Perimetry Principles

Kinetic vs. Static Perimetry

Kinetic perimetry, performed classically with a Goldmann perimeter, moves a target of fixed size and luminance from a non-seeing peripheral area inward until the patient first detects it. The boundary connecting all points of equal sensitivity for that target is an isopter. Testing with several differently sized or differently bright targets produces a set of isopters that together map the sensitivity gradient from the periphery toward fixation. Because the examiner controls target size, speed, and path in real time, kinetic perimetry adapts on the fly: it suits patients with severely constricted fields, complex neurological field defects, or poor fixation and cognitive ability who cannot reliably complete an automated test. It also maps sensitivity well beyond the 24-2 or 30-2 test grid used by standard automated perimeters.

Static automated perimetry, the Humphrey Field Analyzer being the standard example, instead presents stationary stimuli of varying luminance at fixed test locations and determines the threshold (the minimum detectable luminance) at each point individually, using a staircase or Bayesian algorithm. The result is a point-by-point sensitivity map rather than a set of isopter boundaries. This is the clinical gold standard for detecting and monitoring glaucomatous field loss, since it produces quantitative, reproducible threshold data that can be compared visit to visit to track change over time. Kinetic testing, with its examiner-dependent target path, does this much less reliably.

SITA and Automated Threshold Testing

Full threshold testing, measuring every point's threshold from scratch with a classic staircase method, is accurate but slow. The Swedish Interactive Threshold Algorithm (SITA), used in SITA Standard and SITA Fast protocols, uses a Bayesian approach instead: it starts from a prior probability model of what a normal (or already partially tested) visual field should look like, then adjusts each subsequent stimulus based on the patient's actual responses. Total test time drops by roughly half compared to full threshold testing, with comparable accuracy. That tradeoff is why SITA has largely replaced full threshold testing in routine clinical use.

Reliability Indices

Every automated field printout reports reliability indices, and these need to be checked before the result can be trusted. The false positive rate reflects how often the patient presses the response button when no stimulus was actually presented. A high false positive rate marks a "trigger-happy" responder and can artificially inflate the field, making it look better (brighter, more sensitive) than it really is. The false negative rate and fixation loss rate are the complementary checks: false negatives measure failure to respond to a stimulus at a location already shown to be seeable, and fixation losses, tracked via the Heijl-Krakau method using the physiologic blind spot, flag whether the patient held steady fixation throughout the test. Check all three before drawing conclusions from an otherwise abnormal-looking field.

Humphrey Global Indices

Two summary numbers on the Humphrey printout condense the whole field into single values. Mean deviation (MD) is the average deviation, in decibels, of the patient's measured thresholds from an age-corrected normal database across all test points: near 0 dB is normal, roughly −2 to −6 dB reflects mild loss, −6 to −12 dB moderate loss, and below −12 dB severe generalized depression. Pattern standard deviation (PSD) measures something different, the irregularity, or focality, of the loss. PSD rises when a focal defect stands out against a relatively preserved background, as in early glaucoma, even while overall MD is still normal. It stays low in generalized, uniform depression such as media opacity from cataract, where MD may be reduced but the pattern of loss across the field is even rather than focal. Reading MD and PSD together, rather than either alone, is what actually distinguishes "diffuse" loss from "focal" loss.

The glaucoma hemifield test (GHT) is a related, automated interpretation tool: it compares five corresponding sectors in the superior and inferior hemifields to detect asymmetric loss between them. A result of "outside normal limits" flags a significant superior-inferior asymmetry, the pattern expected from nerve fiber bundle damage, which tends to affect one hemifield disproportionately. That asymmetry is characteristic of early glaucomatous field loss, sometimes appearing before MD itself becomes clearly abnormal.

Frequency Doubling Technology

Frequency doubling technology (FDT) perimetry is a separate, screening-oriented test. It uses a low-spatial-frequency sinusoidal grating flickering at a high temporal frequency, which produces an illusory doubling of the number of bars the patient perceives. This illusion is selectively mediated by the magnocellular (M-cell) pathway, large-diameter retinal ganglion cells that are disproportionately affected early in glaucoma. That selectivity is what makes FDT a fast, sensitive glaucoma-screening tool, even though it does not replace standard automated perimetry for definitive diagnosis or monitoring.

Short-wavelength automated perimetry (SWAP) works on the same logic but targets a different pathway. A blue stimulus is presented against a bright yellow background; the yellow light desensitizes the long- and middle-wavelength cones, leaving only the short-wavelength (S-cone, blue-yellow) system to respond. That system feeds a small population of bistratified retinal ganglion cells that also drop out early in glaucoma, so SWAP can pick up field loss before it appears on standard white-on-white testing. The tradeoff is a longer test and results that vary more from visit to visit, which is why SWAP supplements standard automated perimetry rather than replacing it.

Related Topics

Continue exploring Clinical Examination Skills: