dermatosis (Sweet’s syndrome). Association with a hydralazine-induced lupus syndrome.Am J Med. 1986; 81:558–560.
18. Servitje O, Ribera M, Juanola X, et al .Acute neutrophilic dermatosis associated with hydralazine-induced lupus.Arch Dermatol. 1987; 123:1435–1436.
19. Ramsey-Goldman R, Franz T, Solano FX, et al .Hydralazine induced lupus and Sweet’s syndrome. Report and review of the literature.J Rheumatol. 1990; 17:682–684.
20. Cartee TV, Chen SC .Sweet syndrome associated with hydralazine-induced lupus erythematosus.Cutis. 2012; 89:121–124.
21. Barr KL, O’Connell F, Wesson S, et al .Nonbullous neutrophilic dermatosis: Sweet’s syndrome, neonatal lupus erythematosus, or both? Mod Rheumatol. 2009; 19:212–215.
22. Satter EK, High WA .Non-bullous neutrophilic dermatosis within neonatal lupus erythematosus.J Cutan Pathol. 2007; 34:958–960.
23. Fryha R, Matta M, Kurban A .Sweet’s syndrome simulating systemic lupus erythematosus.Dermatologica. 1972; 144:321–324.
24. Ackerman AB, Briggs PL, Bravo F .Differential Diagnosis in Dermatopathology III. 1993 .Philadelphia:Lea & Febiger; 22
25. Ackerman AB .Histologic Diagnosis of Inflammatory Skin Diseases: An Algorithmic Method Based on Pattern Analysis. 1997; 2nd ed.Baltimore:Lippincott Williams & Wilkins; 525–543.
26. Waldman MA, Callen JP .Pyoderma gangrenosum. Preceding the diagnosis of systemic lupus erythematosus.Dermatology. 2005; 210:64–67.
27. Obermoser G, Zelger B, Zangerle R, et al .Extravascular necrotizing palisaded granulomas as the presenting skin sign of systemic lupus erythematosus.Br J Dermatol. 2002; 147:371–374.
28. Owen CE, Malone JC, Callen JP .Sweet-like dermatosis in 2 patients with clinical features of dermatomyositis and underlying autoimmune disease.Arch Dermatol. 2008; 11:1486–1490.
29. Chu P, Connolly MK, LeBoit PE .The histopathologic spectrum of palisaded neutrophilic and granulomatous dermatitis in patients with collagen vascular disease.Arch Dermatol. 1994; 130:1278–1283.
lupus; neutrophils; bullous lupus; Sweet syndrome
Copyright © 2014 by Lippincott Williams & Wilkins