Purchase this article with an account.
A. S. Wang, G. Vizzeri, S. Deokule, A. Tafreshi, L. M. Alencar, R. N. Weinreb, F. A. Medeiros; Repeatability and Inter-Observer Variability of Goldmann Applanation Tonometry, Dynamic Contour Tonometry and Ocular Response Analyzer. Invest. Ophthalmol. Vis. Sci. 2009;50(13):2846.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To evaluate the repeatability and inter-observer variability of Goldmann Applanation Tonometry (GAT), Dynamic Contour Tonometry (DCT) and Ocular Response Analyzer (ORA) in eyes of healthy individuals.
The intraocular pressure (IOP) was measured independently three times by one masked operator and one time each by two masked operators during the same session using GAT, DCT and ORA in 52 eyes (26 OD, 26 OS) of 26 healthy subjects (mean age 36.8±11.8 years, 14 females). IOP measurements were obtained using a Goldmann tonometer masked for IOP readings. Coefficients of variation (CV) and intra-class correlation coefficients (ICC) were calculated and Bland-Altman plots were obtained to assess the reproducibility and agreement within three measurements (trials) taken by the same operator (i.e., intra-observer agreement) and between single measurements taken by different operators (i.e., inter-observer agreement).
The intra-observer CVs and ICCs (95% CI) were 9.7% and 0.79 (0.69 - 0.86), 3.7% and 0.89 (0.83 - 0.93), 9.8% and 0.57 (0.41 - 0.70), for GAT, DCT and ORA corneal compensated IOP (IOPcc), respectively. Corresponding values for the inter-observer CVs and ICCs were 8.7% and 0.81 (0.73 - 0.88), 5.9% and 0.75 (0.64 - 0.84), 11.7% and 0.50 (0.34 - 0.65). In general, the 95% limits of agreement calculated were larger for GAT and ORA than for DCT and larger between observers than within observer (e.g., the intra-observer 95% limits of agreement for trial1 - trial2 were -4.0 to 3.6, -2.1 to 3.5 and -5.6 to 7.3 while the inter-observer 95% limits of agreement for operator1 - operator2 were -3.8 to 4.6, -1.6 to 2.3 and -4.4 to 4.8 for GAT, DCT and ORA IOPcc, respectively).
In general, in this study DCT showed lower CVs and more consistent measurements both within and between observers, suggesting that DCT may provide IOP measurements that are more reproducible and operator independent compared to GAT and ORA.
This PDF is available to Subscribers Only