EyeWorld is the official news magazine of the American Society of Cataract & Refractive Surgery.
Issue link: https://digital.eyeworld.org/i/176967
42 EW REFRACTIVE SURGERY September 2013 Refractive editor's corner of the world How do you use your topographer? by William Tullo, OD R efractive surgeons rely on their diagnostic devices to screen and help determine appropriate candidates for refractive surgery. Corneal topography, for example, is very effective in diagnosing corneal dystrophies such as keratoconus. In screening for refractive surgery, a diagnosis of keratoconus would most likely exclude a prospective candidate from surgery. But what about those individuals who do not have keratoconus but may be at risk? Can topography be effectively used to screen for this? What do we do if a yellow or red flag shows up on these diagnostic instruments? Can a patient proceed with surgery? What does this actually mean? William Tullo, OD, has extensive experience evaluating corneal topography. In this "Refractive editor's corner of the world," he shares his thoughts on this issue. He also provides some additional guidelines for a few common diagnostic devices that surgeons can use to further clarify these dilemmas that are encountered daily in our practices. Thank you, Dr. Tullo, for taking the time to share your thoughts and experiences. Kerry Solomon, MD, refractive editor T he majority of refractive surgeons use corneal topography and corneal tomography to screen their patients before refractive surgery. Working with more than 100 different refractive surgeons has led me to observe the variable criteria utilized to screen patients for refractive surgery. We all agree that avoiding elective corneal surgery in patients who are at high risk for vision loss due to ectasia has led to the emphasis on obtaining "normal" corneal scans prior to surgery. I would like to discuss some of the common misconceptions observed when screening patients. The problem of screening patients for refractive surgery is undoubtably complex. It is important to note, topography and tomography can only show us the presence of a "normal shape" cornea or a shape consistent with patients with an accepted diagnosis of keratoconus. By itself, it will never be able to predict risk of becoming "keratoconic." I think many physicians are asking the wrong question. They ask me all the time "Does this patient have keratoconus?" The better question is, "Can this patient safely have refractive surgery without loss of vision due to corneal destabilization and irregularity?" Part of the problem is also the fact that keratoconus is a spectrum disease and there really is no currently acceptable definition of "suspect keratoconus." We want to identify subclinical not suspect cases of keratoconus. We know that many patients with "abnormal topography/tomography" undergo laser vision correction (LVC) with no loss of vision and some patients with normal topography/tomography undergo LVC and develop ectasia and loss of vision. So it is important to put topography/tomography interpretation in proper context when screening patients for surgery. I have found that many physicians are using topography and tomography as the sole criteria for candidacy. This forces me to consider the statistics of keratoconus. Both topography and tomography have less than 100% sensitivity and specificity. If we ignore medical-legal considerations for a moment, it is interesting to watch how our colleagues use topography and tomography to decide candidacy. It is not uncommon that when I review a case of post-LASIK ectasia, there is no preoperative evidence of keratoconus on topography or tomography. Physicians may be depending on this technology to predict keratoconus, not detect the onset of corneal changes consistent with keratoconus. It is important to remember that all patients who eventually are diagnosed with keratoconus start out with normal shaped corneas, and no topography or tomography device alone can predict changes before they occur. How do we then improve our ability to identify "subclinical" keratoconus? I am excited about the advances in corneal tomography over the past decade. First Bausch + Lomb (Rochester, N.Y.) gave us the Orbscan, allowing us to better understand posterior corneal changes consistent with keratoconus. More recently, the Oculus Pentacam (Arlington, Wash.) has incorporated a normative database and regression analysis in its BAD—Belin/Ambrosio Enhanced Ectasia Display. Unfortunately, I have seen a disturbing trend develop over the past several years with the misuse of the BAD on the Pentacam. Due to the lack of understanding on how multivariate regression analysis works, many physicians are misinterpreting the meaning of the display. Red/yellow individual D flags on BAD scans with normal Final D does not mean increased ectasia risk. The engineers at Oculus designed this display for engineers and statisticians—not practicing physicians. They also did not account for the unique medicolegal environment in the United States. I recommend in the U.S. to take advantage of the software's ability to remove the color coding of the sub-D values to reduce inadvertent medico-legal concerns. The power of a multivariate regression analysis as the final value is the best predictor. The individual continued on page 45 Table 1: Clinical considerations in addition to corneal topography Age less than 30 years Astigmatism asymmetry > 0.75 D Reduced or asymmetric BCVA Family history of keratoconus History of eye rubbing Elevated vertical coma > 0.35 microns @ 6.0 mm Table 2: Orbscan considerations Table 3: Pentacam considerations Anterior elevation ratio > 0.51 Final D > 2.60 Anterior elevation >16 microns Pachymetric asymmetry at thinnest point > 25 microns Posterior elevation > 50 microns Sub-D values > 2.6 SD Thinnest pachymetry < 500 microns ART-Max < 340 Irregularity index @ 3 mm > 1.5 D Irregularity index @ 5 mm > 2.0 D Peripheral pach @ 3 mm < 20 microns thicker than central pach Posterior best fit sphere > 55.0 D Anterior thinnest point elevation > 8 microns Posterior thinnest point elevation > 18 microns Abnormal PTI/CTSP profile Inferior decentered/ elongated elevation patterns Coincident front/back elevation and pachymetric thinnest point Coincident front/back elevation and pachymetric thinnest point Separate hyperopic database