Predicting ectasia

Validity of new metric for ectasia risk in refractive surgery comes under the microscope in congress debate

Roibeard O’hEineachain

Posted: Wednesday, May 2, 2018

Most refractive surgeons agree that a certain amount of residual stromal thickness is necessary to maintain corneal biomechanical integrity and avoid ectasia following LASIK, but whether it is the same in every cornea or depends on the percent of tissue altered (PTA), was debated by two leading experts in the field at the XXXV Congress of the ESCRS in Lisbon.

Marcony R Santhiago MD, Federal University of São Paulo, Brazil, who coined the term PTA, maintained that a high PTA is the most robust risk factor for ectasia after LASIK in patients with normal preoperative corneal topography, because it provides a more individualised measure of biomechanical change.

He explained that PTA is the ratio of flap thickness and ablation depth divided by preoperative central corneal thickness. The formula was developed based on a consideration of biomechanical variables changed as a result of LASIK and photorefractive keratectomy.

According to a number of case-control studies he and his associated conducted, the risk of ectasia rapidly increases with a PTA value greater than 35% (with 100% sensitivity) and peaks its maximum combination of sensitivity and specificity when it is 40% or more.

Moreover, they have found that in eyes with normal preoperative topography, a high PTA value had higher predictive capabilities for ectasia risk than the absolute cut-off value of the RSB (300μm), as well as other predictors of ectasia such as the Ectasia Risk Score System (ERSS) and risk factors such as high myopia and age.

He noted that the minimum residual stromal bed thickness suggested by some authors is based on a flawed interpretation of data that assumes the cohesive tensile strength is uniform throughout the corneal stroma. But, in fact, the stroma weakens as one moves posteriorly, beyond its anterior third region.

“Based on these corneal structural differences it seems that that a ratio or equation would be more representative of the changes that occur after LASIK than specific cut-off values of the residual stromal bed thickness,” he said.

Dr Santhiago noted that a cohort study by Alain Saad MD and his associates appeared to show that a PTA value of 40% was not associated with an increased risk of post-LASIK ectasia in eyes with normal topography. The study included 126 eyes with a PTA value greater than 40%, none of whom developed ectasia during three years’ follow-up.

However, he pointed out that if you assume an incidence of 0.4% for post-LASIK, a cohort study for a risk factor for keratoconus would require a study population of 67,515 patients.

Alain Saad MD

In response, Dr Alain Saad MD, Rothschild Foundation Paris, France, and American University of Beirut, Lebanon, pointed out that PTA has not been validated by other centres, even though it has been four years since Dr Santiago first introduced the concept. A study from Dr Saad and co-authors included 593 eyes with 21% of them (126 eyes) having a PTA value greater than 40%, none of whom developed ectasia. If the PTA concept was applied to this cohort, those patients would have been excluded from LASIK surgery while they could benefit from it without complications.

However, he concurred with Dr Santhiago that case-control studies are necessary to determine the sensitivity of a predictive factor for a given complication. He and his associates therefore conducted a study involving 31 eyes that developed ectasia following LASIK with 80 eyes that didn’t develop the complication.

They found that a PTA below 40% had specificity of 90% but a sensitivity of only 48%. The risk factor would therefore only be present in about half of cases of post-LASIK ectasia. The only other reported external communication describing PTA was done by Lewis Groden during the American Society of Cataract and Refractive Surgery and found a sensitivity below 15% for PTA.

Dr Saad noted that there are several theories to explain the contradiction between the findings of his centre’s studies with those of Dr Santhiago. One is that PTA is only a two-dimensional measurement, which cannot take account of the three-dimensional alteration of tissue that takes place in LASIK.

“PTA assumes that all flaps will have the same diameter and optical zone. But when you do a flap with the same thickness but a larger diameter you might induce a higher amount of biomechanical instability. Or if you’re modifying optical zone. When you go from a 5mm to a 7mm optical zone the volume ablated increases by three-and-half times, even though thickness only increases by 1.5.”

Similarly, using an optical zone of 5mm to treat -11.0D of myopia will induce photo ablation depth of 90µm. However, using a 7mm optical zone to treat -5.0D of myopia will also induce an ablation depth of 90µm. Therefore, even though the PTA is the same, the volume of tissue removed will increase twofold.

Dr Saad noted that another weakness of PTA is that the preliminary study that coined the PTA concept relied only on Placido disc topography to select the “Normal” and “Abnormal” groups, while it is well admitted nowadays that tomography analysis is very important in order to screen corneas. Thus a bias in the group selection of the initial paper introducing the PTA might explain the difficulty to reproduce similar results on external groups. Dr Saad and his associates have developed the SCORE analyser, which is the first screening tool that automatically associates the Placido-based keratometric map, anterior and posterior corneal elevation and corneal thickness profile data into a single number score classification system for corneas.

Their research has shown that this score detects ectasia risk with a sensitivity of 61 to 93% and a specificity of 96 to 100% and when combined with ERSS sensitivity of 80.6% and specificity of 97%.

“We definitely need to improve topographic and tomographic screening to eliminate eyes with keratoconus. Marcony and his co-authors are on the right track with the PTA concept; however, at this stage, with an accurate preoperative topographical and tomographical screening, PTA concept should not be generalised and applied in the screening process,” Dr Saad concluded.