April 2011
Volume 52, Issue 14
ARVO Annual Meeting Abstract  |   April 2011
Comparison Of The Glaucoma Staging System To Mean Deviation In Staging Visual Fields
Author Affiliations & Notes
  • Michael V. Boland
    Wilmer Eye Institute,
    Division of Health Sciences Informatics,
    Johns Hopkins University, Baltimore, Maryland
  • Karun Arora
    Wilmer Eye Institute,
    Johns Hopkins University, Baltimore, Maryland
  • Footnotes
    Commercial Relationships  Michael V. Boland, None; Karun Arora, None
  • Footnotes
    Support  This work was supported by a Clinician Scientist Award from the American Glaucoma Society, and by an institutional award from Research to Prevent Blindness
Investigative Ophthalmology & Visual Science April 2011, Vol.52, 5510. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Michael V. Boland, Karun Arora; Comparison Of The Glaucoma Staging System To Mean Deviation In Staging Visual Fields. Invest. Ophthalmol. Vis. Sci. 2011;52(14):5510.

      Download citation file:

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

  • Supplements

Purpose: : The Glaucoma Staging System (GSS, Mills et al. 2006) has been proposed as a means of assigning disease severity based on visual field results alone. Mean Deviation (MD) is the primary feature used to determine a stage but additional criteria related to global visual field statistics, numbers of depressed points and paracentral defects are also used. The GSS has thus far been used to classify relatively small numbers of fields and its behavior in large numbers of glaucoma subjects has not been assessed.

Methods: : We implemented and validated the GSS algorithm to automate analysis of visual fields. We then identified in our clinical database 9768 visual fields from patients with a diagnosis of open angle glaucoma. For each field, we calculated both the stage based on MD alone as well as the stage using the full GSS algorithm. We then determined which visual fields were moved from the MD-only stage due to the additional GSS stage-specific criteria.

Results: : The visual fields analyzed had an average MD of -7.3 +/- 8.5dB. The GSS algorithm assigned 19%, 39%, 24%, 9%, and 9% of fields to stages 0 through 4 respectively. Overall, 46% of fields were assigned to a stage other than the one that would have been chosen using MD alone. The largest movement was from stage 1 to 0 (19%) for those fields that did not meet any of the additional criteria for stage 1. None of the fields assigned to stages 2 or 3 based on MD alone moved up based on additional criteria. Two of the additional criteria, a central point <0 dB (3B), and 50% of central points <0 dB (4B) were never met by any of our fields. These two and four other criteria (1C, 2B, 2C, 4C) never caused a field to be moved from the stage that would have been assigned by MD alone. We did not find any cases in which the GSS stage could not be assigned because of ambiguity in the algorithm.

Conclusions: : It is possible to simplify the GSS. Of the 12 additional criteria included in the GSS algorithm, half were either never true or had no impact on the determination of GSS stage. Of these six potentially redundant criteria, 5 of them address field loss within the central 5 degrees. This suggests that isolated paracentral field loss, in the absence of other features already included in other GSS criteria, is extremely rare and may not need to be considered in staging disease.

Keywords: visual fields 

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.