Implications of Dermoscopy and Histopathological Correlation in Discoid Lupus Erythematosus in Skin of Color.

Background: Discoid lupus erythematosus (DLE) is chronic dermatosis manifests as scaly indurated plaques with erythema and peripheral hyperpigmentation. Few cases progress to systemic lupus erythematosus. Differentials include lichenoid photo dermatitis, Jessner's lymphocytic infilterate, and polymorphus light eruptions. It is difficult to assess the activity clinically. Histopathology is characteristic and evaluation of disease activity is possible. Dermoscopy is a useful diagnostic method in many dermatoses. Dermoscopy is reflection of histological changes. Hence, dermoscopic features may act as a tool for activity assessment. Here authors have pursued dermoscopic and histopathological correlation in DLE lesions to assess the activity of disease. Aims: To study dermoscopic features in DLE and correlate the patterns with histopathological changes in skin of color. Method: This study was conducted in a tertiary hospital. Clinically suspected and histopathologically proven lesions of DLE were enrolled in this study. The target lesion was marked and sent for biopsy after performing dermoscopy. Activity of the lesion was assessed on the basis of histopathological features. SPSS statistics for windows v20.0 (SPSS Inc, Chicago, USA) was used to analyze data. Chi-square and Fisher's χ2 test was used to statistically signify association. Cohen's kappa coefficient was used to determine the agreement.
Results: Study included 110 patients with Fitzpatrick skin type IV-V having 120 lesions. Follicular keratotic plug [73 (60.8%)] and peri-follicular whitish halo [65(54.1%)] were commonly found in dermoscopy. Blue-gray and brown dots, telangiectasia, follicular red dots, white rosettes and white areas include other features. Interface dermatitis, peri-appendageal infilterate, melanin incontinence, melanophages and fibrosis were noted in histopathology. Perfect agreement was observed in follicular plugs.
Conclusion: Dermoscopy patterns were well correlated with histopathological changes. Thus dermoscopy played an important role in assessing the activity of lesion. Copyright:

Introduction

Discoid lupus erythematosus (DLE) is a chronic and the most common form of cutaneous lupus erythematosus (CLE) commonly occurring in middle-aged individuals.[1] DLE usually involves sun-exposed areas, especially the face and the scalp, though covered areas are involved occasionally. It is associated with scarring and disfigurement and can progress to systemic lupus erythematosus (SLE) in 5%–10% of cases.[2] Clinical features consist of indurated plaques surmounted with adherent scales with hyperpigmented border. Distribution can be localised or generalised.[3]

Clinical diagnosis is generally obvious based on the classical morphology of the lesions. Occasionally, lesions mimic lichenoid photo dermatitis, lichen planus, Jessner's lymphocytic infiltrate, polymorphous light eruptions and sarcoidosis, and other subtypes of CLE, alarming for investigations to arrive at accurate diagnosis. Overlap exists in clinical features between DLE and subacute CLE.[4] Although scoring system is described to check the activity of lesions, definitive measures are still warranted.[5] Thus, limitation and uncertainty prevail in clinical criteria for activity of disease, which is important in treatment prediction and prognosis.

Histopathology is considered as a standard and definitive method for diagnosis. It reveals ortho-hyperkeratosis, thickening of the basement membrane, follicular dilatation and plugging, peri-appendageal lymphocytic infiltration, and involvement of the deep dermis.[67] Activity and stability of a disease can be well studied by histopathology.[17]

Dermoscopy is a non-invasive technique of examination of skin lesions wherein it enables the clinician to appreciate the details of changes that occur at the surface and subsurface levels. It demonstrates precise patterns in a particular condition, thus benefiting in the exact diagnosis.

Dermoscopy features are reflections of histopathological changes.[8] Thus, the correlation of dermoscopic features with their corresponding histopathological changes would be helpful to know the activity or stability. Thus, dermoscopy can be utilised to know the activity in a non-invasive fashion.

Dermoscopy is proven to be a reliable diagnostic method in DLE by demonstrating characteristic features, and an excellent correlation with corresponding histopathological changes can be derived.[9] However, the dermoscopic descriptions of DLE in the Indian population are limited to case reports.[101112] Here, authors pursued to study the dermoscopic patterns in active and chronic/inactive lesions of DLE and correlate features with their corresponding histopathological changes in skin of colour. Consistency of dermoscopic and histopathological correlation was assessed.

Materials and Methods

This study was conducted in a tertiary hospital in Southern India from July 2019 to December 2020. Ethical clearance and patient consent were obtained. It was a cross-sectional and observational study. All patients with age >18 years with clinically suspected and histopathologically proven lesions of DLE were enrolled in this study. Patients who were presently on treatment for DLE or with previous treatment within 1 month of enrolment were excluded from the study. Patients with immunosuppressive therapy and pregnant ladies were not included. Detailed history and patient particulars, including age, sex, location and duration of disease, were noted. Detailed physical examination and thorough haematological investigations were done to assess the systemic involvement.

The target lesion was selected and the area for biopsy was marked. Dermoscopy of the target lesion was done by DermLite Foto plus II Pro with 10 × magnification attached to a digital single-lens reflex camera to capture the images. Minimal pressure was applied during dermoscopy to prevent blanching, and ultrasound gel was used as the interface medium. The site where dermoscopy was done was biopsied with a 3.5-mm biopsy punch and sent for histopathological study. Haematoxylin and eosin staining was used for staining the slides. Dermoscopic (BSA and AG) and histopathological (BPN) features were independently analysed. Based on the dermoscopic and histopathological findings, appropriate treatment was offered to the patient.

In this study, the activity of the lesion was assessed based on histopathological features. Interface dermatitis, basal cell vacuolation and vasodilatation with peri-appendageal infiltrate were considered as features of an active lesion. Focal or diffuse fibrosis in the dermis was considered a feature of inactive lesions. Hyperkeratosis, acanthosis, follicular keratosis, melanin incontinence, thickened basement membrane, reduced number of follicles, and melanophages are the features of both types of lesions.[57] Assessment of clinical activity of a lesion was done arbitrarily based on the morphology of the lesions. The infiltrated and erythematous plaques and atrophic depigmented lesions were considered as active and inactive lesions, respectively.

Statistical analysis

Dermoscopic and histopathological data were tabulated. SPSS statistics for Windows v20.0 (SPSS Inc, Chicago, USA) was used to statistically analyse the data. Continuous variables such as age and duration are described as means ± standard deviations. Discrete variables are shown as percentages.

Chi-square and Fisher's χ2 test was used to statistically signify association between qualitative variables such as presence or absence of dermoscopic features and histopathological features. P < 0.05 was considered statistically significant. Cohen's kappa coefficient was used to determine the agreement between histopathological and dermoscopic diagnosis.

Results

In this study, 110 patients with Fitzpatrick skin type IV–V having 120 lesions who met the inclusion criteria were included in the study. Of the 110 patients, 70 were males and 50 were females. The mean age of patients was 46 ± 11 years. The mean duration of disease was 9 ± 6 months. The face was commonly affected in the case of 60 (50%) lesions. Extremities were the least affected in 10 (8.33%) lesions. Other sites involved were scalp, neck and chest with 30 (25%), 10 (8.33%) and 10 (8.33%) lesions, respectively.

Follicular keratotic plug was the most common finding and was seen in 73 (60.8%) lesions [Figures 1c, 2c and 3c]. Peri-follicular whitish halo was noted in 65 (54.1%) lesions [Figures 1c, 2c and 3c]. Less common finding was blue-grey dots/globules [Figures 3c, 4c and 6] and brown dots/globules/areas [Figures 1c, 2c, 4c, 5c and 6] observed in 15 (12.4%) and 40 (33.3%) lesions, respectively. Other dermoscopic findings observed were telangiectatic vessels [Figures 1c, 3c, 5c and 6], white scales [Figures 1c, 6], focal and structureless white areas [Figures 5c, 6], follicular red dots [Figure 3c] and white rosettes [Figures 1c, 4c]. The various dermoscopic patterns found in the study are depicted in Table 1.

Figure 1

Discoid lupus erythematosus in the active stage: (a) Clinical image. (b) Histopathology shows focal atrophy with hyperkeratosis and interface dermatitis. [H and E, 10×]. (c) Dermoscopy shows follicular plugs (red arrows), perifollicular halo (yellow arrows) and telangiectasia (blue arrow). Blackish pigmentation (yellow star) with follicular (green arrows), white rosettes (yellow box) and interfollicular (red stars) white scales are well appreciated

Figure 2

Discoid lupus erythematosus in the active stage: (a) Clinical image. (b) Histopathology shows lymphocytic infiltrate and focal fibrosis in peri-appendageal location and perifollicular extravasated erythrocytes. [H and E, 10×]. (c) Dermoscopy shows follicular plugs (red arrows), perifollicular halo (yellow arrows) and perifollicular erythema (red box) with red globules and red dots (black box). Perifollicular pigmentation (yellow stars), black dots (white box), follicular red dots (blue box) and white rosettes (yellow box) are well appreciated

Figure 3

Discoid lupus erythematosus in the active stage: (a) Clinical image. (b) Histopathology shows lymphocytic infiltrate and focal fibrosis in peri-appendageal location. [H and E, 10×]. (c) Dermoscopy shows follicular plugs (red arrows), perifollicular halo (yellow arrows) and telangiectasia (blue arrows). Blue-grey dots/globules (orange arrows) and follicular red dots (blue box) are well appreciated

Figure 4

Discoid lupus erythematosus in the inactive stage: (a) Clinical image. (b) Histopathology shows capillary dilatation with diffuse fibrosis. [H and E, 40×]. (c) Dermoscopy shows white rosettes (yellow box), blue-grey globules (orange arrows). Brown dots (white box) and areas (yellow star) with honey-comb pigmentation (white star) are well appreciated

Figure 6

Dermoscopy of discoid lupus erythematosus on the lip shows crusted yellowish scales (red arrows), linear (yellow arrows) and storiform vessels (yellow circles), white scales (red stars). Brown (yellow stars) and blue-grey globules (orange arrows) are noted. Pinkish-white globular strucutres are well appreciated (blue circles)

Figure 5

Discoid lupus erythematosus in the inactive stage: (a) Clinical image. (b) Histopathology shows follicular plugs and acanthosis. [H and E, 40×]. (c) Dermoscopy shows white areas (black stars) with few follicular plugs (red arrow) and unfocused vessels (blue circles). Note the adherent fibres (white arrows), which should not be confused with vessels

Table 1

Dermoscopic findings in patients with discoid lupus erythematosus

Dermoscopic features	Present (%)	Absent (%)
Follicular keratotic plugs	73 (60.8)	47 (39.1)
Perifollicular whitish halo	65 (54.1)	55 (45.8)
Telangiectatic vessels	54 (44.9)	66 (54.9)
White scales	50 (41.6)	70 (58.3)
Structureless white areas	44 (36.6)	76 (63.3)
Follicular red dots	44 (36.6)	76 (63.3)
Pigmentation	35 (29.1)	85 (70.8)
White rosette	30 (24.9)	90 (74.9)
Black or brown dots	40 (33.3)	80 (66.6)
Blue-grey dots	15 (12.4)	105 (87.4)

On histopathology, interface dermatitis [Figure 1b] and peri-appendageal infiltrate [Figures 2b, 3b] were seen in 85 (70.83%) and 80 (66.66) lesions, respectively. Perifollicular fibrosis and hyperkeratosis were noted in 40 (33.3%) lesions each [Figure 2b]. Other histopathological findings such as pigmentary incontinence [Figure 1b], dilated vessels [Figure 4b], fibrosis [Figure 4b] and extravasated red blood cells [Figure 2b] were found in variable frequency [Table 2]. Perifollicular white halo corresponded to peri-follicular fibrosis, whereas follicular plugging corresponded to follicular hyperkeratosis. Whitish scales, blue-grey dots and structureless areas corresponded to hyperkeratosis, pigment incontinence and dermal fibrosis, respectively. Dermoscopic findings and their histopathological correlation are shown in Table 3. It should be noted that dermoscopy helps in the visualisation of a lesion in vertical view and histopathology is in the horizontal view. Thus, discrepancy is expected when correlation of both features is analysed. In this study, dermoscopic patterns such as follicular keratotic plugs, telangiectatic vessels, white scales and structureless white areas were in perfect agreement with their histopathological changes by Cohen's kappa- coefficient analysis. Nevertheless, moderate agreement was found in peri-follicular whitish halo pattern [Table 4]. Some clinical and histopathological features were correlated with specific dermoscopic features and are represented in Table 5.

Table 2

Frequency of histopathological findings in patients with discoid lupus erythematosus

Histological changes	Present (%)	Absent (%)
Perivascular and appendageal lymphocytic infiltrate	80 (66.66)	40 (33.3)
Interface dermatitis	85 (70.83)	35 (29.1)
Peri-follicular fibrosis	40 (33.3)	80 (66.7)
Hyperkeratosis	40 (33.3)	80 (66.7)
Apoptotic keratinocytes	30 (24.9)	90 (74.9)
Thickening of Basement membrane	35 (29.1)	85 (70.8)
Follicular hyperkeratosis	80 (66.7)	40 (33.3)
Pigment incontinence	66 (54.9)	54 (44.9)
Atrophy	53 (44.1)	67 (55.8)
Dilated vessels	65 (54.1)	55 (45.8)
Dermal fibrosis	46 (38.3)	74 (61.6)
Extravasated erythrocytes	40 (33.3)	80 (66.7)
Mucin deposit	45 (37.4)	75 (62.5)

Table 3

Dermoscopic and histopathological correlation of discoid lupus erythematosus lesions

Dermoscopic features	Present	Absent	HPE finding	χ 2	P
Follicular plugging	73	47	Follicular hyperkeratosis (80/40)	0.88	0.17
Peri-follicular halo	65	55	Perifollicular fibrosis (40/80)	10.5	0.001
Telengiectasia	54	66	Dilated vessels (65/55)	2.0	0.07
White scales	50	70	Hyperkeratosis (40/80)	1.7	0.09
Structure-less whitish areas	50	70	Decreased pigmentation with fibrotic foci (46/74)	0.27	0.29
Follicular red dots	44	76	Extravasated erythrocytes (40/80)	0.29	0.30
White rosette	30	90	Infundibular hyperkeratosis (80/49)	34.54	0.001
Bluish-grey Pigmentation	10	8	Pigment incontinence (66/54)	0.02	0.48

Table 4

Analysis and interpretation of Cohen’s kappa coefficient for interobserver variation

Dermoscopic features	Histopathological changes	Cohen’s kappa coefficient (95%CI)	Interpretation
Follicular keratotic plugs	Follicular hyperkeratosis	0.87 (0.785-0.964)	Perfect Agreement
Peri-follicular whitish halo	Peri-follicular fibrosis	0.59 (0.465-0.724)	Moderate Agreement
Telangiectatic vessels	Dilated vessels	0.81 (0.717-0.919)	Perfect Agreement
White scales	Hyperkeratosis	0.82 (0.720-0.927)	Perfect Agreement
Structureless white areas	Fibrosis	0.93 (0.864-0.997)	Perfect Agreement

Table 5

Correlation of clinical and histopathological features with dermoscopic patterns

Activity of lesions	Clinical Features	Dermoscopic Features	Histopathological Features
Active	Scaly and erythematous plaques	Follicular plugging	Follicular hyperkeratosis
	Scaly and erythematous plaques	Perifollicular white halo	Perifollicular fibrosis
	Erythematous plaques or patch	Thick and thin arborizing vessels	Horizontally placed dilated capillaries
	Erythematous plaques or patch	Perifollicular erythema	Dilated capillaries around the follicles
	Erythematous plaques or patch	Follicular red dots	Extravasated erythrocytes and dilated vessels
Inactive	Depigmented patch	White area	Dermal fibrosis
	Depigmented patch	Thin arborizing vessels	Horizontally placed dilated capillaries
Both active and inactive	Scaly and erythematous, or depigmented plaques, or patches	White Scales	Hyperkeratosis
	Scaly and erythematous, or depigmented plaques, or patches	White Rosettes	Hyperkeratosis in the dilated infundibulum
		Pigment globules
	Erythematous, pigmented, or depigmented plaques or patches	Brown	Epidermal melanin
	Erythematous, pigmented, or depigmented plaques or patches	Blue	Dermal melanin or melanophages
	Erythematous, pigmented, or depigmented plaques or patches	Blue-white veil	Compact ortho-hyperkeratosis with heavy melanin in the dermo-epidermal junction and melanophages

Discussion

Dermoscopy is a non-invasive diagnostic modality used in pigmentary and inflammatory dermatosis. However, large-scale studies with histopathological correlation in skin of colour are lacking. Here, dermoscopic findings and histopathological correlation in Indian skin are discussed. DLE begins as a red-purple macule, papule or a plaque which develops hyperkeratotic surface classically. The scarring process starts with inflammatory cells invading the insertion of arrector pili in outer root sheath, also known as follicular bulge, marking the sites where follicular stem cells are located. Later, sebaceous gland, perifollicular sheath and interfollicular area are also damaged causing hyper or hypopigmented atrophic and fibrotic patches without follicular ostia.[13]

Follicular plugging was the most common finding in our study. It appeared like roundish whitish-yellow globular structures in the dilated follicular ostium and was seen in active lesions. It was demonstrated by dermoscopy even if thick scales were present. Plugging is due to follicular hyperkeratosis, which eventually leads to blocking of the follicular lumen.[14] Inactive lesions showed very few follicular plugs. It is in line with previous studies.[915] Thus, the presence of follicular plugs is suggestive of activity and has therapeutic implications in choosing the appropriate treatment.

Perifollicular white halo, as a whitish rim around the follicle, was the second most common finding observed in active lesions. Previously, it was described predominantly in non-scalp lesions.[15] However, we found it in three scalp lesions as well. It corresponded to perifollicular fibrosis in histopathology. This is similar with previous publications.[91415] Therefore, the absence of perifollicular halo implies inactivity of the lesion.

Vascular patterns were noted in 54 lesions and included telangiectatic/linear vessels, branching/arborizing vessels, dotted vessels, perifollicular erythema and red areas. Telangiectasias were thicker in active lesions and thinner in inactive lesions, especially in scalp lesions. The thickness of the vessel was compared to the width of an adjacent terminal hair.[16] Thick and thin arborizing vessels had diameter more and less than that of a hair, respectively. Arborizing vessels were noted predominantly in scalp lesions, and dotted vessels were more obvious in non-scalp lesions. Thin arborizing vessels across dilated follicular plugs gave a characteristic 'red spider on a yellow dot' pattern. Mucosal lesions revealed hairpin-like vessels and storiform pattern. This is in accordance with previous studies.[17]

Different morphology of vascular elements was found to be due to the anatomical location of lesions, variable degree of dilatation and location of vessels in the dermis. Dotted vessels correlated to tips of the tortuous and dilated capillaries in vertical arrangement. Technically, arborizing/linear vessels were suggestive of dilated capillaries that were placed horizontally parallel to the skin surface. Nevertheless, this pattern could not be visualised in histopathology.

Perifollicular erythema is another vascular element that was noted in acute lesions on the face and chest. This is in contrast to previous studies, wherein perifollicular erythema has not been described in non-scalp lesions.[18] This is probably attributed to inclusion of facial lesions in this study. Perifollicular erythema was due to dilated capillaries around the follicles and was suggestive of activity of the lesions.

Follicular red dots were seen in 44 active lesions in both scalp and non-scalp areas. They correlated with dilated infundibulum with keratin plug, surrounded by extravasated erythrocytes and dilated vessels. In non-scalp lesions, they represented extravasation of erythrocytes. Typically, red dots were not blanchable, unlike linear vessels. This was in line with other studies.[19] Although follicular red dots are characteristic of DLE, they are also found in normal scalp.[20] Thus, follicular red dots should be considered as diagnostic in the context of clinical features and with other dermoscopic patterns favouring DLE.

White scales were noted in a diffuse and perifollicular pattern. They represented hyperkeratosis. They were observed in acute and chronic lesions as well. This indicates that scales are not reliable dermoscopic pattern to know the activity of a lesion. Perifollicular scales were found commonly in the scalp lesions as compared to non-scalp lesions owing to higher density and length of hair in the scalp. This is in accordance with previous studies.[21]

Pigmentation was noted in various patterns such as honey-comb pattern, perilesional, speckled and perifollicular pigmentation. This was in line with other studies.[22] Brown pigmentation implied epidermal melanin, and grey-blue pigment was because of dermal melanin. Honey-comb pigmentation was due to heavy melanin in skin colour, which should be considered as normal finding. Speckled pigmentation appeared as dirty-brown or grey-blue dots/globules scattered over whitish areas. Earlier reports conferred that brown dots/globules and blue-grey globules are suggestive of activity and inactivity of the lesion, respectively.[2223] However, we observed pigmented globules both in active and inactive lesions in this study. This is due to robust interface dermatitis with melanin incontinence resulting in blue globules under dermoscopy. Furthermore, blue-grey dots/globules were noted in non-scalp lesions on the lip and chest in this study, which is in contrast to previous studies.[1624] They correlated with melanophages in the interfollicular area in histology.

White rosettes were found both in scalp and facial lesions. They were observed in both acute and inactive lesions. They corresponded to hyperkeratosis in the dilated infundibulum and result from an optical phenomenon under polarised light. They appear as four white dots meeting in the centre to give an appearance of a rosette resembling the leaf of a clover.[1225] Akin to scales, they do not impress upon the duration of given lesion. However, their number would reduce due to reduction in follicles in late stage. Although they are not specific to DLE, their presence was well correlated with histopathological features of infundibular hyperkeratosis with a P value of 0.001. Presence of white rosettes in patches of cicatricial alopecia on the scalp is almost definitive of DLE.[16]

A blue-white veil, an uncommon dermoscopic pattern in DLE, was noted. It appeared as bluish pigmentation over the entire lesion. It correlated with compact ortho-hyperkeratosis with heavy melanin in the dermo-epidermal junction and melanophages. This is similar to a previous study.[26] However, in this study, it was found in active and inactive lesions. This disparity can be explained on the basis of high melanin in the epidermis and dermis due to vigorous basal cell vacuolisation with melanin incontinence.

A white area was noted in 44 chronic lesions as focal or diffuse structureless area. It was found in both scalp and non-scalp lesions. It correlated with dermal fibrosis and was seen in inactive lesions. Nevertheless, inflammatory activity in older lesions imparted pinkish hue to the white area, resulting in pinkish-white globular structures. This was due to increased vascularity. Similar observations were mentioned in previous studies.[927]

To synopsise, interface dermatitis and peri-appendageal infiltrate in histopathology were suggestive of activity of DLE lesion. However, these changes could not be visualised under dermoscopy. On the contrary, dermoscopy revealed follicular plugs, perifollicular halo with pronounced vascular structures, including perifollicular erythema, follicular red dots and thick arborizing vessels in active lesions. Diffuse fibrosis with loss of appendages was found in inactive lesions and was noted as white structureless areas under dermoscopy. Interestingly, brown and blue-grey globules, white rosettes, blue veil and white scales, which respectively represented melanin (epidermis and dermis), follicular hyperkeratosis, compact ortho-hyperkeratosis and hyperkeratosis, were noted in both active and inactive lesions. In terms of variations of dermoscopic patterns in the skin of colour; follicular plugs appeared whitish-yellow rather than yellow, honey-comb pigmentation was intense, blue-white veil, if present, was strikingly conspicuous, and brown pigmentation was obvious in perilesional area.

Limitations

This study is a descriptive study with histopathological correlation. The main limitation of the study is the lack of an age-sex matched control group. Special stains such as periodic acid Schiff and alcian blue can also highlight histopathological changes with dermoscopic findings, but it was not done in the study. Moreover, serial sectioning of tissue was not carried out.

Conclusion

Dermoscopy is a novel-diagnostic tool in diagnosing DLE with accuracy in skin of colour and obviates the need for routine biopsy in most cases. Dermoscopy patterns were well correlated with histopathological changes. Thus, dermoscopy played an important role in assessing the activity of a lesion. This manuscript acts as a paradigm for future research in terms of prospective studies demonstrating lesion progression and response to treatment.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.