The literature search identified 5783 publications through PubMed, 1668 records through Embase, and 4 publications through manual search (Fig. 1). After removal of duplicates and studies with publication dates before January 2012, 3104 records were eligible for abstract and title screening, of which 226 full-text articles were assessed further for eligibility. Most publications excluded at abstract and title screening were because the study outcomes or study population were not of interest and therefore could not provide meaningful data toward the predefined objectives of this literature review. A total of 15 publications met the full eligibility criteria for quantitative synthesis. Of these, seven were excluded from this analysis for not reporting data on associations between vitiligo and autoimmune comorbidities. The remaining eight publications met all selection criteria and included 10,246 unique patients [9, 20,21,22,23,24,25,26]. Six records were cross-sectional studies [9, 20,21,22,23, 25], and two were cohort studies [24, 26] (Table 1). Overall, four studies were rated as high quality [20, 21, 23, 24], three as moderate [9, 25, 26], and one as low quality [22] (Table 1, Supplementary Tables S2 and S3); however, no study was excluded on the basis of the quality assessment.

PRISMA flow diagram. PRISMA Preferred Reporting Items for Systematic Reviews and Meta-Analyses

Seven studies examined autoimmune comorbidities in patients with vitiligo [9, 20,21,22,23,24,25], and the remaining study examined vitiligo as a comorbid disease in patients with alopecia areata [26]. The most common comorbidities across all studies by pooled prevalence estimates (occurring in > 3% of patients) were thyroid diseases (14.2%), psoriasis (5.1%), and rheumatoid arthritis (3.2%; Table 2). On the basis of GRADE assessment, there is a high certainty of evidence that greater vitiligo extent is associated with increasing prevalence of hypothyroidism and type 1 diabetes (Table 3). There is also moderate certainty of evidence that vitiligo in adults is associated with alopecia areata and hyperthyroidism, and that greater vitiligo extent is associated with increasing prevalence of thyroid disease, rheumatoid arthritis, and pernicious anemia.

The pooled prevalence of thyroid disease in adults with vitiligo was 13.6% (95% CI 8.9–19.2) but with substantial heterogeneity across studies (n = 7; I2 = 98.1%) [9, 20,21,22,23,24,25]. The individual pooled prevalences of hypothyroidism (including Hashimoto’s thyroiditis) and hyperthyroidism (including Graves’ disease) were 9.9% (9.1–10.7%; n = 4; I2 = 0%) and 2.1% (1.6–2.6%; n = 2; I2 = 0%), respectively [20, 21, 23, 25]. In addition to reporting crude prevalence values, one included study calculated the risk of developing thyroid disease [25]. This cross-sectional study showed that adults with vitiligo are associated with significantly increased odds of developing hypothyroidism compared with the general population (odds ratio [95% CI] 4.02 [3.29–4.88]; P < 0.05).

Three included studies also reported data indicating that thyroid disease prevalence may vary across patient subgroups. A cross-sectional study reported that hypothyroidism and hyperthyroidism were associated with older age and more extensive vitiligo (measured by affected body surface area, number of body parts involved, and bilaterality), disease duration, and vitiligo-associated itching and burning [22]. In addition, subgroup analyses from two additional cross-sectional studies revealed that thyroid disease was more prevalent in female versus male patients with vitiligo (hypothyroidism, 88% vs 25% [20]; thyroid disease, 18.5% vs 5.1% [P < 0.001] [21]).

The pooled prevalence of psoriasis in adults with vitiligo was 5.1% (95% CI 2.3–7.9), with substantial heterogeneity across studies (n = 4; I2 = 97.0%) [21,22,23,24]. Furthermore, associations between vitiligo and psoriasis may depend on patient race. One included cross-sectional study reported that psoriasis occurred significantly less frequently in Black patients with vitiligo compared with patients of White or other races (P = 0.021) [21].

The pooled prevalence of rheumatoid arthritis among adults with vitiligo was 3.2% (95% CI 1.7–4.6), with substantial heterogeneity across studies (n = 6; I2 = 95.8%) [9, 20,21,22,23,24]. Rheumatoid arthritis prevalence may also differ across patient subgroups on the basis of evidence reported by two included cross-sectional studies. The first study reported that rheumatoid arthritis was observed in a higher percentage of female patients with vitiligo (2.1%) than male (0.9%) [20]. The second study found that rheumatoid arthritis was associated with older age and more extensive, long-standing, or active vitiligo (based on affected body surface area, disease duration, and vitiligo-associated itching and burning) [22].

The pooled prevalence of alopecia areata among adults with vitiligo was 2.7% (95% CI 2.3–3.1), with low heterogeneity (n = 5; I2 = 13.2%) [9, 21,22,23,24]. In addition to reporting crude prevalence values, one included study calculated the risk of developing alopecia areata [26]. The prospective cohort study showed that adults with vitiligo have significantly increased risk of developing alopecia areata compared with patients who have no history of autoimmune disease (multivariable-adjusted hazard ratio [HR] [95% CI] 3.13 [1.08–9.10]; P < 0.05). However, alopecia areata was not shown to be associated with vitiligo extent [22]. The association between vitiligo and alopecia areata is also dependent on patient race, with one included cross-sectional study indicating that alopecia areata occurred less frequently in White patients with vitiligo compared with patients of Black or other races (P = 0.024) [21].

The pooled prevalence of type 1 diabetes mellitus in adults with vitiligo was 1.8% (95% CI 0.9–2.8), with substantial heterogeneity across studies (n = 5; I2 = 92.8%) [9, 21,22,23,24]. No association was identified between vitiligo and type 1 diabetes mellitus on the basis of specific patient or disease characteristics [9, 21,22,23,24].

The pooled prevalence of anemia or pernicious anemia in adults with vitiligo is 1.7% (95% CI 0.6–3.3), with substantial heterogeneity across studies (n = 5; I2 = 94.9%) [9, 20,21,22, 24]. In addition, one included study assessed the effects of vitiligo extent (as measured by affected body surface area, number of body parts involved, and bilaterality), reporting that pernicious anemia was associated with more extensive vitiligo lesions [22].

The pooled prevalence of chronic urticaria in adult patients with vitiligo is 1.6% (95% CI 0.7–2.5), with substantial heterogeneity across studies (n = 2; I2 = 80.6%) [9, 22]. No association was identified between vitiligo and chronic urticaria on the basis of specific patient or disease characteristics [9, 22].

The pooled prevalence of inflammatory bowel disease (IBD) in vitiligo was 1.4% (95% CI 0.8–2.1), with substantial heterogeneity across studies (n = 6; I2 = 85.3%) [9, 20,21,22,23,24]. In addition to heterogeneity among included studies, a cohort study that did not meet the meta-analysis selection criteria reported no association between vitiligo and an increased risk of IBD, including ulcerative colitis (HR [95% CI] 1.14 [0.77–1.68]) and Crohn’s disease (HR [95% CI] 1.45 [0.87–2.41]) [27].

Vitiligo is also associated with other comorbidities with a pooled prevalence of less than 1.5%, including systemic lupus erythematosus (1.2% [126/9750]), Sjögren’s syndrome (1.1% [67/5822]), celiac disease (0.9% [≤ 43/4724]), Addison’s disease (0.8% [38/4724]), multiple sclerosis (0.3% [8/2585]), and myasthenia gravis (0.2% [6/2585]) [9, 20,21,22,23,24]. In a cross-sectional study, systemic lupus erythematosus was only observed in Black patients; however, statistical significance was not reached versus other races [21]. In a prospective cross-sectional survey, Sjögren’s syndrome was associated with more extensive vitiligo (based on affected body surface area) and abdominal lesions [22]. No further associations based on patient or disease characteristics were identified between vitiligo and these other autoimmune conditions [9, 20,21,22,23,24].