Table 1 Summary Table of Systematic Review Findings

Periocular Pigmentary Changes

Poliosis was commonly observed in patients with vitiligo, with reported prevalence of 5.9% [5], 16% [6], and 18% [7]. One study reported that of six patients with both uveitis and vitiligo, 67% also had poliosis [8]. Partial or total brow whitening was noted in 13.4% of vitiligo patients by another study [6].

Periocular depigmentation is another common feature of vitiligo. One study reported lid depigmentation in 48% of the vitiligo group [6] while other studies reported periocular involvement in 47-49% [9, 10]. Presence of periorbital depigmented patches was associated with a 58-fold increased risk for ocular findings [odds ratio (OR) 58.579, 95% CI 3.119–1100.3] [11].

Corneal fluorescein staining (CFS) positivity, reported as significantly higher in vitiligo patients compared to healthy controls, was also found to be more prominent in patients with periocular involvement but this difference was not statistically significant [10]. One study observed reductions in Schirmer test scores when comparing patients with periocular depigmentation to those without periocular depigmentation and to healthy controls. They found that patients with acrofacial vitiligo had significantly lower Schirmer test scores than patients with generalized vitiligo (p = 0.001) [12]. Another study found Schirmer test scores were lower in patients with periocular vitiligo compared to healthy controls, but this difference was not statistically significant [13].

Periocular vitiligo may also be related to measures of dry eye disease, including tear break up time (TBUT), Rose Bengal scores, and tear osmolarity [12,13,14]. Patients with acrofacial vitiligo had significantly lower TBUT (p = 0.001) and significantly higher Rose Bengal scores (p = 0.011) than patients with generalized vitiligo [12]. Compared to healthy controls, patients with periocular involvement also had significantly lower TBUT (p = 0.005) [13]. Moreover, 62% of periocular vitiligo patients compared to 28% of healthy controls had TBUT values less than 10 s, a statistically significant difference (p < 0.001) [13]. In terms of tear osmolarity, patients with vitiligo with ocular involvement had significantly higher tear osmolarity values than those without ocular involvement and healthy controls (p = 0.02, p < 0.001, respectively) [14].

Albert, D.M., J.J. Nordlund, & A.B. Lerner (1979) [6]

Bulbul Baskan, E., et al. (2006) [11]

Dertlioğlu, S.B., et al. (2016) [9]

Dogan, A.S., et al. (2015) [10]

Ayotunde, A. & G. Olakunle (2005) [5]

Nordlund, J.J., et al. (1981) [8]

Güngör, Ş., et al. (2015) [12]

Serin, D., et al. (2012) [13]

Erdur, S.K., et al. (2018) [14]

Gopal, K.V.T., et al. (2007) [7]

Dry Eye Disease

In evaluating for dry eye disease in vitiligo patients, studies overwhelmingly reported that vitiligo patients had lower Schirmer scores compared to healthy controls, but at a difference that was not statistically significant [10, 12,13,14,15]. Only one study reported a statistically significant reduction in Schirmer scores [16] compared to healthy controls.

Other indicators of dry eye disease were more conclusive, such as reports of significantly lower fluorescein breakup (FBU) scores [10], significantly higher Rose Bengal scores [12], and significantly lower tear breakup scores (TBUT) in patients with vitiligo [12,13,14,15]. One study reported that tear osmolarity was significantly higher among vitiligo patients compared to healthy controls [14].

Ocular surface disease index (OSDI) scores were higher in vitiligo patients compared to controls, either approaching significance (p = 0.071) [15] or achieving significance [10, 14], (p < 0.001 and p = 0.001, respectively).

To better understand the potential mechanism of dry eye disease in vitiligo patients, one study used conjunctival impression cytology and observed that 31.7% of patients with periocular vitiligo compared to 8% of healthy controls had grade 2–3 changes per Nelson classification, a statistically significant difference [13].

Of the studies that investigated Meibomian gland dysfunction (MGD) [10, 15], one reported that there was no difference in MGD between vitiligo patients and healthy controls while the other reported a statistically significant difference. Meibomoscores in patients with periocular involvement did not differ significantly from those without periocular involvement [15].

Dogan, A.S., et al. (2015) [10]

Güngör, Ş., et al. (2015) [12]

Serin, D., et al. (2012) [13]

Karadag, R., et al. (2016) [16]

Erdur, S.K., et al. (2018) [14]

Palamar, M., et al. (2017) [15]

Cataracts

Several studies reported an absence of notable slit lamp findings, observing clear ocular media with no corneal infiltrates [17] and no anterior lenticular opacities [17,18,19]. However, one study described three patients in their mid-thirties with posterior subcapsular cataract of no evident cause, [6] and another study described two patients with mild lens nuclear sclerosis in both eyes [19]. Karadag et al. reported a statistically significant difference in prevalence of punctate lenticular opacities in vitiligo patients compared to controls (30.4% and 10.5%, respectively) (p = 0.033)), [16] but it seems that overall, there was no significant difference in the incidence of cataracts in vitiligo patients compared to healthy controls (p = 0.6854) [20, 21].

Albert, D.M., J.J. Nordlund, & A.B. Lerner (1979) [6]

Karadag, R., et al. (2016) [16]

Fouad, Y.A., et al.(2020) [20]

El-Mofty, A.M. and A. El-Mofty (1979) [17]

Khurrum, H., K.M. AlGhamdi, & E. Osman (2016) [21]

Mostafa W.Z., et al. (2015) [18]

Gass, J.D. (1981) [19]

Iris Atrophy & Pigmentary Abnormalities

Several studies reported findings of iris atrophy and/or pigmentary abnormalities [6, 7, 16, 20, 22,23,24,25]. While findings of iris atrophy were overall higher in vitiligo patients compared to healthy controls, the prevalence was variable, ranging from 0%-25.5% [6, 23, 24]. In the study that identified focal iris atrophy in 25.5% of vitiligo patients, findings ranged from subtle transillumination defects to small atrophic patches [23]. One study reported racial differences among vitiligo patients, with white patients having higher prevalence of iris transillumination compared to black patients (23% and 5.5%, respectively) [22].

Iris hypopigmentation in vitiligo patients ranged from 5-23% compared to 1-7.1% in healthy controls [7, 23]. One study reported iritis in 2.68% of vitiligo patients compared to 0% of healthy controls. The iritis was associated with heterochromia [6]. Another study reported iritis in 12.5% of vitiligo patients [25]. There does not appear to be a significant difference in iris color between the vitiligo and control groups [7, 16].

Elevated Intraocular Pressure & Glaucoma

Studies reported no significant difference in intraocular pressure (IOP) among vitiligo patients compared to healthy controls [5, 9, 16, 20]. One study diagnosed glaucoma in 2.2% of patients with vitiligo (2/90) compared to 0% of controls. The 2 patients who had glaucoma had generalized vitiligo [21].

One study did find a statistically significant difference in normal tension glaucoma (NTG) in vitiligo patients compared to healthy controls (18% and 0%, respectively) (p = 0.04). Two-thirds of the patients with NTG had periorbital lesions [9].

Rogosic et al. reported primary open angle glaucoma in 57% (24/42) of their vitiligo patients with the duration of vitiligo nearly two-fold longer in patients with both vitiligo and glaucoma than in those with vitiligo alone (p < 0.001) [26]. They identified a 4.4-fold risk of developing glaucoma in patients aged 56 and older compared to those younger than age 56. When corrected for age, they found a 92% probability of association between duration of vitiligo (greater than 13 years) and development of glaucoma [26].

The only study that investigated retinal nerve fiber layer (RNFL) thickness found that mean RNFL thickness outside the disc margin was significantly lower in the vitiligo group compared to controls (p = 0.02), and that mean RNFL thickness beneath the measuring ellipse in the superior sector of both eyes were significantly lower than controls [27]. The significance of these findings is unclear.

Dertlioğlu, S.B., et al. (2016) [9]

Ayotunde, A. & G. Olakunle (2005) [5]

Karadag, R., et al. (2016) [16]

Fouad, Y.A., et al.(2020) [20]

Khurrum, H., K.M. AlGhamdi, & E. Osman (2016) [21]

Ornek, N., et al. (2013) [27]

Rogosić, V., et al. (2010) [26]

Uveitis

Studies found little evidence of increased rates of uveitis among those with vitiligo, as illustrated by no significant difference in prevalence of ocular inflammation between vitiligo and control groups [21,22,23,24]. A single study did report a slightly increased prevalence of active uveitis– either iridocyclitis, chorioretinitis, or both – among those with vitiligo (9/112 patients with vitiligo vs. 0/372 healthy controls) [6].

Cowan, C.L., Jr., et al.(1986) [22]

Khurrum, H., K.M. AlGhamdi, & E. Osman (2016) [21]

Biswas, G., et al. (2003) [23]

Fleissig, E., et al. (2018) [24]

Retina & Choroid

Several studies reported normal fundal examinations in patients with vitiligo [17, 18, 21, 28]. Nevertheless, observed choroid pigment abnormalities in vitiligo patients included prominent choroidal pattern (36%) [6] and tigroid retina (9.8%) [16]. Prevalence of choroidal nevi varied greatly in the reviewed studies, ranging from 0-32.2% [20, 22, 24]. One study reported that there was a significantly higher prevalence of choroidal nevi in white compared to Black patients [22].

One study reported statistically reduced subfoveal choroidal thickness (p = 0.0002) [20] while another, which utilized OCT imaging, found no statistically significant difference in outer nuclear layer thickness, total macular thickness, or thickness of the macular RNFL, ganglion cell layer, inner nuclear layer, inner plexiform player, or outer plexiform layer between vitiligo patients and controls [29].

Several studies showed evidence of higher retinal pigmentary abnormalities in the vitiligo groups, including focal retinal pigment epithelium (RPE) hyperpigmentation, focal RPE hypopigmentation/atrophy, RPE mottling, and/or chorioretinal degeneration/scarring [20, 24, 30, 31].

Of the studies that reported on patients’ visual acuity, most [5, 16, 17, 29] reported normal visual acuity or no significant difference in visual acuity in vitiligo patients compared to healthy controls. Only one study reported significantly lower corrected visual acuity in vitiligo patients compared to controls (p = 0.0109) but this study was limited by a small sample size [20]. Another study reported that 95% of vitiligo patients (107/112 patients) had a visual acuity of 20/30 or better, and the remaining 5 patients’ impaired visual acuity was attributable to chorioretinitis-like lesions in 3 patients and other disease processes [6]. Similarly, perimetry measurements were reported as normal or not significantly different in vitiligo patients compared to controls [5, 16, 17].

The results of studies utilizing electroretinography (ERG) in vitiligo patients were varied. While some studies reported normal ERG results with no significant change in wave amplitudes and wave implicit times [18, 32], others reported ERG findings suggestive of impaired retinal electrophysiological function in vitiligo patients [19, 29, 33].

Shoeibi et al. concluded that the mean rod response b-wave, standard combined a- and b-waves, single-flash cone response b-wave and the 30-Hz flicker (N1-P1) amplitudes were significantly lower in vitiligo patients compared to age-matched healthy controls [33]. Similarly, Aydin et al. found that the mean mfERG-c P1 and mfERG-p P1 amplitudes were significantly lower in the vitiligo group compared with controls (p = 0.002 and p = 0.006, respectively) [29]. One group identified moderately to severely abnormal rod and cone function in both eyes of all ten vitiligo patients who were tested [19].

Albert, D.M., J.J. Nordlund, & A.B. Lerner (1979) [6]

Ayotunde, A. & G. Olakunle (2005) [5]

Karadag, R., et al. (2016) [16]

Cowan, C.L., Jr., et al.(1986) [22]

Fouad, Y.A., et al.(2020) [20]

Aydin, R., et al. (2018) [29]

El-Mofty, A.M. and A. El-Mofty (1979) [17]

Khurrum, H., K.M. AlGhamdi, & E. Osman (2016) [21]

Biswas, G., et al. (2003) [23]

Fleissig, E., et al. (2018) [24]

Wagoner, M.D., et al. (1983) [30]

Shoeibi, N., et al. (2014) [33]

Gopal, K.V.T., et al. (2007) [7]

Lerner, A.B., J.J Nordlund, & D.M Albert (1977) [25]

Mehran, G., et al. (2014) [34]

Mostafa W.Z., et al. (2015) [18]

Perossini, M., et al. (2010) [28]

Shoeibi, N., et al. (2016) [32]

Gass, J.D. (1981) [19]