Vol 45 No 1 Original Article PDF

Comparative Analysis of Meibomian Gland Dysfunction in Eyes with and without Eyelid Margin Tattoos

Ma. Joanna Carla Z. Garcia, MD1, Edgar U. Leuenberger, MD, DPBO1,2, James Paul S. Gomez, MD, DPBO2, Tommee Lynne T. Tiu, MD, DPBO1,3,4,5, Sharlene I. Noguera, MD, DPBO2

1Department of Ophthalmology, University of the East Ramon Magsaysay Medical Center, Quezon City, Philippines
2Asian Eye Institute, Makati City, Philippines
3Eye Institute, St. Luke’s Medical Center, Quezon City, Philippines
4Clinica Tamesis, Quezon City, Philippines
5Department of Ophthalmology, New Era General Hospital, Quezon City, Philippines

Correspondence: Ma. Joanna Carla Z. Garcia, MD
Department of Ophthalmology, 5th floor Charity Bldg., 64 Aurora Boulevard
Barangay Doña Imelda, Quezon City
e-mail: ma.joannagarciamd@gmail.com

Disclosure: The authors report no conflict of interest.

The prevalence of meibomian gland dysfunction (MGD) in Asians ranges from 40-69% based on population-based studies.1,2 MGD pertains to abnormalities in the quantity and quality of meibomian gland (MG) secretion. This develops when there is terminal duct obstruction, epithelial hyperkeratinization, meibum stasis, and eventual gland drop out. When sufficiently severe, MGD may lead to evaporative dry eye disease.2,3

Development of MGD is multifactorial. Risk factors identified include advanced age, hormonal imbalance or therapy, skin diseases, environmental factors, intake of allergy medications, and contact lens wear.2,3 Additionally, eyelid margin tattoos have been implicated to contribute to MGD development.4,5

Eyelid margin tattoos are permanent cosmetic changes that replace traditional make-up application. During eyelid tattoo application, the upper layers of the eyelid margin skin are infiltrated with pigmented granules with the use of a sterile needle.6,7 Eyelid margin tattoos have been popular since the 1980s but their adverse effects are poorly studied.7,8 The demonstrated effects include decreased tear breakup time (TBUT), increased corneal fluorescein staining, and MG dropout.5 Eyelid margin tattoos may cause MGD from a variety of reasons: needle injury, chronic inflammation, allergenic potential of the pigment, duct obstruction, and blepharitis.5,8,10

Meibography is a diagnostic technique used to assess partial or complete MG dropout. It uses infrared light to visualize the entirety of the MG from orifice to fornix.1 When done in combination with colored photographs of the eyelid margins such as described by Arita et al. in a previous publication, severity of can be assessed.9

An extensive review of online literature using PubMed, Google Scholar, and archives of the Philippine Journal of Ophthalmology yielded only a few case reports and one descriptive study that evaluated the association of eyelid margin tattooing with MGD.4,5,8,10,11,12,13,14 In the present study, we investigated the prevalence ratio of MGD in subjects with and without eyelid margin tattoos with the use of imaging by colored photographs and meibography.


This study was approved by the St. Cabrini Medical Center – Asian Eye Institute (SCMC-AEI) Ethics Review Committee and by the University of the East Ramon Magsaysay Memorial Medical Center, Inc. (UERMMMCI) Research Institute for Health Sciences Ethics Review Committee. Patients were recruited between August 2017 and September 2018 from the outpatient clinics of the Asian Eye Institute and the UERMMMCI. Taking into consideration a 15% dropout rate, the calculated sample size was 15 subjects per group. Written informed consent was obtained from all subjects before examination.

Subjects with and without eyelid margin tattoos were recruited to participate in this study. In the tattoo group, the approximate date of tattoo application and retouches were recorded. Patients who have chronic skin diseases and allergies, history of chemical burns, and are taking post-menopausal hormone therapy were ineligible to join. Contact lens wearers and those on concomitant eye drop use (except for artificial tears) were also excluded. Subjects without eyelid margin tattoos in the control group had no ocular comorbidities and no previous ocular surgeries.

Each subject underwent a comprehensive examination. Colored photographs of the lid margins were taken using a Topcon SL D701 Digital slit lamp with DC-4 digital camera (Topcon and Top Ophthalmic Products and Services Corp., Japan) (Figure 1).

Afterwards, meibography images were taken with the following settings: infrared selector lever tilted to the right, zero light intensity, and 6x magnification. An experienced examiner everted the upper and lower eyelids, then infrared images of the meibomian glands were obtained (Figure 2). The pictures captured were saved digitally in the Topcon DC-4 imaging software.

Grading of the colored photographs was done by a single investigator (JG) and validated by a trained external eye disease consultant (TTT). Blinding was precluded because of the nature of the study. Eyelid margin abnormalities were graded according to the proposed grading scale for MGD described in a previous publication by Arita et al.9 Using this scale, photos were graded based on (1) vascularity of the lid margin – described as hyperemia of the lid margin and the amount of telangiectasia distributed over the full length of the lid margin (Figure 3); (2) plugging of gland orifices – based on the number of abnormal findings such as capping, pouting, and ridging over the full length of the lid margin (Figure 4); (3) lid margin irregularity – based on the number of lid margin irregularities and notching (Figure 5); and (4) thickening – based on the presence of rounding of the lid margin (Figure 6).9

Partial and complete gland dropouts were measured by reviewing the infrared photographs. Partial gland loss was defined as MG showing incomplete loss from the orifice to fornix, based on their number and length, while complete gland dropout was assessed based on the number of MG with complete loss from orifice to fornix (Figures 7 and 8).9

The upper and lower eyelids were graded separately and compared with corresponding reference photos. A grade of zero is given for normal findings, while a higher number in the grading scale denotes an increase in severity of abnormal findings.9

Statistical Analysis

Two-sample independent t-test was used to compare the variables between the two groups. A p<0.05 was considered statistically significant. Prevalence ratio and prevalence odds ratio were calculated to determine the strength of association of having an eyelid margin tattoo and MG dropout.


Thirty-eight Filipino female subjects were included in the study. There were 19 subjects per group, with a mean age of 64.37 years (range: 39-74) in the tattoo group, and 60 years (range: 50-67) in the control group. The difference between the mean ages of the 2 groups was not statistically significant (p=0.07) (Table 1). For those who were able to recall the date of the tattoo application (n=6), the range of tattoo exposure was from 10 to 45 years.

Only thirty-six eyes in the tattoo group were included in the analysis as one subject in the tattoo group had difficulty with upper eyelid eversion. All eyes (n=38) in the control group were included in the analysis.

Analysis of Meibomian Gland Dysfunction Characteristics

Compared to the control group, mean scores on abnormal lid margin vascularity (p=0.000197), lid margin irregularities (p<0.0000001) and lid margin thickening (p<0.0000001) were significantly greater in the upper eyelids of subjects with tattoos (Table 2).

Evaluation of the lower eyelid disclosed significantly higher mean scores for lid margin vascularity (p<0.0000001), lid margin irregularities (p<0.0000001) and lid margin thickening (p<0.0000001) in subjects with eyelid tattoos compared to the control group (Table 2).

Meanwhile, differences in the mean scores on plugged gland orifices was not statistically significant between the two groups in both upper and lower eyelids (p=0.65 and p=0.91, respectively).

The means of scores on partial gland dropout in the upper lids of the control and tattoo group were not statistically different (p=0.07). However, in the lower eyelids, there was a significant difference in both groups (p=0.0000002) with the tattoo group showing poorer scores (Table 3).

In both upper and lower eyelids, complete MG dropout mean scores were significantly greater in the tattoo group versus the control group (p=0.003 and p=0.000001) (Table 3). All subjects with eyelid tattoos had a significant MG dropout both in the upper and lower eyelids, compared to those without eyelid tattoos (9 out of 19 subjects or 47%) (Table 4). Prevalence odds ratio was calculated at 2.13.


Eyelid margin tattoos have gained popularity due to their convenience and permanency. However, there are case reports of early and late adverse reactions to permanent eyelid margin tattoos including pruritus, progressive erythema, eye irritation, eyelid margin swelling and scaling. Recent studies show that eyelid tattoos can cause MGD by mechanical destruction from the sharp needle and from the inflammatory reactions to the toxic pigments.5,12

Eyelid tissues with tattoos examined under light and electron microscopes showed pigment granules dispersed extracellularly between epidermis and dermis. The pigments used in eyelid margin tattoos contain chemical substances which can incite a granulomatous reaction.8

Upon tattoo application, these pigment granules are captured and phagocytosed by macrophages. With time, these pigment-laden macrophages undergo apoptosis and release the pigment particle. Since they are large extracellular particles, they are continuously phagocytosed by neighboring macropages, accounting for their permanency.8,15 It is also possible for the ink pigments to infiltrate the different layers of the eyelid and contiguous structures. Some pigments within the macrophages remain in the dermis, while others migrate to the local lymph nodes.10,12

There have been reports of delayed hypersensitivity reaction postulated to be secondary to localized immune sensitization to certain pigments containing magnesium silicate hydroxide or talc. Histopathologic examination of these cases revealed granulomatous reactions to tattoo pigments characterized by epithelioid granulomas with well-formed epithelioid tubercles, histiocytes, and multinucleated giant cells.8,12

In this study, we compared the severity of MGD in patients with and without eyelid margin tattoos. Although most of the study participants were in the elderly age group, the mean age and the number of eyes studied were similar between the two groups.

We initially sought to determine the association between the duration of tattoo exposure and severity of MGD. Unfortunately, majority of the subjects with eyelid margin tattoo (13 out of 19) failed to recall the date of tattoo application. Hence, the relationship between duration of tattoo exposure and MGD cannot be ascertained.

Eyelid margin abnormalities such as vascularity, irregularity and thickening in both upper and lower eyelids were noted to be greater in the tattoo group. In addition, complete gland dropout was significantly higher in the same group. Prevalence odds ratio was calculated at 2.13. It is likely that having an eyelid margin tattoo leads to increased MG dropout and may aggravate pre-existing MGD. This can be due to the possible effects of the tattoo pigments, causing chronic eyelid irritation and inflammation.8,12,13 Full thickness penetration of the needle during tattoo application can also damage the muscle of Riolan at the gray line leading to irregularities of eyelid margin as well.10,12 Eventually, direct needle penetration and chronic inflammation from the pigments can obliterate the MGs, resulting to MG dropout and tear film lipid layer abnormalities.5,13 Furthermore, other reported more serious adverse effects of eyelid margin tattoos include preseptal cellulitis, delayed hypersensitivity granulomatous reactions and extensive pigment spread in the different layers including muscle fibers.8,11

This study has several limitations. First, due to the cross-sectional design of the study, the presence of MGD prior to tattoo application was not documented and the possibility of MGD exacerbation after eyelid margin tattoo application cannot be ascertained. The documentation of presence of MGD prior to tattoo application is needed to determine the strength of association between eyelid margin tattoos and MG dropout. Second, data regarding tattoo application (i.e. type of tattoo ink used, technique, and depth of application) and presence of any early adverse reactions were not collected. Third, the absence of masking of the grading examiner may have resulted to observer bias. Lastly, we did not assess dry eye disease using other methods like vital dye staining, TBUT, and self-administered questionnaires.

In conclusion, eyelid margin tattoos are associated with several eyelid margin abnormalities and increase the risk of total MG dropout. Our study suggests that female subjects with eyelid margin tattoos are twice more likely to have MG dropout compared to subjects without eyelid margin tattoos. This finding is important in counseling patients who have been diagnosed to have MGD and who are contemplating to have eyelid margin tattoos.


We would like to extend our gratitude to Jennifer M. Nailes, MD, MSPH of the Research Institute for Health Sciences and Department of Preventive and Community Medicine of the UERMMMCI for providing aid in the statistical analysis.



1. Wise RJ, Sobel RK, Allen RC. Meibography: A review of techniques and technologies. Saudi J Ophthalmol.
2. Schaumberg DA, Nichols JJ, Papas EB, et al. International workshop on meibomian gland dysfunction. Inves Ophthalmol Vis Sci. 2011;52:1994-2005.
3. Machalinska A, Zakrzewska A, Safranow K, Wiszniewska B, Machalinski B. Risk factors and symptoms of meibomian gland loss in a healthy population. J Ophthalmol. 2016;1-8.
4. Gomes JA, Azar DT, Baudouin C. TFOS DEWS II iatrogenic report. The Ocular Surface. 2017;14:511-538.
5. Lee YB, Kim JJ, Hyon JY, Wee WR, Shin YJ. Eyelid tattooing induces meibomian gland loss and tear film instability. Cornea. 2015;34(7):750-755.
6. Darby D. Eyeliner tattooing vs dry eye. August 3, 2016. http://www.cosmetictattoo.org/article/eyeliner-tattooingvs-dry-eye.html. (accessed January 5,2017).
7. Petrosyan T. American Optometric Association. Cosmetics and the eye: how your beauty products can be harming your eyes. November 13, 2015. https://www.aoa.org/Documents/CPC%20SUBMISSION%20%20Cosmetics%20and%20th e%20Eye%20How%20Your%20Beauty%20Products%20
Could%20Be%20Harming%20Your%20Eyes.pdf.(accessed August 8, 2017).
8. Vagefi MR, Dragan L, Hughes SM, et al. Adverse reactions to permanent eyeliner tattoo. Ophthalmic Plast Reconstr Surg. 2006;22:48-51.
9. Arita R, Minoura I, Morishige N, et al. Development of definitive and reliable grading scales for meibomian gland dysfunction. Am J Ophthalmol. 2016;169:125-137.
10. Goldman, A & Wollina, U. Severe unexpected adverse effects after permanent eye makeup and their management by Qswitched Nd:YAG laser. Dove Press: Clinical Interventions in Aging. 2014;9:1305-1309.
11. Morrison CJ, Stamm JM. My tattoos caused my dry eye? A new way to look at diagnosis and treatment for patients with tattoo eyeliner. Can J Optom. 2016;78:6-10.
12. Schwarze HP, Giordano-Labadie F, Loche F, et al. Delayed hypersensitivity granulomatous reaction induced by blepharopigmentation with aluminum silicate. J Am Acad Dermatol. 2000;42:888-91.
13. Kojima T, Dogru M, Matsumoto Y, et al. Tear film and ocular surface abnormalities after eyelid tattooing. Ophthalmic Plast Reconstr Surg. 2005;2:69-71.
14. Seol BR, Kwon JW, Wee WR, Han YK. A case of meibomian gland dysfunction after cosmetic eyelid tattooing procedure. J Korean Ophthalmol Soc. 2013;54:1309-1313.
15. Baranska A, Shawker A, Jouve M, et al. Unveiling skin macrophage dynamics explains both tattoo persistence and strenuous removal. J Exp Med. 2018;4:1115-1133.
16. Cornish, R. Statistics: An introduction to sample size calculations. Mathematics Learning Support Centre. 2006. 1-5.