Sustainability of Pain Relief After Corneal Collagen Cross-Linking in Eyes With Bullous Keratopathy

Transcription

1 OIGINA CINICA STUDY Sustainability of Pain elief After Corneal Collagen Cross-inking in Eyes With Bullous Keratopathy Takashi Ono, D, Yosai ori, D, PhD, yohei Nejima, D, PhD, iyuki Ogata, Keiichiro inami, PhD, and Kazunori iyata, D, PhD Purpose: This study aimed to examine the efficiency and sustainability of pain relief produced by corneal collagen cross-linking (CX) in eyes with bullous keratopathy (BK) and to explore the histopathological changes in the stroma by using in vivo confocal microscopy. Design: Prospective observational case series. ethods: ourteen eyes of 14 consecutive BK patients were treated with CX with dehydration of the corneal stroma and followed up for 1 year after treatment. The best-corrected visual acuity (BCVA), intraocular pressure, and central corneal thickness (CCT) were examined before the treatments and up to 1 year after. The intensity and frequency of pain were graded on a scale from 0 (minimum) to 10 (maximum). At 1 year after CX, the corneal stroma was observed using confocal microscopy at depths of 100 and 200 μm. esults: The BCVA and CCT did not change significantly. The mean pain intensity and frequency scores were 5.6 and 4.6, respectively, before treatment. The intensity score significantly decreased at 1 week and onward, and the frequency score significantly decreased over 6 months after treatment. The confocal microscopy images showed that keratocytes and nerve fibers were rare and sparsely distributed in the stroma 1 year after CX. Conclusions: The CX efficiently relieved pain due to BK for 1 year after treatment. The in vivo confocal microscopy observations and unchanged CCT demonstrated that the persistence of pain relief was due to the inadequate regeneration of nerve fibers in the corneal stroma. Key Words: corneal collagen cross-linking, bullous keratopathy, pain relief (Asia-Pac J Ophthalmol 2018;7: ) Corneal collagen cross-linking (CX) creates tight connections among corneal collagen fibers by inducing radicals using riboflavin and ultraviolet (UV) A radiation. Cross-linking has been used to treat keratoconus and postsurgical ectasia to increase the stiffness of the corneal stroma. 1 3 The induction of radicals and increase in corneal stiffness produced by CX are utilized for the treatment of infection, stromal melt, and bullous keratopathy (BK). 3 5 Bullous keratopathy results in accumulation of fluid in the extracellular space between the stromal lamellae because of dysfunction in endothelial pumping, thereby leading to visual impairment and ocular pain. Treatment with CX has been demonstrated to effectively reduce central corneal thickness rom iyata Eye Hospital, iyazaki, Japan. eceived for publication January 17, 2018; accepted June 22, The authors have no funding or conflicts of interest to declare. eprints: Keiichiro inami, PhD, iyata Eye Hospital, 6-3 Kurahara-cho, iyakonojo, iyazaki, , Japan. minami@miyata-med.ne.jp. Copyright 2018 by Asia Pacific Academy of Ophthalmology ISSN: DOI: /APO (CCT) in the edematous stroma, 6 12 improve visual acuity, 8 and relieve pain due to BK 10 in eyes with severe BK. However, these results are only effective for up to 3 months. Previous assessments were conducted on patients with eyes at end-stage BK that were scheduled to undergo keratoplasty. On the other hand, the efficacy of CX in eyes with BK in a relatively early stage has not yet been assessed. We speculated that endothelial pumping would be partially effective and that the effect of CX would be enhanced because of less accumulation of extracellular fluid. urthermore, change in the corneal biomaterial property by CX is facilitated in eyes with thinner corneas, 13,14 so that dehydration of the corneal stroma after deepithelization would enhance the clinical effect of CX and prolong pain relief. This prospective observational case study was conducted to examine the effect and sustainability of pain relief produced by CX in BK eyes for a 1-year period and to observe the superior stroma 1 year after treatment by using in vivo confocal microscopy. ATEIAS AND ETHODS This study was approved by the ethics committee of iyata Eye Hospital (identifier: CS-134) and followed the tenets of the Declaration of Helsinki. Written informed consent was obtained from all patients. The study included 14 eyes of 14 consecutive patients who underwent CX as a palliative therapy for relieving pain until transplantation surgery at iyata Eye Hospital and were followed up for 1 year. Inclusion criteria were BK in a relatively early stage, which was defined as corneal endothelial cell density (ECD) less than 500 cell/mm 2, no significant opacity in the corneal stroma, pain symptoms, and no need of immediate keratoplasty. Eyes with a CCT less than 450 μm (400 μm without the corneal epithelium), a history of corneal incision surgery after the incidence of BK, deficiency in corneal epithelium restoration, corneal melting, or herpetic keratitis were excluded. The ages of the patients ranged from 52 to 98 years, and the mean age was 78.6 ± 10.9 (SD) years. The study included 5 men and 9 women. The primary causes of BK included argon laser iridotomy in 7 eyes, uveitis in 2 eyes, iridocorneal endothelial syndrome in 2 eyes, and penetrating keratoplasty, filtration surgery, and primary angle closure in 1 eye each (Table 1). CX and Examinations The epithelium of the cornea was removed in a region of 8 to 9 mm in diameter. The BK cornea was swollen due to excess absorption of fluid and the CCT was considerably thicker, so the corneal stroma was dehydrated until the CCT was reduced to μm. After that, 0.1% riboflavin was instilled every 2 minutes for 30 minutes. The cornea in the central 8.0 mm diameter was irradiated with UV-A (a wavelength of 365 nm and a power of 3.0 mw/cm 2 ) for 30 minutes by using a CC-365 (PESCHKE Trade, Huenenberg, Switzerland), while continuing the riboflavin 291

2 Ono et al TABE 1. Demographic Data of Patients Sex Age, y / Primary Cause of BK Uveitis, acute glaucoma attack Uveitis Penetrating keratoplasty Primary angle closure Glaucoma filtration surgery Iridocorneal endothelial syndrome Iridocorneal endothelial syndrome IGUE 1. Changes in the intensity and frequency of pain after CX. After treatment, the pain intensity and frequency scores significantly decreased (P < and 0.026, respectively, Steel-Dwass multiple comparison). instillation. An accelerated radiation protocol consisting of 5 minutes of radiation at a power of 18.0 mw/cm 2 was used for 9 of the 14 eyes (#6 onward in Table 1). A contact lens was used until the corneal epithelium covered the corneal surface. The postoperative regimen was 0.5% levofloxacine and 0.1% fluorometholone eye drops 4 times daily for 3 months, then gradually decreased. Neither infection nor the progress of corneal stroma opacity was observed during the observation period. Throughout the observation period, the corneal epithelium layer was stable and corneal disorders such as neuroparalytic keratitis were not observed. The best-corrected visual acuity (BCVA), intraocular pressure (IOP), and CCT were examined before treatment and 1 week; 1, 3, and 6 months; and 1 year after treatment. The BCVA was measured using a 5-meter Snellen chart and converted to logarithm of the minimum angle of resolution (loga) for analysis. The IOP was measured using a noncontact tonometer (NT-4000, Nidek, Gamagori, Japan), and CCT was measured using an anterior-segment optical coherence tomographer (SS-1000, Tomey, Nagoya, Japan). The ECD was measured using noncontact specular microscopy (A-3509, Konan, Nishinomiya, Japan). An endothelial cell image around the center of the cornea was captured, all endothelial cells within the captured image of an area of mm were automatically traced, and ECD (cell/mm 2 ) was determined as described previously. 15 The intensity of pain was assessed using a scale from 0 to 10, with increments of 1, where 0 indicated a complete lack of pain and 10 indicated the most painful 11 ; the frequency of pain was scaled in the same manner, and 10 indicated always painful. At 1 year after CX, the central corneal stroma was observed using a confocal microscope (HT3 ostock cornea module, Heidelberg Engineering, Heidelberg, Germany) at depths of 100 and 200 μm, as demarcation lines are normally found at depths of μm and μm with the use of the standard and accelerated radiation protocols, respectively. 2 icrostructures in the corneal stroma were evaluated. Within the full μm frame of each image, the number of keratocyte nuclei that were in focus and entirely contained within the frame was counted, and the total length of nerve fibers was measured. The densities of keratocytes (cell/mm 2 ) and nerves (mm/mm 2 ) were calculated. 16 Statistical Analysis Changes in the BCVA, IOP, CCT, and intensity and frequency scores of pain symptoms were examined using the Kruskal-Wallis test and then Steel-Dwass multiple comparison. A P value less than 0.05 was considered to indicate a significant difference. The results are expressed as the mean ± standard deviation, unless otherwise specified. ESUTS No postoperative complication such as haze and stromal scars were observed throughout the follow-up period. The BCVA, IOP, and CCT are shown in Table 2. The BCVA did not significantly improve after treatment (P = 0.98, Kruskal-Wallis test), and IOP and CCT did not change from their preoperative levels throughout the postoperative period (P = 0.55 and 0.27, respectively). igure 1 shows changes in the pain intensity and frequency scores. The mean pain intensity score was 5.6 ± 2.9 (range, 0 9) before treatment, and it significantly decreased after treatment (P < 0.022, Steel-Dwass multiple comparison). The pain intensity 1 week after CX was 1.6 ± 2.4 (range, 0 8), which did not change further during the study; the score at 1 year after treatment was 1.5 ± 2.2 (range, 0 6). The frequency score showed no difference before treatment (4.6 ± 3.1) and during 3 months thereafter TABE 2. Changes in BCVA, IOP, and CCT Before CX After 1 wk 1 mo 3 mo 6 mo 1 y BCVA, loga 2.23 ± ± ± ± ± ± 0.53 IOP, mm Hg 16.1 ± ± ± ± ± ± 6.3 CCT, μm 738 ± ± ± ± ± ± 99 Data presented as mean ± standard deviation Asia-Pacific Academy of Ophthalmology

3 Before CX After 12 months BK Pain elief by CX slightly, but the nerve fibers rarely regenerated. DISCUSSION 100 μm depth (P > 0.064). However, significant decreases were found at 6 and 12 months after (P < 0.026); the score at 1 year after treatment was 1.1 ± 1.6 (range, 0 5). All patients eyes except for 1 (#13 in Table 1) were examined using confocal microscopy 1 year after treatment. igure 2 shows typical images of the corneal stroma at depths of 100 and 200 μm from the corneal surface before and 12 months after treatment. Before CX, the keratocytes and nerve fibers in the stroma appeared as bright objects (arrows in ig. 2), and the extracellular matrix was observed in the stroma. Postoperatively, the keratocytes and nerve fibers were rarely observed. igures 3 and 4 show the confocal images of all eyes 1 year after CX at depths of 100 and 200 μm, respectively. At a depth of 100 μm, the gross densities of keratocytes and nerve fibers with respect to all images were 2.4 cells/mm2 and 0, respectively; keratocytes were found in 3 eyes (#3, 8, and 11 in ig. 3). At a depth of 200 μm, these gross densities were 1.4 cells/mm2 and 0.18 mm/mm2, respectively; keratocytes and nerve fibers were found in 3 eyes (#8, 9, and 11 in ig. 4) and 1 (#8) eye, respectively. However, the intensity and frequency of pain symptoms in the eye of patient #8 did not increase. The keratocytes proliferated In CX after stromal dehydration, no significant changes were observed in BCVA and CCT during 1 year of follow-up, whereas pain relief persisted after treatment. In the previous studies of CX treatment for BK eyes, the BCVA improved 1 month after CX but did not differ from the pretreatment BCVA at 6 months,10 and CCT significantly decreased and persisted for 6 months after treatment10,11 but increased later.12 In addition, pain symptoms were significantly relieved at 1 month but recovered to pretreatment levels by 6 months.11 The previous histopathological analysis of corneal buttons revealed that the stroma is compacted by CX, and this compaction alters with the progression of BK.9 The discrepancy of previous results from our results probably comes from the difference in severity of BK and the dehydration of the stroma before CX treatment. We supposed that the differences would allow the sustainability of the pain relief with no change in CCT. Ocular pain in BK is caused by defects in the corneal epithelium resulting from fluid accumulation in the stroma. Therefore, pain can be relieved by reducing fluid accumulation by compacting the stroma.6,9,16 Alternatively, corneal sensitivity can be blocked by ablating the subbasal nerves. In the current case series, pain relief persisted for up to 1 year, whereas CCT remained constant. The subbasal nerves require 12 to 24 months to regenerate to baseline levels after interventions, as indicated by in vivo confocal microscopy observations.17,18 Nerve fibers were observed 1 year after CX at the depth of 200 μm in 1 eye (#8) without an increase in the intensity and frequency of pain symptoms. In this eye, the nerve fibers were under regeneration; however, corneal sensitivity would still be blocked between the subbasal nerves and the depth of 200 μm. These findings suggest that the pain relief after CX persisted because of the blocking of corneal sensitivity. In vivo confocal microscopy 1 year after CX revealed that keratocytes and nerve fibers in the anterior stroma disappeared after treatment. A study of a human eye model and patients with keratoconus demonstrated that CX causes keratocyte apoptosis in the anterior stroma.17,19,20 Jordan et al17 also showed changes in the subbasal nerve plexus in response to CX for keratoconus. IGUE 3. Confocal microscopic images of eyes 1 year after CX at a IGUE 4. Confocal microscopic images of eyes 1 year after CX at a depth of 100 μm. Nerve fibers were absent, whereas keratocytes had regenerated (arrows). depth of 200 μm. Slight regeneration of keratocytes and nerve fibers were identified (arrows). 200 μm depth IGUE 2. In vivo confocal microscopic images of a representative case at depths of 100 and 200 μm before and 1 year after CX. The keratocytes and nerve fibers in the stroma that were observed before CX (arrows) disappeared after CX Asia-Pacific Academy of Ophthalmology 293

4 Ono et al TABE 3. CCT and Pain Intensity/requency Scores Using Conventional and Accelerated Protocols Before CX After 1 y Protocol Used Normal Accelerated Normal Accelerated CCT, μm Pain intensity score Pain frequency score 2.02 ± ± ± ± ± ± ± ± ± ± ± ± 99 Data presented as mean ± standard deviation. Taken together with the findings of the current study, we assumed that most of the nerve could be lost after treatment, and restoring the nerve density to pretreatment levels requires 1 year or longer. Croxatto et al 20 sequentially observed patients for up to 3 years and revealed that nerves continue to regenerate after CX and require 2 3 years to reach pre-cx levels. Although the current study did not examine keratoconus eyes, nerve fibers were still sparse at 1 year. Hence, the slow regeneration of the stromal nerves was hypothesized to facilitate pain relief for up to 1 year. onger-term observations are necessary to verify this speculation. The current cases were treated using 2 kinds of CX protocols. It has been reported that the protocols could result in difference responses in keratoconus cases. 21,22 Differences with the use of the accelerated protocol were examined regarding CCT and intensity and frequency of pain. As shown in Table 3, no significant difference was found (P > 0.12, ann-whitney U test). Although the sample sizes were too small, the results demonstrated that the use of the accelerated protocol would be comparable with the conventional protocol. We supposed that dehydration before CX treatment would minimize the different responses. ore cases are required for verifying the differences. The current study was subject to limitations. irst, the sample size (14 eyes) was not sufficient. We designed the pain relief treatment protocol for patients with BK without a plan for immediate keratoplasty, which limited the number of subjects. Although the above findings demonstrate that CX effectively relieves pain in BK for up to 1 year owing to the slow regeneration of nerves in the stroma, these findings need to be confirmed in a larger patient cohort. Next, the severity of BK was not quantitatively assessed. The durations of pain relief did not coincide with previous reports, 9 12 which could result from the difference in BK severity. However, the qualitative assessment of BK has not yet been standardized. Corneal opacity and endothelial pumping dysfunction need to be quantitatively scaled to improve the precision of analysis. Confocal microscope observations were limited in the central stroma and the images at 2 depths were analyzed. The effect of CX in a BK stroma would be less and not uniform as observed after CX of the keratoconus cases. It was anticipated that the effect in the peripheral area would alter with edematous conditions of surrounding stroma. Although uniformity was not confirmed, confocal microscopy was examined at the central area only. The current IOP values could be affected by CCT 23 and corneal biomechanical properties. 24 As the CCT did not change before and after CX, the CCT would have less influence on IOP measurements. However, the effect of change in the biomechanical properties by CX was not known. The use of IOP measurement compensating for the corneal biomechanical properties would be better for more precise analysis. 24 In conclusion, CX for BK with no significant stromal opacity effectively relieved pain for 1 year. This palliative treatment would be helpful in retaining quality of vision in BK eyes until the need for surgical treatment. In vivo confocal microscopy observations demonstrated that the sustainability of pain relief was because of the slow regeneration of nerves in the stroma. EEENCES 1. Spoerl E, Seiler T. Techniques for stiffening the cornea. J efract Surg. 1999;15: andleman JB, Khandelwal SS, Hafezi. Corneal cross-linking. Surv Ophthalmol. 2015;60: Sorkin N, Varssano D. Corneal collagen crosslinking: a systematic review. Ophthalmologica. 2014;232: Iseli HP, Thiel A, Hafezi, et al. Ultraviolet A/riboflavin corneal crosslinking for infectious keratitis associated with corneal melts. Cornea. 2008; 27: Suri K, Hammersmith K, Nagra PK. Corneal collagen cross-linking: ectasia and beyond. Curr Opin Ophthalmol. 2012;23: Wollensak G, Aurich H, Wirbelauer C, et al. Potential use of riboflavin/uva cross-linking in bullous keratopathy. Ophthalmic es. 2009;41: Bottos K, Hofling-ima A, Barbosa C, et al. Effect of collagen crosslinking in stromal fibril organization in edematous human corneas. Cornea. 2010;29: Ehlers N, Hjortdal J. iboflavin-ultraviolet light induced cross-linking in endothelial decompensation. Acta Ophthalmol. 2008;86: Arora, anudhane A, Saran K, et al. ole of corneal collagen crosslinking in pseudophakic bullous keratopathy: a clinicopathological study. Ophthalmology. 2013;120: Sharma N, oy S, aharana PK, et al. Outcomes of corneal collagen crosslinking in pseudophakic bullous keratopathy. Cornea. 2014;33: Ghanem C, Santhiago, Berti TB, et al. Collagen crosslinking with riboflavin and ultraviolet-a in eyes with pseudophakic bullous keratopathy. J Cataract efract Surg. 2010;36: Cordeiro Barbosa, Barbosa JB Jr, Hirai E, et al. Effect of crosslinking on corneal thickness in patients with corneal edema. Cornea. 2010; 29: Wollensak G, Spoerl E, Seiler T. Stress-strain measurements of human and porcine corneas after riboflavin-ultraviolet-a-induced cross-linking. J Cataract efract Surg. 2003;29: Vantipalli S, i J, Singh, et al. Effects of thickness on corneal biomechanical properties using optical coherence elastography. Optom Vis Sci. 2018;95: Ono T, Iida, Sakisaka T, et al. Effect of laser peripheral iridotomy using argon and neodymium-yag lasers on corneal endothelial cell density: 7-year longitudinal evaluation. Jpn J Ophthalmol. 2018;62: Wollensak G, Aurich H, Pham DT, et al. Hydration behavior of porcine cornea crosslinked with riboflavin and ultraviolet A. J Cataract efract Asia-Pacific Academy of Ophthalmology

5 BK Pain elief by CX Surg. 2007;33: Jordan C, Patel DV, Abeysekera N, et al. In vivo confocal microscopy analyses of corneal microstructural changes in a prospective study of collagen cross-linking in keratoconus. Ophthalmology. 2014;121: Erie JC, caren JW, Hodge DO, et al. ecovery of corneal subbasal nerve density after PK and ASIK. Am J Ophthalmol. 2005;140: Dhaliwal JS, Kaufman SC. Corneal collagen cross-linking: a confocal, electron, and light microscopy study of eye bank corneas. Cornea. 2009;28: Croxatto JO, Tytiun AE, Argento CJ. Sequential in vivo confocal microscopy study of corneal wound healing after cross-linking in patients with keratoconus. J efract Surg. 2010;26: Chan TC, Chow VW, Jhanji V, et al. Different topographic response between mild to moderate and advanced keratoconus after accelerated collagen cross-linking. Cornea. 2015;34: Ng A, Chan TC, Cheng AC. Conventional versus accelerated corneal collagen cross-linking in the treatment of keratoconus. Clin Exp Ophthalmol. 2016;44: Doughty J, Zaman. Human corneal thickness and its impact on intraocular pressure measures: a review and meta-analysis approach. Surv Ophthalmol. 2000;44: edeiros A, Weinreb N. Evaluation of the influence of corneal biomechanical properties on intraocular pressure measurements using the ocular response analyzer. J Glaucoma. 2006;15: Asia-Pacific Academy of Ophthalmology 295