Dry eye tests explained: TBUT, Schirmer and osmolarity

A dry eye assessment is not one test. It is a short sequence of them, run in a particular order, because each one disturbs the tear film a little more than the last. Expect twenty to thirty minutes: some questions, a look at your tear film under a blue light, possibly a drop or a paper strip, and a close look at your lid margins.

The TFOS DEWS II diagnostic methodology report sets out the sequence. Triaging questions first, to exclude conditions that can mimic dry eye. Then symptom screening with the DEQ-5 or the Ocular Surface Disease Index, which confirms that you might have dry eye disease and triggers the rest. Then the diagnostic tests themselves: breakup time, ideally measured non-invasively, tear osmolarity, and ocular surface staining with fluorescein and lissamine green across the cornea, the conjunctiva and the eyelid margin.

Only after that does the report use meibomian gland assessment, lipid layer thickness and tear volume to sub-classify the disease as predominantly evaporative or aqueous deficient. That last step decides your treatment.

Why the order is not arbitrary

Every test that touches the eye changes what is being measured. Fluorescein alters the tear film it is meant to assess, and the cut-off for breakup time shifts with the volume instilled. A Schirmer strip provokes reflex tearing.

So the least invasive measurements come first. The Meibomian Gland Dysfunction Workshop's diagnosis report lists a clinic sequence in the same spirit: questionnaire, blink rate, tear meniscus height, osmolarity if available, then fluorescein and breakup time, then staining, then the Schirmer test, with gland assessment at the end. If your optometrist seems to be working in a fussy order, that is why.

Symptoms, then tear break-up time

The questionnaire is a gate rather than a verdict: it establishes that symptoms are worth pursuing. We cover it in our piece on the dry eye self-test and the OSDI.

Tear break-up time measures stability: how long the tear film holds together after a blink before the first dry spot appears. In the fluorescein version, dye is instilled, you blink a few times and then hold your eyes open while the optometrist watches through a blue exciter filter and a yellow barrier filter. In the non-invasive version, a grid of light is projected onto the tear film and a camera times the moment it distorts. The dye feels like a brief cool sting and tints your tears yellow for a while.

The MGD workshop's diagnosis report tabulates a fluorescein break-up time under 10 seconds as discriminating dry eye with 82% sensitivity and 86% specificity in one study, and grades severity from 10 seconds or more as normal down to instant breakup at the severe end. The catch is repeatability. Break-up time varies between measurements and with how much dye went in, so a single reading is weaker evidence than it looks.

Tear osmolarity

Osmolarity measures how concentrated your tears are. A tiny chip touches the tear meniscus at the lower lid to collect a nanolitre-scale sample. Most people feel nothing at all.

Lemp and colleagues studied 314 consecutive subjects across ten sites, comparing osmolarity against break-up time, corneal and conjunctival staining, the Schirmer test and meibomian gland grading. Of the six, osmolarity had the superior diagnostic performance, with the highest area under the curve at 0.89. The most sensitive threshold between normal and mild or moderate subjects was 308 mOsm/L and the most specific was 315; at a cut-off of 312 mOsm/L, hyperosmolarity showed 73% sensitivity and 92% specificity. Differences between the two eyes correlated with increasing disease severity, and the authors concluded that inter-eye variability is a characteristic of dry eye not seen in normal subjects.

That same study is the plainest published account of what the other tests cannot do. Corneal staining showed 54% sensitivity, conjunctival staining 60% and meibomian gland grading 61%, while break-up time managed only 45% specificity and the Schirmer test 51%.

The Schirmer test and ocular surface staining

The Schirmer test measures tear production. A folded strip of filter paper is hooked over the lower lid margin and left for five minutes, and the wetted length is read off in millimetres. It is the least comfortable test in the set, and many people's eyes water simply because there is a strip in them. That is precisely its weakness, and why some clinics use a finer phenol red thread instead. The MGD workshop's diagnosis report tabulates a Schirmer result under 5.5 mm in five minutes as discriminating dry eye with 85% sensitivity and 83% specificity in one study, while Lemp's cohort found its specificity closer to half.

Ocular surface staining looks for damage rather than function. Fluorescein highlights breaks in the corneal surface; lissamine green picks up devitalised cells on the conjunctiva and along the lid margin. Both sting briefly and both are graded against a published scale such as the Oxford scale, which runs from 0 to 15. Staining is specific but insensitive, so a clean cornea does not mean a healthy tear film.

The lid examination, and what it adds

This is the part that decides whether your dry eye is evaporative, and the part most often skipped.

Gland expression. The optometrist applies firm, steady pressure to the lid margin and watches what comes out of the gland orifices. Standardised instruments exist: the diagnostic expressor described by Korb and Blackie raises the pressure to between 30 and 40 mm Hg and expresses about eight glands at once, roughly a third of the lid's length. The score is how many of those eight yield a fluid secretion. Quality is graded separately across eight central lower-lid glands, from clear through cloudy to a thick toothpaste-like consistency. It feels like firm pressure, briefly uncomfortable rather than painful.

Meibography photographs the glands themselves in silhouette. The lid is everted and imaged with an infrared camera, showing the glands as pale parallel structures and how much of each has shortened or disappeared. Arita's meiboscore grades loss from 0 to 3 in each lid and sums both lids to a scale of 0 to 6. In 236 healthy volunteers aged 4 to 98 that score rose with age and correlated with lid margin abnormality. The MGD workshop tabulates a meiboscore of 3 or more as identifying evaporative dry eye with 83% sensitivity and 90% specificity. We cover this in what meibography shows.

Lipid layer interferometry reads the colour and brightness of the interference pattern on the tear film to estimate lipid layer thickness, reported from 15 to 157 nanometres with a mean near 42. Nothing touches the eye. It marks whether the glands are delivering oil rather than measuring protection directly.

What each test can and cannot tell you

No single number diagnoses dry eye, which is why the sequence exists. What matters at the end is whether the tear film is unstable because the oil has failed, thin because production has failed, or both. Our comparison of evaporative and aqueous-deficient dry eye explains why that distinction drives everything after it.

If the exam points to meibomian gland dysfunction, eyelid warming is among the first-line measures guideline groups recommend, and the Meibocare E-Heated Eye Mask is designed to bring the eyelids to about 42 °C on the recommended setting and timer. If your results point to reduced tear production instead, warming the lids is not the treatment you need, and that is a distinction the tests exist to make.

If you have an active eye infection, have had eye surgery recently, have glaucoma, or notice any change in your vision, see your optometrist or ophthalmologist rather than waiting for a routine appointment, and speak to them before starting heat therapy.

Frequently asked questions

What tests are done for dry eye?

The TFOS DEWS II sequence starts with triaging questions to exclude mimics, then a symptom questionnaire such as the DEQ-5 or OSDI, then tear break-up time measured ideally without touching the eye, tear osmolarity, and ocular surface staining with fluorescein and lissamine green. Meibomian gland assessment, lipid layer thickness and tear volume then sub-classify the disease as evaporative or aqueous deficient.

What is a normal tear break-up time?

Ten seconds or more is generally treated as normal. The Meibomian Gland Dysfunction Workshop's diagnosis report tabulates a fluorescein break-up time under 10 seconds as separating dry eye from normal with 82% sensitivity and 86% specificity in one study, and grades severity downwards from there. Readings vary between measurements and with how much dye is instilled, so one result is weak evidence alone.

What tear osmolarity level indicates dry eye?

In a 314-subject, ten-site study, the most sensitive threshold between normal and mild or moderate dry eye was 308 mOsm/L and the most specific was 315. At a cut-off of 312 mOsm/L, tear hyperosmolarity showed 73% sensitivity and 92% specificity. Osmolarity had the best diagnostic performance of the six tests compared, and differences between the two eyes tracked severity.

Does a dry eye exam hurt?

Mostly no. Osmolarity and non-invasive break-up time are usually not felt at all. Fluorescein and lissamine green drops sting briefly and tint your tears. The Schirmer test is the least comfortable, with a paper strip resting on the lower lid for five minutes, and gland expression involves firm pressure on the lid margin that is uncomfortable rather than painful.

Educational information only, not medical advice. References: Wolffsohn JS, et al. Ocul Surf. 2017;15:539–574; Lemp MA, et al. Am J Ophthalmol. 2011;151:792–798; Tomlinson A, et al. Invest Ophthalmol Vis Sci. 2011;52:2006–2049; Arita R, et al. Ophthalmology. 2008;115:911–915; Willcox MDP, et al. Ocul Surf. 2017;15:366–403; Jones L, et al. Ocul Surf. 2017;15:575–628.