Integrity of Intrastromal Arcuate Keratotomies Performed by Femtosecond Laser.

To the Editor: Femtosecond laser platforms used for cataract surgery are integrated with anterior-segment optical coherence tomography (AS-OCT), allowing for real-time imaging and measurement of the cornea, possibly leading to more precise arcuate keratotomies.1 The femtosecond laser permits creation of either anterior penetrating incisions or intrastromal incisions. The advantages of intrastromal incisions are less risk of infection and possible preservation of corneal nerves.2 Isolated case reports have described incidents of microperforations, full perforation, or vertical gas breakthrough after clear corneal incisions and arcuate keratotomy with femtosecond lasers.3-5 A video review was undertaken for patients undergoing femtosecond laser-assisted cataract surgery with placement of a monofocal non-toric intraocular lens using an OCT-guided femtosecond laser (Catalys; Abbott Medical Optics, Abbott Park, IL) by one surgeon at Surgisite, Boston, Massachusetts. Data were analyzed in such a manner that patients could not be identified directly or through identifiers linked to the patients. The study was hence exempt from human subjects regulations and did not require review by an institutional review board. Review of the videotaped procedures was performed by two observers. All incisions were set within 0.5 mm of the limbus with 20% of the cornea left uncut anteriorly and posteriorly. A microperforation was defined by the appearance of one or more air bubbles either at the epithelial layer or into the overlying liquid interface (Figure 1). Twenty-eight consecutive eyes of 25 patients who underwent planned intrastromal limbal relaxing incisions were reviewed; 64.9% (n = 19) of eyes underwent one limbal relaxing incision and 32.1% (n = 9) underwent two limbal relaxing incisions. Nine incisions showed no signs of microperforations, 17 eyes showed microperforation of one incision, and 2 eyes had microperforations of both planned incisions. No eyes experienced posterior perforation or full-thickness perforation. When comparing both groups (perforated vs non-perforated), perforation status did not seem to affect improvement in corrected distance visual acuity (P = .65) and mean spherical equivalence (P = .19). Such a high incidence of microperforation in planned intrastromal incisions could be explained by a deficiency in adequate imaging or by the inability of the femtosecond laser to deliver its energy in the intended location.5 It is possible that the incidence of perforation would be reduced by leaving a larger percentage of uncut cornea anteriorly. A deeper incision would lead to a loss of refractive effect and require longer arc length incisions with further corneal nerve ablation.2 It is not clear whether our results extend to different laser platforms or whether the incidence noted in our series is machine specific. The loss of integrity of an intrastromal incision may lead to undesired consequences, although no infections or increased pain or foreign body sensation were noted in our series. Intrastromal arcuate keratotomies may be less effective at correcting astigmatism if gas escapes during incision creation. The latter may lead to a reduction in tissue dissection and subsequent loss of refractive effect. Our results show that refractive cylinder as determined by manifest refraction did not change significantly from preoperative measurements. This applied equally to nonperforated and perforated incisions. It is premature