Natural History of Geographic Atrophy in Untreated Eyes with Nonexudative Age-Related Macular Degeneration: A Systematic Review and Meta-analysis

Liangbo Shen; Feimei Liu; Holly Grossetta Nardini; Lucian V Del Priore

doi:10.1016/j.oret.2018.01.019

Natural History of Geographic Atrophy in Untreated Eyes with Nonexudative Age-Related Macular Degeneration: A Systematic Review and Meta-analysis

Ophthalmol Retina. 2018 Sep;2(9):914-921. doi: 10.1016/j.oret.2018.01.019. Epub 2018 Mar 2.

Authors

Liangbo Shen¹, Feimei Liu², Holly Grossetta Nardini¹, Lucian V Del Priore³

Affiliations

¹ Department of Ophthalmology and Visual Science, Yale University School of Medicine, New Haven, Connecticut.
² Department of Biomedical Engineering, Yale University, New Haven, Connecticut.
³ Department of Ophthalmology and Visual Science, Yale University School of Medicine, New Haven, Connecticut. Electronic address: lucian.delpriore@yale.edu.

PMID: 31047226
DOI: 10.1016/j.oret.2018.01.019

Abstract

Topic: Systematic review and meta-analysis of the natural history of geographic atrophy (GA).

Clinical relevance: Several different models have been used to describe the natural history of GA in untreated eyes, and the reported progression rates vary widely across clinical trials.

Methods: We searched in MEDLINE, EMBASE, Cochrane Library, Clinicaltrials.gov, and PubMed for studies that measured GA size in untreated eyes over a follow-up period ranging from the start dates of the databases through June 6, 2017. Data were analyzed using 3 models: (1) the area linear model, in which the lesion area enlarges linearly with time; (2) the radius linear model (RLM), in which the lesion radius expands linearly with time; and (3) the area exponential model, in which the lesion area changes exponentially with time. A horizontal translation factor was added to shift each data set to correct for the differences in participant entry time into the studies. The model that best fit data was determined by performing residual analyses, determining the dependence of growth rate on time, and the predicted age of GA onset. The risk of bias was assessed using the Newcastle-Ottawa scale.

Results: We included 25 studies with data from 2942 eyes. The RLM yielded the best goodness of fit and predictive performance of GA progression. Cumulative data for untreated eyes fit a straight line in the RLM (r² = 0.986) with a randomly dispersed residual plot. The GA radius enlarged at a constant rate of 0.163 mm/year (95% confidence interval, 0.158-0.167 mm/year), which was independent of time (r = -0.108). The RLM predicted the mean age of onset of GA as 67.4±5.2 years. Our analysis also suggested that the GA progression rate may be associated with the age of onset.

Conclusions: In this meta-analysis, the progression pattern of GA was uniform across a wide range of studies, and best fit the RLM. Our analysis may shed light on the natural history of GA and may influence the design of future clinical trials.

Publication types

Review