Secondary Logo

Journal Logo

Original articles

Computerized image analysis of clinical and histopathological melanin changes in perilesional skin compared with normal skin in vitiligo

Khafagy, Naziha H.a; Diab, Heba M.a; Deban, Mohammed A.b; Aid, Shimaa M.a

Author Information
Journal of the Egyptian Women's Dermatologic Society: September 2013 - Volume 10 - Issue 3 - p 166-171
doi: 10.1097/01.EWX.0000433263.24970.32
  • Free

Abstract

Introduction

Vitiligo is an acquired pigmentary disease of the skin manifested by depigmented white patches surrounded by a normal or a hyperpigmented border 1. It affects up to 2% of the population, equally distributed in both sexes and may begin at any age, but in 50% of cases, it begins between the ages of 10 and 30 years 2. Vitiligo is a multifactorial disease with a complex pathogenesis; its precise cause remains unknown. Old theories include complete destruction of melanocytes because of the presence of autoantibodies, the ‘self-destruction’ theory 3. Other studies suggested intrinsic and extrinsic defects of melanocytes, rendering them as nonfunctioning cells 4.

Although vitiligo is categorized as a systemic disease and many studies have focused on vitiliginous skin, studies on perilesional and apparently normal distant skin remain few and controversial 5. Colour changes in these areas are deceivingly perceived as normal skin by the naked eye. The degree of the true involvement of the skin and size of the affected surface area influence strategies of available therapeutic options 6–8.

The aim of this study was to compare and correlate changes in skin colour and histopathological patterns in lesional, perilesional and distant apparently normal skin in vitiligo patients.

Patients and methods

The study included 40 vitiligo patients with different clinical types. They included patients with skin types III, IV and V according to Fitzpatrick’s classification. Patients were selected from the Outpatient Dermatology Clinic, Ain Shams University Hospitals, during the period from June 2010 to July 2011. Informed written consent was obtained from all patients and the study was approved by the Research ethical committee of the Faculty of Medicine, Ain Shams University.

The patients selected had nonsegmental vitiligo, and had experienced no activity of vitiligo disease (increased size or spread of vitiligo patches or appearance of new ones) for at least 6 months. Their vitiligo surface area involvement was examined using Wood’s light and was less than 15% according to the vitiligo body surface area of involvement estimated by the patient’s palm (1% body surface area) 9. Patients were divided according to disease duration into two groups.

Group 1 included 20 patients with disease duration of less than 1 year. Patients were new cases diagnosed for the first time as vitiligo patients.

Group 2 included 20 patients with disease duration of more than 1 year and less than 5 years. All patients did not receive any local treatment for at least 8 weeks or any systemic treatment for at least 4 months before the study.

All patients were subjected to a thorough assessment of history, complete general and dermatological examination and examination by Wood’ s light to detect vitiligo macules or patches not apparent to the naked eye.

Digital photography

Digital photos were taken for every patient documenting three areas, vitiligo patch, perilesional area and apparently normal distant skin 5 cm from any vitiligo patch, after Wood’s light examination to ensure that perilesional and apparently normal distant skin areas are not vitiliginous as perceived by naked eye examination. Photography was performed under standard conditions (fixed distance between the person and the camera, colour of the background, source of illumination). The same camera (Panasonic Lumix, DMC-FX10, 6.0 Megapixels; Sony Corp., Japan) was used throughout the study. All shots were taken at 3.0 Megapixels while disabling the flashlight. Photos were entered and analysed by computerized assisted digital image analysis.

Punch biopsy

From every patient, 3 mm skin biopsy specimens were taken from three areas: vitiligo patch, perilesional area and from apparently normal distant skin (5 cm from any vitiligo patch). A subcutaneous injection of 1 ml xylocaine 2% was administered. Biopsies were fixed in formalin and stained with Haematoxylin and Eosin (H&E) for routine histopathological examination and Fontana-Masson (F-M) stain for staining of melanin pigment. Three punch biopsies were also taken from normal individuals with the same skin types as our patients (III, IV and V) and stained similarly to compare the average melanin density in a normal individual with that in vitiligo patients. Stained slides were then analysed by computerized assisted digital image analysis.

Computerized assisted digital image analysis

The L*a*b* three-dimensional colorimetric system was used for assessment of the intensity of the constitutional skin colour. In this system, skin colour is correlated with L* (luminance value) and b* (degree of tanning) components and is assigned a value. The L* value corresponds to relative lightness ranging from total black (L*=0) to total white (L*=100). The b* value is an indicator of the tanning level; it represents the balance between yellow (positive value) and blue (negative value), showing an overall negative relationship with the L* value. However, the a* value was not used in our study as it refers to the erythema index and it is not relevant because of its sensitivity to external variations such as temperature and posture 10. Optima’s program was used for assessment of the pigmented area/epidermal area ratio (PA/EA ratio), which is an indicator for the degree of melanin pigmentation 11.

Statistical analysis

Data were coded, entered and processed into an SPSS software statistical computer package (version 12; SPSS Inc., Chicago, Illinois, USA). Descriptive data of patients were expressed as mean±SD for continuous data and as number and percentage for discrete data.

The Friedman test was used for comparison of nonparametric data to detect differences across multiple test values. The Kruskal–Wallis test was used to test equality of population medians among different groups. P less than 0.05 was considered the cut-off value for significance.

Results

This study included 40 vitiligo patients ranging in age from 20 to 65 years, mean±SD 32.5±15.2 years. They were 27 women (67.5%) and 13 men (32.5%). They included 12 patients (30%) with skin type III, 22 (55%) with skin type IV and six (15%) with skin type V according to Fitzpatrick’s classification.

For both values of L* and b* measured using computerized image analysis of clinical photos, there was a statistically significant difference between lesional skin and each of perilesional and distant skin in both groups 1 and 2 (P<0.05 for each; Table 1). Also, there was a significant difference on comparing perilesional with distant skin of group 2 (P<0.05), but there was no statistically significant difference between perilesional skin and distant skin in group 1 (P>0.05; Table 2).

Table 1
Table 1:
Comparison between lesional, perilesional and distant skin of group 1 in computerized image analysis of clinical photos
Table 2
Table 2:
Comparison between perilesional and distant skin of group 2 in computerized image analysis of clinical photos

Computerized image analysis of H&E-stained sections showed a decrease in basal pigmentation, with a statistically significant difference between the three examined areas of skin biopsy specimens in group 1 and a statistically significant difference between perilesional and distant skin in group 2, whereas all lesional skin in group 2 showed no basal pigmentation. In terms of comparison of sections of distant skin and sections of normal controls, there was a statistically significant difference in groups 1 and 2 (Figs 1 and 2) (P<0.05). Computerized image analysis of F-M-stained slides also showed the same relations: a statistically significant difference between the three examined areas in group 1 (P<0.05) and a statistically significant difference between perilesional and distant skin of group 2 in melanin reaction, whereas all lesional skin in group 2 showed no melanin pigment (P<0.05) (Tables 3 and 4 and Figs 3 and 4).

Figure 1
Figure 1:
Comparison between H&E-stained slides of a normal individual versus one patient from group 1 (both of skin type III) showing a gradual increase in basal pigmentation away from the lesional skin towards distant skin: (a) normal control skin, (b) vitiligo patient lesional skin, (c) perilesional skin and (d) distant skin of the same vitiligo patient (×400).
Figure 2
Figure 2:
Comparison between H&E-stained slides of a normal individual versus one patient from group 2 (both of skin type III) showing (a) normal control skin, (b) vitiligo patient lesional skin, (c) perilesional skin and (d) distant skin and showing a gradual increase in basal pigmentation away from the lesional skin towards distant skin (×400).
Table 3
Table 3:
Comparison between lesional, perilesional and distant skin of group 1 in computerized image analysis of F-M-stained slides
Table 4
Table 4:
Comparison between perilesional and distant skin of group 2 in computerized image analysis of F-M-stained slides
Figure 3
Figure 3:
Comparison between F-M-stained slides of a normal individual versus one patient from group 1 (both of skin type III) showing (a) normal control skin, (b) vitiligo patient lesional skin, (c) perilesional skin and (d) distant skin and showing a gradual increase in basal pigmentation away from the lesional skin towards distant skin (×400).
Figure 4
Figure 4:
Comparison between F-M-stained slides of a normal individual versus one patient from group 2 (both of skin type III) showing (a) normal control skin, (b) vitiligo patient lesional skin, (c) perilesional skin and (d) distant skin and showing a gradual increase in basal pigmentation away from the lesional skin towards distant skin (×400).

Discussion

During the evaluation process of vitiligo patients, several personal errors are difficult to distinguish by the human eye, even trained eyes, in good contrast because of slight colour variations and gradual colour changes 4,6. Most studies carried out on vitiligo examined histopathological changes alone 12 or colour changes alone 13, and none, to our knowledge, has combined and correlated both colour changes with histopathological changes in lesional, perilesional and apparently normal distant skin. In the present study, we used computerized assisted digital image analysis to evaluate colour changes in digital photos and to assess histopathological melanin changes in biopsies stained by H&E and F-M stains. The use of computerized digital image analysis provided an easy and inexpensive method in addition to its ability to discriminate even small colour differences. Also, its use eliminated the imprecise assessment of colour changes and the errors in different observers’ evaluations 5,13. Further, most studies neglected the effect of disease duration and degree of body surface area involvement. Therefore, we selected patients with limited body surface area (affection up to 15% only) and the two groups represent patients with different disease durations.

In the present study, on comparing the results of digital photos by computerized image analysis using the L*a*b* three-dimensional colorimetric system, there was a high statistically significant difference between lesional skin and both perilesional and distant skin in both groups and on comparison of perilesional and distant skin, there was a significant difference in patients with disease duration more than 1 year and less than 5 years. However, there was no significant difference in patients with disease duration of less than 1 year. Similar results were obtained in one study that considered the relationship between L* and b* values and melanosome size and its melanin content in apparently normal distant skin of vitiligo patients and reported a good linear correlation between L* values and melanosome size, whereas b* values were strongly linked to melanin content 5. Another study 13 compared the L* values and b* values of digital photos taken by a chromometer for vitiligo skin, perilesional skin and apparently normal distant skin and reported that the L* value decreases significantly in relation to increasing distance from the vitiligo spot in contrast to the b* value, which is gradually increased from lesional to perilesional to distant skin. However, these studies were carried out only on colour changes and did not correlate them with histopathological melanin changes.

Some authors studied perilesional skin using light and electron microscopy and reported vacuolization in keratinocytes, especially those of the basal cell layer, mononuclear cell infiltrate, degenerative changes in melanocytes and dermal melanophages 8,10,11. However, these studies focused only on the histopathological changes and excluded the degree of melanization. Melanocytes’ distribution and pattern were studied in vitiligo areas, perilesional and distant skin in comparison with normal skin using melanocyte-specific antibody. They concluded that melanocytes were completely depleted from the lesional vitiligo skin, absent or fragmented from the perilesional epidermis and were similar to those in normal controls in distant skin 7. The latter study did not consider disease duration. The results of H&E-stained sections in the present study showed a significant difference in the melanin content between vitiligo skin and both perilesional and distant skin in both groups (short and long disease duration). On comparing perilesional and distant skin, there was a significant difference in group 2, but there was no significant difference in group 1. Lesional skin of patients with disease duration of less than 1 year had residual scanty melanocytes whereas old lesions with 1–5 years’ duration showed no melanin. Similarly, F-M-stained biopsies showed the same results as H&E-stained biopsies.

In terms of the effect of disease duration on melanin pigmentation, almost all authors of earlier studies concluded that long-standing vitiligo patches, especially those more than 5 years’ duration, show a complete loss of melanin and absence of melanocytes from the basal epidermis 13. This is in agreement with our study, which showed that lesional skin of patients with disease duration of less than 1 year had a slight melanin reaction as detected by computerized analysis of clinical photos, H&E-stained and F-M-stained biopsies, whereas lesions of long disease duration showed no melanin reaction. Another study included an immunohistochemical analysis using a panel of 18 antibodies against melanocytes and studied only long-standing stable vitiligo, concluding that melanocytes are absent in all examined vitiligo patches 14. Their conclusion was not definite for the presence or absence of melanocytes in long-standing vitiligo lesions and could not reflect the actual histopathology of short-duration vitiligo patches nor perilesional or distant skin. Our results imply examination of vitiligo patients under Wood’s light for accurate clinical assessment and early medical intervention in vitiligo patches on first presentation because the more time lapses, the less the function of melanocytes will be and the less the probability of repigmentation.

Conclusion

Perilesional skin and apparently normal distant skin in vitiligo patients are involved in the same pathological process affecting lesional skin and reflected on both morphological and histopathological melanin patterns but with different degrees. Perilesional skin shows a moderate degree of these changes between lesional and distant skin. Therefore, it differs considerably when we consider vitiligo as a generalized disease, not merely a localized disease, irrespective of the size of the lesion.

Acknowledgements

Conflicts of interest

There are no conflicts of interest.

References

1. James WD, Berger TGJames WD, Berger TG, Eleston DM.Disturbances of pigmentation.Andrews’ diseases of the skin: clinical dermatology2006:10th ed..Saunders:Elsevier;223–225.
2. Klaus W, Richard AJ, Dick SKlaus W, Richard AJ, Dick S.Disorders presenting in skin and mucous membranes.Fitzpatrick dermatology atlas2007:7th ed..New York:McGraw-Hill Companies;453–459.
3. Mollet I, Ongenae K, Naeyaert JM.Origin, clinical presentation and diagnosis of hypomelanotic skin disorders.Dermatol Clin2007;25:363–371.
4. Le Poole IC, Luiten RM.Autoimmune etiology of generalized vitiligo.Curr Dir Autoimmun2008;10:227–243.
5. Pugashetti R, Maibach HFarage MA, Miller W, Maibach I.Pigmentation in ethnic groups.Text book of aging skin2010.Berlin Heidelberg:Springer-Verlag;503–508.
6. Tony B, Stephen B, Christopher GTony B, Stephen B, Christopher G, Neil C.Disorders of skin color.Rook text book of dermatology2004:7th ed..Oxford:Blackwell;39.4–59.8.
7. Ortonne JPBolognia J, Jorrizzo J, Rapini R.Vitiligo and other disorders of hypopigmentation.Dermatology2008:7th ed..New York:NY Mosbey;913–938.
8. Freedberg AZ, Eisen K, Wolff KF, Katz SIMosher DB, Fitzpatrick TB, Ortonne JB, Hori Y.Hypomelanosis and hypermelanosis.Fitzpatrick’s dermatology in general medicine1999:5th ed..New York:McGraw-Hill;945–1017.
9. Taïeb A, Picardo M.The definition and assessment of vitiligo: a consensus report of the Vitiligo European Task Force.Pigment Cell Res2007;20:27–35.
10. Hann SK, Park Y-K, Lee K-G, Choi EH, Im S.Epidermal changes in active vitiligo.J Dermatol1992;19:217–222.
11. Alaluf S, Atkins D, Barrett K, Blount M, Carter N, Heath A.The impact of epidermal melanin on objective measurements of human skin colour.Pigment Cell Res2002;15:119–126.
12. Kim YC, Kim YJ, Kang HY, Sohn S, Lee E-S.Histopathologic features in vitiligo.Am J Dermatopathol2008;30:112–116.
13. Brazzelli V, Muzio F, Antoninetti M, Villani S, Donadini F, Altomare A, Borroni G.The perilesional skin in vitiligo: a colorimetric in vivo study of 25 patients.Photodermatolo Photoimmunol Photomed2008;24:314–317.
14. Van Den Wijngaard R, Wankowicz-Kalinska A, Le Poole C, Tigges B, Westerhof W, Das P.Local immune response in skin of generalized vitiligo patients: destruction of melanocytes is associated with the prominent presence of CLA+ T cells at the perilesional site.Lab Invest2000;80:1299–1309.
Keywords:

computer image analysis; distant skin; perilesional; vitiligo

© 2013 Egyptian Women's Dermatologic Society