Visual recovery after corneal crosslinking for keratoconus: a 1-year follow-up study

Purpose

To evaluate visual recovery and intraocular straylight in keratoconus patients 3 months and 1 year after corneal crosslinking (CXL)

Patients and Methods

Thirty-three eyes of 28 consecutive patients with mild to moderate keratoconus were included. The following were assessed at baseline, 3 months and 1 year after CXL: corrected distance visual acuity (CDVA), intraocular straylight, spherical equivalent (SE), keratometry (Kmax and K min (Diopters D and axis), the regularity index and pachymetry. Changes from baseline were calculated using mixed linear regression models.

Results

The CDVA remained unchanged 3 months after CXL (-0.003 (95 % CI: -0.038 to 0.044); p?=?0.880) and improved after 1 year (-0.042 (95 % CI: -0.078 to -0.007; p?=?0.021)). The mean straylight value increased significantly by 0.27 (95 % CI: 0.18 to 0.35; p?<?0.001) 3 months after CX and normalized to preoperative values after 1 year (0.06 (95 % CI: -0.03 to 0.14; p?=?0.215)). SE improved from the mean preoperative value of -2.61 D (95 % CI: -3.83 to -1.39) by 1.95 (95 % CI: 1.03 to 2.86; p?<?0.001) at 3 months and remained stable at the 1-year follow-up visit (2.17 (95 % CI: 1.21 to 3.12; p?<?0.001)). Parameters of of keratometry changed only minimally. The regularity index remained almost unchanged at 3 months (2.45 (95 % CI: -4.97 to 9.88; p?=?0.503)) and decreased by 6.97 (95 % CI: -14.08 to 0.14; p?=?0.054). Pachymetry decreased by 44.0 μm (95 % CI: 56.1 to 31.9; p?<?0.001) at 3 months and almost returned to preoperative values at 12 months (-11.3 μm (95 % CI: -27.9 to 5.3; p?=?0.175)).

Conclusions

In accordance with the decrease in CDVA and patients’ complaints of disability due to glare, intraocular straylight increased 3 months after surgery. One year after CXL, there was an increase in CDVA due to an improved SE and regularity index, and intraocular straylight had normalized.     

Alterations of epithelial adhesion molecules and basement membrane components in lattice corneal dystrophy (LCD)

Background

The aim of the study was to investigate the histopathological and ultrastructural correlate of delayed epithelial healing in eyes with lattice corneal dystrophy (LCD).

Materials and Methods

Corneal buttons from 4 patients with LCD (two with subepithelial, two with stromal amyloid deposits) and 2 control corneas were examined. Cell-matrix adhesion molecules and basement membrane components of the corneal epithelium were analyzed by immunohistochemistry and hemidesmosomes between epithelium and stroma were quantified by transmission electron microscopy (TEM).

Results

By TEM well-developed hemidesmosomes anchored the basal epithelial cells to the underlying basement membrane in all normal and LCD corneas. Hemidesmosome density was not significantly different in subepithelial (224.7?±?34.1/100 µm) and stromal (234.3?±?36.3/100 µm) LCD compared to controls (241.3?±?26.8/100 µm). The basement membrane was interrupted in subepithelial, but continous in stromal LCD. Integrin α6 and ß4 staining formed a continous line along the basal surface of the corneal epithelium in control corneas, whereas it appeared discontinous and patchy both in subepithelial and stromal forms of LCD. Staining for αV integrin showed irregular staining patterns, i.e. enhanced labelling intensity in subepithelial and interrupted pattern in stromal LCD, respectively. Integrins α3, ß1, ß2, and ß5, dystroglycan, and plectin were not markedly different in dystrophic corneas. Type VII collagen showed a discontinuous staining in subepithelial forms of LCD. In stromal forms of LCD, type VII collagen staining occurred in additional patches underneath the epithelial basement membrane zone. Type XVII collagen staining was reduced in subepithelial LCD. Laminin-1, laminin-5 and laminin γ2 showed variable irregular staining patterns in dystrophic corneas with focal interruptions, focal thickenings, and reduplications of basement membrane. Some irregularities in corneas with subepithelial amyloid were observed for collagen types IV, V, and XVIII, laminin α1, α3, and γ1, nidogen-1 and -2, perlecan, fibrillin-1.

Conclusions

Immunohistochemical and electron microscopic evidence of structural alterations was found in LCD compared to normal corneas concerning cell-matrix adhesion molecules and basement membrane components. These alterations were more pronounced in dystrophic corneas with subepithelial amyloid deposits than in those with stromal deposits. Histopathological findings may correspond to reduced cell-matrix interactions and partly explain delayed epithelial healing in patients with lattice corneal dystrophy.     

Accelerated versus conventional corneal collagen cross-linking in patients with keratoconus: an intrapatient comparative study

Purpose

To compare the outcomes of the conventional and accelerated corneal collagen cross-linking (CXL) in patients with bilateral progressive keratoconus (KC).

Methods

Fifteen consecutive patients with bilateral progressive KC were enrolled. In each patient, the fellow eyes were randomly assigned to the conventional CXL (3 mW/cm² for 30 min) or accelerated CXL (ACXL) (9 mW/cm² for 10 min) groups. Manifest refraction; uncorrected and corrected distant visual acuity; maximum and mean keratometry; corneal hysteresis and corneal resistance factor; endothelial cell density and morphology; central corneal thickness; and wavefront aberrations were measured before and 12 months after the CXL.

Results

Manifest refraction spherical equivalent and refractive cylinder improved significantly only in conventional group. Uncorrected and corrected distant visual acuity did not change significantly in either group. Also there was no significant change in the maximum and mean keratometry after 12 months. There was significant decrease in central corneal thickness in both groups which was more prominent in conventional group. Endothelial cell density reduced only in the conventional group which was not statistically significant (P = 0.147). CH, CRF, and wavefront aberrations did not change significantly in either group. We did not observe any significant difference in the changes of the variables between the two groups.

Conclusions

Accelerated CXL with 9 mW/cm² irradiation for 10 min had similar refractive, visual, keratometric, and aberrometric results and less adverse effects on the corneal thickness and endothelial cells as compared with the conventional method after 12 months follow-up. However, randomized clinical trials with longer follow-ups and larger sample sizes are needed.

     

Performing corneal crosslinking under local anaesthesia in patients with Down syndrome

Purpose

To report on the ability to perform corneal crosslinking (CXL) under local anaesthesia for the treatment of keratoconus in patients with Down syndrome.

Methods

Nine eyes of seven patients with both keratoconus and Down syndrome were scheduled for an epithelium-off CXL procedure under local anaesthesia. Exclusion criteria were a corneal thickness under 400 µm and the presence of corneal scars. A standardized clinical decision tool was used to estimate patient cooperation and the likelihood for a successful procedure under local rather than general anaesthesia.

Results

In seven eyes, the CXL was completed successfully. The treatment was aborted in two eyes due to insufficient corneal thickness (<400 µm) prior to ultraviolet-A irradiation, even after employing hypoosmolar riboflavin. No adverse events occurred post-operatively, except for one case of delayed epithelial healing (23 days).

Conclusions

With a proper patient selection, CXL under local anaesthesia can be achieved in patients with Down syndrome.

     

Outcomes of corneal collagen crosslinking using a customized epithelial debridement technique in keratoconic eyes with thin corneas

The purpose of the study was to evaluate the outcomes and possible complications of CXL performed with customized epithelial debridement technique to keratoconic corneas with the thinnest pachymetry values less than 400 µm. Nineteen eyes of 19 patients were included. The uncorrected (UCVA) and best corrected visual acuity (BCVA), flattest and steepest keratometric (K) readings, central corneal thickness at the thinnest point (t-CCT), endothelial cell density (ECD) were assessed before and 12 months after CXL. The mean UCVA was increased (p = 0.001), while the mean BCVA did not show any difference (p > 0.05). The mean flattest and steepest K readings were decreased (p = 0.001). No change was observed in the mean t-CCT (p > 0.05). The mean ECD was decreased (p = 0.001). The mean pre-CXL and post-CXL percentages of polymegathism and pleomorphism did not show any significant difference (p > 0.05). CXL performed with customized epithelial debridement technique is successful in halting the progression of keratoconus in corneas thinner than 400 µm after 12 months of treatment. However, significant endothelial cell loss can occur after this procedure.     

Comparison of corneal sensitivity between FS-LASIK and femtosecond lenticule extraction (ReLEx flex) or small-incision lenticule extraction (ReLEx smile) for myopic eyes

Background

To compare the effect on corneal sensitivity between femtosecond laser-assisted laser in situ keratomileusis (FS-LASIK) and femtosecond lenticule extraction (ReLEx flex) or FS-LASIK and small-incision lenticule extraction (ReLEx smile) surgery.

Methods

Twenty-seven subjects (54 eyes) underwent FS-LASIK, 22 subjects (40 eyes) underwent ReLEx flex, and 32 subjects (61 eyes) underwent ReLEx smile surgery. Cochet-Bonnet esthesiometry (Luneau Ophthalmologie Chartres, Cedex, France) was used to evaluate corneal sensitivity preoperatively as well as at 1 week and 1 and 3 months after surgery.

Results

At 1 week, central, superior, nasal, and temporal corneal sensitivity in the ReLEx flex group was significantly higher than in the FS-LASIK group (P?=?0.007, 0.004, 0.020, 0.004 respectively) and in the central and inferior areas at 3 months (P?=?0.002, 0.009 respectively). A higher corneal sensitivity after ReLEx smile surgery was observed in every quadrant at 1 week and 1 and 3 months compared with FS-LASIK surgery (P?<?0.01). Furthermore, in the ReLEx smile group, there were no statistical differences in the superior and temporal quadrants at 1 month postoperatively compared with preoperatively (5.19?±?0.61 cm, P?=?0.198 and 5.64?±?0.48 cm, P?=?0.330 respectively) and no significant differences in any quadrant at 3 months.

Conclusions

Postoperative corneal sensitivity was not remarkably changed after ReLEx smile surgery compared with FS-LASIK. This might be because ReLEx is a flapless procedure.     

Partial lamellar keratoplasty for peripheral corneal disease using a graft from the glycerin-preserved corneoscleral rim

Background

To evaluate the feasibility of partial lamellar keratoplasty (LK) for treatment of peripheral corneal disease (PCD) using a graft from the corneoscleral rim preserved in glycerin.

Methods

Patients who underwent LK for PCD at Shandong Eye Hospital from January 2006 to December 2012 were included. Corneoscleral rims of donor grafts, out of which the corneal buttons had been punched and used for penetrating keratoplasty, were preserved in glycerin and used for LK procedures. A trephine of 7.5–8.0 mm was used to mark the central cornea. An excision was made according to the size of the lesions, after which a partial ring-shaped corneoscleral graft was transplanted to repair the peripheral cornea. Visual acuity, refractive astigmatism, and complications were monitored.

Results

There were 26 patients (28 eyes), including 15 patients with Terrien’s marginal degeneration, eight with Mooren’s ulcer, two with Wegener’s granulomatosis, and one with peripheral corneal ulcer. The mean follow-up was 15 months (range, 6–36 months). Uncorrected visual acuity was improved by a mean of 3.17 Snellen chart lines (P?=?0.010), and best spectacle-corrected visual acuity was improved by a mean of 0.85 Snellen chart lines (P?=?0.045) at 6 months after surgery. The refractive astigmatism decreased by 5.1?±?8.6 diopters (P?=?0.003). Two patients suffering graft ulcers (recurrence) were successfully treated by a second partial LK. No graft rejection occurred.

Conclusions

Partial LK using a graft from the corneoscleral rim appears to be effective in the treatment of patients with PCD. This approach saves the donor graft, which is important in the countries or regions with scarcity of donor tissue.     

Repeatability and comparability of corneal thickness measurements obtained from Dual Scheimpflug Analyzer and from ultrasonic pachymetry

Background

To assess the repeatability and comparability of central corneal thickness (CCT) measurements obtained from the Galilei Dual Scheimpflug Analyzer (Galilei-DSA) and from ultrasonic pachymetry (USP) in normal eyes and in eyes following laser refractive surgery.

Methods

In this prospective observational study, 77 eyes of 39 patients with normal corneas and 39 eyes of 20 patients with prior laser refractive surgery were enrolled. All 116 eyes had three repeated measurements of central CCT with the Galilei-DSA first and then with USP by the same examiner. Reliability of both devices was assessed statistically, and CCT measurements by each were compared.

Results

The mean CCT measurement of normal eyes was 549.2 ± 30.5 μm and 548.5 ± 33.1 μm by the Galilei-DSA and USP respectively; the difference between the two modalities was ?0.7  ± 7.1 μm (P?=?0.365). In post-refractive surgery eyes, the mean CCT measured by the Galilei-DSA and USP was 500.4 ± 44.8 and 494.3 ± 50.3 respectively; the difference between the two devices was ?6.2 ± 9.9 μm (P?<?0.001). For the three repeated measurements, the mean standard deviation was 2 μm with the Galilei-DSA and 1.8 μm with USP in normal eyes, and 2.4 μm with the Galilei-DSA and 1.9 μm with USP in post-refractive surgery eyes.

Conclusion

The Galilei-DSA has high repeatability and excellent to moderate agreement with ultrasonic pachymetry when measuring CCT in normal and post-refractive surgery eyes.     

Comparison of deep anterior lamellar keratoplasty and penetrating keratoplasty with respect to postoperative corneal sensitivity and tear film function

Purpose

To investigate tear film function, central and peripheral corneal sensitivity and corneal subbasal nerve morphology in the cornea after deep anterior lamellar keratoplasty (DALK) compared with penetrating keratoplasty (PK).

Methods

This prospective study compared the changes in 16 eyes of 16 patients who underwent DALK (DALK group) with those in 28 eyes of 28 patients who underwent PK (PK group). Thirty healthy volunteers were also included as controls. Tear functions were evaluated using tear break-up time (TBUT), tear meniscus height (TMH) and corneal fluorescein staining. Corneal sensation was measured with a Cochet-Bonnet esthesiometer. Corneal subbasal nerve morphology was evaluated using in vivo confocal microscopy (IVCM). The patients were examined 1, 3, 6, 9 and 12 months after keratoplasty.

Results

Postoperatively, TMH recovered significantly faster in the DALK group than in the PK group (p?p?p?p?Conclusions Tear film function was restored more rapidly after DALK compared with PK, but there was no significant difference in corneal sensitivity between PK and DALK.     

Comparison of clear corneal incision injuries between torsional and conventional phacoemulsification

Purpose

To compare thermal and mechanical clear corneal incision (CCI) injuries after conventional and torsional phacoemulsification.

Design

A prospective, randomized clinical study.

Methods

A total of 80 eyes were evaluated for the profile of the incision at 1 day and 3 months postoperatively. The wound architecture was recorded with anterior segment optical coherence tomography (AS-OCT). The incisional thickness increased by thermal injury, Descemet’s membrane detachment (DMD) caused by mechanical injury, the presence of endothelial gaping, loss of coaptation, and wound retraction of the conventional phaco group and the torsional phaco group were compared.

Results

The corneal thickness at the incision site of the torsional ultrasound group was significantly lower than the conventional group compared to at the first postoperative visit (1187.55?±?75.66, 1233.62?±?85.87, P?=?.013). However, this thickness was similar between the two groups at 3 months postoperatively. The central endothelial cell loss was significantly lower in the torsional ultrasound group after 3 months (417?±?143, 322?±?130, P?=?.003). There was a positive correlation between cumulated dissipated energy (CDE) and the incisional corneal thickness change that was observed by AS-OCT. Other OCT outcome parameters (such as endothelial gap, DMD, epithelial gap, and uncoaptation) that may be caused by mechanical injury did not differ significantly between the groups on postoperative day one or after 3 months.

Conclusions

The torsional ultrasound mode may provide a lower level of phacoemulsification time and energy and induce less incisional corneal thickness caused by thermal injury in the early postoperative period. The long-term wound healing appeared the same in both ultrasound mode groups.     

Anti-neovascular effect of chondrocyte-derived extracellular matrix on corneal alkaline burns in rabbits

Purpose

We investigated the effect of a chondrocyte-derived extracellular matrix (CDECM) on experimental corneal alkaline burns in rabbits.

Methods

Corneal neovascularization (NV) was induced by applying an 8-mm filter paper soaked in 1 N NaOH to the right central corneas of rabbits for 1 minute. Ten days later, the rabbits were randomly divided into three groups: the alkaline burn group, the CDECM transplantation group, and the human amniotic membrane (HAM) transplantation group. The left eyes were used as controls. CDECM and HAM were transplanted onto the corneal surface to completely cover the resected area and were subsequently sutured. On the 10th day after transplantation, the structural changes of the cornea were analyzed histologically. We examined the effects of CDECM on clinical NV features and on the expression of corneal NV markers.

Results

The alkaline burn produced significant NV and increased the corneal thickness. On day 10 after transplantation, the thickness, NV and opacity of the cornea were markedly decreased in the CDECM group (p?<?0.001). However, the HAM transplantation group did not exhibit improvements in these clinical parameters, and there were no significant differences relative to the burn group. In addition, the use of CDECM improved the healing of the cornea following the alkaline burn by disrupting the corneal epithelial proliferation and reducing the fibrotic changes of the stroma. The hallmarks of NV were significantly induced in the subepithelium by the alkaline burn, and these levels were also suppressed by CDECM. The CDECM suppressed corneal NV by inhibiting nuclear factor-kappa B (NF-κB) activation by blocking the PKC and Akt signaling pathways.

Conclusions

CDECM transplantation was markedly effective in healing alkali-burned corneas by modulating the translocation of NF-κB to the nucleus, thereby representing a promising material for the noninvasive treatment of ocular surface disease.     

“OSMO-UT-DSAEK” using THIN-C medium

Background

When performing ultra-thin Descemet's stripping automated endothelial keratoplasty (UT-DSAEK), the quality of the stromal interface and stromal thickness seem to be critical for visual outcome. The aim of this study was to investigate whether additional osmotic deswelling prior to UT-DSAEK improves the quality of the cut surface and leads to a more reliable and deeper cut in UT-DSAEK (“OSMO-UT-DSAEK”).

Methods

Seventeen human donor corneas not usable for transplantation were used in this experiment. After standard deswelling with culture Medium II, ten corneas were randomly assigned to be additionally deswollen within THIN-C medium. The other remaining seven corneas were put back into culture Medium II. All corneas were placed in an artificial anterior chamber system (Moria); a double path cutting procedure using a microkeratome (Moria) was then performed. Corneal thickness was measured by ultrasound biomicroscopy and in paraffin-embedded slides, followed by histological grading of the cut surface.

Results

Stromal interface smoothness significantly improved after preconditioning in THIN-C medium (Pearson P?=?0.019). The correlation of the corneal thickness obtained by UBM (mean 706?±?SD 208 μm) and histology (mean 530?±?SD 159 μm) was not significant (Pearson r?=?0.11, P?>?0.05, mean difference 247, 95 % CI [+50;+304]). We found no significant correlation between the microkeratome setting and the actual thickness of the lenticule measured in histological analysis in both media as well as for the first and second cut (first cut: Pearson r?=?0.9, P?=?0.1, 95 % CI [?10;+96], second cut: Pearson r?=?0.9, P?=?0.4, 95 % CI [?10;+22]).

Conclusion

Preconditioning of corneas with THIN-C medium significantly improved the quality of the graft interface in UT-DSAEK, but did not significantly improve the cut precision of the microkeratome.     

Inhibitory effects of topical cyclosporine A 0.05 % on immune-mediated corneal neovascularization in rabbits

Background

We aimed to study the inhibitory effects of topical cyclosporine A (CsA) 0.05 % on immune-mediated corneal neovascularization, and to compare its efficacy with those of dexamethasone 0.1 % and bevacizumab 0.5 %.

Methods

Immune-mediated corneal neovascularization was created in 36 right eyes of 36 rabbits. The rabbits were then randomized into four groups. Group I received CsA 0.05 %, Group II received dexamethasone 0.1 %, Group III received bevacizumab 0.5 %, and Group IV received isotonic saline twice a day for 14 days. The corneal surface covered with neovascular vessels was measured on the photographs. The rabbits were then sacrificed and the corneas excised. Paraffin-embedded sections were stained with hematoxylin-eosin and terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling assay.

Results

The means of percent area of corneal neovascularization in Group I, II, III, and IV were 24.4 %, 5.9 %, 37.1 %, and 44.1 %, respectively. The inhibitory effect of CsA 0.05 % was found to be better than the effect found in the bevacizumab 0.5 % and control groups (p?=?0.03 and p?=?0.02, respectively). CsA 0.05 % was found to have significantly lesser inhibitory effects on corneal neovascularization than dexamethasone 0.1 % (p?<?0.001). Apoptotic cell density was higher in Group III and Group IV than in Group I and Group II. There was no difference between Group I and Group II in terms of apoptotic cell density (p?=?0.7).

Conclusions

Topical CsA 0.05 % was shown to have an inhibitory effect on immune-mediated corneal neovascularization in rabbits.     

Assessment of corneal backward light scattering in the healthy cornea and factors affecting corneal transparency

Purpose

To quantify normative corneal optical density (COD) values of healthy individuals and to investigate alterations in the COD in relation to their ocular and demographic characteristics.

Study design

Prospective, cross-sectional, and observational.

Methods

In total, 588 eyes from 588 healthy participants with clear corneas were consecutively enrolled. Corneal densitometry values expressed in standardized grayscale units (GSU) from different corneal layers and zones obtained using the rotating Pentacam Scheimpflug system were analyzed.

Results

The mean (SD) COD over the 12-mm diameter area was 14.4?±?2.74 GSU. A significant positive correlation was found between the participants’ age and the total COD (r?=?0.756, P?<?.001). No significant differences were found between males and females with respect to COD (P?=?.51). No significant correlation was found between the COD measurements and refractive error, pachymetry, and keratometry (P?>?.05 for all comparisons). On the other hand, COD was negatively correlated with white-to-white (WTW) corneal diameter (r?=???0.415, P?<?.001). Moreover, the WTW corneal diameter was correlated with the COD in any zone and layer of the cornea (P?<?.05 for all comparisons).

Conclusions

Wide variations in COD can be detected even in corneas deemed clinically clear. Age and corneal diameter that seem to influence corneal transparency should be taken into account when evaluating corneal backscatter.

     

Comparison of dynamic contour tonometry and Goldmann applanation tonometry in relation to central corneal thickness in primary congenital glaucoma

Background

To compare intraocular pressure (IOP) measurements obtained with dynamic contour tonometer (DCT) and Goldmann applanation tonometer (GAT), and to investigate their relationship to central corneal thickness (CCT) in primary congenital glaucoma (PCG) eyes.

Methods

Thirty-one eyes of 31 PCG patients (25.7?±?7.2 years old) were examined. PCG was defined as elevated IOP, enlarged corneal diameter (buphthalmos), Haab’s striae and abnormal findings at gonioscopy. The mean of three measurements of GAT, DCT (quality scores 1 and 2), and CCT were obtained and assessed for agreement by means of Bland–Altman plot and for Spearman correlation test.

Results

Mean CCT was 534?±?72.3 μm (range: 430 to 610 μm). Mean IOP measurements were 15.1?±?4.2 mmHg (range: 5.5 to 22.7 mmHg) for DCT and 14.5?±?5.6 mmHg (range: 7.0 to 34.0 mmHg) for GAT (P?=?0.244). Spearman correlation tests showed that IOP difference (DCT ? GAT) was not correlated with CCT (r ²?=?0.023, P?=?0.417). IOP measurements by DCT were weakly but statistically correlated with those obtained with GAT (r²?=?0.213, P?=?0.0089). Bland–Altman analysis revealed poor agreement between DCT and GAT readings, considering the 95 % confidence intervals of ±10.45 mmHg.

Conclusions

The differences between DCT and GAT readings were not influenced by CCT in this series of patients. Considering the weak correlation and the poor agreement observed between GAT and DCT measurements and that they both may be affected by corneal biomechanical changes, these methods should not be used interchangeably, and may possibly give no meaningful IOP values in PCG patients.     

Comparison of corneal haze and visual outcome in primary DSAEK versus DSAEK following failed DMEK

Background

Descemet membrane endothelial keratoplasty (DMEK) is being proposed as the procedure of choice in corneal endothelial disease as it achieves better visual and refractive outcomes than Descemet stripping automated endothelial keratoplasty (DSAEK). Nevertheless, primary graft failure is frequent, especially during the learning curve, and secondary back-up procedure consists on DSAEK. We aim to compare corneal haze and visual acuity of patients undergoing primary DSAEK vs. patients undergoing DSAEK as a back-up procedure after primary DMEK failure.

Methods

This study is a comparative case series that included 19 eyes from 16 patients with early stages of corneal failure and limitation of daily activities after primary DSAEK or secondary DSAEK. A control group of non-operated corneas included 10 aged-matched normal eyes. The study was conducted at University Hospital Ramón y Cajal and Vissum Hospital, Madrid, Spain. Corneal densitometry readings and postoperative best-corrected visual acuity in subjects with primary and secondary DSAEK were recorded 6 months after the surgery using the Pentacam Scheimpflug system (Oculus, inc.,Wetzlar, Germany).

Results

In primary DSAEK median densitometry values (range) were statistically significantly higher (p?<?0.05) than normal subjects for the full thickness, posterior and anterior part of the paracentral cornea; and the anterior part of the central cornea. In secondary DSAEK, median densitometry values were statistically significantly higher than normal subjects at all levels of the central and paracentral cornea. In secondary DSAEK, median densitometry values (range) were statistically significantly higher than in primary DSAEK in the full-thickness, anterior part and interface of the central cornea and in the full-thickness and posterior part of the paracentral cornea. Median visual acuity between groups (p?=?0.47) was statistically better for the primary DSAEK group, which also had a higher percentage of patients achieving BCVA of ≥ 20/40 and ≥20/25 than the secondary DSAEK group (100 % vs. 62 % and 60 % vs. 0 % respectively).

Conclusions

There is an increase in central corneal light scattering after secondary DSAEK performed after a failed DMEK as compared to primary DSAEK. This has a negative impact on final visual acuity that needs to be considered in each patient when starting DMEK surgery.     

Hornhautvernetzung mit hypoosmolarer Riboflavin-L?sung beim Keratokonus mit d��nner Hornhaut

Background

The aim of this study was to evaluate the 1-year results of 32 keratoconic eyes with thin corneas which were treated by hypo-osmolar riboflavin solution and ultraviolet A (UVA) collagen cross-linking (CXL).

Patients and methods

Patients with progressive keratoconus and a corneal thickness (CT) less than 400???m (without epithelium) were included in this study. The CT was measured with an ultrasound device (Tomey SP-3000, Nishi-ku, Nagoya, Japan). An increase in the maximum topographic K-value at the apex of keratoconus and a reduction in corneal thickness with or without changes in visual acuity (VA) within the last year were considered to be progression. A total of 32 eyes with an additional follow-up within 1 year were evaluated before and after the procedure. Examinations consisted of an evaluation of VA, corneal topography, slit-lamp microscopy and corneal thickness measurements.

Results

Preoperatively the mean corneal thickness (with epithelium) was 382.3±41.9???m and after removal of the epithelium the thickness of the cornea was reduced to 337.0±51.9???m. After the application of hypo-osmolar riboflavin solution the mean value increased to 451.8±46.7???m. Preoperatively the mean K-value of the apex of the keratoconus was 65.6±11.2 dopters, and 1 year after treatment this value remained relatively unchanged at 64.9±11.0 diopters (P=0.839). Mean VA at the time of the treatment was 0.63±0.37 logMAR and 1 year after the treatment this value was not statistically different (0.59±0.42 logMAR; P=0.662). In the last follow-up examination 1 year after the procedure all corneas were transparent without any scarring lesions in the stroma.

Conclusions

The results of this study using hypo-osmolar riboflavin solution in a cross-linking procedure for thin corneas showed a stability of keratoconus 1 year after CXL. Application of the hypo-osmolar riboflavin solution prevented cross-linked corneas from developing stromal scars.     

Subbasal nerve morphology, corneal sensation, and tear film evaluation after refractive femtosecond laser lenticule extraction

Background

The purpose of this study was to compare corneal subbasal nerve morphology, corneal sensation, and tear film parameters after femtosecond lenticule extraction (FLEX) and small-incision lenticule extraction (SMILE).

Methods

A prospective, randomized, single-masked, paired-eye design clinical trial of 35 patients treated for moderate to high myopia with FLEX in one eye and SMILE in the other. In both techniques, an intrastromal lenticule was cut by a femtosecond laser and manually extracted. In FLEX, a LASIK-like flap allowed removal of the lenticule, whereas in SMILE, it was removed through a small incision. In-vivo confocal microscopy was used to acquire images of the central corneal subbasal nerve plexus, from which nerve density, total nerve number, and nerve tortuosity were analyzed. Corneal sensation was measured using Cochet–Bonnet esthesiometry. A visual analog scale, tear osmolarity, non-invasive tear film break-up time (keratograph) tear meniscus height (anterior segment OCT), Schirmer's test, and fluorescein tear film break-up time were used to evaluate tear film and ocular surface symptoms. Patients were examined before and 6 months after surgery.

Results

There were no statistically significant differences in baseline parameters between FLEX and SMILE (p?>?0.050). With regard to changes from before to 6 months after surgery, mean reduction in subbasal nerve density was 14.22?±?6.24 mm/mm² in FLEX eyes, and 9.21?±?7.80 mm/mm² in SMILE eyes (p?<?0.05). The total number of nerves decreased more in FLEX eyes than in SMILE eyes (p?<?0.05). No change was found when comparing tortuosity (p?>?0.05). Corneal sensation was reduced with 0.38?±?0.49 cm in FLEX eyes, and 0.10?±?0.34 cm in SMILE eyes (p?<?0.01). No differences were found between FLEX and SMILE in tear film evaluation tests (p?>?0.05). Significantly more patients felt postoperative foreign body sensation in the FLEX eye within the first days after surgery, as compared to the SMILE eye.

Conclusions

Six months after surgery, the less invasive SMILE technique seemed better at sparing the central corneal nerves as compared to FLEX. Corneal sensation was only significantly reduced in FLEX eyes. There were no differences between FLEX and SMILE when comparing tear film evaluation tests 6 months after surgery.     

Narbenbildung nach Hornhautvernetzung beim Keratokonus

Background

Corneal scar development after riboflavin-UVA-induced corneal collagen cross-linking (CXL) was retrospectively evaluated.

Patients and methods

A total of 163 CXL-treated eyes in 127 patients with stage 1–3 keratoconus according to Krumeich’s classification were included in this retrospective analysis. The follow-up period was 1 year. At the first and at all follow-up examinations uncorrected visual acuity (UCVA), best corrected visual acuity (BCVA), biomicroscopic findings, corneal topography and corneal thickness were recorded.

Results

At 1 year following CXL, 149 eyes (91.4%) of 114 patients had a clear cornea without corneal scar (control group), while 14 eyes (8.6%) of 13 patients developed clinically significant corneal scar (scar group). Preoperatively, the mean K value of the apex was 62.1 ± 13.8 D in the control group and 71.1 ± 13.2 D in the scar group (P=.02). The mean value of corneal thickness before the procedure was 478.1 ± 52.4 μm in the control group and 420.0 ± 33.9 μm in the scar group (P=.001). The UCVA and BCVA, which were preoperatively similar between groups (P=.59, P=.75 respectively), were postoperatively improved in the control group (P=.023, P=.001 respectively), but reduced in the scar group (P=.012, P=.004 respectively).

Conclusion

K-values and corneal thickness could be considered as predictive factors for the possible development of corneal scarring after riboflavin-UVA-induced CXL. Advanced keratoconus appears to be associated with a higher risk of corneal scar development due to lower corneal thickness, greater curvature and intrinsic tissue characteristics.     

Corneal crosslinking for keratoconus in Japanese populations: one year outcomes and a comparison between conventional and accelerated procedures

Purpose

We retrospectively investigated the efficacy of corneal crosslinking (CXL) on progressive keratoconus in a Japanese population and compared the outcomes of conventional and accelerated CXL.

Study design

A retrospective cohort study

Methods

A total of 108 consecutive eyes in 95 patients (75 men; 21.9 ± 6.2 years) with progressive keratoconus were enrolled. The epithelium was ablated in all eyes. After presoaking the corneal stroma in riboflavin, UV-A was irradiated at 3.0 mW/cm² (conventional CXL) for 30 min on 23 eyes and 18.0 mW/cm² for 5 min (accelerated CXL) on 85 eyes. Best spectacle-corrected visual acuity (BSCVA), manifest refraction, keratometric value, corneal thickness, corneal endothelial cell density (ECD), intraocular pressure, and complications were evaluated at 1, 3, 6, and 12 months after the procedure.

Results

BSCVA, manifest refraction, ECD, and corneal thickness did not change significantly after both procedures. The keratometric value was significantly decreased from the preoperative value at 12 months (p < 0.001). Progression to more than 1.0 D after CXL was observed in 10 eyes (9.3%). The ΔKmax was negatively associated with preoperative Kmax (p < 0.001) and positively associated with preoperative thinnest corneal thickness (p < 0.001). Both treatment modules showed no significant difference in all parameters.

Conclusion

CXL was as effective in treating keratoconus in Japanese patients as in individuals of other ethnicities. Overall, CXL could be performed using either the conventional or accelerated approach to halt the progression of keratoconus in Japanese populations.