Penetrating Keratoplasty vs.

Epikeratoplasty for the Surgical Treatment of Keratoconus

Michael D. Wagoner, MD; Scott D. Smith, MD, MPH; Wilfried J. Rademaker, MD, MPH; Muneera A. Mahmood, MD

ABSTRACT
PURPOSE: Penetrating keratoplasty and epikeratoplasty have been utilized in the surgical treatment of keratoconus. Comparison of the relative efficacy of each procedure in achieving visual outcomes has not been achieved due to limited numbers of cases and follow-up in previous series. METHODS: All patients who underwent either penetrating keratoplasty or epikeratoplasty for keratoconus between January 1987 and December 1997, and for whom at least 24 months of postoperative follow-up data for visual acuity was documented in the medical record, were included in this retrospective, nonrandomized, sequential comparative trial. The sole criteria for outcome in each group, as well as for comparison of the two groups, was Snellen visual acuity measured at the time of each follow-up with the presenting optical aid. RESULTS: Inclusion criteria were met for 443 eyes treated with penetrating keratoplasty and 161 eyes treated with epikeratoplasty. Mean followup was 4.3 years for penetrating keratoplasty and 4.5 years for epikeratoplasty. In each group, approximately 50% of the patients chose rehabilitation with optical correction with either spectacles or contact lenses and 50% chose no optical correction. Final median logMAR visual acuity for all patients, irrespective of means of visual rehabilitation, was 0.30 (20/40) for penetrating keratoplasty and 0.40 (20/50) for epikeratoplasty (P < .00005). In 209 penetrating keratoplasty and 77 epikeratoplasty eyes with optical correction, the final median logMAR visual acuity was 0.18 (20/30) for penetrating keratoplasty and 0.40 (20/50) for epikeratoplasty (P < .00005). The final median logMAR

visual acuity in 234 penetrating keratoplasty and 84 epikeratoplasty eyes without optical correction was 0.48 (20/60) in both groups (P-value was not statistically significant). CONCLUSIONS: Although penetrating keratoplasty was statistically superior to epikeratoplasty with respect to visual outcome, results with epikeratoplasty were adequate to recommend its use as a surgical alternative in cases when it is not desirable to perform penetrating keratoplasty. [J Refract Surg 2001;17:138-146]

From the Anterior Segment/External Disease Division, Department of Ophthalmology and the Department of Research, King Khaled Eye Specialist Hospital, Riyadh, Kingdom of Saudi Arabia (Rademaker, Mahmood, Smith, Wagoner), the Department of Ophthalmology and Visual Sciences, University of Iowa Hospitals and Clinics, Iowa City, Iowa, (Wagoner), and the Division of Ophthalmology, Department of Surgery, Cleveland Clinic, Cleveland, Ohio (Smith). The authors have no proprietary interest in the materials mentioned herein. Correspondence: Michael D. Wagoner, MD, Department of Ophthalmology and Visual Sciences, University of Iowa Hospitals and Clinics, 200 Hawkins Drive, Iowa City, Iowa, 52242-1091. Received: June 20, 2000 Accepted: December 12, 2000

eratoconus is a bilateral, non-inflammatory central corneal thinning disorder characterized by the development of a corresponding protrusion (cone) with an apex often located in an inferior eccentric position.1 The treatment of keratoconus depends upon the severity of the disease and includes the following options: spectacles, rigid gaspermeable corneal or scleral contact lens, and surgical therapy.2 Lamellar keratoplasty was once the surgical treatment of choice for keratoconus.3 Technical difficulties with lamellar keratoplasty, as well as suboptimal visual outcomes due to interface opacification and irregular astigmatism, led to virtual abandonment of this therapeutic option.4 Penetrating keratoplasty, which is technically easier to perform than lamellar keratoplasty, has a better visual prognosis and has been the surgical treatment of choice for contact lens-intolerant keratoconus.5-10 The visual outcomes of penetrating keratoplasty for keratoconus have been excellent, with uniformly low complication rates in all reported series.5-10 Still, there are concerns about immunologically-mediated graft rejection and failure5-9,11-13, microbial keratitis14-16, and traumatic wound dehiscence with loss of intraocular contents.17 Epikeratoplasty is a form of onlay lamellar keratoplasty in which a lens made of human corneal tissue is sutured onto the anterior surface of the cornea to change its anterior curvature and refractive properties.18 This procedure combines

138

Journal of Refractive Surgery Volume 17 March/April 2001

Penetrating Keratoplasty vs. Epikeratoplasty for Keratoconus/Wagoner et al

some of the advantages of lamellar keratoplasty (extraocular procedure, preservation of host endothelium, reduced or eliminated risk of graft rejection, normal postoperative activity, and elimination of risk of traumatic dehiscence of the globe)3,4 and penetrating keratoplasty (less interface opacification and irregular astigmatism than with lamellar keratoplasty).4-6 Although a number of authors have described short19,20 and long-term21,22 success with epikeratoplasty for keratoconus, and a few nonrandomized studies have compared epikeratoplasty with penetrating keratoplasty for keratoconus23-26, these studies have an inadequate number of cases and/or insufficient long-term follow-up to allow definitive conclusions regarding the potential role of epikeratoplasty in the surgical management of keratoconus. This study analyzes the results of a large number of penetrating keratoplasty and epikeratoplasty procedures performed at a single institution by a relatively small group of surgeons for patients with keratoconus in whom long-term follow-up was available. PATIENTS AND METHODS All patients who underwent either penetrating keratoplasty or epikeratoplasty for keratoconus by members of the Anterior Segment Division at the King Khaled Eye Specialist Hospital (KKESH) between 1 January 1987 and 31 December 1997, and for whom at least 24 months of postoperative visual acuity follow-up data were documented in the medical record, were included in this retrospective, nonrandomized, sequential comparative trial. No patients who met these minimal inclusion criteria were excluded from the study. Study Design This study was a retrospective, nonrandomized, sequential comparative trial. Eyes that underwent a primary penetrating keratoplasty procedure and met the inclusion criteria were categorized in the penetrating keratoplasty group (443 eyes). Eyes that underwent a primary epikeratoplasty procedure and met the inclusion criteria were categorized in the epikeratoplasty group (161 eyes). The study tracked the visual outcome based on the original decision to first perform either a penetrating keratoplasty or an epikeratoplasty. Subsequent procedures and interventions, if any, did not affect the group assignment. Seven penetrating keratoplasty patients who underwent repeat penetrating keratoplasty remained in the penetrat-

ing keratoplasty group. Two epikeratoplasty patients who underwent repeat epikeratoplasty and two epikeratoplasty patients who underwent secondary penetrating keratoplasty remained in the epikeratoplasty group. Because these cases represented a small minority of cases relative to the size of the study group, the decision to include them in the final analysis did not have any impact on the final median Snellen visual acuity at any time point in either group. The criteria for outcome of each group, as well as for comparison of the two groups, was solely Snellen visual acuity as measured at the time of each followup examination by E-optotype orientation under dim examination room conditions with the presenting optical aid (without correction, with spectacles, or with contact lens) being utilized by the patient for visual rehabilitation. The endpoint for observation was the most recent patient examination. Selection of Procedure The choice of surgical intervention (penetrating keratoplasty or epikeratoplasty) was at the discretion of individual surgeons. Some surgeons selected epikeratoplasty in cases of mild to moderate apical corneal scars if they felt that these scars would not significantly affect the final visual outcome (ie, visual acuity of >20/40 expected); other surgeons selected only penetrating keratoplasty in eyes with apical corneal scars. If dense apical corneal scars were present, and the surgeon felt they would adversely affect the visual prognosis (ie, visual acuity of <20/40), penetrating keratoplasty was selected unless there were strong contraindications to penetrating keratoplasty. For example, in a child with Downs syndrome where a significant apical scar was felt to limit the visual prognosis with epikeratoplasty to approximately 20/50, epikeratoplasty was offered because of concerns about serious complications related to noncompliance and eye rubbing following penetrating keratoplasty, where the theoretical visual prognosis in a best case scenario was 20/20. Surgical Technique for Penetrating Keratoplasty Penetrating keratoplasty was performed using standard techniques5, with an average donor graft size of 7.75 mm and an average recipient graft size of 7.50 mm. Suturing techniques included interrupted sutures only (approximately 10%), continuous suture only (approximately 35%), or interrupted-continuous (approximately 55%), depending upon surgeon preference.

Journal of Refractive Surgery Volume 17 March/April 2001

139

Penetrating Keratoplasty vs. Epikeratoplasty for Keratoconus/Wagoner et al

Postoperatively, all patients were treated as inpatients with topical antibiotics until re-epithelialization was complete. Depending on surgeon preference, pressure patching, bandage contact lens, or no occlusive therapy was used to facilitate re-epithelialization. Following discharge the patients were followed as outpatients at 2 weeks, 1, 3, 6, 12 months, and twice yearly thereafter. Topical corticosteroids were used four to six times daily immediately postoperatively and tapered gradually over a 1-year period. Loose sutures were removed when identified. Approximately 20% of eyes had all sutures removed by 12 months, 55% had all sutures removed by 24 months, and 85% had all sutures removed by 36 months. Surgical Technique for Epikeratoplasty Epikeratoplasty lenticules of between 6.5 to 9.5 mm diameter (median, 8.5 mm) and plano power were obtained from two different sources (lyophilized KERATOLENS from Allergan Medical Optics, Irvine, CA from September 1986 to September 1993; fresh BIO-LENS, from Cryo-Optic Inc., Houston, TX from February 1993 to December 1997). Epikeratoplasty lenticules were prepared for surgical use by previously described methods.18 All cases were performed by experienced surgeons of the Anterior Segment Division of KKESH using a previously described technique.19 Postoperatively, all patients were treated as inpatients with topical antibiotics and corticosteroids, and a bandage contact lens, pressure patching, or temporary tarsorrhaphy until re-epithelialization was complete. Following re-epithelialization, they were discharged and followed as outpatients at 2 weeks, 1, 3, 6, 12 months, and twice yearly thereafter. Topical corticosteroids (eg, prednisolone acetate 1% three times a day or a weaker corticosteroid) were used postoperatively to prevent premature suture vascularization and loosening, and were tapered gradually over 3 to 6 months. Topical antibi-

otics were used postoperatively four times a day until re-epithelialization was complete and were tapered over 3 to 6 months. Suture removal was delayed until 3 to 6 months, with the exception of removal of sutures that loosened or vascularized prematurely. Outcome Measures Data collected from medical records included demographic data (age, sex); date of examination closest to the point in time prior to the surgical procedure (preoperative examination); date of the final documented postoperative examination; preoperative visual acuity (with presenting type of correction, where available), presence or absence of apical corneal scars; postoperative visual acuity at 2 years (6 mo), each year (6 mo) thereafter, and the final recorded examination; the means by which the recorded visual acuity was achieved at each examination (with spectacles, contact lens, or no correction); early and late postoperative complications resulting in graft or lenticule failure; and subsequent surgical intervention with final visual outcome. Statistical Analysis Data were analyzed using STATA 6.0 (College Station, TX) statistical software. Visual acuity results at each follow-up time point were converted to logMAR scale and treated as continuous variables. Comparison of the distribution of postoperative visual acuity by procedure type was done using the Mann-Whitney test. Analysis of results, other than visual acuity, was performed using the Fishers exact or chi-square test for categorical variables, and two-sample t-test for continuous variables. RESULTS A total of 604 patients qualified for the study; 443 in the penetrating keratoplasty group and 161 in the epikeratoplasty (Table 1). Ninety-four (21.2%)

Table 1 Comparison of Penetrating Keratoplasty and Epikeratoplasty Patient Groups

Total patients Male (%) Female (%) Apical scar (%) Vernal keratoconjunctivitis (%) Median age (yr) at surgery (%) Median follow-up in years (%)
*Not statistically significant

Penetrating Keratoplasty 443 274 (61.9) 169 (38.1) 439 (99.1) 94 (21.2) 21.5 (8.6) 4.3 (2.0)

Epikeratoplasty 161 99 (61.5) 62 (38.5) 84 25 20.3 4.5 (52.2) (15.5) (5.8) (2.1)

P-value N.S* N.S. .001 N.S. N.S. N.S.

140

Journal of Refractive Surgery Volume 17 March/April 2001

Penetrating Keratoplasty vs. Epikeratoplasty for Keratoconus/Wagoner et al

Table 2 Penetrating Keratoplasty vs. Epikeratoplasty: Postoperative Method of Optical Management

Time Preoperative 2 years 3 years 4 years Final examination N Patients 443 410 297 215 443 Penetrating Keratoplasty None Spectacles Contact Lens (%) (%) (%) 383 (86) 35 (8) 25 (6) 305 (74) 91 (22) 14 (3 185 (62) 85 (29) 27 (9) 117 (54) 76 (35) 22 (10) 234 (53) 160 (36) 49 (11) N Patients 161 137 109 79 161 Epikeratoplasty None Spectacles (%) (%) 88 (55) 29 (18) 71 (52) 54 (39) 52 (48) 47 (43) 41 (52) 29 (37) 84 (52) 66 (41) Contact Lens (%) 44 (27) 12 (9) 10 (9) 9 (11) 11 (7)

Table 3 Penetrating Keratoplasty vs. Epikeratoplasty: Visual Acuity (Percentile)* Irrespective of Method of Optical Management
Time Pre-op 2 years 3 years 4 years Final N Patients 443 410 297 215 443 10th 20/800 20/200 20/200 20/160 20/160 Penetrating Keratoplasty 25th 50th 75th 20/800 20/100 20/90 20/80 20/80 20/800 20/50 20/50 20/40 20/40 20/200 20/40 20/30 20/30 20/30 90th 20/100 20/30 20/25 20/25 20/25 N Patients 161 137 109 79 161 10th 20/2000 20/130 20/125 20/160 20/125 Epikeratoplasty 25th 50th 75th 20/400 20/100 20/100 20/100 20/100 20/125 20/60 20/60 20/60 20/50 20/60 20/40 20/40 20/40 20/40 90th 20/40 20/40 20/30 20/30 20/30 P-value <.00005 N.S. 0.005 0.002 0.00005

*The visual acuity outcomes were stratified from the worst (1st percentile) to the best (99th percentile) at each time point for each group. The visual acuity at the 50th percentile (where 50% of eyes had a better visual acuity; 50% of eyes had a worse visual acuity) represents the median visual acuity. The higher and lower percentiles provide an opportunity to see how patients who did poorly (eg, 10th percentile, where 10% of eyes had worse visual acuity and 90% of eyes had a better visual acuity) or well (eg, 90th percentile, where 90% of eye had worse visual acuity and 10% of eyes had better visual acuity) compared. The P-value represents comparison of the median visual acuity data at each time point between penetrating keratoplasty and epikeratoplasty.

patients in the penetrating keratoplasty group and 25 (15.5%) patients in the epikeratoplasty group had concomitant vernal keratoconjunctivitis. Apical scars were present in 439 (99.1%) eyes in the penetrating keratoplasty group and 84 (52.2%) eyes in the epikeratoplasty group (P < .001). No patients in either group had any other significant ocular abnormalities including glaucoma, cataract, or retinal pathology. There was no statistically significant difference in the gender of the patients, median length of follow-up, or median age at the time of surgery. At least one data point was available for all eyes at or beyond the 2-year follow-up criteria. Visual acuity was not available within the 3-month time interval around the 2-year time point for 24 epikeratoplasty eyes and 33 penetrating keratoplasty eyes, but subsequent documentation of visual outcome was available. Mean follow-up was 4.3 years for the penetrating keratoplasty group and 4.5 years for the epikeratoplasty group. The method of optical management selected by the patient for each group during the study period is summarized in Table 2. During the first 3 postoperative years, patients in the epikeratoplasty group were significantly more likely to choose optical cor-

rection (either spectacles or contact lenses) than those in the penetrating keratoplasty group (P = .019). At 4 postoperative years and at the final follow-up examination, the preferred mode of optical correction was almost identical in both groups; approximately half of each group selected optical correction and half preferred no optical correction. The cumulative visual acuity for each group, irrespective of the chosen method of rehabilitation (no correction, spectacles, or contact lenses), is summarized in Table 3. The median logMAR visual acuity at the final examination was 0.30 (20/40) for penetrating keratoplasty and 0.40 (20/50) for epikeratoplasty (P < .00005). There was no statistically significant difference in visual outcome between the two groups at 2 years. Beginning at 3 years postoperatively, and continuing to the final postoperative examination, the visual outcome was significantly better with penetrating keratoplasty than epikeratoplasty. The difference in statistical significance between penetrating keratoplasty and epikeratoplasty was progressively greater at each time point (3-year, 4-year, and final examination). There was no statistically significant difference in the final visual results with epikeratoplasty utilizing

Journal of Refractive Surgery Volume 17 March/April 2001

141

Penetrating Keratoplasty vs. Epikeratoplasty for Keratoconus/Wagoner et al

Table 4 Penetrating Keratoplasty vs. Epikeratoplasty: Visual Acuity (Percentile)* in Eyes With Spectacles or Contact Lenses
N Patients Pre-op 60 2 years 105 3 years 112 4 years 98 Final 209 Time 10th 20/800 20/80 20/80 20/80 20/60 Penetrating Keratoplasty 25th 50th 75th 90th 20/400 20/40 20/50 20/40 20/40 20/125 20/30 20/30 20/30 20/30 20/60 20/30 20/30 20/25 20/25 20/50 20/20 20/25 20/25 20/20 N Patients 73 66 57 38 77 10th 20/200 20/125 20/125 20/200 20/125 Epikeratoplasty 25th 50th 75th 20/125 20/80 20/80 20/80 20/80 20/50 20/60 20/50 20/40 20/50 20/40 20/40 20/40 20/30 20/40 90th 20/30 20/30 20/30 20/25 20/30 P-value <0.00005 <0.00005 < 0.0005 0.0003 <0.00005

*The visual acuity outcomes were stratified from the worst (1st percentile) to the best (99th percentile) at each time point for each group. The visual acuity at the 50th percentile (where 50% of eyes had a better visual acuity; 50% of eyes had a worse visual acuity) represents the median visual acuity. The higher and lower percentiles provide an opportunity to see how patients who did poorly (eg, 10th percentile, where 10% of eyes had worse visual acuity and 90% of eyes had a better visual acuity) or well (eg, 90th percentile, where 90% of eye had worse visual acuity and 10% of eyes had better visual acuity) compared. The P-value represents comparison of the median visual acuity data at each time point between penetrating keratoplasty and epikeratoplasty.

Table 5 Penetrating Keratoplasty vs. Epikeratoplasty: Visual Acuity (Percentile)* in Eyes Without Optical Correction
N Patients Pre-op 383 2 years 305 3 years 185 4 years 117 Final 234 Time 10th 20/800 20/200 20/200 20/200 20/200 Penetrating Keratoplasty 25th 50th 75th 90th 20/800 20/125 20/125 20/125 20/125 20/800 20/60 20/60 20/60 20/60 20/200 20/40 20/40 20/40 20/40 N Patients 20/160 88 20/30 71 20/30 52 20/30 41 20/30 84 10th 20/2000 20/125 20/125 20/160 20/200 Epikeratoplasty 25th 50th 75th 20/800 20/100 20/100 20/100 20/100 20/400 20/60 20/60 20/80 20/60 20/125 20/40 20/40 20/50 20/50 90th 20/100 20/40 20/30 20/40 20/30 P-value N.S. N.S. N.S. N.S. N.S.

*The visual acuity outcomes were stratified from the worst (1st percentile) to the best (99th percentile) at each time point for each group. The visual acuity at the 50th percentile (where 50% of eyes had a better visual acuity; 50% of eyes had a worse visual acuity) represents the median visual acuity. The higher and lower percentiles provide an opportunity to see how patients who did poorly (eg, 10th percentile, where 10% of eyes had worse visual acuity and 90% of eyes had a better visual acuity) or well (eg, 90th percentile, where 90% of eye had worse visual acuity and 10% of eyes had better visual acuity) compared. The P-value represents comparison of the median visual acuity data at each time point between penetrating keratoplasty and epikeratoplasty. Not statistically significant.

lyophilized vs. fresh tissue or when performed in eyes with or without apical corneal scars (P = .4). The visual acuity for patients in each group who chose rehabilitation with optical aids (spectacles or contact lens) is summarized in Table 4. Among this subset of 209 penetrating keratoplasty and 77 epikeratoplasty eyes, the median logMAR visual acuity at the final follow-up examination was 0.18 (20/30) for penetrating keratoplasty and 0.40 (20/50) for epikeratoplasty (P<.00005). Penetrating keratoplasty was significantly better statistically than epikeratoplasty at all time points from 2 years postoperatively until the final postoperative examination. The visual acuity for patients in each group who chose rehabilitation without optical aids is summarized in Table 5. Among this subset of 234 penetrating keratoplasty and 84 epikeratoplasty eyes, the median logMAR visual acuity was 0.48 (20/60) for both penetrating keratoplasty and epikeratoplasty. There was no statistically significant difference in visual acuity at 2, 3, or 4 years, or at the final

follow-up examination between the two groups. All cases of penetrating keratoplasty and epikeratoplasty failure are summarized in Tables 6 and 7, respectively. There were ten (2.3%) eyes with loss of penetrating keratoplasty clarity and eight (5.0%) eyes with epikeratoplasty lenticule failure. DISCUSSION The most common indication for surgical intervention in patients with keratoconus is contact lens failure. Contact lens wear in a patient with keratoconus is a serious commitment that requires multiple examinations for fitting and frequent lens changes, adjustments, and replacements. The Saudi Arabian population described in this series differs from its Western counterparts in that many patients do not have easy access to tertiary care facilities due to large physical distances. As a result, many patients elect not to embark upon a path of contact lens wear and request surgical intervention when spectacle correction becomes inadequate. The

142

Journal of Refractive Surgery Volume 17 March/April 2001

Penetrating Keratoplasty vs. Epikeratoplasty for Keratoconus/Wagoner et al

Table 6 Penetrating Keratoplasty: Course of Eyes With Graft Failure

Case Etiology Intervention Final Best Spectaclecorrected Visual Acuity Count fingers Count fingers Count fingers 20/30

Table 7 Epikeratoplasty: Course of Eyes With Lenticule Failure

Case Etiology Intervention Final Best Spectaclecorrected Visual Acuity Count fingers Count fingers 20/40 20/80 20/125 20/60 20/30 20/300

1. 2. 3. 4. 5.

6.

7. 8. 9. 10.

Immune rejection Immune rejection Immune rejection Primary graft failure Traumatic dehiscence with endothelial failure Sterile ulceration at graft-host junction Microbial keratitis Microbial keratitis Microbial keratitis Microbial keratitis

None None None Repeat PKP*

1. 2. 3. 4. 5.

Repeat PKP

20/25

Repeat Repeat Repeat Repeat Repeat

PKP PKP PKP PKP PKP

20/160 20/25 20/50 20/60 20/80

6. 7. 8.

Persistent epithelial defect* Persistent epithelial defect* Persistent epithelial defect* Persistent epithelial defect* Sterile Mooren-like ulcer Microbial keratitis High cylinder High cylinder

Remove lenticule; not repeated Remove lenticule; not repeated Remove lenticule; not repeated Remove lenticule; repeat EKP Remove lenticule; not repeated Remove lenticule; repeat EKP Secondary PKP** Secondary PKP

*Penetrating keratoplasty

*Primary lenticule failure due to initial failure to re-epithelialize After successful contact lens fit Epikeratoplasty **Penetrating keratoplasty

decision to proceed with surgical intervention is not always straightforward because the same factors that compromise access to preoperative contact lens therapy also compromise access to postoperative surgical care. Even among those patients with access to contact lens facilities, many Saudi Arabian patients are unwilling to attempt contact lens therapy. Those that do find their comfort and wearing time are compromised by the dry and dusty environment. Many patients who attempt contact lens therapy abandon it relatively early. The lack of acceptance and/or availability of hard contact lenses as a major form of visual rehabilitation for keratoconus in Saudi Arabia creates a dilemma of how best to deal with the patient with keratoconus, especially if the cornea is clear or has visually insignificant apical scarring. The problem is compounded by the fact that many of these patients are still in secondary school and are handicapped by poor functional visual acuity. It has been unclear whether it is better to offer these patients epikeratoplasty with its relative safety but presumptive slightly less visual potential, or penetrating keratoplasty with its greater risks but potential for better visual acuity. Although no randomized prospective study was ever designed at KKESH to systematically address this issue, decisions were made on a case-by-case basis over the 11-year period described in this study,

providing the opportunity to make a retrospective comparison of a large number of patients undergoing either penetrating keratoplasty or epikeratoplasty with the availability of long-term follow-up. Visual Results: Penetrating Keratoplasty The literature on penetrating keratoplasty for keratoconus supports a high rate of graft clarity with stable visual acuity and little tendency for recurrent disease during the first postoperative decade.5-10 Two recent studies demonstrated that good results similar to those achieved in Western centers could be achieved with penetrating keratoplasty in Saudi Arabia, despite concerns about compliance, prompt availability of follow-up care, and a substantial percentage of patients with concomitant vernal keratoconjunctivitis.10,28 The only parameter studied in detail in the present study was final visual outcome, which reflects the cumulative impact of all adverse events that occurred from the primary procedure to the most recent patient follow-up examination. Among the entire keratoconus group, a final median visual acuity of 20/40 was achieved despite the fact that slightly less than half the eyes were rehabilitated with spectacles or contact lenses. Among patients electing to use spectacle or contact lenses, the median visual acuity of 20/30, with 25% of patients

Journal of Refractive Surgery Volume 17 March/April 2001

143

Penetrating Keratoplasty vs. Epikeratoplasty for Keratoconus/Wagoner et al

20/25 or better, was comparable to published series where best spectacle-corrected or contact lenscorrected visual acuity was the major criteria for surgical success.5 Although a median visual acuity of 20/60 was achieved in patients who chose no correction, there was some selection bias toward patients who were satisfied with their visual outcomes and did not require further correction, as well as the small number of patients with major complications and a visual outcome that was not amenable to optical correction. Visual Results: Epikeratoplasty The literature on visual outcomes following epikeratoplasty for keratoconus is less complete than that for penetrating keratoplasty, with most series limited by either small numbers20,21 or limited follow-up.19 The Nationwide Study for Epikeratoplasty, the largest published series to date with 177 cases performed by 69 surgeons, has only follow-up after suture removal of more than 30 days for 82 eyes.19 In a series with median follow-up of 21 months20 and 67 months21, there were only 40 cases and 11 cases reported, respectively.20,21 The current series includes 161 cases performed at a single center in which a minimum of 24 months and a median of 52 months of follow-up was available. Overall, previous visual outcomes have been reported to be satisfactory with epikeratoplasty.19-21 The percentage of eyes with a best spectaclecorrected visual acuity of 20/40 or better ranged from 78% to 92%.19-21 The median best spectaclecorrected visual acuity ranged from 20/25 to 20/30.20,21 A consistent feature of these studies was the infrequency with which a best spectaclecorrected acuity of 20/20 was achieved.19-21 It is difficult to compare the results in the present series to previous studies. Patients with some amount of apical scarring (52% of cases) were included in the present series, whereas this was a contraindication to epikeratoplasty in the previous studies. Although there was no significant difference between visual outcome in eyes with or without apical scarring (P = .4), the inclusion of cases with mild apical scarring undoubtedly contributed to some decrease in the median visual acuity compared to previous series19-22, as well as further reduction in the percentage of eyes that achieved a final visual acuity of 20/20 or better.19-22 Irrespective of the limitations in comparing the results in this study of epikeratoplasty for keratoconus with those previously published, the current

series is sufficiently large with enough patients in each category of visual rehabilitation to draw reasonable conclusions regarding the efficacy of epikeratoplasty in achieving satisfactory visual outcomes. Among all 161 eyes, the final median visual acuity was 20/50, despite the fact that slightly less than half the eyes were rehabilitated with spectacles or contact lenses. Among patients electing to use spectacles or contact lenses, the median visual acuity was also 20/50, and in patients who chose no correction, it was 20/60. Visual Results: Penetrating Keratoplasty vs. Epikeratoplasty Comparative studies of visual outcome of penetrating keratoplasty vs. epikeratoplasty are relatively limited.23-26 As with visual acuity outcomes of epikeratoplasty, these studies have unsatisfactory patient numbers, follow-up period, or both, to draw firm conclusions. Fronterre reported 30 eyes that received either penetrating keratoplasty or epikeratoplasty during the same time interval and in which 36 to 72 months of follow-up was available.24 The patients were matched for age, sex, refractive astigmatism, mean keratometric power, and spherical refractive error. Only those penetrating keratoplasty patients who would have met criteria for inclusion in the epikeratoplasty study were included. Patients in both groups had a similar mean uncorrected visual acuity (penetrating keratoplasty, 20/63; epikeratoplasty, 20/52), best spectacle-corrected visual acuity (penetrating keratoplasty, 20/22; epikeratoplasty, 20/23), and contact lens visual acuity (penetrating keratoplasty, 20/20; epikeratoplasty, 20/21). Steinert compared 20 contemporous eyes that underwent penetrating keratoplasty or epikeratoplasty.25 With a mean follow-up period of 33 months for penetrating keratoplasty and 25 months for epikeratoplasty, the mean best spectacle-corrected visual acuity was 20/27 after penetrating keratoplasty and 20/32 after epikeratoplasty. No eyes achieved a best spectacle-corrected visual acuity of 20/20 or better with epikeratoplasty. Goosey reported 16 nonrandomized eyes treated with penetrating keratoplasty (mean follow-up of 20 mo) and 31 treated with epikeratoplasty (mean follow-up 19 mo).26 Although 93% of eyes in each group had a best spectacle-corrected visual acuity with contact lenses of 20/40 or better, a higher percentage of eyes that had undergone penetrating keratoplasty achieved visual acuity of 20/40 or better

144

Journal of Refractive Surgery Volume 17 March/April 2001

Penetrating Keratoplasty vs. Epikeratoplasty for Keratoconus/Wagoner et al

with spectacles (93% vs. 76%) or a best spectaclecorrected visual acuity of 20/20 or better with either spectacles or contact lens (73% vs. 24%). The current study provides more cases and longer follow-up than previously published series to make a valid comparison of the visual outcome of penetrating keratoplasty vs. epikeratoplasty in the surgical treatment of keratoconus. The data is further strengthened because: (1) the study involved a large number of patients treated by a relatively small number of surgeons at a single institution using similar protocols, (2) variability between the two groups that might have been induced by mitigating circumstances such as compliance, distance from the treating center, and other socioeconomic circumstances were eliminated by selection of patients from the same pool, and (3) the similar percentage of type of ocular rehabilitation chosen by patients in both groups. When the data were analyzed irrespective of the preferred method of visual rehabilitation, there was no significant difference in visual acuity between penetrating keratoplasty and epikeratoplasty for the first 2 postoperative years. Beginning with the third postoperative year, the statistical significance of better visual acuity with penetrating keratoplasty than epikeratoplasty increased with progressively longer follow-up. This may have been due to the relatively low percentage of penetrating keratoplasty eyes with optical correction after 2 years (26%), and the subsequent increase in percentage of penetrating keratoplasty eyes with spectacle or contact lens rehabilitation at the end of the study (47%), as opposed to the relatively fixed percentage of epikeratoplasty eyes with optical correction at 2 years (45%) and at the end of the study (48%). The final median visual acuity was one Snellen line better with penetrating keratoplasty (20/40 vs. 20/50) and was highly statistically significant (P < .00005). The difference in visual outcome between penetrating keratoplasty and epikeratoplasty was even more striking among patients who chose optical correction. Corrected visual outcomes were available for 209 (47%) penetrating keratoplasty eyes and 77 (48%) epikeratoplasty eyes. Although this represents less than 50% of each group, the absolute number of patients available for analysis of longterm outcome is larger than any existing comparative series.24-26 Significantly better median corrected visual acuity was achieved with penetrating keratoplasty at 2 years and maintained thereafter. The final median corrected visual acuity was two

Snellen lines better with penetrating keratoplasty (20/30 vs. 20/50) and was highly statistically significant (P < .00005). As with previous studies, in most patients excellent vision was achieved with penetrating keratoplasty, and satisfactory vision with epikeratoplasty. Although penetrating keratoplasty remains the surgical procedure of choice for keratoconus, epikeratoplasty may be selected as an entry level procedure in some situations. These include cases where (1) professional or athletic demands preclude a prolonged period of restricted activity or pose a threat of wound dehiscence (eg, contact sports such as boxing), (2) the visual demands are reduced and/or safety concerns weigh heavily (eg, Downs syndrome), or (3) the patient is willing to undergo a potential staged approach to visual rehabilitation in return for the added safety and reduced restriction of epikeratoplasty, holding secondary penetrating keratoplasty in reserve if initial results are unsatisfactory. The large number of patients enrolled in the current study, along with the stratification of visual acuity results into percentiles at each time point, provide data that can be used to provide prospective surgical candidates with reasonable expectations of surgical outcome with either procedure. REFERENCES
1. Krachmer JH, Feder FS, Belin MW. Keratoconus and related noninflammatory corneal thinning disorders. Surv Ophthalmol 1984;28:293-322. 2. Lass JH, Lembach RAG, Park SB, Hom DL, Fritz ME, Svilar GM, Nuamah IF, Reinhart WJ, Stocker EG, Keates RH. Clinical management of keratoconus. A multicenter analysis. Ophthalmology 1990;97:433-435. 3. Wood TO. Lamellar keratoplasty in keratoconus. Am J Ophthalmol 1971;86:543-536. 4. Richard JM, Paton D, Gasset AR. A comparison of penetrating keratoplasty and lamellar keratoplasty in the surgical management of keratoconus. Am J Ophthalmol 1978;86:807-810. 5. Kirkness CM, Ficker LA, Steele AD, Rice NS. The success of penetrating keratoplasty for keratoconus. Eye 1990;4: 673-688. 6. Sharif KW, Casey TA. Penetrating keratoplasty for keratoconus: complications and long-term success. Br J Ophthalmol 1991;75:142-146. 7. Tuft SJ, Gregory WM, Davison CR. Bilateral penetrating keratoplasty for keratoconus. Ophthalmology 1995;102: 462-468. 8. Paglen PG, Fine M, Abbott RL, Webster RG. The prognosis for keratoplasty in keratoconus. Ophthalmology 1982;89:651-654. 9. Price FW Jr, Whitson WE, Marks RG. Graft survival in four common groups of patients undergoing penetrating keratoplasty. Ophthalmology 1991;98:322-328. 10. Mahmood MA, Wagoner MD. Penetrating keratoplasty for keratoconus: Long-term results in 38 eyes with and 202 eyes without vernal keratoconjunctivitis. Mid East J Ophthalmol 1999;7:26-35.

Journal of Refractive Surgery Volume 17 March/April 2001

145

Penetrating Keratoplasty vs. Epikeratoplasty for Keratoconus/Wagoner et al

11. Chandler JW, Kaufman HE. Graft rejections after keratoplasty for keratoconus. Am J Ophthalmol 1974;77:543-537. 12. Epstein RJ, Seedor JA, Dreizen NG. Penetrating keratoplasty for herpes simplex keratitis and keratoconus: allograft rejection and survival. Ophthalmology 1987;94: 935-942. 13. Girard LJ, Esnaola N, Rao R, Barnett L, el-Maghraby A, Canizales R. Allograft rejection after penetrating keratoplasty for keratoconus. Ophthalmic Surg 1993;24:40-43. 14. Fong LP, Ormerod LD, Kenyon KR, Foster CS. Microbial keratitis complicating penetrating keratoplasty. Ophthalmology 1988;95:1269-1275. 15. Varley GA, Meisler DM. Complications of penetrating keratoplasty: graft infections. Refract Corneal Surg 1991;7: 62-66. 16. al-Hazaa S, Tabbara K. Bacterial keratitis after penetrating keratoplasty. Ophthalmology 1988;95:1504-1508. 17. Rehany U, Rumelt S. Ocular trauma following penetrating keratoplasty: incidence, outcome, and postoperative recommendations. Arch Ophthalmol 1998;116:1282-1286. 18. Kaufman HE, Werblin TP. Epikeratophakia for the treatment of keratoconus. Am J Ophthalmol 1982;93:342-347. 19. McDonald MB, Kaufman HE, Durrie DS, Keates RH, Sanders DR. Epikeratophakia for keratoconus: the nationwide study. Arch Ophthalmol 1986;104:1294-1300. 20. Fronterre A, Portesani GP. Epikeratoplasty for keratoconus. Report of 40 cases. Cornea 1989;8:236-239.

21. Waller SG, Steinert RF, Wagoner MD. Long-term results of epikeratoplasty for keratoconus. Cornea 1995;14:84-88. 22. Rademaker WJ, Wagoner MD. Epikeratoplasty for keratoconus: long-term results in 114 eyes with and 108 eyes without apical corneal scarring. Middle East J Ophthalmology 1999;7:36-47. 23. McDonald MB, Safir A, Waring GO, Schlichtemeier WR, Kissling GE, Kaufman HE. A preliminary comparative study of epikeratophakia or penetrating keratoplasty for keratoconus. Am J Ophthalmol 1987;103:467. 24. Fronterre A, Portesani GP. Comparison of epikeratoplasty and penetrating keratoplasty for keratoconus. Refract Corneal Surg 1991;7:167-173. 25. Steinert RF, Wagoner MD. Long term comparison of epikeratoplasty and penetrating keratoplasty for keratoconus. Arch Ophthalmol 1988;106:493-496. 26. Goosey JD, Prager TC, Goosey CB, Bird EF, Sanderson JC. A comparison of penetrating keratoplasty to epikeratoplasty in the management of keratoconus. Am J Ophthalmol 1991;111:145-151. 27. Teichmann KD, Wagoner MD. Mooren ulcer after epikeratoplasty for keratoconus. Arch Ophthalmol 1998;116: 1381-1382. 28. Mahmood M, Wagoner MD. Penetrating keratoplasty in eyes with keratoconus and vernal keratoconjunctivitis. Cornea 2000;19:468-470.

146

Journal of Refractive Surgery Volume 17 March/April 2001