The advent of laser in situ keratomileusis (LASIK) has increased the popularity of excimer laser surgery for the correction of refractive errors, as it offers the patient fast, painless visual rehabilitation.1 The use of a lamellar corneal flap preserves the Bowman′s layer and decreases the possibility of cellular communication between the corneal epithelium and stroma. This produces less corneal haze and reduces interface healing, with greater precision and accuracy in refractive correction. However, the use of a microkeratome to create the corneal flap can sometimes result in complications, and previous studies have indicated a flap complication rate of 0.3 - 10%.2 One such complication is the creation of a loose sheet of epithelium or an epithelial defect in the flap. This is reported to occur in 1.6% of procedures.3 While this often does not interfere with completion of the surgical procedure, it may increase the risk of adverse events such as infection,4 diffuse lamellar keratitis (DLK),5 recurrent corneal erosion,6 epithelial ingrowth,7 and decreased accuracy in refractive correction.8 It has also been postulated that the risk of such flap-related problems may be greater with certain microkeratomes, possibly due to design characteristics.9
It is often possible to identify individuals with pre - existing pathology in the cornea who are prone to epithelial problems during LASIK. Such conditions include corneal basement membrane dystrophies, diabetes mellitus, and tear insufficiency.610 A recent study has also identified the following risk factors for epithelial erosions after LASIK: Fitzpatrick skin type I or II, Lancer ethnicity scale 1 or 2, age above 40 years, and light coloured hair and eyes.11 In another study, the authors reported that older patients and eyes with increased corneal pachymetry were at greater risk for epithelial problems during LASIK.4 The use of excessive topical anaesthetics, malfunctioning microkeratomes and faulty blades can also result in this complication.5 We identified a subset of 6 patients with clinically normal corneas, in whom uneventful LASIK resulted in bilateral epithelial defects. In this article, we describe the patient profile of this subgroup, clinical presentation of the epithelial defects, management, and anatomical and functional outcomes. Aetiopathogenesis and management strategies for the treated and fellow eyes are discussed.
Materials and Methods
The surgical procedures in these 6 patients were performed by four experienced surgeons. Due to the logistics of the procedure at our center, different surgeons treated the fellow eyes in three patients. We do not perform bilateral simultaneous LASIK, and the second eye was treated after complete healing of the epithelium in the first eye. The procedure was performed using a Hansatome microkeratome (Hansa Research and Development Inc., Miami, Florida, USA) to create a superior hinged flap, and an Allegretto Wavelight excimer laser (Laser Technologies AG, Erlangen, Germany) for excimer ablation in all patients. Either a new microkeratome blade or a second use blade was used to create the flap. In all cases, the blade was inspected prior to cutting the flap, to ensure that it was defect free. Two microkeratome units were used, # 2034 in 6 eyes and # 2787 in the other eyes.
A standard LASIK procedure was followed, similar to that used in other patients who did not develop flap complications. The periorbital skin around the treatment eye was prepared with 10% betadine, and one drop of 0.5% proparacaine (Paracaine, Sunways, Mumbai, India) was placed in the conjunctival cul-de-sac, along with one drop of 0.25% oxymetazoline hydrochloride eye drops (Oxylin, Allergan, MP, India), 5 minutes prior to creating the flap. At surgery, one more drop of 0.5% proparacaine was placed on the ocular surface after the patient was positioned under the laser. A Barraquer′s speculum was used to separate the eyelids and the surface was washed with balanced salt solution (Krishna Keshav Laboratories, Ahmedabad, India) to remove surface secretions. The corneal surface was marked with radial lines, in the horizontal and vertical meridians, using methylene blue. The 9.5 mm suction ring was used in 8 eyes (4 first and 4 second treated eyes) and the 8.5 mm ring was used in 4 eyes (2 first and 2 second treated eyes), according to surgeon preference. After using a Barraquer tonometer to confirm the desired increase in intraocular pressure, one or two drops of 0.3% hydroxypropylmethyl cellulose (Genteal, Novartis, Mumbai, India) were placed on the corneal surface. The microkeratome was then fitted on to the suction ring and advanced to create the corneal flap. A 160 m head was used in all eyes. The epithelial problem was noted after reversing the microkeratome, and consisted of either a defect or a loose sheet of epithelium attached to the corneal flap by a hinge.
The microkeratome pass was uneventful, and the flap was complete in all 12 eyes, although one patient was not cooperative during the procedure in both eyes (case # 1). The surgeon proceeded with flap lifting and excimer ablation, after which the flap was repositioned. An attempt was made to realign the loose sheet of epithelium in its bed, although it was noted that the epithelial sheet was always larger than the underlying bed. The corneal flap was carefully stroked into position to avoid damaging the loose sheet of epithelium. The presence of the loose epithelium interfered with the surgeon′s ability to properly reposit the corneal flap and ensure that the gutter of the corneal flap was minimal and uniform all around. This was more evident in the first few eyes in this series, and with experience the problem could be addressed more effectively. A bandage contact lens was used in 6 eyes and in the other eyes, the epithelium was reposited and the speculum removed after flap adherence. Although the decision to use a bandage contact lens depended upon the size of the epithelial defect, it was left to the discretion of the operating surgeon and no specific protocol was followed.
If a contact lens was not applied, one drop of 0.3% hydroxypropyl methylcellulose was placed on the corneal surface and the patient was asked to wait with eyes closed. The patient was examined under a slitlamp 20 to 30 minutes after the procedure to ensure that the flap was in satisfactory position. If the patient complained of pain and irritation, one drop of 0.5% proparacaine was instilled in the treated eye. The following drops were used in the treated eye at two-hourly intervals until bedtime - 0.3% ciprofloxacin (Ciplox, Cipla, Mumbai, India), 0.3% carboxymethlycellulose (Refresh Tears, Allergan, Mumbai, India). At review the next morning, ocular symptoms, unaided visual acuity, refraction, best spectacle corrected visual acuity (BSCVA) and slitlamp biomicroscopic findings were noted. Patients were reviewed at appropriate intervals till complete healing of the epithelium, after which the second eye was treated. Ciprofloxacin 0.3% antibiotic ointment (Ciplox, Cipla, Mumbai, India) was used in 6 eyes at bedtime for one month following the procedure, according to the discretion of the surgeon.
The mean age of the 6 patients (5M:1F) in this study was 28.5 7.9 years (range 23 - 44 years). There were no systemic or ocular predisposing factors for intraoperative epithelial problems in these patients. None of the patients were known diabetics and detailed ophthalmic evaluation did not reveal the features suggestive of epithelial basement membrane dystrophy. Following the occurrence of an epithelial defect in the first eye, the cornea was carefully examined in the second eye as well, using a dilated pupil and retroillumination at the slitlamp to detect subtle basement membrane pathology in the cornea. Fellow eyes were treated by different surgeons in 3 patients (cases # 1,5 and 6). Details of the patients are described in Table 1, intraoperative findings in Table 2, and outcomes in Table 3. The average preoperative refractive error was - 7.3 0.7 D (range -4.0 D, -12.25 D). Mean corneal pachymetry was 550 18 m and average keratometry was 44.7 1.5D. One of the patients (case # 1) was a contact lens user prior to surgery. One patient (case # 6) had nystagmus since birth.
A loose sheet of epithelium was seen in 6 eyes (3 first and 3 second treated eyes), and control epithelial defect (2 mm to 6 mm in the longest diameter and averaged 3 mm) was present in 6 eyes (3 first and 3 second treated eyes). The epithelial pathology (loose sheet or defect) was similar in both eyes of 4 patients (cases # 3 to 6) in t his group. All of the defects occurred in the inferior area of the cornea. Concurrence in the site of the epithelial problem was noted in both eyes of all patients. In eyes in which a sheet of epithelium was still attached to the cornea, the hinge was always located superiorly. The surface of the cornea in the region of the epithelial defect and under the loose epithelial flap was smooth. This suggested a separation of the epithelial sheet at the level of the Bowman′s layer although histopathological confirmation was lacking in this study.
The epithelial defect was present in all six eyes on day 1. There was roughening of the epithelium at the site of the epithelial loosening in eyes with an epithelial flap though a defect was not noted in these eyes. The healing time of the epithelial defect was 3 - 7 days (mean 5 days). In the 6 eyes where a bandage contact lens inserted at the conclusion of surgery, healing was uneventful, and the lens was removed one week after LASIK.
Diffuse lamellar keratitis (DLK) was noted in 3 eyes and was stage 212 in all eyes. This resolved by the second week, with the increased frequency of topical corticosteroid. Corneal infection was not noted in any of the eyes. At an average follow-up of 5.5 9.5 months (range 0.25 to 21 months), the unaided visual acuity was 6/9 or better in 10 eyes, and the mean residual error was - 0.68 1.2 D (range -0.25 D, -3.0 D). In case # 3, unaided visual acuity was 6/15 in the right eye and 6/ 24 in the left eye. BSCVA was maintained in 8 eyes, and 4 eyes lost one line of BSCVA (case # 2, right eye, case # 4, both eyes, case # 6, left eye). Episodes of recurrent erosion or epithelial ingrowth were not noted in any of the eyes in the follow-up period of the study.
The incidence of bilateral corneal epithelial defect following LASIK was 1.0% (6 of 605 patients) in our study. Tekwani et al4 have reported an incidence of 7.9%. Other studies have not specified this.31314
Mean corneal thickness in our patient population was 550 18 m which is not different from that reported by Tekwani et al4 in their study (560.5 34.7 μm). They indicated an increasing risk for intraoperative epithelial defects with increasing corneal thickness - every 50 m increase in preoperative corneal thickness was associated with an odds ratio of 2.30 (95% CI, 1.24 - 4.28). We do not think that corneal thickness alone is a causative factor, since we have performed uneventful LASIK in eyes with pachymetry in excess of 600 m. Average corneal pachymetry of the eyes of other patients in this cohort who underwent uneventful LASIK was 547 m. However, in predisposed eyes a thicker cornea may increase the shearing effect of the microkeratome and predispose the case to epithelial problems. Average preoperative keratometry in our patients was 44.7 1.5 D and this was not significantly different from that in those who underwent uneventful LAISK (44.2 D). Schirmer tests were not performed routinely in these patients but there was no clinical evidence of decreased tear function in these eyes in the pre- and post-LASIK period.
Stage 2 DLK occurred in 3 of 12 eyes in this series and resolved without complications after the use of topical corticosteroids. Although flap handling during the procedure was compromised by the epithelial defect, flap striae were not noted in any of the eyes in this series. Anatomically the epithelial pathology resolved in all eyes, without corneal infection, and at a mean follow-up of 5.5 months. As of the last review there were no episodes of recurrent corneal erosion or epithelial ingrowth.
The reason for the epithelial problem during LASIK in this patient group remains unclear. In absence of an overt cause for the same, it seems reasonable to postulate that these eyes have a sub-clinical weakness of adherence of the corneal epithelium which results in the problem when the cornea is subjected to shearing forces during flap creation using a microkeratome. It is interesting that the epithelial pathology occurred in the inferior half of the cornea in all patients. The configuration of the pathology is also consistent with the hinge of the epithelial sheet positioned superiorly.
Although infection is a cause for concern with loss of epithelial integrity, our experience suggests that the use of topical corticosteroids with concurrent topical antibiotic may be important to prevent the occurrence of DLK in these eyes. Others have reported an increased occurrence of DLK (24 times) in eyes with epithelial defects during LASIK.15
Possible alternative options that were not explored in our study could be the use of fine scissors or trim the stretched sheet of epithelium to better fit the bed, since the overlap of the edges may result in greater patient discomfort. This could possibly predispose to displacement of the flap due to lid movement. Smirennaia et al3 recommend removal of such flaps of epithelium, followed by the use of a bandage contact lens if the defect was greater than 3 mm. But we found that retaining the epithelial flap in its bed did not compromise the outcome in our patients. It is possible that retaining the epithelial barrier may protect against microbial infection, stromal hydration, and interface inflammation. The use of minimal topical anaesthetic eye drops in this situation appears to have a role in management, although such use should not be left to the discretion of the patient in view of the serious possibility of anaesthetic abuse keratopathy.16
The use of the HansatomeR microkeratome with the suction released during the reverse pass of the head has been described,4 as this decreases the shearing forces on the loose flap cut by the forward pass of the blade. Releasing the vacuum has the effect of increasing the space available between the head and the corneal flap and may help prevent the problem. However, this is technically a more difficult procedure and it is important that the patient does not move the eye during keratome reversal. Alternatively, those with access to a different microkeratome may choose to use the same in the fellow eye. There is as yet no convincing evidence that the rate of epithelial problems with the HansatomeR is greater than with other microkeratomes.
The configuration of the epithelial flap is these eyes had a consistent morphology with the hinge of the flap located superiorly. The edges of the tear were located inferiorly, and had ragged edges, with the convexity of the edge facing the limbus (Figure). This morphology suggests that the shearing forces are exerted on the epithelium during the forward pass of the microkeratome. The epithelium is first stretched and with the application of increasing force, tends to tear at the inferior portion. If this is accepted, it is not clear how reversing the microkeratome without using suction can help in decreasing the occurrence of this complication. The creation of an epithelial defect or a flap could be related to the degree of weakness in the adherence of the epithelium. Eyes with a very weak adherence may have an epithelial defect, as the shearing force rips off a small area of the loosely adherent epithelium, while in eyes with a slightly firmer attachment, there may be a flap as the forces used are unable to complete the detachment of a piece of epithelium. The angular trailing edge of the HansatomeR microkeratome applanation plate may also contribute to the occurrence of such epithelial problem during LASIK. The zero compression head incorporates a modification in the design of this plate and it remains to be seen if this decreases the occurrence of epithelial defects.4
It is possible that in case # 6, the nystagmus contributed to the creation of the epithelial defect in both eyes. We have performed a similar procedure in two other patients with nystagmus, without complications, and hence are unsure, about the contributory role of the nystagmus.
Lastly, in such eyes, one can consider the use of photorefractive keratectomy, or laser epithelial keratomileusis (LASEK) particularly when the refractive error is minimal. Since we believe that such patients are predisposed to epithelial problems due to an inherent weakness of the adhesion of the epithelium, it may be prudent to use a dry weck cell to push on the corneal epithelium of the fellow eye prior to treatment. If there is a weak adhesion in that eye as well, there will be a shift of the loose epithelium and this may aid in decision-making for the type of refractive surgery, and any surgical modifications.
In conclusion, we report a subset of patients who appear predisposed to bilateral corneal epithelial defects following LASIK. When such problems occur, appropriate management can result in good refractive correction.
1. Farah SG, Azar DT, Gurdal C, Wong J. Laser in situ keratomileusis: literature review of a developing technique J Cataract Refract Surg. 1998;24(989):1006
2. Jacobs MJ, Taravella MJ. Incidence of intraoperative flap complications in laser in situ keratomileusis J Cataract Refract Surg. 2002;28:23–28
3. Smirennaia E, Sheludchenko V, Kourenkova N, Kashnikova O. Management
of corneal epithelial defects following laser in situ keratomileusis J Refract Surg. 2001;17:S196–99
4. Tekwani NH, Huang D. Risk factors for intraoperative epithelial defect in laser in-situ keratomileusis Am J Ophthalmol. 2002;134:311–16
5. Sachdev N, McGhee CN, Craig JP, Week KH, McGhee JJ. Epithelial defect, diffuse lamellar keratitis, and epithelial ingrowth following post-LASIK epithelial toxicity J Cataract Refract Surg. 2002;28:1463–66
6. Ti SE, Tan DTH. Recurrent corneal erosion after laser in situ keratomileusis Cornea. 2001;20:156–58
7. Asano-Kato N, Toda I, Hori-Komai Y, Takano Y, Tsubota K. Epithelial ingrowth after laser in situ keratomileusis: Clinical features and possible mechanisms Am J Ophthalmol. 2002;134:801–807
8. Mulhern MG, Naor J, Rootman DS. The role of epithelial defects in intralamellar inflammation after laser in situ keratomileusis Can J Ophthalmol. 2002;37:409–15
9. Walker MB, Wilson SE. Lower intraoperative flap complication rate with the Hansatome microkeratome compared to the automated corneal shaper J Refract Surg. 2000;16:79–82
10. Fraunfelder FW, Rich LF. Laser-assisted in situ keratomileusis complications in diabetes mellitus Cornea. 2002;21:246–48
11. Bashour M. Risk factors for epithelial erosions in laser in situ keratomileusis J Cataract Refract Surg. 2002;28:1780–88
12. Linebarger EJ, Hardten DR, Lindstrom RL. Diffuse lamellar keratitis: Diagnosis and management
J Cataract Refract Surg. 2000;26:1072–77
13. Ahee JA, Kaufman SC, Samuel M, Bogorad D, Wee C. Decreased incidence of epithelial defects during laser in situ keratomileusis using intraoperative nonpreserved carboxypropyl methylcellulose sodium 0.5% solution J Cataract Refrac Surg. 2002;28:1651–54
14. Johnson JD, Harison-Dagher M, Pineda R, Yoo S, Azar DT. Diffuse lamellar keratitis: Incidence, associations, outcomes, and a new classification system J Cataract Refract Surg. 2001;27:1560–66
15. Shah MN, Misra M, Wilhelmus KR, Koch DD. Diffuse lamellar keratitis associated with epithelial defects after laser in situ keratomileusis J Cataract Refract Surg. 2000;26:1312–18
16. Varga JH, Rubinfeld RS, Wolf TC, Stutzman RD, Peele KA, Clifford WS, et al Topical anesthetic abuse ring keratitis: Report of four cases Cornea. 1997;16:424–29
Proprietary Interest: None