There has been much advancement in the field of cornea and external disease over the last year. This review provides an update of literature published over the past 1 year. This review is not meant to be all-inclusive, but highlights only a selected number of articles listed in PubMed between January 1, 2012, and December 31, 2012, from the following journals: American Journal of Ophthalmology, British Journal of Ophthalmology, Ophthalmology, Investigative Ophthalmology and Visual Science, and Cornea. We included the following topics in our review: corneal transplantation [including anterior lamellar keratoplasty (ALK), penetrating keratoplasty (PK), Descemet stripping automated endothelial keratoplasty (DSAEK), Descemet membrane endothelial keratoplasty (DMEK), Descemet membrane endothelial transfer (DMET)], transplant rejection, ocular surface epithelial transplantation [including limbal epithelial transplantation, cultivated limbal epithelial transplantation, cultivated oral mucosal epithelial transplantation (COMET)], keratoprosthesis (KPro), infectious keratitis, cross-linking, keratoconus, corneal neovascularization (CNV), and advances in corneal imaging, namely, optical coherence tomography (OCT), Pentacam Scheimpflug imaging, and in vivo confocal microscopy (IVCM).
Findings from some of the most clinically relevant articles in our review this year include the following: thinner DSAEK grafts (<120 µm) have better visual outcomes than thicker grafts (≥200 µm); however, graft thickness (GT) accounts for only 5% of the visual outcome.1 In glaucoma patients undergoing DSAEK, prior glaucoma surgery is a significant risk factor for endothelial failure,2 and hypotony in these eyes has been found to have an increased rate of graft dislocation after DSAEK.3 In DMEK, the 1-hour anterior segment OCT (ASOCT) has a good predictive value on the 6-month graft adherence status and may be used to facilitate decision making regarding surgical intervention on graft dislocation after DMEK.4 From a comparative case series, DMEK has a significantly reduced risk of rejection compared with DSAEK and PK.5 A new classification method of differentiating eyes with keratoconus or subclinical keratoconus from normal eyes with high precision, by applying a support vector machine analysis to corneal indices obtained by Scheimpflug camera with Placido corneal topography, has been described.6 Densitometry could be objectively used to monitor infective keratitis and treatment response,7 and IVCM could enhance the early detection of herpetic stromal keratitis (HSK) recurrence.8
Deep Anterior Lamellar Keratoplasty
There is currently an increasing trend toward ALK. Anterior lamellar keratoplasty has the major advantage of retention of healthy endothelium, while replacing epithelium and corneal stroma. The studies in this area have focused either on visual outcomes or surgical techniques and collectively show improved outcomes compared with PK.
Deep Anterior Lamellar Keratoplasty Outcomes
San et al9 conducted a prospective, randomized study comparing the outcomes of PK (75 eyes) with deep anterior lamellar keratoplasty (DALK) (99 eyes; 88 big-bubble technique, 11 manual dissection) in patients with keratoconus. There was comparable best spectacle-corrected visual acuity (BSCVA) (≥20/40 in 85% in PK vs 83% in DALK, P > 0.05), mean spherical equivalent (SE) [−1.5 diopters (D) in PK vs −2.25 D in DALK, P = 0.08], maximum keratometry (Kmax) (46.85 D in PK vs 46.9 D in DALK, P = 0.66), higher-order aberrations (HOAs) (P > 0.05), and contrast sensitivity function under photopic conditions (P > 0.05). Contrast sensitivity function was higher in DALK under mesopic conditions (P = 0.01) due to the smooth interface in the periphery of DALK eyes compared with PK eyes.
Similarly, Borderie et al10 compared the outcomes of 142 DALK eyes (46% big-bubble technique, 54% manual dissection) with 142 matched PK eyes in patients with various corneal diseases. Long-term graft survival and endothelial cell densities (ECDs) were higher after DALK (median graft survival was 49 years in DALK vs 17.3 years in PK, P < 0.0001; mean 5-year endothelial cell loss was 22.3% in DALK vs 50.1% in PK, P < 0.0001). Overall BSCVA was similar comparing DALK with PK; however, the manual dissection subgroup had significantly lower visual acuity (VA) than big-bubble subgroup (average difference, 2.2–2.5 lines) and PK (average difference, 1.0–1.8 lines). The average central corneal thickness (CCT) at 12 months was significantly higher in manual dissection subgroup compared with big-bubble subgroup and PK group (P < 0.001).
Sarnicola et al11 reported high graft survival rate and stable ECD after DALK in a large noncomparative series of 660 eyes, with mean follow-up duration of 4.5 years. The mean graft survival rate was 99.3% (range, 98.5%–100%); average endothelial cell loss was 11% (range, 10%–12%), which occurred mainly during the first 6 months postoperatively, and stabilized after.
Lyall et al12 studied the long-term outcomes of DALK performed for herpes simplex virus (HSV) keratitis-related corneal scarring in 18 eyes, with mean follow-up duration of 56 months. There was a high percentage of postoperative complications; 33% had HSV keratitis recurrence despite a 12-month course of oral acyclovir, 50% developed either stromal or epithelial graft rejection, and 28% developed graft failure. Multivariate binary logistic regression did not find a correlation between graft failure and previous rejection, HSV recurrence, and any postoperative complication. The presence of neovascularization was not found to be predictive of recurrent infection, rejection, or failure. The majority of the eyes required a second surgery.
Nanavaty and Daya13 demonstrated good outcomes following DALK in eyes with previous hydrops in a series of 10 eyes. Deep anterior lamellar keratoplasty was performed using a modified Melles technique of optical recognition with pre–Descemet membrane dissection. Sixty percent had microperforations at the hydrops site that was managed with intracameral air injection; none required conversion to PK. Best spectacle-corrected VA improved from 6/24 or less preoperatively to 6/12 or greater in all eyes at the last follow-up (mean follow-up of 56.4 ± 23.8 months); there was no difference in BSCVA of eyes with or without microperforations. The mean SE was −2.4 ± 4.2 D, keratometric astigmatism was 3.8 ± 1.6 D, and mean keratometry (K) was 43.7 ± 2.6 D.
In a prospective interventional study, Huang et al14 investigated the clinical outcomes of big-bubble DALK in 131 eyes with various corneal diseases. Overall, big bubble was achieved in 65.6%; in 11.5%, DALK was completed with manual dissection; 22.9% had conversion to PK due to Descemet membrane perforation. There was a higher success of big-bubble formation in eyes with advanced keratoconus (80.6%), chemical or thermal burns (73.3%), corneal dystrophy (71.4%), and HSV scar (70%), compared with moderate keratoconus (36.4%) and bacterial keratitis scar (31.3%) (P < 0.05). The factors influencing the different rate of big-bubble formation in various diagnoses were not clear. Best spectacle-corrected VA was similar among all eyes regardless of the diagnosis at a mean follow-up of 21.4 months. Endothelial cell density was lower in cases with stromal rejection (P < 0.05) and multiple attempts at intrastromal air injection (P < 0.05). The rest of the eyes had a small decrease in ECD up to 6 months, which stabilized after.
Bhatt et al15 compared the outcomes of big-bubble DALK (25/46 eyes, 54.3%) and failed big bubble with completion of DALK by manual dissection (21/46 eyes, 45.7%) in a retrospective series. Pachymetry was higher in the manual dissection group (mean, 628.9 vs 564.15 µm, P < 0.0005). Otherwise, there was no significant difference in the other outcome measures between the 2 groups [best-corrected VA (BCVA), SE, interface densitometry, and contrast sensitivity]. Their results seemed to suggest that residual stromal thickness of 65 µm could result in comparable outcomes with Descemet membrane baring technique.
In a randomized, double-blind trial, Baradaran-Rafu et al16 compared the outcomes of 2 DALK techniques (Anwar technique vs Melles technique, 23 vs 25 eyes, respectively). At 15 months after DALK, the 2 groups had comparable corrected distance VA (CDVA) [0.17 ± 0.09 logarithm of the minimum angle of resolution (logMAR) in Anwar group, 0.18 ± 0.11 logMAR in Melles group, P = 0.803], SE (−1.82 ± 2.7 D in Anwar group, −2.69 ± 3.94 D in Melles group, P = 0.155), HOA (P > 0.05), corneal hysteresis (CH) (P = 0.606), corneal resistance factor (CRF) (P = 0.509), endothelial cell loss (1% ± 2% in Anwar group, 1% ± 3% in Melles group, P = 0.869), and CCT (P = 0.155). However, the Anwar group showed better contrast sensitivity (P < 0.05).
Deep Anterior Lamellar Keratoplasty Techniques
The big-bubble technique with Descemet membrane baring is surgically complex and not easily reproducible. Various techniques have been described to aid the achievement of the big-bubble. Riss et al17 described Pentacam-based (Oculus, Wetzlar, Germany) big-bubble DALK, in which initial trephination to a depth of 90% was performed based on preoperative Pentacam imaging at that position, before attempting the big bubble. They performed the technique in 50 eyes with keratoconus, keratoglobus, and anterior stromal scars and achieved successful big bubble in 68%; 16% were converted to PK because of macroperforations.
Ghanem and Ghanem18 described “pachy-bubble” DALK that used intraoperative pachymetry to assist in big-bubble formation. They performed trephination to 400-µm depth, performed ultrasound pachymetry (AccuPach VI; Accutome, Malvern, Pa) at 0.8 mm internally from the groove at 11- to 1-o’clock position, and created a 2-mm incision with a micrometer diamond knife calibrated to 90% depth of the thinnest measurement. They then inserted a cannula into the tunnel and injected air. They performed this technique on 34 eyes and achieved successful big-bubble formation in 88.2%. Deep anterior lamellar keratoplasty was completed by manual dissection in 5.9%, and 5.9% of the eyes were converted to PK because of macroperforations. Microperforations occurred in 14.7% of the eyes.
Buzzonetti et al19 proposed a femtosecond laser-assisted big-bubble DALK technique and reported the early results in 5 pediatric eyes. An IntraLase 60-kHz femtosecond laser (Abbott Medical Optics, Inc., Santa Ana, CA) was used to create mushroom incisions on both donor and recipient corneas, down to 100 µm above the thinnest site in the recipient cornea. The recipient lamella was then removed, a cannula inserted into the residual stroma, and air was injected. The big bubble was achieved in all the eyes, and there were no intraoperative complications. The outcomes at 10 months were good, with mean BCVA of 20/30, mean SE of 1.8 ± 1.2 D, mean refractive astigmatism of 1.8 ± 1.4 D, mean K of 46.2 ± 0.8 D, and mean corneal thinnest point of 581 ± 46 µm.
Cleary et al20 investigated the corneal stromal surfaces produced by deep excimer ablation, femtosecond lamellar cuts, and manual dissection in cadaver eyes and evaluated the effect of excimer laser surface smoothing. They found that femtosecond cuts resulted in large concentric ridges that persisted even after excimer smoothing, and excimer ablation with smoothing produced a significantly smoother surface than excimer ablation alone or manual dissection with or without smoothing. With that, they developed a dual laser-assisted lamellar anterior keratoplasty technique, using deep excimer ablation with smoothing to prepare the recipient cornea, with a manually dissected circumferential pocket at the base, and femtosecond laser to prepare the donor in a modified top-hat design with a tapered brim. With laser-assisted lamellar anterior keratoplasty, they achieved smooth graft-host interface, with closely matched wound edge configuration observed on OCT.
Descemet Stripping Automated Endothelial Keratoplasty
Endothelial keratoplasty (EK) has replaced PK as the preferred technique for treating endothelial disease with several new studies evaluating this technique.
Li et al21 reported 3-year visual outcome data after DSAEK, in a series of 108 eyes with Fuchs endothelial dystrophy (FED) or pseudophakic bullous keratopathy. There was a gradual improvement in vision over time. At 3 years, 70.4% achieved BSCVA of 20/25 (vs 6.5% preoperatively and 36.1% at 6 months), and 47.2% achieved BSCVA of 20/20 (vs 0.9% preoperatively and 11.1% at 6 months).
The relationship between GT and VA remains controversial. Terry et al1 retrospectively analyzed 418 eyes after DSAEK and found a weak correlation between preoperative donor thickness and BSCVA (R = 0.236, P < 0.001) that accounted for only 5% of the visual outcome (R2 = 0.056). Thicker grafts of 200 µm or greater had worse visual outcomes than thinner graft of less than 120 µm (mean BSCVA 20/33 in eyes with thicker graft vs 20/25 in eyes with thinner graft, P = 0.006). Visual outcomes were comparable in grafts with thickness between 125 and 199 µm.
In another Descemet stripping EK (DSEK) series of 51 eyes, Shinton et al22 investigated the impact of GT and found no clear association between central GT (moderate correlation at month 6 rs = 0.46, P = 0.041; no other significant correlations at other time points) and total corneal thickness (no significant correlations at all time points) with VA over time.
Ang et al23 compared the endothelial cell loss and graft survival in eyes that have undergone DSAEK (119 eyes) or PK (87 eyes) for FED, pseudophakic, or aphakic bullous keratopathy, in a retrospective cohort study. Descemet stripping automated endothelial keratoplasty resulted in less endothelial cell loss compared with PK (30% ± 22% vs 37% ± 25% at 1 year, P = 0.045; 36% ± 23% vs 45% ± 33% at 2 year, P = 0.018; 39% ± 24% vs 47% ± 28% at 3 year, P = 0.022). Graft survival was comparable between DSAEK and PK (94% vs 90% at 1 year, 88% vs 85% at 2 years, 87% vs 85% at 3 years).
Price et al24 performed a similar prospective, multicenter DSAEK study using donor and recipient criteria similar to the Cornea Donor Study (CDS) and compared the 3-year outcomes of their DSAEKs (173 eyes) with PKs (1101 eyes) from the CDS. The 3-year survival rate was similar between both groups for FED (96% for both, P = 0.81) and other moderate risk conditions (86% in DSAEK and 84% in PK, P > 0.05). In contrast to the findings by Ang et al,23 there was similar 3-year endothelial cell loss for both groups for FED (46% in DSAEK and 51% in PK, P = 0.33) and other conditions (59% in DSAEK and 61% in PK, P = 0.70). The 3-year probability of a rejection episode was significantly lower in DSAEK compared with PK (9% vs 20%, P = 0.0005). A smaller 3.2-mm DSAEK incision had a significantly higher cell loss compared with a 5-mm incision (60% vs 33%, P = 0.0007) at 3 years, but there was no significant difference in graft survival (P = 0.45).
Li et al25 evaluated the effects of graft rejection after DSAEK on long-term ECD decline and graft survival in 615 eyes with FED. The rejection rate was 7.3%, and the majority of rejections occurred between postoperative months 12 and 18, which corresponded to the tapering of corticosteroid eyedrops. The cumulative probability of graft rejection was 6% after 1 year, 10% after 2 years, and 17% after 4 years. There were no cases of endothelial failure. There was a higher decline in ECD in eyes with rejection episodes compared with eyes without at all time points (P < 0.05 at 2 and 3 years).
Anshu et al2 studied the long-term graft survival and risk factors for graft failure after DSEK in eyes with preexisting glaucoma in a retrospective case-control series of 453 cases [342 without prior glaucoma (C), 65 with preexisting medically managed glaucoma (G), and 46 with prior glaucoma surgery (GS)]. The 5-year graft survival rate was 96% in group C, 90% in group G, and 48% in group GS (P < 0.001); the 5-year survival rates were 25% in eyes with a glaucoma drainage device and 59% in eyes with trabeculectomy. Prior glaucoma surgery (P < 0.0001) and a prior rejection episode (P = 0.0023) were found to be significant risk factors for endothelial failure using multivariate analysis.
Goshe et al3 compared the dislocation rates of DSAEK eyes with previous glaucoma surgery (67 eyes) and those without (787 eyes). They found an increased rate of graft dislocation in eyes with previous glaucoma surgery (9% vs 2%, P = 0.008), which was correlated with postoperative hypotony (in eyes with dislocation, 83% with glaucoma surgery had hypotony vs 9.8% without glaucoma surgery had hypotony; P < 0.001). Measures should be taken to avoid post-DSAEK hypotony, especially in eyes with previous glaucoma surgery.
Interface abnormalities can be a problem in EK. Arundhati et al26 described retained viscoelastic as a cause of reticular interface haze in 5 patients after combined DSEK/phacoemulsification with intraocular lens (IOL) implantation. Other associated signs included graft detachment without overlying corneal edema and a hyperreflective area on ASOCT. Three patients were observed with eventual clearing of the haze; 2 underwent irrigation and aspiration of the interface, with quick resolution of the haze.
Bhogal et al27 investigated the effects of adhesion promoting surgical techniques in DSAEK using an OCT-based model. They showed that interface fluid diminished with time during tamponade at low and high pressures (P < 0.0001), but there was incomplete removal of interface fluid in all cases; increasing the anterior chamber pressure or tamponade duration had no significant effect on graft adhesion (P = >0.05). Removal of interface fluid (P = 0.0001) and stromal roughening (P = 0.0034) resulted in an increase in adhesion strength. Extended intraoperative air tamponade, venting incisions to allow interface fluid egress, and stromal roughening could be used to promote graft adhesion in DSAEK.
In a prospective cohort and cross-sectional study, Patel et al28 compared the anterior corneal HOAs after DSEK for FED with those of age-matched controls. Anterior corneal HOAs, which might contribute to decreased postoperative BCVA, were higher in FED than in controls (P < 0.001) and remained higher 2 years after DSEK. Over a 4-mm optical zone, total HOAs were correlated with recipient age at 12 (r = 0.47; P = 0.01) and 24 months (r = 0.59; P < 0.001), BCVA at 12 (r = 0.30; P = 0.04) and 24 months (r = 0.59; P < 0.001), and subepithelial haze at 24 months (r = 0.41; P = 0.01).
Biggelaar et al29 evaluated the cost-effectiveness of PK (40 eyes), femtosecond laser-assisted DSEK (36 eyes), and DSAEK (42 eyes) at 1 year postoperatively, based on data from a randomized multicenter clinical trial (the Dutch Lamellar Corneal Transplantation Study) and a noncomparative prospective study. The 3 outcome measures for effectiveness were BSCVA improvement, refractive astigmatism of 3 D or less, and improved vision-related quality of life. They found that femtosecond laser-assisted DSEK was the least cost-effective. Descemet stripping automated endothelial keratoplasty was more costly, but more effective compared with PK, with an incremental cost of €4975 per additional clinically improved patient. Depending on the threshold value of cost-effectiveness, DSAEK or PK could be the preferred technique. However, the study provided an indication of the cost-effectiveness only for the short term.
In a longer-term study, Bose et al30 quantified the cost-effectiveness of DSEK and PK for the treatment of FED or bullous keratopathy in a tertiary eye center in Singapore. They studied the increase in quality-adjusted life-years (QALYs) based on BSCVA improvement and the incremental cost-effectiveness ratio 3 years after surgery for 93 DSEK eyes and 171 PK eyes. The adjusted gain in utility relative to no intervention was marginal for PK (0.128 QALYs) compared with DSEK (0.174 QALYs). The average cost-effectiveness ratio relative to no surgical intervention was $56,409 per QALY for PK and $42,904 per QALY for DSEK; the incremental cost-effectiveness ratio for DSEK relative to PK was $5209 per QALY. From a cost-effectiveness perspective, DSEK was the preferred surgery.
There is a shortage of donor corneas, and many countries rely on obtaining tissue from overseas eye banks. Yamazoe et al31 investigated the influence of precutting and long-distance transportation on corneal grafts for DSAEK on 124 donor tissues. They found an acceptable rate of endothelial cell loss associated with precutting (1.75%) and long-distance transportation (3.79%).
Rauen et al32 did a retrospective study to evaluate if precut donor for EK was associated with a change in the prevalence of bacterial or fungal donor rim cultures. They compared 351 “no-cut” donor rim cultures with 278 precut donor rim cultures and found no increased bacterial (8.5% positive culture in no-cut donor vs 4.7% in precut donor, P = 0.05) or fungal (2.3% positive culture in no-cut donor vs 2.5% in precut donor, P = 1.0) contamination in precut donors.
Ultrathin DSAEK may produce similar visual outcomes as DMEK but with the benefits of greater ease of preparation and handling. A major limitation of microkeratome dissection is its poor accuracy in determining the thickness of the dissected tissue. Busin et al33 compared 3 microkeratome-assisted techniques for the preparation of ultrathin DSAEK grafts in 40 cadaver eyes. After initial debulking with a 300-µm microkeratome head, the corneas were divided into 4 groups; group A had manual intrastromal hydration before a 130-µm microkeratome head was used to perform the second cut; group B had osmotic hydration in tissue culture medium for 24 hours before the second cut; group C had a second cut without stromal hydration with either 90-µm microkeratome (if CCT >150 µm) or 50-µm microkeratome (if CCT ≤150 µm); and group D was the control group. All grafts were prepared successfully without perforation; however, there were multiple areas of Descemet detachment in 40% of corneas in group A. There was no significant endothelial cell loss in all groups. Hence, ultrathin DSAEK grafts could be prepared with the described methods for groups B and C.
Femtosecond laser could create more reproducible and thinner grafts compared with microkeratome; however, there are issues with the smoothness of the stromal interface. To address this, Rousseau et al34 tested various precutting profiles with the IntraLase femtosecond laser on cadaver eyes. They found that the 60-kHz laser with double-layer pattern with 2 successive full lamellar cuts (9.0-mm diameter, 350-µm depth, 2.1-µJ energy, and spot size/step 4:4 µm, then 8.3-mm diameter, 150-µm depth, 0.9-µJ energy, and spot size/step 4:4 µm) created the smoothest interface without stromal adherence comparable to laser in situ keratomileusis flap on scanning electron microscopy.
Descemet Membrane Endothelial Keratoplasty
In DMEK, only Descemet membrane and endothelium are transplanted, without the additional stroma as in DSEK/DSAEK. Rudolph et al35 compared visual outcomes and HOAs in 30 DMEK eyes, 20 DSAEK eyes, 20 PK eyes, and 20 control eyes. The mean BSCVA in the DMEK group at 6 months was significantly better than in the DSAEK group (P < 0.001) and PK group (P = 0.005) and was not different from the control group (P = 0.998). Anterior corneal HOAs in the central 4-mm zone were similar in the DMEK, DSAEK, and control groups; but were higher in the PK group (P < 0.05). Posterior corneal HOAs in the central 4-mm zone were similar in the DMEK and control groups but were higher in the DSAEK and PK groups (P < 0.05).
In a retrospective, single-center case series, Laaser et al36 evaluated the outcomes of triple DMEK in 61 eyes. They found that BCVA improvement, ECD loss, and complication rate were similar to DMEK alone. There was significant visual improvement (BCVA 0.6 ± 0.23 logMAR preoperatively vs 0.19 ± 0.22 logMAR at 6 months, P < 0.05). There was significant ECD loss at 1 month postoperatively (ECD 2573 ± 235 cells/mm2 preoperatively vs 1507 ± 213 cells/mm2 at 1 month, P < 0.005), which stabilized with mean ECD loss of 40% at 6 months. There was a small hyperopic shift (mean SE, −0.3 ± 1.1 D preoperatively vs 0.2 ± 1.1 D at 1 month, P < 0.05; vs 0.5 ± 1.8 D at 3 months, P < 0.05; vs 0.9 ± 1.5 D at 6 months, P > 0.05) and significant refractive cylinder change at 1 month (−0.9 ± 1.0 D preoperatively vs −1.5 ± 1.4 D at 1 month, P < 0.05).
Graft detachment is a major complication after DMEK. Yeh et al4 used ASOCT to predict graft attachment status at 6 months after DMEK. They performed serial ASOCTs on 87 eyes at 1 hour, 1 week, 3 months, and 6 months after DMEK. Anterior segment OCT at 1 hour had the best predictive value for graft attachment with positive predictive value of 75% and negative predictive value of 96%; 90% of grafts with complete attachment or less than one third detachment at 1 hour remained stable or improved in 6 months; 75% of grafts with more than one third detachment at 1 hour showed persistent detachment in 6 months. One hundred percent of graft attachment at 1 week correlated with graft attachment at 6 months.
Descemet Membrane Endothelial Transfer
Dirisamer et al37 reported a case of corneal clearance despite complete graft detachment after DMEK. They postulated that re-endothelialization of the recipient posterior stroma occurred by cells transfer from the detached Descemet roll in the anterior chamber and named that DMET. Corneal edema mostly resolved by 6 months, and endothelial cells were first detected with specular microscopy at 2 months, and ECD measured 830 cells/mm2 at 6 months.
The same authors further investigated DMET in a prospective study38 on 7 eyes with FED and 5 with aphakic or pseudophakic bullous keratopathy. Fuchs endothelial dystrophy eyes showed progressive corneal clearance at 3 to 6 months after DMET, with mean ECD of 797 ± 743 cells/mm2 at 6 months. However, eyes with bullous keratopathy showed no improvement, and endothelial cells could not be detected. They postulated that DMET would not have an effect on eyes with an absolute deficiency of endothelial cells; for FED, the remaining rim of host endothelium contributed to the restoration of corneal clarity.
Shah et al39 reported a case of bilateral endothelial dysfunction secondary to histology-proven combined FED and posterior polymorphous membrane dystrophy in the same eye, in which repopulation of central endothelial cells following Descemet membrane stripping without endothelium replacement was observed by serial confocal microscopy, with improved VA and spontaneous corneal clearing.
To identify risk factors for failure in tectonic keratoplasty, Ang et al40 retrospectively analyzed data from the Singapore Corneal Transplant Study of 362 eyes that underwent tectonic keratoplasty (39.2% PK, 35.1% ALK, 25.7% corneoscleral patch graft). Risk factors for tectonic failure were severe lid disease [odds ratio (OR), 6.1%; 95% confidence interval (CI), 1.7–22.1; P = 0.006], central ALK (OR, 7.5; 95% CI, 1.8–32.4; P = 0.007), and peripheral grafts (OR, 5.7; 95% CI, 1.1–28.3; P = 0.035). Risk factors for physiologic graft failures were active corneal inflammation [hazard ratio (HR), 2.5; 95% CI, 1.4–4.4; P = 0.003] and graft sizes of 9 mm or greater (HR, 17.9; 95% CI, 2.3–140.3; P = 0.006). The mean physiological graft survival was 96 months (95% CI, 83–110 months); ALK eyes had better physiologic survival probabilities (10-year survival was 66.8% for ALK and 44.2% for PK).
Femtosecond lasers are capable of creating circular corneal trephinations and multiplanar keratoplasty incisions, which potentially increase graft-host interface surface area and fit. Angunawela et al41 compared manual suction trephination and femtosecond laser trephination for PK in rabbits’ eyes and cadaver corneas. They found that femtosecond laser trephination resulted in a more precise incision with less graft undercutting (mean bevel angle was 16.72 vs 46.86 degrees with manual trephination, P < 0.0001), a more constant intraocular pressure (IOP) during cutting, and less endothelial cell damage (cell loss of 4516 vs 13548–18064 with manual trephination, P < 0.0001).
Gaster et al42 compared the outcomes of 66 keratoconic eyes that underwent femtosecond laser-enable keratoplasty (FLEK) with zigzag incision and 71 keratoconic eyes that underwent mechanical trephination PK in a retrospective series. The FLEK group had faster visual recovery and better outcomes (BSCVA at 3 months ≥20/40 in 42.42% of FLEK eyes vs 9.7% of PK eyes, P = 0.001; lower topographic astigmatism at 3 months in FLEK eyes; P = 0.001). Postoperative complications were similar in both groups.
In a study on cadaver eyes, Maier et al43 examined the effect of femtosecond laser top-hat keratoplasties with different overlaps (0–3 mm) between the anterior and posterior trephination edges on wound closure and wound stability. They found a significant increase in wound stability for all top-hat trephinations compared with straight trephination (P < 0.001). There was no difference when the different overlaps were compared.
Graft refractive surgery (GRS) with relaxing incisions, with or without counter-quadrant compression sutures, has been effectively used to reduce post-PK astigmatism. In a series of 78 eyes, Feizi and Javadi44 studied the change in graft steepness after GRS and assessed the factors associated with the keratometric change. There was a 0.86-D mean increase in K after GRS (P < 0.001). Keratometric coupling ratio was associated with age (R2 = 0.53, P = 0.04) and preoperative K (R2 = 0.61, P = 0.02); the older the patient, and the lower the preoperative K, the higher the myopic shift. Total arc length, achieved correction in vector keratometric astigmatism, number of incisions, and use of compression sutures did not show any association with keratometric coupling ratio.
In a retrospective series, Arundhati et al5 evaluated the risk of rejection in DMEK (141 eyes), DSEK (598 eyes), and PK (30 eyes). They showed that DMEK had a reduced rejection risk compared with DSEK or PK (Kaplan-Meier cumulative probability of a rejection episode at 1 year and 2 years was 1% and 1% for DMEK, 8% and 12% for DSEK, and 14% and 18% for PK; P = 0.004). Descemet membrane endothelial keratoplasty eyes had a 15 times lower risk of rejection than DSEK eyes (95% CI, 2.0–111; P = 0.008) and 20 times lower risk than PK eyes (95% CI, 2.4–166; P = 0.006).
The CDS was a prospective, multicenter, double-masked clinical trial that studied the effect of donor age on the outcomes of PK in patients with moderate-risk. Stulting et al45 assessed the associations between donor and recipient factors and allograft rejection in 1090 eyes from the CDS. Over 5 years, 23% had at least 1 rejection event; the 5-year predicted probability of rejection was 26% ± 3%. Thirty-seven percent of these eyes had graft failure compared with 5% of eyes without rejection (HR, 15.03, 95% CI, 10.34–21.83; P < 0.001). They found a higher graft rejection rate in pseudophakic or aphakic corneal edema compared with FED (34 ± 6% vs 22 ± 4%; HR, 1.56, 95% CI, 1.21–2.03), in phakic Fuchs eyes compared with Fuchs eyes with combined cataract extraction with or without IOL implantation (29% vs 19%; HR, 0.54, 95% CI, 0.36–0.82), and in female recipients compared with male recipients (29% vs 21%; HR, 1.42; 95% CI, 1.08–1.87). Graft rejection was not associated with donor age.
Guilbert et al46 also studied the risk factors for graft rejection and graft failure in a retrospective cohort series of 1438 eyes that underwent corneal transplantation for optical reasons (1209 PK, 165 ALK, 64 DSEK). The cumulative incidence of rejection was 8% at 12 months (95% CI, 6.5%–9.4%), 24.3% at 60 months (95% CI, 21.5%–27%), and 30.1% at 120 months (95% CI, 25.8%–34.4%). Multivariate analysis showed a higher incidence of rejection in patients younger than 20 years and between 41 and 50 years old (P = 0.00002), high-risk eyes with 2 or more quadrants of corneal vascularization or history of irreversible allograft rejection (P = 0.0003), aphakia and eyes with anterior chamber IOL (P = 0.00003), and PK compared with ALK (P = 0.035). Irreversible rejection were more likely to occur in high-risk eyes (P = 0.02), eyes with preoperative hypotony (P = 0.009), PK compared with ALK and DSEK (P = 0.03), eyes with presenting VA worse than 1.30 logMAR (P = 0.002), and diffuse or progressive endothelial rejection and mixed rejection compared with epithelial, subepithelial, and stromal rejection (P = 0.02).
Shimazaki et al47 conducted a prospective trial examining the efficacy and safety of long-term topical corticosteroids after PK and showed that long-term use of topical corticosteroids was beneficial for the prevention of rejection. They randomized 42 patients who were post-PK 1 year into 2 groups, the steroid group (who received topical 0.1% fluorometholone 3 times a day), and no-steroid group. There was a higher incidence of rejection in the no-steroid group compared with the steroid group 12 months later (42.9% vs 5.5%, P = 0.027; Kaplan-Meier analyses log-rank test P = 0.032). There were no significant adverse effects.
Ocular Surface Epithelial Transplantation
Since the development of the corneal epithelial stem cell transplantation, autologous or allogeneic limbal transplantation has been used to reconstruct ocular surfaces. More recently, ex vivo cultivated mucosal epithelial transplantation has been described.
Limbal Epithelial Transplantation
Sangwan et al48 described a single-stage novel technique of limbal transplantation, in 6 patients with unilateral total limbal stem cell deficiency (LSCD). They harvested a 2 × 2-mm strip of donor limbal tissue from the healthy eye, divided the strip into 8 to 10 pieces, and placed them onto a human amniotic membrane (AM), which was secured onto the cornea with fibrin glue. A stable ocular surface was achieved in all eyes by 6 weeks and was maintained over a mean follow-up period of 9.2 ± 1.9 months, with the transplants becoming progressively transparent over time. Best-corrected VA improved from less than 20/200 in all eyes preoperatively to 20/40 or greater in 66.6%. None of the donor eyes developed any complications.
The “Cincinnati procedure” combines a living-related conjunctival-limbal allograft and keratolimbal allograft for eyes with bilateral LSCD. Chan et al49 developed the “modified Cincinnati procedure,” which combined conjunctival-limbal autograft and keratolimbal allograft for the treatment of unilateral LSCD. They performed the procedure on 11 eyes with either chemical and/or thermal burns, with staged PK in 91% of the eyes. That was effective in improving vision (BCVA ≥20/80 in 73% at the final follow-up vs ≤20/400 in all eyes preoperatively) and maintaining long-term ocular surface stability (100% had stable or improved ocular surface). The success rate for PK was 60%. There were no serious adverse events secondary to systemic immunosuppression, and that was tapered off at a mean of 16 months (range, 8–28 months).
Eberwein et al50 sought to evaluate the outcomes of combining AM transplantation and mitomycin C treatment with allogeneic limbokeratoplasty in 20 eyes with severe bilateral LSCD. They found the additional use of AM and mitomycin C to be beneficial for overall graft survival. The success rate was 70% after a follow-up period of 20 months (compared with 40% after 2.5 years after conventional limbokeratoplasty in a previous series by the same authors). There was limbal immunologic rejection in 10% of the eyes and no cases of scleral melt nor delayed wound healing.
Cultivated Limbal Epithelial Transplantation
Basu et al51 reported the long-term outcomes and complications of xeno-free allogeneic (living-related donor) cultivated limbal epithelial transplantation in 28 eyes with bilateral LSCD. A stable ocular surface was maintained in 71.4% with a follow-up of 4.8 ± 2.8 years; the Kaplan-Meier survival rate of allogeneic limbal transplantation was 76.4% ± 8.7% at 1 year, 70.5% ± 8% at 2 years, and 63.9% ± 8.9% at 3 years. Forty-six percent underwent PK; the Kaplan-Meier survival rate for PK was 76.9% ± 11.7% at 1 year, with a median survival of 3.3 years. Limbal transplantation and/or PK was successful in restoring BCVA of 20/60 or greater in 67.8% of the eyes. There were no serious ocular complications.
Cultivated Oral Mucosal Epithelial Transplantation
Hirayama et al52 compared 16 eyes that underwent COMET using substrate-free cell sheets prepared on fibrin-coated dishes with 16 matched eyes that underwent COMET using AM sheets in a case-control study. They found that COMET using substrate-free cells sheets had better clinical outcomes than AM sheets over a mean follow-up duration of 109.8 weeks. In the substrate-free group, 62.5% achieved a successful ocular surface (vs 37.5% in AM group), with significantly better postoperative BCVA at all time points (P < 0.05) and lesser neovascularization (P = 0.023). The graft survival was also significantly better in the substrate-free group with Kaplan-Meier analysis (P = 0.046).
Cultured autologous oral mucosal epithelial cell sheet is a tissue-engineered transparent and rapidly bioadhesive cell sheet, cultured with UpCell-Insert (CellSeed, Inc, Tokyo, Japan). Burillon et al53 performed a prospective, noncomparative, 2-stage clinical trial to evaluate the safety and efficacy of cultured autologous oral mucosal epithelial cell sheet in 26 eyes. Efficacy was measured using a composite criterion based on epithelial defect, punctate epithelial keratopathy, conjunctivalization of the cornea, number of vascular pediculi, and vessel activity. Sixty-five percent was successful after 360 days (all 5 criteria showed improvement or remained stable); 85% showed improvement in at least 1 criterion. There were no serious adverse events related to the product.
Keratoprosthesis or artificial corneas have an important role in cases of multiple failed corneal transplants or severe corneal diseases with high probability of failure with corneal transplant.
Boston type I KPro is the most commonly implanted KPro worldwide. In an international consecutive case series, Aldave et al54 evaluated the outcomes of 113 Boston type I KPro procedures performed in 7 countries outside North America and compared them with 110 procedures performed in the University of California, Los Angeles (UCLA). There were similar VA outcomes in the 2 groups (59% in the international group had CDVA 20/20-20/200 up to 2 years after surgery vs 60% in UCLA group). Retention failure rate was higher in the international group (0.165/eye-year vs 0.100/eye-year). The international group had a higher incidence of endophthalmitis (8.9% vs 1.1%, P = 0.019), whereas the UCLA group had a higher incidence of retroprosthetic membrane (RPM) (48.9% vs 26.7%, P = 0.002), persistent epithelial defect (36.2% vs 9.9%, P < 0.0001), and retinal detachment (16% vs 5%, P = 0.017).
Kang et al55 reported the outcomes of patients who underwent Boston KPro type I (19 eyes) and type II (2 eyes) as a primary procedure, with a follow-up duration of 14.6 months. The most common indications for surgery were chemical or thermal injury, aniridia, and Stevens-Johnson syndrome. At the last follow-up, 71.4% achieved BCVA of 20/200 or greater. The most common complications were RPM formation (47.6%), cystoid macular edema (33.3%), and elevated IOP (23.8%); endophthalmitis occurred in 4.8%. Their retention rate was 90.5%.
Rudnisky et al56 sought to identify risk factors for RPM development in a large multicenter cohort study of 265 eyes that underwent Boston type I KPro; 31.7% developed RPM. Significant risk factors were infectious keratitis as a surgical indication (HR, 3.2; 95% CI, 1.66–6.17) and aniridia (HR, 3.13; 95% CI, 1.1–8.89).
Chan and Holland57 did a retrospective review of infectious keratitis after Boston type I KPro in 126 eyes. The incidence of infectious keratitis was 7.9% (10/126 eyes); 50% were caused by fungus and 20% by bacteria, and 30% were culture-negative. All cases were on topical vancomycin and moxifloxacin prophylaxis, and 2 were on topical amphotericin prophylaxis. Forty percent required KPro removal with therapeutic PK, and 10% had KPro replacement. At the last follow-up, only 20% retained their preinfection BCVA. Risk factors for infectious keratitis included diagnosis of cicatrizing conjunctivitis (P = 0.0003) and persistent epithelial defect (P = 0.0142); contact lens wear, vancomycin use, and systemic immunosuppression were not significant risk factors.
Similarly, Ramchandran et al58 reported their outcomes of infectious endophthalmitis after Boston type I KPro in 141 eyes. The incidence of endophthalmitis was 7.1% (10/141 eyes); 90% were caused by bacteria, and 10% was culture-negative. All cases were on topical fluoroquinolones prophylaxis, and none were on vancomycin. Seventy percent had recurrent endophthalmitis that occurred at a mean of 4 months after resolution of the initial episode. The authors recommended the use of prophylactic vancomycin, as fluoroquinolone was insufficient to prevent endophthalmitis in these eyes.
Robert et al59 reviewed the literature from 2001 onward and reported the prevalence of endophthalmitis following Boston type I KPro was 5.4% over the last 10 years. Gram-positive bacteria were the most common causative microorganisms, whereas gram-negative bacteria and fungi were emerging pathogens. The majority occurred in patients with Stevens-Johnson syndrome, ocular cicatricial pemphigoid, and chemical burns. Other risk factors included periprosthetic cornea infection, or glaucoma drainage device erosion.
Talajic et al60 reported the glaucoma outcomes associated with Boston type I KPro in a series of 38 eyes. Seventy-six percent had preexisting glaucoma, additional 13% developed glaucoma, and 21% had a hypertensive spike in the first 4 weeks. Visual acuity was limited by glaucoma in 37%, and visual fields were limited by glaucoma in 66%. The high incidence of glaucoma progression can compromise visual outcome after KPro, and clinicians should have a low threshold to treat even slightly elevated IOP.
Dokey et al61 described the occurrence of chronic hypotony without an evident cause as a complication following Boston type I KPro in 6 of 68 eyes. The median time to hypotony onset was 18.5 months, and the incidence was 3.7% at 1 year (95% CI, 0.9%–14%) and 13.3% at 2 years (95% CI, 5.5%–30%). An increased risk of chronic hypotony was seen with RPM (P < 0.01) but not with increased age, glaucoma drainage implants, and history of multiple corneal transplants.
VisionGraft Sterile Cornea is a sterile, gamma-irradiated, full-thickness corneal lenticule without viable endothelium that has been precut using a femtosecond laser. Akpek et al62 reported favorable results with its use in Boston type I KPro in 11 eyes, without any donor-related complications over a mean follow-up of 16.5 months.
Robert et al63 did a randomized study that supported the use of frozen corneas as carriers in Boston KPro. Thirty-seven eyes were included in the study; 19 eyes had fresh cornea carriers, and 18 eyes had frozen cornea carriers. Both groups had similar visual outcomes and complications.
Stapleton et al64 conducted a prospective, population-based, case-control study in Australia between 2003 and 2004, encompassing 90 cases of keratitis and 1090 community controls, to establish the risk factors for moderate and severe microbial keratitis among daily contact lens wearers. Independent risk factors identified included poor storage case hygiene 6.4× [population attributable risk (PAR) 49%], infrequent storage case replacement 5.4× (PAR 27%), solution type 7.2× (35%), occasional overnight lens use 6.5× (PAR 23%), high socioeconomic status 4.1× (PAR 31%), and smoking 3.7× (31%).
Lichtinger et al65 reviewed the trends of bacterial keratitis in Toronto from 2000 to 2010. There were 1701 corneal scrapings performed in the 11-year period (mean, 154.6 cases per year); 57.4% were culture-positive, and 91.8% of these were caused by bacteria (76.2% gram-positive isolates, 23.8% gram-negative isolates). There was a decreasing trend of gram-positive bacteria (P < 0.05), with coagulase-negative Staphylococcus being the most common pathogen (48% of gram-positive isolates). Increasing methicillin resistance was seen (28% in the first 4 years vs 38.8% in the last 3 years, P = 0.133); all resistant isolates were found to be sensitive to vancomycin. The increase in gram-negative bacteria (r = 0.7, P = 0.0165) was likely related to the rise in contact lens wear, with Pseudomonas aeruginosa being the most common gram-negative bacteria (43% of gram-negative isolates). The sensitivity of gram-negative isolates was excellent, with greater than 97% response for all tested antimicrobials.
In a 15-year study from 1995 to 2009, Henry et al66 described the clinical outcomes of infective keratitis progressing to endophthalmitis. The endophthalmitis rate was 0.5% (49/9934 eyes). Risk factors included use of topical corticosteroids (76%), fungal keratitis (78%), corneal perforation (35%), and infective keratitis adjacent to a surgical wound (45%). Patients with sequential keratitis and endophthalmitis had poor visual outcomes (only 14% had VA ≥20/50 at last follow-up, 35% had no light perception vision, and 31% underwent evisceration or enucleation).
Goldschmidt et al67 developed a new diagnostic test based on real-time polymerase chain reaction (PCR) high-resolution melting analysis that could detect the equivalent of 0.1 colony-forming units per milliliter of fungus, differentiate filamentous fungi from yeasts, and discriminate among species of yeasts in less than 2.3 hours after DNA extraction. The sensitivity and specificity for this test were 100% for culture-positive specimens. Compared with classic PCR, high-resolution melting has the advantage of minimizing false-positive and false-negative results. Kuo et al68 developed a dot hybridization assay, based on a fungus-specific oligonucleotide probe, with 100% sensitivity and 96.7% specificity for the diagnosis of fungal keratitis. The turnover time for the test was 1 working day.
Ikeda et al69 assessed the diagnostic value of real-time PCR and the prognostic determinants of Acanthamoeba keratitis. Real-time PCR was effective in diagnosing Acanthamoeba (50 times more sensitive than conventional cyst counting, with 100% specificity). There was a strong and positive correlation of Acanthamoeba copy numbers and the infection stage. Logistic regression analysis showed that Acanthamoeba copy numbers (OR per category, 3.48; P < 0.05) and advanced infection stage at the first visit (OR of 2.8 per stage increase; P < 0.05) were the highest risk factors for poor outcomes. A weak reduction in amoebic DNA (<90%) was associated with advanced infection stage (OR of 8.0 per stage, P < 0.05) and previous use of steroids (P < 0.05).
Younger et al70 investigated the utility of corneal biopsy in the evaluation of infective keratitis in a retrospective review of 48 eyes. Corneal biopsies were useful, especially in culture-negative cases (44% had positive biopsy), or were unresponsive to treatment. Culture and histopathologic examinations provided complementary information (65% concordant). Histopathologic examinations were more likely to indicate the causal organism in discordant positive results and more likely to identify fungi and Acanthamoeba.
Treatment and Outcomes
The Steroids for Corneal Ulcers Trial (SCUT) was a randomized controlled, multicenter trial assessing the clinical outcomes of patients with bacterial corneal ulcers who received adjunctive topical corticosteroids versus placebo.
Sy et al71 did a subanalysis of the SCUT data and compared the outcomes of P. aeruginosa ulcers (22%, 110/500 eyes) with other bacterial ulcers. Pseudomonas cases had worse presenting BSCVA (0.24 logMAR difference, P = 0.001), but showed greater improvement with treatment, with no difference in 3-month BSCVA (0.01 logMAR, P = 0.87). There was no difference in adverse events in both groups and no significant benefit or adverse effects with corticosteroid treatment.
Lalitha et al72 also did a subanalysis of the SCUT data, looking at the clinical outcomes of Nocardia keratitis (11%, 55/500 eyes). Nocardia cases had better presenting BSCVA (0.34 logMAR vs 0.86 logMAR in non-Nocardia cases, P < 0.001), but improved less with treatment (BSCVA 0.21 logMAR less improvement, P = 0.001; 0.2-mm less improvement in infiltrate or scar size, P = 0.03). They had worse outcomes with corticosteroids (0.4-mm larger infiltrate or scar size at 3 months, P = 0.03), but there was no difference in 3-month BSCVA, time to re-epithelialization, and perforation between the corticosteroid and placebo arms. Treatment with fluoroquinolones as monotherapy in two thirds of cases resulted in good outcomes (median BSCVA 20/25), despite variable in vitro susceptibility.
Liu et al73 studied the clinical characteristics and outcomes of 57 eyes of 53 patients 16 years or younger, with HSV keratitis and/or HSV blepharoconjunctivitis, seen in the Massachusetts Eye and Ear Infirmary between 2000 and 2009. Sixty-eight percent had keratitis, and 74% had stromal disease. Children with HSV keratitis are at high risk of corneal scarring (79%), recurrence (80%), and vision loss (26% had VA ≤20/40 at the last visit). Oral acyclovir treatment was well tolerated; however, 37% receiving long-term oral acyclovir had recurrent HSV, which could be secondary to a more robust inflammatory reaction to HSV or a suboptimal dosage of acyclovir as the child grows.
The Mycotic Ulcer Treatment Trial Therapeutic Exploratory Study was a randomized, controlled, double-masked clinical trial in South India, investigating whether natamycin or voriconazole resulted in better visual outcomes in fungal keratitis. Lalitha et al74 reported the minimum inhibitory concentration (MIC) of 84 fungal isolates from the trial (52% Fusarium, 20% Aspergillus, 28% others) to natamycin and voriconazole and the correlations with VA, scar size, re-epithelialization time, and perforation. Minimum inhibitory concentration was significantly different for different organisms (P = 0.0001); MIC50 for natamycin was equal or higher than voriconazole for all organisms, and MIC90 for natamycin was higher for all organisms except Fusarium and Bipolaris. They found that a higher MIC was associated with an increased risk of perforation (OR, 2.03; P = 0.04); there was no significant association with BSCVA, infiltrate or scar size, and time to re-epithelialization. Their results suggested that resistance to antifungals (natamycin more than voriconazole) might be associated with worse outcomes.
Intracameral voriconazole may be useful in treating infections caused by fungi with high MIC. Han et al75 investigated the effect of intracameral voriconazole on corneal endothelium in 36 rabbit eyes. Different concentrations of voriconazole (0.03%, 0.1%, 0.25%, 0.5%, and 1%) were injected into 30 eyes. There was no difference in the CCT up to 120 minutes among the groups (P > 0.05) or percentage of dead endothelial cells (P = 0.504). Scanning electron microscopy showed cells borders were blurred in groups receiving voriconazole 0.25% or more, suggestive of cell wall damage with those concentrations of voriconazole.
Corneal Cross-linking in Keratoconus
Vinciguerra et al76 reported their 2-year corneal cross-linking (CXL) results in 40 eyes of 40 patients 18 years or younger with progressive keratoconus. There was a continuous improvement of most indices up to 24 months (simK1/2 and min K). There was also an improvement in refraction and uncorrected VA (UCVA) and BSCVA during the first 6 months with no change in VA after.
Caporossi et al77 analyzed the results of 152 patients 18 years or younger with progressive keratoconus from the Siena CXL Pediatrics trial. They found that CXL was effective in slowing keratoconus progression in pediatric patients, irrespective of preoperative corneal thickness. There was faster functional recovery in patients with lower corneal thickness initially, but at 36 months, there was no significant difference between the 2 groups. There was mean improvement of 0.15 Snellen lines, significant topographic improvement, and coma reduction (P < 0.01).
In a prospective study of 38 eyes with progressive keratoconus, Kymionis et al78 compared the effects of CXL with transepithelial phototherapeutic keratectomy (t-PTK) ablation to a depth of 50 µm (Allegretto Wavelight excimer laser; Wavelight Technologies, Erlangen, Germany) (19 eyes) versus mechanical epithelial debridement (19 eyes). For t-PTK-CXL group, uncorrected distance VA, CDVA, mean corneal astigmatism, and mean steep K all improved. For the other group, there was no significant improvement in the uncorrected distance VA, CDVA, corneal astigmatism, and K. These results were likely due to t-PTK smoothening the anterior corneal stroma, hence decreasing the keratoconus-induced irregular astigmatism. There were no intraoperative complications, and IVCM showed similar postoperative recovery in both groups, without any change to the ECD.
Corneal Cross-linking in Infectious Keratitis
Photoactivated riboflavin is known to cause oxidative damage to microbials. Panda et al79 performed CXL on 7 eyes with infective keratitis that was not responding to medical treatment. Antimicrobial treatment was resumed 2 hours after CXL. In all cases, re-epithelialization was initiated within 48 to 72 hours; complete resolution was achieved between 3 and 5 weeks.
Galperin et al80 evaluated the efficacy of CXL treatment in Fusarium solani keratitis in a rabbit model. Corneas from the CXL group showed less hyphae and inflammatory cells on microbiological and histological assessment on day 7. Corneal cross-linking seemed effective in decreasing the severity of F. solani keratitis.
Sharma et al81 reported persistent corneal edema after CXL secondary to corneal endothelial damage in 2.9% of patients who underwent CXL for progressive keratoconus. Corneal edema persisted for 2 to 3 weeks. Edema resolved in 1 patient and improved in 4; PK was offered to the other 5 patients.
Ghanem et al82 reported 7 cases of ring-like peripheral sterile corneal infiltrates that resolved after a few weeks of treatment with topical corticosteroids. They postulated that the infiltrates could be due to phototoxicity of the corneal stroma or an alteration in protein antigenicity, hence triggering an immune response.
In a prospective study of 72 eyes of 36 patients with progressive keratoconus, Wasilewski et al83 found a significant reduction in corneal sensitivity in eyes that underwent CXL compared with fellow untreated eyes. Corneal sensitivity reduction was greatest in the first week, with partial recovery over 6 months.
Vimalin et al84 studied the effects of CXL on limbal epithelial cells in cadaveric eyes, with one half of the limbus left unprotected (sector A) and the other half covered with a metal shield (sector B). Riboflavin–UV-A treatment caused damage to limbal epithelial cells, including stem cells, and covering the limbal region with a metal shield effectively prevented the damage.
Optimization Model for CXL
Schumacher et al85 developed a theoretical model of the CXL procedure, which predicted an optimal riboflavin concentration for maximizing the increase in corneal stiffness. Further work is required to validate the performance of the model.
In a cross-sectional study, Li et al86 demonstrated the use of RTVue to map the central corneal epithelium thickness. Compared with normal eyes, keratoconic eyes had thinner corneal epithelium inferiorly (P = 0.03), lower minimum thickness (P < 0.0001), greater superior-inferior thickness (P = 0.013), more negative minimum-maximum thickness (P < 0.0001), greater map standard deviation (P < 0.0001), and larger pattern SD (P < 0.0001).
Arbelaez et al6 used a Scheimpflug camera and Placido corneal topography to analyze 877 keratoconic eyes, 426 eyes with subclinical keratoconus, 940 eyes with previous corneal surgery, and 1259 healthy eyes. They found that applying a support vector machine–based algorithm to the corneal indices accurately differentiated normal eyes from eyes with clinical and subclinical keratoconus. The diagnostic accuracy was further improved by including posterior corneal surface and corneal thickness measurements.
Saad and Gatinel87 found that corneal and ocular wavefront analysis enabled the detection of subclinical keratoconus that may not be detected by Placido-based topography. Corneal and ocular tilt, vertical coma, and trefoil were significantly different in the forme fruste keratoconus eyes compared with the normal eyes.
Disease Severity and Progression
Ishii et al88 evaluated the correlation between Pentacam measurements of the anterior and posterior corneal elevations of 86 keratoconic eyes, with keratoconus severity index and Amsler-Krumeich classification. They found that elevation differences had a significant correlation with keratoconus severity index and Amsler-Krumeich classification, and corneal height information was a potential index for defining keratoconus severity. Posterior elevation measurement showed more significant diagnostic value compared with anterior elevation measurement.
In a retrospective study, Leoni-Mesplie et al89 compared the severity of keratoconus at diagnosis and scalability over 2 years. They found that 27.8% of children had stage 4 keratoconus (Krumeich classification) compared with 7.8% adults at diagnosis (P < 0.0001). Keratoconus was not more likely to progress in children. However, in disease progression, the scalability rate was higher in children in terms of SE (P = 0.03), Kmax (P = 0.02), minimum K (P = 0.04), and CRF (P = 0.007).
Choi and Kim90 tracked the longitudinal changes of 94 eyes with mild keratoconus using computerized videokeratography. They found that 25% showed progression, which lasted 3.5 years on average, and the median time from diagnosis to progression was 12 years. Significant predictors of progression included the highest point on the anterior elevation from the anterior best-fit sphere, irregularity index at 3 mm, irregularity index at 5 mm, thinnest pachymetry less than 350 µm at baseline, and yearly change rate of anterior best-fit sphere 0.1 D/y or greater, central K of 0.1 D/y or greater, simulated K in maximum 0.15 D/y or greater, simulated K in minimum 0.2 D/y or greater, and anterior chamber depth (ACD) of 0.0 mm/y or greater. The axial change in the steep meridian from with-the-rule astigmatism toward oblique astigmatism was also associated with keratoconus progression.
Intracorneal Ring Segment
Pena-Garcia et al91 sought to make quantitative predictions of the visual changes after intracorneal ring segment (ICRS) implantation. They divided 58 keratoconic eyes after KeraRing implantation into 2 groups (eyes that gained ≥0.2 in CDVA and eyes that lost >0.15 in CDVA). Significant correlations found were preoperative CDVA, apical K, and a new parameter named K-factor. They predicted that ICRS was very effective for patients with preoperative CDVA of 0.01 to 0.3, and higher values of CDVA were often related to a decrease in CDVA.
Pinero et al92 evaluated the biomechanical changes after femtosecond laser-assisted ICRS KeraRing implantation with the Ocular Response Analyzer. There was an increase in CH at 3 months (P = 0.03) and a decrease in CRF at 6 months (P = 0.02). Corrected distance VA at 1 month was correlated with preoperative difference between CH and CRF and preoperative K. Preoperative biomechanical parameters were correlated with HOA.
The most common method for quantifying CNV is by analyzing photographic images of the cornea. Anijeet et al93 demonstrated in 23 patients that the combined use of fluorescein angiography (FA) and indocyanine green angiography, with image analysis using computer software, provided better vessel delineation than color images. Indocyanine green angiography was able to define CNV most effectively, even for deep vessels in the presence of corneal scarring. Fluorescein angiography was useful to detect leakage, which occurred earlier in CNV of less than 6 months’ duration than those of more than 1-year duration.
In agreement with the above findings, Kirwan et al94 quantified CNV changes in a prospective series of patients with active keratitis. After treatment, there was a reduction in the mean CNV area (P < 0.05) and vessel diameter (P < 0.01), which was more apparent with FA and indocyanine green angiography than color images, and an increase in time to dye leakage on FA.
Cheng et al95 treated 20 eyes with CNV with topical 1.0% bevacizumab for 3 weeks and reported an improvement in CNV area by week 6 and in vessel caliber by week 12. At 24 months, CNV area was reduced by 47.5% and vessel caliber by 36.2%. There were no adverse events.
Benayoun et al96 administered a single subconjunctival injection of bevacizumab to 12 eyes with CNV and found vessel recession at 1 week, which peaked at 1 month, and an increase in CNV thereafter, suggestive of the need for repeat treatment after 1 month. Overall, there was a mean decrease in CNV of 25% at 3 months (P = 0.02). No local or systemic adverse events were observed.
Koenig et al97 evaluated the efficacy and safety of combined feeder vessel coagulation and topical bevacizumab therapy in the treatment of mature CNV in 16 eyes. Nine eyes received an additional subconjunctival injection of bevacizumab at the time of diathermy. Occlusion of all treated vessels without reperfusion occurred in 68.8%, and there was a mean reduction of 46% of CNV. Additional subconjunctival bevacizumab led to a greater CNV reduction.
Sorafenib is an orally active multikinase inhibitor that has been shown to reduce choroidal neovascularization in animal models. Seo et al98 studied the effects of sorafenib in a rat model of CNV and found a reduction of CNV in a dose-dependent manner. At 2 weeks, CNV was reduced by 44% in a low-dose group and by 66% in the high-dose group. Further studies are needed to determine efficacy and safety in humans with CNV.
ADVANCES IN CORNEAL IMAGING
Optical Coherence Tomography
Anterior segment OCT provides an objective and quantitative assessment of the anterior segment. In a population-based study, Ang et al99 analyzed ASOCT images of 438 adult South Asian Indians in Singapore using the ZAP software (Zhongshan Assessment Program) and reported the normative values of ACD, CCT, posterior corneal curvature (PCC), anterior corneal curvature, and posterior corneal arc length (PCAL) of those eyes. Multivariate analysis showed a significant association between PCAL and ACD, and PCC. Tan et al100 did a similar study of 237 adult Malays in Singapore. Multivariate analysis showed a significant association between PCAL and ACD, PCC, and height. Posterior corneal arc length may be used for optimal graft sizing for EK and calculation of the size of an anterior chamber IOL.
Correa-Perez at al101 assessed the reliability of CCT measurements in 77 healthy eyes using Cirrus high-definition OCT and showed reliable intraobserver measurements and consistent agreement between 2 independent examiners. Comparing these to CCT measurements using standard ultrasonic pachymetry, they found a mean difference of −4.5 µm. Cirrus HD-OCT is hence a reliable noncontact pachymeter.
Prakash et al102 went on to study corneal epithelial thickness measurements by Cirrus OCT in a similar series of 210 healthy eyes. They showed that Fourier-domain OCT is reliable and reproducible for epithelial thickness measurements at the vertex
Fukuda et al103 demonstrated the use of Fourier-domain OCT to quantitatively evaluate iridotrabecular contact in 60 eyes after PK. Iridotrabecular contact was seen in 46.7% of the eyes, with a mean area of 5.97 mm2, and mean iridotrabecular contact index of 14%. Iridotrabecular contact was significantly associated with bullous keratopathy, infectious keratitis, pseudophakia, combined open-sky cataract extraction, previous PK, and graft size of 7.75 mm or greater.
Pentacam Scheimpflug Imaging
In a prospective study encompassing 36 eyes with bacterial keratitis and 64 normal eyes, et al7 investigated the use of corneal densitometry measurements with Pentacam in eyes with bacterial keratitis. The mean densitometry value of normal corneas was 12.3 ± 2.4. In infective keratitis, the value was higher, both at the infection site and at the point furthest away from the infection site and both during the active stage and after complete healing beyond 4 weeks. Active infiltrate resulted in a higher value than corneal scar after healing. Densitometry could be objectively used to monitor infective keratitis and treatment response.
In Vivo Confocal Microscopy
Hillenaar et al8 demonstrated the usefulness of supplementary IVCM in the follow-up of HSK as it enhanced early detection of inflammatory activity. They performed slit-lamp biomicroscopy and IVCM on 35 patients with active HSK at diagnosis and after 1, 3, 6, and 12 months and compared the findings to the unaffected fellow eye. On IVCM, disease activity was indicated by the presence of dendriform cells, pseudoguttae, and increased objective corneal backscatter measurement. There were 17 recurrences detected with the use of both slit-lamp biomicroscopy and IVCM; 3 were missed by slit-lamp examination alone, and 6 were missed by IVCM.
In vivo confocal microscopy was also found to be useful in the early detection of silicone oil keratopathy in a study by Le et al.104 In 99 eyes with silicone oil endotamponade after vitreoretinal surgery, cornea abnormalities were identified in only 12.1% with slit-lamp examination and in 40.4% with IVCM. In vivo confocal microscopy findings included decreased endothelium density, increased polymegathism and pleomorphism, and the presence of hyperreflective silicone oil membrane or droplets adhering to the endothelium and in the corneal stroma. Silicone oil keratopathy was more likely to occur in older patients and in pseudophakic or aphakic eyes, indicating that these patients require closer follow-up for the development of silicone oil keratopathy.
Miri et al105 described the IVCM features of normal limbus in 46 eyes. Palisades of Vogt were visible in the superior and inferior limbus as hyperreflective, double-contoured, parallel lines, which alternated with islands of interpalisade rete pegs. Cords of cells extending from the periphery of palisades of Vogt were seen that might represent stem cell crypts. The same authors went on to describe the IVCM features of LSCD in 23 eyes.106 Conjunctivalized corneal epithelial cells appeared hyperreflective with ill-defined borders and were significantly larger, with more variable cell shape, and less dense. Other features included the presence of goblet cells, intraepithelial cystic changes, intraepithelial dendritic cells, and superficial and deep vessels. The subbasal nerve density was significantly lesser in total LSCD compared with partial LSCD.
This review highlights significant cornea literature that is applicable to the practicing ophthalmologist. The selected studies represent but a small proportion of relevant articles published within the year, but nevertheless demonstrate significant developments in the field. Corneal surgery continues to advance, with a major focus on selective lamellar keratoplasty, both DALK and EK, providing better visual and graft survival outcomes, whereas studies on ocular surface transplantation and KPro procedures provide viable alternatives for high-risk cases. Advances in corneal diagnostics include new corneal imaging technologies, improvements in diagnosing infectious keratitis, and in the detection of keratoconus, whereas CXL studies offer new treatment modalities for these common corneal problems. Further research in these new areas is likely to improve our ability to diagnose and treat corneal disorders.
1. Terry MA, Straiko MD, Goshe JM, et al. Descemet’s stripping automated endothelial keratoplasty
: the tenuous relationship between donor thickness and postoperative vision. Ophthalmology. 2012; 119: 1988–1996.
2. Anshu A, Price MO, Price FW. Descemet’s stripping endothelial keratoplasty
: long-term graft survival and risk factors for failure in eyes with preexisting glaucoma. Ophthalmology. 2012; 119: 1982–1987.
3. Goshe JM, Terry MA, Li JY, et al. Graft dislocation and hypotony after Descemet’s stripping automated endothelial keratoplasty
in patients with previous glaucoma surgery. Ophthalmology. 2012; 119: 1130–1133.
4. Yeh RY, Quilendrino R, Musa FU, et al. Predictive value of optical coherence tomography
in graft attachment after Descemet’s membrane endothelial keratoplasty
[published online ahead of print 2012]. Ophthalmology. 2013; 120: 240–245.
5. Arundhati A, Price MO, Price FW Jr. Risk of corneal transplant
rejection significantly reduced with Descemet’s membrane endothelial keratoplasty
. Ophthalmology. 2012; 119: 536–540.
6. Arbelaez MC, Versaci F, Vestri G, et al. Use of a support vector machine for keratoconus
and subclinical keratoconus
detection by topographic and tomographic data. Ophthalmology. 2012; 119: 2231–2238.
7. Otri AM, Fares MA, Al-Aqaba U, et al. Corneal densitometry as an indicator of corneal health. Ophthalmology. 2012; 119: 501–508.
8. Hillenaar T, Cleynenbreugel HV, Verjans GMGM, et al. Monitoring the inflammatory process in herpetic stromal keratitis
: the role of in vivo confocal microscopy
. Ophthalmology. 2012; 119: 1102–1110.
9. San ES, Kubaloglu A, Unal M, et al. Penetrating keratoplasty
versus deep anterior lamellar keratoplasty
: comparison of optical and visual quality outcomes. Br J Ophthalmol. 2012; 96: 1063–1067.
10. Borderie VM, Sandali O, Bullet J, et al. Long-term results of deep anterior lamellar versus penetrating keratoplasty
. Ophthalmology. 2012; 119: 249–255.
11. Sarnicola V, Toro P, Sarnicola C, et al. Long-term graft survival in deep anterior lamellar keratoplasty
. 2012; 31: 621–626.
12. Lyall DAM, Tarafdar S, Gilhooley MJ, et al. Long-term visual outcomes, graft survival and complications of deep anterior lamellar keratoplasty
in patients with herpes simplex related corneal scarring. Br J Ophthalmol. 2012; 96: 1200–1203.
13. Nanavaty MA, Daya SM. Outcomes of deep anterior lamellar keratoplasty
in keratoconic eyes with previous hydrops. Br J Ophthalmol. 2012; 96: 1304–1309.
14. Huang T, Zhang X, Wang Y, et al. Outcomes of deep anterior lamellar keratoplasty
using the big-bubble technique in various corneal diseases. Am J Ophthalmol. 2012; 154: 282–289.
15. Bhatt UK, Fares U, Rahman I, et al. Outcomes of deep anterior lamellar keratoplasty
following successful and failed ‘big bubble’. Br J Ophthalmol. 2012; 96: 564–569.
16. Baradaran-Rafu A, Eslani M, Sadoughi M, et al. Anwar versus Melles deep anterior lamellar keratoplasty
– a prospective randomized clinical trial [published online ahead of print October 2012]. Ophthalmology. 2013; 120: 252–259.
17. Riss S, Heindl LM, Bachmann BO, et al. Pentacam-based big bubble deep anterior lamellar keratoplasty
in patients with keratoconus
. 2012; 31: 627–632.
18. Ghanem RC, Ghanem MA. Pachymetry-guided intrastromal air injection (“pachy-bubble”) for deep anterior lamellar keratoplasty
. 2012; 31: 1087–1091.
19. Buzzonetti L, Petrocelli G, Valente P. Big-bubble deep anterior lamellar keratoplasty
assisted by femtosecond laser in children. Cornea
. 2012; 31: 1083–1086.
20. Cleary C, Song JC, Tang M, et al. Dual laser-assisted lamellar anterior keratoplasty
with top hat graft: a laboratory study. Cornea
. 2012; 31: 791–797.
21. Li JY, Terry MA, Goshe J, et al. Three-year visual acuity outcomes after Descemet’s stripping automated endothelial keratoplasty
. Ophthalmology. 2012; 119: 1126–1129.
22. Shinton AJ, Tsatsos M, Konstantopoulos A, et al. Impact of graft thickness on visual acuity after Descemet’s stripping endothelial keratoplasty
. Br J Ophthalmol. 2012; 96: 246–249.
23. Ang M, Mehta JS, Lim F, et al. Endothelial cell loss and graft survival after Descemet’s stripping automated endothelial keratoplasty
and penetrating keratoplasty
. Ophthalmology. 2012; 119: 2239–2244.
24. Price MO, Gorovoy M, Price FW Jr, et al. Descemet’s stripping automated endothelial keratoplasty
: three-year graft and endothelial cell survival compared with penetrating keratoplasty
[published online ahead of print 2012]. Ophthalmology. 2013; 120: 246–251.
25. Li JY, Terry MA, Goshe J, et al. Graft rejection after Descemet’s stripping automated endothelial keratoplasty
—graft survival and endothelial cell loss. Ophthalmology. 2012; 119: 90–94.
26. Arundhati A, Planchard B, Price MO, et al. A cause of reticular interface haze and its management after Descemet stripping endothelial keratoplasty
. 2012; 31: 1365–1368.
27. Bhogal MS, Angunawela RI, Bilotti E, et al. Theoretical, experimental, and optical coherence tomography
(OCT) studies of graft apposition and adhesion in Descemets stripping automated endothelial keratoplasty
. Invest Ophthalmol Vis Sci. 2012; 53: 3839–3846.
28. Patel SV, Baratz KH, Maguire LJ, et al. Anterior corneal aberrations after Descemet’s stripping endothelial keratoplasty
for Fuchs’ endothelial dystrophy. Ophthalmology. 2012; 119: 1522–1529.
29. Biggelaar FVD, Cheng Y, Nuijts R, et al. Economic evaluation of endothelial keratoplasty
techniques and penetrating keratoplasty
in the Netherlands. Am J Ophthalmol. 2012; 154: 272–281.
30. Bose S, Ang M, Mehta JS, et al. Cost-effectiveness of Descemet’s stripping endothelial keratoplasty
versus penetrating keratoplasty
[published online ahead of print 2012]. Ophthalmology. 2013; 120: 464–470.
31. Yamazoe K, Yamazoe K, Shinozaki N, et al. Influence of the precutting and overseas transportation of corneal grafts for Descemet stripping automated endothelial keratoplasty
on donor endothelial cell loss. Cornea
. 2012; 0: 1–4.
32. Rauen MP, Goins KM, Sutphin JE, et al. Impact of eye bank lamellar tissue cutting for endothelial keratoplasty
on bacterial and fungal corneoscleral donor rim cultures after corneal transplantation. Cornea
. 2012; 31: 376–379.
33. Busin M, Patel AK, Scorcia V, et al. Microkeratome-assisted preparation of ultrathin grafts for Descemet stripping automated endothelial keratoplasty
. Invest Ophthalmol Vis Sci. 2012; 53: 521–524.
34. Rousseau A, Bensalem A, Garnier V, et al. Interface quality of endothelial keratoplasty
buttons obtained with optimized femtosecond laser settings. Br J Ophthalmol. 2012; 96: 122–127.
35. Rudolph M, Laaser K, Bachmann BO, et al. Corneal higher-order aberrations after Descemet’s membrane endothelial keratoplasty
. Ophthalmology. 2012; 119: 528–535.
36. Laaser K, Bachmann BO, Horn FK, et al. Descemet membrane endothelial keratoplasty
combined with phacoemulsification and intraocular lens implantation: advanced triple procedure. Am J Ophthalmol. 2012; 154: 47–55.
37. Dirisamer M, Ham L, Dapena I, et al. Descemet membrane endothelial transfer: “free-floating” donor Descemet implantation as a potential alternative to “keratoplasty
. 2012; 31: 194–197.
38. Dirisamer M, Yeh RY, Dijk KV, et al. Recipient endothelium may relate to corneal clearance in Descemet membrane endothelial transfer. Am J Ophthalmol. 2012; 154: 290–296.
39. Shah RD, Randleman JB, Grossniklaus HE, et al. Spontaneous corneal clearing after Descemet’s stripping without endothelial replacement. Ophthalmology. 2012; 119: 256–260.
40. Ang M, Mehta JS, Sng CCA, et al. Indications, outcomes, and risk factors for failure in tetonic keratoplasty
. Ophthalmology. 2012; 119: 1311–1319.
41. Angunawela RI, Riau A, Chaurasia SS, et al. Manual suction versus femtosecond laser trephination for penetrating keratoplasty
: intraocular pressure, endothelial cell damage, incision geometry, and wound healing responses. Invest Ophthalmol Vis Sci. 2012; 53: 2571–2579.
42. Gaster RN, Dumitrascu O, Rabinowitz YS. Penetrating keratoplasty
using femtosecond laser-enable keratoplasty
with zig-zag incisions versus a mechanical trephine in patients with keratoconus
. Br J Ophthalmol. 2012; 96: 1195–1199.
43. Maier P, Bohringer F, Birnbaum D, et al. Improved wound stability of top-hat profiled femtosecond laser-assisted penetrating keratoplasty
in vitro. Cornea
. 2012; 31: 963–966.
44. Feizi S, Javadi MA. Corneal graft curvature change after relaxing incisions for post-penetrating keratoplasty
. 2012; 31: 1023–1027.
45. Stulting RD, Sugar A, Beck R, et al. Effect of donor and recipient factors on corneal graft rejection. Cornea
. 2012; 31: 1141–1147.
46. Guilbert E, Bullet J, Sandali O, et al. Long-term rejection incidence and reversibility after penetrating and lamellar keratoplasty
[published online ahead of print December 6, 2012]. Am J Ophthalmol. 2013; 155: 560–569.
47. Shimazaki J, Iseda A, Satake Y, et al. Efficacy and safety of long-term corticosteroid eye drops after penetrating keratoplasty
—a prospective, randomized, clinical trial. Ophthalmology. 2012; 119: 668–673.
48. Sangwan VS, Basu S, MacNeil S, et al. Simple limbal epithelial transplantation (SLET): a novel surgical technique for the treatment of unilateral limbal stem cell deficiency. Br J Ophthalmol. 2012; 96: 931–934.
49. Chan CC, Biber EJ, Holland JM. The modified Cincinnati procedure: combined conjunctival limbal autografts and keratolimbal allografts for severe unilateral ocular surface failure. Cornea
. 2012; 31: 1264–1272.
50. Eberwein P, Bohringer J, Schwartzkopff D, et al. Allogenic limbo-keratoplasty
with conjunctivoplasty, mitomycin C, and amniotic membrane for bilateral limbal stem cell deficiency. Ophthalmology. 2012; 119: 930–937.
51. Basu S, Fernandez S, Das MM, et al. Clinical outcomes of xeno-free allogenic cultivated limbal epithelial transplantation for bilateral limbal stem cell deficiency. Br J Ophthalmol. 2012; 96: 1504–1509.
52. Hirayama M, Satake K, Higa Y, et al. Transplantation of cultivated oral mucosal epithelium prepared in fibrin-coated culture dishes. Invest Ophthalmol Vis Sci. 2012; 53: 1602–1609.
53. Burillon C, Huot V, Justin L, et al. Cultured autologous oral mucosal epithelial cell sheet (CAOMECS) transplantation for the treatment of corneal limbal epithelial stem cell deficiency. Invest Ophthalmol Vis Sci. 2012; 53: 1325–1331.
54. Aldave AJ, Sangwan S, Basu VS, et al. International results with the Boston type I keratoprosthesis
. Ophthalmology. 2012; 119: 1530–1538.
55. Kang JJ, Cruz MS, Cortina JDL. Visual outcomes of Boston keratoprosthesis
implantation as the primary penetrating corneal procedure. Cornea
. 2012; 31: 1436–1440.
56. Rudnisky CJ, Belin A, Todani MW, et al. Risk factors for the development of retroprosthetic membranes with Boston keratoprosthesis
type 1—multicenter study results. Ophthalmology. 2012; 119: 951–955.
57. Chan CC, Holland EJ. Infectious keratitis
after Boston type 1 keratoprosthesis
. 2012; 31: 1128–1134.
58. Ramchandran RS, DiLoreto MM, Chung DA, et al. Infectious endophthalmitis in adult eyes receiving Boston type I keratoprosthesis
. Ophthalmology. 2012; 119: 674–681.
59. Robert MC, Moussally M, Harissi-Dagher K. Review
of endophthalmitis following Boston keratoprosthesis
type I. Br J Ophthalmol. 2012; 96: 776–780.
60. Talajic JC, Agoumi Y, Gagne S, et al. Prevalence, progression, and impact of glaucoma on vision after Boston type 1 keratoprosthesis
surgery. Am J Ophthalmol. 2012; 153: 267–274.
61. Dokey A, Ramulu CA, Utine PY, et al. Chronic hypotony associated with the Boston type 1 keratoprosthesis
. Am J Ophthalmol. 2012; 154: 266–271.
62. Akpek EK, Aldave JV, Aquavella AJ. The use of precut, gamma irradiated corneal lenticules in Boston type 1 keratoprosthesis
implantation. Am J Ophthalmol. 2012; 154: 495–498.
63. Robert MC, Biernacki M, Harissi-Dagher K. Boston keratoprosthesis
type 1 surgery: use of frozen versus fresh corneal donor carriers. Cornea
. 2012; 31: 339–345.
64. Stapleton F, Edwards L, Keay K, et al. Risk factors for moderate and severe microbial keratitis
in daily wear contact lens users. Ophthalmology. 2012; 119: 1516–1512.
65. Lichtinger A, Yeung P, Kim SN, et al. Shifting trends in bacterial keratitis
in Toronto—an 11-year review
. Ophthalmology. 2012; 119: 1785–1790.
66. Henry CR, Flynn D, Miller HW, et al. Infectious keratitis
progressing to endophthalmitis—a 15-year study of microbiology, associated factors, and clinical outcomes. Ophthalmology. 2012; 119: 2443–2449.
67. Goldschmidt P, Degorge O, Semoun S, et al. New strategy for rapid diagnosis and characterization of keratomycosis. Ophthalmology. 2012; 119: 945–950.
68. Kuo MT, Chang CK, Cheng HC, et al. A highly sensitive method for molecular diagnosis of fungal keratitis
—a dot hybridization assay. Ophthalmology. 2012; 119: 2434–2442.
69. Ikeda Y, Miyazaki K, Yakura D, et al. Assessment of real-time polymerase chain reaction detection of Acanthamoeba
and prognosis determinants of Acanthamoeba keratitis
. Ophthalmology. 2012; 119: 1111–1119.
70. Younger JR, Johnson GN, Holland D, et al. Microbiologic and histopathologic assessment of corneal biopsies in the evaluation of microbial keratitis
. Am J Ophthalmol. 2012; 154: 512–519.
71. Sy A, Srinivasan J, Mascarenhas M, et al. Pseudomonas aeruginosa keratitis
: outcomes and response to corticosteroid treatment. Invest Ophthalmol Vis Sci. 2012; 53: 267–272.
72. Lalitha P, Srinivasan R, Rajaraman M, et al. Nocardia keratitis
: clinical course and effect of corticosteroids. Am J Ophthalmol. 2012; 154: 934–939.
73. Liu S, Pavan-Langston KA, Colby D. Pediatric Herpes simplex of the anterior segment—characteristics, treatment and outcomes. Ophthalmology. 2012; 119: 2003–2008.
74. Lalitha P, Prajna NV, Oldenburg CE, et al. Organism, minimum inhibitory concentration, and outcome in a fungal corneal ulcer clinical trial. Cornea
. 2012; 31: 662–667.
75. Han SB, Yang HK, Hyon JY, et al. Toxicity of voriconazole on corneal endothelial cells in an animal model. Br J Opthalmol. 2012; 96: 905–908.
76. Vinciguerra P, Albe E, Frueh BE, et al. Two-year corneal cross-linking
results in patients younger than 18 years with documented progressive keratoconus
. Am J Ophthalmol. 2012; 154: 520–526.
77. Caporossi A, Mazzotta C, Baiocchi S, et al. Riboflavin-UVA–induced corneal collagen cross-linking
in pediatric patients. Cornea
. 2012; 31: 227–231.
78. Kymionis GD, Grentzelos MA, Kounis GA, et al. Combined transepithelial phototherapeutic keratectomy and corneal collagen cross-linking
for progressive keratoconus
. Ophthalmology. 2012; 119: 1777–1784.
79. Panda A, Krishna SN, Kumar S. Photo-activated riboflavin therapy for refractory corneal ulcers. Cornea
. 2012; 31: 1210–1213.
80. Galperin G, Berra M, Tau J, et al. Treatment of fungal keratitis
infection by corneal cross-linking
. 2012; 31: 176–180.
81. Sharma A, Nottage JM, Mirchia K, et al. Persistent corneal edema after collagen cross-linking
. Am J Ophthalmol. 2012; 154: 922–926.
82. Ghanem RC, Netto MV, Ghanem VC, et al. Peripheral sterile corneal ring infiltrate after riboflavin-UVA collagen cross-linking
. 2012; 31: 702–705.
83. Wasilewski D, Mello GHR, Moreira H. Impact of collagen crosslinking on corneal sensitivity in keratoconus
. 2012; 0: 1–4.
84. Vimalin J, Gupta N, Jambulingam M, et al. The effect of riboflavin-UVA treatment on corneal limbal epithelial cells—a study on human cadaver eyes. Cornea
. 2012; 31: 1052–1059.
85. Schumacher S, Mrochen M, Wernli J, et al. Optimization model for UV-riboflavin corneal cross-linking
. Invest Ophthalmol Vis Sci. 2012; 53: 762–769.
86. Li Y, Tan O, Brass R, et al. Corneal epithelial thickness mapping by Fourier-domain optical coherence tomography
in normal and keratoconic eyes. Ophthalmology. 2012; 119: 2425–2433.
87. Saad A, Gatinel D. Evaluation of total and corneal wavefront high order aberrations for the detection of forme fruste keratoconus
. Invest Ophthalmol Vis Sci. 2012; 53: 2978–2992.
88. Ishii R, Kamiya K, Igarashi A, et al. Correlation of corneal elevation with severity of keratoconus
by means of anterior and posterior topographic analysis. Cornea
. 2012; 31: 253–258.
89. Leoni-Mesplie S, Mortemousque B, Touboul D, et al. Scalability and severity of keratoconus
in children. Am J Ophthalmol. 2012; 154: 56–62.
90. Choi JA, Kim MS. Progression of keratoconus
by longitudinal assessment with corneal topography. Invest Ophthalmol Vis Sci. 2012; 53: 927–935.
91. Pena-Garcia P, Vega-Estrada A, Barraquer RI, et al. Intracorneal ring segment in keratoconus
: a model to predict visual changes induced by the surgery. Invest Ophthalmol Vis Sci. 2012; 53: 8447–8457.
92. Pinero DP, Alio JL, Barraquer RI, et al. Corneal biomechanical changes after intracorneal ring segment implantation in keratoconus
. 2012; 31: 491–499.
93. Anijeet DR, Zheng Y, Tey A, et al. Imaging and evaluation of corneal vascularization using fluorescein and indocyanine green angiography. Invest Ophthalmol Vis Sci. 2012; 53: 650–658.
94. Kirwan RP, Zheng Y, Tey A, et al. Quantifying changes in corneal neovascularization
using fluorescein and indocyanine green angiography. Am J Ophthalmol. 2012; 154: 850–858.
95. Cheng SF, Dastjerdi MH, Ferrari G, et al. Short-term topical bevacizumab in the treatment of stable corneal neovascularization
. Am J Ophthalmol. 2012; 154: 940–948.
96. Benayoun Y, Adenis JP, Casse G, et al. Effects of subconjunctival Bevacizumab on corneal neovascularization
: results of a prospective study. Cornea
. 2012; 31: 937–944.
97. Koenig Y, Bock F, Kruse FE, et al. Angioregressive pretreatment of mature corneal blood vessels before keratoplasty
: fine-needle vessel coagulation combined with anti-VEGFs. Cornea
. 2012; 31: 887–892.
98. Seo JW, Chung JS, Choi SH, et al. Inhibition of corneal neovascularization
in rats by systemic administration of Sorafenib. Cornea
. 2012; 31: 907–912.
99. Ang M, Chong WT, Tay W, et al. Anterior segment optical coherence tomography
study of the cornea
and anterior segment in adult ethnic South Asian Indian eyes. Invest Ophthalmol Vis Sci. 2012; 53: 120–125.
100. Tan DKL, Chong W, Tay WT, et al. Anterior chamber dimensions and posterior corneal arc length in Malay eyes: an anterior segment optical coherence tomography
study. Invest Ophthalmol Vis Sci. 2012; 53: 4860–4867.
101. Correa-Perez ME, Lopez-Miguel S, Miranda-Anta A, et al. Precision of high definition spectral-domain optical coherence tomography
for measuring central corneal thickness. Invest Ophthalmol Vis Sci. 2012; 53: 1752–1757.
102. Prakash G, Agarwal AI, Mazhari A, et al. Reliability and reproducibility of assessment of corneal epithelial thickness by Fourier domain optical coherence tomography
. Invest Ophthalmol Vis Sci. 2012; 53: 2580–2585.
103. Fukuda R, Usui A, Tomidokoro T, et al. Noninvasive observations of peripheral angle in eyes after penetrating keratoplasty
using anterior segment Fourier-domain optical coherence tomography
. 2012; 31: 259–263.
104. Le Q, Wang X, Lv J, et al. In vivo laser scanning confocal microscopy of the cornea
in patients with silicone oil tamponade after vitreoretinal surgery. Cornea
. 2012; 31: 876–882.
105. Miri A, Al-Aqaba AM, Otri M, et al. In vivo confocal microscopic features of normal limbus. Br J Ophthalmol. 2012; 96: 530–536.
106. Miri A, Alomar M, Nubile T, et al. In vivo confocal microscopic findings in patients with limbal stem cell deficiency. Br J Ophthalmol. 2012; 96: 523–529.