Blink Analysis

Blinking is not simply a count. Completeness, timing and duration together describe how effectively blinking restores and protects the tear film.

How to do the test

The patient should look straight ahead at a distant target and remain relaxed. The room should be quiet, with minimal or no air draft, and there should be no unnecessary movement in front of the patient. Room lights may remain on, but should not be excessively bright.

OSDx uses infrared (IR) illumination for blink analysis. This allows blink behaviour to be observed without visible illumination that could itself stimulate the eye and alter the natural blink pattern.

1. Capture the closed-eye image

At the start of the test, ask the patient to close the eye and click Capture.

For best results, position the patient so that the eye, from canthus to canthus, fills the frame. A good closed-eye reference image is important for reliable blink detection.

If the closed-eye image needs to be retaken, click Capture again.

2. Allow normal blinking to become established

After capturing the closed-eye image, ask the patient to open the eyes, continue looking at the distant target and blink normally.

Allow a short period for the patient to relax and for spontaneous blinking to stabilise. Avoid talking to the patient or creating visual distractions during this period.

3. Start the blink study

Once the patient is relaxed and blinking naturally, click Start test.

OSDx records the blink pattern for the selected study duration of 30 or 60 seconds. The study ends automatically when the selected duration has elapsed.

4. Check the quality of the recording

The blink graph provides an immediate visual indication of whether the recording is suitable for analysis.

In a good-quality study, distinct blink spikes should be visible against a relatively quiet baseline around the mean line.

If the baseline around the mean is filled with disturbances or substantial background activity, the recording may be unreliable. This can occur, for example, when the patient is moving or roving the eyes instead of maintaining steady fixation. In this situation, the study should be repeated.

Repeating or restarting the test

To interrupt and restart an ongoing blink study, click Start test again.
To retake the closed-eye reference image, click Capture again.

Tip: Natural blink behaviour is easily influenced by attention, conversation, visual activity and the surrounding environment. A relaxed patient, steady distance fixation and a quiet, draft-free environment help obtain a representative spontaneous blink recording.

Why analyse blinking?

Each blink redistributes the tear film, assists lipid spreading and briefly brings the upper and lower lid margins together. In dry eye disease, blink behaviour may change in response to ocular-surface discomfort or visual demand. A simple blink rate therefore provides only part of the picture: two patients with the same number of blinks per minute may have very different proportions of incomplete blinks, interblink intervals and lid-contact times.^1–3

OSDx therefore evaluates the pattern of blinking in addition to its frequency.

Complete and incomplete blinks

Complete blink

A blink in which the lids achieve complete closure. Complete blinking permits effective lid apposition and tear-film redistribution across the exposed ocular surface.

Incomplete blink

The lids approach one another but fail to close completely. Frequent incomplete blinking can leave part of the cornea repeatedly exposed and may impair tear-film redistribution. Incomplete blinking has been associated with shorter tear break-up time, greater dry-eye symptoms and meibomian-gland dysfunction.⁴

Why the distinction matters: blink rate alone can be misleading. A high blink rate does not necessarily mean effective ocular-surface protection if many of those blinks are incomplete. The complete-to-incomplete blink pattern is therefore clinically more informative than blink count alone.

Timing parameters

IBI

Interblink Interval is the time between successive blinks. A shorter mean IBI corresponds to more frequent blinking and may reflect reduced ability to maintain a stable, comfortable ocular surface between blinks. In one controlled visual-task study, mean IBI was 5.97 s in normal subjects versus 2.56 s in dry-eye subjects (P=0.004). Mean IBI showed an ROC AUC of 0.85 for distinguishing the two groups.¹

IBI SD

Standard deviation of IBI describes the regularity or variability of the blink rhythm. Blink timing is naturally variable, and the distribution contains information that is lost when only blink rate is reported. In the same pilot study, IBI variability differed significantly between normal and dry-eye groups and IBI SD alone achieved an ROC AUC of 0.75.¹ Thus an unusually repetitive or irregular blink sequence should be interpreted together with the mean IBI rather than judged from a single interval.

Blink duration

Blink duration measures how long a blink lasts. Prolonged lid closure may represent an adaptive response during ocular-surface stress and can influence visual function. Studies of extended blinks have found longer blink durations in dry-eye subjects; one study reported a median extended-blink duration 2.53 times longer in dry eye than in controls.² Another study found substantially greater mean lid-closure duration and total lid-contact time in dry-eye subjects.³

Duration SD

Standard deviation of blink duration indicates how consistent individual blink durations are during the recording. A larger SD means that short and prolonged closures are intermixed. This can reveal episodic extended or squeeze-like blinks that would be obscured by the mean duration alone. There is not yet a universally accepted diagnostic cut-off for blink-duration SD, so it should be regarded as a descriptive pattern metric and interpreted with the individual blink durations and clinical findings.^2,3

Lid Contact Time (LCT)

LCT represents the period during a complete blink for which the upper and lower lids remain in contact. It provides information that neither blink rate nor IBI can provide: how much effective closed-lid time is actually occurring.

This distinction can be clinically relevant. In a study comparing dry-eye and normal subjects, mean duration of full lid closures was 0.565 s versus 0.080 s, respectively, and total lid-contact time was markedly greater in the dry-eye group. The authors proposed lid-contact time as a potentially meaningful descriptor of altered blink behaviour and its effect on visual function.³

OSDx reports LCT as part of the blink pattern rather than treating every detected blink as equivalent. A patient may therefore have a seemingly acceptable blink rate but still show inefficient blinking because of incomplete closures, or an altered pattern because of prolonged lid contact.

Important: published blink-duration and lid-contact values depend strongly on the recording method, frame rate, blink definition and visual task. LCT should therefore be interpreted as a physiological pattern parameter and followed longitudinally under comparable test conditions rather than using a rigid universal normal cut-off.

Reading the OSDx blink result

Look first at the number and proportion of complete versus incomplete blinks.
Use mean IBI to assess how frequently the ocular surface requires a new blink.
Use IBI SD to assess the temporal pattern rather than relying only on average blink rate.
Review mean blink duration and its SD for unusually prolonged or heterogeneous closures.
Consider LCT as an additional measure of effective lid apposition during complete blinking.
Interpret blink findings together with tear-film stability, lipid-layer assessment, meibomian-gland findings and ocular-surface staining.

References

1. Johnston PR, Rodriguez J, Lane KJ, Ousler G, Abelson MB. The interblink interval in normal and dry eye subjects. Clin Ophthalmol. 2013;7:253–259. PMID: 23403736.

2. Rodriguez JD, Lane KJ, Ousler GW III, Angjeli E, Smith LM, Abelson MB. Investigation of extended blinks and interblink intervals in subjects with and without dry eye. Clin Ophthalmol. 2013;7:337–342. PMID: 23439943.

3. Rodriguez JD, Lane KJ, Ousler GW III, Angjeli E, Smith LM, Abelson MB. Blink patterns and lid-contact times in dry-eye and normal subjects. Clin Ophthalmol. 2013;7:1181–1189. PMCID: PMC4015796.

4. Wang MTM, Tien L, Han A, Lee JM, Kim D, Markoulli M, Craig JP. Evaluation of incomplete blinking as a measurement of dry eye disease. Ocul Surf. 2019. PMID: 31152804.