Advertisement

Is Hyperpigmentation in Systemic Sclerosis a Perivascular Dermal Tattoo?

  • Pauline Henrot
    Correspondence
    Corresponding author
    Affiliations
    University of Bordeaux, Inserm, BMGIC, UMR1035, Bordeaux, France

    Department of Rheumatology, National Reference Center for Systemic Autoimmune Rare Diseases, Hopital Pellegrin, Bordeaux, France
    Search for articles by this author
  • François Moisan
    Affiliations
    University of Bordeaux, Inserm, BMGIC, UMR1035, Bordeaux, France
    Search for articles by this author
  • Julien Seneschal
    Affiliations
    University of Bordeaux, Inserm, BMGIC, UMR1035, Bordeaux, France

    Department of Dermatology and Pediatric Dermatology, National Center for Rare Skin Disorders, Hôpital Saint André, Bordeaux, France
    Search for articles by this author
  • Alain Taïeb
    Affiliations
    University of Bordeaux, Inserm, BMGIC, UMR1035, Bordeaux, France

    Department of Dermatology and Pediatric Dermatology, National Center for Rare Skin Disorders, Hôpital Saint André, Bordeaux, France
    Search for articles by this author
  • Author Footnotes
    5 These authors contributed equally to this work.
    Muriel Cario
    Footnotes
    5 These authors contributed equally to this work.
    Affiliations
    University of Bordeaux, Inserm, BMGIC, UMR1035, Bordeaux, France

    Department of Dermatology and Pediatric Dermatology, National Center for Rare Skin Disorders, Hôpital Saint André, Bordeaux, France
    Search for articles by this author
  • Author Footnotes
    5 These authors contributed equally to this work.
    Marie-Elise Truchetet
    Footnotes
    5 These authors contributed equally to this work.
    Affiliations
    Department of Rheumatology, National Reference Center for Systemic Autoimmune Rare Diseases, Hopital Pellegrin, Bordeaux, France

    University of Bordeaux, CNRS, Immunoconcept, UMR 5164, Bordeaux, France
    Search for articles by this author
  • Author Footnotes
    5 These authors contributed equally to this work.
Open ArchivePublished:March 28, 2020DOI:https://doi.org/10.1016/j.jid.2020.03.937

      Abbreviation:

      SSc (systemic sclerosis)
      To the Editor
      Systemic sclerosis (SSc) is a rare but severe disease whose pathophysiology remains partly unknown, combining autoimmune features, vasculopathy, and systemic fibrosis (
      • Gabrielli A.
      • Avvedimento E.V.
      • Krieg T.
      Scleroderma.
      ). Vascular involvement is a major issue. Digital ulcers are a biomarker of disease severity and higher mortality (
      • Hughes M.
      • Herrick A.L.
      Digital ulcers in systemic sclerosis.
      ,
      • Meunier P.
      • Dequidt L.
      • Barnetche T.
      • Lazaro E.
      • Duffau P.
      • Richez C.
      • et al.
      Increased risk of mortality in systemic sclerosis-associated digital ulcers: a systematic review and meta-analysis.
      ); pulmonary hypertension and renal crisis are potentially lethal and silent at an early stage. Thus, better characterizing patients at risk is a crucial matter. In this context, we focused on pigmentary disorders in SSc, which affect about half of the patients and can impact their quality of life (
      • Jewett L.R.
      • Hudson M.
      • Malcarne V.L.
      • Baron M.
      • Thombs B.D.
      Canadian Scleroderma Research Group. Sociodemographic and disease correlates of body image distress among patients with systemic sclerosis.
      ). Recently, our group observed a link between the presence of diffuse hyperpigmentation and vascular involvement in SSc, especially digital ulcers (
      • Leroy V.
      • Henrot P.
      • Barnetche T.
      • Cario M.
      • Darrigade A.S.
      • Manicki P.
      • et al.
      Association of skin hyperpigmentation disorders with digital ulcers in systemic sclerosis: analysis of a cohort of 239 patients.
      ). Here, we reveal microscopic distribution of melanin in SSc skin among patients presenting hyperpigmentation or not.
      In this observational cross-sectional study, we characterized arm skin melanin distribution in 26 patients with SSc patients in a blinded manner. Written informed consent was obtained from all participants and the study was approved by the institutional ethical committee (CPP, 2012 A00081-42, Aquitaine). Patients with SSc were recruited in a case-control manner to compare melanin distribution between patients with hyperpigmented or non-hyperpigmented SSc. Demographic and disease characteristics of patients with SSc are presented in Table 1. Diffuse hyperpigmentation was present in 14 (54%) patients with SSc (Figure 1a), accompanied with hypopigmented macules in seven patients. The hyperpigmented group included a significantly higher proportion of diffuse cutaneous SSc forms as previously reported (
      • Leroy V.
      • Henrot P.
      • Barnetche T.
      • Cario M.
      • Darrigade A.S.
      • Manicki P.
      • et al.
      Association of skin hyperpigmentation disorders with digital ulcers in systemic sclerosis: analysis of a cohort of 239 patients.
      ), possibly because of the association between diffuse cutaneous SSc and dark phototypes (
      • Gelber A.C.
      • Manno R.L.
      • Shah A.A.
      • Woods A.
      • Le E.N.
      • Boin F.
      • et al.
      Race and association with disease manifestations and mortality in scleroderma: a 20-year experience at the Johns Hopkins Scleroderma Center and review of the literature.
      ); indeed, this group tended to include darker phototypes than the no pigmentary disorders group without reaching statistical significance (P = 0.0809), which is consistent with the literature (
      • Reveille J.D.
      • Fischbach M.
      • McNearney T.
      • Friedman A.W.
      • Aguilar M.B.
      • Lisse J.
      • et al.
      Systemic sclerosis in 3 US ethnic groups: a comparison of clinical, sociodemographic, serologic, and immunogenetic determinants.
      ). The proportion of patients with clinical vascular involvement was not statistically different between the two groups, which allows for the avoidance of a potential bias related to the previously reported association between hyperpigmentation and vascular involvement (
      • Leroy V.
      • Henrot P.
      • Barnetche T.
      • Cario M.
      • Darrigade A.S.
      • Manicki P.
      • et al.
      Association of skin hyperpigmentation disorders with digital ulcers in systemic sclerosis: analysis of a cohort of 239 patients.
      ).
      Table 1Characteristics of Included Patients with SSc
      CharacteristicsPatients with SSc (n = 26)P-value
      No pigmentary disorders (n = 12)Hyperpigmentation (n = 14)
      Demographic characteristics
      Age, years, mean (SD)60.58 (9.42)57.43 (9.57)0.4070
      Female, n (%)9 (75)6 (43)0.1302
      Phototype I–III, n (%)11 (92)8 (57)0.0809
      Phototype IV–VI, n (%)1 (8)6 (43)
      Disease characteristics
      DcSSc, n (%)3 (25)10 (71)0.0472
      P-value is significant (<0.05).
      Anticentromere, n (%)7 (58)4 (29)0.2329
      Anti–topoisomerase I, n (%)2 (17)6 (43)0.2164
      Anti–RNA polymerase III, n (%)0 (0)0 (0)NA
      Disease duration, years, mean (SD)5.58 (5.66)5.79 (5.49)0.9272
      Cutaneous involvement
      mRSS, mean (SD)10.75 (9.97)21.42 (11.85)0.0214
      P-value is significant (<0.05).
      Vascular involvement
      Digital ulcers, n (%)7 (58)9 (64)1.000
      PAH, n (%)2 (17)2 (14)1.000
      SSc renal crisis, n (%)0 (0)1 (7)1.000
      Pulmonary involvement
      ILD, n (%)5 (42)6 (43)1.000
      Pulmonary fibrosis, n (%)1 (8)2 (14)1.000
      Gastrointestinal involvement
      Esophageal involvement, n (%)9 (75)10 (71)1.000
      Intestinal involvement, n (%)1 (8)2 (14)1.000
      Musculoskeletal involvement
      Synovitis, n (%)4 (33.3)4 (29)1.000
      Joint contractures, n (%)2 (17)2 (14)1.000
      Potential photosensitizing therapies
      Phototherapy, n (%)0 (0)0 (0)NA
      Hydroxychloroquine, n (%)2 (17)6 (43)0.2164
      Nifedipin, n (%)2 (17)3 (21)1.000
      NSAID, n (%)1 (8)0 (0)0.4615
      Amiodarone, n (%)0 (0)1 (7)1.000
      Cyclines/quinolones, n (%)1 (8)0 (0)0.04615
      Other systemic therapies
      Methotrexate, n (%)5 (42)6 (43)1.000
      Immunosuppressive drugs, n (%)2 (17)5 (36)0.3913
      Abbreviations: DcSSc, diffuse cutaneous systemic sclerosis; ILD, interstitial lung disease; mRSS, modified Rodnan skin score; NA, not applicable; NSAID, nonsteroidal anti-inflammatory drug; PAH, pulmonary arterial hypertension; SSc, systemic sclerosis.
      Phototype: according to Fitzpatrick’s classification. Systemic therapies: current or past (whatever the duration). NSAID: if long-term treatment.
      1 P-value is significant (<0.05).
      Figure thumbnail gr1
      Figure 1Patients with SSc presenting hyperpigmentation and vascular involvement specifically exhibit melanin around dermal vessels. (a) Illustration of one patient with SSc with diffuse hyperpigmentation (male, 61 years old, <1 year of disease duration). Note the homogeneous hyperpigmentation even in non–sun-exposed areas, as well as the peculiar copper-like skin color. The patient gave informed consent for the publication of this image. (b) Brightfield image shows brown pigment deposit near a dermal vessel (red arrows) in SSc skin. CD31 staining indicates that the pigment is specifically located around dermal vessels but not in endothelial cells (CD31+, green; DAPI [cell nuclei], blue). Black (or white) line indicates the dermo-epidermal junction. Bar = 50 μm. (c) Fontana-Masson stain confirms the presence of melanin (black pigment) around dermal vessels (orange square). Bar = 50 μm. High-power view (×100 objective) of the dermal vessel shows perivascular melanin granules (red arrows). Bar = 20 μm. (d) CD68 staining shows melanin inside dermal macrophages or so-called melanophages (CD68+, green) (white arrow). Some cells exhibit melanin inside their cytoplasm but are CD68− (red arrow). Bar = 20 μm. (e) Flow-chart of repartition of patients with SSc in our cohort. Vascular involvement was defined as presence of DUs and/or pulmonary hypertension and/or renal crisis. Among patients with hyperpigmented SSc, the presence of this histological pattern is significantly associated with vascular involvement (P = 0.0063, chi-square test). DU, digital ulcer; NVasc, no vascular involvement; SSc, systemic sclerosis; Vasc, vascular involvement.
      In brightfield images, we observed a striking pattern of dark brown pigment deposit in the dermis of eight patients with SSc, particularly lining dermal vessels. We could not easily evidence any pigmented deposit in the superficial dermis as its thickness was greatly reduced (
      • Henrot P.
      • Moisan F.
      • Laurent P.
      • Manicki P.
      • Kaulanjan-Checkmodine P.
      • Jolivel V.
      • et al.
      Decreased CCN3 in systemic sclerosis endothelial cells contributes to impaired angiogenesis.
      ). CD31 staining allowing the identification of endothelial cells confirmed that pigment was specifically located near the outside layer of dermal vessels (Figure 1b). We did not find any evidence of perivascular pigment around lymphatic vessels (Supplementary Figure S1a). Fontana-Masson stain showing black pigment pointed out that the dermal pigment was at least partly constituted by melanin (Figure 1c), as supported by high-power view showing melanin granules (Figure 1c and Supplementary Figure S1b). Perls and Sudan black B stains were negative, ruling out hemosiderin and lipofuscin presence, respectively, in the perivascular pigmented deposit (Supplementary Figure S1c and d). Finally, bleaching with hydrogen peroxide completely washed out the perivascular pigment, confirming the melanin content (Supplementary Figure S1e). CD68 immunostaining identifying macrophages showed that, in some cases, the pigment was inside the cytoplasm of dermal melanophages, but not always (Figure 1d).
      Among patients with SSc, those that exhibited such a pattern of perivascular pigmentation all belonged to a subgroup associating hyperpigmentation and vascular involvement (defined as the presence of digital ulcers and/or pulmonary hypertension and/or renal crisis) (Figure 1e). Among patients with hyperpigmented SSc, the presence of this histological pattern was significantly associated with vascular involvement (P = 0.0063, chi-square test). Overall, the specificity of this histological pattern to indicate vascular involvement was 100% in our cohort; moreover, among patients with SSc with clinical hyperpigmentation, the sensitivity of the pattern to indicate vascular involvement was 80%. Demographic and disease characteristics of the eight patients with SSc that presented this pattern are presented in Supplementary Table S1.
      This finding raises several points. First, such dermal pigmentation could account for the persistent tanning that some patients report, maybe owing to the absence of pigment clearance once in the dermis (dermal tattoo); it could also explain the peculiar skin color (copper-like) observed in some patients. In this case, clinical hyperpigmentation would be explained by dermal pigmentation rather than increased epidermal pigmentation, and some samples even exhibited a marked decrease in epidermal pigmentation (Supplementary Figure S1f).
      Second, this histological pattern probably is not restricted to SSc skin, and dermal pigment is a common phenomenon among dark phototypes (
      • Ortonne J.P.
      • Bissett D.L.
      Latest insights into skin hyperpigmentation.
      ), including in the case of post-inflammatory hyperpigmentation, although the perivascular location has not been commonly described. However, in our case, dermal pigmentation was also observed in fair phototypes; moreover, such pigment deposit was only incidentally seen in a cohort of 26 healthy control skin samples matched on sample location and phototype. Overall, intensity of pigment deposit around dermal vessels was significantly higher in patients with SSc compared with healthy controls (Supplementary Figure S1g; P = 0.0337).
      Third, the specific perivascular localization of pigment deposit suggests, in an original way, a pathophysiological relationship between vasculature and pigmentation, as seen in other diseases such as melasma (
      • Passeron T.
      Long-lasting effect of vascular targeted therapy of melasma.
      ) and solar lentigo (
      • Hasegawa K.
      • Fujiwara R.
      • Sato K.
      • Park J.Y.
      • Kim S.J.
      • Kim M.
      • et al.
      Increased blood flow and vasculature in solar lentigo.
      ), and recently highlighted by our group in SSc (
      • Leroy V.
      • Henrot P.
      • Barnetche T.
      • Cario M.
      • Darrigade A.S.
      • Manicki P.
      • et al.
      Association of skin hyperpigmentation disorders with digital ulcers in systemic sclerosis: analysis of a cohort of 239 patients.
      ). This also raises the question of the origin of the pigment deposit. It is unlikely that the observed deposits are the result of drugs inducing dermal pigmentation, such as hydroxychloroquine; indeed, pigmentary changes usually differ clinically (bluish-green macules) and microscopically (pigmented deposit Perls-positive and in the superficial dermis) (
      • Jallouli M.
      • Francès C.
      • Piette J.C.
      • Huong du L.T.
      • Moguelet P.
      • Factor C.
      • et al.
      Hydroxychloroquine-induced pigmentation in patients with systemic lupus erythematosus: a case-control study.
      ), although a Perls-negative perivascular deposit has already been reported (
      • Puri P.K.
      • Lountzis N.I.
      • Tyler W.
      • Ferringer T.
      Hydroxychloroquine-induced hyperpigmentation: the staining pattern.
      ). However, and most importantly, of the eight patients that presented this pattern, three had no history of current or past drug uptake known to modify skin pigmentation, including hydroxychloroquine. In any case, the similar distribution suggests a shared mechanism, possibly triggered by a circulating factor.
      Thus, one could imagine an enhanced secretion of melanogenic factors by SSc endothelial cells, such as endothelin-1 (
      • Regazzetti C.
      • De Donatis G.M.
      • Ghorbel H.H.
      • Cardot-Leccia N.
      • Ambrosetti D.
      • Bahadoran P.
      • et al.
      Endothelial cells promote pigmentation through endothelin receptor B activation.
      ), combined with the loss of epidermal melanocytes because of an adhesion defect, subsequently falling into the dermis, similar to vitiligo (
      • Picardo M.
      • Bastonini E.
      A new view of vitiligo: looking at normal-appearing skin.
      ). Dermal melanin would then be phagocyted by macrophages (Figure 1d) migrating toward the nearest vessel. However, this hypothesis does not fully explain the presence of melanin inside the cytoplasm of CD68-negative perivascular cells. The nature of these cells remains to be elucidated; notably, one should consider the possibility of a local melanin production by transdifferentiation of perivascular cells.
      Altogether, perivascular melanin could help to detect vascular involvement, specifically among patients with SSc presenting clinical hyperpigmentation. However, this finding needs to be assessed in larger cohorts of patients with hyperpigmented SSc to validate the association with clinical vascular involvement. Moreover, the cross-sectional design of the study does not allow determining if the perivascular melanin deposit occurred before or after the development of the vascular injury. Prospective studies are warranted to address the possibility of using this histological pattern as a predictive marker. Despite various durations of pigmentary changes in patients with SSc, we did not see any correlation between the intensity of the pigmented deposit and disease duration (data not shown). Nevertheless, this does not refute the hypothesis that the pigmented deposit occurs early in the development of the disease. Future research should also focus on the mechanism of melanin deposit in the dermis, particularly this peculiar localization around dermal vessels, to further establish the link between hyperpigmentation and vasculopathy.

      Data availability statement

      No datasets were generated or analyzed during this study.

      Conflict of Interest

      The authors state no conflict of interest.

      Acknowledgments

      This work was funded by Association des Sclérodermiques de France and Société Française de Dermatologie . The authors wish to thank Dr Hugues Begueret, from the Pathology Department of Bordeaux University Hospital, for providing the Perls stain, as well as Catherine Pain for technical assistance, and all the clinicians and patients who took part in the SSc-VISS cohort.

      Author Contributions

      Conceptualization: PH, MET, MC; Formal Analysis: PH; Investigation: PH, MC, FM; Resources: MET, JS; Supervision: MET, AT, MC; Visualization: PH, FM; Writing - Original Draft Preparation: PH, MET, MC; Writing - Review and Editing: PH, FM, JS, AT, MC, MET

      Supplementary Material

      Figure thumbnail fx1
      Supplementary Figure S1Dermal pigment observed in the skin of patients with SSc is localized preferentially around dermal blood vessels and consists in melanin. (a) Double CD31/vWF immunostaining shows that melanin is located around capillaries rather than lymphatic vessels (all identified vessels were double positive). Representative image of one SSc sample is shown. Bar = 50 μm. (b) High-power view of H&E stain shows perivascular brown melanin granules. Bar = 20 μm. (c) Perls stain shows the absence of hemosiderin around dermal vessels in the skin of patients with SSc (iron, or hemosiderin, presence should be associated with a blue staining). Left panel, patient with SSc, bar = 50 μm; right panel, positive control (spleen), bar = 100 μm. (d) Black Sudan B stain shows the absence of lipofuscin in the perivascular deposit, which remains brown (red arrows). Black dots (black arrow) indicate a lipofuscin-positive cell in the same image. Bar = 20 μm. (e) Bleaching of a skin sample from a patient with SSc with 10% H2O2 for 2 hours shows complete clearance of the perivascular pigment, confirming the melanin content. Bar = 50 μm. (f) Fontana-Masson staining of one SSc sample. Black pigment (melanin) is spotted around one dermal vessel (white arrows), whereas no pigment is seen in the epidermis in this section. Bar = 50 μm. (g) Quantification of pigment deposit around dermal vessels in all SSc samples (n = 26) and in location- and phototype-matched HC samples (n = 26) shows that the intensity of pigment is significantly higher in patients with SSc than in HCs (P = 0.0337, unpaired t-test with Welch’s correction). Three representative brightfield images of dermal vessels per patient were analyzed and mean number of black pixels (after converting images to 8-bit format) is represented. HC, healthy control; SSc, systemic sclerosis; vWF, von Willebrand factor.
      Supplementary Table S1Characteristics of the Patients with SSc that Presented the Histological Perivascular Pattern
      CharacteristicsSSc patients with dermal perivascular pigment (n=8)
      Demographic characteristics
      Age, years, mean (SD)54.88 (8.53)
      Female, n (%)5 (63)
      Phototype I–III, n (%)4 (50)
      Phototype IV–VI, n (%)4 (50)
      Disease characteristics
      DcSSc, n (%)5 (63)
      Anticentromere, n (%)3 (38)
      Anti–topoisomerase I, n (%)4 (50)
      Anti–RNA polymerase III, n (%)0 (0)
      Disease duration, years, mean (SD)6.25 (5.42)
      Cutaneous involvement
      mRSS, mean (SD)20.5 (13.2)
      Vascular involvement
      Digital ulcers, n (%)7 (88)
      PAH, n (%)1 (13)
      SSc renal crisis, n (%)0 (0)
      Pulmonary involvement
      ILD, n (%)4 (50)
      Pulmonary fibrosis, n (%)2 (25)
      Gastrointestinal involvement
      Esophageal involvement, n (%)5 (63)
      Intestinal involvement, n (%)2 (25)
      Musculoskeletal involvement
      Synovitis, n (%)1 (13)
      Joint contractures, n (%)2 (25)
      Potential photosensitizing therapies
      Phototherapy, n (%)0 (0)
      Hydroxychloroquine, n (%)4 (50)
      Nifedipin, n (%)2 (25)
      NSAID, n (%)0 (0)
      Amiodarone, n (%)0 (0)
      Cyclines/quinolones, n (%)0 (0)
      Other systemic therapies
      Methotrexate, n (%)3 (38)
      Immunosuppressive drugs, n (%)3 (38)
      Abbreviations: DcSSc, diffuse cutaneous systemic sclerosis; ILD, interstitial lung disease; mRSS, modified Rodnan skin score; NSAID, nonsteroidal anti-inflammatory drug; PAH, pulmonary arterial hypertension; SSc, systemic sclerosis.

      References

        • Gabrielli A.
        • Avvedimento E.V.
        • Krieg T.
        Scleroderma.
        N Engl J Med. 2009; 360: 1989-2003
        • Gelber A.C.
        • Manno R.L.
        • Shah A.A.
        • Woods A.
        • Le E.N.
        • Boin F.
        • et al.
        Race and association with disease manifestations and mortality in scleroderma: a 20-year experience at the Johns Hopkins Scleroderma Center and review of the literature.
        Med (Baltim). 2013; 92: 191-205
        • Hasegawa K.
        • Fujiwara R.
        • Sato K.
        • Park J.Y.
        • Kim S.J.
        • Kim M.
        • et al.
        Increased blood flow and vasculature in solar lentigo.
        J Dermatol. 2016; 43: 1209-1213
        • Henrot P.
        • Moisan F.
        • Laurent P.
        • Manicki P.
        • Kaulanjan-Checkmodine P.
        • Jolivel V.
        • et al.
        Decreased CCN3 in systemic sclerosis endothelial cells contributes to impaired angiogenesis.
        J Invest Dermatol. 2020; 140: 1427-1434.e5
        • Hughes M.
        • Herrick A.L.
        Digital ulcers in systemic sclerosis.
        Rheumatology. 2017; 56: 14-25
        • Jallouli M.
        • Francès C.
        • Piette J.C.
        • Huong du L.T.
        • Moguelet P.
        • Factor C.
        • et al.
        Hydroxychloroquine-induced pigmentation in patients with systemic lupus erythematosus: a case-control study.
        JAMA Dermatol. 2013; 149: 935-940
        • Jewett L.R.
        • Hudson M.
        • Malcarne V.L.
        • Baron M.
        • Thombs B.D.
        Canadian Scleroderma Research Group. Sociodemographic and disease correlates of body image distress among patients with systemic sclerosis.
        PLoS ONE. 2012; 7e33281
        • Leroy V.
        • Henrot P.
        • Barnetche T.
        • Cario M.
        • Darrigade A.S.
        • Manicki P.
        • et al.
        Association of skin hyperpigmentation disorders with digital ulcers in systemic sclerosis: analysis of a cohort of 239 patients.
        J Am Acad Dermatol. 2019; 80: 478-484
        • Meunier P.
        • Dequidt L.
        • Barnetche T.
        • Lazaro E.
        • Duffau P.
        • Richez C.
        • et al.
        Increased risk of mortality in systemic sclerosis-associated digital ulcers: a systematic review and meta-analysis.
        J Eur Acad Dermatol Venereol. 2019; 33: 405-409
        • Ortonne J.P.
        • Bissett D.L.
        Latest insights into skin hyperpigmentation.
        J Investig Dermatol Symp Proc. 2008; 13: 10-14
        • Passeron T.
        Long-lasting effect of vascular targeted therapy of melasma.
        J Am Acad Dermatol. 2013; 69: e141-e142
        • Picardo M.
        • Bastonini E.
        A new view of vitiligo: looking at normal-appearing skin.
        J Invest Dermatol. 2015; 135: 1713-1714
        • Puri P.K.
        • Lountzis N.I.
        • Tyler W.
        • Ferringer T.
        Hydroxychloroquine-induced hyperpigmentation: the staining pattern.
        J Cutan Pathol. 2008; 35: 1134-1137
        • Regazzetti C.
        • De Donatis G.M.
        • Ghorbel H.H.
        • Cardot-Leccia N.
        • Ambrosetti D.
        • Bahadoran P.
        • et al.
        Endothelial cells promote pigmentation through endothelin receptor B activation.
        J Invest Dermatol. 2015; 135: 3096-3104
        • Reveille J.D.
        • Fischbach M.
        • McNearney T.
        • Friedman A.W.
        • Aguilar M.B.
        • Lisse J.
        • et al.
        Systemic sclerosis in 3 US ethnic groups: a comparison of clinical, sociodemographic, serologic, and immunogenetic determinants.
        Semin Arthritis Rheum. 2001; 30: 332-346