|Home | About | Journals | Submit | Contact Us | Français|
To determine the refractive outcomes and complications of retreatment after aborted primary laser in situ keratomileusis (LASIK) due to flap complications.
This retrospective study evaluated 50 retreated eyes that had flap complications during primary LASIK at the Eye Consultants Center in Riyadh, Saudi Arabia. Data were analyzed for patients with at least 3 months follow-up post retreatment.
Thirty-three eyes of 31 consecutive patients with 3 months follow-up or later post retreatment were included. The primary LASIK was aborted due to incomplete flaps in 22 eyes (66.7%), buttonhole flaps in 7 eyes (21.2%), free partial flaps in 3 eyes (9.1%), and a free complete flap in 1 eye (3.0%). Twenty-two eyes (66.7%) were retreated with LASIK, and 11 eyes (33.3%) were retreated with surface ablation. The mean spherical equivalent (SE) was –0.23 ± 0.72 D, the mean astigmatism was –0.65 ± 0.89 D, and the mean loss of the best corrected visual acuity (BCVA) was 0.78 lines at the final postoperative visit. At the last postoperative visit, 20/30 or better BCVA was achieved in 90.1% of eyes that underwent retreatment with LASIK and in 91% of eyes that were retreated with surface ablation. There was no statistical difference in postoperative SE between eyes retreated with LASIK and eyes retreated with surface ablation (P = 0.610). There was no statistical difference in postoperative BCVA between eyes retreated with LASIK and those retreated with surface ablation (P = 0.756). There were no intraoperative complications and no eyes required a second retreatment.
Creation of a flap after a previous intraoperative flap complication was not associated with any complications. The refractive outcomes of retreatment with LASIK or surface ablation were comparable and reasonably favorable.
Laser in situ keratomileusis (LASIK) is the preferred surgical technique for the correction of ametropia.1–3 However, the success of LASIK depends on excellent flap creation. Although proper care and meticulous handling of the microkeratome can prevent most flap complications, they have been reported in 0.04–10% of cases.4–13 Prior to surgery and during the informed consent process, it is the responsibility of the refractive surgeon to inform patients about the possibility of flap complications. Flap complications can delay initial visual recovery or result in considerable patient dissatisfaction. When flap complications occur, the surgeon has to decide whether to proceed with stromal ablation with the excimer laser or to abort the procedure and retreat 3 months or later. Two retreatment alternatives are repeating the LASIK procedure and performing surface ablation.
The etiopathogenesis of a poor keratectomy is likely multifactorial. Incomplete flaps can be caused by jamming of the microkeratome from mechanical obstacles such as the speculum, drape, lid, lashes, loose epithelium, and gear-assembly or electrical failure.13,14 Other causes include damaged microkeratome blades, balanced salt solution deposits on the blade during keratectomy in the second eye, and surgeon inexperience.9 Buttonhole flaps occur when the microkeratome blade exits through the epithelium at mid-incision and subsequently re-enters to complete the flap. Research suggests steep corneas, small corneal diameters, deep eye sockets, loss of suction ring pressure, and conjunctival incarceration in the suction port leading to suction loss during the keratectomy can all cause buttonhole flaps.5,14 Additionally, the more friable epithelium of the second eye due to longer anesthetic action15 and aspects related to the microkeratome (e.g., blunted blades, microflaws of blades, and poor oscillation)16 have been suggested as predisposing factors for buttonholes. Free flaps (partial or complete) may occur due to a loss of suction during the microkeratome pass that causes shallow engagement of the keratome on the corneal surface, allowing the blade to skim the top of the cornea and produce a free cap.7
In this study, we report the surgical and refractive outcomes of the retreatment (LASIK or surface ablation) of patients whose primary LASIK procedure was aborted due to flap complications.
A chart review was conducted to retrospectively identify patients who were retreated after a primary LASIK procedure that was aborted due to flap complications. Data were collected on subjects who underwent retreatment with LASIK, laser subepithelial keratomileusis (LASEK) or photorefractive keratectomy (PRK). The primary LASIK and retreatments were performed between June 1999 and July 2008 at the Eye Consultants Center in Riyadh, Saudi Arabia. Fifty eyes underwent retreatment after primary LASIK. Only subjects with 3 months or longer follow-up after retreatment were included in the study. Data were collected at 3 months or later to ensure stability of visual outcomes.
Uniform surgical techniques were not possible as eight refractive surgeons performed primary LASIK. However, there were some common techniques followed by all surgeons during LASIK. For example, the right eye was always treated first in simultaneous, bilateral surgery cases. All LASIK flaps were created using either the Hansatome microkeratome (Bausch and Lomb Inc., Rochester, NY, USA) or the Moria LSK2 Carriazo-Barraquer manual microkeratome (Moria SA, Anthony, France). For the Hansatome microkeratome, superiorly placed hinged corneal flaps were created with a 160- or 180-μm microkeratome head and an 8.5- or 9.5-mm suction ring. For the Moria microkeratome, a 9.0-mm superior hinged flap was created with a 130-μm plate. For the Moria microkeratome, –1 ring was used for corneas with a mean keratometry of 41 diopters (D), a 0 ring was used for corneas with a mean keratometry of 41–43 D, a +1 ring was used for corneas with a mean keratometry of 43–46 D, and a +1 or +2 ring was used for corneas with mean keratometry greater than 46 D.
Prior to commencement of the procedure, the microkeratome was assembled by a surgical nurse assistant, and the suction ring and microkeratome blade were inspected by the surgeon under a microscope. The same blade was used for bilateral surgeries. The microkeratomes were serviced regularly in accordance with company recommendations or whenever difficulties were noted. All eyes underwent refractive ablation with the NIDEK EC-5000 Excimer Laser (Nidek Co. Ltd., Gamagori, Japan). The surgeon determined the type of surgical approach for retreatment.
Data on age, sex, laterality of the eye, refraction, keratometry, pachymetry, and the best corrected visual acuity (BCVA), pertinent to the primary and retreatment surgeries were recorded. Surgical details including microkeratome specification (style, ring size, depth of cut), type of flap complication, and the retreatment technique were also collected. Additionally, the interval between primary LASIK and retreatment as well as the postoperative follow-up after retreatment were collected. The Mann–Whitney test was used to compare the final SE (SE at final postoperative visit after retreatment) and final logarithm of the minimum angle of resolution (LogMAR) BCVA between the eyes that underwent retreatment with LASIK and the eyes that were retreated with surface ablation. A P value less than 0.05 was considered statistically significant.
Postoperative data at 3 months or later were available on 33 eyes of 31 consecutive patients. The remaining subjects were lost to follow-up. The mean age of the 31 patients (18 men, 13 women) in this series was 27.9 ± 8.1 years (range, 18–43 years). Of the 33 eyes, 14 were right eyes and 19 were left eyes (two cases developed bilateral flap complications).
Three different types of intraoperative flap complications occurred during primary LASIK: incomplete flaps in 22 eyes (66.7%), buttonhole flaps in 7 eyes (21.2%), free partial flaps in 3 eyes (9.1%), and a free complete flap in 1 eye (3.0%). All buttonhole flaps occurred centrally and occupied approximately 10-20% of the surface area of the flap.
Retreatment with LASIK was performed in 22 eyes (66.7%) [Table 1]. The remaining 11 eyes (33.3%) underwent surface ablation [Table 2]: 10 eyes underwent LASEK (with the use of mitomycin-C in case 1) [Table 2] and 1 eye (case 2) [Table 2] underwent PRK. For each type of flap complication, most of the patients were retreated with LASIK with the exception of cases of free partial flaps [Tables [Tables11 and and22].
Prior to the primary LASIK procedure, the mean spherical equivalent (SE) was –4.00 ± 2.80 D (range, –14.75 to –1.00 D), the mean astigmatism was –0.62 ± 0.86 D (range, –4.25 to 0 D), and the mean keratometry was 43.5 ± 1.7 D. Immediately preceding retreatment, the mean loss of BCVA was 0.43 lines of Snellen visual acuity, the mean SE was –3.94 ± 2.86 D (range, –13.75 to 0 D), and the mean astigmatism was –0.54 ± 0.77 D (range, –2.50 to +1.5 D). There was a myopic change in refraction after primary LASIK and just prior to retreatment of –0.35 ± 0.24 D in 14 eyes. There was a hyperopic change in refraction after primary LASIK and just prior to retreatment of +0.58 ± 0.41 D in 13 eyes. The mean time to retreatment was 3.9 months (range, 1–25 months).
The mean follow-up after retreatment was 23 months (range, 3–115.3 months). The final mean SE refraction was –0.23 ± 0.72 D (range, –2.00 to +1.50 D), the mean astigmatism was –0.65 ± 0.89 D (range, –4.25 to 0 D), and the mean final loss of BCVA was 0.78 lines after retreatment.
For eyes that underwent retreatment with LASIK, the mean pre-retreatment SE was –4.57 ± 2.77 D (range, –13.25 to –1.25 D) and the final SE was –0.1 ± 0.72 D (range, –2.00 to +1.50 D); the mean pre-retreatment astigmatism was –0.52 0.70 D (range, –2.50 to 0 D) and the final mean astigmatism was –0.72 ± 0.1 D (range, –3.00 to 0 D). The final mean loss of BCVA was 0.9 lines. There were no intraoperative or postoperative complications for all eyes that underwent retreatment with LASIK and no eyes required a second retreatment. Details of the microkeratome used in both surgeries (primary and retreatment) as well as the flap complication during primary LASIK are summarized in Table 3. Of the 33 eyes that had flap complications during primary LASIK, the Hansatome microkeratome was used on 22 eyes (66.7%), and the Moria microkeratome was used on 11 eyes (33.3%). For the 22 eyes that underwent retreatment, the Hansatome microkeratome was used on 13 eyes (59.1%), and the Moria microkeratome was used on 9 eyes (40.9%).
For the 11 eyes that underwent retreatment with surface ablation, the mean pre-retreatment SE was -2.7 ± 2.5 D (range, –8.25 to 0 D) and the final mean SE was –0.50 ± 0.61 D (range, –1.75 to 0 D); the mean pre-retreatment astigmatism was –0.60 ± 0.92 D (range, –2.00 to +1.5 D) and the final mean astigmatism was –0.47 ± 0.52 D (range, –1.5 to 0 D). The final mean loss of BCVA was 0.55 lines.
There was no statistical difference in the final SE between eyes that underwent retreatment with LASIK and those that underwent retreatment with surface ablation (P = 0.610). There was no statistical difference in the final LogMAR BCVA between eyes that underwent retreatment with LASIK and those that underwent retreatment with surface ablation (P = 0.756).
Prior to primary LASIK, 88% of eyes had a BCVA of 20/20 or better. After the flap complication and immediately preceding the retreatment, 68.7% of eyes achieved a BCVA of 20/20 or better and 93.7% achieved a BCVA of 20/40 or better. Two eyes lost 3 lines or more of BCVA, secondary to a central corneal scar due to a buttonhole flap. After a minimum of 3 months follow-up, 90.1% of eyes that underwent retreatment with LASIK achieved a final BCVA of 20/30 or better and 54.5% of eyes achieved a final BCVA of 20/20 or better. A loss of >2 lines of BCVA occurred in 2 (9.1%) of the 22 eyes that underwent retreatment with LASIK. Of these, one eye had a significant linear corneal scar that was secondary to a free partial flap bisecting the visual axis and the other lost 6 lines of BCVA that was secondary to a significant central corneal scar, which was secondary to a buttonhole flap. By comparison, 91% of eyes that underwent surface ablation achieved a final BCVA of 20/30 or better and 45.4% of eyes that underwent surface ablation achieved a final BCVA of 20/20 or better. The mean loss of final BCVA for the group of eyes that had buttonhole LASIK flaps was 1.85 lines, whereas the mean loss of final BCVA of eyes that had incomplete flaps was 0.65 lines.
The aim of this retrospective, observational case series was to determine the outcomes of LASIK or surface ablation retreatment in patients with intraoperative flap complications during the primary LASIK procedure. In our series, recutting the cornea was not associated with another flap complication. Generally, the visual outcomes for retreatment with LASIK or surface ablation were reasonably favorable. The least loss of BCVA was achieved in retreated eyes that had incomplete flaps. By comparison, retreated eyes with buttonhole flaps or partially free flaps involving the visual axis or the paracentral area attained the worst BCVA.
After incomplete, free (partial or complete), or buttonhole flaps occur, the conventional management approach involves inspection of the abnormal lamellar section and carefully replacing the abnormal corneal tissue to achieve the best realignment. Placement of therapeutic bandage contact lenses is recommended to protect the irregular flap of corneal tissue from the eyelids and to promote healing of the damaged corneal tissue. After complete corneal healing, the surgeon can consider retreatment (at least 3 months after the flap complication) to achieve better final visual outcome.17
Our study concurs with previous reports7,9,18–20 that indicate recutting the cornea is not associated with flap complications. In contrast, other studies have reported a higher risk of intraoperative flap complications during retreatment.17,21 We agree with Tham and Maloney17 who suggest a minimum of 3 months prior to retreatment with LASIK, as the original flap is likely to be more adherent after this interval which enables a safer repeat keratectomy.
According to Jabbur et al.,18 a different microkeratome suction ring and/or plate should be selected to cut a larger, deeper, and more temporal or inferior flap compared to the original flap. Our findings (same thickness flaps, 9 eyes; thinner flaps, 9 eyes; thicker flaps, 4 eyes) indicate that theoretically most corneal flap thicknesses during retreatment were either the same or thinner compared to the original flap [Table 2]. However, we do not routinely perform intraoperative pachymetry; therefore, the actual flap thickness may differ compared to the intended flap thickness in our study. Hence, we cannot determine whether the creation of a thicker flap is an important factor for avoiding flap complications during LASIK retreatment.
Although none of our cases had flap complications during LASIK retreatment, the potential for complications remains. Recutting with a mechanical microkeratome is subject to the same risks as the primary procedure.22,23 Furthermore, the depth of the initial cut created by a microkeratome varies from the periphery to the center (the center is thinner than the periphery). This variation in corneal thickness over the entire flap makes it difficult to accurately plan the necessary depth of the second mechanical microkeratome cut.24,25
The improved predictability of flap thickness with the femtosecond laser26–28 makes it ideal for recutting procedures. The femtosecond laser produces a uniform flap regardless of position (center or peripheral). This feature might be important if the previous LASIK flap was complicated by a buttonhole perforation or an incomplete or free flap; the repeat flap is designed to cut below the original flap.28 Additionally, the rate of epithelial ingrowth after LASIK retreatment is statistically significantly lower in eyes with femtosecond laser keratectomies compared with keratectomies with a mechanical microkeratome.29 A similar observation was reported by Kamburoglu and Ertan.30 It has been hypothesized that mechanical microkeratomes produce a tapered flap edge, which allows for easier access of the epithelium to the stromal interface. On the other hand, femtosecond lasers create a nearly vertical flap edge resulting in a more effective barrier against epithelial migration.29 Additionally, mechanical microkeratomes cut tissue by sliding the blade on the stromal interface, which increases the risk of inoculating the epithelium.29
Most flap complications, such as incomplete flaps, usually do not cause long-term loss of vision. Flap complications that cut through Bowman's layer, such as buttonhole flaps and free flaps, tend to be more serious (albeit less common). This can be compounded if the abnormal flap involves or extends over the entrance pupil as this can result in irregular astigmatism and substantial loss of BCVA.7,20 In this study, the mean loss of BCVA in the group of eyes that had buttonhole flaps was 3 times higher than in the group of eyes that had incomplete flaps. In addition, the worst visual outcome was achieved in eyes that had buttonhole flaps or free partial flaps involving the visual axis (case 5 and case 7 of the repeat LASIK group, respectively) [Table 1].
In conclusion, recutting the cornea after primary LASIK was not associated with flap complications. The visual outcomes for retreatment by either repeat LASIK or surface ablation were comparable and reasonably favorable. Incomplete flaps caused the least loss of BCVA in retreated eyes. By comparison, retreated eyes with previous buttonhole flaps or partially free flaps involving the visual axis attained the worst BCVA.
Source of Support: Nil,
Conflict of Interest: None declared.