PreserFlo™ MicroShunt Versus Trabeculectomy in Patients With Moderate to Advanced Open-Angle Glaucoma: 12-Month Follow-Up of a Single-Center Prospective Study.

PURPOSE: The study compares the efficacy and safety of PreserFlo™ MicroShunt (Santen, Osaka, Japan) and trabeculectomy in eyes with moderate to advanced open-angle glaucoma. METHODS: In this prospective study, 300 patients (300 eyes) with moderate to advanced open-angle glaucoma were treated with either PreserFlo MicroShunt or trabeculectomy. The implantation of the PreserFlo MicroShunt was performed in 150 eyes (group A) and trabeculectomy was performed in 150 eyes (group B). The efficacy was evaluated by estimating the absolute and qualified success rates using the Kaplan-Meier survival analysis.  Results: During the 12-month follow-up period, 81.33% of eyes in group A and 94% in group B reduced intraocular pressure (IOP) >20% without glaucoma eye drops. The average IOP 12 months after surgery was 12.89±3.4 mmHg in group A and 11.39±4.5 mmHg in group B, which was significantly lower than the baseline intraocular pressure (group A: 23.47±8.36 mmHg, p=0.00053; group B: 22.03±5.2 mmHg, p= 0.0006). The number of topical medications administered 12 months after ocular surgery was 0.4±0.8 in group A and 0 in group B, compared to 2.5±1.2 in group A and 2.7±0.9 in group B at baseline (group A: p= 0.00091; group B: p= 0.00072). Ten eyes (6.67%) in group A and four eyes (2.67%) in group B were referred for bleb revision due to bleb fibrosis and consequent acute postoperative rise in IOP. Four eyes of group A and two eyes of group B were treated with transscleral cyclophotocoagulation. Five eyes in group B received re-trabeculectomy because of dysfunction of the primary bleb. CONCLUSION: PreserFlo MicroShunt reduced the number of antiglaucoma agents compared to baseline, but was inferior to trabeculectomy at reducing IOP after 12 months. Additionally, trabeculectomy produced better absolute success rates after 12 months than PreserFlo MicroShunt in the treatment of moderate to advanced open-angle glaucoma.

PreserFlo™ MicroShunt Combined with Phacoemulsification versus PreserFlo™ MicroShunt as a Standalone Procedure in Patients with Medically Resistant Open-Angle Glaucoma.

Purpose: To compare the efficacy and safety of PreserFlo™ MicroShunt (Santen, Osaka, Japan) combined with phacoemulsification to PreserFlo™ MicroShunt as a standalone procedure in eyes with moderate to advanced open-angle glaucoma. Methods: In an observatory, prospective, clinical study, 30 patients (30 eyes) with moderate to advanced angle glaucoma were allocated to either PreserFlo™ MicroShunt combined with phacoemulsification (15 eyes; Group A) or PreserFlo™ MicroShunt as a standalone procedure (15 eyes; Group B). The follow-up time of the study was 12 months. Results: Average intraocular pressure (IOP) at 12 months was 11.62 ± 1.6 mmHg in Group A and 13.8 ± 3.6 mmHg in Group B, which was significantly lower than baseline IOP (Group A: 23.47 ± 8.99 mmHg, P < 0.001; Group B: 23.4 ± 8.68 mmHg, P < 0.001). The absolute reduction of IOP within the 12 postoperative months was not significantly different between the two groups (P = 0.056). The number of the topical medications that were administered 12 months after ocular surgery was 0 in Group A and 0.6 ± 0.8 in Group B, compared to 3.13 ± 1.02 in Group A (P < 0.001) and 2.4 ± 1.45 in Group B (P = 0.004) at baseline. Phacoemulsification combined with PreserFlo™ MicroShunt significantly reduced the number of antiglaucoma agents after 12 months compared to the standalone procedure (P = 0.026). One eye in Group A was referred for bleb revision due to bleb fibrosis and a consequent acute postoperative rise in IOP. One eye in Group A required transscleral cyclophotocoagulation with MicroPulse® laser. One bleb revision was also necessary in Group B at the 4th postoperative week. Endothelial cell density did not significantly change over 12 months in either group (Group A: baseline, 2017.3 ± 346.8 cells/mm[2]; 12 months, 1968.5 ± 385.6 cells/mm[2]; P = 0.38; Group B: baseline, 2134.1 ± 382.6 cells/mm[2]; 12 months, 2094.4 ± 373.3 cells/mm[2], P = 0.42). The PreserFlo™ MicroShunt combined with phacoemulsification produced higher absolute success rates after 12 months in patients with moderate to advanced open-angle glaucoma than the PreserFlo™ MicroShunt as standalone procedure (Group A: 80% and Group B: 60%, P = 0.022). Conclusions: In eyes with moderate to advanced open-angle glaucoma, PreserFlo™ MicroShunt with or without phacoemulsification is effective in reducing IOP and the number of the antiglaucoma agents with a very small incidence of complications and subsequent glaucoma surgeries. However, adding phacoemulsification to PreserFlo™ MicroShunt successfully reduces IOP without the need for ongoing topical medications as are needed after the standalone procedure.

Treatment of Chronic and Painful Corneal Defects with a Complete or Partial Conjunctival Flap. / Partielle oder komplette Bindehautdeckung ­ Indikation und operatives Vorgehen bei chronischen therapieresistenten Hornhautoberflächenproblemen.

BACKGROUND: The coverage of chronic and painful corneal defects with a complete or partial conjunctival flap is a surgical practice that is still applied to advanced stage of keratopathies, which insist despite repeated treatments with a therapeutic contact lens, corneal transplantation or amniotic membrane transplantation. MATERIAL AND METHODS: 37 cases of refractory corneal defects were treated between January 2006 and December 2019 with conjunctival flap. 27 cases underwent a complete and 10 cases a partial conjunctival flap. The mean age of the patients (n = 37, 23 women and 14 men subjects) was 73 years (23 - 87 years). Unsuccessful efforts of intensive conservative treatment and repetitive surgical procedures for several weeks or even months ended up to the indication of performance of a conjunctival flap. Conjunctival flap was performed using the Gundersen technique. A very thin flap of the upper bulbar conjunctiva is dissected downward towards the limbus without creating buttonholes. When the limbus is reached, the resultant flap is freed and without tension it is pulled over the corneal defect and sutured with nylon sutures placed at the limbus (complete conjunctival flap) or at the stroma of healthy cornea (partial conjunctival flap). RESULTS: This kind of conjunctival flap has low rates of complication. The postoperative follow-up after the partial or complete conjunctival flap was satisfying in 34 cases and showed a good healing process without major complications. Partial conjunctival flap shrinkage was observed in 3 cases, which could be revised in 2 of them. One case underwent enucleation, because of recurrent severe keratopathy. CONCLUSION: Conjunctival flap is an established surgical procedure that is applied rarely. The more frequent performance of this technique is suggested as it consists the ultima ratio for the treatment of refractory neurotrophic keratitis and non-healing epithelial defects.

Tectonic Eccentric Penetrating Sclerokeratoplasty à chaud as Treatment for Peripheral Perforated or Predescemetal Corneal Ulceration. / Tektonische exzentrische Sklerokeratoplastik à chaud zur Behandlung von perforierten oder prädescemetalen peripheren Hornhautulzerationen.

BACKGROUND: Tectonic eccentric sclerokeratoplasty à chaud is a very complex corneal operation and aims to preserve the eyeball and possibly to improve visual acuity in patients with peripheral corneal ulcerations or perforating corneal trauma. METHODS: 1650 eyes received corneal transplants in our clinic between 2006 and 2019. 14 of these 1650 eyes (0.8%) had a perforated or predescemetal peripheral corneal ulceration and these were treated between 2006 and 2019 with a tectonic eccentric sclerokeratoplasty à chaud. The mean age of the 8 women and 6 men included in this case series was 63 years (22 - 91 years). RESULTS: In the postoperative follow-up period, most complications and reoperations occurred in the first 6 months. Re-ulceration was observed in 2 cases as well as early suture loosening in 12 patients. These 2 cases were treated with a partial conjunctival flap. Additionally, prolonged superficial wound healing process was shown in 12 patients. Therefore 27 amniotic membrane transplantations, 12 suture replacements, and 6 partial conjunctival flaps were performed in total. No re-keratoplasties were required in this case series. All eyes could be stabilised through the sclerokeratoplasty as well as subsequent operations and intensive local or even systemic treatment. After removal of the corneal suture, the best-corrected visual acuity improved only insignificantly to hand movement in 1 case, finger counting in 3 cases, 0.1 in 5 cases, 0.2 in 4 cases and 0.5 in 1 case. Preservation of the eyeball with pain relief was achieved in all 14 eyes. CONCLUSIONS: Eccentric sclerokeratoplasty is a very demanding surgical procedure and is extremely rarely performed. The goal is to preserve the eye.

Transscleral cyclophotocoagulation with MicroPulse® laser versus cyclophotocoagulation with continuous diode laser in patients with open-angle glaucoma.

PURPOSE: To present the efficacy and safety of surgical treatment of moderate to advanced stage of open-angle glaucoma using the surgical method of transscleral cyclophotocoagulation with MicroPulse® laser (Iridex, Silicon Valley, California, USA) and to compare these results with those of continuous transscleral cyclophotocoagulation with diode laser. METHODS: In a prospective observational clinical study 22 patients (30 eyes) with moderate and advanced open-angle glaucoma underwent transscleral cyclophotocoagulation. 15 eyes were treated with transscleral cyclophotocoagulation with MicroPulse® laser with 2500 mW (group A) and 15 eyes with continuous diode laser 810 nm and power between 2000 and 2300 mW (group B). The follow-up period of the study was 12 months. RESULTS: A reduction of the intraocular pressure > 30% was achieved in 53.3% of the group A and in 60% of the group B during the follow-up period of 12 months. A statistically significant decrease of the number of antiglaucoma agents was observed in group A. One eye (6.6%) of group A underwent additional antiglaucoma procedures to achieve target intraocular pressure. The corresponding percentage is 40% in group B. The postoperative inflammatory response was significantly more limited in group A compared to group B. CONCLUSIONS: The transscleral cyclophotocoagulation with MicroPulse® diode laser with 2500 mW achieves a similar decrease of the intraocular pressure as cyclophotocoagulation with continuous diode laser for the first postoperative months, but more satisfactory reduction of the number of eye drops as well as more limited postoperative inflammatory response.

Preliminary results of Preserflo Microshunt versus Preserflo Microshunt and Ologen implantation.

PURPOSE: To report preliminary 6-month results on the use of the Preserflo Microshunt implant with and without Ologen in 50 pseudophakic eyes with moderate to advanced primary open-angle glaucoma (POAG). METHODS: Fifty pseudophakic eyes underwent ab externo Preserflo Microshunt implantation. Data was gathered retrospectively and two groups were then created, group A with application of MMC 0.2 mg/ml and group B with MMC 0.2 mg/ml and Ologen collagen matrix (OCM) implantation. Absolute success was regarded as the percentage of eyes achieving: a) 5 ≤ intraocular pressure (IOP) ≤ 13 mmHg, b) 5 ≤ IOP ≤ 16 mmHg, and c) 5 ≤ IOP ≤ 21 mmHg without additional medication or surgery and qualified success was regarded as the percentage of eyes achieving a) IOP ≤ 13 mmHg, b) IOP ≤ 16 mmHg, and c) IOP ≤ 21 mmHg with or without medication. Evaluation was performed using a log-rank Kaplan-Meier test. A scatterplot analysis presented the treatment effect data of all eyes with a minimum of 20% IOP reduction per case. Failure was defined as requiring additional surgery, IOP greater than 21 mmHg with or without medication and failure to reach 20% IOP reduction. RESULTS: Mean postoperative IOP was significantly lower in both groups. IOP decreased by 49.06% in group A and by 53.01% in group B at 6 months (P < 0.88), respectively. Medication use was lower in both groups (Wilcoxon test, P < 0.001). The absolute and qualified success rates were not statistically significant between the groups (all P > 0.05). Cumulative IOP results per case were not statistically different in group A compared with group B. One revision surgery in group A (4% failure rate) and three in group B (12% failure rate) were performed. CONCLUSIONS: Both groups showed equal results in terms of cumulative and mean IOP reduction, medication reduction as well as in absolute and qualified success rates. No significant difference was found in any parameters tested between Preserflo Microshunt with MMC 0.2 mg/ml and with or without OCM implantation at 6 months. Long-term follow-up is required to further evaluate this data.

The Efficacy of Ab Externo Canaloplasty by Using a Nonabsorbable Suture and No Viscoelastics; IOP Results, Intraoperative and Early Complication Rates within 6-Month Follow-Up Period.

BACKGROUND: To report the efficacy of ab externo canaloplasty by exclusively using only a nonabsorbable suture as a standalone operation or combined with phacoemulsification in terms of mean intraocular pressure (IOP), medication reduction, and intra- and postoperative complication rates as well as additional glaucoma surgery. PATIENTS AND METHODS: Forty-eight patients with primary open-angle glaucoma (POAG) and 50 eyes were divided into two groups according to the operation performed (Group A canaloplasty, n = 34 and Group B canaloplasty combined with phaco, n = 16), respectively. The success criteria were evaluated between three IOP ranges: a) IOP ≤ 16 mmHg, b) ≤ 18 mmHg, and c) ≤ 21 mmHg. Complete success was considered the percentage of eyes reaching these goals with no medication and qualified success with or without medication. Groups A and B were compared with the Kaplan-Meier test. Mean IOP and medication reduction as well as additional glaucoma surgery were also evaluated. The follow-up time was 6 months. Statistical importance was given at p < 0.05. RESULTS: Fail rate and intraoperative conversion of canaloplasty to other surgery (trabeculotomy due to extensive cheese-wiring) accounted for 12% (6 eyes), while a mandatory usage of the iTrack 250A microcatheter to successfully overcome intracanalicular bridging of the Schlemm's canal accounted for 20% (10 eyes). The most common early postoperative complications in both groups were hyphema, cheese wiring, and transient IOP rising. The mean IOP in Group A reduced from 23.67 ± 6.06 mmHg to 13.35 ± 4.23 mmHg and in Group B, it reduced from 23.55 ± 6.19 mmHg to 12.75 ± 1.5 mmHg at 6 months (p < 0.001). Group A's absolute success rates accounted for a) 44%, b) 47.1%, and c) 52.9% in comparison to Group B [a) 50%, b) 50%, and c) 56.3%] at 6 months (p < 0.05). Qualified success with or without medication showed no statistical differences between the groups. The mean medication reduction was significantly lower (p < 0.001) in both groups (Group A from baseline 3.2 ± 1.23 to 0.95 ± 1.35 and Group B from 3.12 ± 0.88 to 1.12 ± 0.75 at 6 months, respectively). Five eyes (10%) required additional glaucoma surgery. CONCLUSIONS: Ab externo canaloplasty by exclusively using only a nonabsorbable suture as a standalone operation or combined with phacoemulsification seems to lower the mean IOP and medication sufficiently. However, they pose a greater surgical challenge and intraoperative complication rate. Additional glaucoma surgery or medication following both procedures may be necessary to achieve an optimal target IOP.

The correct graft orientation during descemet membrane endothelial keratoplasty (DMEK) using the "bubble-tap" technique.

PURPOSE: To demonstrate a novel, alternative endothelium Descemet membrane layer (EDM) orientation method in Descemet membrane endothelial keratoplasty (DMEK) that does not involve prior ink or trepanation marking of the graft, or intraoperative ocular coherence tomography (OCT) guided graft implantation during surgery, thus preventing the occurrence of an upside-down graft implantation that leads with certainty to primary graft failure. METHODS: From 2017 to early 2020, 500 eyes underwent DMEK operation using the "bubble-tap" technique first described from Dr. Perdikakis in the department of ophthalmology of St.-Johannes- Hospital in Dortmund in Germany. Primary graft failure and re-bubbling results following "bubble-tap" assisted DMEK are presented. RESULTS: Primary graft failure due to upside-down graft implantation was not observed in any patient. Re-bubbling was performed in 4.8% of the eyes. In 1.8% of the cases, a re-DMEK was performed due to a graft failure. CONCLUSION: The "bubble-tap" technique is a novel, reliable and easy to master orientation method of EDM that enables the surgeon to perform DMEK with safety even in complicated cases with poor visibility in the anterior chamber, while it yields superior or equal results in comparison to other graft marking methods or intraoperative OCT-guided graft implantation.

Transscleral cyclophotocoagulation with MicroPulse® laser versus Ahmed valve implantation in patients with advanced primary open-angle glaucoma.

PURPOSE: To compare the efficacy of transscleral cyclophotocoagulation with MicroPulse® laser (Iridex, Silicon Valley, California, USA) with 3000 mW to Ahmed valve implantation in eyes with advanced stage of primary open-angle glaucoma. METHODS: In a prospective observational clinical study, 30 patients (30 eyes) with advanced open-angle glaucoma were randomized for either micropulse transscleral cyclophotocoagulation with 3000 mW or Ahmed valve implantation. Fifteen eyes were treated with transscleral cyclophotocoagulation with MicroPulse® laser with 3000 mW (group A) and 15 eyes with Ahmed valve implantation (group B). As inclusion criteria are included the diagnosis of advanced primary open-angle glaucoma, an intraocular pressure above 21 mmHg, cup-to-disk ratio 0.9-1.0, failure to meet the target IOP with either maximal tolerated local medical therapy (2-4 antiglaucoma agents) or systemic therapy (acetazolamide). The follow-up time of the study was 12 months. An absolute success was defined the achievement of IOP between 6 and 15 mmHg and at least 30% reduction of the IOP from baseline under reduced or the same number of antiglaucoma agents after the surgical procedure without following glaucoma surgeries and as qualified success the achievement of IOP between 6 and 18 mmHg and at least 20% reduction of the IOP from baseline regardless of the number of postoperative antiglaucoma agents. The efficacy was evaluated by estimating the absolute success rate and the qualified success rate using the Kaplan-Meier survival analysis. RESULTS: A reduction of the intraocular pressure > 30% was achieved in 33.3% of group A and in 73.3% of group B during the follow-up period of 12 months. A statistically significant decrease in the number of eye drops was observed in both groups (p < 0.01). 53.3% of the eyes of group A underwent additional anti-glaucoma procedures to achieve target intraocular pressure. The number of the local medications that were administered 12 months after the ocular surgery was 2 (± 1.3) in group A and 0.57 (± 0.9) in group B, compared to 3.2 (± 0.78) in group A and 3.33 (± 0.7) in group B administered prior to the surgery (p: 0.016). 8 eyes (53.33%) in group A were referred for additional treatment due to an acute postoperative rise in IOP (5 eyes: mTS-CPC, 1 eye: canaloplasty, 2 eyes: AGV implantation). No further anti-glaucoma procedures were necessary in group B. CONCLUSION: The Ahmed valve implantation achieves a more efficient decrease of the intraocular pressure as well as of the number of antiglaucoma agents than the transscleral cyclophotocoagulation with MicroPulse® diode laser 3000 mW. Additionally, the Ahmed valve implantation showed better results in terms of absolute and qualified success rates in the treatment of advanced primary open-angle glaucoma.

Estimation of risk-benefit ratio and comparison of post-operative efficacy results between trabeculectomy and canaloplasty.

PURPOSE: To estimate a risk-benefit ratio by comparing the efficacy of canaloplasty to trabeculectomy exclusively in pseudophakic eyes with primary open angle glaucoma. PATIENTS AND METHODS: One hundred four eyes that underwent ab externo canaloplasty and 136 eyes that underwent trabeculectomy with mitomycin C 0.02% and collagen matrix implantation were retrospectively compared. The efficacy was evaluated by evaluating the absolute success rate (5 ⩽ intraocular pressure ⩽ 15 mmHg) and the qualified success rate (intraocular pressure ⩽15 mmHg) using the Kaplan-Meier survival analysis. A meta-analysis to evaluate the relative risk of both procedures in relation to post-operative interventions was performed. RESULTS: Mean intraocular pressure was significantly lower in both groups. Intraocular pressure decreased by 32.17% in the canaloplasty group and by 55.04% in the trabeculectomy group at 12 months (analysis of variance, p < 0.001). Medication use was lower in both groups (analysis of variance, p < 0.001) by the 12th month. The absolute success rate for canaloplasty was 20.19% of eyes compared to 52.21% of eyes with trabeculectomy (p < 0.0001). The qualified success rate was not statistically different between groups (p = 0.15). The relative risk ratio was not statistically different between groups (relative risk of 0.01 and weight of 49.65% for group A and relative risk of 0.0005 and weight of 50.35% for group B; p = 0.5). The hospitalization length was longer in trabeculectomy-treated patients (t-test, p < 0.0001). CONCLUSION: The trabeculectomy group showed better results in terms of absolute success rate. However, canaloplasty may provide a better risk-benefit ratio in terms of qualified success rate, hospitalization time, and required post-operative interventions, since canaloplasty yielded equal or superior results compared to trabeculectomy.

[Explantation of the CyPass implant in a case series of patients with corneal decompensation]. / CyPass-Explantation bei einer Fallserie von Patienten mit Hornhautdekompensation.

BACKGROUND: In microinvasive glaucoma surgery (MIGS) the CyPass microstent (Alcon Inc.) was implanted in the suprachoroidal space to increase the uveoscleral outflow, thereby lowering the intraocular pressure. Due to the excessive loss of corneal endothelial cells, the implant was withdrawn from the global market in August 2018. METHODS: A loss of corneal endothelial cells was observed in 6 patients (8 eyes in total) with early to moderate primary open-angle glaucoma 3 years after CyPass implantation. One eye developed a bullous keratopathy. In all patients an implant migration in the anterior chamber compared to the original postoperative position was detected, with more than one ring of the CyPass implant being visible in the anterior chamber. RESULTS: In all eight cases the CyPass implant was removed. The local therapy with cortisone eye drops in combination with a 5% sodium chloride solution did not improve chronic corneal edema in one patient. Therefore, a Descemet membrane endothelial keratoplasty (DMEK) was performed with postoperative improvement of visual acuity. CONCLUSION: In patients with more than one visible ring of the CyPass implant in the anterior chamber and subsequent localized loss of corneal endothelial cells, CyPass trimming or removal is recommended to avoid the progression of generalized corneal endothelial cell reduction and the onset of refractory, chronic corneal edema.

[DMEK in complicated cases of bullous keratopathy after multiple intraocular surgeries]. / DMEK bei komplizierten Fällen einer bullösen Keratopathie nach multiplen Augenoperationen.

BACKGROUND: During the last decade Descemet membrane endothelial keratoplasty (DMEK) has been established as a surgical treatment even for complex cases of bullous keratopathy. MATERIAL AND METHODS: In a case series 9 eyes from 8 patients with chronic corneal edema caused by multiple intraocular operations underwent DMEK surgery. It was the sole surgical procedure in three eyes. In one case DMEK was combined with an intravitreal injection of bevacizumab, with opacified intraocular lens exchange in a second case, with transscleral cyclophotocoagulation with Iridex micropulse laser (Iridex, Silicon Valley, CA, USA) and intravitreal injection of bevacizumab in a third case and with the Tutopatch® (DMR srl, Italy) coverage of a scleral perforation in a fourth case. Additionally, DMEK was combined with trimming of the Ahmed valve tube length in two cases. The follow-up period was 12 months. RESULTS: Out of 9 eyes 2 showed a persistent partial or total detachment of the graft with concomitant recurrence of bullous keratopathy despite repeated injection of 20% sulfur hexafluoride (SF6) in the anterior chamber. These 2 eyes were treated with penetrating keratoplasty and achieved satisfactory postoperative results including pain relief. Visual acuity did not improve in these 2 eyes due to optic atrophy. Visual acuity and pain improved in the first postoperative month in 7 eyes. Donor endothelial cell density decreased from 2465±147/mm2 to 1295 ± 254/mm2 and 1180 ± 197/mm2 (p < 0.001, n = 7) after 6 and 12 months, respectively. Central corneal thickness decreased from 880 ± 232 µm to 571 ± 15 µm (p = 0.001, n = 7) after 12 months. Multiple rebubblings were performed in 2 of the 9 eyes because of a persistent corneal graft detachment. CONCLUSION: Although DMEK is a technically demanding surgical treatment for complicated cases of severe bullous keratopathy, it provides satisfactory results in terms of the improvement of visual acuity, pain relief and a shorter postoperative period.

Histological Corneal Alterations in Keratoconus After Crosslinking-Expansion of Findings.

PURPOSE: To investigate histopathologic, immunohistochemical, and electron microscopic findings in 8 keratoplasty specimens with a history of corneal collagen crosslinking (CXL) for keratoconus. Five new (hitherto unreported) and 3 previously published specimens were analyzed. METHODS: Corneal buttons of 8 keratoconus corneas (5-114 months after CXL) were compared with 5 keratoconus specimens without CXL and 5 normal corneas for morphological alterations. Corneal buttons were evaluated by light microscopy and immunohistochemistry using antibodies against CD34, PGP 9.5, nestin, telomerase reverse transcriptase, and Ki67 as well as by transmission electron microscopy. RESULTS: Keratoconus corneas after CXL showed a significant keratocyte loss (except 1 specimen with an increased keratocyte number), whereas keratoconus corneas without CXL revealed a higher keratocyte density compared with healthy controls. Keratocyte loss could be clinically correlated with corneal opacification and corneal perforation. In corneas after CXL, the remaining keratocytes appeared more polymorphic and revealed a different expression of surface markers similar to keratocytes in corneal scars. The presence of proteoglycans, nerves, and endothelial cells was unaffected by CXL. CONCLUSIONS: CXL may cause permanent keratocyte loss or repopulation of altered keratocytes, resulting in clinical complications such as corneal opacification or perforation. Despite its good safety profile and high effectiveness in progressive keratoconus, CXL should be performed in accordance with current guidelines strictly adhering to protocol and safety standards.