To the Editor: Vitiligo is an autoimmune depigmenting disorder. Vitamin B12 (B12) and folic acid (FA) supplementation can benefit vitiligo patients.1Don P. Iuga A. Dacko A. Hardick K. Treatment of vitiligo with broadband ultraviolet B and vitamins.Int J Dermatol. 2006; 45: 63-65Crossref PubMed Scopus (21) Google Scholar B12/FA deficiencies cause homocysteine (HC) elevations; supplementation reduces homocysteine. Homocysteine may mediate melanocyte destruction via increased oxidative damage, interleukin 6 production and nuclear factor κB (NFKB) activation. We examined the association between HC and vitiligo. Institutional review board (IRB)–approved chart review was conducted of patients with new-onset vitiligo (January 2003-July 2006) at St. Luke’s-Roosevelt Hospital. Complete blood cell count, B12, FA, and HC were collected. Body surface area (BSA) was estimated by using palm size, approximately 2 inches,2Shaker O.G. El-Tahlawi S.M. Is there a relationship between homocysteine and vitiligo? A pilot study.Br J Dermatol. 2008; 159: 720-724PubMed Google Scholar to estimate 1% BSA. Distribution and number of body parts involved were recorded. Categorical variables were examined by chi-square tests. All continuous variables were not normally distributed and were analyzed by non-parametric Mann-Whitney U or Kruskal-Wallis tests. Linear and logistic regression was performed for continuous and binary dependent variables, respectively. Log transformation was used for nonlinear data. Data were analyzed with SAS 9.2 Software (SAS Institute Inc., Cary, NC). HC levels were available for 56 patients; 46 also had B12 and FA. No vegetarians, or B12 (<300 pg/mL)/ FA deficiencies (<5.4 ng/mL) were noted. Linear regression models, controlling for age, gender, and ethnicity, found significant associations between HC and B12 (b = -0.01, P = .0003) (Fig 1), log vitiligo BSA (b = 0.16, exponentiated b = 1.17, P = .003) and number of vitiligo-affected body parts (b = 0.41, P = .004) (Fig 2).Fig 2Association between vitiligo extent and serum homocysteine levels. A, Body surface area, (B) number of body parts involved, (C) bilaterality of lesions, but not (D) distribution of lesions (dependent variables) are associated with serum homocysteine levels (independent variables). Linear regression models were constructed both as parsimonious models and as confounding models including age, gender, ethnicity, serum vitamin B12 levels, and supplementation with oral vitamin B12 and folic acid. Outliers were excluded from statistical analysis and are indicated as open circles. Regression coefficients for confounding models and P values are presented above, except for panel C, where results of parsimonious model are presented.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Logistic regression models, controlling for age, gender, ethnicity, B12, and oral B12/FA supplementation, found significant associations between HC and face and body (generalized) vitiligo lesions, compared with face or body alone (odds ratio [OR] = 12.00; 95% confidence interval [CI], 1.43-100.78; P = .02). Parsimonious logistic regression models found significant associations between HC and lesion bilaterality (OR = 1.21; 95% CI, 1.10-1.33; P < .0001) (Fig 2, C), but were no longer significant when age, gender, ethnicity, serum B12 levels, and B12/ FA supplementation were included in the model (P = .81). HC was not associated with type of generalized vitiligo (Wilcoxon signed-rank test, P = .16) (Fig 2, D). B12 and FA levels were not associated with BSA, distribution, body part number, or bilaterality of lesions (P = 0.43-0.89) nor supplementation of B12 and FA (P = 0.29-0.97). We demonstrate an association between serum HC and extent of vitiligo, suggesting HC as a new biomarker of vitiligo extent. Prior published studies of the association between vitiligo and HC found conflicting results. An age-matched case-controlled study of Egyptian patients demonstrated elevations of HC.2Shaker O.G. El-Tahlawi S.M. Is there a relationship between homocysteine and vitiligo? A pilot study.Br J Dermatol. 2008; 159: 720-724PubMed Google Scholar Validity of parametric tests for between-group comparisons was not assessed; groups were compared by using Pearson correlation, limiting the utility of the statistics and did not control for B12/FA. Extraordinarily elevated HC in cases and controls were not consistent with our population levels. Balci et al3Balci D.D. Yonden Z. Yenin J.Z. Okumus N. Serum homocysteine, folic acid and vitamin B12 levels in vitiligo.Eur J Dermatol. 2009; 19: 382-383PubMed Google Scholar performed a Turkish age/gender-matched case-control study, finding no association between HC, B12/FA, and vitiligo. Their study did not control for vitiligo duration, age, or gender. Differences between HC and vitamin levels may represent regional or ethnic variation. Nevertheless, the results of the present study demonstrate elevated serum HC levels in extensive vitiligo. We strongly recommend the inclusion of HC as a severity marker on initial examination for patients with vitiligo.
Read full abstract