Purpose
To compare the ocular characteristics and visual outcomes of eyes with corneal ectasia that were fitted with the Prosthetic Replacement of the Ocular Surface Ecosystem (PROSE) scleral device to those that underwent keratoplasty.
Design
Retrospective, comparative case series.
Methods
We reviewed the charts of consecutive patients with corneal ectasia that were evaluated for PROSE or underwent keratoplasty at our institution. Clinical data, topographic indices, and corneal thickness were reviewed, and eyes were stratified according to the Amsler-Krumeich classification for severity of ectasia. Only the more severe eye of each patient was included in the study. We compared visual acuity before and after PROSE fitting or keratoplasty. For PROSE evaluations, achievement of satisfactory fit and continued wear at 1 year of follow-up were recorded.
Results
From 2010 to 2012, 36 patients underwent PROSE evaluation for corneal ectasia while 37 patients underwent keratoplasty for the same indication. All eyes were successfully fitted with the PROSE device. Eyes in the keratoplasty group had more severe ectasia than eyes in the PROSE group ( P = .038). Visual acuity was achieved more rapidly in the PROSE cohort compared to keratoplasty, and mean visual acuity was significantly better for all eyes ( P < .0001) and when including only eyes with stage 4 ectasia ( P < .001). More eyes with stage 4 ectasia achieved 20/25 visual acuity after PROSE than after keratoplasty ( P = .003). At 1 year follow-up in the PROSE cohort, Snellen acuity was 20/28 ( P = .108 vs keratoplasty), improving to 20/25 with over-refraction ( P = .006 vs keratoplasty).
Conclusions
Eyes with advanced corneal ectasia can be successfully fitted with the PROSE device, and the visual acuity outcome for stage 4 ectasia was better and more rapid compared to keratoplasty. The acuity remained excellent with 1 year of follow-up. PROSE evaluation should be considered in patients with advanced corneal ectasia before proceeding to keratoplasty, especially if the ectasia is deemed stable.
Visual acuity in corneal ectasia, including keratoconus (KCN), pellucid marginal degeneration, and ectasia after laser in situ keratomileusis (LASIK), may be reduced because of both irregular astigmatism and corneal scarring, and management primarily consists of optical correction to maximize visual function. Although most patients with ectasia can be managed with spectacles and contact lenses, 12%–20% of patients with keratoconus will progress to corneal transplantation because of difficulty achieving satisfactory vision or comfort with contact lenses. Although keratoplasty is often successful in increasing visual acuity, the potential exists for significant intraoperative and postoperative morbidity, especially given the expected survivorship of patients with KCN.
Recently, large-diameter lenses that vault the cornea and limbus and rest on the sclera have emerged as a treatment option to improve the visual function of patients with severe corneal ectasia. The Prosthetic Replacement of the Ocular Surface Ecosystem (PROSE) is a custom-fitted device developed at the Boston Foundation for Sight (Needham, Massachusetts, USA) to restore vision, reduce symptoms, and improve quality of life for patients suffering from complex corneal disease including corneal ectasia. Other authors have previously described use of the PROSE device to improve visual function of patients with corneal ectasia whose vision is not adequately corrected with spectacles or other types of contact lenses, and it has been suggested that many of these patients may otherwise require corneal transplantation. However, no previous studies have evaluated the success of the PROSE device based on the severity of corneal ectasia, or directly compared PROSE patients to those who underwent corneal transplantation. Therefore, the purpose of the current study was to compare the demographics, ocular characteristics, and visual outcomes of consecutive eyes with corneal ectasia referred for PROSE evaluation at our institution to consecutive eyes with corneal ectasia that underwent keratoplasty during the same time period.
Materials and Methods
The University of Michigan Institutional Review Board approved this retrospective study, which was conducted in accordance with the tenets of the Declaration of Helsinki guidelines and all applicable federal and state laws. We included consecutive eyes from January 2010 to December 2012 that were referred for PROSE evaluation for corneal ectasia, or that underwent keratoplasty for the same indication during the same time period, at the W.K. Kellogg Eye Center, University of Michigan, Ann Arbor, Michigan, USA. For the current study, corneal ectasia included a clinical diagnosis made by a fellowship-trained corneal specialist of KCN, pellucid marginal degeneration, or post-LASIK ectasia. All PROSE evaluations and fittings were completed by 1 provider (K.D.) who completed a PROSE clinical fellowship at the Boston Foundation for Sight. For each patient, only the eye with more severe ectasia was included in this study. Five surgeons performed the corneal transplants, with surgical technique varying according to surgeon preference; both penetrating keratoplasty (PKP) and deep anterior lamellar keratoplasty (DALK) were included. The decision to pursue PROSE evaluation or keratoplasty was at the surgeon’s discretion.
From the medical records we recorded the corrected distance visual acuity (CDVA), including the method used to achieve the acuity, in the 3 months prior to PROSE evaluation or keratoplasty. We also recorded habitually corrected visual acuity (HCVA), defined as the visual acuity in correction that a patient utilized on a regular basis. This term was defined since some patients with corneal ectasia can refract better with spectacle correction but cannot tolerate the full prescription, or can achieve improved acuity with a contact lens but are unable to wear it because of comfort or fit.
Topographic indices including average central keratometry, and simulated minimum and maximum keratometry as measured by Humphrey Atlas 993 corneal topographer (Carl Zeiss Meditec, Jena, Germany), Orbscan IIz (Bausch and Lomb, Rochester, New York, USA), or Pentacam HR (Oculus, Wetzlar, Germany) at the time of clinical evaluation, were recorded. Apical corneal thickness as assessed by ultrasound pachymetry (Pachette 2; DGH Technology Inc, Exton, Pennsylvania, USA), Orbscan IIz, or Pentacam HR, was also recorded.
In order to assess the severity of corneal ectasia, we used the Amsler-Krumeich classification ( Table 1 ), which uses the refraction, clinical findings on slit lamp examination, corneal curvature, and corneal thickness. For eyes that underwent successful PROSE fitting, we recorded the number of visits and the length of time needed to achieve a fit, as well as visual acuity after completion of the fitting process and at 1 year of follow-up. Assessment of continued wear was determined at the 1 year follow-up visit by patient-reported number of hours of average wear time per day. Complications related to PROSE wear were recorded. For eyes in the keratoplasty group, complications as well as CDVA with method to achieve acuity at 3, 6, 9, and 12 months postoperatively were recorded.
| Stage 1 |
| Eccentric steepening |
| Myopia, induced astigmatism, or both <5.00 D |
| Mean central K reading <48.00 D |
| Stage 2 |
| Myopia, induced astigmatism, or both from 5.00 to 8.00 D |
| Mean central K readings <53.00 D |
| Absence of scarring |
| Minimum corneal thickness >400 μm |
| Stage 3 |
| Myopia, induced astigmatism, or both from 8.00 to 10.00 D |
| Mean central K readings >53.00 D |
| Absence of scarring |
| Minimum corneal thickness 300–400 μm |
| Stage 4 |
| Refraction not measurable |
| Mean central K readings >55.00 D |
| Central corneal scarring |
| Minimum corneal thickness 200 μm |
Comparisons between the demographics, ocular characteristics, and visual outcomes of both groups were made using 2-tailed t tests and Fisher exact test. Snellen acuity was converted to logarithm of the minimal angle of resolution (logMAR) for statistical analysis. All statistical analyses were performed with a Microsoft Office Excel statistical package (Redmond, Washington, USA) with a P value <.05 considered significant.
Results
We included 36 eyes of 36 patients that underwent PROSE evaluation and 37 eyes of 37 patients that underwent keratoplasty. For eyes in the PROSE group, 34 eyes had a clinical diagnosis of KCN or pellucid marginal degeneration and 2 eyes had a diagnosis of post-LASIK ectasia. All 37 eyes in the keratoplasty group had a diagnosis of KCN or pellucid marginal degeneration ( P = .146); 34 eyes underwent PKP and 3 eyes underwent DALK. All eyes in the PROSE group were able to be successfully fitted with the device.
In both the PROSE and keratoplasty cohort, a majority had tried contact lenses to improve visual acuity and function. For eyes in the PROSE cohort, all had tried and failed a rigid gas-permeable contact lens (RGP), 7 (19.4%) had tried and failed a soft contact lens, 13 (36.1%) a piggyback lens, 14 (38.9%) a hybrid lens, and 10 (27.8%) a commercially available scleral lens. For eyes that underwent keratoplasty, 20 (54.1%) had tried and failed an RGP, 3 (8.1%) a hybrid lens, and 2 (5.4%) a commercially available scleral lens. Twelve eyes (32.4%) in the keratoplasty group were not fitted for a contact lens, or it was unknown if they tried a contact lens. None had tried and failed the PROSE device; 2 patients were referred for PROSE but declined fitting and elected to proceed with keratoplasty, 1 patient was evaluated for PROSE but declined to proceed with fitting despite adequate visual rehabilitation, and 1 patient did not undergo PROSE evaluation owing to lack of insurance coverage.
Demographic information, ocular characteristics including keratometric parameters, and baseline visual acuity are displayed in Table 2 . There were no significant differences between eyes in the keratoplasty cohort and those in the PROSE cohort.
| PROSE (N = 36) | Keratoplasty (N = 37) | P Value | |
|---|---|---|---|
| Age (y) | 42.0 ± 13.8 | 39.3 ± 14.2 | .404 |
| Sex (female:male) | 11:25 | 11:26 | .939 |
| Average central K (D) | 56.49 ± 8.92 (n = 30) | 53.65 ± 5.98 (n = 22) | .202 |
| Simulated minimum K (D) | 53.16 ± 10.75 (n = 26) | 47.09 ± 5.83 (n = 9) | .119 |
| Simulated maximum K (D) | 59.75 ± 9.37 (n = 26) | 54.68 ± 7.10 (n = 9) | .151 |
| Keratometric astigmatism (D) | 6.83 ± 4.32 (n = 26) | 7.59 ± 2.63 (n = 9) | .623 |
| Central corneal thickness (μm) | 426.7 ± 103.4 (n = 28) | 464.4 ± 86.1 (n = 21) | .185 |
| Manifest refraction sphere (D) | −8.91 ± 6.12 (n = 33) | −10.30 ± 6.66 (n = 27) | .406 |
| Manifest refraction astigmatism (D) | 5.12 ± 3.01 (n = 33) | 5.69 ± 2.99 (n = 27) | .472 |
| Baseline CDVA (logMAR) | 0.711 ± 0.491 (20/103 Snellen) | 0.886 ± 0.487 (20/154 Snellen) | .131 |
| Baseline HCVA (logMAR) | 1.031 ± 0.642 (20/215 Snellen) | 0.963 ± 0.484 (20/184 Snellen) | .615 |
Table 3 displays the Amsler-Krumeich classification for all eyes. Patients in the keratoplasty group were more severe in distribution than eyes in the PROSE group ( P = .038), with all eyes in the keratoplasty group classified as stage 3 or 4. Stage 4 ectasia was present in 25 eyes (69.4%) in the PROSE group and 33 eyes (89.2%) in the keratoplasty group.
| PROSE (N = 36) | Keratoplasty (N = 37) | P Value | |
|---|---|---|---|
| Amsler-Krumeich stage, n (%) | .038 | ||
| Stage 1 | 0 | 0 | |
| Stage 2 | 5 (13.9%) | 0 | |
| Stage 3 | 6 (16.7%) | 4 (10.8%) | |
| Stage 4 | 25 (69.4%) | 33 (89.2%) |
Visual acuity improved in all eyes after PROSE fitting or keratoplasty ( Figures 1 and 2 ), except for 1 eye with light perception vision after keratoplasty that was not included in any analyses of visual acuity (see complications below). Visual acuity was significantly better immediately after PROSE fitting compared to keratoplasty when including all eyes ( P < .0001), as well as only eyes with stage 4 ectasia ( P < .001). Considering only eyes with mean keratometry readings >60 diopters (D) (n = 10), mean visual acuity was 20/25 after PROSE fitting.
Figure 3 displays the percentage of all eyes, as well as stage 4 eyes only, that achieved 20/40 or 20/25 Snellen acuity. There were no significant differences in the percentage of eyes that attained 20/40 acuity between the PROSE and keratoplasty cohort. However, the PROSE group had a higher percentage of eyes with 20/25 acuity when considering all eyes, as well as those with stage 4 ectasia ( P = .001 and P = .003, respectively).
Figure 4 displays the visual acuity over time after PROSE evaluation or keratoplasty. After keratoplasty, mean acuity continued to improve until the 12-month follow-up visit. The mean spherical equivalent spectacle correction was −1.68 D 1 year after keratoplasty, with 3.04 D of refractive astigmatism; 10 patients (27%) required an RGP to achieve their best acuity. PROSE device fitting took 2.38 ± 1.41 months and 6.49 ± 2.13 visits, at which time best acuity was achieved. One year of follow-up was available for 34 patients fitted for PROSE; of those, 5 patients had discontinued device wear. Two discontinued wear because of unsatisfactory vision, 1 had difficulty inserting and removing the device, 1 had irregular sleeping habits that prevented device wear, and 1 had acute corneal hydrops. If the 2 patients who did not return for follow-up were assumed to have discontinued wearing PROSE, the success rate for continued wear at 1 year would be 80.6% (29/36). In the 29 patients that were still wearing PROSE at 1 year, the average wear time was 13.0 ± 3.3 hours per day and the mean acuity was 20/28 ( P = .108 vs keratoplasty), improving to 20/25 with over-refraction ( P = .006 vs keratoplasty). Eight of 10 eyes with mean keratometry >60 D were wearing PROSE at 1 year; mean acuity was 20/35, improving to 20/30 with over-refraction.
