Femtosecond lasers have been used to create arcuate incisions for the correction of astigmatism after penetrating keratoplasty. Femtosecond laser systems designed for cataract surgery are also being used to perform limbal relaxing incisions (LRIs) at the time of laser cataract surgery. To the best of my knowledge, these applications of femtosecond laser technology involve the use of penetrating incisions.
At Paracelsus Medical University in Salzburg, Austria, my colleagues and I conducted a prospective, nonrandomized, single-center study of nonpenetrating intrastromal astigmatic keratotomy (ISAK) in 21 patients with corneal astigmatism of 0.75 to 7.00 D. The procedure was performed with the iFS Advanced Femtosecond Laser (Abbott Medical Optics, Inc.).
Patients included in the study had normal keratometry and central pachymetry greater than 480 μm. Patients with diabetes, autoimmune disease, keratoconus, irregular astigmatism, retinal disease, or glaucoma were excluded, as were patients who were pregnant or on systemic or ocular steroids. ISAK was performed in one eye only, with the fellow eye serving as a control.
Three types of patients were included in the study. The majority (n = 15) were patients with lenticular changes who were expected to undergo cataract surgery within the next 6 months. The goal for these patients was to reduce astigmatism prior to cataract surgery. A second group included two phakic patients with healthy crystalline lenses who needed astigmatic correction to achieve emmetropia. In a third group, four patients had residual astigmatism following cataract/IOL surgery. For the latter two groups, good UCVA was the goal.
Preoperative cylinder in the study ranged from 0.75 to 3.50 D. Pseudophakic patients in the study had a mean preoperative astigmatic refraction of 1.25 D, and phakic patients had a mean preoperative cylinder of 1.40 D. For this study, we employed a variety of patterns (ie, various optical zones and inclination angles) to establish the efficacy of the procedure and to begin developing appropriate nomograms based on age, degree of astigmatism, keratometry, and other factors.
The axis of astigmatism was marked preoperatively and verified intraoperatively with high-precision topography (Keratron Scout; Optikon). Immediately before performing ISAK, ultrasound pachymetry measurements were recorded in the center and all four quadrants of the cornea to confirm the intended incision depth. The goal was to make the arcuate incisions at least 100 μm away from Descemet membrane. Pachymetry often changes slightly due to variations in hydration or to swelling following the use of topical anesthetics. Remeasuring just before the treatment was applied (and reprogramming the laser as needed) compensated for such changes and added to the precision of the arc placement.
The software provided for use with the iFS laser in corneal transplants is flexible. To accommodate the wide variety of cuts now in use or under investigation for corneal transplant, the iFS laser can be programmed to make cuts of virtually any configuration by customizing the depth, length, and angle of inclination. In all cases, paired 90° arcs were created at a depth of 100 μm from Descemet membrane. Much like a traditional astigmatic keratotomy (AK), these incisions cut the lamellae, achieving a relaxing effect along the steep axis. Unlike a traditional AK, however, the cuts were completely intrastromal; there was no penetration of the anterior corneal surface, so there was no wound gape, epithelial ingrowth, or risk of infection, and the incisions stopped 100 μm short of Descemet membrane.
Postoperatively, the cylinder was reduced in all eyes (Figure 1). At 6 months, half the eyes had gained at least 1 line of UCVA. Of the four arc patterns attempted in this study, a 30° inclination proved to be the most accurate and was subsequently used in the majority of the cases. ISAK performed with this pattern achieved about a 1.00 D of reduction in cylinder (Table 1). With 6 to 12 months of follow-up, the effect appeared to be stable over time, with little change in refraction.
Patient-reported satisfaction was high. On a scale of 0 to 10 (with 0 indicating extremely satisfied), the average satisfaction score was 1. Eighty-five percent of patients rated their satisfaction as 2 or better, and there were no scores greater than 5. All but one patient said he or she would have the procedure again. When asked to compare the treated eye with the untreated fellow eye, patients reported better quality and clarity of vision in the treated eye and no difference in glare, halos, photophobia, dry eye, or foreignbody sensation between the two eyes.
No significant adverse events were seen, but there were two minor intraoperative complications. In one early case, there was a slight 0.5-mm decentration, but at 6 months this eye achieved UCVA of 20/20 with a refraction of -0.50 D of sphere. In a second case, there was an intraoperative loss of suction when the patient fell asleep and moved his head. The laser stopped automatically, suction was reapplied, and the procedure was completed as planned. Images generated with optical coherence tomography show an appropriate incision with no damage due to the suction loss (Figure 2). There was no statistically significant loss of endothelial cells in the treated eyes compared with the fellow eyes.
The ISAK procedure is potentially useful for the correction of astigmatism in a number of settings. Its use may reduce the need for toric IOL implantation, improve the precision of LRIs, and enhance postoperative refractive outcomes. From a practical standpoint, it is convenient and cost-effective to identify additional applications for a well-established laser platform that is already used for refractive surgery. In our clinic, the iFS laser is located next to the excimer laser and on the same floor as the cataract operating suites, so it can be used for both refractive and cataract patients.
Clinically, ISAK with a femtosecond laser is less invasive than traditional AK, in which the incisions penetrate the anterior corneal surface. Although the effect achieved currently with ISAK is less robust than that of traditional AK, we believe that outcomes will improve as nomograms are further developed. Moreover, because ISAK incisions are entirely intrastromal, there is less concern about their opening up if an excimer laser procedure is needed after ISAK. Over-correction, which sometimes occurs with AK incisions and may require suturing, is unlikely with ISAK. In addition to reducing these risks associated with penetrating AK incisions, we anticipate being able to achieve more predictable results with ISAK because of the precision of the laser. Once the nomograms are perfected, it will be possible to program the laser to perform incisions that are more accurate than one can create manually.
This small, initial study demonstrates that a predictable, stable effect can be achieved with femtosecond intrastromal arcuate incisions. We are now embarking on a more extensive randomized, multicenter study in which 60 eyes will be treated with ISAK patterns and intrastromal AK will be compared with penetrating AK. Additionally, more work remains to be done to adjust the nomogram for higher levels of astigmatic correction, which may require shorter or concentric arcs.
Günther Grabner, MD, is Professor of Ophthalmology and Chairman of the University Eye Clinic at Paracelsus Medical University in Salzburg, Austria. Professor Grabner did not provide financial disclosure information. He may be reached at tel: +43 662 4482 3700; e-mail: firstname.lastname@example.org.