ERROR BUDGET ANALYSIS OF CORNEAL LASER REFRACTIVE SURGERY

Error budget analysis of corneal laser refractive surgery procedures suggests several targets for achieving better clinical outcomes in the future, said Michael Mrochen PhD, at the 2012 annual meeting of the Association for Research in Vision and Ophthalmology. Through a systematic evaluation of factors that influence the refractive and quality of vision outcomes after corneal laser surgery, Dr Mrochen reviewed specific technical and physiological sources for error, and he arrived at the following conclusions regarding their resolution.

“Achieving improved outcomes after corneal refractive surgery will require reconsideration of subjective refraction as the gold standard for treatment planning and outcome evaluation, along with improved diagnostics to determine the relevance of epithelial smoothing and the biomechanical response, and the development of eye-specific assumptions and models for transferring the attempted optical correction onto the cornea,†said Dr Mrochen, IROC Science to Innovation, Zurich, Switzerland.

Explaining the foundation of his error budget analysis for corneal laser surgery, Dr Mrochen noted first that the field encompasses four categories – spherocylindrical corrections, presbyopia treatment, correction of higher order aberrations (HOAs), and therapeutic treatments. Taking all of these treatments into account, he identified four key areas of focus – refractive predictability, predictability of HOA correction, visual quality/depth of field, and biological factors. Then, he reviewed sources of error within each as well as obstacles to their removal.

Refractive predictability

Dr Mrochen observed that currently, refractive predictability of modern wavefrontoptimised or wavefront-guided laser surgery is excellent considering that achieved SE is within 0.5 D of target in 90 per cent to 95 per cent of eyes. Although ideally, 99 per cent to 100 per cent of eyes would fall within this predictability range, the ability to achieve that goal is limited by the measurement reproducibility of subjective refraction, said Dr Mrochen. He explained this limitation by citing a study by Raasch et al. [Ophthalmic Physiol Opt 2001;21:376-83], which analysed test-retest refraction data, and showed that even in normal eyes, there was repeatability within 0.5 D only 92 per cent of the time. In irregular eyes with keratoconus, repeatability of the two measurements was much worse.

“If the reproducibility of the refraction itself is similar to that of the laser refractive outcome, we need to question whether subjective refraction is precise enough to use for a gold standard as we attempt to further optimise the predictability of our refractive outcomes. In other words, the reproducibility of the subjective refraction is masking our outcome analysis,†he said. Data on outcomes from wavefront-guided ablations show that in eyes with preoperatively low levels of HOAs, the surgical treatment usually induces HOAs whereas some HOA correction can be achieved in eyes with higher pre-existing levels. Nevertheless, HOAs in the latter eyes remain under corrected, indicating room for improvement in HOA treatment predictability. As an added source of error, it has been shown that in eyes with higher HOA levels, the predictability of HOA correction is coupled with the refractive component of the correction and worsens as the amount of spherical correction increases. Dr Mrochen noted that correction of HOAs affects visual performance through its association with quality of vision. In addition, it has relevance to presbyopia treatments considering the current interest in ablations designed to induce spherical aberration as a means to increase depth of field. “While in theory, inducing spherical aberration is a viable method for improving near vision in presbyopes, its clinical feasibility is limited by the unpredictability of inducing specific HOAs,†said Dr Mrochen.

Biological factors

An effect of epithelial remodelling on deviation from the desired outcome must also be taken into account in aiming to improve results of laser refractive surgery, said Dr Mrochen, and he credited Dan Z Reinstein MD, for his extensive work in this area. “We still do not know if epithelial remodelling after stromal ablation has clinical relevance, but there is good indication it may be a factor influencing the predictability of correcting HOAs. Smoothing of the cornea by epithelial remodelling over the ablated surface may be masking our intended correction, and at the present time, we cannot overcome that issue, because of the limitations in diagnostics,†he said. Within the realm of biological factors that can affect outcome predictability of corneal laser refractive surgery, the biomechanical response of the cornea is another issue that needs to be considered. However, there are also uncertainties in this area as currently, it is neither possible to accurately predict the biomechanical response to surgery in the individual eye nor to reliably measure it using available diagnostic systems.

“The biomechanical response of the cornea might lead to aberrations, such as spherical aberration, or create an unstable and unpredictable situation, such as when performing a topography-guided therapeutic treatment in an eye where residual stromal bed thickness may be less than 300 microns. However, at the present time, we cannot reliably model these factors and predict their impact on outcome,†Dr Mrochen said. As a final point, Dr Mrochen reviewed the weaknesses in developing the laser algorithm and transferring the planned ablation profile to the cornea. “Achieving better outcomes with corneal laser surgery in the future will require the ability to model the ablation so that it is specific for an individual eye,†Dr Mrochen said.