Treatment of Epithelial Ingrowth After Laser In Situ KeratomileusisWith Mechanical Debridement and Flap Suturing

Objective  To evaluate the efficacy and safety of mechanical debridement and suturingof the laser in situ keratomileusis (LASIK) flap in the treatment of clinicallysignificant epithelial ingrowth after LASIK.

Methods  In a retrospective study, 20 eyes (n = 19 patients) in which clinicallysignificant epithelial ingrowth developed after LASIK were treated with liftingof the flap, scraping of the epithelial ingrowth, and flap suturing. Primaryoutcome measurements including recurrence of ingrowth, uncorrected visualacuity (VA), manifest refraction, best spectacle-corrected VA, and complicationswere evaluated at the last postoperative examination.

Results  At the last postoperative examination (mean ± SD, 10.5 ±14.3 months; range, 1.5-64 months), 100% of eyes had no recurrence of clinicallysignificant epithelial ingrowth. The uncorrected VA changed from 20/20 orbetter in 7 eyes (35%) and 20/40 or better in 15 eyes (75%) preoperativelyto 20/20 or better in 9 eyes (45%) and 20/40 or better in 16 eyes (80%) atthe last follow-up examination. There was no significant change in the meanlogarithm of the minimum angle of resolution (logMAR) uncorrected VA before(mean ± SD, 0.3 ± 0.5; range, −0.1 to 1.7) and after surgery(mean ± SD, 0.2 ± 0.4; range, −0.1 to 1.7) (P = .40). Mean ± SD spherical equivalent changed from –0.21± 0.82 diopters (D) (range, −1.25 to 1.00 D) preoperatively to−0.53 ± 0.89 D (range, −2.50 to 0.38 D) at last follow-up(P = .30). No eyes lost 2 or more lines of best spectacle-correctedVA, and there were no complications associated with the treatment.

Conclusions  Suturing the LASIK flap in addition to mechanical debridement of epithelialingrowth is a safe and effective treatment for clinically significant epithelialingrowth after LASIK.

Epithelial ingrowth after laser in situ keratomileusis (LASIK) has beenreported to occur in 0% to 20% of cases,^1-9 witha cumulative mean of 4.3% in a review of LASIK publications.¹⁰ Mostcases are self-limited, occurring at the edge of the flap and extending inwardless than 0.5 mm, and cause no adverse effect on the results of surgery.¹¹ However, ingrowth becomes clinically significantand requires removal when it extends into the entrance pupil and decreasesvision, approaches the edge of the pupil and reduces vision or induces nighttimeglare, causes keratolysis, induces astigmatism by raising an area of the flap,or causes epithelial irregularity with fluorescein staining at the edge ofthe flap and foreign-body sensation.¹¹

Clinically significant ingrowth requiring removal has been reportedin 0.92% to 3.2% of patients undergoing LASIK.^4,11 Reportedtechniques for removing epithelial ingrowth include lifting the flap and scrapingthe epithelial ingrowth,^3,11,12 withor without adjunctive treatments such as ethanol,¹³ mitomycin,phototherapeutic keratectomy,¹² or suturingof the flap.^11,14 The recurrencerate after scraping alone has been reported as 44%.¹¹ Adjunctivetreatments to lower this recurrence rate, however, may cause adverse effects.Treatments with ethanol and mitomycin have toxic effects,^11,15 andphototherapeutic keratectomy can cause irregular astigmatism and unpredictableshifts in refraction.^12,16 Suturingthe flap to create a tight apposition between the flap and the stromal bedhas been proposed to be effective at preventing the recurrence of ingrowthwithout the complications of other adjunctive treatments. This study evaluatesthe efficacy and safety of suturing the LASIK flap after removal of epithelialingrowth in the treatment of clinically significant epithelial ingrowth afterLASIK.

A retrospective review was conducted of 20 eyes in 19 patients treatedfor clinically significant epithelial ingrowth after LASIK with lifting ofthe flap, scraping of the ingrowth, and suturing of the flap to the stromalbed. All eyes were diagnosed with clinically significant epithelial ingrowthafter LASIK and were treated owing to evidence of progressive epithelial ingrowth,peripheral flap melting, topography changes, refractive changes, and/or chronicirritation. The preoperative LASIK examination, LASIK surgical report, postoperativecourse, epithelial ingrowth treatment report, and postepithelial ingrowthtreatment examinations were reviewed.

Treatment of epithelial ingrowth was performed by a single surgeon (E.E.M.).After providing informed consent, patients were given topical 0.5% proparacainehydrochloride (Ophthetic; Allergan Inc, Irvine, Calif), 0.1% diclofenac sodium(Voltaren; Ciba Vision Ophthalmics, Duluth, Ga), and 0.3% ciprofloxacin hydrochloride(Ciloxan; Alcon Laboratories Inc, Fort Worth, Tex) in the operated-on eyeimmediately before the procedure. Patients were taken to the laser and reclinedin a supine position. A sterile drape and a wire lid speculum were placedin the operated-on eye. The flap was lifted using a LASIK spatula. The epithelialingrowth was scraped from the posterior surface of the flap and from the keratectomybed using a blunt photorefractive keratectomy spatula (AE-2767 Maloney spatula;ASICO, Westmont, Ill). After repositioning of the flap, the interface wasirrigated with balanced saline solution. The flap was sutured into place with10-0 nylon, using a total of 5 to 11 interrupted sutures. The eye was thenirrigated with balanced salt solution, followed by 0.3% ciprofloxacin. Alsoadministered were 0.1% diclofenac and 1% prednisolone acetate (Pred Forte;Allergan Inc), and a bandage contact lens was placed. Patients were treatedpostoperatively with 0.3% ciprofloxacin hydrochloride 4 times a day and 1%prednisolone acetate 4 times a day. Once the epithelium was healed, the bandagecontact lens was removed, and 0.1% fluorometholone acetate (Flarex; AlconLaboratories Inc) was given 2 times a day for 2 weeks.

Outcome measures included recurrence of epithelial ingrowth on slitlampbiomicroscopy results, uncorrected visual acuity (UCVA), manifest refraction,and best spectacle-corrected visual acuity (BSCVA). Postoperative examinationswere performed at 1 day. The frequency of subsequent follow-up examinationsvaried according to the healing response and was determined by the surgeon.Statistical analysis was performed with the paired t testusing Microsoft Excel 2000 (Microsoft Corp, Redmond, Wash). A P value of .05 was considered statistically significant. For statisticalanalysis of UCVA and BSCVA, Snellen visual acuity was converted to logarithmof the minimum angle of resolution (logMAR) notation. Unless otherwise indicated,data are expressed as mean ± SD.

Twenty eyes of 19 patients were treated for clinically significant epithelialingrowth after LASIK with removal of the ingrowth and suturing of the LASIKflap (Table 1 and Table 2). Nine patients were men (10 eyes [50%]) and 10 were women(10 eyes [50%]). The mean age was 47.5 ± 9.9 years (range, 30-60 years).All eyes underwent primary LASIK, and 13 eyes (65%) had a history of LASIKretreatment. All eyes received a diagnosis of clinically significant epithelialingrowth (Figure 1). Six eyes (30%)had evidence of progressive, peripheral flap melting on examination. Ten eyes(50%) were referred from outside institutions for treatment.

Epithelial ingrowth was treated with mechanical debridement and suturingof the flap at a mean of 10.3 ± 7.9 months (range, 3 weeks to 27 months)after the most recent LASIK procedure (Figure2). All suture knots were buried at the time of surgery. Ten eyes(50%) had previous treatment before suturing of the flap was performed. Foureyes (20%) had more than 1 previous treatment. Previous treatments includedlifting the flap and scraping the epithelial ingrowth in 10 eyes (50%), liftingthe flap, scraping the ingrowth, and treating with alcohol in 3 eyes (15%),and lifting the flap, scraping the ingrowth, and suturing the flap down in2 eyes (10%). The treatments of the 2 eyes that underwent previous suturingof the flap were performed at outside institutions. Sutures were removed ata mean of 1.9 ± 0.6 weeks (range, 1-3 weeks). The mean last follow-upexamination after treatment was 10.5 ± 14.3 months (range, 1.5-64 months).

After treatment, there was no recurrence of clinically significant epithelialingrowth (Figure 3). One eye hadrecurrence of ingrowth at the flap margin that was not considered clinicallysignificant, did not progress, and did not require further removal.

The UCVA changed from 20/20 or better in 7 eyes (35%) and 20/40 or betterin 15 eyes (75%) preoperatively to 20/20 or better in 9 eyes (45%) and 20/40or better in 16 eyes (80%) at the last postoperative examination. There wasno significant change in the mean logMAR UCVA before (0.3 ± 0.5; range,−0.1 to 1.7) and after (0.2 ± 0.4; range, −0.1 to 1.7)(P = .40) surgery. No eyes lost 2 or more lines ofBSCVA. For eyes with documented BSCVA before and after treatment (n = 10),mean logMAR BSCVA before treatment was 0.0 ± 0.1 (range, −0.1to 0.3) and after treatment was 0.0 ± 0.1 (range, −0.1 to 0.1)(P = .60).

For eyes that had documented manifest refractions before and after treatment(n = 10), mean spherical equivalent changed from −0.21 ± 0.82diopters (D) (range, −1.25 to 1.00 D) before treatment to −0.53± 0.89 D (range, −2.50 to 0.38 D) at last follow-up (P = .30). The mean refractive shift was −0.31 ± 0.98 D(range, −1.75 to 1.25 D). Three eyes had hyperopic shifts (range, 0.13-1.25D), 6 eyes had myopic shifts (range, −0.25 to −1.75 D), and 1eye had no change in spherical equivalent.

There were no intraoperative or postoperative complications relatedto the epithelial ingrowth debridement and suturing of the flap. Specifically,there was no evidence of flap striae or diffuse lamellar keratitis as a resultof lifting the flap, scraping the keratectomy bed, and suturing the keratectomyflap. Three eyes had sequelae resulting from the epithelial ingrowth. Oneeye with preoperative flap melting and irregular astigmatism had postoperativeirregular astigmatism and also had minimal stromal edema associated with corneaguttata. The UCVA was 20/200 and corrected to 20/25 with a rigid gas-permeablecontact lens at 21 months after treatment for clinically significant epithelialingrowth. One eye with preoperative flap melting had slight haze in the interfacepostoperatively with a UCVA of 20/25. One eye had a history of a segmentedflap from multiple microkeratome cuts that resulted in central scarring anda BSCVA of 20/160. This eye was the eye that had a recurrence of clinicallyinsignificant epithelial ingrowth that did not require treatment. Becauseof the central scarring from the segmented flap, this eye underwent penetratingkeratoplasty 58 months after the epithelial ingrowth had been treated.

Most cases of epithelial ingrowth after LASIK are clinically benign,self-limited, and do not require treatment. Clinically significant epithelialingrowth is an uncommon but serious complication that can result in loss ofvision, astigmatism, and keratolysis if left untreated. With appropriate treatment,however, the final best-corrected visual acuity is usually not affected.¹¹ This study suggests that suturing the LASIK flapin addition to mechanical debridement of epithelial ingrowth is a safe treatmentfor clinically significant epithelial ingrowth after LASIK, and a treatmenteffective at preventing recurrence, loss of vision, and flap melting.

Two hypotheses exist on the pathogenesis of epithelial ingrowth afterLASIK.^3,11,12 Thefirst suggests that epithelial cells are implanted in the lamellar interfaceby the microkeratome blade or during irrigation of the stromal bed. However,isolated epithelial cells do not appear to spread.¹¹ Thesecond hypothesis is that epithelial cells grow under the edge of the flapand progress into the interface. This ingrowth generally stops, forming ademarcation line of mild fibrosis, and causes remodeling of the adjacent stromawith melting of the flap edge.¹¹

The continuity of the epithelial ingrowth with the surface epitheliummakes the second hypothesis more likely.^11,17 Ithas therefore been suggested that epithelial ingrowth consists of an epithelialfistula underneath the flap with a tract extending to the edge of the flap.¹¹ A fistula may form owing to poor flap adhesion, whichallows for surface epithelial cells to enter into the lamellar interface.Epithelial defects, basement membrane dystrophy, excessive hydration of theflap intraoperatively, or unrecognized factors may cause poor flap adhesionand predispose to fistula formation.¹¹ Oncethe fistula forms, remodeling of the stroma along the tract causes keratolysis.It has been recommended to treat clinically significant epithelial ingrowthas early as possible to prevent maturation of the fistula.¹¹

Based on the theory of ingrowth occurring in the form of a fistula extendingfrom the edge of the LASIK flap, the fistula must be closed to prevent recurrenceof the ingrowth. Scraping the epithelial ingrowth removes the contents ofthe fistula but does not close it, and recurrence of epithelial ingrowth aftertreatment suggests that the passive apposition of the LASIK flap to the stromalbed may not always be enough to close the fistula. Treatment by lifting theflap, scraping the posterior surface of the flap and the stromal bed, andreplacing the flap with a bandage lens to encourage tight flap adhesion resultedin a 44% recurrence rate of epithelial ingrowth, which included ingrowth thatwas clinically significant (23%) and that which was not (21%).¹¹

Adjunctive treatments including mitomycin, ethanol, and phototherapeutickeratectomy have been suggested as ways to reduce the high recurrence rateafter epithelial ingrowth removal. These treatments have potential complicationsand also do not close the fistula. Mitomycin may inhibit corneal wound healing.¹⁸ Ethanol has been shown to have damaging effects onkeratocytes¹⁹ and to cause diffuse lamellarkeratitis,¹³ and it can result in total flapmelting after application to the interface to treat recurrent epithelial ingrowth.¹⁵ Phototherapeutic keratectomy may shift the refractionand induce irregular astigmatism.^12,16 Theonly adjunctive treatment that closes the fistula is the placement of suturesat the site of ingrowth.^14,20

Our study shows that suturing the LASIK flap is an effective and safeadjunct to epithelial ingrowth removal after LASIK. There were no recurrencesof clinically significant epithelial ingrowth in the study eyes. One eye hada recurrence of ingrowth that did not become clinically significant, did notprogress, and did not require further treatment. Suturing of the flap didnot cause a significant change in the spherical equivalent manifest refractionat the last follow-up after the sutures had been removed. The treatment alsoshowed no adverse effect on the final UCVA or BSCVA, and no eyes lost 2 ormore lines of BSCVA. There were no complications resulting from the treatment.

Discussion has occurred about whether suturing is appropriate as a primaryform of treatment.²⁰ The 44% recurrence ratereported after epithelial ingrowth removal has been attributed to less aggressivetreatment (treatment after ingrowth was present for 1 month), which allowedthe fistula to become established.^11,20 Anearlier approach of treating clinically significant ingrowth if present atthe 3-week examination after surgery has reduced the recurrence rate, andsuturing has only been necessary in a limited number of cases out of severalhundred eyes.²⁰

Further study is needed to investigate the results of suturing of theLASIK flap as an adjunct to removal of epithelial ingrowth. Comparison ofsuturing as a primary form of treatment with its use as treatment for recurrentcases may help define its role in the removal of clinically significant epithelialingrowth. A larger sample size with longer postoperative follow-up would furthercharacterize refractive changes, stability of postoperative refraction, andrecurrence rates of epithelial ingrowth. This preliminary study suggests thatsuturing of the flap is a safe and effective treatment option for clinicallysignificant epithelial ingrowth after LASIK, and prevents loss of BSCVA dueto astigmatism and keratolysis that can be caused by progressive epithelialingrowth.

Correspondence: Edward E. Manche, MD, Department of Ophthalmology,Stanford University School of Medicine, 900 Blake Wilbur Dr, Stanford, CA94305 (edward.manche@stanford.edu).

Submitted for publication July 14, 2003; final revision received November10, 2003; accepted December 3, 2003.

This study was presented at the American Society of Cataract and RefractiveSurgery 2003 Symposium on Cataract, IOL and Refractive Surgery; April 13,2003; San Francisco, Calif.