Morphological characterization and clinical effects of stromal alterations after intracorneal ring segment implantation in keratoconus

Purpose To analyze the histological and (ultra)structural stromal tissue changes after femtosecond (Fs) laser–assisted intracorneal ring segment (ICRS) implantation and their refractive and topographic effects in patients with keratoconus. Methods This monocentric retrospective case series included 15 consecutive patients with clinical peri-segmental lamellar channel deposits after treatment with Fs-ICRS implantation for keratoconus. The stromal changes were investigated using in vivo confocal microscopy. Two patients underwent a penetrating keratoplasty after the Fs-ICRS implantation; the explanted corneas were processed for histopathology and transmission electron microscopy (TEM). Refractive and topographic effects were investigated comparing the uncorrected (UDVA) and corrected (CDVA) distance visual acuity, spherical equivalent (SE), flat (K1), steep (K2), and steepest (Kmax) keratometry before and after detection of lamellar channel deposits. Results In vivo confocal microscopy revealed diffuse linear and focal granular hyperreflective structures. Histologically, there was mild proliferation of fibroblasts and fibrosis. TEM demonstrated focal accumulations of degenerated keratocytes with cytoplasmic lipid inclusions. There were no significant changes for UDVA (Δ = 0.0 ± 0.2 logMAR; p = 0.67), CDVA (Δ = 0.0 ± 0.1 logMAR; p = 0.32), SE (Δ 0.1 ± 0.9 D; p = 0.22), K1 (Δ = 0.3 ± 1.0 D; p = 0.28), K2 (Δ = 0.1 ± 0.9 D; p = 0.51), and Kmax (Δ = 0.3 ± 1.5 D; p = 0.17). Conclusions Two types of structural stromal changes were identified: (1) diffuse peri-segmental fibrosis and (2) lamellar channel deposits. These structural changes showed no evidence of a relevant refractive or topographic effect.


Introduction
Intracorneal ring segments (ICRS) are crescent-shaped arcs of polymethylmethacrylate (PMMA) developed to be surgically inserted into the deep corneal stroma for the purpose of remodeling the corneal curvature. Originally developed to reversibly treat mild myopia [1][2][3][4], ICRS have been first introduced in 2000 by Colin et al. as an option to treat keratoconus (KC) patients [5]. Since 2004 and the approval of the Food and Drug Administration (FDA) for Intacs (Addition Technology Inc., Des Plaines, IL, USA) [6], the surgical procedure has been proven effective in improving the refractive and topographic outcomes of patients with KC [7][8][9][10][11], and the method has been extended to a larger spectrum of corneal ectasia, such as pellucid marginal degeneration (PMD) [12][13][14] and corneal ectasias after laser vision correction (LVC), e.g., after laser in situ keratomileusis (LASIK) [15,16].
Currently, there are several ICRS available on the market, designed to be implanted into the 5-6 mm, 6-7 mm, or 7-8 mm optical zone of the cornea. The Intacs SK (for "Severe Keratoconus") (Addition Technology Inc., Des Plaines, IL, USA) was developed for the 6-7 mm optical zone with the aim to correct larger myopic and astigmatic refractive errors in more advanced forms of corneal ectasia [17]. More recently, new designs with asymmetric progressive thickness emerged to treat KC with specific asymmetric phenotypes in corneal topography such as "duck" and "snowman" phenotypes [18,19].
ICRS implantation is associated with potential complications such as infectious keratitis, asymmetric or superficial segment displacement, segment extrusion, posterior corneal perforations, corneal stromal edema around the incision, extension of the incision towards the central visual axis, halos, and glares [10,20]. These complications have become rarer since the tunnel creation for the insertion of ICRS is no longer performed mechanically but commonly carried out using femtosecond laser (Fs laser) [20,21]. Besides these complications, intrastromal structural changes such as peri-segmental fibrosis [22] and intrastromal lamellar channel deposits have been observed with slit lamp, in vivo confocal microscopy and optical coherence tomography [4,[23][24][25][26][27]. Still, the exact nature and mechanism of these structural changes remain currently unclear.
The purposes of this study were: (1) To characterize the morphological and (ultra)structural tissue changes of stromal alterations and deposits after Fs-ICRS implantation, including slit lamp examination, in vivo confocal microscopy, histopathology, and transmission electron microscopy (TEM).
(2) To analyze the potential effect of stromal lamellar channel deposits on the refractive and topographic outcomes of the procedure.
The (measured using a Scheimpflug camera Pentacam HR (OCULUS GmbH, Wetzlar, Germany)) were compared for all 15 patients between before (absence of) and after (presence of) visible intrastromal lamellar channel deposits on slit lamp examination using a Wilcoxon signed-rank test. Statistical analysis was performed with SPSS Version 20.0.0 for Windows (SPSS Inc., Chicago, IL, USA). Values are expressed as mean ± SD (minimum-maximum). P-values < 0.05 were considered significant.
In vivo confocal imaging of the cornea was performed for 5 (33.3%) out of 15 patients, at the day of slit lamp diagnosis, using a confocal scanning laser ophthalmoscope Heidelberg Retina Tomograph HRT 3 with a Rostock Cornea Module for confocal cornea microscopy (Heidelberg Engineering GmbH, Heidelberg, Germany).
Two of these patients underwent a penetrating excimer laser-assisted keratoplasty [30,31] (both 8.0/8.1 mm diameter, double cross stitch suture [32,33]) 3.1 and 7.0 years after Fs-ICRS implantation due to high residual irregular astigmatism with decreasing visual acuity. The procedures were uncomplicated. We recovered the 2 explanted recipient corneal tissues with implanted ICRS segments for analysis.
The first explanted cornea was entirely fixed in neutral buffered 4% formaldehyde and processed for histopathological analysis. After embedding in paraffin, sections were obtained and stained with hematoxylin/eosin (H/E) and masson-trichrome staining [34].
The second explanted cornea was entirely fixed in 3% cacodylate-buffered glutaraldehyde and processed for TEM analysis. After post-fixation in 2% buffered osmium tetroxide for 1 h, the tissue was dehydrated and embedded in epoxy resin (Epon). Semi-thin and ultra-thin sections were cut on a Reichert-Ultracut (Cambridge Instruments, Nussloch, Germany), stained with toluidine blue, contrasted with uranyl acetate/lead citrate, and examined with an electron microscope (EM 906E; Carl Zeiss Microscopy, Oberkochen, Germany) [35].
Comparing the postoperative values in the absence and presence of lamellar channel deposits (Fig. 2)   The in vivo confocal microscopy showed two types of hyperreflective structures (Fig. 3A-D): (1) focal granular highly hyperreflective structures following a peri-segmental lamellar formation, compatible with lamellar channel deposits; (2) diffuse linear mildly hyperreflective structures all around the ICRS, compatible with peri-segmental fibrosis.
The histopathological examination of the first explanted cornea showed a proliferation of fibroblasts with mild fibrosis. As expected, no lipids could be visualized as these were dissolved during the preparation process (Fig. 4A-D).
TEM of the second explanted cornea showed perisegmental fibrotic stromal changes within a narrow zone of 3 to 15 µm. Within this zone, amorphous and vacuolar materials were found to be interspersed between the condensed collagen fibers (Fig. 5A-E). Few degenerative keratocytes containing cytoplasmic lipid inclusions were occasionally observed in this zone (Fig. 5F). In regions of clinical lamellar channel deposits, focal accumulations of degenerative keratocytes containing large amounts of cytoplasmic lipid inclusions could be visualized (Fig. 6A-D).

Discussion
Since 1991 and the first ICRS implanted in humans [36], significant improvements in design [25], tunnel creation [37], and segment insertion techniques have led to better postsurgical outcomes [10] and reduction of complications [21].
Nevertheless, unavoidable tissue alterations remain such as peri-segmental lamellar channel deposits or fibrosis. These alterations were already described at the premises of the development of PMMA corneal implants in animal models, where crystalline deposits were observed [38,39]. The hypothesis for those lamellar channel deposits was an abnormal cholesterol production induced by post-surgical stress on keratocytes. These deposits were also observed during the first human clinical studies [40]. Ruckhofer et al. analyzed these deposits using in vivo confocal imaging and scanning electron microscopy on explanted ICRS [24] and supported the hypothesis of lipid deposits in the free space between the ICRS and the stromal tissue. This hypothesis is reinforced as these deposits tend to disappear after removal of the segment [4]. Nevertheless, Ruckhofer et al. did not defend the hypothesis of lipid synthesis by stressed keratocytes but assumed a phagocytosis of lipids accumulated in the peri-segmental space by the keratocytes [25]. Twa et al. also analyzed those lamellar channel deposits in an explanted human cornea with history of ICRS implantation in KC, using histology (oil red O staining), TEM, and immunohistochemistry. They found evidence of saturated and unsaturated lipid droplets of cholesterol ester and triglycerides in the keratocytes adjacent to the ICRS, but no manifest extracellular accumulation [26].
The present study also reveals highly hyperreflective granular inclusions in the free space between the ICRS and the stroma in the in vivo confocal microscopy, compatible with lipid deposits, supporting the observations of  [23][24][25]. TEM analysis performed on the explanted cornea also showed intracellular lipid droplets, as well as extracellular vacuolar inclusions, which could represent-as demonstrated by Twa et al. [26,39]-areas of previous cholesterol inclusions, extracted during the preparation of the tissue for TEM. Histological analysis did not reveal any lipid inclusion, probably due to the dissolution of these during the preparation process for light microscopy [41].
In addition to the lamellar channel deposits, we also pointed out the diffuse presence of condensed and linear fibrotic tissue around the ICRS in in vivo confocal microscopy, histopathology, and TEM. This fibrosis was already described after ICRS implantation but also after other types of refractive surgery such as myopic photorefractive keratectomy (PRK) [42] or laser in situ keratomileusis (LASIK) [43], or after corneal injury [44,45]. It is supposedly a wound-healing reaction with remodeling of corneal stroma and collagen synthesis [46]. Samimi et al. characterized these fibers as type IV collagen [41].
Stromal fibrosis and lamellar channel deposits lead to structural changes in the stroma which could theoretically influence the postoperative refractive and topographic results. This issue has already been investigated by some authors who demonstrated a lack of significant impairment of outcomes [23,25]. The present study showed no significant difference between postoperative examinations in absence or in presence of lamellar channel deposits for UDVA, CDVA, SE, K1, K2, and Kmax, remaining consistent with the existing literature.
Nevertheless, these structural changes exist and extreme cases of tissue alterations after ICRS implantation were reported [47]. This must raise concerns about the "reversibility" of ICRS implantation, a term historically used to promote the procedure as an option to treat mild myopia. ICRS implantation has already proven to be reversible regarding the refractive and topographic changes after ICRS explantation [48,49]. Structural changes also appear to be partially reversible. Spontaneous reversibility of lamellar channel deposits has been observed, generally from 24 months postoperatively [50]. In the case of ICRS explantation, the lamellar channel deposits disappear [4] and the peri-segmental fibrotic tissue tends to normalize [41].
The present findings are consistent with the existing literature and suggest a comparable pathophysiology of stromal tissue alterations in corneas treated with (Fs-)ICRS implantation for mild myopia and for KC. The potential stiffening effects of these structural changes are still unclear and subject to controversy. Further studies, including biomechanical analysis, are necessary to evaluate the potential role of peri-segmental fibrosis on KC stabilization, even without additional riboflavin UV-A crosslinking. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.