Purchase this article with an account.
Anna Maria M. Stevens, Philippe A. Kestelyn, Jr., Philippe G. Kestelyn, Sr.; Using The OHTS - EGPS Risk Calculator With The Octopus Visual Field Indices. Invest. Ophthalmol. Vis. Sci. 2011;52(14):4171.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The OHT study identified risk factors for the conversion from ocular hypertension to POAG and a multivariate model for quantitative risk assessment was developed and validated (Weinreb and Medeiros). The EGP study compared the risk factors identified in the OHT study and refined the risk model (http://ohts.wustl.edu/risk/) Independent risk factors for the conversion to POAG in both studies were: age, IOP, CCT, vertical C/D, and PSD. Whereas the OHT study relied on Humphrey visual field testing, 25 % of patients in the EGP study performed Octopus visual fields. The authors of the EGP study took the square root of the loss of variance of the Octopus to obtain the equivalent of Humphrey PSD. They attributed the difference in odds ratio for PSD between the OHT study and the EGP study to the 25% of patients tested with Octopus perimetry. To convert LV (Octopus) to PSD (Humphrey) 2 algorithms are available in the literature (Monhart, 2006; Zeyen et al, Graefes Arch Clin Exp Ophthalmol, 1995). For an individual patient tested with Octopus perimetry the risk of conversion to glaucoma varies widely depending on which of the 3 method of conversion from LV to PSD is used.The purpose of our study was to obtain a more reliable method of conversion from LV (Octopus G1) to the equivalent Humphrey 30-2 PSD value.
Fifty consecutive ocular hypertensive patients were enrolled according to the inclusion criteria of the OHT study. All patients had experience with automated perimetry (> 3 VFtests). Most patients were followed using Octopus perimetry. Within 4 weeks the patients underwent VF testing on both Octopus and Humphrey machines. A conversion formula was derived from the data using trend analysis.
square root LV = 0.41 PSD + 1.05correlation coefficient r = 0.43
With the Monhart algorithm based on 26 eyes the risk of conversion is always lower than when taking the square root of LV, as done in the EGP study. With our algorithm the risk of conversion is lower for LV values up to 3 and higher for values > 3 . A potential bias in our study is the fact that most patients were familiar with the Octopus machine. The discordant results with the different conversion methods indicate that we need larger data sets to solve the contradictions.
This PDF is available to Subscribers Only