OHTS and EGPS: Glaucoma detection using Confocal Scanning Laser Ophthalmology

Summary

Principal Investigator: Linda Zangwill
Abstract: DESCRIPTION (provided by applicant): The overall goal of this study is to pool confocal scanning laser ophthalmoscopy (CSLO) data from two large randomized trials, the Ocular Hypertension Treatment Study (OHTS) and the European Glaucoma Prevention Study (EGPS), to improve our ability to predict the development of primary open angle glaucoma (POAG) in ocular hypertensive patients and to enhance our understanding of the effect of ocular hypotensive treatment on optic disc topography. Moreover, the proposed study will use the EGPS dataset to independently test the prediction model developed in the CSLO Ancillary Study to the OHTS and to increase the statistical power for detection of disc features that may be predictive of POAG. OHTS and EGPS share similarities in study design and measurement protocols including ascertainment of POAG, and cause-specific attribution of abnormality. Datasets in a common format have been constructed for each study and merged. Independent analyses by each Coordinating Center have confirmed the OHTS prediction model for the development of POAG. Both studies included CSLO imaging on a subset of their participants with similar image acquisition and processing protocols. Specifically, the CSLO Ancillary Study to the OHTS has analyzable good quality CSLO data on 439 participants and the EGPS has analyzable CSLO data on 470 participants. This proposal will merge the CSLO data with the existing merged EGPS and OHTS dataset and will include CSLO data from 909 ocular hypertensive participants with a median follow-up of more than 4 years and a total of 84 POAG endpoints. The merged sample, the largest longitudinal dataset of ocular hypertensive subjects with CSLO imaging will provide statistical power to test hypotheses that might not be possible with one dataset alone. This collaboration will not interfere with each study's ability to analyze its own dataset. The Specific Aims are: 1) Improving Prediction Models using Baseline HRT indices. To develop statistical models that use baseline Heidelberg Retina Tomograph (HRT) topographic optic disc parameters to predict development of a POAG endpoint detected by reproducible visual field abnormalities and/or optic disc deterioration. 2) Improving Prediction Models Using Baseline and Follow-up HRT indices. To determine the predictive accuracy of baseline HRT indices or change in these indices during follow-up for developing POAG. 3) Predicting the first sign of glaucoma. To evaluate whether baseline HRT alone or in combination with other baseline ocular and demographic factors are predictive of whether a POAG endpoint is first detected by visual field changes or optic disc changes. 4) Comparison of stereophotography and confocal scanning laser ophthalmoscopy for detecting glaucomatous change over time. To compare the detection of optic disc changes using HRT automated change detection algorithms to expert qualitative assessment of stereophotographs. PUBLIC HEALTH RELEVANCE: The overall goal of this study is to pool confocal scanning laser ophthalmoscopy (CSLO) data from two large randomized trials, the Ocular Hypertension Treatment Study (OHTS) and the European Glaucoma Prevention Study (EGPS), to improve our ability to predict the development of primary open angle glaucoma (POAG) in ocular hypertensive patients and to enhance our understanding of the effect of ocular hypotensive treatment on optic disc topography. Because CSLO measures are reproducible, objective, quantitative, non-invasive, and require minimal patient cooperation, their potential as an early diagnostic test for POAG has broad public health significance. The combined EGPS/OHTS sample has more than 909 randomized participants with longitudinal CSLO measurements making it the largest sample of prospectively obtained CSLO measurements in ocular hypertensive individuals.
Funding Period: 2008-09-30 - 2010-08-31
more information: NIH RePORT