I salute Dr. Amoils and coauthors for yet another important contribution to our knowledge about refractive surgery in their paper on iatrogenic keratectasia after LASIK. However, I think their conclusions fall short in one very important way. Looking at the preoperative topographies, I would not have offered LASIK to most of these patients under today's criteria. Certainly, at the time these patients were treated, I might have attempted LASIK. I do not mean to criticize Dr. Amoils and his coauthors for the care that these patients were given, although in these patients and some others in the literature, useful conclusions can be drawn from preoperative topography to avoid these predicaments. These conclusions are lacking in the paper. Moreover, it is important to ascribe these adverse outcomes to the preoperative patient condition and not to LASIK as a procedure. Perhaps most worrisome is the notion that only 37 μm of tissue ablation can lead to iatrogenic keratectasia. I believe that in this patient and all the others it is entirely an issue of preoperative patient screening and what follows is a case-by-case analysis of why.
Table 2 identifies only the right eye of patient 5 and both eyes of patient 6 as having FFK. While there are several ways to assign this as a diagnostic category, I think for my purposes I would call all these patients at least suspicious for FFK and offer them photorefractive keratectomy (PRK) rather than LASIK.
The topography of patient 1 was done in the absolute scale; that is, the color range extends from 9.00 to 96.00 diopters (D). For that reason, it is almost entirely uninformative. However, some inferior steepening is noticed even though the gradations of color value are spread across this very wide range. Such a topography map on a standardized scale would show alarming inferior steepening. The same is true for patient 2. The preoperative topography of the left eye is very uninformative since so little color is shown on the map and this was done on an absolute color scale. I find it suspicious that any asymmetry is detectable between the 2 eyes. There is only a small fleck of red on the left eye's topography, but it represents a superior to inferior discrepancy in places of perhaps as much as 3.00 to 4.50 D. For this reason, a standardized scale is necessary. I would consider such a patient as having had LASIK done almost without preoperative topography. The patient apparently had a preoperative central pachymetry of 542 μm and a 91 μm ablation, which should leave a residual corneal thickness of 0.451 μm. However, it measured 0.342 μm. Certainly, with her developing ectasia, thinning could have occurred, but one wonders whether, accompanied by this inferior steepening, there may have been an error in preoperative central pachymetry.
Most commonly, the cornea is thinnest in the center and becomes thicker toward the periphery. The extent to which this happens is very illuminating. Failure of the cornea to thicken by more than 20 μm in the midperiphery as compared to the central cornea is extremely suspicious, and I would encourage practitioners to measure both central and midperipheral corneal thickness for that reason. I have found, although it has not been published elsewhere, that patients with suspicious topography who have PRK instead of LASIK based on other criteria often have a show of failure to thicken in the midperiphery.
In patient 3, while there is no inferior steepening, there is no way for me to tell the possible asymmetry of the topographic astigmatism between the 2 eyes. One hallmark of keratoconus or a tendency toward it is an asymmetry in cylinder between the 2 eyes. The only suspicious finding I see here is the discrepancy between refractive cylinder and topographic cylinder. The postoperative central pachymetry by my computation would be expected to be 0.459 μm and was 0.335 μm. While thinning may have occurred, it is an extreme amount for that period of time. The patient is wearing 1.50 D of cylinder in his glasses but measures 2.45 D topographically.
Patient 4 had a very steep cornea to begin with, with portions of the cornea in the 49.00 D range. Fortunately, it was a normalized topography. Again, the topographies of the 2 eyes are not available for comparison, something which is very important in analyzing unexpected postoperative ectasia. Still, I consider any portion of the cornea in the 49.00 D range a very suspicious finding.
I agree with the authors that patients 5 and 6 represent FFK. However, the fact that patient 5 developed postoperative ectasia with only a 37 μm ablation is, again, important to attribute to the patient's own preoperative FFK and not to the depth of the ablation. To conclude otherwise instills unnecessary fear.
Patient 7 suffers from having an absolute scale used on topography. There are very few colors representing the corneal steepness, but the ones that we see show a steepness in some parts of the cornea in the 49.00 D range. There is also some asymmetry between the 2 eyes, and the right eye is far from a classic bow tie. Moreover, while the axis of the cylinder in the eyes usually adds up to 180 degrees, this is not true in this case. In my experience, this level of steepening and asymmetry is often accompanied by a failure of the midperipheral cornea to thicken by more than 20 μm compared to the central cornea. Patient 8 would violate what we know in the year 2000 as a rule of thumb of leaving 250 μm of posterior stroma behind. The corneal topography as printed is not all that informative. I certainly might like to see a normalized scale from beyond 39.5 to 44.5 D rather than seeing it extend from 38.00 to 50.00 D, but the red tones did not print well in the article. Nevertheless, it was a thin cornea and I am again left wondering whether it thickened to the normal amount in the midperiphery.
Overall, I think it is important to try to predict a given patient's risk of iatrogenic keratectasia from topography and pachymetric data preoperatively. I do not think it is reasonable to conclude from the data presented here that LASIK can cause iatrogenic keratectasia in seemingly random cases of otherwise well-selected patients.
Robert J.S Mack MD
aHoffman Estates, Illinois, USA