Abstract
Background: Keratoconus affects the anterior segment of the eye, which directly affects the refractive state of the eye. There are three components for the measurement of a corneal curvature or central corneal power (CCP) of the eye, namely, the power along the flat meridian, the power along the steep meridian and the axis of the flat meridian. Traditionally, CCP is analysed using univariate methods that processes each component separately; however, because of the trivariate nature of CCP, the use of multivariate methods and statistics may be beneficial.Aim: The aim of this study was to compare the short-term variation of the anterior and posterior CCP in eyes with and without keratoconus using multivariate methods of analysis.Setting: Data were extracted from a doctoral study by the first author. The group with keratoconus (KC) was obtained from patients attending a university-based contact lens clinic in Johannesburg, South Africa.Methods: A total of 28 eyes with KC and 28 eyes of 28 healthy control eyes without KC were included in this prospective quantitative study. Measurements were taken with the Oculus Pentacam (Wetzlar, Germany) and data related to the anterior and posterior CCP were analysed using multivariate methods and analysis.Results: For both KC and control groups, short-term variation of CCP of the anterior corneal surfaces was significantly greater than that for the posterior corneal surfaces. Whilst short-term variation was similar for both corneal surfaces in the KC group, variation of the posterior corneal surfaces was significantly different from that of the anterior corneal surfaces for the control group.Conclusion: Multivariate analysis of short-term variation of CCP of both surfaces of the cornea in eyes with or without KC contributed towards a more complete understanding of the disease.
Highlights
Between the late seventies and eighties of the 20th century, a combination of efforts from Long,[1] Keating[2,3,4] and Harris[5] led to the formalisation of a mathematically and scientifically meaningful representation of dioptric power via the concept of the symmetric, square 2 × 2 matrix
Measurements were taken with the Oculus Pentacam (Wetzlar, Germany) and data related to the anterior and posterior central corneal power (CCP) were analysed using multivariate methods and analysis
Whilst shortterm variation was similar for both corneal surfaces in the KC group, variation of the posterior corneal surfaces was significantly different from that of the anterior corneal surfaces for the control group
Summary
Between the late seventies and eighties of the 20th century, a combination of efforts from Long,[1] Keating[2,3,4] and Harris[5] led to the formalisation of a mathematically and scientifically meaningful representation of dioptric power via the concept of the symmetric, square 2 × 2 matrix. The matrix representation of dioptric power has enabled scientific methods of analysis to be conducted on critical optometric and ophthalmologic data that otherwise could not be properly understood either qualitatively or quantitatively These methods include adding, averaging and squaring sphero-cylindrical powers in their entirety and calculating variances, standard deviations and a host of other univariate and multivariate statistical functions[6] that were once thought to be impossible for such data.[29] These methods of analysing refractive and keratometric data have been used very frequently in South Africa and elsewhere for the study of a multitude of clinical and research-oriented issues of critical importance to avoid, for example, unnecessary vision impairment and its consequences. CCP is analysed using univariate methods that processes each component separately; because of the trivariate nature of CCP, the use of multivariate methods and statistics may be beneficial
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
