Techniques and variables in visual field testing

A number of factors can affect a visual field test other than the disease being studied. It is important to minimize the influence of these variables as completely as possible in order to assess accurately and document precisely the abnormalities present so that deviations from normal and future changes can be recognized easily. Fluctuation, the combination of normal physiologic variability and measurement error, complicates the recognition of pathologic change. If all other possible variables are eliminated, then change in the disease must be responsible for any alteration in the visual field. Unfortunately it is impossible to eliminate all other variables. Awareness of factors influencing the visual field, however, can help minimize these variables and improve interpretation ( Box 9-1 ).

Box 9-1

Examination artifacts

  • Technician: results vary with different technicians

  • Equipment: results vary with different equipment

  • Test: results vary with different types of tests

  • Software: results vary with different testing or interpretation algorithms

Eye artifacts

  • Refraction: should have distance prescription with proper addition for near vision

  • Pupil size: should be 3 mm or more; must be consistent

  • Fixation: results vary with quality of fixation control

  • Media opacity: visual acuity should be recorded

Patient artifacts

  • Misunderstanding the test

  • Fatigue

  • Inattentiveness

  • Physiologic/pathologic/psychologic/mental status

  • Systemic illness, hangover, anxiety, and so on

Analysis artifacts

  • Is the visual field normal? Requires standards for normal

  • Has the visual field changed? Requires knowledge of fluctuation

  • Misinterpretation

Some artifacts that affect visual field results



With age, the visual field has a linear decrease in sensitivity and the slope steepens. The increase in fluctuation that occurs as the test moves toward the periphery is also greater with age. The combined effect of these variables is a field with an increasingly steep slope as one moves away from fixation. The effect of age on the central field is gradual and can usually be ignored in the evaluation of individual patients. Mean sensitivity of the visual field decreases approximately 0.58–1.0 dB per decade. Increased age may be associated with increased variability in repeated test results over time. The standard deviation of the mean sensitivity of points tested ranges from about 1.0 to 2.0 dB in the normal central field of a young patient. Patients older than 60 years of age may have standard deviations up to twice that amount centrally and up to 10 dB per point at 30° eccentricity.


Most patients maintain acceptable fixation on the central target of the perimeter. Patients with poor fixation can be encouraged to stabilize their fixation, but this does not always prevent them from looking around. Technicians and computerized systems usually monitor or grade patient fixation in some way. Machines that monitor fixation continuously, generally by some form of eye movement or pupillary reflex assessment, often have algorithms that disregard responses generated during fixation losses. These machines return to the same test location later during the examination and present the stimulus again. This programming feature helps ensure that the responses recorded by the machine occur during periods of steady fixation. Other machines use a monitoring system with blind spot fixation in which stimuli are projected on the physiologic blind spot at intervals throughout the test. If fixation is steady these stimuli will continue to land on the blind spot and will not be detected. If fixation has shifted the stimuli will land on photoreceptors and be detected. The machine records and/or alerts the operator that a fixation loss has occurred. If the patient generates fixation losses more than 20–30% of the time, the test can be considered only an approximation of the true visual field.


In addition to monitoring fixation, patient reliability should be graded as good or poor by the technician. Computerized machines can provide some index of reliability based on false-positive or false-negative responses and fixation losses. Fatigue, drugs, age, and illness can all affect patient reliability and must be considered when assessing the accuracy of a given test. Even in well-rested patients, the test itself can be fatiguing, so that reliability tends to decrease with prolonged or extensive examination sessions.

Although long test sessions can fatigue a patient, experience with the machine usually decreases variability over repeated testing sessions (learning curve). Thus the first visual field may be the least accurate. Patients with experience on manual perimeters may have less of a learning curve effect. We tend to repeat the initial test if the results are abnormal in any way. Although computerized machines are automated, they are not automatic. Patient/technician interaction can have a substantial impact on the reliability of the examination and may also aid in patient satisfaction.



A pupillary diameter of less than 3 mm can cause generalized depression of the visual field. It is usually best to test the field with a pupil that is at least 3 mm in diameter. If it is not possible to dilate the pupil to 3 mm, the test should be performed with a pupil that is no smaller than that which existed during previous tests.


Any opacity of the ocular media can cause a localized or generalized depression in the visual field. This is particularly problematic when following a glaucoma patient who is developing cataracts. As the lens opacity become denser, field defects may appear to enlarge or become denser because of the reduced amount of light reaching the retina or because of image distortion or light scattering. Patterns of localized loss tend to remain consistent before and after cataract extraction, however. Visual acuity, refraction, and the appearance of the lens can help in determining the influence of cataract on the field. If acuity has dropped by more than one line on the Snellen chart, the examiner should suspect that the cataract is accentuating the appearance of visual field defects. Some analysis programs compensate for this reduction by factoring out generalized depression from the visual field so that scotomata are exposed ( Fig. 9-1 ). However, the pattern deviation is not perfect in identifying purely localized defects if the cataract is significant. The most reliable criteria for identifying glaucomatous loss appear to be a glaucoma hemifield test ‘outside normal limits’, two hemifield clusters that fall below the 5% population limit, and four abnormal points (<5%) in a hemifield on the pattern standard deviation plot.

Fig. 9-1

Developing cataract in a glaucoma patient. Note the increasing depression of the visual field in the left grey-scale printout and in the total deviation graphic presentation (third from the left). The pattern deviation that is presented in the far right column shows little change over time. Cataract extraction with intraocular lens implantation was performed prior to the last field test. Note that the generalized depression and the total deviation have reversed while the pattern deviation remains similar. This methodology allows improved ability to follow glaucoma patients in the presence of developing cataracts.


Proper refraction with appropriate correction for presbyopia and patient age are required for accurate testing. In one series of experiments, overcorrection of +1.00 D in the sphere reduced mean sensitivity 3.6 ± 0.8 dB, and overcorrection of +2.00 D caused a reduction of 5.3 ± 0.9 dB. Another study found a decrease in threshold sensitivity as high as 7.6 dB with +6.00 D overcorrection.



It is virtually impossible for two technicians to administer a manual visual field examination in precisely the same manner. Even the same technician inadvertently varies technique at least slightly from one examination to another. Thus to accurately interpret visual fields, the interpreter must be familiar with the skills and variations of the technician performing the tests. The technician can improve patient performance by monitoring the patient consistently during the examination, but this has few advantages over intermittent monitoring following a brief introductory orientation. Computerized perimetry has a great advantage over manual perimetry because it allows repeated performance of a standardized test. However, technical supervision is mandatory during the entire test to ensure the best possible results.


The level of background illumination affects the contour of the hill of vision and thus the appearance of the visual field. Brighter background illumination increases the slope of the central field and may influence the appearance of field defects. Different perimeters have different backgrounds: the Humphrey, Goldmann, and more recent Octopus perimeters use 31.5 apostilbs of background illumination. By contrast, older Octopus machines used 4 apostilbs of background illumination.


Obviously, increasing the size or intensity (brightness) of the stimulus projects more light onto the retina, whereas the opposite is true with a smaller or dimmer stimulus. For comparison with previous visual fields it is important to use the same stimuli as were used initially. This is relatively straightforward with kinetic perimetry because the size and intensity of the stimulus defines the isopter. Mistakes can occur, however, if the technician uses the same color ink to represent different isopters on various occasions. Standardization is crucial in visual field testing both for obtaining and recording results.

Static automated perimetry reports resemble each other superficially regardless of the stimulus size used during the test. A change in stimulus size can produce a different test result, however. The authors use a Goldmann size III stimulus for virtually all conventional automated perimetric testing. The small benefit gained by using non-standard testing parameters is rarely worth the risk of forgetting to reset the machine to standard parameters before administering the next test. However, if the visual field at a size III test object is almost totally black, some useful information and monitoring potential can be obtained with a size V test stimulus. If an apparent large change occurs between tests, however, the examiner should check the stimulus used in order to ensure that identical parameters were employed ( Fig. 9-2 ).

Fig. 9-2

These field tests were taken 1 month apart. Field (A) stimulus size III was used first and shows a dense arcuate defect superonasally.

Field (B) stimulus size V shows a subtle field defect superonasally in the grey scale. Care must be taken not to be misled by the change in stimulus size when interpreting a field change.


Temporal summation, the ability of the visual apparatus to accumulate information over time, can influence visual field testing for stimulus exposure times less than 0.5 seconds ( Fig. 9-3 ). This principle is used in frequency-doubling perimetry where the time between a pair of stimuli is in the range of 50–100 ms. Most manual perimetry is performed with exposure times of approximately 1 second, so temporal summation has little influence.

Fig. 9-3

Temporal summation related to background luminance. Using the 10 dB background curve, approximately a 3 dB increase ({) can be anticipated by increasing stimulus exposure time from 0.1 to 0.2 seconds.

Modified from Aulhorn E, Harms H. In: Jameson D, Hurvich LM: editors: Handbook of sensory physiology, vol 7, New York, Springer-Verlag, 1972. With kind permission of Springer Science and Business Media.


To compare visual field charts, the same region of the visual field must be tested during serial examinations ( Fig. 9-4 ). For most purposes, tests that examine alongside vertical and horizontal meridians are more useful than are tests that examine on the meridian (the latter are rarely used today).

Fig. 9-4

Although these two visual fields appear to reveal tremendous change (see Fig. 9.4B ), the tests were performed on the same day. (A) The 30-1 visual field positions the spot on the horizontal and vertical midlines and then positions subsequent spots 6° off these midlines.

(B) The 30-2 program positions the spots 3° to either side of the midlines. The 30-2 program is better for recognizing change along horizontal and vertical meridians such as those that occur in patients with glaucoma and neurologic deficits.



Regardless of the equipment used, there are certain fundamental requirements for accurate visual field testing. Accurate distance refraction, with the appropriate addition for the distance from patient to stimulus, should be used. Because accommodative capacity varies with age, the amount of addition should be adjusted for the patient’s age and the instrument used ( Table 9-1 ).

Table 9-1

Addition for near, bowl perimetry

Age Octopus, Model 500 Octopus, Model 201 Humphrey analyzers Goldmann perimeter
30–39 Plano Plano +1.00 +0.50
40–44 +1.00 +0.50 +1.50 +1.00
45–49 +1.25 +1.00 +2.00 +1.50
50–54 +1.75 +2.00 +2.50 +2.25
55–59 +2.00 +2.00 +3.00 +3.00
>59 +2.00 +2.00 +3.00 +3.50

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Feb 12, 2019 | Posted by in OPHTHALMOLOGY | Comments Off on Techniques and variables in visual field testing

Full access? Get Clinical Tree

Get Clinical Tree app for offline access