- Email: [email protected]
Anatomic Predictive Factors of Acute Corneal Hydrops in Keratoconus
Anatomic Predictive Factors of Acute Corneal Hydrops in Keratoconus An Optical Coherence Tomography Study Esteban Fuentes, MD, Otman Sandali, MD, Mohamed El Sanharawi, MD, MPH, Elena Basli, MD, Taous Hamiche, BOpt, Isabelle Goemaere, BOpt, Vincent Borderie, MD, PhD, Nacim Bouheraoua, MD, Laurent Laroche, MD Purpose: To define the optical coherence tomography (OCT) corneal changes predisposing to acute corneal hydrops among patients with advanced keratoconus. Design: Retrospective cohort study. Participants: A total of 191 advanced keratoconic eyes from 191 patients with advanced keratoconus cases were studied. Methods: Data collected from patients with advanced keratoconus cases were studied during a minimum period of 24 months of follow-up. High-resolution Fourier-domain corneal OCT (5 mm of axial resolution) and corneal topography were performed every 4 months during the follow-up. Several anatomic features at the keratoconus cone were analyzed with OCT, including epithelial and stromal thicknesses, the aspect of Bowman’s layer, the presence of Vogt’s striae, and stromal opacities. A comparative analysis between anatomic corneal features in eyes that developed corneal hydrops and those that did not develop this complication during the follow-up was performed. Main Outcome Measures: Evaluation of anatomic corneal changes at risk of developing a corneal hydrops on the basis of OCT findings. Results: Eleven cases of corneal hydrops (5.8%) occurred in our series during a mean follow-up of 30 months (24e36 months). All of these patients were male and younger (23.75.9 years) than patients with no acute keratoconus (32.711.3 years). Increased epithelial thickening with stromal thinning at the conus and the presence of anterior hyperreflectives at the Bowman’s layer level were significantly associated with corneal hydrops, whereas the presence of corneal scarring was a preventive factor. At the healing stage, a pan-stromal scar occurs, with a significant stromal thickening and cornea flattening. Conclusions: Increased epithelial thickening, stromal thinning at the keratoconus cone, anterior hyperreflectives at the Bowman’s layer level, and the absence of stromal scarring are associated with a high risk of developing corneal hydrops. These aspects should be taken into account by the clinician in the evaluation of keratoconus eyes and in the planning of corneal keratoplasty. Ophthalmology 2015;122:1653-1659 ª 2015 by the American Academy of Ophthalmology.
Acute corneal hydrops is a condition characterized by marked corneal edema after a break in Descemet’s membrane. This complication typically affects young individuals with progressive disease and occurs in approximately 3% of patients with keratoconus.1e4 The patients generally present with a sudden onset of decrease in visual acuity, ocular irritation or pain, and photophobia.5,6 Acute keratoconus has been associated with various clinical risk factors, including young age at onset, eye rubbing, atopy, vernal keratoconjunctivitis, and Down syndrome.3,5e8 However, no anatomic predictive risk factors for developing corneal hydrops have been described to date. The detection of patients with keratoconus with a higher risk of corneal hydrops is important because the visual 2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
prognosis and surgical outcomes after deep anterior lamellar keratoplasty (DALK) are better in patients without a history of acute keratoconus and in whom the big bubble technique allowing a complete stromal dissection is contraindicated.9e11 Penetrating keratoplasty is still possible in eyes with a history of hydrops with good visual recovery after surgery. Although DALK is a physiologic approach preserving the healthy endothelial layer in eyes with keratoconus, an Australian graft registry study demonstrated contradictory results, attributing a better prognosis to penetrating keratoplasty over DALK.12 This multicenter study shows “real-life” results from several surgeons with various levels of experience, which differ from selected monocentric studies performed by 1 experienced surgeon.13,14 http://dx.doi.org/10.1016/j.ophtha.2015.04.031 ISSN 0161-6420/15
1653
Ophthalmology Volume 122, Number 8, August 2015 Table 1. AmslereKrumeich Keratoconus Classification Stage I II III IV
Characteristics Eccentric steepening Induced myopia and/or astigmatism of 5.00 D Keratometric reading 48.00 D Vogt’s lines, typical topography Induced myopia and/or astigmatism >5.00 to 8.00 D Keratometric reading 53.00 D Pachymetry 400 mm Induced myopia and/or astigmatism >8.00 to 10.00 D Keratometric reading >53.00 D Pachymetry 200 to 400 mm Refraction not measurable Keratometric reading >55.00 D Central scars Pachymetry 200 mm
D ¼ diopters. Stage is determined if 1 of the characteristics applies. Pachymetry is measured at the thinnest site of the cone.
High-resolution optical coherence tomography (OCT) is a noncontact technique permitting high-resolution scans of different corneal layers in a few seconds. We recently established an OCT keratoconus classification containing 5 distinct keratoconus stages on the basis of structural corneal changes during the evolution of the disease.15 Several authors have reported the value of OCT in the diagnosis and management of patients who developed corneal hydrops.11,16e21 However, to our knowledge, probably because of the rarity of this complication, no OCT study addresses the problem from a preventive perspective in advanced keratoconus cases before the occurrence of corneal hydrops. Furthermore, the anatomic outcomes of corneas that developed acute corneal hydrops have not been extensively studied. The purpose of this study was to describe corneal OCT aspects predisposing to corneal hydrops and to study the anatomic outcomes of keratoconus cases that developed corneal hydrops.
Methods Patients We analyzed retrospectively collected data of consecutive patients followed from July 2011 to July 2014 in a reference center for keratoconus at the Quinze-Vingts National Ophthalmology Hospital (Paris, France). In accordance with French law, institutional review board and ethics committee approval was not required for this study because no modifications to French standards of treatment or follow-up were made. Eyes were assessed and patients were followed up according to the standard operative procedures of the center. The inclusion criteria were patients with advanced keratoconus defined by stage 3 or more regarding AmslereKrumeich classification (Table 1).22 To avoid bias, only the most advanced eye was selected for each patient. A minimum follow-up of 24 months was required. Exclusion criteria comprised any type of prior ocular surgery, including corneal collagen cross-linking, corneal rings, or keratoplasty, and patients who showed signs of acute or healed corneal hydrops confirmed by OCT at the first consultation. Each eye had a characteristic keratoconic appearance on the topographic map (asymmetric bowtie with skewed radial axis, central or inferior steep zone, or claw shape). The keratoconic slit-lamp findings included Munson’s sign, Vogt’s striae, Fleischer ring, apical scar, apical thinning, Rizutti sign, and corneal scars.
Optical Coherence Tomography A Fourier-domain OCT system (RTVue; Optovue, Inc., Fremont, CA) was used in this study. The system works at an 830-nm wavelength, with a scan speed of 26.000 axial scans per second and a depth resolution capacity of 5 mm in tissue. The corneal adaptor module provides the 6-mm scan diameter pachymetry map and the minimum corneal thickness. For each patient, 3 highresolution scans were made across the conus to evaluate structural corneal changes. Corneal epithelial thickness and stromal thickness were measured manually as the distances between the airetear and the epitheliumeBowman layer interfaces and between the Bowman layerestroma and the stromaeDescemet’s membrane interfaces, respectively. The cursors were placed perpendicular to the anterior ocular surface at the point of measurement. The OCT examinations were performed every 4 months at the follow-up. To evaluate interobserver variability, all measurements were made by 2 different examiners (T.H. and I.G.). All OCT scans
Table 2. Optical Coherence Tomography Keratoconus Classification Stage 1 2 3
4 5
1654
Characteristic Thinning of epithelial and stromal layers at the conus. Corneal layers have a normal aspect. Hyperreflective anomalies occurring at the Bowman’s layer level and epithelial thickening at the conus: 2a, clear stroma; 2b, stromal opacities. Posterior displacement of the hyperreflective structures occurring at the Bowman’s layer level with increased epithelial thickening and stromal thinning: 3a, clear stroma; 3b, stromal opacities. Pan-stromal scar Hydrops stage: 5a, acute onset, characterized by the rupture of Descemet’s membrane with delamination of collagen lamellae, large fluid-filled intrastromal cysts, and the formation of epithelial edema; 5b, healing stage, pan-stromal scarring with a remaining aspect of Descemet’s membrane rupture.
Fuentes et al
OCT Anatomic Risk Factors of Corneal Hydrops
Table 3. Baseline Characteristics of Patients Grouped According to Optical Coherence Tomography Keratoconus Classification
Age, yrs (mean, SD) Range Hydrops observed during the study (n) OCT minimum pachymetry, mm (mean, SD) Range OCT epithelial thickness at the conus, mm (mean, SD) Range OCT stromal thickness at the conus, mm (mean, SD) Range OCT epithelial-to-stromal thickness ratio (mean, SD) Range
Stage 1 (n [ 101)
Stage 2 (n [ 60)
Stage 3 (n [ 20)
Stage 4 (n [ 10)
33 (13) (17e73) 0 402 (39) (323e499) 41 (6) (30e55) 367 (44) (272e464) 0.11 (0.03) (0.07e0.19)
33 (10) (22e58) 0 305 (67) (216e419) 74 (20) (52e109) 242 (63) (116e367) 0.34 (0.15) (0.13e0.64)
26 (8) (16e43) 11 274 (51) (149e375) 128 (30) (80e178) 161 (47) (62e231) 0.9 (0.48) (0.44e2.44)
32 (6) (20e37) 0 333 (45) (300e434) 101 (7) (92e113) 232 (22) (196e258) 0.44 (0.05) (0.4e0.53)
P Value 0.0426y <0.0001* <0.0001y <0.0001y <0.0001y <0.0001y
OCT ¼ optical coherence tomography; SD ¼ standard deviation. *Pearson’s chi-square test. y KruskaleWallis 1-way analysis of variance comparing the 4 groups together.
were classified according to our OCT keratoconus classification (Table 2).15 Several anatomic features at the keratoconus cone were analyzed, including the epithelial thickness (microns), stromal thickness (microns), epithelial-to-stromal thickness ratio, presence of hyperreflective anomalies occurring at the Bowman’s layer level, and presence of Vogt’s striae and stromal opacities.
Topography Corneal topography was obtained with the Orbscan IIz (Bausch & Lomb Surgical, Rochester, NY). The steepest and the mean simulated keratometry values were recorded.
Statistical Analysis Results are presented as mean standard deviation for continuous variables and as proportions (percentages) for categoric variables. Nonparametric ManneWhitney and KruskaleWallis tests were used to compare continuous data as appropriate. The Wilcoxon signed-rank test was used to statistically evaluate comparisons between continuous data before and after hydrops. For binary outcomes, the stratified Cochran chi-square test and the Fisher exact test were used for intergroup comparisons of proportions when appropriate. Linear regression, with evaluation of the Spearman coefficient, was used to analyze associations between 2
Figure 1. Optical coherence tomography (OCT) scans show the evolution of 3 patients who developed corneal hydrops (before hydrops, acute onset, and healing stage). A, D, G, Stage 3a before hydrops; important epithelium thickening, stromal thinning, and anterior hyperreflective anomalies occurring at the Bowman’s layer level with clear stroma. Arrows indicate Vogt’s striae. B, E, H, Corneal hydrops (stage 5a); Descemet’s membrane rupture with delamination of collagen lamellae, large fluid-filled intrastromal cysts, and epithelial edema formation. C, F, I, Healing stage (stage 5b); pan-stromal scarring with a remaining aspect of Descemet’s membrane rupture. All scale bars represent 250 mm.
1655
1656
0.0037 61 (3)/55 (4) 62/56 65/55 56/51 59/52 61/57
D ¼ diopters; L ¼ left; M ¼ male; OCT ¼ optical coherence tomography; R ¼ right; SD ¼ standard deviation. *Wilcoxon signed-rank test. Comparison before and after hydrops.
60/53 58/55 60/56 63/59 60/48
66/53 67/60
67/60
0.0038 67 (3)/58 (4) 66/60 70/58 63/53 67/62 63/60 66/57 72/63 65/51
0.8/0.1 0.9/0.1
69/63
0.0038 1 (0.6)/0.2 (0.1) 0.7/0.1 0.9/0.2 0.8/0.2 0.4/0.2 0.6/0.2 1.6/0.3 0.6/0.3 1.4/0.3
204/524 198/547
2.4/0.2
0.001 164 (52)/381 (111) 180/450 160/320 152/348 231/450 190/470 89/306 210/231 115/308
160/73 176/73
73/239
142 (26)/72 (12) 120/66 145/72 127/83 102/75 114/76 141/80 130/60 165/90
M 16 L 3 365/385 M 16 R 3 338/328
Sex Age (yrs) Eye OCT stage OCT minimum pachymetry (mm) (before/ after hydrops) OCT epithelial thickness at the conus (mm) (before/after hydrops) OCT stromal thickness at the conus (mm) (before/after hydrops) OCT epithelial-to-stromal thickness ratio (before/after hydrops) Steepest keratometry (D) (before/after hydrops) Mean keratometry (D) (before/after hydrops)
178/47
0.0049 285 (51)/318 (48)
23.7 (5.9)
M 23 L 3 285/326 M 22 R 3 240/290 M 28 L 3 360/391 M 28 L 3 295/300 M 34 L 3 287/290 M 20 L 3 227/241 M 31 R 3 259/278 M 23 L 3 262/375
9 8 7 6 5 4 3 2 1 Case
Table 4. Characteristics of Patients before and after Corneal Hydrops
A total of 267 eyes of 267 keratoconic patients were in stage 3 or 4 of the AmslereKrumeich classification of keratoconus. However, 73 eyes underwent corneal surgery (corneal crosslinking, n ¼ 11; corneal rings, n ¼ 15; and keratoplasty, n ¼ 47) during the followup, and 3 eyes showed clinical signs of corneal hydrops at the first consultation and were removed from our series. The remaining 191 eyes were included in the study. The analysis of OCT images yielded highly reproducible epithelial and stromal thickness measurements, evidenced by excellent ICC parameters (0.92 for epithelial thickness and 0.97 for stromal thickness). The repartition of all patients according to our OCT classification and their anatomic characteristics are summarized in Table 3. Eleven cases (5.8%) in our cohort developed a corneal hydrops during a mean follow-up of 30 months (range, 24e36 months). In acute hydrops onset, OCT showed a typical aspect of Descemet’s membrane rupture with delaminations of collagen lamellae, large fluid-filled intrastromal cysts, and epithelial edema formation. The Descemet’s membrane rupture and the edematous corneal changes were mainly located at the conus area. At the healing stage, panstromal scarring occurred with a remaining aspect of Descemet’s membrane rupture (Fig 1). All 11 patients who developed corneal hydrops in our study were classified as OCT stage 3 during their last visit before the acute phase. Indeed, this stage was associated with the occurrence of corneal hydrops (P < 0.0001, Pearson’s chi-square test). The characteristics of patients before and after the corneal hydrops are summarized in Table 4. All these patients were male and younger than patients who did not develop hydrops in the study period. Before corneal hydrops, the average epithelial and stromal thicknesses at the cone were 142 (26) mm and 164 (52) mm, respectively. At the scarring stage, the mean values were 72 (12) mm and 381 (111) mm for epithelial and stromal thicknesses, respectively, demonstrating a stromal thickening and epithelial thinning at the conus after hydrops. The average steepest keratometry decreased from 67 (3) diopters before hydrops to 58 (4) diopters at the healing stage (6 months after hydrops). The comparative clinical and anatomic characteristics between those with hydrops and those with no hydrops are shown in Table 5. Before developing corneal hydrops, the minimal corneal and stromal thicknesses at the conus were significantly thinner and the epithelial thickness was significantly thicker in the hydrops group compared with the no hydrops group. The epithelium-to-stroma thicknesses ratio was significantly higher in patients who developed corneal hydrops. In addition, the presence of hyperreflectivity at the Bowman’s layer and Vogt’s
10
Results
M 20 R 3 221/290
11
Mean ± (SD)
P Value*
continuous variables. Interobserver agreement for the measurement of epithelial and stromal thickness using OCT images was assessed by the intraclass correlation coefficient (ICC), and the limits of agreement were assessed using the BlandeAltman method. The ICC was considered reliable if the values were between 0.4 and 0.75, and values >0.75 were considered excellent. Because many comparisons were performed, the Bonferroni correction was used to correct the P value. P values 0.05 were considered statistically significant. Univariate comparisons by logistic regression were performed to identify associations between the occurrence of hydrops and other variables under study. Variables found to be associated with a P value < 0.05 were then subjected to multivariate logistic regression. P 0.05 was considered statistically significant. Statistical analysis was carried out using SPSS for Windows version 16.0 (SPSS, Inc., Chicago, IL).
0.001
Ophthalmology Volume 122, Number 8, August 2015
Fuentes et al
OCT Anatomic Risk Factors of Corneal Hydrops
Table 5. Comparative Clinical and Anatomic Characteristics between the Hydrops and No Hydrops Groups Hydrops (n [ 11)
No Hydrops (n [ 180)
23.7 (5.9) (16e34) 11/0 285 (51) (221e365) 142 (26) (102e178) 164 (52) (73e231) 1 (0.6) (0.4e2.4) 11 8 67 (2.8) (63e72)
32.7 (11.3) (17e73) 115/65 355 (73) (80e500) 62 (31) (30e181) 300 (92) (62e464) 0.26 (0.23) (0.07e1.4) 52 31 57.6 (7.4) (45e77)
Age, yrs (mean, SD) Range Sex (M/F) OCT minimum pachymetry, mm (mean, SD) Range OCT epithelial thickness at the conus, mm (mean, SD) Range OCT stromal thickness at the conus, mm (mean, SD) Range OCT epithelial-to-stromal thicknesses ratio, mm (mean, SD) Range OCT hyperreflectivity at Bowman’s layer OCT Vogt’s striae visualization Steepest keratometry values, D (mean, SD) Range
P Value 0.0108* 0.0171y 0.0044* <0.0001* 0.0004* 0.0123* <0.001y <0.001y 0.7441*
D ¼ diopters; F ¼ female; M ¼ male; OCT ¼ optical coherence tomography; SD ¼ standard deviation. *ManneWhitney test. y Fisher exact test.
striae visualized on OCT were significantly more frequent in the hydrops group. With respect to the steepest keratometry, there was no significant difference between both groups. To define patients with a higher risk of hydrops among stage 3 OCT cases (n ¼ 20), an additional intragroup comparative analysis was performed between patients who developed hydrops and patients who did not develop acute keratoconus during the follow-up, as shown in Table 6. Of note, patients with higher epithelial thickness (P ¼ 0.042) and without stromal scarring (or stage 3a) (P ¼ 0.007) were at higher risk of developing corneal hydrops. However, no case of acute keratoconus occurred in eyes with stromal opacities (Fig 2). Variables associated with a P value < 0.05 were then subjected to logistic regression analysis. Of note, an increased epithelial thickness at the conus was the only risk factor for the occurrence of hydrops corresponding to patients classified as stage 3 OCT (P ¼ 0.02).
Discussion Corneal hydrops is a rare event and has been reported at 3% in the literature.1e3 The incidence of acute hydrops in our study was higher and could be explained by selection bias in
our study, which considered more severe keratoconus cases than those in the literature. The predominance of male patients and young patients, which occurred in our series, has been reported.1e3 This study describes new anatomic risk factors of corneal hydrops that should be considered when evaluating patients with keratoconus. Indeed, surgical and visual prognoses are worse in patients who developed acute keratoconus because DALK is difficult and complete dissection is rarely possible.9,10 The morphologic stage 3a appearance of the OCT classification indicated a higher risk to develop corneal hydrops. The principal OCT anatomic findings at this stage are hyperreflective anomalies occurring at the Bowman’s layer level with marked epithelial thickening and stromal thinning. Anterior hyperreflectives at the Bowman’s layer level were found in all patients before developing corneal hydrops. In a recent confocal microscopy study performed on 10 keratoconic eyes that developed corneal hydrops, a similar hyperreflective band was found in all eyes.23 These anomalies correspond to fibrillar degeneration and fibroblastic
Table 6. Comparison between Hydrops and No Hydrops Groups, among Patients with Stage 3 on Optical Coherence Tomography Hydrops Group (n [ 11) OCT minimum corneal pachymetry, mm (mean, SD) OCT epithelial thickness at the conus, mm (mean, SD) OCT stromal thickness at the conus, mm (mean, SD) OCT epithelial-to-stromal thicknesses ratio, mm (mean, SD) OCT Vogt’s striae visualization Absence of stromal scar (stage 3a OCT) Steepest keratometry values, D (Mean, SD)
285 142 164 1
(51) (26) (52) (0.6) 8 11 67 (2.8)
No Hydrops Group (n [ 9) 260 111 158 0.8
(50) (29) (42) (0.3) 1 2 65 (10)
P Value 0.5737* 0.042* 0.6454* 0.2445* 0.0098y 0.007y 0.7302*
D ¼ diopters; OCT ¼ optical coherence tomography; SD ¼ standard deviation. *ManneWhitney test. y Fisher exact test.
1657
Ophthalmology Volume 122, Number 8, August 2015
Figure 2. The optical coherence tomography scans show aspects of advanced keratoconus with stromal opacities. A, Stage 3b; important epithelium thickening, stromal thinning, and stromal opacity. B, Stage 4; pan-stromal scar. All scale bars represent 250 mm.
accumulation in the anterior stroma beneath Bowman layer breaks and would translate into an active inflammatory disease factor.15,24,25 This activation is thought to be triggered by cytokines (including interleukin-1 and Fas-ligand) released from the damaged epithelial cells.26e29 The epithelium-to-stromal thicknesses ratio at the conus was associated with a high risk for corneal hydrops development. Stromal thickness, especially that of the anterior stroma, increase corneal rigidity and account for maintenance of corneal curvature.30,31 Epithelial thickening in advanced keratoconus cases seems to be an important indirect predictive factor of corneal fragility and therefore hydrops. This is a consequence of the impressive capacity of corneal epithelium to compensate irregularities caused by the extreme thinning of the corneal stroma in keratoconic patients.15,32e36 Furthermore, when analyzing the morphologic appearance of prehydrops stage 3 OCT images, there is a marked compensatory epithelial hyperplasia in a specific area of extreme stromal thinning corresponding to the keratoconus cone. This area corresponds to the weakest zone of the cornea, explaining the occurrence of Descemet’s membrane rupture and the predominance of the edematous corneal changes at the conus area. Vogt’s striae have the aspect of dark parallel lines running through the entire stromal thickness, similar to those observed in confocal microscopy.15,25 These lines have been suggested to represent collagen lamellae under stress. We found that the presence of Vogt’s striae increases the risk of corneal hydrops when they are present in keratoconus with thin stromas, as found in OCT stage 3a. However, the presence of Vogt’s striae was not significantly associated with corneal hydrops occurrence in the multivariate analysis. Of note, no patient with stromal opacities (3b and 4 stages) developed this complication. We hypothesize that the presence of stromal scarring increases corneal rigidity and is a preventive factor of corneal hydrops, as found in our series. The OCT stage 3 corresponds to advanced keratoconus, and the stromal thickness is generally inferior to 300 mm, not allowing cross-linking or corneal ring implantation procedures.15 Deep anterior lamellar keratoplasty is generally indicated and would also prevent the corneal hydrops, especially in patients with other nonanatomic risk factors, such as young age and atopy. At the healing stage of corneal hydrops, significant stromal thickening and cornea flattening occur.1,4 The stromal thickening would be the result of the inflammatory
1658
and healing processes triggered after hydrops.23 As described by Stone et al37 in a histopathologic study, the rupture and detachment of Descemet’s membrane result in the formation of extensive ledges with a fibrous metaplasia of the endothelium with secretion of extensive fibrillar and basement membrane material. However, the beneficial effect of this “natural cross-linking” is limited by the residual pan-stromal corneal scar, which is located centrally or paracentrally at the corneal cone, in most cases limiting visual acuity rehabilitation.1,16
Study Limitations Potential limitations of our study may be the small number of patients (n ¼ 11) who developed corneal hydrops. However, this complication is rare, and patients usually consult once corneal hydrops occurs and not before. Furthermore, all patients in our series had a close corneal OCT follow-up permitting an accurate structural analysis of anatomic profiles at risk of corneal hydrops. In conclusion, we describe new anatomic corneal aspects predisposing to corneal hydrops. Stage 3a of the OCT keratoconus classification (characterized by increased epithelial thickening, stromal thinning at the conus, anterior hyperreflectives at the Bowman’s layer level, and absence of stromal scarring) is associated with a high risk for developing corneal hydrops and should be taken into account by the clinician in the evaluation of eyes with keratoconus and in the planning of corneal keratoplasty. Additional studies with longer follow-up are needed to confirm our findings.
References 1. Tuft SJ, Gregory WM, Buckley RJ. Acute corneal hydrops in keratoconus. Ophthalmology 1994;101:1738–44. 2. Basu S, Vaddavalli PK, Ramappa M, et al. Intracameral perfluoropropane gas in the treatment of acute corneal hydrops. Ophthalmology 2011;118:934–9. 3. Grewal S, Laibson PR, Cohen EJ, Rapuano CJ. Acute hydrops in the corneal ectasias: associated factors and outcomes. Trans Am Ophthalmol Soc 1999;97:187–203. 4. Amsler MM. Quelques données du problème du kératocône. Bull Soc Belge Ophtalmol 1961;128:331–54. 5. Sharma R, Titiyal JS, Prakash G, et al. Clinical profile and risk factors for keratoplasty and development of hydrops in north Indian patients with keratoconus. Cornea 2009;28:367–70.
Fuentes et al
OCT Anatomic Risk Factors of Corneal Hydrops
6. Stoiber J, Muss W, Ruckhofer J, Grabner G. Acute keratoconus with perforation in a patient with Down’s syndrome. Br J Ophthalmol 2003;87:120. 7. Koenig SB, Smith RW. Keratoconus and corneal hydrops associated with compulsive eye rubbing. Refract Corneal Surg 1993;9:383–4. 8. Rehany U, Rumelt S. Corneal hydrops associated with vernal conjunctivitis as a presenting sign of keratoconus in children. Ophthalmology 1995;102:2046–9. 9. Das S, Dua N, Ramamurthy B. Deep lamellar keratoplasty in keratoconus with healed hydrops. Cornea 2007;26:1156–7. 10. Chew A, Mehta J, Tan D. Deep lamellar keratoplasty after resolution of hydrops in keratoconus. Cornea 2011;30:454–9. 11. Fan Gaskin JC, Patel DV, McGhee CNJ. Acute corneal hydrops in keratoconus - new perspectives. Am J Ophthalmol 2014;157:921–8. 12. Coster DJ, Lowe MT, Keane MC, Williams KA; for the Australian Corneal Graft Registry Contributors. A comparison of lamellar and penetrating keratoplasty outcome: a registry study. Ophthalmology 2014;121:979–87. 13. Borderie VM, Sandali O, Bullet J, et al. Long-term results of deep anterior lamellar versus penetrating keratoplasty. Ophthalmology 2012;119:249–55. 14. Borderie VM, Guilbert E, Touzeau O, Laroche L. Graft rejection and graft failure after anterior lamellar versus penetrating keratoplasty. Am J Ophthalmol 2011;151:1024–9. 15. Sandali O, El Sanharawi M, Temstet C, et al. Fourier-domain optical coherence tomography imaging in keratoconus: a corneal structural classification. Ophthalmology 2013;120: 2403–12. 16. Maharana PK, Sharma N, Vajpayee RB. Acute corneal hydrops in keratoconus. Indian J Ophthalmol 2013;61:461–4. 17. Basu S, Vaddavalli PK, Vemuganti GK, et al. Anterior segment optical coherence tomography features of acute corneal hydrops. Cornea 2012;31:479–85. 18. Kallel S, Tahiri Joutei Hassani R, Liang H, et al. “En face” anterior segment optical coherence tomography findings in acute corneal hydrops [in French]. J Fr Ophtalmol 2014;37: 605–12. 19. Kucumen BR, Yenerel NM, Gorgun E, Dinc UA. Anterior segment optical coherence tomography findings of acute hydrops in a patient with keratoconus. Ophthalmic Surg Lasers Imaging 2010;41(suppl):114–6. 20. Ueno H, Matuzawa A, Kumagai Y, et al. Imaging of a severe case of acute hydrops in a patient with keratoconus using anterior segment optical coherence tomography. Case Rep Ophthalmol 2012;3:304–10. 21. Basu S, Mukhopadhyay S. Acute corneal hydrops. Ophthalmology 2012;119:2197–2197.e1; author reply 2198.
22. Krumeich JH, Kezirian GM. Circular keratotomy to reduce astigmatism and improve vision in stage I and II keratoconus. J Refract Surg 2009;25:357–65. 23. Lockington D, Fan Gaskin JC, McGhee CNJ, Patel DV. A prospective study of acute corneal hydrops by in vivo confocal microscopy in a New Zealand population with keratoconus. Br J Ophthalmol 2014;98:1296–302. 24. Chi HH, Katzin HM, Teng CC. Histopathology of keratoconus. Am J Ophthalmol 1956;42:847–60. 25. Efron N, Hollingsworth JG. New perspectives on keratoconus as revealed by corneal confocal microscopy. Clin Exp Optom 2008;91:34–55. 26. Jun AS, Cope L, Speck C, et al. Subnormal cytokine profile in the tear fluid of keratoconus patients. PLoS One 2011; e16437:1e8. 27. Wilson SE, He YG, Weng J, et al. Epithelial injury induces keratocyte apoptosis: hypothesized role for the interleukin-1 system in the modulation of corneal tissue organization and wound healing. Exp Eye Res 1996;62:325–7. 28. Fabre EJ, Bureau J, Pouliquen Y, Lorans G. Binding sites for human interleukin 1 alpha, gamma interferon and tumor necrosis factor on cultured fibroblasts of normal cornea and keratoconus. Curr Eye Res 1991;10:585–92. 29. Lema I, Durán JA. Inflammatory molecules in the tears of patients with keratoconus. Ophthalmology 2005;112:654–9. 30. Müller LJ, Pels E, Vrensen GF. The specific architecture of the anterior stroma accounts for maintenance of corneal curvature. Br J Ophthalmol 2001;85:437–43. 31. Maurice DM, Monroe F. Cohesive strength of corneal lamellae. Exp Eye Res 1990;50:59–63. 32. Reinstein DZ, Archer TJ, Gobbe M. Corneal epithelial thickness profile in the diagnosis of keratoconus. J Refract Surg 2009;25:604–10. 33. Reinstein DZ, Gobbe M, Archer TJ, et al. Epithelial, stromal, and total corneal thickness in keratoconus: three-dimensional display with artemis very-high frequency digital ultrasound. J Refract Surg 2010;26:259–71. 34. 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: 2425–33. 35. Haque S, Simpson T, Jones L. Corneal and epithelial thickness in keratoconus: a comparison of ultrasonic pachymetry, Orbscan II, and optical coherence tomography. J Refract Surg 2006;22:486–93. 36. Sykakis E, Carley F, Irion L, et al. An in depth analysis of histopathological characteristics found in keratoconus. Pathology 2012;44:234–9. 37. Stone DL, Kenyon KR, Stark WJ. Ultrastructure of keratoconus with healed hydrops. Am J Ophthalmol 1976;82:450–8.
Footnotes and Financial Disclosures Originally received: January 26, 2015. Final revision: April 21, 2015. Accepted: April 23, 2015. Available online: June 1, 2015.
Analysis and interpretation: Fuentes, Sandali, El Sanharawi, Basli, Borderie, Bouheraoua, Laroche Obtained funding: Not applicable Manuscript no. 2015-133.
Centre Hospitalier National d’Ophtalmologie des XV-XX, Pierre & Marie Curie University Paris 6, Paris, France. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Author Contributions: Conception and design: Sandali
Overall responsibility: Fuentes, Sandali, Borderie, Bouheraoua, Laroche Abbreviations and Acronyms: DALK ¼ deep anterior lamellar keratoplasty; ICC ¼ intraclass correlation coefficient; OCT ¼ optical coherence tomography. Correspondence: Otman Sandali, MD, Centre Hospitalier National d’Ophtalmologie des XV-XX, 28 Rue de Charenton, 75571 Paris, France. E-mail: [email protected].
Data collection: Fuentes, Hamiche, Goemaere
1659
Acute corneal hydrops is a condition characterized by marked corneal edema after a break in Descemet’s membrane. This complication typically affects young individuals with progressive disease and occurs in approximately 3% of patients with keratoconus.1e4 The patients generally present with a sudden onset of decrease in visual acuity, ocular irritation or pain, and photophobia.5,6 Acute keratoconus has been associated with various clinical risk factors, including young age at onset, eye rubbing, atopy, vernal keratoconjunctivitis, and Down syndrome.3,5e8 However, no anatomic predictive risk factors for developing corneal hydrops have been described to date. The detection of patients with keratoconus with a higher risk of corneal hydrops is important because the visual 2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
prognosis and surgical outcomes after deep anterior lamellar keratoplasty (DALK) are better in patients without a history of acute keratoconus and in whom the big bubble technique allowing a complete stromal dissection is contraindicated.9e11 Penetrating keratoplasty is still possible in eyes with a history of hydrops with good visual recovery after surgery. Although DALK is a physiologic approach preserving the healthy endothelial layer in eyes with keratoconus, an Australian graft registry study demonstrated contradictory results, attributing a better prognosis to penetrating keratoplasty over DALK.12 This multicenter study shows “real-life” results from several surgeons with various levels of experience, which differ from selected monocentric studies performed by 1 experienced surgeon.13,14 http://dx.doi.org/10.1016/j.ophtha.2015.04.031 ISSN 0161-6420/15
1653
Ophthalmology Volume 122, Number 8, August 2015 Table 1. AmslereKrumeich Keratoconus Classification Stage I II III IV
Characteristics Eccentric steepening Induced myopia and/or astigmatism of 5.00 D Keratometric reading 48.00 D Vogt’s lines, typical topography Induced myopia and/or astigmatism >5.00 to 8.00 D Keratometric reading 53.00 D Pachymetry 400 mm Induced myopia and/or astigmatism >8.00 to 10.00 D Keratometric reading >53.00 D Pachymetry 200 to 400 mm Refraction not measurable Keratometric reading >55.00 D Central scars Pachymetry 200 mm
D ¼ diopters. Stage is determined if 1 of the characteristics applies. Pachymetry is measured at the thinnest site of the cone.
High-resolution optical coherence tomography (OCT) is a noncontact technique permitting high-resolution scans of different corneal layers in a few seconds. We recently established an OCT keratoconus classification containing 5 distinct keratoconus stages on the basis of structural corneal changes during the evolution of the disease.15 Several authors have reported the value of OCT in the diagnosis and management of patients who developed corneal hydrops.11,16e21 However, to our knowledge, probably because of the rarity of this complication, no OCT study addresses the problem from a preventive perspective in advanced keratoconus cases before the occurrence of corneal hydrops. Furthermore, the anatomic outcomes of corneas that developed acute corneal hydrops have not been extensively studied. The purpose of this study was to describe corneal OCT aspects predisposing to corneal hydrops and to study the anatomic outcomes of keratoconus cases that developed corneal hydrops.
Methods Patients We analyzed retrospectively collected data of consecutive patients followed from July 2011 to July 2014 in a reference center for keratoconus at the Quinze-Vingts National Ophthalmology Hospital (Paris, France). In accordance with French law, institutional review board and ethics committee approval was not required for this study because no modifications to French standards of treatment or follow-up were made. Eyes were assessed and patients were followed up according to the standard operative procedures of the center. The inclusion criteria were patients with advanced keratoconus defined by stage 3 or more regarding AmslereKrumeich classification (Table 1).22 To avoid bias, only the most advanced eye was selected for each patient. A minimum follow-up of 24 months was required. Exclusion criteria comprised any type of prior ocular surgery, including corneal collagen cross-linking, corneal rings, or keratoplasty, and patients who showed signs of acute or healed corneal hydrops confirmed by OCT at the first consultation. Each eye had a characteristic keratoconic appearance on the topographic map (asymmetric bowtie with skewed radial axis, central or inferior steep zone, or claw shape). The keratoconic slit-lamp findings included Munson’s sign, Vogt’s striae, Fleischer ring, apical scar, apical thinning, Rizutti sign, and corneal scars.
Optical Coherence Tomography A Fourier-domain OCT system (RTVue; Optovue, Inc., Fremont, CA) was used in this study. The system works at an 830-nm wavelength, with a scan speed of 26.000 axial scans per second and a depth resolution capacity of 5 mm in tissue. The corneal adaptor module provides the 6-mm scan diameter pachymetry map and the minimum corneal thickness. For each patient, 3 highresolution scans were made across the conus to evaluate structural corneal changes. Corneal epithelial thickness and stromal thickness were measured manually as the distances between the airetear and the epitheliumeBowman layer interfaces and between the Bowman layerestroma and the stromaeDescemet’s membrane interfaces, respectively. The cursors were placed perpendicular to the anterior ocular surface at the point of measurement. The OCT examinations were performed every 4 months at the follow-up. To evaluate interobserver variability, all measurements were made by 2 different examiners (T.H. and I.G.). All OCT scans
Table 2. Optical Coherence Tomography Keratoconus Classification Stage 1 2 3
4 5
1654
Characteristic Thinning of epithelial and stromal layers at the conus. Corneal layers have a normal aspect. Hyperreflective anomalies occurring at the Bowman’s layer level and epithelial thickening at the conus: 2a, clear stroma; 2b, stromal opacities. Posterior displacement of the hyperreflective structures occurring at the Bowman’s layer level with increased epithelial thickening and stromal thinning: 3a, clear stroma; 3b, stromal opacities. Pan-stromal scar Hydrops stage: 5a, acute onset, characterized by the rupture of Descemet’s membrane with delamination of collagen lamellae, large fluid-filled intrastromal cysts, and the formation of epithelial edema; 5b, healing stage, pan-stromal scarring with a remaining aspect of Descemet’s membrane rupture.
Fuentes et al
OCT Anatomic Risk Factors of Corneal Hydrops
Table 3. Baseline Characteristics of Patients Grouped According to Optical Coherence Tomography Keratoconus Classification
Age, yrs (mean, SD) Range Hydrops observed during the study (n) OCT minimum pachymetry, mm (mean, SD) Range OCT epithelial thickness at the conus, mm (mean, SD) Range OCT stromal thickness at the conus, mm (mean, SD) Range OCT epithelial-to-stromal thickness ratio (mean, SD) Range
Stage 1 (n [ 101)
Stage 2 (n [ 60)
Stage 3 (n [ 20)
Stage 4 (n [ 10)
33 (13) (17e73) 0 402 (39) (323e499) 41 (6) (30e55) 367 (44) (272e464) 0.11 (0.03) (0.07e0.19)
33 (10) (22e58) 0 305 (67) (216e419) 74 (20) (52e109) 242 (63) (116e367) 0.34 (0.15) (0.13e0.64)
26 (8) (16e43) 11 274 (51) (149e375) 128 (30) (80e178) 161 (47) (62e231) 0.9 (0.48) (0.44e2.44)
32 (6) (20e37) 0 333 (45) (300e434) 101 (7) (92e113) 232 (22) (196e258) 0.44 (0.05) (0.4e0.53)
P Value 0.0426y <0.0001* <0.0001y <0.0001y <0.0001y <0.0001y
OCT ¼ optical coherence tomography; SD ¼ standard deviation. *Pearson’s chi-square test. y KruskaleWallis 1-way analysis of variance comparing the 4 groups together.
were classified according to our OCT keratoconus classification (Table 2).15 Several anatomic features at the keratoconus cone were analyzed, including the epithelial thickness (microns), stromal thickness (microns), epithelial-to-stromal thickness ratio, presence of hyperreflective anomalies occurring at the Bowman’s layer level, and presence of Vogt’s striae and stromal opacities.
Topography Corneal topography was obtained with the Orbscan IIz (Bausch & Lomb Surgical, Rochester, NY). The steepest and the mean simulated keratometry values were recorded.
Statistical Analysis Results are presented as mean standard deviation for continuous variables and as proportions (percentages) for categoric variables. Nonparametric ManneWhitney and KruskaleWallis tests were used to compare continuous data as appropriate. The Wilcoxon signed-rank test was used to statistically evaluate comparisons between continuous data before and after hydrops. For binary outcomes, the stratified Cochran chi-square test and the Fisher exact test were used for intergroup comparisons of proportions when appropriate. Linear regression, with evaluation of the Spearman coefficient, was used to analyze associations between 2
Figure 1. Optical coherence tomography (OCT) scans show the evolution of 3 patients who developed corneal hydrops (before hydrops, acute onset, and healing stage). A, D, G, Stage 3a before hydrops; important epithelium thickening, stromal thinning, and anterior hyperreflective anomalies occurring at the Bowman’s layer level with clear stroma. Arrows indicate Vogt’s striae. B, E, H, Corneal hydrops (stage 5a); Descemet’s membrane rupture with delamination of collagen lamellae, large fluid-filled intrastromal cysts, and epithelial edema formation. C, F, I, Healing stage (stage 5b); pan-stromal scarring with a remaining aspect of Descemet’s membrane rupture. All scale bars represent 250 mm.
1655
1656
0.0037 61 (3)/55 (4) 62/56 65/55 56/51 59/52 61/57
D ¼ diopters; L ¼ left; M ¼ male; OCT ¼ optical coherence tomography; R ¼ right; SD ¼ standard deviation. *Wilcoxon signed-rank test. Comparison before and after hydrops.
60/53 58/55 60/56 63/59 60/48
66/53 67/60
67/60
0.0038 67 (3)/58 (4) 66/60 70/58 63/53 67/62 63/60 66/57 72/63 65/51
0.8/0.1 0.9/0.1
69/63
0.0038 1 (0.6)/0.2 (0.1) 0.7/0.1 0.9/0.2 0.8/0.2 0.4/0.2 0.6/0.2 1.6/0.3 0.6/0.3 1.4/0.3
204/524 198/547
2.4/0.2
0.001 164 (52)/381 (111) 180/450 160/320 152/348 231/450 190/470 89/306 210/231 115/308
160/73 176/73
73/239
142 (26)/72 (12) 120/66 145/72 127/83 102/75 114/76 141/80 130/60 165/90
M 16 L 3 365/385 M 16 R 3 338/328
Sex Age (yrs) Eye OCT stage OCT minimum pachymetry (mm) (before/ after hydrops) OCT epithelial thickness at the conus (mm) (before/after hydrops) OCT stromal thickness at the conus (mm) (before/after hydrops) OCT epithelial-to-stromal thickness ratio (before/after hydrops) Steepest keratometry (D) (before/after hydrops) Mean keratometry (D) (before/after hydrops)
178/47
0.0049 285 (51)/318 (48)
23.7 (5.9)
M 23 L 3 285/326 M 22 R 3 240/290 M 28 L 3 360/391 M 28 L 3 295/300 M 34 L 3 287/290 M 20 L 3 227/241 M 31 R 3 259/278 M 23 L 3 262/375
9 8 7 6 5 4 3 2 1 Case
Table 4. Characteristics of Patients before and after Corneal Hydrops
A total of 267 eyes of 267 keratoconic patients were in stage 3 or 4 of the AmslereKrumeich classification of keratoconus. However, 73 eyes underwent corneal surgery (corneal crosslinking, n ¼ 11; corneal rings, n ¼ 15; and keratoplasty, n ¼ 47) during the followup, and 3 eyes showed clinical signs of corneal hydrops at the first consultation and were removed from our series. The remaining 191 eyes were included in the study. The analysis of OCT images yielded highly reproducible epithelial and stromal thickness measurements, evidenced by excellent ICC parameters (0.92 for epithelial thickness and 0.97 for stromal thickness). The repartition of all patients according to our OCT classification and their anatomic characteristics are summarized in Table 3. Eleven cases (5.8%) in our cohort developed a corneal hydrops during a mean follow-up of 30 months (range, 24e36 months). In acute hydrops onset, OCT showed a typical aspect of Descemet’s membrane rupture with delaminations of collagen lamellae, large fluid-filled intrastromal cysts, and epithelial edema formation. The Descemet’s membrane rupture and the edematous corneal changes were mainly located at the conus area. At the healing stage, panstromal scarring occurred with a remaining aspect of Descemet’s membrane rupture (Fig 1). All 11 patients who developed corneal hydrops in our study were classified as OCT stage 3 during their last visit before the acute phase. Indeed, this stage was associated with the occurrence of corneal hydrops (P < 0.0001, Pearson’s chi-square test). The characteristics of patients before and after the corneal hydrops are summarized in Table 4. All these patients were male and younger than patients who did not develop hydrops in the study period. Before corneal hydrops, the average epithelial and stromal thicknesses at the cone were 142 (26) mm and 164 (52) mm, respectively. At the scarring stage, the mean values were 72 (12) mm and 381 (111) mm for epithelial and stromal thicknesses, respectively, demonstrating a stromal thickening and epithelial thinning at the conus after hydrops. The average steepest keratometry decreased from 67 (3) diopters before hydrops to 58 (4) diopters at the healing stage (6 months after hydrops). The comparative clinical and anatomic characteristics between those with hydrops and those with no hydrops are shown in Table 5. Before developing corneal hydrops, the minimal corneal and stromal thicknesses at the conus were significantly thinner and the epithelial thickness was significantly thicker in the hydrops group compared with the no hydrops group. The epithelium-to-stroma thicknesses ratio was significantly higher in patients who developed corneal hydrops. In addition, the presence of hyperreflectivity at the Bowman’s layer and Vogt’s
10
Results
M 20 R 3 221/290
11
Mean ± (SD)
P Value*
continuous variables. Interobserver agreement for the measurement of epithelial and stromal thickness using OCT images was assessed by the intraclass correlation coefficient (ICC), and the limits of agreement were assessed using the BlandeAltman method. The ICC was considered reliable if the values were between 0.4 and 0.75, and values >0.75 were considered excellent. Because many comparisons were performed, the Bonferroni correction was used to correct the P value. P values 0.05 were considered statistically significant. Univariate comparisons by logistic regression were performed to identify associations between the occurrence of hydrops and other variables under study. Variables found to be associated with a P value < 0.05 were then subjected to multivariate logistic regression. P 0.05 was considered statistically significant. Statistical analysis was carried out using SPSS for Windows version 16.0 (SPSS, Inc., Chicago, IL).
0.001
Ophthalmology Volume 122, Number 8, August 2015
Fuentes et al
OCT Anatomic Risk Factors of Corneal Hydrops
Table 5. Comparative Clinical and Anatomic Characteristics between the Hydrops and No Hydrops Groups Hydrops (n [ 11)
No Hydrops (n [ 180)
23.7 (5.9) (16e34) 11/0 285 (51) (221e365) 142 (26) (102e178) 164 (52) (73e231) 1 (0.6) (0.4e2.4) 11 8 67 (2.8) (63e72)
32.7 (11.3) (17e73) 115/65 355 (73) (80e500) 62 (31) (30e181) 300 (92) (62e464) 0.26 (0.23) (0.07e1.4) 52 31 57.6 (7.4) (45e77)
Age, yrs (mean, SD) Range Sex (M/F) OCT minimum pachymetry, mm (mean, SD) Range OCT epithelial thickness at the conus, mm (mean, SD) Range OCT stromal thickness at the conus, mm (mean, SD) Range OCT epithelial-to-stromal thicknesses ratio, mm (mean, SD) Range OCT hyperreflectivity at Bowman’s layer OCT Vogt’s striae visualization Steepest keratometry values, D (mean, SD) Range
P Value 0.0108* 0.0171y 0.0044* <0.0001* 0.0004* 0.0123* <0.001y <0.001y 0.7441*
D ¼ diopters; F ¼ female; M ¼ male; OCT ¼ optical coherence tomography; SD ¼ standard deviation. *ManneWhitney test. y Fisher exact test.
striae visualized on OCT were significantly more frequent in the hydrops group. With respect to the steepest keratometry, there was no significant difference between both groups. To define patients with a higher risk of hydrops among stage 3 OCT cases (n ¼ 20), an additional intragroup comparative analysis was performed between patients who developed hydrops and patients who did not develop acute keratoconus during the follow-up, as shown in Table 6. Of note, patients with higher epithelial thickness (P ¼ 0.042) and without stromal scarring (or stage 3a) (P ¼ 0.007) were at higher risk of developing corneal hydrops. However, no case of acute keratoconus occurred in eyes with stromal opacities (Fig 2). Variables associated with a P value < 0.05 were then subjected to logistic regression analysis. Of note, an increased epithelial thickness at the conus was the only risk factor for the occurrence of hydrops corresponding to patients classified as stage 3 OCT (P ¼ 0.02).
Discussion Corneal hydrops is a rare event and has been reported at 3% in the literature.1e3 The incidence of acute hydrops in our study was higher and could be explained by selection bias in
our study, which considered more severe keratoconus cases than those in the literature. The predominance of male patients and young patients, which occurred in our series, has been reported.1e3 This study describes new anatomic risk factors of corneal hydrops that should be considered when evaluating patients with keratoconus. Indeed, surgical and visual prognoses are worse in patients who developed acute keratoconus because DALK is difficult and complete dissection is rarely possible.9,10 The morphologic stage 3a appearance of the OCT classification indicated a higher risk to develop corneal hydrops. The principal OCT anatomic findings at this stage are hyperreflective anomalies occurring at the Bowman’s layer level with marked epithelial thickening and stromal thinning. Anterior hyperreflectives at the Bowman’s layer level were found in all patients before developing corneal hydrops. In a recent confocal microscopy study performed on 10 keratoconic eyes that developed corneal hydrops, a similar hyperreflective band was found in all eyes.23 These anomalies correspond to fibrillar degeneration and fibroblastic
Table 6. Comparison between Hydrops and No Hydrops Groups, among Patients with Stage 3 on Optical Coherence Tomography Hydrops Group (n [ 11) OCT minimum corneal pachymetry, mm (mean, SD) OCT epithelial thickness at the conus, mm (mean, SD) OCT stromal thickness at the conus, mm (mean, SD) OCT epithelial-to-stromal thicknesses ratio, mm (mean, SD) OCT Vogt’s striae visualization Absence of stromal scar (stage 3a OCT) Steepest keratometry values, D (Mean, SD)
285 142 164 1
(51) (26) (52) (0.6) 8 11 67 (2.8)
No Hydrops Group (n [ 9) 260 111 158 0.8
(50) (29) (42) (0.3) 1 2 65 (10)
P Value 0.5737* 0.042* 0.6454* 0.2445* 0.0098y 0.007y 0.7302*
D ¼ diopters; OCT ¼ optical coherence tomography; SD ¼ standard deviation. *ManneWhitney test. y Fisher exact test.
1657
Ophthalmology Volume 122, Number 8, August 2015
Figure 2. The optical coherence tomography scans show aspects of advanced keratoconus with stromal opacities. A, Stage 3b; important epithelium thickening, stromal thinning, and stromal opacity. B, Stage 4; pan-stromal scar. All scale bars represent 250 mm.
accumulation in the anterior stroma beneath Bowman layer breaks and would translate into an active inflammatory disease factor.15,24,25 This activation is thought to be triggered by cytokines (including interleukin-1 and Fas-ligand) released from the damaged epithelial cells.26e29 The epithelium-to-stromal thicknesses ratio at the conus was associated with a high risk for corneal hydrops development. Stromal thickness, especially that of the anterior stroma, increase corneal rigidity and account for maintenance of corneal curvature.30,31 Epithelial thickening in advanced keratoconus cases seems to be an important indirect predictive factor of corneal fragility and therefore hydrops. This is a consequence of the impressive capacity of corneal epithelium to compensate irregularities caused by the extreme thinning of the corneal stroma in keratoconic patients.15,32e36 Furthermore, when analyzing the morphologic appearance of prehydrops stage 3 OCT images, there is a marked compensatory epithelial hyperplasia in a specific area of extreme stromal thinning corresponding to the keratoconus cone. This area corresponds to the weakest zone of the cornea, explaining the occurrence of Descemet’s membrane rupture and the predominance of the edematous corneal changes at the conus area. Vogt’s striae have the aspect of dark parallel lines running through the entire stromal thickness, similar to those observed in confocal microscopy.15,25 These lines have been suggested to represent collagen lamellae under stress. We found that the presence of Vogt’s striae increases the risk of corneal hydrops when they are present in keratoconus with thin stromas, as found in OCT stage 3a. However, the presence of Vogt’s striae was not significantly associated with corneal hydrops occurrence in the multivariate analysis. Of note, no patient with stromal opacities (3b and 4 stages) developed this complication. We hypothesize that the presence of stromal scarring increases corneal rigidity and is a preventive factor of corneal hydrops, as found in our series. The OCT stage 3 corresponds to advanced keratoconus, and the stromal thickness is generally inferior to 300 mm, not allowing cross-linking or corneal ring implantation procedures.15 Deep anterior lamellar keratoplasty is generally indicated and would also prevent the corneal hydrops, especially in patients with other nonanatomic risk factors, such as young age and atopy. At the healing stage of corneal hydrops, significant stromal thickening and cornea flattening occur.1,4 The stromal thickening would be the result of the inflammatory
1658
and healing processes triggered after hydrops.23 As described by Stone et al37 in a histopathologic study, the rupture and detachment of Descemet’s membrane result in the formation of extensive ledges with a fibrous metaplasia of the endothelium with secretion of extensive fibrillar and basement membrane material. However, the beneficial effect of this “natural cross-linking” is limited by the residual pan-stromal corneal scar, which is located centrally or paracentrally at the corneal cone, in most cases limiting visual acuity rehabilitation.1,16
Study Limitations Potential limitations of our study may be the small number of patients (n ¼ 11) who developed corneal hydrops. However, this complication is rare, and patients usually consult once corneal hydrops occurs and not before. Furthermore, all patients in our series had a close corneal OCT follow-up permitting an accurate structural analysis of anatomic profiles at risk of corneal hydrops. In conclusion, we describe new anatomic corneal aspects predisposing to corneal hydrops. Stage 3a of the OCT keratoconus classification (characterized by increased epithelial thickening, stromal thinning at the conus, anterior hyperreflectives at the Bowman’s layer level, and absence of stromal scarring) is associated with a high risk for developing corneal hydrops and should be taken into account by the clinician in the evaluation of eyes with keratoconus and in the planning of corneal keratoplasty. Additional studies with longer follow-up are needed to confirm our findings.
References 1. Tuft SJ, Gregory WM, Buckley RJ. Acute corneal hydrops in keratoconus. Ophthalmology 1994;101:1738–44. 2. Basu S, Vaddavalli PK, Ramappa M, et al. Intracameral perfluoropropane gas in the treatment of acute corneal hydrops. Ophthalmology 2011;118:934–9. 3. Grewal S, Laibson PR, Cohen EJ, Rapuano CJ. Acute hydrops in the corneal ectasias: associated factors and outcomes. Trans Am Ophthalmol Soc 1999;97:187–203. 4. Amsler MM. Quelques données du problème du kératocône. Bull Soc Belge Ophtalmol 1961;128:331–54. 5. Sharma R, Titiyal JS, Prakash G, et al. Clinical profile and risk factors for keratoplasty and development of hydrops in north Indian patients with keratoconus. Cornea 2009;28:367–70.
Fuentes et al
OCT Anatomic Risk Factors of Corneal Hydrops
6. Stoiber J, Muss W, Ruckhofer J, Grabner G. Acute keratoconus with perforation in a patient with Down’s syndrome. Br J Ophthalmol 2003;87:120. 7. Koenig SB, Smith RW. Keratoconus and corneal hydrops associated with compulsive eye rubbing. Refract Corneal Surg 1993;9:383–4. 8. Rehany U, Rumelt S. Corneal hydrops associated with vernal conjunctivitis as a presenting sign of keratoconus in children. Ophthalmology 1995;102:2046–9. 9. Das S, Dua N, Ramamurthy B. Deep lamellar keratoplasty in keratoconus with healed hydrops. Cornea 2007;26:1156–7. 10. Chew A, Mehta J, Tan D. Deep lamellar keratoplasty after resolution of hydrops in keratoconus. Cornea 2011;30:454–9. 11. Fan Gaskin JC, Patel DV, McGhee CNJ. Acute corneal hydrops in keratoconus - new perspectives. Am J Ophthalmol 2014;157:921–8. 12. Coster DJ, Lowe MT, Keane MC, Williams KA; for the Australian Corneal Graft Registry Contributors. A comparison of lamellar and penetrating keratoplasty outcome: a registry study. Ophthalmology 2014;121:979–87. 13. Borderie VM, Sandali O, Bullet J, et al. Long-term results of deep anterior lamellar versus penetrating keratoplasty. Ophthalmology 2012;119:249–55. 14. Borderie VM, Guilbert E, Touzeau O, Laroche L. Graft rejection and graft failure after anterior lamellar versus penetrating keratoplasty. Am J Ophthalmol 2011;151:1024–9. 15. Sandali O, El Sanharawi M, Temstet C, et al. Fourier-domain optical coherence tomography imaging in keratoconus: a corneal structural classification. Ophthalmology 2013;120: 2403–12. 16. Maharana PK, Sharma N, Vajpayee RB. Acute corneal hydrops in keratoconus. Indian J Ophthalmol 2013;61:461–4. 17. Basu S, Vaddavalli PK, Vemuganti GK, et al. Anterior segment optical coherence tomography features of acute corneal hydrops. Cornea 2012;31:479–85. 18. Kallel S, Tahiri Joutei Hassani R, Liang H, et al. “En face” anterior segment optical coherence tomography findings in acute corneal hydrops [in French]. J Fr Ophtalmol 2014;37: 605–12. 19. Kucumen BR, Yenerel NM, Gorgun E, Dinc UA. Anterior segment optical coherence tomography findings of acute hydrops in a patient with keratoconus. Ophthalmic Surg Lasers Imaging 2010;41(suppl):114–6. 20. Ueno H, Matuzawa A, Kumagai Y, et al. Imaging of a severe case of acute hydrops in a patient with keratoconus using anterior segment optical coherence tomography. Case Rep Ophthalmol 2012;3:304–10. 21. Basu S, Mukhopadhyay S. Acute corneal hydrops. Ophthalmology 2012;119:2197–2197.e1; author reply 2198.
22. Krumeich JH, Kezirian GM. Circular keratotomy to reduce astigmatism and improve vision in stage I and II keratoconus. J Refract Surg 2009;25:357–65. 23. Lockington D, Fan Gaskin JC, McGhee CNJ, Patel DV. A prospective study of acute corneal hydrops by in vivo confocal microscopy in a New Zealand population with keratoconus. Br J Ophthalmol 2014;98:1296–302. 24. Chi HH, Katzin HM, Teng CC. Histopathology of keratoconus. Am J Ophthalmol 1956;42:847–60. 25. Efron N, Hollingsworth JG. New perspectives on keratoconus as revealed by corneal confocal microscopy. Clin Exp Optom 2008;91:34–55. 26. Jun AS, Cope L, Speck C, et al. Subnormal cytokine profile in the tear fluid of keratoconus patients. PLoS One 2011; e16437:1e8. 27. Wilson SE, He YG, Weng J, et al. Epithelial injury induces keratocyte apoptosis: hypothesized role for the interleukin-1 system in the modulation of corneal tissue organization and wound healing. Exp Eye Res 1996;62:325–7. 28. Fabre EJ, Bureau J, Pouliquen Y, Lorans G. Binding sites for human interleukin 1 alpha, gamma interferon and tumor necrosis factor on cultured fibroblasts of normal cornea and keratoconus. Curr Eye Res 1991;10:585–92. 29. Lema I, Durán JA. Inflammatory molecules in the tears of patients with keratoconus. Ophthalmology 2005;112:654–9. 30. Müller LJ, Pels E, Vrensen GF. The specific architecture of the anterior stroma accounts for maintenance of corneal curvature. Br J Ophthalmol 2001;85:437–43. 31. Maurice DM, Monroe F. Cohesive strength of corneal lamellae. Exp Eye Res 1990;50:59–63. 32. Reinstein DZ, Archer TJ, Gobbe M. Corneal epithelial thickness profile in the diagnosis of keratoconus. J Refract Surg 2009;25:604–10. 33. Reinstein DZ, Gobbe M, Archer TJ, et al. Epithelial, stromal, and total corneal thickness in keratoconus: three-dimensional display with artemis very-high frequency digital ultrasound. J Refract Surg 2010;26:259–71. 34. 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: 2425–33. 35. Haque S, Simpson T, Jones L. Corneal and epithelial thickness in keratoconus: a comparison of ultrasonic pachymetry, Orbscan II, and optical coherence tomography. J Refract Surg 2006;22:486–93. 36. Sykakis E, Carley F, Irion L, et al. An in depth analysis of histopathological characteristics found in keratoconus. Pathology 2012;44:234–9. 37. Stone DL, Kenyon KR, Stark WJ. Ultrastructure of keratoconus with healed hydrops. Am J Ophthalmol 1976;82:450–8.
Footnotes and Financial Disclosures Originally received: January 26, 2015. Final revision: April 21, 2015. Accepted: April 23, 2015. Available online: June 1, 2015.
Analysis and interpretation: Fuentes, Sandali, El Sanharawi, Basli, Borderie, Bouheraoua, Laroche Obtained funding: Not applicable Manuscript no. 2015-133.
Centre Hospitalier National d’Ophtalmologie des XV-XX, Pierre & Marie Curie University Paris 6, Paris, France. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Author Contributions: Conception and design: Sandali
Overall responsibility: Fuentes, Sandali, Borderie, Bouheraoua, Laroche Abbreviations and Acronyms: DALK ¼ deep anterior lamellar keratoplasty; ICC ¼ intraclass correlation coefficient; OCT ¼ optical coherence tomography. Correspondence: Otman Sandali, MD, Centre Hospitalier National d’Ophtalmologie des XV-XX, 28 Rue de Charenton, 75571 Paris, France. E-mail: [email protected].
Data collection: Fuentes, Hamiche, Goemaere
1659