Familial aggregation of vitiligo in the French West Indies (Isle of Martinique)

ARTICLE

Vitiligo is a skin disorder characterized by the destruction and/or inactivation of melanocytes, resulting in patches of depigmentation. Vitiligo affects individuals of all ethnic origins with an estimated prevalence of 0.5% of the world population. Genetic factors are thought to be involved in the ethiopathogenesis of vitiligo, since familial history has been reported in 20 to 30% of cases [1]. Genetic models for vitiligo including multiallelic autosomal transmission have been proposed [2]. However, the genetic transmission of vitiligo follows a multifactorial pattern, probably involving genes that predispose to the expression of the disease after interaction with environmental factors [3]. The genetic dissection of complex traits that do not exhibit the classic Mendelian inheritance attributable to a single gene locus is a difficult task [4]. Before undertaking long and expensive DNA-based studies, there is a need for studying the pattern of the disease's incidence in families and populations. It is necessary for segregation analysis to fit a general model for the inheritance pattern of traits in pedigrees [4]. In the case of vitiligo, most families only exhibit two affected members and multiplex families with more than 3 affected members are the exception [5, 6], thus rendering segregation analysis difficult. Pedigree studies concerning the largest number of families possible are needed to avoid the risk of developing impaired models. Moreover, the influence of environmental factors should be explored, and one possible means could be to study families of different extraction, with different environmental risk factors. Most family studies on vitiligo were performed in occidental countries or in India, mainly because the prevalence in other areas was unknown [7, 8]. Our purpose was to study the familial aggregation of vitiligo in the Isle of Martinique (French West Indies), where the prevalence of vitiligo has been recently demonstrated to be close to 0.34%, with 34% of family history [9].

Materials and methods

Geography and population

The Isle of Martinique is located in the Caribbean area, at 14.5°N and 63°W. The population was estimated to be 359,579 in 1992 [10]. Ninety-six per cent of the population is of African-European descent (black Carribeans), and less than 4% is of European descent (whites).

Family data sources

Families were ascertained through a single affected proband. The probands were affected individuals with generalized vitiligo, who presented spontaneously at consultation rooms and medical wards of the Dermatology Department of the Centre Hospitalier Universitaire de Fort-de-France (Martinique, French West Indies), between October 1995 and October 1999. Family data were collected by questionning the probands. Clinical examination of nuclear families was performed by a dermatologist. Information was taken regarding sex, current age and age of onset of vitiligo of proband and affected relatives, in a set of relatives including children, grandchildren, spouse(s), parents, siblings, grandparents, uncles and aunts, nephews and nieces, and cousins. There was no occurrence of two probands in the same family. No patient had segmental vitiligo.

Controls

Data looking at sex and age of onset of vitiligo, were compared to a control population of 36 patients with sporadic generalized vitiligo, recruited from the Dermatology Department of the Centre Hospitalier Universitaire de Fort-de-France (Martinique, French West Indies), during the same period of time. Sporadic vitiligo was defined by the absence of known cases in the family history and the absence of clinical lesions of vitiligo in the nuclear family after examination by a dermatologist.

Environmental and trigggering factors

Patients with familial and sporadic vitiligo were questionned about triggering and environmental factors, including professionnal activity, exposure to chemicals or physical injuries, leisures and sport activities, thyroid diseases, stress and emotional factors. Data were compared between the two populations using the khi² test.

Epidemiological data

Prevalence of vitiligo in the affected families was determined by the ratio of the number of affected individuals and affected relatives over the total number of relatives in the studied families. It was compared to the prevalence in the general population by the khi² test. As mentioned earlier, the prevalence of vitiligo in the general population from which the families were sampled is 0.34% [9]. The sex ratio was the male/female ratio. The sex and age of onset in family cases were compared by the khi² test to data in sporadic cases.

Results

Data were collected and analyzed from 16 families. The total number of relatives in the 16 families was 546, with 38 vitiligo affected subjects. The prevalence of vitiligo among relatives of patients was 7%. The difference with the prevalence in the general population in Martinique (0.34%) was highly significant (p < 0.001). Sex ratio was 0.65 (60% of females) in probands and affected relatives and 0.44 (70% of females) in controls (insignificant difference). The age of onset of vitiligo was 31 years (ranging from 2 to 53 years) in probands and affected relatives and 33 years (ranging from 1 to 72 years) in controls (insignificant difference). Vitiligo occurred before the age of 20 in 13 controls (sporadic vitiligo) (36%) and in 3 patients among probands and affected relatives (19%). The age distribution of onset in probands, affected relatives and controls is shown in Figure 1. The number of affected relatives among families and controls is shown in table I. The detailed informative pieces about affected relatives are presented in table II. No significant difference was observed in environmental and triggering factors between familial and sporadic vitiligo. Stress and emotional factor was noticed as a precipitating or worsening cause by 55% of patients.

Discussion

The clear preponderance of vitiligo among relatives of patients (7%) as compared to the prevalence in the general population (0.34%) was previously reported in other countries and confirms the involvement of genetic mechanisms in the pathophysiology of the disease [11]. One of the major problems in forming concrete genetic models in vitiligo is the lack of multiplex families, as most families, ascertained through probands, have no more than 2 affected members [2, 8]. Our data in Martinique, with only 75% of families exhibiting two affected relatives, are in agreement with such observations. Moreover, some relatives could carry the predisposing alleles and not be affected by the disease because of a lack of exposure to triggering factors [2]. The assumption that environmental factors may differ from one area to another and so can affect the phenotypal expression of the disease, may lead to changes in the proportion of familial cases among different countries. But, in fact, the observations performed in our country did not differ from the literature data in other areas and failed to show any differences in environmental or triggering factors in sporadic or familial cases. Nevertheless, healthy relatives, other than belonging to the nuclear families, were not systematically examined by a dermatologist, for practical reasons, which could have lead to bias in our study, as some types of vitiligo are very discrete (i.e.: genital involvement). For these reasons, in family studies, some authors only considered nuclear families able to be used in performing segregation analysis, i.e.: when all members had been clinically examined by a dermatologist [7]. In Martinique, the small number of involved families and the lack of multiplex families with no more than 3 affected relatives did not allow us to carry out segregation analysis [7].

In our population, data concerning the age of onset of vitiligo are not in agreement with most of the previous reports, which showed vitiligo appearing earlier in family cases than in sporadic cases [12, 13]. Our study failed to show any statistical difference between the age of onset among probands and affected relatives (31 years) and controls (33 years). Moreover, a large proportion of family cases exhibited no vitiligo before the age of 20 (36%). This could be due to a bias related to the small size of our group or be a particularity of the populations studied, who are all suject to the same environmental factors, living on a very small island.

Article accepted on 9/7/01

REFERENCES

1. Nath SK, Majumder PP, Nordlund JJ. Genetic epidemiology of vitiligo: multilocus recessivity cross-validated. Am J Hum Genet 1994; 55: 981-90.

2. Majumder PP, Nordlund JJ, Nath SK. Pattern of familial aggregation of vitiligo. Arch Dermatol 1993; 129: 994-8.

3. Le Poole IC, Das PK, Van den Wijngaard RMJGJ, Bos JD, Westerhof W. Review of the ethiopathomechanism of vitiligo: a convergence theory. Exp Dermatol 1993; 2: 145-53.

4. Lander ES, Schork. Genetic dissection of complex traits. Science 1994; 265: 2037-48.

5. Merelender J, Rywlyn J. Heredity of acquired vitiligo (vitiligo in 3 generations). Acta Derm Venereol 1940; 121: 583.

6. Nordlund JJ, Halder RM, Grimes P. Management of vitiligo. Dermatol Clin 1993; 11: 27-33.

7. Nath SK, Majumder PP, Nordlund JJ. Genetic epidemiology of vitiligo: multilocus recessivity cross-validated. Am J Hum Genet 1994; 55: 981-90.

8. Das SK, Majumder PP, Chakraborty R, Majumdar TK, Haldar B. Studies on vitiligo II. Familial aggregation and genetics. Genet Epidemiol 1985; 2: 255-62.

9. Boisseau-Garsaud AM, Garsaud P, Calès-Quist D, Hélénon R, Quénéhervé C, Charles-Sainte-Claire R. Epidemiology of vitiligo in the French West Indies (Isle of Martinique). Int J Dermatol 2000; 39: 18-20.

10. Service statistique régional, ministère des Affaires sociales, de la Santé et de la Ville. Statistiques Santé-Affaires sociales, région Sanitaire Antilles-Guyane. Atlier Eds, Fort-de-France, 1994; 1-47.

11. Bhatia PS, Mohan L, Pandey ON, Singh KK, Arora SK, Mukhija RD. Genetic nature of vitiligo. J Dermatol Sci 1992; 4: 180-4.

12. Ando I, Chi HI, Nakagawa H, Otsuka F. Differences in clinical features and HLA antigens between familial and non-familial vitiligo of non-segmantal type. Br J Dermatol 1993; 129: 408-10.

13. Schallreuter KU, Lemke R, Brandt O, Schwartz R, Westerhof W, Montz R, Berger J. Vitiligo and other diseases: coexistence or true association? Hamburg study on 321 patients. Dermatology 1994; 188: 269-75.