Multiple regression analysis in myopic wavefront laser in situ keratomileusis nomogram development.

PURPOSE: To develop a systematic method for quantifying pretreatment adjustments to the treatment sphere in patients having myopic wavefront laser in situ keratomileusis (LASIK).

SETTING: Moorfields Eye Hospital, London, United Kingdom.

DESIGN: Comparative case series.

METHODS: Preoperative and 3-month postoperative data from consecutive cases of myopic wavefront LASIK (<10.0 diopters [D] sphere; <4.5 D cylinder) treated in 2008-2009 were tabulated. Multiple regression modeling was used to derive a weighted relationship between the achieved manifest refraction spherical equivalent (MRSE) and several possible influences. A new nomogram was derived from the resultant regression equation in which the percentage treatment boost was based on the partial correlation coefficient for the 4.0 mm pupil wavefront refraction spherical equivalent (SE), and pretreatment adjustment to the treatment sphere was based on the sum of the other weighted independent variables. The MRSE results in eyes treated using the new nomogram in 2009-2010 were compared with those from the original data set.

RESULTS: The variance in postoperative MRSE error was significantly reduced using the new nomogram (0.10 D in 2009-2010 versus 0.16 D in 2008-2009; P<.0001, Bartlett test). The R(2) measure of linear fit increased from 0.967 in 2008-2009 to 0.983 in 2009-2010.

CONCLUSIONS: Multiple regression modeling can be used to take account of several possible influences in enhancing the accuracy of wavefront LASIK. Nomograms developed through multiple regression modeling can be used to derive eye-specific pretreatment adjustments to the treatment sphere.

FINANCIAL DISCLOSURE: Neither author has a financial or proprietary interest in any material or method mentioned.