Management of Unsuccessful LASIK Surgery

Renato Ambrósio, JR, MD, PhD; Daniela Jardim, MD; Marcelo V.Netto, MD; Steven E. Wilson, MD


Compr Ophthalmol Update. 2007;8(3):125-141. 

In This Article

Surgical Management

Overcorrection, undercorrection, and residual or induced astigmatism are commonly noted after LASIK.[7,8] Although unpredictable wound healing11 and corneal biomechanical changes[12] are thought to underlie corrections that deviate from the preoperative plan, in most cases, the specific etiology is not clear. Often, low levels of residual refractive error are compatible with relatively good visual function, and the potential risks of re-treatment (for example, the uncorrected vision becoming worse than before re-treatment) outweigh the potential benefits. However, many patients seek re-treatment to improve uncorrected distance vision despite concerns expressed by the surgeon. Ultimately, the surgeon must decide whether additional surgical treatment is advisable and, if it is not, he or she must be willing to educate the patient regarding concerns and must insist on the patient following the appropriate course of action. In some cases, even a relatively small undercorrection or regression, such as -0.5 diopter, can be safely corrected with LASIK enhancement and markedly improve patient satisfaction. If there are no contraindications, such as insufficient posterior residual bed or severe LASIK-induced dry eye, we often provide such re-treatments with excellent results.

Re-treatment rates reported in the literature vary from 3% to 37.9%.[5,6,7,8,32,71] There are two possible techniques for re-treatment of eyes that have had LASIK: re-lifting the original flap or re-cutting a new flap. Reports have suggested that both techniques are safe and effective,[72] but it is the opinion of the authors that re-treatment should be performed by re-lifting the flap to avoid serious flap complications, such as intersecting lamellar cuts that have the potential to seriously compromise vision.

The first consideration when thinking about LASIK re-treatment is whether there is sufficient residual posterior stromal tissue for additional tissue removal to avoid progressive iatrogenic ectasia.[73] Any intervention should leave a minimum of 250 micrometers of residual central stroma, although as much tissue as possible should remain at the end of the procedure, as rare corneas that have greater than this amount have developed ectasia.[7] At a minimum, the corneal topographic pattern and central corneal thickness should be evaluated to avoid re-treatments in corneas that already show signs of ectasia. Regardless of how carefully the surgeon evaluates these cases, however, there remains a risk of ectasia, and the patient should be advised of this risk.

In the future, corneal biomechanics measurements, such as CH and CRF provided by the Ocular Response Analyzer, may play an important role in this decision-making process,[12] but at present, there are limited clinical data to support this concept.

Once the refraction is stable for several months, typically between 3 months and 6 months after the original LASIK surgery, we perform the re-treatment. If the eye has consecutive hyperopia after treatment of myopia, we typically subtract at least 0.5 diopter from the cycloplegic spherical equivalent for the laser treatments because of the well-known tendency to obtain greater effect than expected in such enhancements (see more specific guidelines following).

Our technique is to use a Sinskey hook to delineate a 1-clock-hour arc of flap edge at 180º from the hinge and then to carefully use toothed forceps to peel back the flap. This technique tends to produce a very clean flap edge similar to a capsulorhexis in cataract surgery. Once the laser is applied, we reposition the flap, taking special care to not entrap epithelium at the edge that would predispose to epithelial ingrowth. If the edge of the flap is smooth, we avoid a contact lens. If there is significant epithelial trauma at the flap edge, we typically use a bandage contact lens for the first 24 hours.

The risk of progressive ectasia after multiple re-treatments has prompted some surgeons to propose undersurface ablation of the flap for LASIK re-treatment for eyes with insufficient posterior stroma.[74] However, we advise against this approach, as there are inherent difficulties related to centration of the treatment when the pupil is not utilized during the ablation.

Surface ablation with intraoperative mitomycin C (MMC) to prevent haze is frequently a safer option in cases where there is concern that the posterior residual bed is insufficient.[31,75,76] Surface ablation may provide better registration for custom treatments and, theoretically, should have less biomechanical effect on the corneal architecture.

Surface re-treatments should be considered as a last option before corneal transplantation, even if the cornea will be thin (Figures 2 and 8). Thus, Vinciguerra et al,[77] reported a series of 31 corneas with ultrasonic pachymetry ranging from 258 to 466 ìm that were stable after 5 years of follow-up following surface retreatment with phototherapeutic keratectomy. There are some reports of progressive ectasia after surface ablation.[78,79,80] Therefore, the patient should be informed that there is a possibility that corneal transplantation will still be needed to restore vision. Intensive topical corticosteroid treatment should be given whenever surface ablation is performed in an eye with a prior LASIK flap because of the high risk that diffuse lamellar keratitis will be triggered. Our approach is to give hourly 1% prednisolone acetate or 0.5% loteprednol etabonate for the first 2 to 3 days after surgery and then to taper slowly over the next week to 10 days.

Customized excimer laser re-treatment may improve quality of vision and symptoms such as halos and ghosting.[32] As the technology continues to improve, more and more eyes with irregular corneal topography can be improved to the point that functional vision is restored. Some cases with irregular topography—for example, decentered ablations (Figures 9, 10, and 11)—require customized ablation for a surgical approach to symptoms.

There are several important factors that must be considered when planning therapeutic wavefrontguided custom ablations ( Table 3 ). Whole eye wavefront measurements are difficult to obtain in many cases, and they are frequently not possible due to severe irregularity that confounds the measurements. In some of these cases, customized ablations can be based on corneal topography. In other cases, an irregularity debulking procedure, such as phototherapeutic keratectomy (PTK) with a masking agent, must be performed to reduce irregularity so that meaningful wavefront analysis or topography can be obtained for customized excimer laser ablation.[81,82,83,84] A common masking agent used by many surgeons is a sterile 10% solution of sodium hyaluronate diluted in physiological saline (although some surgeons have reported good results with nonviscous artificial tears or even physiological saline alone). The procedure is performed by covering the area of the cornea to be ablated with a thin layer of masking agent. The PTK is then performed with a broad beam laser over the masking agent while a spatula is repeatedly passed over the corneal surface to respread the masking agent. In this way, the peaks that protrude above the masking agent are ablated, the valleys are protected, and, therefore, little change in overall corneal thickness tends to occur. If done properly, surprisingly high amounts of irregularity can be markedly reduced without much in the way of hyperopic shift associated with PTK alone.

The best approach to customized re-treatment appears to be through integration of whole eye wavefront analysis with topographic corneal elevation data. Elevation topography provides important information regarding corneal surface shape within the pupil that guides the surgeon in his or her decision as to whether customized ablation can be used directly, as opposed to performing a preliminary smoothing procedure to eliminate features that preclude meaningful wavefront analysis or customized laser ablation. Thus, in cases where the surface contour is so irregular that meaningful topography cannot be obtained, a staged approach in which PTKsmoothing precedes customized excimer laser ablation is more likely to be successful. Topographic data are also essential for planning treatment outside the pupil—for example, when custom expansion of the functional optical zone is indicated or when the surgeon must add a peripheral transition zone.

The correlation between the total wavefront analysis and corneal topography may also provide important prognostic information. Thus, optimal results of therapeutic wavefront-guided custom treatments are more likely when the high order total wavefront analysis matches the corneal elevation results obtained from corneal topography (Urbano and Nosé, unpublished data, 2004).

The wavefront-derived autorefraction should be compared to the manifest and cycloplegic refractions. We recommend performing subjective manual refraction using the objective wavefront parameters as a beginning point to determine whether there is a reasonable correlation between these two measures. The greater the disparity between the wavefront analysis-derived refraction and the manifest/cycloplegic refraction, the more concern the surgeon should have about the outcome of a custom therapeutic treatment yielding an unexpected result that does not improve vision in the eye. If the disparity is great, the best approach may be to perform PTK-smoothing initially, followed by re-evaluation and custom treatment as a secondary procedure, based on parameters that are hopefully more reliable.

It is important to follow manufacturer's recommendations when obtaining wavefront measurements that will be used for customized therapeutic treatment. Thus, if the wavefront instrument is designed for use with the pupil dilated, the measurements should be obtained with the eye dilated. Nomograms are, for the most part, meaningless for customized therapeutic treatments because of the great variability among different eyes. However, the concept of there being an enhanced effect of a hyperopic ablation in an eye with overcorrection for myopia[85] typically holds true for therapeutic procedures as well. Thus, for levels of hyperopia overcorrection between +0.75 and 2.0 diopters, we recommend a decrease of 0.5 diopters in the attempted retreatment. For levels above +2.0 diopters, we recommend a reduction of 15% to 20%.

Ectasia has emerged as a rare but devastating complication of LASIK surgery. Progressive steepening of the corneal contour, myopia, irregular astigmatism, loss of best spectaclecorrected visual acuity, and poor vision quality characterize this condition. Lowering IOP with medications and avoiding eye rubbing are measures that may decrease progression in some patients.[86] Some cases have been reported in which there was a decrease of corneal curvature in response to lowering IOP.[86] Rigid gas-permeable contact lenses[70] often restore best spectacle-corrected vision in patients with ectasia. One should attempt to treat the patient with contact lenses before considering additional surgical measures. Surgical approaches to improving corneas in eyes that cannot be fit with gas-permeable contact lenses, or in cases where the patient cannot successfully wear the contact lenses, include insertion of corneal intrastromal rings (e.g., Ferrara rings, Keraring [Rockmed B.V., Oirschot, the Netherlands, Intacs®, Addition Technology, Des Plaines, IL])[87,88] and lamellar keratoplasty or penetrating keratoplasty.

An experimental approach to progressive ectasia is collagen crosslinking. The current method for cross-linking under investigation is application of riboflavin, followed by ultraviolet A (UVA) light treatment.[89] Riboflavin-UVA crosslinking has stopped progression of the ectatic process in patients with keratoconus who have not had LASIK.[89] In one report,[89] the ectatic process appeared to be reversed in some patients with keratoconus. Early experience with riboflavin-UVA cross-linking in LASIK-induced ectasia suggests promising results (Jankov, personal communication during the ESCRS meeting, London 2006), but definitive studies are not yet available.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.