Laser in situ keratomileusis (LASIK) is an effective procedure to treat a wide range of myopia.1 The advantages of LASIK over photorefractive keratectomy (PRK) are rapid visual recovery, lower risk of corneal haze, greater regression of myopia, and less postoperative pain.1- 3 However, LASIK requires more skillful surgical technique and more instrumentation than PRK. Thus, inaccurate and inadequate procedures during LASIK have a higher potential of complications.
The complications of LASIK include severing of the corneal flap, epithelial ingrowth, flap wrinkling, corneal astigmatism, and corneal infection.2- 4 To our knowledge, only one case report by Pallikaris and Siganos5 has been published of corneal perforation following LASIK; it was a survey of their early 43 patients. Unfortunately, the treatment and the clinical course of the corneal perforation was not reported. We describe the treatment and clinical course of a patient with a LASIK-induced corneal perforation that affected the final visual acuity. We believe that this report on the treatment and recovery of the corneal perforation will be valuable information for refractive surgeons.
A 33-year-old man was referred to us with a complaint of decreased visual acuity in the left eye after bilateral simultaneous LASIK performed 3 days previously at a different facility. A corneal perforation was noticed in the left eye during LASIK laser ablation in this eye. The surgeon stated that, before surgery, the patient had myopia in both eyes (−9.5 diopters [D] −5.0 D × 65° OD and −9.0 D −5.5 D × 110° OS) and his best-corrected visual acuity was 10/20 OU. The corneal flap was intended to be 160 µm thick, otherwise, precise information on the intended ablation depth could not be obtained from the surgeon. However, the surgeon found that the remaining stromal bed was thin due to excessive thickness of the corneal flap.
Our initial examination showed best-corrected visual acuity of 20/60 OD and 20/500 OS. Slitlamp examination of the right eye revealed that the corneal flap was missing and that the stromal bed was thin; in the left eye, the corneal flap was very edematous and a space was visible between the corneal flap and the stromal bed (Figure 1, left). The anterior chamber was very shallow, and aqueous humor was observed to leak onto the ocular surface with blinking (Figure 1, right). A round, 0.25-mm diameter perforation site was observed in the center of the stromal bed by slitlamp examination.
Left eye. Three days after bilateral laser in situ keratomileusis. Left, The corneal flap is edematous and a space is evident between the flap and the stromal bed (arrow). The anterior chamber is very shallow. Right, Following a blink, aqueous humor is visible on the ocular surface (arrows).
We treated the corneal perforation by applying a therapeutic soft contact lens with topical antibiotics, oral carbonic anhydrase inhibitors, and eye patching. After 1 week of treatment (10th day postoperatively), the aqueous humor leakage had ceased, the anterior chamber depth was normal, and the corneal flap was closer to the stromal bed. However, the flap remained edematous. Topical medications were discontinued but the therapeutic contact lens and eye patching were continued for 3 more weeks. On the 32nd postoperative day, the edema and diffuse opacification of the corneal flap suddenly resolved with no space observed between the flap and the stromal bed although one had been present on the previous day. The patient's best spectacle-corrected visual acuity was still decreased at 30/50 OS due to residual areas of corneal opacification and irregular astigmatism. In addition, neovascularization was noted between the corneal flap and the stromal bed and epithelial ingrowth was present focally at the temporal edge of the flap (Figure 2).
Left eye. Thirty-two days after laser in situ keratomileusis and following treatment as described in the "Report of a Case" section. The cornea is generally clear and anterior chamber depth is normal, but areas of corneal opacification and irregular corneal astigmatism remained. Neovascularization is evident between the corneal flap and the stromal bed (arrows) and epithelial ingrowth was localized at the temporal edge of the flap (arrowheads).
No standard treatment has been established for corneal perforation with leakage of aqueous humor onto the ocular surface after LASIK. Because this is the first case of corneal perforation due to LASIK presenting to our institution, medical treatment was intuitive. Nevertheless, it was effective; the perforation closed, the edema cleared, and visual acuity improved to 30/50 OS.
The surgeon did not mention whether the patient had keratoconus; however, preoperatively the patient had high myopia and astigmatism, with a best-corrected visual acuity of 10/20 OS. In addition, the stromal bed in the unperforated right eye appeared very thin. These data suggest that the patient may have had keratoconus. In cases of keratoconus, the cornea is thin and excessive thickness of the flap may be created by the microkeratome because of distortion of the cornea, resulting in a thin stromal bed. We presume that this may have led to perforation with photoablation in this case.
The most interesting observation was that abrupt clearing of the corneal edema and the recovery of transparency occurred approximately 1 month after the injury by LASIK. This phenomenon suggests that wound closure and functional recovery of the endothelium leads to a prompt normalization of the corneal thickness. The site of perforation in the LASIK surgery was 0.25 mm in diameter and, therefore, may require approximately 1 month for recovery of the endothelial function.
Despite recovery of the cornea, high corneal irregular astigmatism and residual opacification remained, and this patient will eventually require lamellar or penetrating keratoplasty for better vision. One may question whether surgery should have been immediately performed. However, suturing of the flap to promote wound closure carries a high risk of inducing additional high irregular astigmatism. In our patient who had no suturing, the corneal edema disappeared completely in 1 month. We can then choose the most appropriate treatment after residual astigmatism and opacification have resolved. In our patient, it was more effective to use a therapeutic contact lens than to suture the corneal flap, because the suturing method has a high possibility of inducing new irregular astigmatism. This case suggests that eye surgeons should treat corneal perforations by conservative medical therapy, waiting at least 1 month before determining the best course of possible subsequent surgical therapy.
This investigation was supported in part by grant-in-aid 09671800 for Scientific Research from the Japanese Ministry of Education, Science, and Culture, Tokyo (Dr Watanabe) and by the Osaka Eye Bank Association Fund, Osaka, Japan (Drs Watanabe, Maeda, Inoue, and Shimomura).
Corresponding author: Hitoshi Watanabe, MD, Department of Ophthalmology, Osaka University Medical School, Room E7, 2-2 Yamadaoka, 565-0871 Suita, Japan.
Hori Y, Wantanabe H, Maeda N, Inoue Y, Shimomura Y, Tano Y. Medical Treatment of Operative Corneal Perforation Caused by Laser In Situ Keratomileusis. Arch Ophthalmol. 1999;117(10):1422-1423. doi: