Frequency Doubling Technology

From EyeWiki
Original article contributed by: Helen Jiang, MD, Reena Garg, MD
All contributors: Helen Jiang, MD and John Davis Akkara
Assigned editor:
Review: Not reviewed
{| cellspacing="5"

|-!align="right" |Lead Editors: |add |- !align="right" |Contributing Editors: |add |-


Summary[edit | edit source]

Various testing modalities are currently available for detecting visual field loss. This page will focus specifically on the role of frequency doubling technology in assessing visual fields.

Frequency doubling technology (FDT) perimetry is based on a flicker illusion created by counterphase flickering of a low spatial frequency sinusoidal grating at a high temporal frequency.[1] This phenomenon essentially creates an image that appears double its actual spatial frequency.

The frequency doubling illusion has previously been attributed specifically to the nonlinear response of the My cells in the magnocellular layer.[2] This is significant because it suggests the ability of FDT to selectively detect loss of function in certain parts of the nerve fiber layer. However, several studies have questioned whether such a separate group of nonlinear ganglion cells truly exist, and it has instead been proposed that frequency doubling may be due to mechanisms elsewhere in the visual pathway (such as the cortex) rather than non-linear cells based in the retina.[3],[4]

The field of vision and characteristics of visual field loss in glaucoma[edit | edit source]

The normal field of vision encompasses approximately 50 degrees nasal and superior, 70 degrees inferior, and 90 degrees temporal.[5] Sensitivity is greatest in the middle and declines toward the periphery, commonly referred to as the “hill of vision”. Visual field defects are considered to be any significant change in sensitivity from this normal hill of vision.

Early glaucomatous nerve damage is thought to be due to loss of large-diameter retinal ganglion cells, namely the My cells of the magnocellular layer.[6],[7] These cells are responsible for a small proportion of the total retinal ganglion cells and have minimal functional redundancy, making them a good target for detecting early glaucoma damage.[7],[8]

The earliest visual field defects in glaucoma have been described as the nasal step or paracentral scotoma, followed by extension of these defects centrally over time.[9] The inferior and superior nerve sectors have been reported to have the greatest amount of optic nerve damage in glaucoma. It has been postulated that the structure of the lamina cribrosa lends axons passing through these poles to be more susceptible to distortions caused by elevated IOP.[10]

Hart and Becker have described three phases of glaucomatous nerve damage.[9] Occult damage occurs in the first phase without detectable visual field loss. This may be due to the redundancy of the nerve fibers.[11] In the second phase, visual field loss is barely detectable, though it has been noted that deficits in color vision and decreased contrast sensitivity may be detected. In the third phase, visual field defects become manifest and extend to cover an entire Bjerrum, or arcuate region as they progress.[9]

Devices and Algorithm[edit | edit source]

The 2 FDT perimeters available are the first generation Frequency Doubling Technology (Welch Allyn, Skaneateles, NY) and the second generation Humphrey Matrix (Carl Ziess Meditec, Inc., Dublin, CA). The machines are portable and can be set on a flat surface at sitting level for use.

1st generation FDT perimetry
There are two different presentation patterns using the original FDT perimeter. The C-20 tests 17 different points within a central 20-degree radius (4 targets in each quadrant and one central target), while the N-30 incorporates two additional nasal targets above and below the horizontal meridian.[12],[13],[14] Both patterns can be used in screening or threshold testing. Good sensitivity and specificity for detecting glaucomatous change has been described using both screening and threshold testing.[15],[16],[17]

In the first generation FDT perimeter, each target is displayed as a 10-degree-diameter square, with the exception of the central stimulus which is presented as a 5-degree-diameter circle. The 0.25 cycles/degree sinusoidal grating undergoes counterphase flicker at 25 Hz and lasts for 720msec. There is first a 160msec increase in contrast, then 400msec of steady presentation at the set contrast, followed by a 160msec decrease in contrast. Consecutive stimuli are separated by up to 500msec of variable pause in order to reduce the rhythmic responses. A button press by the patient between 100msec and one second after presentation is counted as stimuli detection, at which time the current stimulus will be discontinued to order to shorten testing time.[15]

The original FDT perimeter threshold testing utilizes a four-reversal staircase procedure known as the modified binary search algorithm (MOBS).[12] In this way, contrast is decreased if the stimulus is detected at a higher level, or contrast is increased until a stimulus is detected.[15]

Different types of screening protocols are available which vary by contrast level.

The N30-1 screening test presents stimuli that can be detected by 99% of the normal population adjusted for age.[14] If the initial target is not detected, the same stimulus is presented again. If the stimulus is again not detected, the target will be presented at a level detected by 99.5% of the normal population. If this target is not seen the stimulus is presented as maximum contrast. Sensitivity for the N30-1 test is been reported to be between 78-92% and specificity between 85-100%.[13] The high specificity associated with this test suggests it may be a useful option for large population screening.

The N30-5 screening test presents stimuli that can be detected by 95% of the age-adjusted normal population.[14] If the initial target is not detected, the same stimulus is presented again. It the stimulus is again not detected, the target will be presented at a level detected by 98% of the normal population. If this target is not seen the stimulus is presented at the 99% detection level. The sensitivity for this test is reported to be 85-95% and the sensitivity between 80-90%.[13] The high sensitivity associated with this screening test suggests it to be better for detecting earlier visual field loss in glaucoma patients, rather than as a screening tool for the general population.

Both the N30-1 and N30-5 tests take about half a minute in each normal eye and up to 2 minutes for eyes with visual field defects.[13] Tests with targets at higher contrast levels are more specific for detecting patients with visual field defects while tests with lower contrast levels are more sensitive.

Second generation Humphrey Matrix FDT perimetry
The second generation Humphrey Matrix FDT Perimeter introduced several new features, including the ability to monitor eye fixation. Threshold testing on the Humphrey Matrix uses smaller targets that are presented along a grid. In addition, greater spatial resolution is available with 24-2, 30-2, 10-2, and macular threshold tests. The 24-2 and 30-2 threshold tests present 5 degree-diameter targets at a higher frequency 0.50 cycles/degree sinusoidal grating that undergoes counterphase flicker at 18Hz.[13] The 10-2 and macular tests present 2-degree-diameter targets at a counterphase frequency of 12Hz.[12] The N-30 threshold testing utilizes a two-reversal MOBS algorithm while the other threshold tests unique to the Humphrey Matrix utilize a Bayesian threshold estimation strategy (ZEST) similar to the Swedish Interactive Threshold Algorithm (SITA) used in the Humphrey Field Analyzer.[13] The 24-2 screening protocol, available on the Humphrey Matrix, utilizes the same principle as the N30 tests but presents 54 targets in a grid bracketing the vertical and horizontal meridians.[13]

Testing methods[edit | edit source]

During FDT perimetry testing, the patient is presented different flicker targets and instructed to press the response button at the appearance of each stimulus. Due to the relatively large targets used in FDT perimetry, refractive errors up to 6 diopters are not considered to significantly influence test results.[14] Patients are able to wear their own corrective lenses while performing the test.

As described in the above section, the FDT perimeter can be used for screening or threshold testing. Screening protocols vary by contrast level. Tests with targets at higher contrast levels are more specific for detecting patients with visual field defects while tests with lower contrast levels are more sensitive. Both the N30-1 and N30-5 screening tests take about half a minute in each normal eye and up to 2 minutes for eyes with visual field defects.[13] Threshold testing gives the lowest contrast sensitivity required to detect the FDT stimulus at each point. Threshold testing takes approximately 5 minutes per eye using the Humphrey Matrix 24-2 protocol.[13]

Interpretation[edit | edit source]

Results from FDT perimetry are reported using decibels of sensitivity. FDT perimetry has a manufacturer-provided internal normative database. The global indices that are reported are the mean deviation (MD) and pattern standard deviation (PSD). Mean deviation (MD) reports the amount the entire field deviates from normal, and pattern standard deviation (PSD) demonstrates irregularities within the field such as a localized defect.[5]

Quigley noted that the presence of two or more defective locations on FDT screening testing gave the best performance criterion with sensitivity of 91% and specificity of 94% for detecting patients with glaucomatous field loss.[16] Patel et al. created an algorithm that gave more weight to locations near the center as well as denser field defects and reported sensitivity of 80.5% and specificity 95% for detecting glaucomatous changes using this formula.[18] Another study looked at previously suggested algorithms for FDT interpretation in the literature and found no significant difference in diagnostic performance.[19]

Reliability indices
False-positives are detected when a patient responds to a stimulus of zero contrast. False-negatives are recorded when a patient does not respond to a stimulus of maximum contrast, though such a result may be recorded in patients with significant vision loss who are responding reliably.[14] Fixation losses are noted when a patient detects a 1 degree target at 50% contrast in the area of the physiological blind spot. As the first generation FDT perimeter does not have the ability to monitor fixation, it is possible that patients with stable fixation outside the expected position range will be recorded as having fixation loss.[14] Due to the fact that only three trials are used to calculate each reliability index, it has been recommended that tests with even one false-positive be repeated to ensure more accurate results.[14]

FDT vs SAP[edit | edit source]

Many studies evaluating FDT use in glaucoma have used standard automated perimetry (SAP) as the gold standard to compare visual field loss detection. FDT results have been reported to correlate strongly with defects detected using SAP.[20],[21] A study comparing second generation FDT 24-2 testing with SAP 24-2 testing demonstrated strong correlation in patients with early to moderate glaucomatous field loss.[22]

Beyond simple correlation with SAP, several studies have shown FDT to be more sensitive for detecting early visual field loss compared to SAP when structural optic disc abnormality or RNFL appearance are used to define glaucomatous neuropathy.[23],[24],[25]

A longitudinal study by Medeiros et al. followed a group of glaucoma suspects with baseline normal visual fields using SAP testing.[26] The authors found that 59% of the patients who subsequently developed defects on SAP had FDT abnormalities that preceded the SAP visual field losses by up to 4 years. However, the study also reported that 18% of SAP converters were not found to have repeatable abnormalities on FDT.[26]

One advantage of FDT perimetry is that it can be completed faster than using the SAP SITA-FAST strategy.[27] The FDT machine is also more portable compared to the Humphrey Field Analyzer. FDT has been reported to have both lower intra- and inter-test variability compared to SAP, which suggests it may be a useful test to monitor long-term progression of visual field loss.[28]

Uses in clinical practice[edit | edit source]

Frequency doubling technology has been presented by many as an effective method for detecting glaucomatous changes given the relatively inexpensive, efficient, and non-operator dependent nature of the test. FDT also allows subjects to wear their own spectacle correction and is also unaffected by blur up to 6 or 7 diopters. The FDT machines are relatively portable (Humphrey FDT weight 19 pounds and Humphrey Matrix 30 pounds)[29] and thus may lend itself to use in community screenings for glaucoma.

Several studies in the literature have demonstrated the FDT perimeter to be a reliable and efficient way to screen for glaucomatous changes.[7],[30] Quigley reported the average testing time for one eye using FDT perimetry to be 1.8 +/- 0.7 minutes, supporting the efficiency of this test.[16] The same author also noted that the presence of two or more defective locations on FDT testing gave the best performance criterion with sensitivity of 91% and specificity of 94% for detecting patients with glaucomatous field loss.[16] A prospective trial performed by Cello et al. compared FDT perimetry results from 254 normal eyes with 230 eyes with early, moderate, or advanced glaucoma.[15] The authors demonstrated sensitivity and specificity levels of over 97% when using FDT perimetry to detect moderate to advanced glaucoma damage. Sensitivity and specificity for detecting early glaucomatous changes were reported to be 85% and 90%, respectively.[15]

Studies have reported that FDT use for glaucoma mass screening demonstrates reasonable performance given the resulting positive predictive value.[31],[32]

Beyond screening for and monitoring the progression of glaucoma, FDT perimetry may also be applied for use in other ophthalmic pathologies. FDT has also been shown to correlate with SAP in detecting visual field defects secondary to neuro-ophthalmologic disorders.[33] Reduced sensitivity on FDT perimetry has been reported in patients with diabetes compared to age matched controls,[34],[35],[36] suggesting a role for FDT perimetry in screening for diabetic retinopathy.

References[edit | edit source]

  1. Kelly DH. Frequency doubling in visual responses. J Opt Soc Am. 1966;56:1628–1633.
  2. Maddess T, Henry GH. Performance of nonlinear visual units in ocular hypertension and glaucoma. Clinical Vision Science. 1992;7:371-383.
  3. Zeppieri M, Demirel S, Kent K, Johnson CA. Perceived spatial frequency of sinusoidal gratings. Optometry and Vision Science. 2008;85:318-329.
  4. White AJR, Sun H, Swanson WH, Lee BB. An examination of physiological mechanisms underlying the frequency-doubling illusion. Invest Ophthalmol Vis Sci. 2002;43:3590-3599.
  5. 5.0 5.1 Heijl A, Patella VM. Essential Perimetry: The Field Analyzer Primer. 3rd ed. Dublin, CA: Carl Zeiss Meditec;2002.
  6. Quigley HA, Dunkelberger GR, Green WR. Chronic human glaucoma causing selectively greater loss of large optic nerve fibers. Ophthalmology. 1988;95:357-363.
  7. 7.0 7.1 7.2 Johnson CA, Samules SJ. Screening for glaucomatous visual field loss with frequency-doubling perimetry. Investigative Ophthalmology & Visual Science. 1997;38:413-425.
  8. Johnson CA. Selective versus nonselective losses in glaucoma. J Glaucoma (Suppl).1994;3:S32-S44.
  9. 9.0 9.1 9.2 Hart WM, Becker B. The onset and evolution of glaucomatous visual field defects. Ophthalmology. 1982;89:268-279.
  10. Quigley HA, Addicks EM, Green WR. Optic nerve damage in human glaucoma III. Quantitative correlation of nerve fiber loss and visual field defect in glaucoma, ischemic neuropathy, papilledema, and toxic neuropathy. Arch Ophthalmol. 1982;100:135-146.
  11. Kerrigan-Baumrind LA, Quigley HA, Pease ME, Kerrigan DF, Mitchell RS. Number of ganglion cells in glaucoma eyes compared with threshold visual field tests in the same persons. Invest Ophthalmol Vis Sci. 2004;41:741-748.
  12. 12.0 12.1 12.2 Zeppieri M, Johnson CA. Frequency doubling technology (FDT) perimetry. Published 2008. Accessed October 1, 2015.
  13. 13.0 13.1 13.2 13.3 13.4 13.5 13.6 13.7 13.8 Johnson CA. FDT perimetry for the detection of glaucomatous visual field loss. Glaucoma Today. 2008:26-28
  14. 14.0 14.1 14.2 14.3 14.4 14.5 14.6 Anderson AJ, Johnson CA. Frequency-doubling technology perimetry. Ophthalmol Clin N Am. 2003;16:213-225.
  15. 15.0 15.1 15.2 15.3 15.4 Cello KE, Nelson-Quigg JM, Johnson CA. Frequency doubling technology perimetry for detection of glaucomatous field loss. American Journal of Ophthalmology 2000;129:314-322.
  16. 16.0 16.1 16.2 16.3 Quigley HA. Identification of glaucoma-related visual field abnormality with the screening protocol of frequency doubling technology. American Journal of Ophthalmology. 1998;125:819-829.
  17. Burnstein Y, Ellish NJ, Magbalon M, Higginbotham EJ. Comparison of frequency doubling perimetry with humphrey visual field analysis in a glaucoma practice. American Journal of Ophthalmology.2000;129:328-333.
  18. Patel SC, Friedman DS, Varadkar P, Robin AL. Algorithm for interpreting the results of frequency doubling perimetry. American Journal of Ophthalmology.2000;129:323-327.
  19. Muskens RPHM, Heeg GP, Jansonius NM. An evaluation of algorithms designed to classify the results from frequency doubling perimetry. Ophthal Physiol Opt. 2004;24:498-503.
  20. Casson R, James B, Rubinstein A, Ali H. Clinical comparison of frequency doubling technology perimetry and Humphrey perimetry. Br J Ophthalmol..2001;85:360-362.
  21. Artes PH, Hutchison DM, Nicolela MT, et al. Threshold and variability properties of matrix frequency-doubling technology and standard automated perimetry in glaucoma. Invest Ophthalmol Vis Sci. 2005;46:2451–2457.
  22. Leeprechanon N, Giangiacomo A, Fontana H, Hoffman D, Caprioli J. Frequency-doubling perimetry: comparison with standard automated perimetry to detect glaucoma. Am J Ophthalmol. 2007;143:263-271.
  23. Sample PA, Bosworth CF, Blumenthal EZ, Girkin C, Weinreb RN. Visual function-specific perimetry for indirect comparison of different ganglion cell populations in glaucoma. Invest Ophthalmol Vis Sci.2000;41:1783-1790.
  24. Brusini P, Salvetat ML, Zeppieri M, Parisi L. Frequency doubling technology perimetry with the humphrey matrix 30-2 test. J Glaucoma. 2006;15:77-83.
  25. Racette L, Medeiros, FA, Zangwill LM, Ng D, Weinreb RN, Sample PA. Diagnostic accuracy of the matrix 24-2 and original N-30 frequency-doubling technology tests compared with standard automated perimetry. Invest Ophthalmol Vis Sci. 2008;49:954-960.
  26. 26.0 26.1 Medeiros, FA, Sample PA, Weinreb RN. Frequency doubling technology perimetry abnormalities as predictors of glaucomatous visual field loss. American Journal of Ophthalmology. 2004;137:863-871.
  27. Wadood AC, Azuara-Blanco A, Aspinall P, Taguri A, King AJW. Sensitivity and specificity of frequency-doubling technology, tendency-oriented perimetry, and humphrey swedish interactive threshold algorithm-fast perimetry in a glaucoma practice. American Journal of Ophthalmology.2002;133:327-332.
  28. Spry PGD, Johnson CA, McKendrick AM, Turpin A. Variability components of standard automated perimetry and frequency-doubling technology perimetry. Invest Ophthalmol Vis Sci.2001;42:1404-1410.
  30. Trible JR, Schultz RO, Robinson JC, Rothe TL. Accuracy of glaucoma detection with frequency-doubling perimetry. Am J Ophthalmol 2000;129:740-745.
  31. Iawsaki A, Sugita M. Performance of glaucoma mass screening with only a visual field test using frequency-doubling technology perimetry. American Journal of Ophthalmology.2002;134:529-537.
  32. Allen, CS, Sponsel WE, Trigo Y, Dirks MS, Flynn WJ. Comparison of the frequency doubling technology screening algorithm and the humphrey 24-2 SITA-FAST in a large eye screening. Clinical and Experimental Ophthalmology 2002;30:8-14.
  33. Wall M, Neahring RK, Woodward KR. Sensitivity and specificity of frequency doubling perimetry in neuro-ophthalmic disorders: a comparison with conventional automated perimetry. Invest Ophthalmol Vis Sci 2002;43:1277-1283.
  34. Jackson GR, Scott IU, Quillen DA et al. Inner retinal visual dysfunction is a sensitive marker of non-proliferative diabetic retinopathy. Br J Ophthalmol 2012;96:699-703.
  35. Parikh R, Naik M, Mathai A et al. Fole of frequency doubling technology perimetry in screening of diabetic retinopathy. Indian J Ophthal 2006;54:17-22.
  36. Parravano M, Oddone F, Mineo D et al. The role of Humphrey Matrix testing in the early diagnosis of retinopathy in type 1 diabetes. Br J Ophthalmol 2008;92:1656-1660.