AUTOMATED PERIMETRY DR.JYOTI SHETTY MEDICAL DIRECTOR BANGALORE WEST LIONS EYE HOSPITAL, BANGALORE.

AUTOMATED PERIMETRY

DR.JYOTI SHETTYMEDICAL DIRECTOR

BANGALORE WEST LIONS EYE HOSPITAL, BANGALORE

BASIC CONCEPTS• Traquair's has defined the visual field as been a hill

island of vision in a sea of darkness testing along X-Y axes of this 3 dimensional area determines the location in the visual field and along the Z axis identifies the visibility threshold.

• X - Y axis - kinetic perimetry• Z axis - static perimetry • Automated Perimetry - " Differential light threshold –

Ability to differentiate an illuminated target against an illuminated background."

• Threshold Perimetry - Modality of choice

BASIC CONCEPTS --- contd

THRESHOLD • Luminance of stimuli that is seen 50% of times it

is presented• Logarithmic unit dB ( dB prop. 1 / brightness )• Bracketing strategy ( 4 - 2 - 2 algorithm )• Supra threshold - 95 % chance a stimulus is seen.• Infra threshold - 5% chance a stimulus is seen.

BASIC MACHINE DESIGN

• Illuminated hemispherical bowl 33 cm away with target of fixation

• Stimuli - spot of light - LED / Projection system / Comp. Video monitor

• HFA - II ( 700 Series ) Aspherical bowl 30 cms away ,smaller ,more ergonomic stimuli in periphery more closer,

programmed to decrease stimuli brightness (4dB).

FIXATION CONTROL• CC TV monitor• Heijl - Krakau Blind spot method• Gaze tracker

– Full time two variable Gaze monitor– Image analysis– Errors - upward / downward– Fixation checked 100 % of stim. Time– No testing time for fixation check

Basic software design

• Strategies for threshold detection – • Intensity of the stimulus presented at a

given point is related to the normal threshold at that stimulus site.

• Bracketing strategies to define threshold at any point.

4-2-2 algorithm SITA

INFORMATION DISPLAY• Numeric data display

actual dB value at each point

• Gray scale – range of decibels and their corresponding luminance

• Difference / Depth defect – actual value is arithmetically subtracted from a presumed expected field.

Parameters recommended for testing

• Foveal fixation target – small and large diamond with yellow lights.

• Goldmann size III target for stimuli & blind spot check. If excessive fixation loss it can be decreased to II or I or if vision less than 6/36 than it can be increased to V.

• White stimulus colour• Normal testing speed. can be slow down if patient is

slow to response.• Foveal threshold - ON / OFF

Threshold tests• Central 30-2 – 76 points are tested . Each point 6 deg

apart. Straddling the horizontal and vertical axis so that the 2 inner most test points are 3 deg from fixation point.

• Central 24-2 – 56 points are tested . Avoids rim artifacts.

• Central 10-2 – 68 points space 2 deg apart. Useful in advance disease with spilt fixation.

• Macular threshold test – square grid of 16 points each 2 deg apart , with each point thresholded 3 times.

30 – 2 24 – 2 Macular threshold

INTERPRETATION

Factors for consistency in testing • Best Refractive correction used. Contact lens to avoid

rim artifacts.• Pupil Diameter – at least 3.5 mm in size.• Visual Acuity• Date & Time of testing• Age-For comparison with normative data• Short term fluctuation-Fluctuation occurring within

the test. Should be <3dB.

INTERPRETATION ----contd.Reliability of patient • Fatigue, anxiety and learning effect• Fixation loss – should be less than 20%• False positive and negative response should

be less than 33%.

FIXATION LOSSES: 10/12 XXFALSE POS ERRORS : 0/7 FALSE NEG ERRORS :0/6 TEST DURATION : 16:20

Statistical global indices

• MD – mean deviation – sensitive to total loss

• PSD – pattern standard deviation – sensitive to localized loss.

• CPSD – corrected pattern standard deviation – PSD corrected for short term fluctuation. Very sensitive index.

Glaucoma defect with automated perimetry- Anderson's Criteria

• 3 or more cont.non edge points with >= 5 dB loss

• 2 or more cont. non edge points with >=10 dB loss

• Diff. of 10 dB across nasal hor. meridian at 2 or more adj. points ( nasal step.)

• GHT - ONL• PSD plot - >= 3 pts , p< 5%

of which one < 1%• CPSD ( p <5% ) GHT ONL

INTERPRETATION ----contd.• Progression of defect• Test parameters comparable• Defect - increased in size / depth• >= 7 dB increase in depth of existing defect• >= 9 dB depression adj. to abnormal point• >= 11 dB depression of a normal point ( New Defect )• Box plot change analysis• Overview• Glaucoma change probability analysis

SWAP• Tests subset of Ganglions affected earlier & selectively --

Blue / Yellow• Reduces the redundancy of responsiveness to stimuli• Intense yellow background - bleaches green / red cones• Blue stim. ( 440nm ) - isolates blue cones• Adaptation - 3 mts. Room illumination - minimal• Stimulus size & BS check size V• Mean threshold values lower than SAP - Gray scale darker• Stat Pac probability plots more reliable

SWAP -- contd

• Field defect precedes SAP by >= 3 yrs • Once abnormal - remain abnormal ( no recovery of damaged

blue cones )• No role in advanced POAG / advanced lenticular changes /

colour vision abnormalities• Most useful in younger Glaucoma suspects, OHT , POAG

with mild to mod.damage• Time consuming - SITA optimised for SWAP / Fast Pac can

be used