Histopathological finding of vesiculobullous lesions of skin in relation to their clinical presentation

Prospective study from a tertiary care center

Abstract

Background: 

Vesiculobullous disorders (VBDs) are extant with diverse clinical manifestations. Vesicles and bullae are fluid-filled cavities present within or beneath the epidermis. They are autoimmune blistering disorders in which autoantibodies are directed against target antigens present in the epidermis and dermo-epidermal junction.

Objective: 

Evaluation of the various clinicodemographic profile of patients with a pattern of distribution (subtypes) of VBDs of the skin and assess the association between clinical aspects and histological changes in vesiculobullous lesions of the skin.

Materials and Methods: 

The study material constituted 93 cases of VBDs out of 936 skin biopsies reported over two and a half years (January 2016 to June 2018) from the tertiary care center. A detailed history of the patients was taken, and a complete physical and dermatological examination with findings including clinical diagnosis was recorded. Histopathological examination (incisional/excisional/punch biopsy) was done in each case. The clinico-demographic evaluation was done and the results were correlated with histopathological findings.

Results: 

Vesiculobullous lesions constituted 10.06% of all skin biopsies. The majority of cases were of pemphigus vulgaris (PV) 30 (32.25%) followed by 16 (17.2%) of bullous pemphigoid. In 83 cases (89.24%) histopathology findings were consistent with clinical diagnosis. Out of 34 cases that were diagnosed clinically as PV, the histopathological study proved 30 cases (88.23%) as PV.

Conclusion: 

Vesiculobullous lesions of the skin are a heterogeneous group of disorders. It is essential to differentiate each pattern of subtype based on clinical examination and histopathological findings. Histopathological diagnosis with clinical correlation plays a major role in arriving at the diagnosis.

INTRODUCTION

Vesiculobullous diseases (VBDs) are a group of disorders in which the primary lesion is a vesicle or a bulla, on the skin or mucous membrane, or both. In most cases, the blisters (vesicles and bullae) are due to a primary disease, like pemphigus. As a secondary phenomenon, it also occurs in many unrelated conditions.[¹] Five principal ways which can result in blister formation are Genetic derangement, physical, immunological, inflammatory damage, and drug reaction.[²] Of all these causes, immunological reaction accounts for most of the dermatological disease-producing blisters.

Pathological evaluation of a blistering disorder involves systematic analysis, which includes the blister separation plane, the mechanism of blister formation, and the character of the inflammatory infiltrate, including its presence or absence. Among the various dermatological conditions, vesiculobullous lesions (VBLs) form one of the most frequent clinical problems. Though some of the VBLs have a characteristic clinical presentation, many times it may not be possible to make a definite diagnosis on physical examination alone. Therefore, a histopathological examination is essential for arriving at a diagnosis and classification. This study is undertaken to evaluate the clinical features and histopathology of various VBD of skin for the diagnostic potential and to correlate them for greater diagnostic accuracy.[³]

MATERIALS AND METHODS

The present study was conducted after approval from the institutional ethical committee in the Department of Pathology, Sri Aurobindo Medical College and Post Graduate Institute, Indore, Madhya Pradesh. The study design was a cross-sectional observational study. Out of 936 skin biopsies, 93 cases of VBDs were reported over two and a half years (January 2016 to June 2018) were included in the study.

For the present study, all the cases presenting with vesicobullous lesions in the outpatient department or inpatient department during the study period and underwent skin biopsy for histopathology were included in the study. Patients presenting with vesicobullous lesions but not undergoing skin biopsy, patients undergoing skin biopsy for skin lesions other than vesicobullous lesions, and inconclusive skin biopsies on histopathology were excluded.

A detailed clinical history and general and local examination with particular reference to the mode of onset, characteristic, and distribution pattern of the lesions was done. Informed written consent was obtained from these patients, before the biopsy. Ideal lesion sites were chosen with intact vesicles or bulla. Under local anesthesia, skin biopsies were taken using punch biopsy needles of 3–5 mm diameter, or an excisional biopsy was performed. One biopsy was of the lesion from an intact vesicle or bullae and the second biopsy was from the perilesional site (a few mm to 1 cm from the lesion preferably uninvolved skin). Intact vesicle or bulla was preserved in 10% buffered formalin and subjected to processing. Histopathological sections were evaluated after staining by Hematoxylin and Eosin technique.[⁴]

Staining

Hematoxylin solution (Harris’s)

Hematoxylin––2.5 g, absolute alcohol––25 mL, potash alum––50 g, mercuric oxide––1.25 g, distilled water––500 mL, and glacial acetic acid––20 mL.

Eosin

Eosin––1 g and distilled water––100 mL.

Staining procedure for histological section

Xylene––15 min; hydrate through 95% alcohol––2 min; 80% alcohol––10 dips; 70% alcohol––10 dips; 50% alcohol––10 dips; bring to water––10 dips; and stain in alum hematoxylin––15 min. Wash in running tap water. Differentiate 1% acid alcohol––3 to 4 dips. Wash in running tap water till sections blue. Ammonia––1 min. Stain in 1% eosin solution––8 to 10 min. Wash in running tap water––for 1 to 5 min. Dehydrate through 50% alcohol––10 dips, 70% alcohol 10 dips, 90% alcohol 10 dips, 95% alcohol 10 dips, Absolute alcohol 10 dips, slides were dried and mounted and further studied under a microscope along with clinical data. The histopathological diagnosis was made and reviewed by at least two consultant histopathologists. The separation plane of the blister, the mechanism of blister formation, and the character of the inflammatory infiltrate were observed. The results obtained were analyzed to derive the percent proportion of cases based on gender, age of presentation, duration of disease, site of occurrence, morphological diagnosis, level of vesicle formation, the content of blister, and mechanism of vesicle formation. Also, we have studied the specific histological features of different VBDs. The results were compared with previous similar studies.

Descriptive data were presented as percentages and frequencies. Data were collected, compiled, and analyzed using Microsoft Excel.

RESULTS

A total of 936 skin biopsies were received in the Department of Pathology in the study period, out of which 93 cases were diagnosed as VBLs (10.06%) with male predominance (59%). Table 1 shows the distribution of cases as per the type of various VBLs. The majority of cases were of pemphigus vulgaris (PV) 30 (32.25%) followed by bullous pemphigoid (BP) (17.2%) and dermatitis herpetiformis (DH) (7.52%). A maximum number of patients diagnosed with PV belonged to the age group 61–70 years (23.3%). BP (31.25%) and bullous drug eruption (BDE) (67%) were presented in the age group of 51–60 years [Figure 1]. Approximately 33% of the total cases had an erythematous base. PV (16.67%), BP (18.75%), pemphigus foliaceous (PF) (25%), erythema multiforme (EM) (33.34%), varicella (50%), and Stevens–Johnson syndrome (SJS) (33.34%) had an erythematous base. Nikolsky’s sign was present in 6.45% with PV (16.67%) followed by toxic epidermal necrolysis (TEN) (50%). Bulla spread sign (BSS) was seen in only PV (6.67%).

Table 1: Distribution of cases as per the type of various vesiculobullous lesions (VBLs)

Figure 1: Age and lesion-wise distribution of cases

Table 2 shows the distribution of various VBLs according to the morphology of secondary lesions. Approximately 100% of cases of TEN, epidermal bullosa aquisita (EBA), 73% PV, 67% SJS, and BDE showed erosion. Crusts were noted in 100% of cases of TEN, Herpes simplex 1 (HSV), and papulovesicular tuberculid (PT) followed by 75% of cases of PF. Plaque formation was seen in 100% of cases of HSV, Hailey–Hailey (HH), drug-induced exanthematous reaction (DIER), and insect bite reaction (IBR). Papule was noted in 100% of cases of HSV, vasculitis, IBR, and PT followed by 50% cases of PF and EM. Pigmentation was seen in 100% of cases of HH and Tinea with Id eruptions and 50% of cases of TEN. Nodules were shown in 16.6% of cases of EM and 14.3% of cases of DH. Scaling was seen only in 50% of cases.

Table 2: Distribution of various VBLs according to the morphology of secondary lesions

As seen in Table 3, the most common presentation of PV was over the trunk (56.67%) followed by upper limb (UL) (50%) and lower limb (LL) (33.34%), lesions all over the body (13.33%), and over the scalp (13.33%). Most of the patients with PF presented with lesions over UL, LL, and face (75%).

Table 3: Distribution of various VBLs according to site involvement

Table 4 shows histopathological and clinical concordance. Out of 34 cases that were diagnosed clinically as PV, the histopathological study proved 30 cases (88.23%) as PV. Out of 14 clinically diagnosed cases of BP, 13 cases (92.85%) were proven on histology while one case was reported as BSL with blistering. Three additional cases which were diagnosed as BP on histopathology were initially diagnosed clinically as PV, TEN, and BDE.

Table 4: Showing clinical and histopathological concordance

Table 5 and Figure 2 show the distribution of cases of the individual VBD according to the level of blister formation. All the patients of PV (100%) presented with suprabasal blister formation. PF showed subcorneal blistering (75%) while only one case (25%) showed supracorneal blistering. Subepidermal blistering was most commonly seen with BP (93.75%) and EM (66.66%).

Table 5: Distribution of cases of individual VBDs according to the level of blister formation

Figure 2: Distribution of vesiculobullous lesions according to the level of blister formation microscopy

Table 6 shows the content of the blister cavity, PV shown in 93.33% of cases of acantholytic cells, 73.33% of cases of neutrophils, 13.33% of cases-eosinophils and mixed inflammatory cells infiltrate, and 6.66% of cases of lymphocytes in the blister cavity. PF and BP showed predominantly neutrophilic infiltrate.

Table 6: Distribution of VBL according to contents of blister cavity on microscopy

PV showed acantholysis (90%), and a row of tombstone appearance (76.66%) as the major epidermal change followed by orthokeratosis (56.66%). PF showed most commonly acanthosis (100%) followed by spongiosis (75%). Acantholysis and parakeratosis were absent in PF. The most common feature in BP was acanthosis (87.5%) followed by orthokeratosis (75%) [Table 7].

Table 7: Distribution of types of VBL as per epidermal changes

PV showed perivascular infiltration in 93.33% of cases and dermal edema in 53.33% of cases. Approximately 30% cases showed melanin incontinence and only 3.33% cases showed adnexal infiltrate. PF showed 100% perivascular infiltrate and 75% cases with melanin incontinence. Only 25% of the patient showed dermal edema. Perivascular infiltrate was seen in 93.75% of cases of BP. Approximately 18.75% also showed adnexal infiltrate. Also, it showed dermal edema in 81.25% of cases and 31.25% of cases showed melanin incontinence. Papillary microabscess was a significant feature seen in DH (42.85% cases). Other than it only one case (6.25%) of BP showed this feature.

DISCUSSION

Autoimmune blistering diseases are a group of bullous disorders characterized by pathogenic antibodies directed at target antigens in the epidermis or dermo-epidermal junction. For confirmation of diagnosis, along with routine histopathological examination, immunofluorescence studies are essential. In the present study majority of the patients presented in the age group 31–40 years (age range 10–78 years) similar to studies conducted by Madhavi et al.[⁵] and Murthy et al.[⁶] The mean age was 43.18 years in this study. This is in concordance with a study conducted by Mittal et al.[⁷] (47.1 years). Males were affected slightly more than females similar to studies by Chanabasayaa et al., and Mohanty et al.[⁸⁹] However, some studies by Deepti et al. and Selvaraj et al. showed female predominance.[¹⁰¹¹] In this study, BP was common in an older individual with a mean age of 56.31 years which is similar to that reported by Selvaraj et al.[¹¹] The mean age of presentation in patients with PV in this study was 45.4 years similar to Mittal et al.[⁷] (42 years).

Spectrum of vesiculobullous disorders

In this study, PV was the most common VBD constituting 32.25% followed by BP (17.2%). This is similar to various other studies.[⁷⁹¹⁰¹²] Chanabasayaa et al. showed BP as the most common disorder followed by PV.[⁸] Most of the authors have not included disorders like vasculitis (V), traumatic bulla (TrB), IBR, tinea with Id eruption (T.ID), benign spindle lesion with blister (BSL), DIER, PT, epidermolysis bullosa simplex (EBS), and chronic dermatitis (CD) in VBL as in this study. These entities are often confused with immunobullous disorders clinically and can only be histologically differentiated. In this study, PV was the most common VBD constituting 32.35% followed by BP (17.2%) and PF (4.3%). The pattern of frequency of different types of VBD correlates closely to the study done by Deepti et al.,[¹⁰] whereas a study done by Chanabasayya et al. showed BP (46.15%) as the most common VBD followed by PV (18.68%).[⁸] The incidence of BDE, EM, DH, Varicella, and SJS were 3.2%, 6.45%, 7.5%, 4.3%, and 3.2%, respectively, in our study which correlates closely with other studies.[⁷¹⁰]

In the present study, oral mucosal involvement was seen in 46.67% cases of PV and 18.75% of BP, and no oral involvement in patients with PF, similar to the study conducted by Mittal et al.[⁷] Approximately 64.2%cases of PV and 10% of cases BP showed oral mucosal involvement. In the current study, Nikolsky’s sign was positive in 16.67% of patients with PV and 25% of patients with PF. This positivity rate was low in cases of PV and was comparable in cases of PF when compared to studies conducted by Deepti et al. and Arya et al.[¹⁰¹²] In our study, 54.83% of cases presented cutaneous blisters. In comparison to Deepti et al. and Gupta et al., it is slightly lower.[¹⁰¹³] This could be because of the inclusion of entities like CD, EM, and DH which less frequently presented with cutaneous blisters in the present study.

Table 8 shows an analysis of types of VBDs and their comparison with other studies. In our study, PV was the most common VBD constituting 32.35% (30 out of 93 cases) followed by BP constituting 17.2% (16 out of 93 cases) of the total cases of VBD. Pemphigus foliaceus constituted 4.3% of the cases. The pattern of frequency of different types of VBD correlates closely to study done by Mittal et al., Mohanty et al., and Deepti et al., which also showed PV as the commonest entity, whereas a study done by Chanabasayya et al. showed BP (46.15%) as the most common VBD followed by PV (18.68%).[⁷⁸⁹¹⁰]

Table 8: Analysis of types of vesiculobullous disorders

Pemphigus vulgaris

Table 9 shows a comparison of Histopathological features of PV with other studies. The present study showed PV as the major type of pemphigus (32.25%) followed by PF (4.2%) [Figure 3]. This was in concordance with studies conducted by Kannan et al. and Kanwar et al. but the incidence is relatively lesser in comparison.[¹⁴¹⁵] This could be because of the large sample size taken in the present study. All the patients had classical features of PV, that is, flaccid VBLs. Mucosal involvement was present at one time or the other during the disease course in 46.67% of the cases which is comparable to a study done by Chowdhury et al.[¹⁶] Oral mucosa was the most common site of mucosal involvement in PV similar to the study by Deepti et al.[¹⁰] The most common presentation of PV was over the trunk (56.67%) followed by UL (50%) and LL (33.34%). This was similar to the study done by Mittal et al. and Kushtagi et al.[⁷¹⁷]

Table 9: Comparison of histopathological features of pemphigus vulgaris with other studies

Figure 3: Photomicrograph––pemphigus vulgaris––roof, floor, and contents of blister (acantholytic cells and inflammatory cells) of the bulla and dermis showing perivascular infiltrate (10× H and E)

The main histopathologic features observed in cases of PV were acantholysis with a suprabasal blister which was seen in the majority of the patients. Acantholysis is the hallmark and a prerequisite for the diagnosis of pemphigus. Lever first described it and later on it was confirmed by several other authors like Handa et al. and Arya et al.[¹²¹⁸] Suprabasal bulla was present in 100% of our cases which is similar to the findings of Deepti et al. (88.2%), Arundhati et al. (93.2%), and contrary to Chowdhury et al. (62.5%) had relatively lesser incidence.[¹⁰¹⁶¹⁹]

Pemphigus foliaceous

Table 10 shows clinical features and histopathological findings in PF and its comparison with other studies. The disease was seen in four cases constituting 4.3% of the total cases of autoimmune VBD [Figure 4]. This is slightly lower compared to other studies.[⁸⁹¹⁰¹⁶] This could be because of the large sample size of this study. In PF cases the patients presented predominantly with extensive flaccid VBL (25%). The other predominant lesions seen were extensive crusts and erosions. Nikolsky’s sign was positive in 25% of cases and it was lower as compared to other studies.[⁷¹⁹]Mucosal membrane involvement was absent in PF and is similar to the study by Mittal et al. Subcorneal bulla were present in 75% of the cases which was similar to Mittal et al. and lower than the study by Arundhati et al. Acantholytic cells were seen in 25% cases which are lesser compared to other studies.[⁷¹⁹]

Table 10: Clinical features and histopathological findings in pemphigus foliaceous

Figure 4: Microphotograph pemphigus foliaceous showing subcorneal bulla with roof, floor, and contents (10× H and E)

Bullous pemphigoid

In the present study, BP constituted (16 cases) 17.2% of the study cases with an M: F ratio of 4.3:1 which is similar to studies by Murthy et al.[⁶] Subepidermal blister was seen in 93.75%of cases of BP similar to other studies[⁶⁷⁸⁹] [Figure 5]. Predominant inflammatory cells were eosinophils similar to studies that of Arya et al. and Kushtagi et al.[¹²¹⁷] The most common age group for BP was 51–70 years in the present study which was similar to that of Chanabasayya et al. and other studies.[⁷⁸¹⁰] There was a male predominance in the present study with a male-to-female ratio (M:F) of 4.3:1, which is similar to the study done by Chanabasayya et al. and against the observations made in the above studies.[⁷⁸¹⁰] Upper and lower extremities were the most common sites involved in BP which was similar to the study by Mittal et al. whereas, in a study by Deepti et al., lesions were present all over the body.[⁷¹⁰] Histopathological examination revealed subepidermal bulla in all the cases of BP. Predominant eosinophilic infiltration in the dermis was more common than predominant neutrophilic infiltration, which corresponds to the above studies.[⁸¹⁰¹⁹]

Figure 5: Bullous pemphigoid- microphotograph showing subepidermal blister with roof and floor of the blister (10× H and E)

Stevens–Johnson Syndrome

In the present study, we had three cases of SJS. The patients were in the first to the second decade of life with male predominance. This is similar to the study by Mittal etal. 66.6% (2/3) of cases had extensive mucosal involvement in form of conjunctivitis and generalized erosions which is similar to the study by other studies.[⁷] Histopathology showed the presence of subepidermal blister in 33.33% of cases. Blister contains necrolytic keratinocytes and a few neutrophils. Chung et al. study showed that keratinocyte apoptosis followed by necrosis is the pathogenic basis of the widespread epidermal detachment observed in SJS/TEN[²⁰²¹] [Figure 6].

Figure 6: Microphotograph (SJS) showing dermo-epidermal separation with adjacent epidermis showing acantholysis, spongiosis, and orthokeratosis with perivascular neutrophilic infiltrate in the dermis (10× H and E)

Dermatitis herpetiformis

There were 7 cases of DH in the present study. They most commonly presented with grouped erythematous papules and a few vesicles distributed predominantly over extensor aspects of the leg, thighs, and trunk. There was no history of gluten sensitivity. There was no mucosal involvement. Surrounding skin showed erythema and scarring. Histopathology showed subepidermal blister filled with neutrophils, dermis showing papillary microabscess and perivascular inflammatory infiltrate. Subepidermal bullae (71.42%)and papillary microabscess (42.85%) were present. In a study done by Deepti et al.[¹⁰] Approximately 100% of cases showed subepidermal bulla and papillary microabscess.

Erythemamultiforme

In the present study, we had six cases of EM, This was a slightly higher incidence compared to other studies.[¹⁰¹⁷] This may be due to the large sample size of our study. The present study showed an equal M:F ratio in EM cases, whereas Deepti et al. and Kushtagi et al. showed slight female predominance.[¹⁰¹⁷] In the present study, the most common age group affected in EM was 41–50 years similar to Kushtagi et al. and higher than the study by Deepti et al.[¹⁰¹⁷]This shows the geographical variation of the disease. In the present study, only 33.33% of cases presented with cutaneous blisters, and Nikolsky’s sign and BSS were absent, which was similar to the study by Deepti et al.[¹⁰]16.67% cases had oral mucosa involvement in our study whereas, in Deepti et al. study, oral mucosal involvement was 100%. This difference may be due to exacerbation and remission course of the disease. Subepidermal blisters were found in 66.66% of the cases. This was similar to the findings of Arundhati et al. (50%) whereas Kushtagi et al. showed subepidermal blisters in 100%.[¹⁷¹⁹]The present study showed Orthokeratosis (100%), acanthosis (66.66%), and spongiosis (66.66%) in the epidermis which was similar to other studies.[¹⁹] In the present study perivascular infiltrate and dermal edema were the most common dermal features, this was similar to the findings of Deepti et al.[¹⁰] [Figure 7].

Figure 7: Microphotograph of erythema multiforme showing dermo-epidermal separation with blister cavity showing acantholytic cells and mixed inflammatory cells (10× H and E)

Bullous drug eruption

Three cases with BDE were noted in the present study in the age group of 51–60 years which is slightly higher than Deepti et al. (30–49 years). Two were males and one was female. In a study done by Deepti et al., there was male predominance. Blister was seen in 66.67% and mucosal involvement was seen in 33.34%. Histopathology showed bulla in dermo-epidermal junction (66.66%) and perivascular infiltration was predominant. Blister showed predominantly acantholytic cells (66.66%) and eosinophilic infiltration (33.33%). Deepti et al. also had similar findings.[¹⁰]

Epidermal bullosa acquisita

In the present study, there was one case of EBA. A 20-year-old female presented with bulla over the upper and lower extremities. Histopathologically, there was a subepidermal blister with acanthocytes in the blister cavity. Mohanty et al. also had one case with a similar presentation. This is similar to a study done by Arundhati et al.[⁹¹⁹]

Varicella

In the present study, we had four cases of varicella, which was found to be the most common viral infection with a total incidence of 4.3%. M:F ratio was 1:1 and the mean age was 33.33 years. Mittal et al. had male predominance but the mean age (28.67%) was similar to the present study. Gupta et al. studied the pediatric age group and hence had a younger age group.[⁷¹³]All the patients presented with small vesicles all over the body. Crusting and erosions were the most common secondary lesions. This was similar to other studies.[¹³] In the present study, 33.33% of cases showed intraepidermal blisters which were lesser in comparison to Mittal et al. (100%). This may be due to the aging of the blister or due to the wrong selection of the area of the biopsy. Giant cells and acantholytic cells were the most common content of the blister. This was similar to other studies.[⁷]

Hailey–Hailey disease

In the present study, there was one case. A 60-year-old man presented with plaques. Histopathologically, there was an interspinous blister. This was different from Mohanty et al. and Arundhati et al. (suprabasal blister). Acantholytic cells were present in the blister cavity. This was similar to other studies. The epidermis showed acanthosis, spongiosis, and orthokeratosis. This was similar to the study done by Mittal et al.[⁷⁹¹⁹]

The limitation of our study is the small sample size due to time constraints, unavailability of the immunofluorescence technique, and socioeconomic status of the patients.

CONCLUSION

PV constituted the most common VBD in the present study. Clinical and histopathological concordance in the diagnosis of VBL was seen in approximately 90% of cases. Clinicopathological correlation is very important in the diagnosis of VBD rather than considering either of them alone. Even though clinical examination has its importance, both histopathological and immunofluorescence testing is a must as it confirms the diagnosis and subtyping. Considering the socioeconomic situations of the patients, and the unavailability of immunofluorescence techniques widely, histopathological diagnosis with clinical correlation still plays a major role in arriving at the diagnosis.

Ethical policy and institutional review board statement

Approval for this study was obtained from the Institutional Ethics Committee of Sri Aurobindo Medical College and Postgraduate Institute, Indore, Madhya Pradesh, India with letter no. SAMC/IEC/16/38 dated December 7, 2016.

Financial support and sponsorship

Not applicable.

Conflicts of interest

There are no conflicts of interest.