Purchase this article with an account.
Li Lian Foo, Monisha E. Nongpiur, Mingguang He, John C. Allen, Jr., Renyi Wu, Yingfeng Zheng, Seang M. Saw, Tien Y. Wong, Tin Aung; Novel Prediction Model for Angle Width in Chinese Singaporeans. Invest. Ophthalmol. Vis. Sci. 2011;52(14):3064.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To investigate the association of recently identified anterior-segment optical coherence tomography (ASOCT) parameters with angle width; and to describe a novel mathematical model to predict angle width.
Subjects were recruited from an on-going population based cross-sectional study of Chinese persons aged 40 years and older in Singapore. Participants underwent gonioscopy, A-scan biometry, and ASOCT (Carl Zeiss Meditec, Dublin, CA). Customized software (Zhongshan Angle Assessment Program, Guangzhou, China) was used to measure the ASOCT parameters. Linear regression modeling using the R-square, best subsets selection method was performed with trabecular-iris space area at 750µm (TISA750) and angle opening distance at 750µm (AOD750) as the dependent angle width variables. The optimal model consisted of the best ASOCT and A-scan subset of six predictors. Subsequently, a forward selection regression algorithm was used to test the improvement in the total R-square with the addition of each variable. Only right eye data of each subject was evaluated.
Complete data were available for 1067 subjects. The mean (standard deviation) age was 56.9 (8.5) years and 50.2% were male. Using TISA750 as the angle width parameter, the single best predictors among the ASOCT and A-scan independent variables were anterior chamber volume (ACV, R2 = 0.51), followed by anterior chamber area (ACA, R2=0.49) and lens vault (LV, R2=0.47). The best subset of 6 variables explained 81.4% of the variability in TISA750, and the fitted equation is given as TISA750=1.80065+0.01183*ACV-0.13922*ACW-0.29330*iris thickness at 750µm (IT750)-0.07677*ACA+0.09424* iris area (IAREA)-0.00011*LV. The best subset of 5 variables explained 78.2% of the variation in TISA750. The single best predictors of AOD750 as the angle width parameter was LV (R2 = 0.56), followed by ACA (R2=0.55) and ACV (R2=0.54). The best subset of 6 variables explained 85.3% of the variability in AOD750, and the fitted equation is given as AOD750=3.29363-0.01822*ACA 0.26610*ACW- 0.71949*IT750 0.01822*ACV-0.000215*LV+0.15463*IAREA.
ACV and LV were the most important single variables for predicting angle width. The fitted regression equations obtained for predicting angle width warrants future research as a possible screening tool for narrow angles.
This PDF is available to Subscribers Only