In 1916, Darier introduced the term "erythema annulare centrifugum" for a dermatitis that he thought was characterized by annular lesions with an elevated, firm, nonscaly border and typified histopathologically by a superficial and deep, dense infiltrate of lymphocytes without involvement of the epidermis.[1] In the same article, he made mention of another condition with an annular scaly border characterized by spongiosis histopathologically. In the following decades, the term erythema annulare centrifugum came to be used for both diseases, which were then interpreted to represent manifestations of the very same pathologic process.

It was only in 1974, that W.H. Clark and coworkers published an article in Human Pathology about what they called "the lymphocytic infiltrates of the skin."[2] In it, they attempted to separate patterns of infiltrates made up largely of lymphocytes. In their conception, Darier reported on two entirely unrelated conditions: one with a superficial infiltrate accompanied by epidermal changes, to which they suggested to give the name "erythema annulare centrifugum" and one with a deep infiltrate devoid of epidermal changes, which they suggested naming "palpable migratory and arciform erythema." We will address this condition in a separate paragraph below.

In 1978, in a volume called Histologic Diagnosis of Inflammatory Skin Diseases, A.B. Ackerman addressed the same subject under the heading "gyrate erythema" thus: "Gyrate erythemas are of two types: (1) superficial and (2) deep. Deep lesions, unlike the superficial ones, show neither focal epidermal spongiosis nor parakeratosis. Clinically, the lesions of the deep type are firmer than those of the superficial type, and they lack peripheral scales."[3] Ackerman added to the conception of Clark and coworkers the precise description of the epidermal changes in what he called the superficial type of gyrate erythemas, to wit, focal spongiosis and parakeratosis. He emphasized also that this condition had histopathologic changes very similar to those seen in pityriasis rosea, whereas the deep type of gyrate erythema was typified by an entirely different pattern, namely, dense infiltrates of lymphocytes around the superficial and deep vascular plexus in the dermis unaccompanied by any epidermal changes.

In 1981, Bressler and Jones reviewed the history of erythema annulare centrifugum and summarized their observations as follows: "Darier, who originally described the disease, described an annular, indurated, erythematous lesion without a scale that histologically was characterized by a superficial and deep lymphohistiocytic infiltrate and normal epidermis. Ackerman suggested that there are two types of gyrate erythema, a superficial type showing a scale, and the deep type as described by Darier. After reviewing the literature and studying patients with gyrate erythemas, it seems that there are two distinct types best termed the superficial and deep forms of gyrate erythema. Dermatologists use the term erythema annulare centrifugum to denote both of these forms. Perhaps that term should be discarded."[4]

Over the past decades, the designations "superficial type" and "deep type" of erythema annulare centrifugum, "superficial type" and "deep type" of gyrate erythema, or "superficial type" and "deep type" of figurate erythema have been used in articles and textbooks to refer to two conditions that have in common an annular or gyrate appearance of lesions clinically. All these designations, however, do not convey clearly that the two diseases are totally unrelated. For that reason, Ackerman et al. introduced in 1995 the term "erythema figuratum" as a designation for what others called the "deep type." In their book, Resolving Quandaries in Dermatology, Pathology and Dermatopathology (I. Volume), they write: "In times past, we divided 'figurate erythemas' into two types, namely, 'superficial' and 'deep' based in part on whether one or both vascular plexuses were surrounded by lymphocytes. We considered the scaly erythema annulare centrifugum to be synonymous with the superficial type, but accorded no appellation to the nonscaly (because of the lack of epidermal involvement) deep type. That latter condition consists clinically of elevated, smooth-surfaced, red lesions with prominent slopes and arcuate, annular, and polycyclic configurations . . . we now employ the term erythema annulare centrifugum instead of the superficial type of figurate erythema, and erythema figuratum in lieu of the deep type." [5] The term "erythema figuratum" as a designation for a distinctive disease, however, has not found entrance into textbooks of dermatology and dermatopathology. In most of the current textbooks, erythema annulare centrifugum is still discussed as having two types that are considered to be variants of one and the same process but not two entirely unrelated conditions, as can be seen from the following quotations:

"There are two distinct patterns, a superficial type and a deep type. In the deep variant there is a moderately dense infiltrate of lymphocytes, histiocytes and rarely eosinophils involving both the superficial and deep vascular plexuses. The epidermis is normal."[6]

"Erythema annulare centrifugum is also known as gyrate erythema and represents a hypersensitivity reaction manifesting as arcuate and polycyclic areas of erythema. The condition has been categorized into superficial and deep variants. The deep form was originally described by Darier and is characterized by annular areas of palpable erythema with central clearing and absence of surface changes . . . The presence of increased deposits of connective tissue mucin distinguishes tumid LE from deep gyrate erythema."[7]

"Erythema annulare centrifugum [EAC] (synonyms superficial and deep gyrate erythema, erythema perstans, erythema gyratum repens) has been applied to a broad spectrum of clinical findings . . . However Ackerman prefers to call the two histologically different forms (superficial and deep), of EAC "superficial gyrate erythema" and deep gyrate erythema" . . . In the deep variant of EAC, the epidermis is usually unremarkable and a mononuclear cell infiltrate with a sharply demarcated perivascular arrangement is present in the mid and lower dermis."[8]

"Bressler and Jones have suggested dropping the term erythema annulare centrifugum, and instead referring either to deep gyrate erythema that has a firm yet indurated border, lacks scale and is rarely pruritic, or to superficial gyrate erythema that has an indistinct border, has trailing scale, and is more often pruritic. Although this subdivision is reasonable, it does not help in identifying possible etiologic agents . . . The typical deep gyrate erythema lack epidermal damage and have intense lymphohistiocytic cuffing about both the superficial and deep vessels."[9]

In summary, Darier, in his first description of what he called erythema annulare centrifugum, already mentioned that there are two different pathologic processes that share the shape of lesions clinically but that have completely different histopathologic findings.[1] In the years that followed, several authors have attempted to stress that two unrelated conditions should not be referred to by the same term.[2,3,4] Nevertheless, the nomenclature is confusing to date because both conditions are still discussed in most textbooks in one and the same paragraph under one and the same heading. The etiology of both conditions is still opaque.

W.H. Clark and coworkers, in their article of 1974 "The Lymphocytic Infiltrates of the Skin," distinguished histopathologically between four patterns of infiltrates made up of lymphocytes and they thought each of them to be associated consistently with a specific presentation clinically.[2] They designated and described these conditions thus:

(1) Erythema anulare centrifugum: " . . . annular and . . . bizarre erythemas that tend to affect the trunk and extremities, but usually spare the face. . . . the infiltrate is the sparsest [of all lymphocytic infiltrates] . . . cells are disposed about the superficial vascular plexus, show some diffusion into the papillary dermis, and usually involve the epidermis"

(2) Palpable migratory and arciform erythema: " . . . affects the skin of the trunk or buttocks. [lesions] tend to disappear spontaneously only to reappear after some weeks . . . [There is a] compact dense cuff of lymphocytes about the vascular bed of the reticular dermis . . . the papillary dermis and epidermis are generally spared."

(3) Discoid palpable erythema of the face (Jessner and Kanof): " . . . peculiar reddish discoid lesion usually affecting the face; the process . . . is often confused with discoid lupus erythematosus . . . [the] infiltrate [of the reticular dermis] may be sparse and is associated with mucinosis of collagen."

(4) Cutaneous lymphoplasia: " . . . the face is the common site of this red nodule . . . [a] tumor-like process of the reticular dermis, showing lymphocytes with follicle formation, at times admixed with numerous eosinophils."

Although the article was designated by the authors as"a clinicopathological classification," it was unaccompanied by a single photograph of lesions clinically, that making it difficult for a reader to estimate fully the observations made by the authors. As mentioned earlier, Clark and coworkers suggested using the term "erythema annulare centrifugum" exclusively for an annular dermatitis with changes in the epidermis histopathologically (their category 1). For another condition without epidermal involvement and typified by denser and deeper infiltrates (their category 2), they introduced a new term, namely, "palpable migratory and arciform erythema," and they sought to convey by it the clinical appearance (which they deemed to be arciform) and the dynamic of the disease (which they had observed to be migratory). They noted that this malady was also distinct from yet another condition, in which, in addition to an infiltrate of lymphocytes without epidermal changes, deposits of mucin were found in the dermis (their category 3). The disease typified by deposits of mucin was regarded by Clark et al. to be the very same one described by Jessner and Kanof under the designation "lymphocytic infiltration of the skin," which we will address separately below.

Subsequent to the publication by Clark, the term "palpable migratory and arciform erythema" was used rarely and almost never in chapters and articles addressing the issue of figurate and annular erythemas. In very few articles, authors reported on patients in whom they had made the diagnosis of palpable migratory and arciform erythema: In 1990, Lorisch reported on a man with recurrent polycyclic lesions on the chest typified by dense perivascular infiltrates histopathologically.[10] Abeck et al, in 1997, told of their experience with three patients, all of whom had recurrent lesions on the trunk that had appeared rather quickly.[11] The authors described the findings histopathologically thus: "The epidermis was unchanged. A dense lymphocytic infiltrate was located around epithelial structures of adnexa and around blood vessels. There was no interstitial distribution of inflammatory cells, and biopsy specimens were depleted of mucin. Plasma cells were absent as well. . . . " Steinemann et al. in 1999, reported on a patient with recurrent lesions on the face, neck, and back, accompanied by pruritus and burning sensations, in whom they had made a diagnosis of palpable migratory and arciform erythema.[12] Histopathologically, they found a dense infiltrate of lymphocytes around vessels and adnexa, together with deposits of mucin. Although these findings are quite suggestive for the tumid expression of lupus erythematosus, the authors made a diagnosis of palpable migratory and arciform erythema. In 2004, J. M. Muche et al. told of their experience with a patient who had large, elevated, reddish annular lesions localized on the upper trunk. Histologically dense infiltrates in the dermis consisted of CD8+ lymphocytes with polyclonal T-cell receptor (TCR) gene rearrangement.[13] They considered palpable arciform migratory erythema to be a T-cell pseudolymphoma.

In most textbooks of dermatology and dermatopathology, palpable arciform migratory erythema is not mentioned.[6,7,9] Only in the textbook edited by Bolognia and coworkers, the condition is described thus: "Palpable migratory arciform erythema is an uncommon condition with a dermal lymphoid infiltrate in which irregularly spreading plaques are present."[8]

No study has ever been published about a larger group of patients in whom a diagnosis of palpable migratory arciform erythema had been made. In recent articles, it is not mentioned that the term palpable arciform migratory erythema had been introduced by Clark et al. for the deep type erythema annulare centrifugum, the same condition to which others gave the names deep gyrate erythema, deep figurate erythema, and erythema figuratum.

The term "lymphocytic infiltration of the skin" was introduced at a meeting of the Bronx Dermatological Society, in 1953, where Jessner and Kanof reported on a group of patients that they believed to suffer from a distinctive condition.[14] In the meeting report, Jessner described his findings thus: "The lesions [of lymphocytic infiltration] are flat, discoid, more or less elevated, pinkish to reddish, brown, starting as small papules, expanding peripherally, sometimes clearing in the center, sometimes showing circinate arrangement. The surface is smooth, occasionally uneven. There is no follicular hyperkeratosis . . . . The face is obviously the area of predilection . . . . The epidermis may be stretched but is otherwise uninvolved. . . . Distributed through the cutis are rather sharply circumscribed lymphocytic infiltrations, sometimes extending to the subcutaneous fat, frequently, but not always, around vessels and/or appendages. The infiltrates consist of lymphocytes, mostly small ones, and usually a few histiocytes and plasma cells. The infiltrates are often enmeshed in a fine reticulum. There are no eosinophils, no germinal centers or germinal center-like formations. . . . These cases must be distinguished from cases of chronic discoid lupus erythematosus, under which label they are invariably described."

The report did not include any clinical pictures or photomicrographs.

The concept of lymphocytic infiltration (Jessner and Kanof) was reinforced by W.H. Clark and colleagues who, in 1974, in their article "The Lymphocytic Infiltrates of the Skin," categorized the condition described by Jessner and Kanof as a distinctive clinicopathologic entity.[2] Just as Jessner and Kanof did, Clark regarded the condition to be different from lupus erythematosus because of the lack of changes at the dermoepidermal junction. In their publications, neither Jessner and Kanof nor Clark et al referred back to publications by colleagues from Germany and France published decades before on a manifestation of lupus erythematosus that typically was devoid of epidermal changes, to which they had given the designation lupus erythematosus tumidus. This manifestation of lupus erythematosus will be addressed separately below.

In the years that followed, considerable confusion existed in regard to differentiation of lymphocytic infiltration of Jessner and Kanof from lupus erythematosus on one hand and from palpable migratory and arciform erythema on the other.[15--26] In 1989, Toonstra and colleagues studied 100 patients in whom they had made the diagnosis of lymphocytic infiltration of Jessner and Kanof.[21] In their opinion, the condition represented "a single entity," because "Progression into polymorphous light eruption, discoid lupus erythematosus, or lymphoma was not observed." In 1990, Lorisch and coworkers concluded that "[palpable migratory and arciform erythema] can . . . not be differentiated immunohistologically from the lymphocytic infiltration of Jessner-Kanof."[10] In contrast, Cerio and colleagues were convinced that "Jessner's lymphocytic infiltration is a heterogeneous disorder with at least two separate immunophenotypes. The finding of perivascular follicular center differentiation in more than half the specimens suggests a possible relationship in these cases to other benign lymphoid hyperplasias (pseudolymphomas)."[22] No mention was made in any of these articles of lupus erythematosus tumidus, indicating that it was not a universally accepted concept that lupus erythematosus could present itself devoid of epidermal involvement.

In 1995, Ackerman and coworkers suggested that "lymphocytic infiltration of Jessner is a synonym for tumid discoid lupus erythematosus" and they described the latter condition to be typified by lymphocytic infiltrates in the dermis together with deposits of mucin but with no changes at the dermoepidermal junction.[5] Abeck et al, in 1997, stated that "palpable migratory arciform erythema shows distinctive differences in clinical and pathological features . . . in contrast to other diseases with cutaneous lymphocytic infiltrates, including lymphocytic infiltration of Jessner and Kanof."[12] Steinemann et al., in 1999, designated palpable migratory arciform erythema as a "T-cell pseudolymphoma" and stated that it "is not a unique entity, but a special clinical presentation of LIS [lymphocytic infiltration of the skin of Jessner and Kanof]." [11] But, in 2001, Weber and coworkers confirmed by way of phototesting that lymphocytic infiltration of Jessner is likely to be a manifestation of lupus erythematosus because "in 10 patients with active LIS [lymphocytic infiltration] . . . UVA and/or UVB elicited abnormal papular phototest reactions resembling lesions of LIS both clinically and histologically . . . Lesions typically developed 3--6 days . . . after the first UV exposure . . . This characteristic latency interval . . . indicates that LIS is a photosensitive disorder distinct from polymorphic light eruption but indistinct from lupus erythematosus (LE)."[23]

Nevertheless, Dippel et al., in 2002, stated that "Lymphocytic infiltration of the skin (Jessner and Kanof) is a T-cell pseudolymphoma . . . ." [24] Poenitz and coworkers, in 2003 were convinced that "Jessner's lymphocytic infiltration of the skin (JLIS) is a clinically and histologically distinct disease entity" that should not be separated from palpable migratory arciform erythema.[25] Only recently, some authors mentioned that "Lymphocytic infiltration frequently develops central clearing that gives an annular appearance similar to that of erythema chronicum migrans."[26] They suggested that lymphocytic infiltration of Jessner may sometimes be a manifestation of borreliosis.

In most current editions of textbooks on dermatology and dermatopathology, lymphocytic infiltration of the skin is addressed separately, but criteria for diagnosis of it are largely indistinguishable from those mentioned in the same volumes for diagnosis of tumid lupus erythematosus.

Weedon, in his textbook Skin Pathology writes that "Jessner's infiltrate of the skin is a relatively uncommon condition . . . there is usually a moderate dense, perivascular infiltrate of small lymphocytes involving the superficial and deep vascular plexuses of the dermis . . . a small amount of mucin may be seen between the collagen bundles; the epidermis is usually normal, with no evidence of atrophy, basal vacuolar change or follicular plugging. These features together with the negative immunofluorescence are said to distinguish this condition from lupus erythematosus." [6]

In contrast, a paragraph in Fitzpatrick's Dermatology in General Medicine reads as follows: " The etiology of JLIS [Jessner's lymphocytic infiltration of the skin] is unknown . . . some regard it as a distinct entity, whereas others consider it to be a variant of lupus erythematosus, or less commonly polymorphous light eruption, CLH, or lymphocytic lymphoma. The tumid variant of LE is typified by abundant dermal mucin and subtle alterations at the dermal-epidermal interface, and these changes are not seen in JLIS." [9]

In a book titled Dermatology edited by Bolognia et al., the condition is addressed thus: "JLIS [Jessner's lymphocytic infiltration of the skin] is a poorly described entity whose very existence remains controversial. Some authors believe it to be a variant of lupus erythematosus, cutaneous lymphoid hyperplasia, or polymorphous light eruption. Others believe it to represent an infectious process, possible caused by Borrelia burgdorferi infection. Still other raise the possibility that JLIS ia an early evolving cutaneous T cell lymphoma . . . The differential diagnosis includes tumid lesions of LE."[8]

In the 9th edition of Lever's textbook, where it is written that lymphocytic infiltration of the skin "is not a well-understood entity . . . The histologic differential diagnosis includes: lupus erythematosus, polymorphous light eruption, lymphocytoma cutis, and lymphoma . . . " the authors continue to say that "Approximately 10% of cases of lupus that lack interface alterations and that show negative direct immunofluorescence may initially be placed into the category of lymphocytic infiltrates of the skin." [7]

In sum, the controversy about the nature of lymphocytic infiltration of Jessner and its relationship to palpable migratory arciform erythema, tumid lupus erythematosus, erythema migrans, and lymphomatous infiltrates continues to this date. Confusion reigns in articles and textbooks about the relationship of all these conditions to each other because criteria set forth for diagnosis of them seem to be merging.

The problems of diagnosis and differential diagnosis of the conditions addressed above cannot be understood without elucidating the historical background of tumid lupus erythematosus.

The first description of lupus erythematosus tumidus dates back as early as 1909, when E. Hoffman published a report in a German journal on his experience with patients who presented themselves with lesions on the face that were reminiscent of those seen in lupus erythematosus but largely devoid of epidermal involvement in the form of scale or crust.[27] Histopathologic investigations were not performed by Hofman, but the designation lupus erythematosus tumidus was suggested by him because of the firm infiltration of the skin.

More than 20 years later, at a meeting of the Société de dermatologie et de syphiligraphie in 1930, Gougerot and Burnier told of two patients with an unusual form of lupus erythematosus that seemed to be rare and not well known. Lesions consisted of one or more round or oval lesions covered by nonscaly epidermis.[28] In one patient, lesions had been present behind the ears; in the other lesions were situated on the neck, and the patient had previously been diagnosed with typical lesions of lupus erythematosus of the face. At other sessions of the same society in 1931 Gougerot and Burnier [29] as well as Gougerot and Carteaud, [30] presented reports on two other patients with the same condition. Unfortunately, no biopsies were performed in these patients.

In the following years, few reports on similar patients appeared in the German and French literature of dermatology, [31–33] but no larger studies were undertaken. In the 1970s, French authors noted the similarities between the tumid expression of lupus erythematosus and what colleagues in the United States referred to as lymphocytic infiltration of the skin.[34] Only slowly, the concept of tumid lupus erythematosus found entry in publications written in English.[35,36] In 1999, Ruiz and Sanchez told of their experience with four patients with tumid lupus erythematosus. They concluded that "the clinical findings of concurrent lesions of TLE [tumid lupus erythematosus ] and classic DLE [discoid lupus erythematosus] in the same patient, the presence of lesions of TLE in patients with SLE [systemic lupus erythematosus], and the histologic findings demonstrate that TLE is a distinct manifestation of DLE, and as such, a cutaneous expression of SLE."[37] The first larger series of 40 patients with the condition was reported only in the year 2000 [38] followed by several articles on larger groups of patients reported by the same group of colleagues.[39–43]

Some authors emphasized that "The histopathological change that primarily distinguishes LET [tumid lupus erythematosus ] from other variants of cutaneous lupus erythematosus is the lack of alterations of the dermo-epidermal junction and epidermis." [44] More recently, however, Vieira et al. observed that lupus erythematosus tumidus may have "differing degrees of mucin deposition" and "minimal epidermal changes" in about 50% of the cases. Direct immunofluorescence tests in their patients were often negative and the clinical course was usually benign without systemic manifestations. The authors concluded that "The main controversies are the acceptance of LET as a separate subset of CCLE [chronic cutaneous lupus erythematosus], and the histopathologic diagnostic features, mainly the presence or absence of epidermal and dermal-epidermal changes in these lesions."[45] The relationship of tumid lupus erythematosus to discoid lupus erythematosus as well as to systemic lupus erythematosis, however, remained to be subject of controversy.[46–48]

In recent textbooks on dermatology and dermatopathology, tumid lupus erythematosus appears to be a widely accepted concept, as can be seen from the following quotations:

"In tumid lesions [of lupus erythematosus] there is an increased dermal mucin often accompanied by dermal edema. Some cases have a sparse inflammatory cell infiltrate, while others have a heavy infiltrate of lymphocytes and less mucin . . . epidermal involvement is uncommon."[6]

"In tumid LE [lupus erythematosus], the dermal changes of DLE [discoid lupus erythematosus] are present, . . . a variable perivascular mononuclear infiltrate with prominent mucin between the collagen bundles; the classical epidermal changes that occur in other forms of LE are not present."[9]

"The epidermis appears to be uninvolved , although there is an intense dermal inflammatory infiltrate . . . morphologically the lesions are similar to those of Jessner´s lymphocytic infiltrate."[8]

"The dermal form of LE [lupus erythematosus] without surface/epithelial changes is known as tumid LE. Clinically affected patients display indurated papules, plaques and nodules without erythema, atrophy or ulceration of the surface. Histologically they have superficial and deep dermal perivascular, interstitial and periappendageal lymphocytic infiltrates associated with stromal mucin deposition." [7]

In summary, it is universally acknowledged today that a tumid expression of lupus erythematosus exists. It is characterized by superficial and deep perivascular and periadnexal infiltrates of lymphocytes in company with deposits of mucin and largely devoid of interface changes. The criteria set forth originally by Jessner and Kanof for the condition designated by them "lymphocytic infiltration of the skin" do not permit differentiation from tumid lupus erythematosus. What they termed "lymphocytes embedded in a fine reticulum" most likely refers to lymphocytic infiltrates accompanied by abundant mucin in the reticular dermis. Clark and coworkers refering to the disease described by Jessner and Kanof also stated that typically deposits of mucin were encountered in the dermis together with lymphocytic infiltrates, a combination of findings that today would be diagnosed as tumid lupus erythematosus.