Laser correction of low astigmatism reveals tendency for astigmatic overcorrection

Issue: May 25, 2013

Using wavefront-optimized LASIK to treat eyes with a preoperative cylinder of 0.5 D or less resulted in significant astigmatic overcorrection, according to a study.

Perspective from Noel A. Alpins, MD

“One of our quality benchmarks is to continually evaluate our treatment outcome,” coauthor Stephan J. Linke, MD, senior consultant and head of the Refractive Surgery Team at University Medical Center Hamburg-Eppendorf, Germany, said. “Because patients with very low amounts of ametropia are usually more demanding and sensitive, otherwise they would not have asked for elective refractive surgery, we felt that current literature has not sufficiently answered whether very low cylinder magnitudes can be treated in terms of high predictability, safety and efficacy.”

Results were published in Journal of Cataract & Refractive Surgery.

Preoperative cylinder

The retrospective study included 448 eyes of 448 patients; mean age was 37.8 years. Preoperatively, all patients had a refractive cylinder of 0.75 D or less and a subjective sphere between −2.75 D and −11.5 D.

Patients were divided into three subgroups of preoperative refractive cylinder magnitude: 0.25 D, 0.5 D and 0.75 D. Ablation was performed with the WaveLight Allegretto (Alcon) excimer laser platform.

At 4-month follow-up, the mean uncorrected distance visual acuity was 0.10 logMAR, and mean manifest refraction spherical equivalent was 0.68 D. There was no statistically significant difference in efficacy or safety of treatment between the three groups, according to the authors.

“The Alpins vector method was applied to investigate the amount of surgically induced astigmatism,” Linke said. “This vector analysis has been used in various studies to investigate cylindric changes after laser refractive surgery.”

Patients in the 0.25 D and 0.5 D subgroups experienced astigmatic overcorrection, as reflected by the correction index, magnitude of error, index of success and the flattening index. The postoperative mean manifest cylinder increased in eyes with a preoperative refractive cylinder of 0.5 D or less.

Unexpected results

“Because we treat a wide range of ametropic conditions with absolutely highly predictable, safe and efficient treatment results, we were initially somewhat surprised by the fact that, in the current study, we had obviously been faced with less convincing treatment results,” Linke said. “However, we are highly aware that the current results may have been negatively biased by drop-out of happy patients who have not attended to follow-up examinations.”

Linke said he and his colleagues can only speculate as to why full astigmatic correction of low myopic astigmatism favored astigmatic overcorrection. Three potential mitigating factors are flap creation, inappropriate patient alignment and irregularities in the laser beam.

The same surgical protocol was used in all eyes at the nine study sites; postoperatively, 88.2% (395 eyes) were within 1 D of the intended spherical equivalent correction.

“Based on the present data, caution is recommended whenever treating very low amounts of astigmatism,” Linke said. “Because we currently cannot indicate whether the results of our current study are due to treatment setting problems or, which is more likely, due to loss of follow-up of happy patients, we thus can only advise refractive surgeons to closely monitor their outcomes and consecutively adopt the nomogram,” Linke said. – by Bob Kronemyer

Reference:

Frings A, et al. J Cataract Refract Surg. 2013;doi:10.1016/j.jcrs.2012.09.024.

For more information:

Stephan J. Linke, MD, can be reached at Department of Ophthalmology, University Medical Center Hamburg-Eppendorf, Martinistraße 52, 20246 Hamburg, Germany; 49-40-7410-53314; email: slinke@uke.de.
Disclosures: Linke has no relevant financial interest in the study. Lead author Andreas Frings, MD, was supported by a grant from Wavelight GmbH (Erlangen, Germany).

Perspective

Noel A. Alpins, MD

This paper quantifies the phenomenon of low astigmatism treatments of less than 1 D target-induced astigmatism vector (TIA) being less accurate proportionally than larger treatments. The well-directed analyses employs 0.25 D of cylinder stratification steps (perhaps spectacle rather than preferred corneal plane) and a reference list that is comprehensive, making the paper a very useful guide for publication of astigmatism treatments and analysis.

Particularly worthy of note are the polar vector graphs, which demonstrate well their clinical relevance as a model for standardized graphical reporting of astigmatism outcomes. It appears that summated vector mean was not included, which misses an opportunity to view overall vectorial trends.

The diminishing correction indices of 2.65 (0.25 D TIA), 1.8 (0.50 D) and 1.19 (0.75 D) with increasing treatment, although confirming less accuracy with lower treatments, might suggest that various mathematical and treatment anomalies may also be significantly at play influencing reported results. These are very small TIA denominators frequently cause excess ratios for indices; measurement variability once astigmatism is below 0.5 D; and when spherical LASIK or PRK is being performed, coexisting unplanned surgically induced astigmatism vector (SIA) often occurs concurrently (around 0.4 D). Here, trending close to the same treatment meridian can significantly increase an index such as SIA (correction index) and flattening effect (flattening index), and can create variability of the distance vision (index of success) when compared to the TIA. All these idiosyncrasies of small treatments could have contributed to the observed effect, suggesting overstatement of actual inaccuracy.

Noel A. Alpins, MD

OSN Refractive Surgery Board Member

Disclosures: Alpins has financial interest in the Assort surgical management system.