Predictive factors of femtosecond laser flap thickness measured by online optical coherence pachymetry subtraction in sub-Bowman keratomileusis

PURPOSE: To evaluate possible factors responsible for the difference between predicted and measured parameters during 100 mu m flap creation with a femtosecond laser (IntraLase FS30) using online optical coherence pachymetry (OCP). SETTING: AugenVersorgungsZentrum, Weilheim, and the Technical University of Munich, Munich, Germany. METHODS: In this nonrandomized prospective interventional case study, 287 eyes of 146 consecutive patients were monitored by online OCP before and after flap creation with the femtosecond laser. The laser-specific settings were held constant during the study to attempt a 100 pm flap in all eyes. A multiple linear regression model with backward variable selection procedure was applied to evaluate possible multivariable explanatory powers of several covariates. In addition, very thin and very thick flaps (ie, lower and upper quartiles of flap thickness distribution) were analyzed separately in a logistic regression model. RESULTS: Central flap thickness measured with online OCP subtraction varied according to a Gaussian distribution from 57 to 138 pm, with a mean of 100.4 mu m +/- 13.6 (SD). Regression analysis between predicted and measured flap thickness showed no predictive power of 11 variables including the keratometry value of the cornea, preoperative corneal thickness, and patient age. CONCLUSION: The piano applanation interface of the IntraLase FS30 femtosecond laser produced ultrathin flaps for sub-Bowman keratomileusis that were independent of some preoperative and surgical factors known to affect outcomes with mechanical microkeratomes.