Histopathologic Analysis of Explanted KAMRA Corneal Inlays Demonstrating Adherent Fibroconnective Tissue Scar Formation

Corneal inlays have emerged as an attractive alternative surgical treatment for presbyopia, the age-related decrease in natural accommodative amplitudes and uncorrected near visual acuity typically compensated for by reading glasses. Corneal inlays are devices implanted within a surgical pocket or flap in the corneal stroma of a patient’s nondominant eye. There are various types of corneal inlays that each act on a different principle to address presbyopia. The KAMRA inlay (AcuFocus, Inc., Irvine, CA, USA) works by improving depth of focus via a pinhole effect and was the first corneal inlay to gain approval by the United States Food and Drug Administration (FDA) in April 2015. These inlays are made of polyvinylidene fluoride and carbon.

Inlay proponents assert the reversibility of the technique as a major advantage of corneal inlays compared to laser refractive surgery or presbyopia-correcting intraocular lenses. Dissatisfied patients may have their corneal inlays surgically explanted, e.g., due to undesirable glare and halo visual effects or unintended postoperative refractive shifts [1]. Of the >20,000 inlays that have been implanted in Europe, 362 inlays, or 1.81%, have been removed due to consumer dissatisfaction [2]. Relatively little is known about explanted corneal inlays and the resultant effect on the cornea and visual status. Limited studies on patients with explanted inlays suggest that patients return to within one diopter of preoperative refraction without significant loss of corrected acuity [1, 3-7]. Several groups have reported the development of corneal stromal deposits and/or haze after implantation of hydrogel-based inlays [8, 9]. In this study, explanted KAMRA inlays were examined via histopathologic and immuno-staining methods to investigate the cellular composition of these membranes.

This is an observational case series of 11 KAMRA corneal inlay specimens explanted due to the presence of adherent opaque membranes associated with peri-inlay stromal haze. Each inlay was bisected and embedded on edge in HistoGel (Thermo Scientific, Kalamazoo, MI, USA) and subsequently in paraffin. Histologic sections were cut from each paraffin-embedded block for staining with hematoxylin and eosin and for immunohistochemistry. The cellular composition of the tissue adherent to the inlays was examined via light microscopy. Inlays with tissue membranes were further studied for immunohistochemical expression of smooth muscle actin, keratocyte markers (CD34), and macrophage markers (CD68).

Within a 1-year period in 2010–2011, 11 specimens labeled as explanted KAMRA corneal inlays were received by our ophthalmic pathologist (G.J.H.). These cases were sent by the inlay manufacturer from various cities across the United States, including St. Louis, Chicago, San Diego, Indianapolis, and Los Angeles. Of these specimens, four corneal inlays had sufficient cellular material to be identified after histologic processing. Three other specimens were too hypocellular for histologic analysis, one specimen did not contain any cellular material, two specimens did not survive histologic processing, and one specimen jar arrived empty.

The corneal inlay histologic specimens with sufficient cellular material demonstrated adherent fibroconnective tissue, which was found to be composed of myofibroblasts rather than normal keratocytes (Fig. 1). On hematoxylin and eosin stain, fibroconnective tissue was adherent to the inlays in all cases and filled the inlay fenestrations in some areas. Immunostaining for the keratocyte marker CD34 [10] was negative, whereas the myofibroblast marker smooth muscle actin was strongly positive in the cellular component, suggesting a reactive fibroconnective tissue process. No granulomatous foreign-body reaction was specifically identified as the macrophage marker CD68 was negative in one case that was examined (data not shown), while in a second case in which CD68 staining was attempted, the tissue did not survive processing.

In this study, fibroconnective tissue with myofibroblasts was found on explanted corneal inlays. Potential pathophysiologic mechanisms include keratocyte-to-myofibroblast transdifferentiation versus fibrous ingrowth. It is unclear whether this reactive scar formation may impact visual potential in patients after inlay explantation. This finding of fibrous tissue formation is perhaps not surprising in light of historical evidence of myofibroblast-related corneal haze following photorefractive keratectomy [11]. Older models of inlays have similarly led to corneal opacities demonstrating fibrous encapsulation of inlays, epithelial cell implantation and ingrowth on inlays, or progressive stromal deposits that persisted even after inlay explantation [8, 9]. Additionally, abdominal meshes made of a similar polyvinylidene fluoride-based material as the KAMRA corneal inlay, utilized for laparoscopic repair of incisional hernia, sometimes require explantation due to adhesions, with the explants histologically exhibiting myofibroblastic proliferation at the adhesion sites [12].

In the present study, following histologic processing, only four inlay specimens maintained sufficient cellular material for analysis, while four were acellular/hypocellular and two did not survive processing at all, thereby demonstrating the challenges in processing these very small specimens containing relatively little cellular material in proportion to the prosthetic material. While it is our impression that such a small specimen’s survival of histologic processing is rather unpredictable, it remains a hypothetical possibility that the degree of cellularity of the explants may have been influenced by surgical technique or perhaps by the extent/severity of the peri-inlay membrane formation in vivo.

When counseling prospective patients for surgery, even remote risks are typically mentioned for balanced consent. Similarly, patients considering corneal inlay implantation should be informed that any incisional ocular surgery involving implantable devices may not always be completely reversible upon explantation, ranging from subclinical reactive scar formation to keratitis and vision-threatening complications. A number of non-peer-reviewed articles in various ophthalmic magazines and news outlets, such as EyeNet and Ophthalmology Times, suggest that corneal inlay implantation is completely reversible without loss of visual acuity and/or is risk-free. Of 21 magazine reports on this subject between 2013 and 2016 that we reviewed, 38% claimed complete procedure reversibility, and 86% cited experts who disclosed financial interest in inlays. Yet, the professional use information put forth by AcuFocus, Inc. for the KAMRA inlay does mention the potential loss of uncorrected and best-corrected visual acuity after inlay removal, and that this should be disclosed to all patients in the informed consent [13].

In general, the few studies that have examined visual function after inlay explantation report minimal if any change in visual acuity within a short time frame. A small case series of 10 patients with explanted KAMRA inlays showed a 1–2 Snellen line reduction in several mean visual acuities (uncorrected distance, uncorrected near, and corrected near) without change in corrected distance at 6 months post explantation as compared with pre implantation [4]. Another study that assessed a longer average follow-up of 32 months included four patients with explanted KAMRA inlays with no loss of mean corrected distance visual acuity, but it did not report whether mean corrected near visual acuity was compromised [3]. These studies included patients who underwent inlay explantation between 1 and 17 months after implantation [3, 4], so it is unclear what visual outcomes might be expected for eyes that undergo explantation at a later time.

Notably, the clinical trial outcome data on device effectiveness presented to the FDA approval committee excluded 44 patients or 8.7% who had their KAMRA inlays removed [2]. A primary safety endpoint at 12 months reported that <1% of eyes had clinically significant haze on slit-lamp evaluation associated with corrected distance visual acuity loss >2 lines [2], but this statistic did not include those patients who had their inlays explanted. In addition, postoperative satisfaction scores reported to the FDA indicated that 71.8% of patients expressed moderate-to-high satisfaction with uncorrected near vision at 12 months [2], but did not comment on the reasons why the remaining 28.2% of patients were less than satisfied. It is possible that corneal haze may have contributed to visual dissatisfaction in this latter group. A recent comparison of the FDA safety and efficacy data for the KAM-RA and Raindrop inlays (ReVision Optics, Lake Forest, CA, USA) reported a 7–8% explantation rate for both inlays within the 24-month study period, with over a third of the Raindrop inlays explanted due to corneal haze and 44% with persistent loss of at least 1 line of monocular uncorrected distance visual acuity at 6 months post explantation (these percentages were not reported for KAMRA) [14]. Similarly, older hydrogel inlay designs were associated with progressive stromal deposits and/or haze that persisted up to 6 months after explantation [8, 9]. The stromal deposits on the inlay-stromal interface progressed for several years, with a fibrous layer eventually encapsulating the inlay prior to explantation [9]. Longer duration of KAMRA inlay implant appears to correlate with development of stromal haze and changes in corneal topography and aberrometry [3, 4], suggesting that earlier removal may minimize the persistence of corneal haze.

This limited study focused on the histopathology of explanted corneal inlays. It is inherently biased toward problematic inlays resulting in explantation and was not designed to investigate the frequency of the complication described herein, but rather was intended to increase the awareness of potential inlay-related complications. As the inlays in this study were explanted by surgeons with financial relationships and confidentiality agreements with the manufacturer, the clinical data on these explanted specimens could not be included in the present study without risk of industry bias. It is important to raise awareness of the potential complication described in the present study in order to prompt unbiased future studies that correlate histologic findings with clinical data. The frequency of clinically detectable peri-inlay membranous haze leading to inlay explantation, and the question of how often this process is associated with any permanent change in vision or corneal health following explantation, needs to be investigated and reported. It is also worth noting that the presence of microscopic alterations and/or clinical haze does not necessarily translate into visual issues, and that the few small studies to date on explanted inlays suggest that most eyes do return to preoperative corneal states and visual function. Nevertheless, patients with inlays should be counseled and monitored for various complications such as those seen here and in older inlay models, and these events should be reported in the literature. This study highlights the need for further research on the effect of inlays on the cornea before and after explantation in order to more appropriately counsel prospective patients and possibly prevent occurrence of visually significant opacities.

The authors have no ethical conflicts to disclose.

The authors have no conflicts of interest to declare.

This work was supported in part by an unrestricted grant from Research to Prevent Blindness to the Department of Ophthalmology and Visual Sciences at Washington University. The funding organization had no role in the design or conduct of this research.

Study design: G.L. Paley, G.J. Harocopos. Data analysis: G.J. Harocopos. Manuscript preparation: G.L. Paley, G.J. Harocopos. G.J. Harocopos had full access to all the data and takes responsibility for the integrity and accuracy of the data presented.