ARTICLE
Corneal wavefront–guided customized laser in situ keratomileusis after penetrating keratoplasty Serhat Imamoglu, MD, Vedat Kaya, MD, Deniz Oral, MD, Irfan Perente, MD, Berna Basarir, MD, Omer Faruk Yilmaz, MD
PURPOSE: To assess the efficacy and safety of corneal wavefront-guided custom laser in situ keratomileusis (LASIK) to correct refractive errors and higher-order aberrations (HOAs) after penetrating keratoplasty (PKP). SETTING: Beyoglu Eye Training and Research Hospital, Istanbul, Turkey. DESIGN: Noncomparative case series. METHODS: The study comprised consecutive patients who were unable to tolerate spectacles or contact lenses for the correction of anisometropia after PKP and had corneal wavefront-guided custom LASIK. The uncorrected (UDVA) and corrected (CDVA) distance visual acuities, manifest refraction spherical equivalent (MRSE), and corneal HOAs were recorded before and after LASIK. RESULTS: The study evaluated 11 eyes (10 patients). The mean follow-up after LASIK was 24.3 months G 11.7 (SD) (range 9 to 36 months). The mean MRSE was 2.97 G 1.66 diopters (D) (range 0.50 to 5.38 D) preoperatively and 0.88 G 0.96 D (range 2.75 to 0.5 D) postoperatively. The mean total higher-order root mean square (RMS) was 4.65 G 1.14 mm (range 2.26 to 5.94 mm) preoperatively and 2.71 G 1.31 mm (range 1.22 to 5.33 mm) postoperatively. Postoperatively, the UDVA improved in 7 eyes and remained unchanged in 4 eyes. The CDVA improved in all eyes postoperatively even cases in which the attempted astigmatic correction was not totally achieved. CONCLUSIONS: Corneal wavefront-guided custom LASIK after PKP did not totally correct both refractive errors and HOAs because of the high volume of laser ablation required and inadequate corneal stromal thickness. Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2014; 40:785–792 Q 2014 ASCRS and ESCRS
Visual outcomes after penetrating keratoplasty (PKP) are unpredictable because the surgery often induces significant astigmatism. Astigmatism after keratoplasty is typically irregular. In addition, high levels of spherical refractive errors and higher-order aberrations (HOAs) are often present.1 On the other hand, patients with good corrected distance visual acuity (CDVA) often have problems related to anisometropia. When irregular astigmatism or anisometropia is present, spectacles provide limited visual rehabilitation while rigid gas-permeable (RGP) contact lenses typically give better results. However, the abnormal corneal shape and eyelid problems that are often present in eyes after PKP usually limit the use of contact lenses and prompt patients to seek surgical treatment. Q 2014 ASCRS and ESCRS Published by Elsevier Inc.
Until excimer lasers became available, incisional refractive surgical procedures, such as astigmatic keratotomy and wedge resection, were used with limited success to correct astigmatism after keratoplasty.2 Photorefractive keratectomy (PRK) has been used successfully to correct spherical and cylindrical refractive errors; however, haze formation and loss of CDVA are frequent postoperative complications.3 Although laser in situ keratomileusis (LASIK) brings the added risk for flap-related and wound-related complications in eyes with corneal grafts, it has largely replaced PRK for the correction of refractive errors after keratoplasty.4,5 Although standard excimer laser ablation patterns can address spherocylindrical refractive errors only, 0886-3350/$ - see front matter http://dx.doi.org/10.1016/j.jcrs.2013.10.042
785
786
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
the advent of topography-guided ablation systems made it possible to perform custom treatments and correct irregular astigmatism. The differences in curvature, thickness, and size between the graft and the recipient cornea might lead to misalignment at the graft–host interface. Also, variable tension of the sutures induces irregular astigmatism by increasing HOAs after PKP.6 After keratoplasty, eyes have nearly 5.5 times more HOAs than normal eyes. Trefoil is reportedly the most dominant HOA.1 By addressing both lower-order aberrations (LOAs) and HOAs, wavefront-guided LASIK has the potential to effectively correct irregular astigmatism and maximize visual improvement after keratoplasty. The corneal wavefront is derived from corneal topography. Corneal wavefront–guided treatment is basically advanced topography-guided ablation. The technique gives the surgeon the ability to selectively correct corneal HOAs along with refractive errors. After PKP, anterior and posterior corneal HOAs are reported to be significantly higher than in normal eyes.7 In this study, we assessed the results of corneal wavefront–guided custom LASIK for the correction of LOAs and HOAs after PKP.
Surgical Planning On the day of the surgery, corneal topography maps were taken and the best map was chosen with respect to reliability indices of the maps. The Placido-disk topography unit used in this study provides corneal elevation data using the arcstep method. The device compares the data obtained from the corneal elevation map with the reference ideal eye model with a Q-value of 0.25. The elevation difference between the 2 surfaces creates an optical path difference, which is expressed by Zernike polynomials (up to the 7th order) to represent the corneal wavefront. The Optimized Refractive Keratectomy-Custom Ablation Manager software was used to plan the treatment with the Schwind Esiris excimer laser platform (both Schwind eyetech-solutions GmbH and Co. KG). This platform uses a 0.8 mm para-Gaussian profile flying-spot laser at 200 Hz and a 330 Hz eye tracker. The software uses topographic data (obtained with the Placido-disk topography device) and other clinical parameters (manifest refraction, pachymetry, and flap thickness) to determine the most appropriate custom ablation profile to correct LOAs and HOAs. Using the software, the surgeon can change the optical zone diameter and select the aberrations to be treated. The ablation zone diameter was determined according to the scotopic pupil size, and full correction of LOAs and HOAs were targeted in all eyes. The ablations were planned to leave a minimum postoperative stromal bed thickness of 250 mm or 50% of the preoperative thickness. The aim of the treatment was to decrease HOAs and anisometropia primarily.
PATIENTS AND METHODS This retrospective noncomparative clinical case series comprised consecutive patients who had corneal wavefront–guided custom LASIK between December 2005 and February 2008 to correct refractive errors and HOAs after PKP. All patients were unable to tolerate spectacle correction. Rigid gas-permeable contact lens correction was attempted in 4 eyes without success; the other patients refused to try the lenses. The study inclusion criteria were a clear full-thickness corneal graft with an endothelial cell count (ECC) above 1500 cells/mm2, a stable refractive error for at least 12 months after all sutures had been removed, and the absence of ocular pathology other than the indication for PKP. All patients received a detailed explanation about wavefront-guided LASIK, including potential risks and benefits, after which they provided informed consent. The same surgeon (V.K.) performed all LASIK and PKP procedures.
Submitted: August 4, 2012. Final revision submitted: October 5, 2013. Accepted: October 14, 2013. From the Eye Clinic (Imamoglu), Haydarpasa Numune Education and Research Hospital, a private practice (Kaya), the Department of Ophthalmology (Oral), Faculty of Medicine, Yeditepe University, and Beyoglu Eye Training and Research Hospital (Perente, Basarir, Yilmaz), Istanbul, Turkey. Corresponding author: Serhat Imamoglu, MD, Tıbbiye Caddesi € €udar, Istanbul, _ Number: 40, Usk Turkey. E-mail: ophserhat@ hotmail.com.
Surgical Technique In all eyes, a Carriazo-Pendular microkeratome (Schwind eye-tech-solutions GmbH and Co. KG) was used to create a superior, hinged 160 mm thickness after the flap had been lifted. After excimer laser ablation, the flap was placed back on the stromal bed and a plano soft contact lens was placed over the cornea until the first postoperative visit. Prednisolone acetate 1% and tobramycin 0.3% eyedrops 4 times a day and preservative-free artificial tears every hour were prescribed. Tobramycin drops were used for 10 days, and the prednisolone acetate drops tapered over 6 weeks.
Patient Evaluation The uncorrected distance visual acuity (UDVA), manifest refraction, CDVA, and corneal wavefront aberrations as well as subjective complaints were evaluated preoperatively and at the last postoperative visit. The preoperative workup also included tonometry, slitlamp biomicroscopy, fundus examination, pupillometry (Colvard pupillometer, Oasis Medical, Inc.), scanning-slit topography (Orbscan IIz, Orbtek, Inc.), and Placido-disk topography (Keratron Scout, Optikon 2000 SpA). Ultrasonic pachymetry (Tomey AL-3000, Tomey Corp.) measurements were taken at 9 points over the corneal graft including the center and 4 points in 3.0 mm and 7.0 mm zones. Endothelial cell counts were performed with a noncontact specular microscope (Noncon Robo FA-3509, Konan Medical). The change in astigmatism achieved at the last follow-up was evaluated using Alpins vectorial analysis.8 The following vectors were determined and evaluated: targetinduced astigmatism (TIA) as the vector of the intended
J CATARACT REFRACT SURG - VOL 40, MAY 2014
787
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
Table 1. Patient characteristics. Parameter Eyes (n) Age (y) Mean Range Sex (n) Male Female Transplant indication (n) Keratoconus Corneal dystrophy
Table 2. Visual and refractive outcomes. Value 11 32.3 22, 53 6 4 8 2
change in cylinder for each treatment and surgically induced astigmatism (SIA) as the vector of the actual change achieved.
Statistical Analysis Preoperative and postoperative data were compared by using paired t test, and P values less than 0.05 were considered statistically significant. Snellen visual acuity scores were used for statistical analysis.
Parameter UDVA (Snellen lines) Mean G SD Range Cylinder (D) Mean G SD Range MRSE (D) Mean G SD Range CDVA (Snellen lines) Mean G SD Range
Preoperative Postoperative P Value 0.19 G 0.12 0.05, 0.40
0.52 G 0.27 0.05, 0.80
.002
5.20 G 1.89 1.00, 7.50
3.11 G 1.61 0.00, 5.00
.011
2.97 G 1.66 0.50, 5.38
0.88 G 0.96 2.75, 0.50
.002
0.50 G 0.18 0.30, 0.80
0.90 G 0.15 0.60, 1.00
!.001
CDVA Z corrected distance visual acuity; MRSE Z manifest refraction spherical equivalent; UDVA Z uncorrected distance visual acuity
(range 518 to 660 mm). The mean laser ablation depth was 71 G 11 mm (range 47.9 to 85.3 mm), and the mean postoperative theoretical RSB thickness was 320 G 43 mm (range 279 to 431.2 mm). Visual Acuity and Refraction
RESULTS The study included 11 eyes of 10 patients. Table 1 shows the patients' demographics. The mean interval from PKP to LASIK was 66.8 months G 30.2 (SD) (range 23 to 96 months). One patient had bilateral treatment. The mean follow-up after LASIK was 42.9 G 21.8 months (range 10 to 72 months). In 6 of the 11 eyes, the theoretical ablation depth exceeded the thickness of corneal lenticule available for ablation due to a high amount of aberrations. In these eyes, the ablation zone diameter was decreased or undercorrection was targeted to maintain the minimum residual stromal bed (RSB) thickness. The mean preoperative graft thickness was 552 G 42 mm
Table 2 shows the visual and refractive outcomes. Preoperatively, all eyes had a UDVA of worse than 20/40 and 5 (45%) eyes had a UDVA of 20/200 or worse. After LASIK, 1 (9%) eye had a UDVA of worse than 20/200 and 7 (63%) eyes had a UDVA of 20/40 or better (Figure 1). The mean postoperative UDVA was statistically significantly better than the mean preoperative UDVA (PZ.002). After LASIK, the UDVA improved in 7 (63%) eyes and remained unchanged in 4 eyes (36%). One eye (9%) gained 1 line, 1 eye (9%) gained 4 lines, 2 eyes (18%) gained 5 lines, 1 eye (9%) gained 6 lines, 1 eye (9%) gained 7 lines, and 1 eye (9%) gained 8 lines of UDVA. There was also a statistically significant decrease in the mean manifest refraction spherical equivalent (MRSE) (PZ.002) and in the mean refractive astigmatism (PZ.011) between preoperatively and postoperatively. The attempted astigmatic correction was not totally achieved in all cases (Table 3). The decrease in the mean refractive astigmatism was statistically significant. Predictability
Figure 1. Percentage of eyes with a certain level of cumulative preoperative Snellen CDVA and of postoperative UDVA at last follow-up (CDVA Z corrected distance visual acuity; UDVA Z uncorrected distance visual acuity).
The postoperative achieved correction was within G1.00 D of the intended correction in 7 eyes (64%). The mean magnitude of the TIA was 5.73 G 2.52 D (range 1.01 to 8.24 D), and the mean magnitude of the SIA at last follow-up was 3.96 G 3.32 D (range 0.07 to 11.1 D). The difference between the TIA and
J CATARACT REFRACT SURG - VOL 40, MAY 2014
788
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
Table 3. Values of refractive astigmatism in each eye. Astigmatism (D) Eye 1 2 3 4 5 6 7 8 9 10 11
Preoperative
Postoperative
Achieved Change
6 120 5 131 6.5 100 5.5 156 7.5 96 4 28 4 25 1 82 4 18 7.5 150 7.5 34
3.25 180 3.75 140 3.00 90 5.00 155 5.00 150 2.00 25 0.00 90 1.00 80 2.75 5 4.00 165 4.50 25
2.75 1.25 3.50 0.50 3.00 2.00 4.00 0.00 1.25 3.50 3.00
the SIA were statistically significant (P!.05). Therefore, there was a trend toward undercorrection of the refractive astigmatism after wavefront-guided LASIK for irregular astigmatism (Figure 2). In an ideal complete correction, the SIA and the TIA would be identical.
Higher-Order Aberrations Figure 3 shows the changes in the mean corneal HOAs. The mean preoperative total root mean square (RMS) was 4.65 G 1.14 mm (range 2.26 to 5.94 mm). The postoperative mean total RMS decreased to 2.71 G 1.31 mm (range 1.22 to 5.33 mm), showing a statistically significant improvement (PZ.001). The mean preoperative coma was 0.77 G 0.23 mm (range 0.48 to 1.20 mm). The postoperative mean coma was 0.43 G 0.21 mm (range 0.04 to 0.79 mm). This represented a statistically significant reduction (PZ.002). The mean preoperative trefoil was 1.18 G 0.52 mm (range 0.36 to 2.19 mm). The postoperative mean trefoil was 0.92 G 0.32 mm (range 0.41 to 1.34 mm). The change was not statistically significant (PO.05). The mean preoperative spherical aberration was 0.39 G 0.23 mm (range 0.01 to 0.77 mm). The postoperative mean spherical aberration was 0.40 G 0.25 mm (range 0.11 to 0.94 mm). The change was not statistically significant (PO.05). Complications and Enhancement
Safety Postoperatively, no eye lost CDVA lines. All eyes had a CDVA better than 20/40, and 9 eyes (82%) had a CDVA of 20/25 or better. After corneal wavefront-guided LASIK, the CDVA improved in all eyes. Two eyes (18%) gained 1 line, 4 eyes (36%) gained 3 lines, and 5 eyes (45%) gained 6 lines of CDVA. There was a statistically significant improvement in the mean CDVA between preoperatively and postoperatively (P!.001). No eye had corneal ectasia or recurrent keratoconus on topographic examination.
Figure 2. Relationship between the achieved postoperative spherical equivalent (SE) correction and the attempted SE correction.
Limited epithelial ingrowth that did not require intervention was observed at the flap border in 2 eyes (18%). No enhancements were performed. Case Report A 25-year-old man had PKP in the right eye for keratoconus 7 years previously. The patient presented with contact lens intolerance and 4.00 D of mixed astigmatism. He reported reduced vision, halos, glare, and sunburst at night. The preoperative UDVA was 0.1, and the manifest refraction was 1.75 4.00 with a CDVA of 0.4. Corneal wavefront-guided custom LASIK was performed to correct both HOA-related symptoms and anisometropia. The preoperative corneal pachymetry of the graft was
Figure 3. Higher-order corneal aberrations in eyes that had corneal wavefront-guided custom LASIK after PKP (* Z P!.05; HOA Z higher-order aberrations; SA Z spherical aberration).
J CATARACT REFRACT SURG - VOL 40, MAY 2014
789
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
Table 4. Comparison of preoperative and postoperative data in 11 eyes with corneal wavefront-guided LASIK. Before LASIK Pt/Eye
Dx
1/L 2/L 3/R 4/L 5/L 6/R 7/R 8/R 9/R 10/R 10/L
Dys Kc Kc Kc Dys Kc Kc Kc Kc Kc Kc
After LASIK
MRSE Coma Trefoil SA Total RMS Coma Trefoil SA Total RMS (D) UDVA CDVA (mm) (mm) (mm) (mm) MRSE UDVA CDVA (mm) (mm) (mm) (mm) 0.50 1.50 3.00 1.25 5.00 5.00 5.00 0.50 3.75 3.75 3.75
20/200 20/100 20/200 20/200 20/400 20/100 20/100 20/50 20/200 20/70 20/50
20/70 20/40 20/50 20/70 20/70 20/50 20/30 20/25 20/50 20/30 20/30
0.95 0.49 0.61 0.48 0.81 0.8 1.01 1.2 0.68 0.87 0.61
1.1 1.2 1.5 1.4 1.2 1.3 0.4 0.7 1.6 2.2 0.5
0.22 0.77 0.23 0.1 0.41 0.45 0.61 0.53 0.01 0.57 0.4
3.39 5.82 4.87 3.66 4.87 4.64 5.94 2.26 5.31 5.7 4.7
0.38 0.63 2.75 1.00 1.50 0.50 0.25 1.50 1.63 0.75 0.5
20/32 20/100 20/32 20/30 20/400 20/30 20/20 20/50 20/25 20/50 20/70
20/25 20/20 20/20 20/30 20/32 20/25 20/20 20/25 20/20 20/20 20/20
0.57 0.53 0.23 0.59 0.47 0.04 0.79 0.58 0.19 0.4 0.39
0.41 1.2 0.76 1.29 0.93 0.76 0.68 0.69 1.32 1.34 0.7
0.18 0.94 0.23 0.22 0.4 0.28 0.53 0.32 0.11 0.64 0.55
1.63 5.33 1.96 3.33 2.54 1.64 1.64 1.22 2.46 4.2 3.87
CDVA Z corrected distance visual acuity (Snellen); Dx Z diagnosis; Dys Z corneal dystrophy; Kc Z keratoconus; LASIK Z laser in situ keratomileusis; MRSE Z manifest refraction spherical equivalent; Pt Z patient; RMS Z root mean square; SA Z spherical aberration; UDVA Z uncorrected distance visual acuity (Snellen)
660 mm. The 160 mm flap was lifted, and a 68.8 mm ablation was performed in a 5.5 mm optical zone. The total HOA RMS decreased from 5.31 mm preoperatively to 2.46 mm postoperatively. Postoperatively, the UDVA in the eye improved to 0.8 and the CDVA to 1.0 with 0.25 2.75 manifest refraction. Table 4 shows the patients' individual preoperative and postoperative data. DISCUSSION Irregular astigmatism, which is almost always present after keratoplasty, translates into very high levels of HOAs that are difficult to quantify with total wavefront analysis. Aberrations in eyes after keratoplasty might even exceed the capabilities of currently available wavefront analyzers or lead to faulty measurements.9 Limited data are available in the current literature on quantification of wavefront aberrations in eyes with corneal grafts. Although various studies have used different devices for wavefront measurements, the common finding is the presence of high levels of HOAs compared with the levels in normal eyes.1,10 Using a large-dynamic-range Hartmann-Shack wavefront sensor, Pantanelli et al.1 reported 5.5 times more higher-order RMS (2.25 G 0.75 mm) in postkeratoplasty eyes than normal eyes over a 6.0 mm pupil. The most dominant HOA was trefoil followed by coma and spherical aberration. Shah et al.10 compared wavefront measurements in normal, keratoconic, and PKP eyes and reported higher RMS, coma, and spherical aberration in eyes with
keratoconus and PKP. However McLaren et al.11 did not find a correlation between corneal higherorder wavefront errors and visual function in postkeratoplasty eyes. The corneal wavefront is derived from corneal topography and reflects only the aberrations resulting from the anterior corneal surface, excluding the aberrations originating from internal structures. Unlike total wavefront, pupil dilation is not necessary and accommodation does not influence the measurements. Reliable corneal wavefront measurements can be obtained in eyes with corneal irregularity and high astigmatism. In normal eyes, the cornea is the main source of the ocular aberrations that degrade the retinal image. With the presence of a graft and surgically induced changes, the cornea becomes a much more significant source of aberrations in postkeratoplasty eyes. Corneal wavefront–guided treatment is well suited to eyes with highly irregular corneas because the aberrations are corrected exactly where they originate. In our study, although significant improvements were seen in RMS and coma values after corneal wavefront–guided LASIK, the slight improvement achieved in trefoil values did not reach statistical significance and spherical aberration remained essentially unchanged. The wavefront errors located close to the center of the Zernike pyramid (coma and spherical aberration) have a more adverse effect on visual acuity than errors located near the edge (trefoil and tetrafoil).12 We speculate that the reduction in coma might have been the reason for the significant improvement in CDVA in all study eyes. In our study, we used the Optimized Refractive Keratectomy-
J CATARACT REFRACT SURG - VOL 40, MAY 2014
790
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
Custom Ablation Manager system. By treating corneal HOAs, the system has proved effective in improving the sight of patients who have different corneal alterations.13 In our study, corneal wavefront-guided LASIK effectively reduced spherical and astigmatic refractive errors and significantly improved UDVA. There was a gain of 1 to 8 lines of UDVA in 63% of the eyes. In addition, there was a statistically significant improvement in the postoperative mean CDVA over preoperative levels. Two previous studies5,14 report a loss of CDVA of 2 lines or more in 6.5% of eyes and 15.0% of eyes, respectively, when LASIK was performed to correct refractive errors after keratoplasty. There was no loss of CDVA lines in our series; the CDVA improved in all eyes. In 64% of the eyes in our series, the achieved correction postoperatively was within G1.00 D of the intended correction. Similar to our results, Barraquer and Rodriquez-Barraquer5 found 67% of cases and Ali o et al.14 72% cases were within G1.00 D of emmetropia after LASIK in eyes with corneal grafts. We reduced the ablation zone diameter when the theoretical ablation depth exceeded the thickness of the corneal lenticule available for ablation. The ablation zone diameters in our patients ranged from 4.0 mm to 6.0 mm to 9.0 mm. Although ablation diameters were often smaller than the patients' scotopic pupil size, no patients reported optical side effects. Pupil-related optical aberrations might have been overshadowed by the visual degradation caused by the irregularities in the periphery of the corneal grafts. Formation of fluid-filled cysts in the interface, poor flap adherence, and postoperative flap displacement have been reported after LASIK in grafts with low endothelial cell function.4,15 Although LASIK has no significant effect on corneal endothelial density in the long term,16 transplanted corneal grafts are known to lose endothelial cells at a faster rate than normal eyes.17 To avoid potential flap complications, we did not perform LASIK in grafts with an ECC less than 1500/mm2. Moreover, expected survival of the grafts with limited endothelial reserve are short and regrafting will most likely be necessary; this might be complicated by the changes caused by LASIK in the host corneal rim. Correction of refractive errors after keratoplasty with surface laser ablation dates back to the early 1990s.18 However, haze formation is common and may be accompanied by regression in eyes with corneal grafts.3 The risk for haze formation is reportedly correlated with the ablation depth, as in normal corneas.5 High spherical and cylindrical refractive errors frequently present after keratoplasty require
deep ablations, increasing the risk for haze formation. Intraoperative mitomycin-C has been used to decrease the risk for haze formation after surface ablation in postkeratoplasty eyes.19,20 Although haze formation is usually not a concern with LASIK, Malecha and Holland21 report stromal haze formation leading to a loss of 7 lines of CDVA in a postkeratoplasty eye; the cause was persistent diffuse lamellar keratitis (DLK). However, the patient had systemic lupus erythematosus and was on systemic immunosuppressive therapy. Although DLK was encountered in 3 of the 20 eyes in the same study, DLK is not a common complication of LASIK after keratoplasty. In our series, there were no cases of DLK or interface haze formation. In addition to the risk for flap-related and graft-related complications, another concern about LASIK in postkeratoplasty eyes is the extension of the microkeratome cut into the host cornea, which could lead to problems should regrafting be necessary. In a study reporting the results of LASIK after PKP, Barraquer and Rodriquez-Barraquer5 analyzed 30 eyes that had keratoplasty for keratoconus and found refractive results to be stable 5 years postoperatively. All but 2 eyes in our series had PKP for keratoconus. We have aimed to leave a postoperative stromal bed thickness of 50% of the preoperative pachymetric reading in all eyes. We did not encounter ectatic changes during our relatively short follow-up; however, a much longer follow-up should be performed to assess the safety of LASIK after PKP. Performing LASIK as a 2-step procedure after keratoplasty has been advocated by some because cutting a lamellar flap may reduce astigmatism in the graft.22 In this study, our preference was to perform LASIK as a 1-step procedure because previous studies failed to show significantly better visual results with the 2-step approach. We also had concerns about potential flap and graft complications, especially the epithelial ingrowth that could result from relifting of the flap. All complications associated with LASIK in normal eyes can also occur with LASIK after keratoplasty. Reported LASIK complications in eyes after keratoplasty include epithelial ingrowth, sterile interface inflammation, flap striae, recurrent herpes simplex keratitis, flap displacement, flap edema and retraction, corneal edema, punctate keratitis,4,15,21,23 graft rejection,24,25 stromal haze formation secondary to persistent DLK,20 buttonhole flap formation,20,26 and formation of fluid-filled pockets in the interface.4 As one of the most dreaded complications of keratoplasty, graft dehiscence can theoretically occur with LASIK in postkeratoplasty eyes
J CATARACT REFRACT SURG - VOL 40, MAY 2014
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
secondary to increased intraocular pressure during suction. Although dehiscence of the graft–host junction has been reported in an eye that had LASIK 3 years after PKP, the dehiscence did not occur during LASIK but was caused by moderate eye rubbing 2 weeks after surgery.27 Most authors advocate waiting 3 to 6 months after the last suture removal or any other refractive procedure before performing LASIK in postkeratoplasty eyes to allow time to achieve adequate wound strength and refractive stability.26 We did not encounter significant flap- or graftrelated complications other than limited epithelial ingrowth in 2 eyes (18%); the ingrowth did not require removal. Regarding corneal aberration, no different behavior was observed between a femtosecond laser and a microkeratome for flap creation.28 The pendular microkeratome functioned without problems in all eyes. The incidence of epithelial ingrowth after LASIK in postkeratoplasty eyes has been reported between 11% and 16%.4,25,26 To avoid the risk for wound dehiscence, we waited for at least 12 months after all sutures had been removed and LASIK was performed only in eyes with a visible fibrotic scar at the graft–host interface preoperatively. During surgery, we also tried to keep the suction time at a minimum. This study is limited by the small sample and lack of follow-up data for all eyes at regular postoperative intervals. However, because the shortest time after LASIK was 9 months in our study, the refractive results would likely have stabilized by that time. Although refractive surgery in normal eyes aims to reduce the patient's dependency on corrective lenses, refractive surgical procedures after PKP mainly aim to resolve anisometropia, thus making spectacle correction tolerable to patients. Insufficient stromal bed thickness did not allow full correction of refractive errors and HOAs in most eyes in our study. However, significant decreases in total RMS, coma, myopia, and astigmatism resulted in significant improvement in UDVA and CDVA. Although visual results were good and patient satisfaction high with this method, the risk for long-term corneal biomechanical problems should be kept in mind. To minimize this risk, creating a LASIK flap that does not extend onto the host cornea with the use of a femtosecond laser instead of a mechanical keratome could be considered. Use of a femtosecond laser might decrease the risk for wound dehiscence as well. Our limited experience in this study indicates that corneal wavefront–guided LASIK is a surgical option for the correction of refractive errors and HOAs after keratoplasty.
791
WHAT WAS KNOWN Postoperative HOAs and refractive error after PKP are common findings. Although LASIK and PRK after PKP to correct refractive error have been reported, wavefront-guided LASIK to correct HOAs after PKP has not been studied. WHAT THIS PAPER ADDS Correcting both HOAs and refractive error with corneal wavefront–guided custom LASIK after PKP improved UDVA and CDVA without significant surgical complications, although the attempted astigmatic correction was not totally achieved. This technique did not result in significant refractive regression in any eye during the long-term follow-up.
REFERENCES 1. Pantanelli S, MacRae S, Jeong TM, Yoon G. Characterizing the wave aberration in eyes with keratoconus or penetrating keratoplasty using a high-dynamic range wavefront sensor. Ophthalmology 2007; 114:2013–2021 2. Troutman RC, Swinger C. Relaxing incision for control of postoperative astigmatism following keratoplasty. Ophthalmic Surg 1980; 11:117–120 € lu B. Photorefrac3. Bilgihan K, Ozdek S‚C, Akata F, Hasanreisog tive keratectomy for post-penetrating keratoplasty myopia and astigmatism. J Cataract Refract Surg 2000; 26:1590–1595 4. Hardten DR, Chittcharus A, Lindstrom RL. Long term analysis of LASIK for the correction of refractive errors after penetrating keratoplasty. Cornea 2004; 23:479–489. Also in: Trans Am Ophthalmol Soc 2002; 100:143 150; discussion 150 152. Available at: http://www.aosonline.org/xactions/2002/15456110_v100_p143.pdf. Accessed February 1, 2014 5. Barraquer CC, Rodriquez-Barraquer T. Five-year results of laser in-situ keratomileusis (LASIK) after penetrating keratoplasty. Cornea 2004; 23:243–248 6. Serdarevic ON. Refractive corneal transplantation: control of astigmatism and ametropia during penetrating keratoplasty. Int Ophthalmol Clin 1994; 34(4):13–33 7. Muftuoglu O, Prasher P, Bowman RW, McCulley JP, Mootha VV. Corneal higher-order aberrations after Descemet’s stripping automated endothelial keratoplasty. Ophthalmology 2010; 117:878–884.e6 8. Alpins N. Astigmatism analysis by the Alpins method. J Cataract Refract Surg 2001; 27:31–49 9. Chalita MR, Krueger RR. Shack-Hartmann aberrometry: historical principles and clinical applications. In: Krueger RR, Applegate RA, MacRae S, eds, Wavefront Customized Visual Correction; the Quest for Super Vision II. Thorofare, NJ, Slack, 2004; 127–130 10. Shah S, Naroo S, Hosking S, Gherghel D, Mantry S, Bannerjee S, Pedwell K, Bains HS. Nidek OPD-scan analysis of normal, keratoconic, and penetrating keratoplasty eyes. J Refract Surg 2003; 19:S255–S259 11. McLaren JW, Patel SV, Bourne WM, Baratz KH. Corneal wavefront errors 24 months after deep lamellar endothelial
J CATARACT REFRACT SURG - VOL 40, MAY 2014
792
12. 13.
14.
15.
16.
17.
18.
19.
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
keratoplasty and penetrating keratoplasty. Am J Ophthalmol 2009; 147:959–965 Applegate RA, Sarver EJ, Khemsara V. Are all aberrations equal? J Refract Surg 2002; 18:S556–S562 J, Galal A, Montalba n R, Pin ~ero D. Corneal wavefrontAlio guided LASIK retreatments for correction of highly aberrated corneas following refractive surgery. J Refract Surg 2007; 23:760–773 JL, Javaloy J, Osman AA, Galvis V, Tello A, Haroun HE. Alio Laser in situ keratomileusis to correct post-keratoplasty astigmatism: 1-step versus 2-step procedure. J Cataract Refract Surg 2004; 30:2303–2310 Solomon R, Donnenfeld ED, Perry HD, Nirankari VS. Post-LASIK corneal flap displacement following penetrating keratoplasty for bullous keratopathy. Cornea 2005; 24:874–878 Patel SV, Bourne WM. Corneal endothelial cell loss 9 years after excimer laser keratorefractive surgery. Arch Ophthalmol 2009; 127:1423–1427. Available at: http://archopht.jamanetwork. com/data/Journals/OPHTH/10132/ecs90022_1423_1427.pdf. Accessed February 1, 2014 Bourne WM, Hodge DO, Nelson LR. Corneal endothelium five years after transplantation. Am J Ophthalmol 1994; 118:185–196 Campos M, Hertzog L, Garbus J, Lee M, McDonnell PJ. Photorefractive keratectomy for severe postkeratoplasty astigmatism. Am J Ophthalmol 1992; 114:429–436 Rajan MS, O’Brart DP, Patel P, Falcon MG, Marshall J. Topography-guided customized laser-assisted subepithelial keratectomy for the treatment of postkeratoplasty astigmatism. J Cataract Refract Surg 2006; 32:949–957
20. Solomon R, Donnenfeld ED, Perry HD. Photorefractive keratectomy with mitomycin C for the management of a LASIK flap complication following a penetrating keratoplasty. Cornea 2004; 23:403–405 21. Malecha MA, Holland EJ. Correction of myopia and astigmatism after penetrating keratoplasty with laser in situ keratomileusis. Cornea 2002; 21:564–569 22. Dada T, Vajpayee RB, Gupta V, Sharma N, Dada VK. Microkeratome-induced reduction of astigmatism after penetrating keratoplasty. Am J Ophthalmol 2001; 131:507–508 23. Chan CC, Rootman DS. Corneal lamellar flap retraction after LASIK following penetrating keratoplasty. Cornea 2004; 23:643–646 24. Hersh PS, Jordan AJ, Mayers M. Corneal graft rejection episode after excimer laser phototherapeutic keratectomy. Arch Ophthalmol 1993; 111:735–736 25. Kovoor TA, Mohamed E, Cavanagh HD, Bowman RW. Outcomes of LASIK and PRK in previous penetrating corneal transplant recipients. Eye Contact Lens 2009; 35:242–245 26. Kwitko S, Marinho DR, Rymer S, Ramos Filho S. Laser in situ keratomileusis after penetrating keratoplasty. J Cataract Refract Surg 2001; 27:374–379 27. Ranchod TM, McLeod SD. Wound dehiscence in a patient with keratoconus after penetrating keratoplasty and LASIK. Arch Ophthalmol 2004; 122:920–921. Available at: http://archopht. jamanetwork.com/data/Journals/OPHTH/9925/ecr0604-4-1.pdf. Accessed February 1, 2014 JL, Pin ~ ero D. Very high-frequency digital ultrasound mea28. Alio surement of the LASIK flap thickness profile using the IntraLase femtosecond laser and M2 and Carriazo-Pendular microkeratomes. J Refract Surg 2008; 24:12–23
J CATARACT REFRACT SURG - VOL 40, MAY 2014
Corneal wavefront–guided customized laser in situ keratomileusis after penetrating keratoplasty Serhat Imamoglu, MD, Vedat Kaya, MD, Deniz Oral, MD, Irfan Perente, MD, Berna Basarir, MD, Omer Faruk Yilmaz, MD
PURPOSE: To assess the efficacy and safety of corneal wavefront-guided custom laser in situ keratomileusis (LASIK) to correct refractive errors and higher-order aberrations (HOAs) after penetrating keratoplasty (PKP). SETTING: Beyoglu Eye Training and Research Hospital, Istanbul, Turkey. DESIGN: Noncomparative case series. METHODS: The study comprised consecutive patients who were unable to tolerate spectacles or contact lenses for the correction of anisometropia after PKP and had corneal wavefront-guided custom LASIK. The uncorrected (UDVA) and corrected (CDVA) distance visual acuities, manifest refraction spherical equivalent (MRSE), and corneal HOAs were recorded before and after LASIK. RESULTS: The study evaluated 11 eyes (10 patients). The mean follow-up after LASIK was 24.3 months G 11.7 (SD) (range 9 to 36 months). The mean MRSE was 2.97 G 1.66 diopters (D) (range 0.50 to 5.38 D) preoperatively and 0.88 G 0.96 D (range 2.75 to 0.5 D) postoperatively. The mean total higher-order root mean square (RMS) was 4.65 G 1.14 mm (range 2.26 to 5.94 mm) preoperatively and 2.71 G 1.31 mm (range 1.22 to 5.33 mm) postoperatively. Postoperatively, the UDVA improved in 7 eyes and remained unchanged in 4 eyes. The CDVA improved in all eyes postoperatively even cases in which the attempted astigmatic correction was not totally achieved. CONCLUSIONS: Corneal wavefront-guided custom LASIK after PKP did not totally correct both refractive errors and HOAs because of the high volume of laser ablation required and inadequate corneal stromal thickness. Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2014; 40:785–792 Q 2014 ASCRS and ESCRS
Visual outcomes after penetrating keratoplasty (PKP) are unpredictable because the surgery often induces significant astigmatism. Astigmatism after keratoplasty is typically irregular. In addition, high levels of spherical refractive errors and higher-order aberrations (HOAs) are often present.1 On the other hand, patients with good corrected distance visual acuity (CDVA) often have problems related to anisometropia. When irregular astigmatism or anisometropia is present, spectacles provide limited visual rehabilitation while rigid gas-permeable (RGP) contact lenses typically give better results. However, the abnormal corneal shape and eyelid problems that are often present in eyes after PKP usually limit the use of contact lenses and prompt patients to seek surgical treatment. Q 2014 ASCRS and ESCRS Published by Elsevier Inc.
Until excimer lasers became available, incisional refractive surgical procedures, such as astigmatic keratotomy and wedge resection, were used with limited success to correct astigmatism after keratoplasty.2 Photorefractive keratectomy (PRK) has been used successfully to correct spherical and cylindrical refractive errors; however, haze formation and loss of CDVA are frequent postoperative complications.3 Although laser in situ keratomileusis (LASIK) brings the added risk for flap-related and wound-related complications in eyes with corneal grafts, it has largely replaced PRK for the correction of refractive errors after keratoplasty.4,5 Although standard excimer laser ablation patterns can address spherocylindrical refractive errors only, 0886-3350/$ - see front matter http://dx.doi.org/10.1016/j.jcrs.2013.10.042
785
786
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
the advent of topography-guided ablation systems made it possible to perform custom treatments and correct irregular astigmatism. The differences in curvature, thickness, and size between the graft and the recipient cornea might lead to misalignment at the graft–host interface. Also, variable tension of the sutures induces irregular astigmatism by increasing HOAs after PKP.6 After keratoplasty, eyes have nearly 5.5 times more HOAs than normal eyes. Trefoil is reportedly the most dominant HOA.1 By addressing both lower-order aberrations (LOAs) and HOAs, wavefront-guided LASIK has the potential to effectively correct irregular astigmatism and maximize visual improvement after keratoplasty. The corneal wavefront is derived from corneal topography. Corneal wavefront–guided treatment is basically advanced topography-guided ablation. The technique gives the surgeon the ability to selectively correct corneal HOAs along with refractive errors. After PKP, anterior and posterior corneal HOAs are reported to be significantly higher than in normal eyes.7 In this study, we assessed the results of corneal wavefront–guided custom LASIK for the correction of LOAs and HOAs after PKP.
Surgical Planning On the day of the surgery, corneal topography maps were taken and the best map was chosen with respect to reliability indices of the maps. The Placido-disk topography unit used in this study provides corneal elevation data using the arcstep method. The device compares the data obtained from the corneal elevation map with the reference ideal eye model with a Q-value of 0.25. The elevation difference between the 2 surfaces creates an optical path difference, which is expressed by Zernike polynomials (up to the 7th order) to represent the corneal wavefront. The Optimized Refractive Keratectomy-Custom Ablation Manager software was used to plan the treatment with the Schwind Esiris excimer laser platform (both Schwind eyetech-solutions GmbH and Co. KG). This platform uses a 0.8 mm para-Gaussian profile flying-spot laser at 200 Hz and a 330 Hz eye tracker. The software uses topographic data (obtained with the Placido-disk topography device) and other clinical parameters (manifest refraction, pachymetry, and flap thickness) to determine the most appropriate custom ablation profile to correct LOAs and HOAs. Using the software, the surgeon can change the optical zone diameter and select the aberrations to be treated. The ablation zone diameter was determined according to the scotopic pupil size, and full correction of LOAs and HOAs were targeted in all eyes. The ablations were planned to leave a minimum postoperative stromal bed thickness of 250 mm or 50% of the preoperative thickness. The aim of the treatment was to decrease HOAs and anisometropia primarily.
PATIENTS AND METHODS This retrospective noncomparative clinical case series comprised consecutive patients who had corneal wavefront–guided custom LASIK between December 2005 and February 2008 to correct refractive errors and HOAs after PKP. All patients were unable to tolerate spectacle correction. Rigid gas-permeable contact lens correction was attempted in 4 eyes without success; the other patients refused to try the lenses. The study inclusion criteria were a clear full-thickness corneal graft with an endothelial cell count (ECC) above 1500 cells/mm2, a stable refractive error for at least 12 months after all sutures had been removed, and the absence of ocular pathology other than the indication for PKP. All patients received a detailed explanation about wavefront-guided LASIK, including potential risks and benefits, after which they provided informed consent. The same surgeon (V.K.) performed all LASIK and PKP procedures.
Submitted: August 4, 2012. Final revision submitted: October 5, 2013. Accepted: October 14, 2013. From the Eye Clinic (Imamoglu), Haydarpasa Numune Education and Research Hospital, a private practice (Kaya), the Department of Ophthalmology (Oral), Faculty of Medicine, Yeditepe University, and Beyoglu Eye Training and Research Hospital (Perente, Basarir, Yilmaz), Istanbul, Turkey. Corresponding author: Serhat Imamoglu, MD, Tıbbiye Caddesi € €udar, Istanbul, _ Number: 40, Usk Turkey. E-mail: ophserhat@ hotmail.com.
Surgical Technique In all eyes, a Carriazo-Pendular microkeratome (Schwind eye-tech-solutions GmbH and Co. KG) was used to create a superior, hinged 160 mm thickness after the flap had been lifted. After excimer laser ablation, the flap was placed back on the stromal bed and a plano soft contact lens was placed over the cornea until the first postoperative visit. Prednisolone acetate 1% and tobramycin 0.3% eyedrops 4 times a day and preservative-free artificial tears every hour were prescribed. Tobramycin drops were used for 10 days, and the prednisolone acetate drops tapered over 6 weeks.
Patient Evaluation The uncorrected distance visual acuity (UDVA), manifest refraction, CDVA, and corneal wavefront aberrations as well as subjective complaints were evaluated preoperatively and at the last postoperative visit. The preoperative workup also included tonometry, slitlamp biomicroscopy, fundus examination, pupillometry (Colvard pupillometer, Oasis Medical, Inc.), scanning-slit topography (Orbscan IIz, Orbtek, Inc.), and Placido-disk topography (Keratron Scout, Optikon 2000 SpA). Ultrasonic pachymetry (Tomey AL-3000, Tomey Corp.) measurements were taken at 9 points over the corneal graft including the center and 4 points in 3.0 mm and 7.0 mm zones. Endothelial cell counts were performed with a noncontact specular microscope (Noncon Robo FA-3509, Konan Medical). The change in astigmatism achieved at the last follow-up was evaluated using Alpins vectorial analysis.8 The following vectors were determined and evaluated: targetinduced astigmatism (TIA) as the vector of the intended
J CATARACT REFRACT SURG - VOL 40, MAY 2014
787
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
Table 1. Patient characteristics. Parameter Eyes (n) Age (y) Mean Range Sex (n) Male Female Transplant indication (n) Keratoconus Corneal dystrophy
Table 2. Visual and refractive outcomes. Value 11 32.3 22, 53 6 4 8 2
change in cylinder for each treatment and surgically induced astigmatism (SIA) as the vector of the actual change achieved.
Statistical Analysis Preoperative and postoperative data were compared by using paired t test, and P values less than 0.05 were considered statistically significant. Snellen visual acuity scores were used for statistical analysis.
Parameter UDVA (Snellen lines) Mean G SD Range Cylinder (D) Mean G SD Range MRSE (D) Mean G SD Range CDVA (Snellen lines) Mean G SD Range
Preoperative Postoperative P Value 0.19 G 0.12 0.05, 0.40
0.52 G 0.27 0.05, 0.80
.002
5.20 G 1.89 1.00, 7.50
3.11 G 1.61 0.00, 5.00
.011
2.97 G 1.66 0.50, 5.38
0.88 G 0.96 2.75, 0.50
.002
0.50 G 0.18 0.30, 0.80
0.90 G 0.15 0.60, 1.00
!.001
CDVA Z corrected distance visual acuity; MRSE Z manifest refraction spherical equivalent; UDVA Z uncorrected distance visual acuity
(range 518 to 660 mm). The mean laser ablation depth was 71 G 11 mm (range 47.9 to 85.3 mm), and the mean postoperative theoretical RSB thickness was 320 G 43 mm (range 279 to 431.2 mm). Visual Acuity and Refraction
RESULTS The study included 11 eyes of 10 patients. Table 1 shows the patients' demographics. The mean interval from PKP to LASIK was 66.8 months G 30.2 (SD) (range 23 to 96 months). One patient had bilateral treatment. The mean follow-up after LASIK was 42.9 G 21.8 months (range 10 to 72 months). In 6 of the 11 eyes, the theoretical ablation depth exceeded the thickness of corneal lenticule available for ablation due to a high amount of aberrations. In these eyes, the ablation zone diameter was decreased or undercorrection was targeted to maintain the minimum residual stromal bed (RSB) thickness. The mean preoperative graft thickness was 552 G 42 mm
Table 2 shows the visual and refractive outcomes. Preoperatively, all eyes had a UDVA of worse than 20/40 and 5 (45%) eyes had a UDVA of 20/200 or worse. After LASIK, 1 (9%) eye had a UDVA of worse than 20/200 and 7 (63%) eyes had a UDVA of 20/40 or better (Figure 1). The mean postoperative UDVA was statistically significantly better than the mean preoperative UDVA (PZ.002). After LASIK, the UDVA improved in 7 (63%) eyes and remained unchanged in 4 eyes (36%). One eye (9%) gained 1 line, 1 eye (9%) gained 4 lines, 2 eyes (18%) gained 5 lines, 1 eye (9%) gained 6 lines, 1 eye (9%) gained 7 lines, and 1 eye (9%) gained 8 lines of UDVA. There was also a statistically significant decrease in the mean manifest refraction spherical equivalent (MRSE) (PZ.002) and in the mean refractive astigmatism (PZ.011) between preoperatively and postoperatively. The attempted astigmatic correction was not totally achieved in all cases (Table 3). The decrease in the mean refractive astigmatism was statistically significant. Predictability
Figure 1. Percentage of eyes with a certain level of cumulative preoperative Snellen CDVA and of postoperative UDVA at last follow-up (CDVA Z corrected distance visual acuity; UDVA Z uncorrected distance visual acuity).
The postoperative achieved correction was within G1.00 D of the intended correction in 7 eyes (64%). The mean magnitude of the TIA was 5.73 G 2.52 D (range 1.01 to 8.24 D), and the mean magnitude of the SIA at last follow-up was 3.96 G 3.32 D (range 0.07 to 11.1 D). The difference between the TIA and
J CATARACT REFRACT SURG - VOL 40, MAY 2014
788
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
Table 3. Values of refractive astigmatism in each eye. Astigmatism (D) Eye 1 2 3 4 5 6 7 8 9 10 11
Preoperative
Postoperative
Achieved Change
6 120 5 131 6.5 100 5.5 156 7.5 96 4 28 4 25 1 82 4 18 7.5 150 7.5 34
3.25 180 3.75 140 3.00 90 5.00 155 5.00 150 2.00 25 0.00 90 1.00 80 2.75 5 4.00 165 4.50 25
2.75 1.25 3.50 0.50 3.00 2.00 4.00 0.00 1.25 3.50 3.00
the SIA were statistically significant (P!.05). Therefore, there was a trend toward undercorrection of the refractive astigmatism after wavefront-guided LASIK for irregular astigmatism (Figure 2). In an ideal complete correction, the SIA and the TIA would be identical.
Higher-Order Aberrations Figure 3 shows the changes in the mean corneal HOAs. The mean preoperative total root mean square (RMS) was 4.65 G 1.14 mm (range 2.26 to 5.94 mm). The postoperative mean total RMS decreased to 2.71 G 1.31 mm (range 1.22 to 5.33 mm), showing a statistically significant improvement (PZ.001). The mean preoperative coma was 0.77 G 0.23 mm (range 0.48 to 1.20 mm). The postoperative mean coma was 0.43 G 0.21 mm (range 0.04 to 0.79 mm). This represented a statistically significant reduction (PZ.002). The mean preoperative trefoil was 1.18 G 0.52 mm (range 0.36 to 2.19 mm). The postoperative mean trefoil was 0.92 G 0.32 mm (range 0.41 to 1.34 mm). The change was not statistically significant (PO.05). The mean preoperative spherical aberration was 0.39 G 0.23 mm (range 0.01 to 0.77 mm). The postoperative mean spherical aberration was 0.40 G 0.25 mm (range 0.11 to 0.94 mm). The change was not statistically significant (PO.05). Complications and Enhancement
Safety Postoperatively, no eye lost CDVA lines. All eyes had a CDVA better than 20/40, and 9 eyes (82%) had a CDVA of 20/25 or better. After corneal wavefront-guided LASIK, the CDVA improved in all eyes. Two eyes (18%) gained 1 line, 4 eyes (36%) gained 3 lines, and 5 eyes (45%) gained 6 lines of CDVA. There was a statistically significant improvement in the mean CDVA between preoperatively and postoperatively (P!.001). No eye had corneal ectasia or recurrent keratoconus on topographic examination.
Figure 2. Relationship between the achieved postoperative spherical equivalent (SE) correction and the attempted SE correction.
Limited epithelial ingrowth that did not require intervention was observed at the flap border in 2 eyes (18%). No enhancements were performed. Case Report A 25-year-old man had PKP in the right eye for keratoconus 7 years previously. The patient presented with contact lens intolerance and 4.00 D of mixed astigmatism. He reported reduced vision, halos, glare, and sunburst at night. The preoperative UDVA was 0.1, and the manifest refraction was 1.75 4.00 with a CDVA of 0.4. Corneal wavefront-guided custom LASIK was performed to correct both HOA-related symptoms and anisometropia. The preoperative corneal pachymetry of the graft was
Figure 3. Higher-order corneal aberrations in eyes that had corneal wavefront-guided custom LASIK after PKP (* Z P!.05; HOA Z higher-order aberrations; SA Z spherical aberration).
J CATARACT REFRACT SURG - VOL 40, MAY 2014
789
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
Table 4. Comparison of preoperative and postoperative data in 11 eyes with corneal wavefront-guided LASIK. Before LASIK Pt/Eye
Dx
1/L 2/L 3/R 4/L 5/L 6/R 7/R 8/R 9/R 10/R 10/L
Dys Kc Kc Kc Dys Kc Kc Kc Kc Kc Kc
After LASIK
MRSE Coma Trefoil SA Total RMS Coma Trefoil SA Total RMS (D) UDVA CDVA (mm) (mm) (mm) (mm) MRSE UDVA CDVA (mm) (mm) (mm) (mm) 0.50 1.50 3.00 1.25 5.00 5.00 5.00 0.50 3.75 3.75 3.75
20/200 20/100 20/200 20/200 20/400 20/100 20/100 20/50 20/200 20/70 20/50
20/70 20/40 20/50 20/70 20/70 20/50 20/30 20/25 20/50 20/30 20/30
0.95 0.49 0.61 0.48 0.81 0.8 1.01 1.2 0.68 0.87 0.61
1.1 1.2 1.5 1.4 1.2 1.3 0.4 0.7 1.6 2.2 0.5
0.22 0.77 0.23 0.1 0.41 0.45 0.61 0.53 0.01 0.57 0.4
3.39 5.82 4.87 3.66 4.87 4.64 5.94 2.26 5.31 5.7 4.7
0.38 0.63 2.75 1.00 1.50 0.50 0.25 1.50 1.63 0.75 0.5
20/32 20/100 20/32 20/30 20/400 20/30 20/20 20/50 20/25 20/50 20/70
20/25 20/20 20/20 20/30 20/32 20/25 20/20 20/25 20/20 20/20 20/20
0.57 0.53 0.23 0.59 0.47 0.04 0.79 0.58 0.19 0.4 0.39
0.41 1.2 0.76 1.29 0.93 0.76 0.68 0.69 1.32 1.34 0.7
0.18 0.94 0.23 0.22 0.4 0.28 0.53 0.32 0.11 0.64 0.55
1.63 5.33 1.96 3.33 2.54 1.64 1.64 1.22 2.46 4.2 3.87
CDVA Z corrected distance visual acuity (Snellen); Dx Z diagnosis; Dys Z corneal dystrophy; Kc Z keratoconus; LASIK Z laser in situ keratomileusis; MRSE Z manifest refraction spherical equivalent; Pt Z patient; RMS Z root mean square; SA Z spherical aberration; UDVA Z uncorrected distance visual acuity (Snellen)
660 mm. The 160 mm flap was lifted, and a 68.8 mm ablation was performed in a 5.5 mm optical zone. The total HOA RMS decreased from 5.31 mm preoperatively to 2.46 mm postoperatively. Postoperatively, the UDVA in the eye improved to 0.8 and the CDVA to 1.0 with 0.25 2.75 manifest refraction. Table 4 shows the patients' individual preoperative and postoperative data. DISCUSSION Irregular astigmatism, which is almost always present after keratoplasty, translates into very high levels of HOAs that are difficult to quantify with total wavefront analysis. Aberrations in eyes after keratoplasty might even exceed the capabilities of currently available wavefront analyzers or lead to faulty measurements.9 Limited data are available in the current literature on quantification of wavefront aberrations in eyes with corneal grafts. Although various studies have used different devices for wavefront measurements, the common finding is the presence of high levels of HOAs compared with the levels in normal eyes.1,10 Using a large-dynamic-range Hartmann-Shack wavefront sensor, Pantanelli et al.1 reported 5.5 times more higher-order RMS (2.25 G 0.75 mm) in postkeratoplasty eyes than normal eyes over a 6.0 mm pupil. The most dominant HOA was trefoil followed by coma and spherical aberration. Shah et al.10 compared wavefront measurements in normal, keratoconic, and PKP eyes and reported higher RMS, coma, and spherical aberration in eyes with
keratoconus and PKP. However McLaren et al.11 did not find a correlation between corneal higherorder wavefront errors and visual function in postkeratoplasty eyes. The corneal wavefront is derived from corneal topography and reflects only the aberrations resulting from the anterior corneal surface, excluding the aberrations originating from internal structures. Unlike total wavefront, pupil dilation is not necessary and accommodation does not influence the measurements. Reliable corneal wavefront measurements can be obtained in eyes with corneal irregularity and high astigmatism. In normal eyes, the cornea is the main source of the ocular aberrations that degrade the retinal image. With the presence of a graft and surgically induced changes, the cornea becomes a much more significant source of aberrations in postkeratoplasty eyes. Corneal wavefront–guided treatment is well suited to eyes with highly irregular corneas because the aberrations are corrected exactly where they originate. In our study, although significant improvements were seen in RMS and coma values after corneal wavefront–guided LASIK, the slight improvement achieved in trefoil values did not reach statistical significance and spherical aberration remained essentially unchanged. The wavefront errors located close to the center of the Zernike pyramid (coma and spherical aberration) have a more adverse effect on visual acuity than errors located near the edge (trefoil and tetrafoil).12 We speculate that the reduction in coma might have been the reason for the significant improvement in CDVA in all study eyes. In our study, we used the Optimized Refractive Keratectomy-
J CATARACT REFRACT SURG - VOL 40, MAY 2014
790
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
Custom Ablation Manager system. By treating corneal HOAs, the system has proved effective in improving the sight of patients who have different corneal alterations.13 In our study, corneal wavefront-guided LASIK effectively reduced spherical and astigmatic refractive errors and significantly improved UDVA. There was a gain of 1 to 8 lines of UDVA in 63% of the eyes. In addition, there was a statistically significant improvement in the postoperative mean CDVA over preoperative levels. Two previous studies5,14 report a loss of CDVA of 2 lines or more in 6.5% of eyes and 15.0% of eyes, respectively, when LASIK was performed to correct refractive errors after keratoplasty. There was no loss of CDVA lines in our series; the CDVA improved in all eyes. In 64% of the eyes in our series, the achieved correction postoperatively was within G1.00 D of the intended correction. Similar to our results, Barraquer and Rodriquez-Barraquer5 found 67% of cases and Ali o et al.14 72% cases were within G1.00 D of emmetropia after LASIK in eyes with corneal grafts. We reduced the ablation zone diameter when the theoretical ablation depth exceeded the thickness of the corneal lenticule available for ablation. The ablation zone diameters in our patients ranged from 4.0 mm to 6.0 mm to 9.0 mm. Although ablation diameters were often smaller than the patients' scotopic pupil size, no patients reported optical side effects. Pupil-related optical aberrations might have been overshadowed by the visual degradation caused by the irregularities in the periphery of the corneal grafts. Formation of fluid-filled cysts in the interface, poor flap adherence, and postoperative flap displacement have been reported after LASIK in grafts with low endothelial cell function.4,15 Although LASIK has no significant effect on corneal endothelial density in the long term,16 transplanted corneal grafts are known to lose endothelial cells at a faster rate than normal eyes.17 To avoid potential flap complications, we did not perform LASIK in grafts with an ECC less than 1500/mm2. Moreover, expected survival of the grafts with limited endothelial reserve are short and regrafting will most likely be necessary; this might be complicated by the changes caused by LASIK in the host corneal rim. Correction of refractive errors after keratoplasty with surface laser ablation dates back to the early 1990s.18 However, haze formation is common and may be accompanied by regression in eyes with corneal grafts.3 The risk for haze formation is reportedly correlated with the ablation depth, as in normal corneas.5 High spherical and cylindrical refractive errors frequently present after keratoplasty require
deep ablations, increasing the risk for haze formation. Intraoperative mitomycin-C has been used to decrease the risk for haze formation after surface ablation in postkeratoplasty eyes.19,20 Although haze formation is usually not a concern with LASIK, Malecha and Holland21 report stromal haze formation leading to a loss of 7 lines of CDVA in a postkeratoplasty eye; the cause was persistent diffuse lamellar keratitis (DLK). However, the patient had systemic lupus erythematosus and was on systemic immunosuppressive therapy. Although DLK was encountered in 3 of the 20 eyes in the same study, DLK is not a common complication of LASIK after keratoplasty. In our series, there were no cases of DLK or interface haze formation. In addition to the risk for flap-related and graft-related complications, another concern about LASIK in postkeratoplasty eyes is the extension of the microkeratome cut into the host cornea, which could lead to problems should regrafting be necessary. In a study reporting the results of LASIK after PKP, Barraquer and Rodriquez-Barraquer5 analyzed 30 eyes that had keratoplasty for keratoconus and found refractive results to be stable 5 years postoperatively. All but 2 eyes in our series had PKP for keratoconus. We have aimed to leave a postoperative stromal bed thickness of 50% of the preoperative pachymetric reading in all eyes. We did not encounter ectatic changes during our relatively short follow-up; however, a much longer follow-up should be performed to assess the safety of LASIK after PKP. Performing LASIK as a 2-step procedure after keratoplasty has been advocated by some because cutting a lamellar flap may reduce astigmatism in the graft.22 In this study, our preference was to perform LASIK as a 1-step procedure because previous studies failed to show significantly better visual results with the 2-step approach. We also had concerns about potential flap and graft complications, especially the epithelial ingrowth that could result from relifting of the flap. All complications associated with LASIK in normal eyes can also occur with LASIK after keratoplasty. Reported LASIK complications in eyes after keratoplasty include epithelial ingrowth, sterile interface inflammation, flap striae, recurrent herpes simplex keratitis, flap displacement, flap edema and retraction, corneal edema, punctate keratitis,4,15,21,23 graft rejection,24,25 stromal haze formation secondary to persistent DLK,20 buttonhole flap formation,20,26 and formation of fluid-filled pockets in the interface.4 As one of the most dreaded complications of keratoplasty, graft dehiscence can theoretically occur with LASIK in postkeratoplasty eyes
J CATARACT REFRACT SURG - VOL 40, MAY 2014
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
secondary to increased intraocular pressure during suction. Although dehiscence of the graft–host junction has been reported in an eye that had LASIK 3 years after PKP, the dehiscence did not occur during LASIK but was caused by moderate eye rubbing 2 weeks after surgery.27 Most authors advocate waiting 3 to 6 months after the last suture removal or any other refractive procedure before performing LASIK in postkeratoplasty eyes to allow time to achieve adequate wound strength and refractive stability.26 We did not encounter significant flap- or graftrelated complications other than limited epithelial ingrowth in 2 eyes (18%); the ingrowth did not require removal. Regarding corneal aberration, no different behavior was observed between a femtosecond laser and a microkeratome for flap creation.28 The pendular microkeratome functioned without problems in all eyes. The incidence of epithelial ingrowth after LASIK in postkeratoplasty eyes has been reported between 11% and 16%.4,25,26 To avoid the risk for wound dehiscence, we waited for at least 12 months after all sutures had been removed and LASIK was performed only in eyes with a visible fibrotic scar at the graft–host interface preoperatively. During surgery, we also tried to keep the suction time at a minimum. This study is limited by the small sample and lack of follow-up data for all eyes at regular postoperative intervals. However, because the shortest time after LASIK was 9 months in our study, the refractive results would likely have stabilized by that time. Although refractive surgery in normal eyes aims to reduce the patient's dependency on corrective lenses, refractive surgical procedures after PKP mainly aim to resolve anisometropia, thus making spectacle correction tolerable to patients. Insufficient stromal bed thickness did not allow full correction of refractive errors and HOAs in most eyes in our study. However, significant decreases in total RMS, coma, myopia, and astigmatism resulted in significant improvement in UDVA and CDVA. Although visual results were good and patient satisfaction high with this method, the risk for long-term corneal biomechanical problems should be kept in mind. To minimize this risk, creating a LASIK flap that does not extend onto the host cornea with the use of a femtosecond laser instead of a mechanical keratome could be considered. Use of a femtosecond laser might decrease the risk for wound dehiscence as well. Our limited experience in this study indicates that corneal wavefront–guided LASIK is a surgical option for the correction of refractive errors and HOAs after keratoplasty.
791
WHAT WAS KNOWN Postoperative HOAs and refractive error after PKP are common findings. Although LASIK and PRK after PKP to correct refractive error have been reported, wavefront-guided LASIK to correct HOAs after PKP has not been studied. WHAT THIS PAPER ADDS Correcting both HOAs and refractive error with corneal wavefront–guided custom LASIK after PKP improved UDVA and CDVA without significant surgical complications, although the attempted astigmatic correction was not totally achieved. This technique did not result in significant refractive regression in any eye during the long-term follow-up.
REFERENCES 1. Pantanelli S, MacRae S, Jeong TM, Yoon G. Characterizing the wave aberration in eyes with keratoconus or penetrating keratoplasty using a high-dynamic range wavefront sensor. Ophthalmology 2007; 114:2013–2021 2. Troutman RC, Swinger C. Relaxing incision for control of postoperative astigmatism following keratoplasty. Ophthalmic Surg 1980; 11:117–120 € lu B. Photorefrac3. Bilgihan K, Ozdek S‚C, Akata F, Hasanreisog tive keratectomy for post-penetrating keratoplasty myopia and astigmatism. J Cataract Refract Surg 2000; 26:1590–1595 4. Hardten DR, Chittcharus A, Lindstrom RL. Long term analysis of LASIK for the correction of refractive errors after penetrating keratoplasty. Cornea 2004; 23:479–489. Also in: Trans Am Ophthalmol Soc 2002; 100:143 150; discussion 150 152. Available at: http://www.aosonline.org/xactions/2002/15456110_v100_p143.pdf. Accessed February 1, 2014 5. Barraquer CC, Rodriquez-Barraquer T. Five-year results of laser in-situ keratomileusis (LASIK) after penetrating keratoplasty. Cornea 2004; 23:243–248 6. Serdarevic ON. Refractive corneal transplantation: control of astigmatism and ametropia during penetrating keratoplasty. Int Ophthalmol Clin 1994; 34(4):13–33 7. Muftuoglu O, Prasher P, Bowman RW, McCulley JP, Mootha VV. Corneal higher-order aberrations after Descemet’s stripping automated endothelial keratoplasty. Ophthalmology 2010; 117:878–884.e6 8. Alpins N. Astigmatism analysis by the Alpins method. J Cataract Refract Surg 2001; 27:31–49 9. Chalita MR, Krueger RR. Shack-Hartmann aberrometry: historical principles and clinical applications. In: Krueger RR, Applegate RA, MacRae S, eds, Wavefront Customized Visual Correction; the Quest for Super Vision II. Thorofare, NJ, Slack, 2004; 127–130 10. Shah S, Naroo S, Hosking S, Gherghel D, Mantry S, Bannerjee S, Pedwell K, Bains HS. Nidek OPD-scan analysis of normal, keratoconic, and penetrating keratoplasty eyes. J Refract Surg 2003; 19:S255–S259 11. McLaren JW, Patel SV, Bourne WM, Baratz KH. Corneal wavefront errors 24 months after deep lamellar endothelial
J CATARACT REFRACT SURG - VOL 40, MAY 2014
792
12. 13.
14.
15.
16.
17.
18.
19.
WAVEFRONT-GUIDED LASIK AFTER KERATOPLASTY
keratoplasty and penetrating keratoplasty. Am J Ophthalmol 2009; 147:959–965 Applegate RA, Sarver EJ, Khemsara V. Are all aberrations equal? J Refract Surg 2002; 18:S556–S562 J, Galal A, Montalba n R, Pin ~ero D. Corneal wavefrontAlio guided LASIK retreatments for correction of highly aberrated corneas following refractive surgery. J Refract Surg 2007; 23:760–773 JL, Javaloy J, Osman AA, Galvis V, Tello A, Haroun HE. Alio Laser in situ keratomileusis to correct post-keratoplasty astigmatism: 1-step versus 2-step procedure. J Cataract Refract Surg 2004; 30:2303–2310 Solomon R, Donnenfeld ED, Perry HD, Nirankari VS. Post-LASIK corneal flap displacement following penetrating keratoplasty for bullous keratopathy. Cornea 2005; 24:874–878 Patel SV, Bourne WM. Corneal endothelial cell loss 9 years after excimer laser keratorefractive surgery. Arch Ophthalmol 2009; 127:1423–1427. Available at: http://archopht.jamanetwork. com/data/Journals/OPHTH/10132/ecs90022_1423_1427.pdf. Accessed February 1, 2014 Bourne WM, Hodge DO, Nelson LR. Corneal endothelium five years after transplantation. Am J Ophthalmol 1994; 118:185–196 Campos M, Hertzog L, Garbus J, Lee M, McDonnell PJ. Photorefractive keratectomy for severe postkeratoplasty astigmatism. Am J Ophthalmol 1992; 114:429–436 Rajan MS, O’Brart DP, Patel P, Falcon MG, Marshall J. Topography-guided customized laser-assisted subepithelial keratectomy for the treatment of postkeratoplasty astigmatism. J Cataract Refract Surg 2006; 32:949–957
20. Solomon R, Donnenfeld ED, Perry HD. Photorefractive keratectomy with mitomycin C for the management of a LASIK flap complication following a penetrating keratoplasty. Cornea 2004; 23:403–405 21. Malecha MA, Holland EJ. Correction of myopia and astigmatism after penetrating keratoplasty with laser in situ keratomileusis. Cornea 2002; 21:564–569 22. Dada T, Vajpayee RB, Gupta V, Sharma N, Dada VK. Microkeratome-induced reduction of astigmatism after penetrating keratoplasty. Am J Ophthalmol 2001; 131:507–508 23. Chan CC, Rootman DS. Corneal lamellar flap retraction after LASIK following penetrating keratoplasty. Cornea 2004; 23:643–646 24. Hersh PS, Jordan AJ, Mayers M. Corneal graft rejection episode after excimer laser phototherapeutic keratectomy. Arch Ophthalmol 1993; 111:735–736 25. Kovoor TA, Mohamed E, Cavanagh HD, Bowman RW. Outcomes of LASIK and PRK in previous penetrating corneal transplant recipients. Eye Contact Lens 2009; 35:242–245 26. Kwitko S, Marinho DR, Rymer S, Ramos Filho S. Laser in situ keratomileusis after penetrating keratoplasty. J Cataract Refract Surg 2001; 27:374–379 27. Ranchod TM, McLeod SD. Wound dehiscence in a patient with keratoconus after penetrating keratoplasty and LASIK. Arch Ophthalmol 2004; 122:920–921. Available at: http://archopht. jamanetwork.com/data/Journals/OPHTH/9925/ecr0604-4-1.pdf. Accessed February 1, 2014 JL, Pin ~ ero D. Very high-frequency digital ultrasound mea28. Alio surement of the LASIK flap thickness profile using the IntraLase femtosecond laser and M2 and Carriazo-Pendular microkeratomes. J Refract Surg 2008; 24:12–23
J CATARACT REFRACT SURG - VOL 40, MAY 2014
