December 2002
Volume 43, Issue 13
ARVO Annual Meeting Abstract  |   December 2002
Polarimetric Analysis of the Peripapillary Hyperpigmentation in Glaucoma Patients
Author Affiliations & Notes
  • MB Mellem Kairala
    Schepens Eye Research Institute and Harvard Medical School Boston MA
  • AE Elsner
    Schepens Eye Research Institute and Harvard Medical School Boston MA
  • SA Burns
    Schepens Eye Research Institute and Harvard Medical School Boston MA
  • RB Simmons
    Ophthalmic Consultants of Boston and Harvard Medical School Boston MA
  • Footnotes
    Commercial Relationships   M.B. Mellem Kairala, None; A.E. Elsner, None; S.A. Burns, None; R.B. Simmons, None. Grant Identification: Support: Saltonstall Foundation EYO7624
Investigative Ophthalmology & Visual Science December 2002, Vol.43, 245. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      MB Mellem Kairala, AE Elsner, SA Burns, RB Simmons; Polarimetric Analysis of the Peripapillary Hyperpigmentation in Glaucoma Patients . Invest. Ophthalmol. Vis. Sci. 2002;43(13):245.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Abstract: : Purpose: To investigate the deeper layers of the peripapillary region in glaucomatous eyes, using a novel image processing technique applied to data from polarimetry. Method: Ten patients with clinically detectable peripapillary hyperpigmentation in one or both eyes (6 females, 4 males, mean age = 56 yr) were tested. Visual acuity was better than 20/60 (ametropia < 6 diopters) and IOP was <25 mm Hg. Exclusion criteria were diabetic retinopathy, AMD, and eye trauma or surgery - except for cataract extraction. Five patients had glaucomatous visual field loss, and 5 were Glaucoma Suspects. All patients underwent scanning laser polarimetry (GDx, LDT, San Diego, CA) and stereo color fundus photography. Each GDx series has 20 images, from 2 detectors, differing in illumination polarization. Custom software (MatLab, Mathworks, Inc.) computed images by separating light returning from the retina into 3 components: 2 that retain polarization information (random vs. parallel) and 1 that is depolarized and consists of multiply scattered light from the deeper layers. We computed the area and location by quadrant for hyperpigmentation in the color images, considering an average distance from the optic disc border of 570 µm. To determine whether the software could be useful for detecting small regions of peripapillary change, one patient with atrophy ≷ 30,000 µm2 was excluded from the statistics. We calculated the difference in grayscale for on vs. off pigment and Michaelson contrast in the 3 computed images and R, G, and B channels of the color images. Results: The mean area of hyperpigmentation as measured in color photos was 2.15±1.96 % of the peripapillary region sampled. Peripapillary pigment was distributed as follows: Temporal=162 ± 154, Superior 59 ± 149, Inferior 88 ± 101, and Nasal=0 µm2. The grayscale values were significantly different for on vs. off pigment only for the depolarized light images (p=0.014), and not for the other 2 GDx computed images, or the R, G, or B images from the color slides. Michaelson contrasts were significantly different only for depolarized images (-0.16 ± 0.20, p=0.005) and for the Blue channel (-0.07 ± 0.010, p=0.046). Conclusion: The increased scattering measured in association with peripapillary hyperpigmentation suggests that there is increased tissue disorder at these locations, and not merely absorption changes due to melanin. Polarimetric analysis of images provides additional information concerning the peripapillary region, and is potentially useful in the study of pathophysiology of glaucoma.

Keywords: 429 image processing • 430 imaging/image analysis: clinical • 432 imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) 

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.