Portal de Periódicos da CAPES

Carta Acesso aberto Produção Nacional Revisado por pares

BIBTEX RIS

Prevalence of vitiligo in Brazil—A population survey

2017; Wiley; Volume: 31; Issue: 3 Linguagem: Inglês

10.1111/pcmr.12681

1755-148X

Caio César Silva de Castro, Hélio Amante Miot,

Olfactory and Sensory Function Studies

The prevalence of vitiligo is known to vary with age, gender, and ethnic groups. It has its onset mostly during the 2nd and 3rd decade of life and affects females with a slight predilection over males. Although found globally, vitiligo tends to be more frequent among Asians and Africans (Tarle, Nascimento, Mira, & Castro, 2014; Zhang et al., 2016). In Mexico, a multicentre survey with 50,000 individuals estimated the prevalence to be 0.21% (Estrada Castañón et al., 1992), but data for South America are relatively sparse. A small Brazilian study revealed self-reported vitiligo in one case among 515 adult workers (Ishiy, Silva, Penha, Handel, & Miot, 2014), and an examination of 9,955 schoolchildren from Amazonia resulted in a prevalence of 0.04% (Bechelli et al., 1981). It is important to note, however, that prevalence data do not only depend on study population but also on study methodology. For instance, a meta-analysis has shown a pooled prevalence of 1.8% from 22 hospital-based studies but only 0.2% from 82 community-based studies, with the latter showing evidence of publication bias (Zhang et al., 2016). Given the sparsity of data for Brazil, we initiated a study covering a random sample of individuals selected from 87 Brazilian municipalities with more than 300,000 inhabitants each, representing together a total of close to 40% of the Brazilian population. To obtain a representative sample, we randomly selected residential phone numbers and called them between 9 a.m and 8 p.m from Monday–Saturday during the months of January–June 2017 (Romiti, Amone, Menter, & Miot, 2017). Of those who could be reached, about 85% were willing to participate in the interview. Heads of households were asked whether any member had been diagnosed with vitiligo by a physician. We reasoned that by asking the question in this way, the diagnostic accuracy of self-reported vitiligo was high, with the caveat, of course, that the data might be influenced by the density of physicians, particularly dermatologists. This is the reason why we selected the larger, mostly urban municipalities (they tend to have a higher density of dermatologists compared to smaller, more rural municipalities) and why we correlated our results also with the density of dermatologists in a municipality. Interviews were completed for a total of 6,048 residences, representing 17,004 inhabitants. Prevalence data were analyzed according to city latitude, density of dermatologists and proportion of self-reported race. Vitiligo was reported in 97 residents (prevalence: 0.57%; CI 95%: 0.46%-0.68%), with 49 (50.5%) of them being females, and ranging in age from 28 to 68 years (with a mean of 48 years). As shown in Figure 1, the prevalence was 0.52% (CI95%: 0.39%-0.66%) for women and 0.63% (CI 95%: 0.46%-0.83%) for men, although this difference was not statistically significant (p = .32). Likewise, when stratified by region, no significant overall differences were found (p = .30). There was, however, a significant progressive increase with age (p < .01): below the age of 30, the prevalence was 0.34% (CI95%: 0.20%-0.49%); between 30 and 60, it was 0.69% (CI95%: 0.50%-0.88%); and over 60, it was 0.85% (CI95%: 0.54%-1.2%). Bivariate and multivariate analyses of the prevalence of vitiligo according to dermatologic workforce, self-declared race, and latitude of each municipality were then performed using a generalized linear regression model (for methodological details, see Table 1 legend). This analysis revealed a clear and direct association of vitiligo prevalence with self-declared race and latitude, whereby low latitude and non-European and Amerindian races showed higher prevalence. The following considerations may help to appreciate these findings. The increase in the number of cases with age, for instance, may simply reflect the chronic nature of the disease and the time needed to establish a correct diagnosis. Hence, the current increase in life expectancy should alert health authorities to the impact of vitiligo for public health policies. The differences in prevalence according to latitude reached statistical significance only when adjusted by race and may reflect the overall lifetime sun exposure, with southern latitudes being a proxy for lower exposure. As ultraviolet radiation is immunosuppressive and can lead to repigmentation of vitiligo, it is conceivable that lower sun exposure may be associated with a higher frequency of manifest disease. We may also see the regional differences to be due to the ethnic composition of the surveyed population because the different regions of Brazil were differently colonized by Europeans, Amerindians, and Africans. Vitiligo prevalence was clearly associated with non-European self-declared race, especially Amerindian. Consistent with this observation, vitiligo prevalence has previously been found to be higher in dark-skinned individuals (Sheth, Guo, & Qureshi, 2013). Low prevalence among Europeans is reinforced by large studies from Italy (0.17%), France (0.28%), and Denmark (0.12%), while higher prevalence was found in the United States (0.49%), Mexico (0.21%), Egypt (1.22%), Iran (0.60%), India (1.13%), China (1.9%)], and Australia (1.2%) (Kruger & Schallreuter, 2012). To our knowledge, this is the first study suggesting an association between Amerindian ancestry and vitiligo. Interestingly, Amerindian ancestry also affects the frequency of other diseases. For instance, Amerindian ancestry is associated with an increased frequency of lupus erythematosus (Alarcon-Riquelme et al., 2016), but Amerindians have a lower risk of contracting leprosy, psoriasis, or Alzheimer's disease (Benedet et al., 2012; Garcia et al., 2013; Parisi, Symmons, Griffiths, & Ashcroft, 2013). Also, they differ from Europeans in 18 adaptive immune response genes (Lindenau et al., 2013). Additional studies are required, however, to assess the relative contributions of genetic and epigenetic (e.g., lifestyle) causes for these different disease frequencies. Our finding of an association of genetic ancestry and latitude would seem to warrant further studies to reliably estimate the prevalence of vitiligo in all of Brazil and all other Latin American countries, whereby future studies should also take into account the clinical subtypes of vitiligo. In fact, we are currently performing a detailed interview of the 97 vitiligo patients identified in this survey. We believe that because vitiligo significantly affects the quality of life and incurs substantial economic burden on healthcare systems (Elbuluk & Ezzedine, 2017), such prevalence studies provide an invaluable resource for physicians and public health officials alike. Caio Castro and Helio Miot declare no conflict of interests.