|
 
 |
| REVIEW ARTICLE |
|
| Year : 2019 | Volume
: 6
| Issue : 1 | Page : 14-24 |
|
“Interface” Dermatoses: Revisited
Sasi Kiran Attili
Visakha Institute of Skin and Allergy, Visakhapatnam, Andhra Pradesh, India
| Date of Web Publication | 13-Jun-2019 |
Correspondence Address:
Dr. Sasi Kiran Attili
Visakha Institute of Skin and Allergy, Maharanipeta, Visakhapatnam, Andhra Pradesh
India
 Source of Support: None, Conflict of Interest: None  | Check |
DOI: 10.4103/ijdpdd.ijdpdd_17_19
“Interface dermatoses” are defined as inflammatory dermatoses primarily affecting the interface elements along the dermoepidermal junction, resulting in epidermal basal cell damage. This epidermal basal cell damage can be of two forms: vacuolar/hydropic degeneration and filamentous degeneration (resulting in Civatte bodies). This may be associated with a range of secondary epidermal (atrophy to hypertrophy) and dermal changes (varying from simple effacement of the rete ridges to the diffuse dermal fibrosis/sclerosis). With so many variables, “Interface dermatitis” per se is not a diagnosis by itself but simply a reaction pattern, similar to “psoriasiform dermatitis” or “spongiotic dermatitis.” One should never use these terms as the final sign out diagnosis in reports, though in difficult cases one may allude to the reaction pattern and discuss the possible differentials. This article analyzes and discusses in detail, the basic pathogenesis behind the primary/secondary features associated with interface dermatoses and the steps involved in their histopathological evaluation; thorough knowledge of which may help one arrive at a specific diagnosis.
Keywords: Filamentous degeneration, interface dermatoses, vacuolar degeneration
How to cite this article:
Attili SK. “Interface” Dermatoses: Revisited. Indian J Dermatopathol Diagn Dermatol 2019;6:14-24 |
| Introduction | |  |
The term “interface,” by definition, refers to the surface forming a common boundary between two portions of matter or space. Such an interface extends along the basement membrane at the dermoepidermal junction in the skin as it follows the cutaneous adnexae down to their origins from deep within the dermis and often fat (follicular units in the scalp). This “interface” is composed of the basement membrane, basal epidermal keratinocytes overlying it, and the connective tissue elements adherent to its undersurface. Thus, if one takes the term literally, all diseases affecting these structures, including subepidermal immunobullous diseases, should be classified under “interface dermatoses”! However, affection of the interface elements alone does not suffice. One more crucial factor is needed to label something as “interface dermatitis.” The following is an attempt to simplify the definition and discuss in detail, the pathogenesis behind the various histological manifestations of interface dermatoses.
Over the decades, the terms interface, vacuolar, and lichenoid have been used interchangeably and indiscriminately. Some authors prefer to use these terms synonymously, not making a distinction between “lichenoid”, “vacuolar”, and “interface” dermatoses. Weedon prefers to use the term “lichenoid” for all the disease groups in this spectrum.[1] Calonje et al. however prefer to use the term “lichenoid”, where the band-like infiltrate hugs and often obscures the dermoepidermal interface, while infiltrates associated with “basal cell vacuolation (hydropic change)” are termed “interface” dermatoses.[2] The problem with this classification is that even so-called “lichenoid” dermatoses invariably demonstrate “basal cell vacuolation.” Hence, Ackerman et al. state that all dermatoses associated with basal cell damage should be termed as “interface dermatoses.”[3] The author finds this reasoning much more appealing and uses this in practice.
Epidermal basal cell damage therefore is the sine qua non for making the diagnosis and is universally present in all interface dermatoses. This may either be associated with a paucicellular or a polycellular inflammatory infiltrate, which is the only distinction between the two subtypes of interface dermatoses: lichenoid and vacuolar.
- “Lichenoid” interface dermatitis or lichenoid dermatitis is used to describe polycellular interface dermatoses characterized by a:
- Band-like lymphohistiocytic infiltrate in the upper dermis
- Hugging (the prototype being lichen planus) [Figure 1]
- Often obscuring the dermoepidermal junction [Figure 2]. seen in lichen planus, fixed drug eruption, paraneoplastic pemphigus, some cases of subacute lupus erythematosus, erythema multiforme, and acute pityriasis lichenoides.[1]
- Vacuolar interface dermatitis, on the other hand, refers to a paucicellular (relatively) interface dermatitis, where a patchy or confluent sprinkling of lymphocytes is found along the dermoepidermal junction (with or without obscuring it), in conjunction with a few apoptotic keratinocytes undergoing “vacuolar degeneration,” e.g., graft-versus-host disease, erythema multiforme, lupus erythematosus, and inflammatory vitiligo, [Figure 3]. Note that vacuolar degeneration is not exclusive to “vacuolar interface dermatoses” and is seen in lichenoid dermatoses too. The term simply implies that vacuolar basal cell degeneration is the feature predominantly visible in this group of diseases, due to the paucity of inflammatory cells; which in lichenoid (poly-inflammatory) dermatoses is overrun by the inflammatory cells.
 | Figure 1: Hypertrophic lichen planus with a dense, band-like lymphocytic infiltrate hugging the dermoepidermal junction (H and E, ×100)
Click here to view |
 | Figure 2: (a) Lichen planus with a dense band-like infiltrate obscuring the dermoepidermal junction/interface (H and E, ×100). (b) Pityriasis lichenoides with a dense lichenoid infiltrate, obscuring the dermoepidermal interface. Note the confluent parakeratosis (H and E, ×40)
Click here to view |
 | Figure 3: Inflammatory vitiligo demonstrating focal, patchy, mild vacuolar interface change (H and E, ×200)
Click here to view |
Therefore, the definition of interface, lichenoid, and vacuolar dermatitis may be simplified as follows:
- Interface dermatitis: Inflammatory dermatoses primarily affecting the interface elements along the dermoepidermal junction, resulting in epidermal basal cell damage.
- Lichenoid interface dermatitis: Interface dermatitis with intense inflammation overrunning the vacuolar interface change
- Vacuolar interface dermatitis: Paucicellular interface dermatitis with prominent basal cell vacuolization.
However, the states, lichenoid and vacuolar, are not really exclusive or disease specific. Depending on the stage of disease and intensity of inflammation, one may see lichen planus with predominant vacuolar change (late/burnt out stages) [Figure 4], or conversely, one may see vacuolar interface dermatoses with intense inflammation, e.g., acute pityriasis lichenoides and erythema multiforme [Figure 2]b. Therefore, using these terms serves little purpose apart from implying that the disease is pauci-inflammatory or poly-inflammatory. The term “interface dermatoses” is much more apt as it covers the pauci- and poly-inflammatory states of the diseases under this classification. Regardless, as convention has rooted these terms firmly within the echelons of dermatopathology, the author continues to use these terms in his sign-outs, with the caveat that these descriptive terms bear no implication on the final diagnosis! | Figure 4: Atrophic lichen planus in the late/post inflammatory phase showing marked epidermal atrophy in association with focal vacuolar interface change in association with a few lymphocytes (not lichenoid!), pigment incontinence and marked dermal scarring. Note the irregular basement membrane reduplication/thickening (smudged appearance) (H and E, ×200)
Click here to view |
| Epidermal Basal Cell Damage | | |
Epidermal basal cell damage, as discussed earlier, is universal to interface dermatoses. Such damaged/dead cells may be eliminated either transepidermally through the stratum corneum or expelled through the basement membrane into the papillary dermis, depending on their maturational stage.[4] Great controversy exists as to whether the epidermal basal cell damage/cell death occurs as a result of apoptosis or necrosis,[3] the discussion of which is beyond the scope of this article. What we do know is that epidermal basal cell damage manifests in two main forms (visible on microscopy).
- Vacuolar (hydropic or ballooning or liquefactive – all mean the same!!) degeneration: This is seen in all interface dermatoses (whether “lichenoid” or “vacuolar”) where the keratinocytes appear swollen due to intracellular edema, leading to the formation of intracytoplasmic vacuoles. Nuclei of such cells appear pyknotic (shrunken and hyperchromatic due to condensation of nuclear chromatin) and eventually undergo karyorrhexis (nuclear fragmentation/disintegration). These intracellular vacuoles caused by vacuolar degeneration should not be confused with three other forms of extracellular vacuoles seen in interface dermatoses:
- Suprabasal: These are formed when the “apoptotic” cells eventually get detached from the damaged basement membrane. “Clefting”/dermoepidermal separation seen in lichen planus [Figure 5]a, lupus erythematosus, etc., (Max-Joseph Josees) is a histological phenomenon due to this exaggerated vacuolar basement membrane alteration, resulting in artifactual separation of the basal epidermal cells from the basement membrane, consequent to shrinkage of tissue during fixation and processing[3]
- Within the basement membrane: Vacuolar degeneration is often accompanied by damage/fragmentation of the basement membrane itself, leading to formation of vacuoles/clefting within it
- Infrabasal: Artifactual clefting/vacuolation due to separation of the basement membrane from the scarred papillary dermis, e.g., lichen sclerosus [Figure 5]b.
 | Figure 5: (a) Atrophic lichen planus showing a dense lichenoid infiltrate and an artifactual subepidermal cleft (Max-Joseph space) (H and E, ×200). (b) Lichen sclerosus showing focal artifactual clefts between the basement membrane and the scarred papillary dermis (H and E, ×200)
Click here to view |
That these extracellular clefts are artifactual is apparent from the fact that these spaces contain neither serum nor inflammatory cells. Distinction has to be made from intense inflammatory edema at the dermoepidermal junction, leading to true blistering (apparent by the presence of inflammatory exudates within the blister cavity), seen in bullous variants of lichen planus, erythema multiforme, lichen sclerosus, lupus erythematosus [Figure 6], etc.[3] It is interesting that basement membrane vacuolation can also be seen in immunobullous diseases such as bullous pemphigoid. However, such vacuolation is not accompanied by epidermal basal cell damage (the sine qua non for inclusion under “interface dermatoses”). | Figure 6: Bullous lupus erythematosus demonstrating a subepidermal blister with numerous neutrophils and serous exudate within the blister cavity (true blister) (H and E, ×400)
Click here to view |
Epidermal cell damage in interface dermatoses is not restricted to the basal cells alone. Single apoptotic cells when found high up in the epidermis, often with no visible lymphocytes in the vicinity, are called “high apoptotic keratinocytes” [the term satellite-cell necrosis (also called lymphocyte-associated apoptosis), used to describe the finding of two or more lymphocytes in proximity to a degenerate keratinocyte, was thought to be specific for graft-versus-host disease. This has subsequently proved to be nonspecific as it has also been observed in regressing plane warts, subacute radiation dermatitis, erythema multiforme, and some drug reactions. Lymphocyte-associated apoptosis of melanocytes in vitiligo has also been described.[5] They are eventually disposed off via transepidermal elimination. These cells are usually seen in the phototoxic reactions, adult-onset Still's disease, and acrokeratosis paraneoplastica.[1] However, they are not specific and are also found in the acute cytotoxic type of interface dermatoses caused via T-cell-mediated cytotoxic damage, as defined by LeBoit.[4] These include diseases such as the erythema multiforme [Figure 7] – Steven–Johnson syndrome – toxic epidermal necrolysis spectrum, acute lupus erythematosus, pityriasis lichenoides (acute) [Figure 8]a, fixed drug eruption, paraneoplastic pemphigus, morbilliform drug eruptions, viral exanthems, and acute graft-versus-host disease. In these diseases, one observes pan-epidermal (with the intensity maximum along the basal layers) ballooning degeneration manifesting as swollen, vacuolated keratinocytes with clear cytoplasm. Severe ballooning resulting in cell rupture and lacy/feathery appearance of the epidermis is called reticular alteration. (Note that reticular alteration of the epidermal keratinocytes is also seen in spongiotic vesiculation. However, in true spongiotic vesiculation, basal epidermal damage/dyskeratotic cells are absent. This distinguishes the reticular alteration associated with spongiotic dermatoses, from interface dermatoses). Confluence of such necrotic keratinocytes throughout the upper reaches of the epidermis leads to epidermal necrosis and blistering, as seen in erythema multiforme [Figure 7]. | Figure 7: Erythema multiforme demonstrating normal basket weave orthokeratosis associated with marked ballooning degeneration of the epidermis, leading to reticular alteration and epidermal necrosis; resulting in an intraepidermal blister (H and E, ×200)
Click here to view |
 | Figure 8: (a) Pityriasis lichenoides – note the dyskeratotic cells/colloid bodies throughout the epidermis (high apoptotic keratinocytes) and squamatization of the basal layer (H and E, ×400). (b) Lichen planus – note the dyskeratotic keratinocyte with eosinophilic condensed cytoplasm and pyknotic nuclear remnants (Civatte cell) (H and E, ×400)
Click here to view |
Thus, one can try and differentiate vacuolar interface dermatoses based on the affection of basal epidermis alone [Figure 3] (chronic graft-vs.-host disease, vitiligo, lichen sclerosus, discoid lupus, dermatomyositis) or affection of the entire epidermis [Figure 7] and [Figure 8]a (acute phases of the cytotoxic-mediated interface dermatoses as discussed above). However, as is apparent, these states are not really disease specific but dependent on the intensity of inflammation/stage of the disease. For example, acute cutaneous lupus and acute graft-versus-host disease demonstrate pan-epidermal involvement, while in late phases demonstrate only basal involvement.
- Filamentous degeneration[6] (seen primarily in the lichenoid group of diseases): visible as shrunken/apoptotic/dyskeratotic keratinocytes, [the term “dyskeratotic” implies disordered or premature keratinization, also called dysmaturation. One has to realize that the natural life cycle of a keratinocyte involves keratinization and in that process “cell death.” Hence, when one describes “maturation of keratinocytes,” one is actually referring to the natural process of cell death in keratinocytes. Premature keratinization or dyskeratosis or dysmaturation in this context therefore refers to the premature maturation (leading to death) of keratinocytes, as a result of the apoptotic mechanisms induced by the inflammatory milieu. Thus, instead of the mature cornified cells being naturally present in the corneal layer, dysmaturation leads to dyskeratotic cells being present in the basal layers itself!] often with eosinophilic condensed cytoplasm and pyknotic nuclear remnants [Figure 8]b.
The term Civatte bodies [the term “Civatte cell” is used to describe dyskeratotic cells with eosinophilic cytoplasmic condensation, where the nucleus is still retained. Once the nucleus is extruded/disintegrated, it is usually referred to as a Civatte body (also called colloid body or hyaline globule). However, a number of authors ignore the nuclear status and use these terms synonymously] has been used to describe these dyskeratotic/necrotic keratinocytes, after Achille Civatte who first described them in 1953, while describing the histological features of lichen planus.[7] The pyknotic nuclear remnants of these degenerate cells eventually undergo karyorrhexis. When such cells lose their nuclei, coalesce together, and drop down below the dermoepidermal junction, they are visible as homogenous anucleate pink globules (colloid bodies or hyaline bodies). Hashimoto in 1976 introduced the term “filamentous degeneration” to describe such degeneration of keratinocytes/formation of colloid bodies. Subsequently Kumakari and Hashimoto[8] while studying the pathophysiology of cutaneous-derived amyloid via electron microscopy, concluded that filamentous cells are a special type of prematurely dying cells which attained as much differentiation (keratinization) as possible. These are what we call “dyskeratotic cells.” They observed a sequence of changes involved in the derivation of cutaneous amyloid and thus suggested that colloid bodies are precursors to cutaneous amyloid. The first four of the six steps they described explain the process of colloid body formation: (i) fibrillar changes in the cytoplasm of epidermal keratinocytes, (ii) bundle formation of tonofilaments and degeneration of cell organelles, (iii) filamentous cells (when surrounded by a plasma membrane; apparent on histopathological evaluation as pyknotic cells with eosinophilic cytoplasm, i.e., Civatte cells), (iv) filamentous masses (free lying hyaline globules at the dermoepidermal junction, i.e., hyaline globules), (v) early amyloid islands in the upper dermis, and (vi) typical amyloid islands in the deeper papillary dermis and reticular dermis.
Although it is common knowledge that Civatte cells (filamentous cells) are morphologically different from necrotic/apoptotic keratinocytes dying via hydropic degeneration, the distinction between these two forms of keratinocyte necrosis/degeneration in interface dermatoses has not been widely acknowledged. It is important to recognize that basal cells exhibiting vacuolar degeneration alone are not called dyskeratotic, as they do not exhibit disordered keratinization/filamentous degeneration! The process of keratinization is probably completely halted in these necrotic/apoptotic cells. However, intracytoplasmic vacuoles (perhaps pathogenetically different?) may also be present in cells exhibiting filamentous degeneration [Figure 8]a.[8]
Thus, although intracytoplasmic vacuolation is universally seen in all interface dermatoses and in fact accompanies filamentous degeneration typical of the lichenoid group of diseases, vacuolar interface change is not always accompanied by filamentous degeneration. This may be one feature that differentiates '“vacuolar'” from '“lichenoid' dermatoses. Whether the mechanisms for vacuolar degeneration are different from filamentous degeneration and whether the pathogenesis of the intracytoplasmic vacuoles seen in filamentous cells is different from that of hydropic/vacuolar degeneration, needs to be investigated.
| Evaluation of Interface Dermatoses | | |
A number of eminent authors have already published various algorithms for the deduction of interface dermatoses.[1],[2],[3],[5] There is an inherent flaw in a strict algorithmic approach as most diseases have not read these algorithms and there is a great deal of overlap, even within the same disease spectrum! Rather than relying on a set algorithm, understanding the basic pathogenesis behind the various secondary histological features and clinicopathological correlation is imperative to arrive at a correct diagnosis. The following is an attempt to explain the same, so that readers are able to find their own method that suits their practice. [Table 1] summarizes various steps involved in the evaluation of interface dermatoses, which the author follows. | Table 1: Summary: Steps involved in the evaluation of interface dermatoses
Click here to view |
As already discussed, there is only one feature primary to interface dermatoses: epidermal basal cell damage (vacuolar or filamentous). The primary feature being constant, the various manifestations of interface dermatoses primarily depend on a number of secondary factors that lead to epidermal and dermal changes:
- Intensity of inflammation (lichenoid vs. vacuolar, i.e., poly- vs. pauci-cellular interface dermatoses)
- Primary immunological target (basal keratinocytes in the epidermis vs. follicular unit vs. eccrine duct vs melanocyte, etc.)
- Rate of onset/progression of the disease
- Time of biopsy/stage in the evolution of the disease process (discussed already)
- Affection of a special site, e.g., photoexposed sites in lupus erythematosus, scalp in lichen planus pilaris, acral sites – common for hypertrophic lichen planus, genital/mucosal lesions in lichen sclerosus, etc.
Primary versus secondary interface affection
The first step is to determine whether affection of the interface is primary or secondary! Most dermatoses (e.g., psoriasis, eczema) tend to involve the epidermis as well as the dermis. In some cases, inflammatory cells are invariably cited at the dermoepidermal interface, often with associated focal mild vacuolar interface change. However, these are all not termed interface dermatoses. Interface vacuolar change may be seen in diseases as varied as pityriasis rosea and psoriasis! [Figure 9]. Chau et al. reported junctional vacuolar alteration (76%), spongiosis (76%), necrotic keratinocytes (67%), hypergranulosis (65%), and lichenoid infiltrate (25%) in 51 cases of classical psoriasis they evaluated.[9] Thus, when one refers to interface dermatoses, one intends to convey that the dermatoses predominantly affect the dermoepidermal junction, rather than incidental interface dermatitis due to antigen unmasking. | Figure 9: Early psoriatic lesion showing subcorneal/intraepidermal neutrophils, parakeratosis, hypogranulosis and vacuolar interface dermatitis (H and E, ×200)
Click here to view |
Evaluation on low-power/scanning magnification
Most diseases are diagnosable on scanning magnification alone, and this is of primary importance in the evaluation of any dermatoses.[3]
- Superficial: Lichen planus, lichen sclerosus, vitiligo, fixed drug eruption
- Superficial and deep: Lupus erythematosus, drug reactions, pityriasis lichenoides, morphea
- Perifollicular: Lichen planopilaris, lichen striatus
- Perieccrine: Lichen planoporitis
- Focal/single papilla affected: Lichen nitidus, but rarely also in papular granuloma annulare, papular sarcoidosis, and lichen scrofulosorum.
Evaluation on higher power
Epidermal features
- Epidermal thickness:
This may vary markedly between various interface dermatoses often within the same disease and indeed within the same section [Figure 2]a, depending at the stage at which the biopsy was taken and/or the degree of inflammation. Thus, this is a nonspecific feature.
- Acanthosis: It is usually caused by low–moderate-grade chronic inflammation stimulating epidermopoesis and/or inducing pruritus which may in turn cause reactive acanthosis, e.g., lichen planus. “Saw-toothing of rete ridges” typically described in lichen planus is simply a result of inflammatory cells eating away the acanthotic rete pegs, resulting in them appearing “chewed up!” Lichen planus when associated with severe acanthosis, i.e., lichen simplex chronicus superimposed on classical lichen planus, is termed “hypertrophic” lichen planus [Figure 1].[3] The pseudoepitheliomatous appearance of such lesions can be very difficult to distinguish from well-differentiated squamous cell carcinoma
- Atrophy/effacement of the rete ridge pattern: Often in the late stages of interface dermatoses, inflammation may be absent and the only clue may be the effacement/attenuation of rete ridges, associated with epidermal atrophy and melanophages [Figure 4] and [Figure 5]b. Squamatization of the basal layer may also serve as a clue, wherein, apoptotic basal keratinocytes are totally lost or transform/acquire polygonal shapes with abundant pink cytoplasm (resembling the spinous cells), giving the appearance of the spinous layer directly sitting on the papillary dermis [Figure 8]b.[6],[7] Atrophy is also seen in the acute/intense phase of diseases where epidermopoesis cannot keep up with the necrosis caused by the inflammatory milieu [Figure 5]a. Chronic interface dermatoses are also sometimes associated with atrophy (atrophic lichen planus [Figure 4], discoid lupus erythematosus, lichen sclerosus, and pityriasis lichenoides), presumably due to inhibition/destruction of the progenitor basal keratinocytes (stem cells) themselves
- Parakeratosis: It is usually seen in diseases associated with spongiosis/high keratinocyte damage, where the inflammatory milieu stimulates epidermopoiesis, i.e., there is not enough time for keratinocytes to mature and lose their nuclei as they ascend up, thus resulting in parakeratosis. Note that parakeratosis is usually absent in lichen planus, despite the presence of acanthosis. This is probably because the epidermal keratinocyte damage is primarily basal, leading to dropping down (rather than ascent) of these effete cells. Confluent parakeratosis is a clue to pityriasis lichenoides [readers may have noticed that the author has avoided using the terms pityriasis lichenoides chronica and pityriasis lichenoides et varioliformis acuta and has used the term pityriasis lichenoides alone. This is because often the histological features of both entities overlap and the author does not believe in (nor is it histologically possible)separating out these entities. Late lesions of pityriasis lichenoides et varioliformis acuta may show lesser degrees of epidermal necrosis with more superficial infiltrates, and early lesions of pityriasis lichenides chronica may show florid interface dermatitis with keratinocyte necrosis. Moreover, pityriasis lichenoides et varioliformis acuta can be chronic and recur for years together! Ackerman prefers to use the term Mucha-Habermann disease, to encompass both these entities.[7] However, technically, this is not right either, as Mucha and Habermann described only the acute form of pityriasis lichenoides, i.e., pityriasis lichenoides et varioliformis acuta] [Figure 2]b, where epidermal spongiosis is usually seen throughout, overlying a wedge-shaped dermal lymphocytic infiltrate. However, parakeratosis per se is not specific and may be seen in any interface dermatitis, especially when secondary eczematization (lichen planus hypertrophicus, vulval lichen planus) or mycelial/yeast colonization is present
- Orthokeratosis: It suggests acute disease, where there was not enough time for the inflammatory milieu to cause keratinocyte degeneration/dysmaturation, which is responsible for changes in the stratum corneum, e.g., hyperacute stages of the cytotoxic type of interface dermatoses such as erythema multiforme [Figure 7], fixed drug eruption, and toxic epidermal necrolysis.
- A lot of importance is usually given to the presence of hypergranulosis in lichen planus. However, it is imperative to first understand the biology of the stratum granulosum so that one does not rely too much on this feature alone! As the epidermal keratinocytes mature from the basal layer upward, keratohyalin granules form. These granules are filled with histidine-rich and cysteine-rich proteins and serve to bind the keratin filaments together. Such keratohyalin laden keratinocytes together form the stratum granulosum. Any condition that leads to epidermal acanthosis, e.g., lichen simplex chronicus and lichen planus, will result in an increase in such granules. The “wedge-shaped hypergranulosis” classically described is due to the hypergranulosis concentrating around the infundibular/eccrine duct openings and the epidermis in between the rete ridges, where the epidermis is slightly more acanthotic [Figure 1]. Hypergranulosis, therefore, is not a very sensitive clue and is not a feature of the atrophic variants of lichen planus [Figure 4] and [Figure 5]b or other lichenoid disorders not associated with epidermal acanthosis.
- As already discussed, cytotoxic interface dermatoses cause both intracellular and extracellular edema [Figure 7]. This may be quite marked in some cases, resulting in spongiotic vesciculation! However, this is not specific and may be seen to varying degrees in most interface dermatoses. As explained earlier, this may also be seen when secondary eczematization/excoriation or yeast colonization (mucosal candidiasis may be associated with vulval lichen sclerosus) is present.
- Dysplasia unlike dyskeratosis (which as explained previously is an acquired maturation disorder) refers to various changes associated with malignant transformation, i.e., nuclear hyperchromatoism, cellular pleomorphism, and loss of polarity. Such an abnormality results from an aberration in the embryological anlage, wherein the progenitor/stem cells of a particular lineage differentiate and proliferate abnormally.[7] In contrast, dyskeratotic cells just mature abnormally and do not beget! Such genetic aberrations, even if present in dyskeratotic cells, do not manifest as malignancy, as dyskeratoses itself are a process of cell death! Indeed, in chronic inflammatory dermatoses, an occasional dyskeratotic cell may escape cell death (wherein, it starts to beget), leading to neoplastic transformation (e.g., squamous cell carcinoma developing within lichen planus hypertrophicus or lichen sclerosus): Epidermal tumors both benign and malignant can be associated with varying amounts of reactive interface infiltrates, e.g., lichen planus-like keratosis a.k.a. lichenoid keratosis, where there is solitary benign keratotic epidermal hyperplasia with typical lichen planus-like histological features - presumed to be either a seborrheic keratosis or solar lentigo with secondary inflammation/lichenoid change. Such reactive infiltrates are also quite common in malignant tumors, including squamous cell carcinoma, basal cell carcinoma, and melanoma. Moreover, reactive epidermal hyperplasia/regeneration, in the presence of a chronic intense lichenoid infiltrate, can appear quite dysplastic and be extremely difficult to distinguish from true malignancy which may also indeed occur, e.g., in hypertrophic lichen planus (mentioned earlier) and oral lichen planus [Figure 10]. Close clinicopathological correlation/follow-up ± repeat biopsies may sometimes be essential to avoid overdiagnosing/missing malignant transformation.
 | Figure 10: Oral lichen planus – note the marked keratinocyte dysplasia with nuclear hyperchromatism and cellular pleomorphism (H and E, ×400)
Click here to view |
Basement membrane
- Destruction/fragmentation: As a rule, all interface dermatoses are associated with basement membrane damage/fragmentation, along with damage to the basal keratinocytes [Figure 8]a and [Figure 8]b. Such basement membrane destruction is particularly apparent in interface obscuring dermatoses such as fixed drug eruption and acute pityriasis lichenoides.
- Thickening: Paradoxical thickening/reduplication, as a result of regenerative mechanisms overwhelming the inflammatory destruction of basement membrane, is however seen in some circumstances, i.e., mainly in discoid lupus erythematosus but also in dermatomyositis and in late cases of lichen sclerosus. Often, such a reduplicated, irregularly thickened basement membrane appears “blurred” or “smudged.” It is interesting that this is primarily seen in chronic cases, rather than in the acute inflammatory phase, where basement membrane destruction is seen. [Figure 4] demonstrates basement membrane thickening in a case of atrophic lichen planus.
Dermal features
- Nature of infiltrate: Lymphocytes causing epidermal basal cell damage are of course seen as a universal feature. However, the presence of additional cells can serve as a clue:
- Lymphocytes + histiocytes ± giant cells: Seen in lichen nitidus, papular granuloma annulare, papular sarcoidosis, and lichen scrofulosorum. The former three entities can be extremely difficult to differentiate as often they look similar. The problem is exemplified when a single papilla alone is affected and the “claw-like” hyperplasia of the surrounding epidermis typical of lichen nitidus is incomplete [Figure 11]. Close clinicopathological correlation is required in such cases. Histiocytes may occasionally be seen in cases of lichen striatus too.
 | Figure 11: Lichen nitidus showing a granulomatous, lichenoid infiltrate occupying a single dermal papilla, in the absence of claw-like encapsulation by the surrounding rete pegs (H and E, ×40)
Click here to view |
A special mention has to be made regarding interface change in Hansen's disease. It is the author's experience that interface dermatitis, either focal or diffuse, is invariably seen in Hansen's disease, across the spectrum [Figure 12]. Although the prominent periadnexal/perineural/granulomatous infiltrate gives away the diagnosis, the author has seen cases, wherein the absence of clinical correlation, one may easily miss the diagnosis (such a lichenoid granulomatous pattern may also be seen in other infective conditions such as lupus vulgaris, granulomatous secondary syphilis, and some deep fungal infections as well as in noninfective conditions such as drug eruptions. However, they do not present as interface dermatoses per se). | Figure 12: Borderline tuberculoid Hansen's with a dense lymphohistiocytic infiltrate and numerous giant cells, associated with interface vacuolar basal cell degeneration (H and E, ×200)
Click here to view |
- Lymphocytes + neutrophils ± eosinophils: Usually seen in drug reactions
- Predominantly plasma cells: Consider acrodermatitis chronica atrophicans and syphilis (histiocytes present too)
- Atypical lymphocytes: Pre/early mycosis fungoides where epidermotropism is not yet present can be extremely difficult to diagnose, especially when one has to differentiate this from inflammatory vitiligo [Figure 3], as true vacuolar interface change is present in both.[10] To complicate matters further, atypical lymphocytes in mycosis fungoides have a perinuclear halo[1] [Figure 13], the cause of which is yet unknown even to-date![11] These halo'ed lymphocytes can be extremely difficult to differentiate from vacuolated basal keratinocytes with their pyknotic nuclei (simulating the abnormal lymphocytic nuclei). Fortunately, for dermatopathologists, missing/delaying the diagnosis of early mycosis fungoides has no adverse prognostic implications!
 | Figure 13: Hypopigmented mycosis fungoides (patch stage) showing epidermotropism of atypical lymphocytes with a perinuclear halo and vacuolar interface dermatitis. Without clinicopathological correlation, such cases can be extremely difficult to distinguish from inflammatory vitiligo, especially when epidermotropism is absent (H and E, ×200)
Click here to view |
- Interstitial inflammatory infiltrates composed of lymphocytes and histiocytes ± eosinophils ± neutrophils: The presence of interstitial granuloma annulare-like changes with or without signs of collagenolysis, in conjunction with interface change, is a clue to interstitial granulomatous dermatitis (lichenoid and granulomatous dermatitis). Neutrophils are prominent in some cases (palisaded neutrophilic and granulomatous dermatitis).
Melanin incontinence/melanophages
In skin of color, pigment incontinence is inevitable in any dermatoses affecting the epidermis and is not specific to interface dermatoses alone.[12] However, the degree of pigment incontinence is far greater in interface dermatoses. This is due to a variety of factors taking place simultaneously: stimulation of melanogenesis, defective transfer of melanin, and keratinocyte destruction. Free lying dermal pigment when phagocytosed by histiocytes results in melanophages. Marked pigment incontinence is a clue to fixed drug eruption, although clinicopathological correlation is imperative. Often pigment incontinence/melanophages (±effacement of rete ridges) may be the only clue, particularly in postinflammatory/end stages of interface dermatoses, e.g., lichen planus pigmentosus [Figure 13]. [Note that pigmentary incontinence/melanophages alone do not determine the clinical pigmentation status of the lesion. One may see hypopigmented lesions in association with melanin incontinence (lichen sclerosus) or conversely one may see clinically hyperpigmented lesions with disproportionately less dermal melanin (lichen planus hypertrophicus). This is because the color of the lesions also depends on the thickness of the epidermis and the extent of keratinocyte melanization, which in turn depends on the balance between epidermopoesis and keratinocyte apoptosis), lichen sclerosus, and lichen planus].
Dermal scarring
Most dermatoses causing dermal scarring are associated with follicular plugging and adnexal destruction. Lichen sclerosus is the disease commonly associated with papillary dermal scarring. This is usually preceded/associated with hyalinization (glassy change) due to mucopolysaccharide deposition. Fibrin deposition in the papillary dermis can also be seen in early inflammatory cases of lichen planus. In later phases, however, dermal scarring sets in [Figure 4]. Predominantly, perifollicular scarring is a clue to lichen planopilaris (and discoid lupus erythematosus). In late stages of these diseases, scarring may sometimes be quite marked, resulting in morphea-like changes. Similarly, chronic graft-versus-host disease can be associated with sclerodermatous changes.
Vascular changes
- Perivascular lymphocytic cuffing: Although true vasculitis is not seen in interface dermatoses, perivascular lymphocytic cuffing ± intramural lymphocytes (which some people controversially term “lymphocytic vasculitis”) may be seen in a number of interface dermatoses, including lupus erythematosus, drug reactions, pityriasis lichenoides, and pigmented purpuric dermatoses. The former three are also the conditions usually associated with superficial and deep perivascular inflammatory infiltrates. Perivascular lymphocytic cuffing when accompanied by extravasated red blood cells ± hemosiderin is a clue to the diagnosis of pigmented purpuric dermatoses (capillaritis). Although extravasation of red blood cells may also occur in pityriasis lichenoides, suprabasal epidermal keratinocyte necrosis and spongiosis as in pityriasis lichenoides are not usually seen in capillaritis. Moreover, vascular changes often with fibrinoid necrosis (±leukocytoclastic vasculitis) are much more pronounced in pityriasis lichenoides.
Telangiectasia
Although not entirely specific, the presence of an atrophic epidermis along with patchy vacuolar interface change and telangiectatic vessels in the upper dermis is a clue to the diagnosis of poikilodermas. Late stages of graft-versus-host disease and lupus/dermatomyositis may however also show this feature.
Interstitial mucin
It is a feature classically described but not always seen in cutaneous lupus erythematosus. However, it is a helpful clue in the presence of superficial and deep dermal infiltrates, when differentiation from pityriasis lichenoides is difficult. Interstitial mucin may also be present in interstitial granulomatous dermatitis.
Involvement of the subcutis/panniculum
Pannicular involvement in the presence of interface dermatitis merits consideration of two diseases: morphea profunda where the involvement of the panniculum is predominantly septal; and, lupus panniculitis which is usually associated with lobular lymphocytic infiltrates.
| Conclusion | | |
It is important to recognize that the secondary changes described above, though commonly seen in the respective interface dermatoses, should only be used as pointers. None of these features should be used in isolation to arrive at a specific diagnosis. In day-to-day practice, a significant number of cases do not follow set algorithms or conform to the existing disease classifications and have to be termed simply as “lichenoid” or “vacuolar” interface dermatoses. Further, interface basal degeneration may not always be obvious as explained earlier. Subtle clues in the form of attenuation of rete ridges, epidermal atrophy, squamatization of the basal layer, melanin incontinence, and papillary dermal fibrosis/scarring [Figure 4] and [Figure 14] may be the only tell-tale signs obvious in the late phases and in pauci-inflammatory interface dermatoses such as vitiligo and poikiloderma. Clinicopathological correlation is paramount. | Figure 14: Lichen planus pigmentosus showing postinflammatory features only, i.e., epidermal thinning with effacement/attenuation of rete ridges and pigment incontinence (H and E, ×100)
Click here to view |
Apart from basal keratinocyte damage, interface dermatoses share nothing in common. Although the secondary epidermal and dermal features allow differentiation between the various interface dermatoses, they bear no etiological resemblance to each other. In fact, this applies even to subtypes within the same disease spectrum. Take for example, lichen planopilaris and lichen planoporitis. Are these really part of the same disease spectrum, pathogenically related? Do they have anything really in common apart from exhibiting a “lichenoid” infiltrate? Lichen planus pigmentosus bears no resemblance to lichen planus either clinically or histologically; neither is classical lichen planus usually associated with lichen planus pigmentosus. Moreover, lichen planus pigmentosus is usually associated with epidermal thinning and a vacuolar (pauci-inflammatory) dermatitis, though lichenoid (dense) infiltrates are occasionally seen.[1],[2]
The lichen planus group of diseases are as varied as the “pemphigus” group of diseases in etiology and presentation, though they share some similarities as far as prognosis/therapeutic response is concerned. The author believes that just as pemphigus foliaceous cannot be called a variant of classical pemphigus (vulgaris), it is perhaps not appropriate to call lichen planus pigmentosus, etc., true variants of lichen planus. A better term would be “lichenoid group of diseases.” Alternatively, a more balanced view would be to call them all “interface dermatoses” and resist from lumping them into specific entities, unless definite etiopathogenic continuity is proven. Future research should concentrate on elucidating the antigenic targets of these dermatoses and enable us to understand their pathogenesis further.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
| References | | |
| 1. | Weedon D, editor. The lichenoid reaction pattern. In: Skin Pathology. 2 nd ed. New York: Churchill Livingstone; 2002. p. 31-74.  |
| 2. | Calonje E, Brenn T, Lazar A, McKee PH. McKee's Pathology of the Skin: With Clinical Correlations. 4 th ed. Edinburgh: Elsevier/Saunders; 2012. |
| 3. | Ackerman AB, Boer A, Bennin B, Gottlieb G. Histologic Diagnosis of Inflammatory Skin Diseases: An Algorithmic Method Based on Pattern Analysis. 3 rd ed. New York: Ardor Scribendi; 2005. |
| 4. | LeBoit PE. Interface dermatitis. How specific are its histopathologic features? Arch Dermatol 1993;129:1324-8. |
| 5. | Attili VR, Attili SK. Lichenoid inflammation in vitiligo – A clinical and histopathologic review of 210 cases. Int J Dermatol 2008;47:663-9. |
| 6. | Hashimoto K. Apoptosis in lichen planus and several other dermatoses. Intra-epidermal cell death with filamentous degeneration. Acta Derm Venereol 1976;56:187-210. |
| 7. | Ackerman B, Mones J. Resolving Quandaries in Dermatology, Pathology, & Dermatopathology. Vol. 2. New York: Ardor Scribendi; 2001. |
| 8. | Kumakiri M, Hashimoto K. Histogenesis of primary localized cutaneous amyloidosis: Sequential change of epidermal keratinocytes to amyloid via filamentous degeneration. J Invest Dermatol 1979;73:150-62. |
| 9. | Chau T, Parsi KK, Ogawa T, Kiuru M, Konia T, Li CS, et al. Psoriasis or not? Review of 51 clinically confirmed cases reveals an expanded histopathologic spectrum of psoriasis. J Cutan Pathol 2017;44:1018-26. |
| 10. | Soro LA, Gust AJ, Purcell SM. Inflammatory vitiligo versus hypopigmented mycosis fungoides in a 58-year-old Indian female. Indian Dermatol Online J 2013;4:321-5. [ PUBMED] [Full text] |
| 11. | El Darouti M, Marzouk SA, Horn TD. Failure of detection of mucin in the clear halos around the epidermotropic lymphocytes in mycosis fungoides. J Cutan Pathol 2000;27:183-5. |
| 12. | Joshi RS. Description of a proposed simple semi-objective histological scale for the assessment of dermal melanophages in inflammatory skin diseases. Indian J Dermatopathol Diagn Dermatol 2018;5:34-8. [Full text] |
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9], [Figure 10], [Figure 11], [Figure 12], [Figure 13], [Figure 14]
[Table 1]
|
Search Pubmed for