April 2010
Volume 51, Issue 13
ARVO Annual Meeting Abstract  |   April 2010
Corneal Haze Measured From Confocal Image Brightness
Author Affiliations & Notes
  • J. W. McLaren
    Ophthalmology, Mayo Clinic, Rochester, Minnesota
  • K. M. Kittleson
    Ophthalmology, Mayo Clinic, Rochester, Minnesota
  • W. M. Bourne
    Ophthalmology, Mayo Clinic, Rochester, Minnesota
  • S. V. Patel
    Ophthalmology, Mayo Clinic, Rochester, Minnesota
  • Footnotes
    Commercial Relationships  J.W. McLaren, None; K.M. Kittleson, None; W.M. Bourne, None; S.V. Patel, None.
  • Footnotes
    Support  NIH Grant EY02037, Research to Prevent Blindness, and Mayo Foundation
Investigative Ophthalmology & Visual Science April 2010, Vol.51, 5675. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      J. W. McLaren, K. M. Kittleson, W. M. Bourne, S. V. Patel; Corneal Haze Measured From Confocal Image Brightness. Invest. Ophthalmol. Vis. Sci. 2010;51(13):5675.

      Download citation file:

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

  • Supplements

Purpose: : Back-scattered and reflected light, seen clinically as corneal haze, determine the brightness of images in clinical confocal microscopy. In this study, we developed a method to standardize confocal image brightness to measure corneal haze. This method was tested by comparing stromal haze in corneas after LASIK to stromal haze in normal corneas.

Methods: : Thirty-five corneas from 18 patients 3 years after LASIK and 36 corneas from 18 untreated normal volunteers were examined with a ConfoScan 4 confocal microscope with a z-ring adapter (Nidek Technologies). Mean image brightness and depth from each frame was exported from the operating program. Brightness was adjusted for drift in the illumination and detection systems of the microscope by using the mean brightness of a standard scattering solution (Amco Clear, GFS Chemicals, Columbus, OH) scanned immediately after examination of the cornea. Image brightness from the cornea was expressed as scatter units (SU), the concentration of Amco Clear in nephelometric turbidity units (NTU) that produced the equivalent image intensity, determined from scans of serial dilutions of Amco Clear. Mean image brightness was determined in 20 consecutive layers of 5 percentile stromal depth through the full thickness of the stroma. Mean image brightness in each layer after LASIK was compared to that in equivalent layers of the control corneas by using generalized estimating equation models to account for possible correlation between fellow eyes of the same subject..

Results: : The confocal image brightness from scans of Amco Clear solutions increased linearly with concentrations between 200 and 1,000 NTU. In the mid stroma, image brightness was 742 ± 121 SU after LASIK and 787 ± 146 SU (mean ± SD) in control corneas (p=0.26). After LASIK, the surgical interface was visible at 18 ± 4% of the stromal depth. At 10% of stromal depth (equivalent to the mid flap after LASIK), image brightness was 770 ± 191 SU after LASIK and 1,117 ± 249 SU in control corneas (p<0.001). In all layers anterior to 25% of the stromal thickness after LASIK, including all of the flap, image brightness was significantly lower than it was in the corresponding layers of control corneas (p<0.001). In all layers deeper than 25% of stromal thickness, image brightness was not significantly different from that of control corneas (p>0.1).

Conclusions: : This method of standardizing confocal image brightness provides a means of comparing corneal haze in longitudinal studies and across laboratories. It demonstrates a subtle decrease in haze of the stromal flap measurable 3 years after LASIK.

Clinical Trial: : www.clinicaltrials.gov NCT00350246

Keywords: imaging/image analysis: clinical • cornea: stroma and keratocytes • refractive surgery: LASIK 

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.