Confetti-like depigmentation: A potential sign of rapidly progressing vitiligo Juan Jes us Sosa, BS,a Sharif D. Currimbhoy, MD,a Uzoamaka Ukoha, BS,a Samantha Sirignano, BS,a Ryan O’Leary, MD,d Travis Vandergriff, MD,a Linda S. Hynan, PhD,b,c and Amit G. Pandya, MDa Dallas, Texas, and Los Angeles, California Background: Confetti-like depigmentation was noted in patients reporting recent worsening of vitiligo. Objective: We sought to determine if confetti-like depigmentation is a marker of rapidly progressing vitiligo. Methods: Review of patient records and images of patients from a vitiligo registry resulted in 7 patients with 12 images that fit inclusion criteria and were evaluated for percent depigmentation by 3 independent reviewers. The Vitiligo Disease Activity Score and the Koebner Phenomenon in Vitiligo Score in an additional cohort of patients with confetti-like lesions were compared with patients who had vitiligo without confetti-like lesions. Results: The mean percentage of depigmentation at baseline was 19.2%, which increased to 43.9% in images obtained at a mean of 16 weeks of follow-up. Vitiligo Disease Activity Score and Koebner Phenomenon in Vitiligo Score were significantly higher in the patients with confetti-like lesions compared with those without confetti-like lesions. A skin biopsy specimen of a confetti-like lesion in 1 patient revealed an inflammatory infiltrate in the papillary dermis with CD81 T cells localized to the dermoepidermal junction. Limitations: Small, single-center retrospective review and lack of full-body photographs are limitations. Conclusions: A confetti-like pattern of depigmentation may be a negative prognostic indicator for patients with rapidly progressing vitiligo. Further, prospective studies to evaluate this physical finding should be performed. ( J Am Acad Dermatol http://dx.doi.org/10.1016/j.jaad.2015.05.014.) Key words: depigmentation; leukoderma; melanocyte; pigmentation; skin of color; vitiligo.
itiligo has an unpredictable course and variable response to treatment. Determining the stability of vitiligo is important to establish prognosis and select appropriate treatment. Patient history is the most commonly used method to ascertain stability.1 The Vitiligo Disease Activity Score (VDAS) is a subjective scoring system developed by Njoo et al2 but it depends on patient recall and is subject to recall bias. Other markers of activity include the isomorphic (Koebner) phenomenon and trichrome vitiligo, also known as marginal
hypopigmentation.1-3 The Koebner Phenomenon in Vitiligo Score (K-VSCOR) is a recently developed tool to quantify the likelihood that vitiligo will develop at sites prone to trauma.4 Because the Koebner phenomenon and stability of vitiligo are linked, the K-VSCOR has been useful for clinicians to determine disease stability and activity, which can, in turn, guide therapy and estimate prognosis.4 Despite these tools, clinicians may still be uncertain as to the level of vitiligo activity in an individual patient and additional clinical markers would be useful in clinical practice
From the Departments of Dermatology,a Clinical Sciences (Biostatistics),b and Psychiatry,c University of Texas Southwestern Medical Center; and Department of Dermatology, David Geffen School of Medicine at the University of California, Los Angeles.d Funding sources: None. Conflicts of interest: None declared. Accepted for publication May 10, 2015. Reprints not available from the authors.
Correspondence to: Amit G. Pandya, MD, Department of Dermatology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390-9190. E-mail: [email protected]
utsouthwestern.edu. Published online June 5, 2015. 0190-9622/$36.00 Ó 2015 by the American Academy of Dermatology, Inc. http://dx.doi.org/10.1016/j.jaad.2015.05.014
2 Sosa et al
J AM ACAD DERMATOL
and for research purposes. Furthermore, patients use and no phototherapy in the previous 3 months. scored with the VDAS or K-VSCOR have not been Sporadic or inconsistent (mean # 3 times/wk) use of followed up longitudinally to correlate scores with topical tacrolimus and pimecrolimus was allowed, depigmentation over time. because this is unlikely to affect the course of vitiligo. We noted the presence of confetti-like macules In all, 77 patients without confetti-like lesions also fit on patients reporting recent rapidly progressing the inclusion criteria. Determination of the VDAS and vitiligo. The definition of confetti-like macules is K-VSCOR is routine for patients in the Dallas vitiligo the presence of numerous 1registry and was obtained in to 5-mm depigmented macall patients. Clinical and deCAPSULE SUMMARY ules in groups, usually at the mographic characteristics of borders of existing lesions the 2 groups of patients were The clinical course of vitiligo is and not predominantly in a compared, including age of unpredictable. follicular or perifollicular onset, gender, race, ethnicity, location. The purpose of Confetti-like depigmentation was found site of onset, family history of this pilot study was to investo be associated with rapidly progressing vitiligo, and history of autotigate this finding in a cohort vitiligo. immune diseases. of patients and its relationPatients with vitiligo who present with ship to disease activity over Statistical analysis confetti-like lesions may require more time. Part 1. Three evaluators aggressive therapy to avoid rapid reviewed the 12 images from depigmentation. METHODS 2 visits (baseline and followPart 1 up) for all 7 patients. For the A single-center, retrospective cross-sectional 2 patients with multiple lesions (one with 5 and the study of 178 patient records, including clinical other with 2), the percent depigmentation using the images, was performed after institutional review Vitiligo Area Scoring Index across lesions was board approval from the University of Texas averaged for each evaluator, time point, and image. Southwestern Medical Center. All patients seen as For the remaining 5 patients with 1 lesion, the outpatients at least twice in the last 2 years who had percent depigmentation for each evaluator by time photographic documentation of their lesions were point was included in the analyses. To examine the included. Fifteen patients with a confetti-like pattern effects of time, a repeated measures analysis of of depigmentation were identified. Seven fit the variance using factors of evaluator and time (baseline inclusion criteria for the study: no treatment in the and follow-up 1) was performed. Means and 95% previous year, baseline and follow-up images, and confidence intervals are reported. For 2 patients lack of treatment between visits. Images of 12 lesions with images at a second follow-up period from these 7 patients were evaluated for percent reviewed by 3 evaluators, 1 patient had 3 images depigmentation by 3 independent reviewers. Each evaluated and the other patient had 1 image lesion was scored from 0 to 100 using the percentage evaluated. The average percent depigmentation component of the Vitiligo Area Scoring Index.5 was calculated for the patient with 3 lesions by Because confetti-like macules have been described evaluator; this was then averaged with the patient with chemical leukoderma, information about expowho had 1 lesion. sure to cleaning solutions, hair dyes, or similar agents Part 2. Median and range is reported for VDAS on a consistent basis was obtained from each patient. and K-VSCOR. Mann-Whitney U tests were used to compare patients with and without confettilike lesions. This nonparametric test was used Part 2 because the assumptions of normality and equal All patients with nonsegmental vitiligo enrolled variance were violated. Software was used in all in the institutional review boardeapproved Dallas analyses (SPSS, Version 21, IBM Corp, Armonk, vitiligo registry at the University of Texas SouthNY); results were considered significant if P was western Medical Center were entered into this crossless than .05. sectional study. The inclusion and exclusion criteria for the Dallas vitiligo registry are broad and include every patient with a diagnosis of vitiligo who RESULTS consents to enrollment. A total of 141 patients were Part 1 reviewed, of whom 16 had confetti-like lesions on The areas of involvement in the 7 patients with skin examination. Of these, 15 patients fit the confetti-like lesions were the face, hands, arms, following inclusion criteria: no daily topical steroid and feet. The median time between baseline and d
Sosa et al 3
J AM ACAD DERMATOL
VOLUME jj, NUMBER j
Fig 1. Vitiligo of the hand with multiple small areas of confetti-like depigmentation, including nonfollicular locations.
Fig 2. Vitiligo of the hand 16 weeks later with extension of depigmentation in previous areas of confetti-like depigmentation. There are also new areas of confetti-like lesions at the periphery of the larger macules of depigmentation.
follow-up images was 16 weeks. The mean percent depigmentation at baseline was 19.2% (95% confidence interval 12.9%-25.5%), which increased to 42.9% (95% confidence interval 28.3%-59.4%) at 16 weeks, an increase of 23.7% (P = .0097). Two patients had additional follow-up images at 13 and 62 weeks, respectively, with mean depigmentation of 67.2%, an increase of 45.6% from baseline. Figs 1 and 2 are examples of a patient with confetti-like lesions. Exposure information was available for 5 of the 7 patients. One patient had exposure to home cleaning solutions on the affected areas of the hands and forearms whereas the other 4 reported no exposures at home or work. A biopsy specimen of a confetti-like lesion in 1 patient demonstrated absence of melanin and melanocytes in the affected area on histopathologic examination (Fig 3). Further staining revealed a mixed CD41 and CD81 infiltrate, with CD81 cells at the dermoepidermal junction (Fig 4). Part 2 The median VDAS in patients with confetti-like lesions was 4.0 (range: 2.0-4.0) compared with 1.0 (range: 1.0 to 4.0) in those without confetti-like lesions, a difference of 3.0 (P \ .0001). Median
Fig 3. Vitiligo. Fontana-Masson stain of confetti-like lesion of vitiligo showing lack of pigmentation. Arrows point to the edges of confetti-like lesion. (Original magnification: 3100.)
Fig 4. Vitiligo. Immunoperoxidase stain of vitiligo lesion for CD8 receptors showing CD81 T cells at the dermoepidermal junction in area where melanocytes are located. (Original magnification: 3100.)
K-VSCOR in patients with confetti-like lesions was 48.0 (range: 42.0-14.0) compared with a mean KVSCOR of 27.0 (range: 0.0-56.0) in patients without confetti-like lesions, a difference of 21.0 (P = .0001). There were no significant differences between the 2 groups in mean age, age of onset, gender, family history, site of onset, or presence of autoimmune diseases.
DISCUSSION Confetti-like and pinpoint depigmented macules have been described previously in association with chemical leukoderma.6 The authors indicated that the presence of confetti-like macules rules out vitiligo. The relationship to chemical exposure is supported by the finding of positive patch testing to paraphenylenediamine in hair dye in patients with leukoderma of the scalp.7 A detailed history with respect to chemical exposure was obtained in 5 of our patients. Four had no history of exposure, appearing to rule out chemical leukoderma. One patient reported exposure to cleaning liquids on the
4 Sosa et al
hands, which had confetti-like depigmentation. However, she had a similar pattern of depigmentation on the feet, neck, and face, where there had been no contact with cleaning solutions. Pinpoint depigmented macules in perifollicular skin were described by Menchini et al3 as a sign of worsening vitiligo. Our patients had lesions on areas of the skin with and without hair follicles, including the fingers (Figs 1 and 2). Some lesions were pinpoint in size but others were several millimeters in diameter. Therefore, we prefer the term ‘‘confettilike’’ depigmentation. On histopathologic examination of 1 lesion, a lymphocytic infiltrate associated with lack of melanin and CD81 T cells at the dermoepidermal junction was observed, which is consistent with recent investigations into the pathogenesis of vitiligo.8 These studies have shown activation of the innate immune system in patients with worsening vitiligo, leading to antigen-specific CD81 T cells that destroy melanocytes. The presence of CD81 T cells adjacent to melanocytic areas of the epidermis in the biopsy specimen of a confetti-like lesion in one of our patients is consistent with this model. A biomarker that accurately reflects vitiligo activity would be a welcome advance for clinicians. Meanwhile, the VDAS and K-VSCOR will continue to be the most widely used scoring methods. The VDAS relies on patient history to determine the activity of the disease and ranges from 1 (stable disease with improvement) to 4 (new/expanding vitiligo lesions in past 6 weeks). Patients with confetti lesions in this study had a higher VDAS score and a higher K-VSCOR compared with those with nonconfetti lesions, supporting the likelihood that confetti lesions are a marker of unstable, progressive vitiligo. A limitation of the study was lack of full-body photographs of all patients to fully assess progression of confetti-like lesions. In addition, the study is small and requires more patients from a variety of backgrounds for confirmation. More biopsies should be performed on confetti-like lesions to confirm our findings in 1 patient.
J AM ACAD DERMATOL
Vitiligo has been associated with several autoimmune diseases, particularly thyroid disease.9 It is possible that confetti-like lesions, as a marker of rapidly progressive vitiligo, could be associated with autoimmunity against other organs. However, the size of the study was too small to determine if such a relationship exists. The results of this pilot study suggest that confettilike lesions may be a negative prognostic sign in patients with vitiligo, indicating rapidly progressing disease. They may also be a sign of inflammatory vitiligo. If these findings are confirmed in larger cohorts and other centers, consideration should be given to aggressive treatment of patients with confetti-like lesions to prevent rapid and potentially permanent loss of pigmentation. REFERENCES 1. Sahni K, Parsad D. Stability in vitiligo: is there a perfect way to predict it? J Cutan Aesth Surg. 2013;6:77-82. 2. Njoo MD, Das PK, Bos JD, Westerhof W. Association of the Koebner phenomenon with disease activity and therapeutic responsiveness in vitiligo vulgaris. Arch Dermatol. 1999;135: 407-413. 3. Menchini G, Comacchi C, Cappugi P, Torchia D. Depigmentation patterns of nonsegmental vitiligo. Am J Clin Dermatol. 2013;14:55-59. 4. Diallo A, Boniface K, Ezzedine K, et al. Development and validation of the K-VSCOR for scoring Koebner’s phenomenon in vitiligo/non-segmental vitiligo. Pigment Cell Melanoma Res. 2013;26:405-407. 5. Hamzavi I, Jain H, McLean D, et al. Parametric modeling of narrowband UV-B phototherapy for vitiligo using a novel quantitative tool: the vitiligo area scoring index. Arch Dermatol. 2004;140:677-683. 6. Ghosh S, Mukhopadhyay S. Chemical leukoderma: a clinico-etiological study of 864 cases in the perspective of a developing country. Br J Dermatol. 2009;160:40-47. 7. Taylor JS, Maibach HI, Fisher AA, Bergfeld WF. Contact leukoderma associated with the use of hair colors. Cutis. 1993;52:273-280. 8. Rashighi M, Agarwal P, Richmond JM, et al. CXCL10 is critical for the progression and maintenance of depigmentation in a mouse model of vitiligo. Sci Transl Med. 2014;6: 223ra23. 9. Silverberg JI, Silverberg NB. Clinical features of vitiligo associated with comorbid autoimmune disease: a prospective survey. J Am Acad Dermatol. 2013;69:824-826.