Purchase this article with an account.
M. Aoyama, K. Sugiyama, M. Sugitani, T. Yamamoto, G. Tomita, Y. Kitazawa; The Probability of Detecting Disk Hemorrhage Over Time in Primary Open-Angle Glaucoma and Normal-Tension Glaucoma . Invest. Ophthalmol. Vis. Sci. 2003;44(13):4375.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: To investigate the probability of detecting a disk hemorrhage and its relationship to intraocular pressure (IOP) in primary open-angle glaucoma (POAG) and normal-tension glaucoma (NTG). Methods: We retrospectively reviewed the charts of 246 patients with NTG, and 240 patients with POAG. Only one eye of each patient was used for the analyses. We applied a Kaplan-Meier life-table analysis of survival data regarding non-detection of disk hemorrhage. Several clinical factors were investigated to find a possible association with disk hemorrhage using a Cox proportional hazards model. Results: The life-table analysis revealed that the cumulative probability of detecting a 1st disk hemorrhage increased gradually over time and reached a plateau by 8 years in NTG and 6 years in POAG. The final probability was 38.4 %, which was significantly higher than that in POAG (16.9%). The probability of detecting a 2nd (i.e., recurrent) hemorrhage in patients with a previous hemorrhage was higher than that of the 1st hemorrhage. Likewise, the probability of detecting a 3rd or 4th hemorrhage was higher than that of the 1st hemorrhage. The final probabilities of recurrent hemorrhages were 51.6%(2nd), 68.6%(3rd), and 79.2%(4th) in NTG patients, and 79.6%(2nd), 73.3%(3rd), and 64.0%(4th) in the POAG patients. The Cox model analysis identified low IOP at the time of the 1st hemorrhage as a significant factor related to the 1st disk hemorrhage as a significant factor related to a 1st disk hemorrhage. Conclusions: The probability of detecting disk hemorrhage increases with time and is considered to reach a maximum limit. Recurrent hemorrhages occurred more frequently than initial hemorrhages, and reached the higher limits of their probabilities for detecting a hemorrhage. This supports the concept of two populations in open-angle glaucoma patients: one with disk hemorrhage and the other without.
This PDF is available to Subscribers Only