Understanding Your Test Results

Modern dry-eye diagnosis can be thought of in two broad stages.

First: is there evidence that tear-film or ocular-surface homeostasis has been lost? Then: what is contributing to that loss of homeostasis in this particular person?

These are related questions. But they are not the same question. This distinction explains why some tests help establish dry eye disease, while others mainly help investigate why it may be happening.

According to the TFOS DEWS III diagnostic framework, dry eye requires compatible symptoms together with objective evidence of loss of tear-film and/or ocular-surface homeostasis, after appropriate consideration of other conditions that could explain the symptoms. The three principal homeostasis markers are:

• •tear-film stability
• •ocular-surface staining
• •tear osmolarity

These are the tests asking: is the tear-film/ocular-surface system behaving abnormally?

Once loss of homeostasis has been identified, other tests can help explain what may be contributing to it. For example:

• •Schirmer helps assess whether reduced aqueous tear availability may be part of the picture
• •Tear meniscus height gives information about tear volume present on the eye
• •Meibomian gland expression examines gland function and the secretion that can be expressed
• •Meibography examines gland structure and morphology
• •Lipid-layer interferometry examines properties of the lipid component of the tear film
• •Blink and eyelid assessment looks for blink abnormalities, incomplete closure and exposure
• •Corneal sensitivity testing examines one aspect of corneal sensory function

These tests can be extremely important. But an abnormal result on one of them does not automatically equal a dry-eye diagnosis.

Instead of reading your results as a list of abnormal values, try asking what part of the system each result describes. A simplified map looks like this:

• •Meibography → gland structure
• •Expression → gland function / expressible secretion
• •Interferometry → lipid layer at the tear surface
• •Schirmer → aqueous tear availability under test conditions
• •Tear meniscus → tear volume present
• •NIBUT/TBUT → tear-film stability
• •Staining → ocular-surface disruption
• •Osmolarity → concentration/homeostasis of sampled tears
• •Blink assessment → tear redistribution and eyelid coverage
• •Corneal sensitivity → sensory function

Now the results begin to form a story rather than a collection of diagnoses.

Suppose your meibography shows substantial gland dropout. That tells us something important about gland morphology. But it does not automatically tell us how much meibum is being secreted, how stable the tear film is, how severe your symptoms should be, or whether the structural finding is the dominant driver of your current problem.

Or imagine your Schirmer result is low. That raises the possibility of aqueous deficiency. But it does not automatically prove “My lacrimal glands are damaged.” The test itself cannot establish that conclusion.

The same principle applies repeatedly throughout dry-eye diagnostics: a finding is not always the same thing as an explanation.

This is probably the most useful question you can ask about any dry-eye test. Not “Is this number bad?” but “What did this test actually measure?” For example:

A short NIBUT means the tear film became unstable relatively quickly under the conditions of the test. It does not mean “My meibomian glands caused my dry eye.”

A low Schirmer means relatively little strip wetting occurred during that test. It does not automatically mean “My lacrimal glands have stopped working.”

Meibography showing dropout means normal gland tissue was partially or completely non-visible in areas of that image. It does not automatically mean “Exactly this percentage of my glands is permanently dead.”

High osmolarity means the sampled tear fluid was relatively concentrated. It does not mean “This number is my dry-eye severity score.”

The wording matters.

This question is just as important. For every test, there is a limit to what the result can tell us.

• •Short NIBUT does not identify the cause of instability
• •Staining does not tell us exactly what caused the surface damage
• •Low Schirmer does not prove lacrimal-gland failure
• •Low tear meniscus does not directly measure tear production
• •Poor meibomian gland expression does not tell us the complete structural state of the gland
• •Gland dropout on meibography does not directly measure secretion
• •Thin lipid layer does not directly measure tear evaporation
• •High osmolarity does not identify the dry-eye subtype
• •Incomplete blinking does not prove that blinking caused the disease
• •Severe symptoms with mild signs do not automatically diagnose neuropathic pain

Knowing what a test cannot establish protects you from overinterpreting it.

That is common. And it does not necessarily mean that somebody made a mistake. You may have:

• •short NIBUT but relatively normal staining
• •low Schirmer but reasonable tear meniscus height
• •significant gland dropout but some glands that express well
• •poor expression despite relatively preserved meibography
• •a thin lipid layer without dramatic gland loss
• •normal osmolarity despite other positive dry-eye findings
• •severe symptoms despite relatively mild conventional signs

Different tests examine different parts of the system. So disagreement can sometimes be clinically informative. The question becomes: why don’t these findings match?

Another reason results may not fit perfectly is that your eyes are not static. The tear film changes between blinks, throughout the day, with environment, with visual tasks, with humidity, with sleep, with medications and with treatment.

Some tests are particularly sensitive to measurement conditions. A tear sample is taken at one moment. A blink pattern is observed during one activity. A Schirmer test is performed under particular conditions. A meibography image captures a particular eyelid area with a particular device. So many dry-eye tests are best understood as snapshots of a dynamic system.

Even the examination itself can change later measurements. Bright light can influence blinking. Touching the ocular surface can stimulate reflex tearing. Fluorescein can alter the tear film. Manipulating the eyelids can affect subsequent measurements.

For this reason, TFOS DEWS III recommends performing tests in an order that generally moves from least invasive to more invasive. This is another reason why two examinations performed using different protocols may not produce identical results.

Diagnostic thresholds are useful. Doctors need decision points. But biology does not suddenly change when a measurement crosses a single number.

For example, if a diagnostic threshold is 10 seconds, results of 9.9 seconds and 10.1 seconds do not represent two completely different biological states. Likewise, an osmolarity result just below and just above a threshold should not be interpreted as healthy versus diseased without context.

Thresholds help classify measurements. They do not replace clinical reasoning.

Some dry-eye measurements depend substantially on the instrument being used. We have already seen this clearly with meibography. The same eyelids measured with different imaging systems can produce meaningfully different apparent dropout percentages. Lipid-layer measurements also vary between interferometry technologies. Osmolarity systems are not necessarily interchangeable. Even NIBUT can differ depending on the device and algorithm used.

So if you are monitoring your condition over time, ideally compare the same test, using the same method, preferably with the same device, under reasonably similar conditions. Otherwise, an apparent change may partly reflect measurement rather than biology.

Imagine Clinic A reports NIBUT: 5 seconds, and six months later Clinic B reports TBUT: 8 seconds. It may be tempting to conclude “My tear film improved by three seconds.” But NIBUT and fluorescein TBUT are not interchangeable measurements.

Or: one clinic reports 25% gland dropout and another reports 40%. If different devices, eyelid areas or analysis methods were used, those numbers cannot automatically be interpreted as disease progression.

Always ask: are we actually comparing the same measurement?

This is one of the most important things to understand about dry eye. Symptoms and objective signs show only modest relationships across populations.

A patient may have severe discomfort with relatively limited visible abnormalities. Another may have significant tear-film or ocular-surface abnormalities with surprisingly few symptoms.

That does not mean the first person’s symptoms are imaginary or the second person’s eyes are healthy. Sensory function, disease mechanisms, timing and many other factors influence what a patient feels. Symptoms and signs are different kinds of information. Both matter.

This can happen. A particular biological parameter may improve before symptoms do. Or the parameter that improved may not have been the dominant driver of symptoms.

For example, meibum may become easier to express while tear-film stability remains poor. Or surface staining may improve while pain processing remains altered. Or one driver may improve while another remains active.

An improved test is good information. But it does not guarantee immediate symptom improvement.

This can happen too. Symptoms can improve even when a particular structural measurement changes little.

Meibography is an obvious example. A structural image may remain relatively similar while meibum quality improves, tear-film stability improves, surface inflammation decreases, exposure is reduced, or sensory processing changes.

So symptom improvement is not invalid simply because one machine still produces an abnormal number.

There is no universal answer. The most important finding is usually the one that helps answer a clinically useful question. For example:

• •Do I actually meet the criteria for dry eye disease?
• •What part of tear-film homeostasis is abnormal?
• •Which driver appears most important?
• •Is there another condition that could explain my symptoms?
• •Is there something that needs treatment or further investigation?
• •Are we measuring something that can realistically change?

A dramatic-looking test is not automatically the most clinically important test.

Modern diagnostic devices can detect many abnormalities. But detecting something does not automatically mean it needs to be treated. Some findings may be mild, age-related, incidental, stable, or simply not the main driver of symptoms.

Treatment decisions should therefore not be based simply on “The machine found something abnormal.” A better question is: “Is there good reason to believe this finding is contributing meaningfully to my current problem?”

Instead of leaving only with numbers, try to leave with a model of what your doctor thinks is happening. Useful questions include:

• •Do I meet the criteria for dry eye disease?
• •What objective evidence of loss of homeostasis did you find?
• •Which findings are diagnostic, and which are possible drivers?
• •What do you think my main driver or drivers are?
• •Which findings are you less certain about?
• •Could any other condition explain my symptoms?
• •Which result is actually guiding the treatment recommendation?
• •How will we know whether the treatment is working?

These questions turn testing into clinical reasoning.