Secondary Logo

Journal Logo

Article

Risk factors for epithelial defect formation during laser in situ keratomileusis

Randleman, Bradley J. MD; Lynn, Michael J. MS; Banning, Christopher S. MD; Stulting, Doyle R. MD, PhD

Author Information
Journal of Cataract & Refractive Surgery: October 2007 - Volume 33 - Issue 10 - p 1738-1743
doi: 10.1016/j.jcrs.2007.05.038
  • Free

Abstract

Laser in situ keratomileusis (LASIK) is currently the most commonly performed surgical procedure for the correction of refractive error. Intraoperative complications are rare and usually related to the creation of the lamellar corneal flap.1 The most common intraoperative complication is epithelial defect formation, which reportedly occurs in up to 14% of LASIK cases.2–8 Intraoperative epithelial defect formation can significantly complicate the postoperative course and lead to an increased incidence of recurrent erosions,9 flap edema,10 epithelial ingrowth,11 diffuse lamellar keratitis,12–15 ametropia2, and, rarely, flap melt.16

Reported risk factors for epithelial defect formation during LASIK include age,3,4,17 thicker corneas preoperatively,3 epithelial basement membrane dystrophy,10 preoperative drop regimen,3 skin type and ethnicity,4 and diabetes mellitus.18 An intraoperative epithelial defect in the first eye also increases the risk for an epithelial defect in the fellow eye.3,4,19 Malfunction of the microkeratome or blade defects can lead to intraoperative epithelial defects.20 In addition, microkeratome brand, design, and head characteristics have been shown to affect the formation of epithelial defects during LASIK.8,17,19,21

We reviewed a large consecutive series of primary LASIK cases performed with a single microkeratome brand to identify factors predictive of intraoperative epithelial defects.

PATIENTS AND METHODS

The electronic database of all patients who had primary LASIK at Emory Vision, Atlanta, Georgia, from January 1, 2001, to May 10, 2002, was reviewed. Cases were excluded from analysis if insufficient data were available in the database. As this was a retrospective review with data collection completed before June 2002, no institutional review board approval was necessary.

Data collected included surgery date; patient age, sex, and preoperative spherical equivalent (SE) refraction; preoperative drop regimen; type of laser used for treatment; and the presence or absence of an intraoperative epithelial defect.

All patients received identical surgical preparation with the exception of preoperative mydriatic drop use in patients having treatment with the LADARVision excimer laser (Alcon Laboratories). Phenylephrine hydrochloride 2.5% and cyclopentolate hydrochloride 1% drops were administered approximately 30 minutes before surgery to all patients being treated with the LADARVision laser, while patients having treatment with the EC-5000 laser (Nidek Technologies) received no mydriatic drops. All patients received topical proparacaine hydrochloride 0.5%, naphazoline hydrochloride 0.025% (Naphcon A), and ketorolac tromethamine 0.5% (Acular) preoperatively. Patients were then prepared for surgery with povidone–iodine 10% solution (Iodophor PVP swabstick) and draped with an adhesive drape.

Surgery proceeded in an identical fashion in all patients. A wire eyelid speculum was inserted, and a radial keratotomy marker coated with methylene blue was used to make fiduciary lines on the cornea. The suction ring of a Hansatome microkeratome (Bausch & Lomb) was applied to the surface of the conjunctiva and activated. A Barraquer applanation tonometer verified that the intraocular pressure was greater than 65 mm Hg. The corneal surface was moistened with proparacaine. The Hansatome microkeratome with a 180 μm plate was used. Simultaneous bilateral surgery was performed in most cases. The same blade was used for the second eye unless there was a thin flap or flap complication, other than epithelial defect, in the first eye.

After laser ablation, the flap was gently replaced over the stromal bed and lubricant drops were administered. If an intraoperative epithelial defect had been created, a therapeutic bandage contact lens was placed intraoperatively. All patients were examined with slitlamp biomicroscopy before departing from the surgical center.

The percentage of cases with an epithelial defect was compared between levels of categorical factors using a chi-square test. Pair-wise comparisons of the percentages between levels were done using the Tukey multiple comparison procedure. Means of continuous factors were compared between the 2 laser groups using an independent-groups t test. The relationship between the probability of an epithelial defect and age, preoperative refraction, and sex was assessed using logistic regression.

RESULTS

The experimental population included 6984 eyes. Sex was identified in the database in 6931 (99.2%) eyes, which included 3382 of men (48.8%) and 3549 of women (51.2%). The mean age was 41.8 years (range 17 to 70 years). The mean preoperative SE refractive error was −3.55 diopters (D) ± 3.25 (SD) (range +7.75 to −14.75 D). Of the eyes, 6067 (86.9%) were treated for myopia, including 5283 (75.6%) with the EC-5000 laser and 784 (11%) with the LADARVision laser. Nine hundred seventeen eyes (13.1%) were treated for hyperopia; all hyperopic cases were performed with the LADARVision laser because the EC-5000 laser was not approved by the U.S. Food and Drug Administration for hyperopic treatment during the time of this study.

Intraoperative epithelial defects occurred in 651 cases (9.3%). There were no significant differences in the overall incidence of epithelial defect based on sex; however, significant differences were noted based on age and preoperative hyperopic refraction (Table 1). In the overall population, there was a positive correlation between increasing age and an increased incidence of epithelial defects (Figure 1). The most significant increase in incidence occurred in patients 40 years and older, and this trend continued through the older age ranges. There was also a positive correlation between increasing hyperopia and an increased incidence of epithelial defect in the overall population (Figure 2). There was little variability in the rate of epithelial defect in the myopic cases; however, there was a significant increase in epithelial defect formation in hyperopic cases, and this trend continued with increasing hyperopic refractions.

Table 1
Table 1:
Epidemiology of intraoperative epithelial defects.
Figure 1
Figure 1:
Epithelial defect rate by age.
Figure 2
Figure 2:
Epithelial defect rate by preoperative refraction.

Because most cases were performed with the EC-5000 laser and all hyperopic cases were performed with the LADARVision laser, it became necessary to stratify the overall population by laser model to make further comparisons and analyses.

Patient Characteristics and Epithelial Defect Rates by Laser

Patients treated with the EC-5000 laser were younger and more myopic than patients treated with the LADARVision laser (Table 2).

Table 2
Table 2:
Patient characteristics by laser model.

EC-5000 Laser

Of eyes treated with the EC-5000 laser, 400 (7.6%) sustained an epithelial defect. Neither sex nor preoperative refractive error was a significant factor for epithelial defect in this population (Table 3). Preoperative refraction was also not statistically significant (P = .15) when included as the lone continuous factor in a logistic regression model. Increasing age was associated with a higher risk for epithelial defect. The percentages of eyes with an epithelial defect were statistically significantly different between decade categories (P<.0001). Comparisons between the categories showed that the percentages were not different between the 2 decade categories comprising patients younger than 40 years or between the 3 decade categories comprising patients 40 years and older; however, percentages in the 2 younger categories were different from those in all 3 of the older decade categories. With age categorized into 2 groups (younger than 40 years versus 40 years and older), eyes of patients older than 40 years were found to be at statistically significantly higher risk for epithelial defects (3.7% versus 11.3%, P<.0001; odds ratio [OR], 3.3; 95% confidence interval [CI], 2.6-4.2). In a multivariate logistic regression model including age (2 categories: <40 years and ≥40 years), preoperative refraction (continuous), and sex, age was the only statistically significant factor in the EC-5000 laser group (P<.0001).

Table 3
Table 3:
Univariate associations between patient characteristics and by laser model.

LADARVision Laser

Of eyes treated with the LADARVision laser, 247 (14.5%) had an epithelial defect. The percentage of eyes with an epithelial defect did not differ statistically by sex (Table 3). Preoperative hyperopia was associated with an increased risk for epithelial defect. The percentage of eyes with an epithelial defect differed significantly between the categories of preoperative refraction grouped by 2.0 D increments (P<.0001). A comparison between the categories showed that the 2 highest hyperopic categories (2.0 to 3.9 D: 19.7%; ≥4.0 D: 24.1%) differed statistically from the myopic category (−3.9 to 2.0 D: 8.3%) (P<.05). Hyperopic eyes had a statistically significantly higher risk for epithelial defect (<2.0 D: 152/1243 [12.2%]; ≥2.0 D: 95/458 [20.7%]) (P<.0001; OR, 1.9; 95% CI, 1.4-2.5).

With the LADARVision laser, the effect of age was similar to that with the EC-5000 laser, with eyes of older patients at higher risk for an epithelial defect (Table 3). The percentages of eyes with an epithelial defect were statistically significantly different between decade categories (P<.0001). The percentages were not different between the 2 decade categories comprising patients younger than 40 years or between the 3 decade categories comprising patients 40 years and older; however, percentages in the 2 younger categories were different from those in all 3 of the older decade categories. When age was categorized into 2 groups split at age 40 years, eyes of older patients had a statistically significantly higher risk for epithelial defect (<40 years: 97/2611 [3.7%]; ≥40 years: 303/2672 [11.3%]) (P<.0001; OR, 3.3; 95% CI, 2.6-4.2).

An evaluation of age and preoperative refraction together found that both factors were significantly related to the occurrence of an epithelial defect, especially when comparing age groups split at 40 years and preoperative refraction groups split between myopia and hyperopia (Figure 3). Patients 40 years or older, whether hyperopic or myopic, sustained epithelial defects at a higher rate, while hyperopic eyes had a higher rate of epithelial defect in all age categories. In a multivariate logistic regression model, older age and greater hyperopia conferred a greater risk for epithelial defect, with age remaining more significant than refractive error (Table 4). Sex was not statistically significant when added to the model (P = .43).

Figure 3
Figure 3:
Percentage of cases with an epithelial defect according to age and preoperative refraction for patients operated on with the LADARVision laser.
Table 4
Table 4:
Logistical regression analysis and OR results in patients treated with the LADARVision excimer laser.

Comparison of EC-5000 and LADARVision Lasers

Given the extreme difference in preoperative refraction between patients operated on with the EC-5000 laser and those operated on with the LADARVision laser (Table 2), the occurrence of epithelial defect formation with the 2 lasers was directly compared by analyzing the incidence of epithelial defects in eyes with low myopia (0 to −1.9 D) and in eyes with moderate myopia (−2.0 to −9.0 D) for each laser. There was a difference between lasers in the low myopia group (EC-5000, 7.6%; LADARVision, 12.9%) (P = .008). However, in the moderate myopia subset, the percentages with an epithelial defect were similar (EC-5000, 7.6%; LADARVision, 9.2%) (P = .29) and there were no significant differences by laser in a logistic regression analysis that included laser (P = .75), age (2 categories, <40 years and ≥40 years; P<.0001), and preoperative refraction (continuous; P = .15).

DISCUSSION

In this study, age and preoperative hyperopic refraction were significant risk factors for intraoperative epithelial defects during LASIK. To our knowledge, a relationship between preoperative hyperopia and epithelial defect formation has not been reported.

Our overall epithelial defect rate of 9.3% in this large series agrees with rates in previous publications.2–8 This rate did not vary significantly by sex, as has been reported.3,4,17

Factors inherent to epithelial structure and integrity can predispose patients to epithelial breakdown during LASIK.22–24 Compromise of the adherence between basal epithelium and its basement membrane increases the risk for epithelial disruption in the presence of compressive and shearing forces. Patients with epithelial basement membrane dystrophy have abnormal hemidesmosome and tight junction complexes and suffer epithelial sloughing with increased frequency during LASIK.10 Advanced age also places the corneal epithelium at risk for epithelial defect during LASIK through loss of the integrity of anchoring fibrils from thickening of the epithelial basement membrane with age23 as well as from altered aqueous tear production and composition.24 In this study, age was the most significant risk factor, with 13.4% of patients older than 40 years developing an epithelial defect compared with 4.5% of patients 40 years or younger. This finding supports findings in previous studies.3,4,17 Conditions that affect superficial epithelial cell cohesiveness, including dry eye, diabetic epitheliopathy,18 and exposure to preserved eyedrops and anesthetic agents,3 also make patients susceptible to disruption of the epithelium.

In this study, all eyes treated with the LADARVision excimer laser were exposed to preoperative mydriatic drops, while eyes treated with the EC-5000 laser were not. Although this seemed to minimally affect epithelial defect rates, it is potentially relevant today as some laser systems require preoperative mydriasis for treatment and others do not. Certain ergonomic factors could also explain any minimal differences in epithelial defect rates between lasers.

Preoperative hyperopic refractive error was an independent risk factor for intraoperative epithelial defect. The hyperopic eye differs structurally from the myopic eye in axial length, anterior chamber depth,25–27 and corneal steepness.26–29 Corneas in eyes with a hyperopic refraction may be thicker, which also may predispose these corneas to epithelial defects.3 Thicker corneas exert more frictional force on the traversing microkeratome head, increasing shearing forces on the epithelial surface from compression and friction, which increases the potential separate basal epithelial cells from the underlying stroma. Corneal thickness was not recorded in this database selection, so we are unable to directly comment on the effect of corneal thickness on epithelial defect formation in this population. Further studies of other biomechanical differences between the myopic corneas and hyperopic corneas may reveal other factors that potentially predispose hyperopic eyes to epithelial defect occurrence during LASIK.

Forces applied to the epithelium during the microkeratome pass are considered the primary cause of epithelial defect formation during LASIK,3,19 and several measures to reduce the occurrence of epithelial defects during LASIK have been suggested; these include altering preoperative drop regimens, increasing intraoperative lubrication, and altering microkeratome techniques. Minimizing the use of preserved eyedrops and anesthetic agents preoperatively and lubricants immediately before the microkeratome pass reduces the likelihood of epithelial defects.3,5 Lenton and Albietz22 show that the coefficient of friction on the ocular surface is reduced by the addition of intraoperative lubricants during LASIK. The occurrence of epithelial defect may also be significantly reduced with the application of nonpreserved carboxymethylcellulose before the microkeratome pass.5

Reducing microkeratome head compression has also been associated with reduced epithelial defect rates during LASIK.19,21 Jabbur and O'Brien17 report decreased rates of intraoperative epithelial defect with the Amadeus microkeratome (Advanced Medical Optics) compared with the Hansatome microkeratome. The use of a microkeratome without rotational action or a compressive head may also be warranted in patients at risk for epithelial defect formation. All surgeries in our study were also performed before the acquisition of the Hansatome zero-compression head, which has reduced the occurrence of epithelial defects during LASIK.30 We now exclusively use the Amadeus microkeratome system at our institution. Some authors advocate cessation of suction and manual return of the microkeratome after the forward pass to avoid further compression and contact between the trailing blade and the epithelium.3,30 Femtosecond laser–assisted flap creation may eliminate the shearing and compressive forces inherent to traversing microkeratomes and reduce the incidence of epithelial defect during LASIK.

In summary, the risk for intraoperative epithelial defect during LASIK is highest in patients older than 40 years and in those with a preoperative hyperopic refraction.

REFERENCES

1. Gimbel HV, Anderson Penno EE, van Westenbruggee JA, et al. Incidence and management of intraoperative and early postoperative complications in 1000 consecutive laser in situ keratomileusis cases. Ophthalmology. 1998;105:1839-1847. discussion by TE Clinch, 1847–1848.
2. Smirennaia E, Sheludchenko V, Kourenkova N, Kashnikova O. Management of corneal epithelial defects following laser in situ keratomileusis. J Refract Surg. 2001;17:S196-S199.
3. Tekwani NH, Huang D. Risk factors for intraoperative epithelial defect in laser in-situ keratomileusis. Am J Ophthalmol. 2002;134:311-316.
4. Bashour M. Risk factors for epithelial erosions in laser in situ keratomileusis. J Cataract Refract Surg. 2002;28:1780-1788.
5. Ahee JA, Kaufman SC, Samuel MA, et al. Decreased incidence of epithelial defects during laser in situ keratomileusis using intraoperative nonpreserved carboxymethylcellulose sodium 0.5% solution. J Cataract Refract Surg. 2002;28:1651-1654.
6. Kenyon KR, Paz H, Greiner JV, Gipson IK. Corneal epithelial adhesion abnormalities associated with LASIK. Ophthalmology. 2004;111:11-17.
7. Carrillo C, Chayet AS, Dougherty PJ, et al. Incidence of complications during flap creation in LASIK using the NIDEK MK-2000 microkeratome in 26,600 cases. J Refract Surg. 2005;21:S655-S657.
8. Khachikian SS, Morason RT, Belin MW, Mishra G. Thin head and single use microkeratomes reduce epithelial defects during LASIK. J Refract Surg. 2006;22:482-485.
9. Ti S-E, Tan DTH. Recurrent corneal erosion after laser in situ keratomileusis. Cornea. 2001;20:156-158.
10. Dastgheib KA, Clinch TE, Manche EE, et al. Sloughing of corneal epithelium and wound healing complications associated with laser in situ keratomileusis in patients with epithelial basement membrane dystrophy. Am J Ophthalmol. 2000;130:297-303.
11. Wang MY, Maloney RK. Epithelial ingrowth after laser in situ keratomileusis. Am J Ophthalmol. 2000;129:746-751.
12. 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;26:1312-1318.
13. Haw WW, Manche EE. Late onset diffuse lamellar keratitis associated with an epithelial defect in six eyes. J Refract Surg. 2000;16:744-748.
14. Mulhern MG, Naor J, Rootman DS. The role of epithelial defects in intralamellar inflammation after laser in situ keratomileusis. Can J Ophthalmol. 2002;37:409-415.
15. Esquenazi S, Bui V. Long-term refractive results of myopic LASIK complicated with intraoperative epithelial defects. J Refract Surg. 2006;22:54-60.
16. Castillo A, Diaz-Valle D, Gutierrez AR, et al. Peripheral melt of flap after laser in situ keratomileusis. J Refract Surg. 1998;14:61-63.
17. Jabbur NS, O'Brien TP. Incidence of intraoperative corneal abrasions and correlation with age using the Hansatome and Amadeus microkeratomes during laser in situ keratomileusis. J Cataract Refract Surg. 2003;29:1174-1178.
18. Fraunfelder FW, Rich LF. Laser-assisted in situ keratomileusis complications in diabetes mellitus. Cornea. 2002;21:246-248.
19. Kohnen T, Terzi E, Mirshahi A, Bühren J. Intraindividual comparison of epithelial defects during laser in situ keratomileusis using standard and zero-compression Hansatome microkeratome heads. J Cataract Refract Surg. 2004;30:123-126.
20. Sachdev N, McGhee CN, Craig JP, et al. Epithelial defect, diffuse lamellar keratitis, and epithelial ingrowth following post-LASIK epithelial toxicity. J Cataract Refract Surg. 2002;28:1463-1466.
21. Polk EE, Wexler SA, Kymes S. Incidence of corneal epithelial defects with the standard and zero-compression Hansatome microkeratomes. J Refract Surg. 2005;21:359-364.
22. Lenton LM, Albietz JM. Coefficient of friction of ocular surface lubricants for laser in situ keratomileusis. J Refract Surg. 2001;17:327-333.
23. Alvarado J, Murphy C, Juster R. Age-related changes in the basement membrane of the human corneal epithelium. Invest Ophthalmol Vis Sci. 1983;24:1015-1028.
24. Faragher RGA, Mulholland B, Tuft SJ, et al. Aging and the cornea. Br J Ophthalmol. 1997;81:814-817.
25. Stenström S. Variations and correlations of the optical components of the eye, in: Sorsby A, editor. Modern Trends in Ophthalmology, vol. II, Hoeber, New York, NY, 1947, pp. 87-102.
26. Grosvenor T. High axial length/corneal radius ratio as a risk factor in the development of myopia. Am J Optom Physiol Opt. 1988;65:689-696.
27. Llorente L, Barbero S, Cano D, et al. Myopic versus hyperopic eyes: axial length, corneal shape and optical aberrations. J Vision. 4. 2004. 288-298. Available at: http://www.journalofvision.org/4/4/5. Accessed June 16, 2007.
28. Levi DM, Klein S. Differences in Vernier discrimination for gratings between strabismic and anisometropic amblyopes. Invest Ophthalmol Vis Sci. 1982;23:398-407.
29. Grosvenor T, Scott R. Three-year changes in refraction and its components in youth-onset and early adult-onset myopia. Optom Vis Sci. 1993;70:677-683.
30. Bashour M., 2004. Adhesion abnormalities associated with LASIK [letter], Ophthalmology, 111, 2145-2146.
© 2007 by Lippincott Williams & Wilkins, Inc.