Meibography: what it shows about your meibomian glands

Meibography is an infrared photograph of the oil glands inside your eyelid. The lid is turned outwards, infrared light is used instead of visible light, and the meibomian glands appear as a row of pale, vertical, comb-like structures against darker tissue. What the image shows is structure: how many glands are present, how far back they run, and how much of the row has been lost.

What it cannot show is whether those glands are working. A gland can look intact in the picture and still be plugged at its opening, and a lid with obvious gland loss can still deliver useful oil from what survives. Meibography sits alongside gland expression and tear film testing rather than replacing them.

What the image is actually showing

Meibomian glands sit in vertical rows inside the firm tarsal plate of each eyelid and make meibum, the oily layer that slows evaporation. You cannot see them under white light, but infrared reaches them: lipid reflects infrared differently from the tissue around it, so gland tissue comes out bright and the space where a gland used to be comes out dark.

Older meibography needed a transilluminating probe pressed against the outside of the lid. The method most clinics use now was described by Arita and colleagues in 2008: a slit lamp fitted with an infrared video camera and an infrared transmitting filter, the eyelid simply turned over, no probe required. In their series of 236 healthy volunteers aged 4 to 98, clear images were obtained in every subject, children included.

Dropout, shortening and distortion

Dropout is gland tissue that is no longer there. On the image it is a dark gap in the row: sometimes one missing gland, more often a wedge of lid where the glands have shortened back from the margin and the tips have gone. It is generally read as the end point of long-standing obstruction rather than something that happened last month.

Two other patterns matter, and neither is loss. Shortening is a gland still present but no longer running its full length. Dilation and tortuosity describe glands that look swollen, kinked, or wandering out of the vertical line, suggesting a gland under pressure from oil that is not getting out. These are worth treating hardest, because there is still gland tissue to work with.

Age changes the reading. Arita's group found a significant positive correlation between age and gland loss in their normal population (R = 0.428), holding separately in men and in women, alongside weaker negative correlations between age and both break-up time and the Schirmer test. Gland loss also tracked with lid margin abnormality (R = 0.359). Some change with age is ordinary. The useful questions are whether you have more than your age would predict, and whether it is moving.

How the picture becomes a score

Arita's grading, still the common one, scores each eyelid from grade 0, no loss of glands, to grade 3, where the lost area is more than two thirds of the total gland area. That per-lid figure is the meiboscore. A later paper from the same group summed upper and lower lids into a single meibo-score running from 0 to 6.

How reliable is the number? Pult and Riede-Pult tested it in 2013, having three experienced observers grade meibography photographs from 24 people twice each, using a four-step scale, a five-step pictorial scale, and computerised area measurement on a 100-point scale. Observed gland loss ranged from 0% to 69%. Agreement was best for the computerised measurement, then the five-step scale, then the four-step scale, and even the computerised method showed differences between observers of roughly 13 to 26 points on that 100-point scale.

A one-grade shift between visits means little on its own, particularly when images came from different devices or graders. Serial meibography is most informative from the same practice, on the same instrument, with the images side by side.

What your result changes, and what it does not

Less on its own than most people expect, and more in combination. Arita and colleagues compared 53 eyes with obstructive meibomian gland dysfunction against 60 healthy control eyes in 2009. Symptom score, lid margin abnormality, meibo-score, meibum score and superficial punctate keratopathy were all higher in the affected group, and break-up time was shorter. Yet the ocular symptom score had the highest diagnostic power as a single parameter, followed by lid margin abnormality, then the meibo-score. Their proposal was to use three together: suspect obstructive disease when any two of those three are abnormal, and treat it as very likely when all three are.

Function is assessed separately, by pressing gently along the lid and counting how many glands yield liquid oil. Korb and Blackie showed in 2008 that this count in the lower lid correlated with dry eye symptoms, and is not evenly distributed: the nasal third averaged 3.10 glands yielding liquid secretion, the central third 2.14, the temporal third 0.27. No photograph substitutes for the pressing. Blackie and colleagues' 2010 review adds that obstructive disease is often nonobvious, with signs absent unless specific techniques are used, and that treatment succeeds or fails on whether the obstruction is relieved.

Meibography also will not tell you why. Wang and colleagues found people with clinically detectable incomplete blinking had markedly more gland dropout than matched controls, 41.3% against 27.5%, and around twice the odds of meeting the dry eye criteria, so blink quality and screen habits are worth asking about early. Nor will it tell you whether you also have a tear volume problem; our comparison of evaporative and aqueous deficient dry eye explains why that distinction changes the plan, and the wider sequence is in our guide to dry eye tests.

Can the picture improve?

We have less evidence than the question deserves. One retrospective study of 78 eyes found that patients who actually did the eyelid hygiene they were told to do showed a measured reduction in gland dropout of about 5% after a month, and its authors concluded that dropout was reversible to some extent. A 5% shift sits close to the disagreement between graders described above, so read it as encouraging rather than settled. Our fuller account of what that study can and cannot support is in whether meibomian glands grow back.

The safer framing is that keeping the glands you still have is a realistic goal, and a poor score is a reason to start rather than a verdict.

Having meibography done in New Zealand

Meibography is not part of a standard eye examination. It is offered by optometry practices running a dedicated dry eye clinic, by some ophthalmology practices, and occasionally through hospital eye departments. It is usually bundled into a longer dry eye assessment rather than sold on its own, and fees vary between practices. Assume you will be paying for it yourself, and ask what the appointment includes when you book. Ask for a copy of your images and your grade: a scan you can compare against in two years is worth far more than one you cannot find.

What the image usually changes, if it changes anything, is commitment to the daily part of treatment: warming the lids so the oil flows, cleaning the lid margin, and keeping it up long enough to matter. The Meibocare E-Heated Eye Mask was designed for that job, and is intended to bring the eyelids to about 42 °C on the recommended setting and timer, with three settings each paired with a timer: 20 minutes on Low, 15 on Medium, 10 on High. It is notified to Medsafe on the New Zealand WAND database (240927-WAND-746QNT), a register that records notification rather than assessing performance. If you have an active eye infection, have had recent eye surgery, have glaucoma, or notice any change in your vision, speak to your optometrist or ophthalmologist before starting or continuing heat therapy.

Frequently asked questions

What does meibography show?

It shows the structure of the meibomian glands inside the eyelid, photographed under infrared light with the lid turned outwards. Intact glands appear as bright vertical bands and missing gland tissue appears as dark gaps, so the image reveals how much of the gland row has been lost, whether glands have shortened, and whether they look swollen or distorted.

Does gland dropout mean my glands are gone for good?

Not necessarily, though the evidence is thin. One retrospective study of 78 eyes with moderate to severe gland dysfunction found that patients who complied with eyelid hygiene showed roughly a 5% reduction in dropout after a month, and the authors concluded that dropout was reversible to some extent. That is one short study, so treat protecting the glands you have as the realistic goal.

Do I need meibography to be diagnosed with meibomian gland dysfunction?

No. In the study proposing diagnostic criteria for obstructive gland dysfunction, the ocular symptom score had more diagnostic power on its own than the meibography score did. The recommendation there was to combine symptom score, lid margin abnormality and meibography, suspecting the condition when any two are abnormal. Meibography strengthens the picture rather than making it.

How often should meibography be repeated?

Once a year is plenty for most people, and only if it will change something. Grading is not precise enough for short-interval comparison: even computerised measurement varied by roughly 13 to 26 points on a 100-point scale between observers in one repeatability study. Keep repeat scans at the same practice, on the same instrument, and compare the images directly rather than the scores.

Educational information only, not medical advice. References: Arita R, et al. Ophthalmology. 2008;115:911–915; Arita R, et al. Ophthalmology. 2009;116:2058–2063; Pult H, et al. Cont Lens Anterior Eye. 2013;36:22–27; Korb DR, et al. Cornea. 2008;27:1142–1147; Blackie CA, et al. Cornea. 2010;29:1333–1345; Yin Y, et al. Cornea. 2017;36:332–337; Wang MTM, et al. Ocul Surf. 2018;16:424–429.