Department of Ophthalmology and Visual Sciences, University of Illinois at Chicago, Illinois Eye & Ear Infirmary, Chicago, IL 60612, USAUnited States
Also available as PDF (figures and tables appear in the PDF).
This mini-review provides an overview of recent trends in our understanding of the general structure of the cornea, the importance of corneal nerves, their function and distribution in the different layers of the cornea based on reports obtained by imaging techniques such as in vivo confocal microscopy (IVCM) and immunohistochemistry (IHC) analysis. Recent data on corneal nerve status in conditions such as Dry Eye Disease (DED) and Diabetes is also discussed. There is additional emphasis on corneal nerve damage due to injury especially during surgical interventions and underlying disease states as well as translational research on corneal nerve regeneration. Information on recent clinical studies on the effect of laser corneal surgery and its impact on corneal nerves is also presented.
Keywords: Corneal nerves; Immunohistochemistry (IHC); Dry Eye Disease (DED); Corneal nerves; Immunohistochemistry (IHC); Dry Eye Disease (DED)
The eye is one of the most fascinating organs of the human body and has numerous parts each of which plays a critical role in providing vision (sight). As the well-known saying goes “the eyes are the window to the soul”, for the layman, the cornea is indeed that part of this window through which light enters the eye and along with the lens is focused onto the retina. The retina in turn absorbs and converts the light into electrochemical impulses which are then transferred to the brain via the optic nerve. The cornea is the outermost transparent, clear and avascular connective tissue layer that forms the front part of the eye. It is dome-shaped and covers the pupil, iris and the anterior chamber and acts as a structural barrier for primary infections to the eye. Structurally and anatomically the main layers of the human cornea include, the epithelial layer or epithelium, the Bowman’s membrane, the stromal layer or stroma, the recently identified Pre-Descemet’s layer known as Dua’s layer, the Descemet’s membrane and the endothelial 1-3 layer or endothelium.1-3 The cornea stands out as being one of the most densely innervated tissues in the body4 and our current understanding of the cornea and corneal diseases in general is based on the seminal works of numerous scientists and clinicians in the field of ophthalmology. The purpose of this short review is to highlight some important studies and findings in the field
thalmic branch) and enter the corneal stroma after which they form a subbasal plexus below the 6,7 epithelium, and extend into thinner nerves containing nociceptors at the corneal surface. The
Sarkar and Ruth Zelkha based on Linda JM et al, 2003 and Tomas-Juan J et al, 2014). innervations of the cornea and bulbar conjunctiva is contributed mainly by the sensory fibers of the ophthalmic branch of the trigeminal nerve and by the less numerous sympathetic and parasympathetic nerve fibers.8 In addition to their important sensory function, corneal nerves also play an important role in providing protection and are involved in trophic functions. Corneal nerves have also been known to be involved in the regulation of corneal epithelial integrity, wound healing and cell proliferation.9 A renewed interest in corneal neurobiology arose recently because of the pivotal role these nerves play in maintaining a healthy ocular surface, which is especially important today due to the damage corneal nerves incur from refractive surgery, corneal transplants and herpetic infections (Figure 1).
One major condition that causes epithelial abnormalities is dry eye syndrome. It is has been estimated that nearly 10% of the U.S. population suffers from dry eye syndrome, which in turn significantly affects quality of life in these patients. Dry Eye Disease (DED) is considered to be a disease mainly of the tears and ocular surface causing symptoms of discomfort, disturbance in vision and instability of the tear film with possible damage to the ocular surface.1 DED can also occur due to disturbance of the lacrimal glands, the ocular surface and eyelids, and the sensory and motor nerves that connect them. A stimulation of corneal nerves followed by nerve alterations has been postulated as one of the core pathophysiological mechanisms in DED.
Aging has been implicated in the increased incidence of dry eye10 and although some risk factors have been identified, not
4,46 much is known about the causes. The role of aging and how it affects nerve architecture of the cornea is a very important focus of recent research. Early studies on corneal nerves in animals and humans were purely based on light or electron microscopic evaluations. Schimmelpfennig et al11 studied fresh central corneal buttons from keratoplasties and enucleations by staining with gold chloride and provided one of the first comprehensive 9 11 descriptions of corneal epithelial nerves.11 This has been followed by numerous fixation and staining methodologies and this gamut of analytical procedures have greatly contributed to enhancing our knowledge of the morphology, ultrastructural organization, density, and corneal nerve alterations after injury or 4,6 12,13 death. Studies by two prominent groups mention that ‘aging’ is accompanied by structural and functional changes in the cornea involving corneal steepening and increased thickness of the Descemet’s membrane.14 Aging has also been shown to impact corneal wound healing as well as affect visual acuity and 15-18 refractive outcomes after laser refractive surgery.
For the longest time, visualization of the human cornea and the different layers within has remaineda pipe-dream. In recent years, the introduction of in vivo confocal microscopy (IVCM) has provided a new method for high resolution corneal examination 8,9 19,20 in living patients. It has broadened the scope for non-surgical 21-26 intervention and cellular examination of live corneas in vivo. However, despite breakthroughs in imaging techniques, the distribution of corneal nerves is not completely deciphered as yet and the reasons for this lack of understanding is because of the difficulty in obtaining detailed innervations in the different corneal layers since conventional histology requires fresh corneas. Secondly, transmission electron microscopy (TEM) images are
restricted to very tiny areas of the corneal surface (0.1 mm2). Finally, IVCM images of the human cornea are captured from the corneal apex. Also these microscopes have a limitation in that they cannot image branching nerves and their terminals of diameters <0.5 mm.
In a recent paper published by Haydee Bazan’s group at LSU, New Orleans, LA, USA,27 the authors have introduced a novel tissue preparation technique to study and image the exact location of nerve fibers. Studies in the past used cross-sections which failed to show detailed corneal innervations. Studies by Müller et al have implicated nerve degeneration to be the main reason for lack of innervation data as seen via electron microscopy techniques demonstrating significant nerve degeneration 4,6 within ~12-13.5 h of death. The modified technique used by Bazan’s team for this study allowed for observation of new nerve structure features and, for the first time, provided a complete view of the human corneal nerve architecture. Our study reveals that aging decreases the number of central epithelial nerve terminals, and increases the presence of irregular anomalies beneath the basal layer.
Linna et al28 have shown that corneal areas with short, unconnected nerve fiber bundles are associated with lower sensitivities than corneal areas with long nerve fiber bundles with or without interconnections. Laser-assisted in situ keratomileusis (LASIK)-induced alterations of subbasal nerve morphology can be visualized via in vivo confocal microscopy. This allows the observer to make a direct comparison of corneal sensory innervation and sensitivity to touch, pain, heat, cold, etc.
These days, confocal microscopy is employed on a wide scale for studying and evaluating corneal changes in diabetic pa- 23,29-32 tients. There are numerous reports of changes in the subbasal nerve plexus of diabetic patients due to epithelial loss and 30-32 corneal hypoesthesia. Increased light scattering due to abnormalities of the basement membrane has also been reported in 31,32 a study by Morishige et al. The importance of stromal nerve changes in diabetic patient corneas is unclear at the moment due to the lack of extensive studies using confocal microscopy. Since patients with diabetes have reduced corneal sensitivity, they are more susceptible to corneal trauma. Studies by Müller et al in diabetic rats demonstrated altered morphology of corneal nerves using light and electron microscopy. In addition to the observation of polymorphism in epithelium and endothelium, Busted 36 37 et al36 and Pierro et al37 reported increased corneal thickness in diabetic patients.
Early studies by Frueh et al39 examined the corneas of 10 Type 1 Diabetes, 10 Type 2 Diabetes and 10 Non-diabetic patients by confocal microscopy and found epithelium and endothelium polymorphisms and abnormal stromal nerves in only two patients with Type 1 Diabetes. No specific observations on the subbasal nerves or corneal sensitivity were reported.40 A cor-
relation between corneal light-scattering index and stages of dia- 31,32 betic retinopathy was published by Morishige et al although nerve morphology was not described. Confocal microscopy studies on skin biopsy specimens have revealed that the number of epidermal nerve fibers per unit surface area in patients with diabetic polyneuropathy is reduced.41 Confocal microscopy appears to allow early detection of beginning neuropathy, because decreases in nerve fiber bundle counts precede impairment of corneal sensitivity.42
Apparently, the cornea becomes thicker in a relatively early stage of diabetes but does not further change with the degree of neuropathy. A reduction in neurotrophic stimuli in severe neuropathy may induce a thin epithelium that may lead to recurrent erosions.
In Photorefractive keratectomy (PRK), photoablation causes severing of the subbasal nerve plexus and anterior stromal 43,44 nerves. Tandem scanning confocal microscopy studies by Erie have shown that subbasal nerve fiber density was 98% less than pre-operatively43 and the ablation zone center showed complete absence of branched nerve fibers, 3 months post-surgery. Both Moilanen and Erie have demonstrated that subbasal nerve density was reduced by 87%, 75% and 60%, (at 3, 6 and 12 months respectively) after PRK, and returned to preoperative 43,45 levels at 2 and 3 years postoperatively. In another study using confocal microscopy, Erie’s team proved faster recovery of subbasal nerve density in the central cornea in PRK as compared to LASIK.44 Hanneken’s group have recently published an excellent review on corneal regeneration after PRK wherein they elucidate how corneal wounding develops following PRK. They also reviewed the influence of intra-operative application of mitomycin C, bandage contact lenses, anti-inflammatory and other drugs in preventing corneal haze post-PRK. Laser in situ keratomileusis (LASIK) is a procedure that utilizes either a bladeless femtosecond laser (FS-LASIK or F-LASIK) or a traditional mechanical microkeratome (MS-LASIK) to create a corneal flap, followed by stromal ablation using an excimer laser. Femtosecond laser technology was first developed in the early 1990s by Dr. Kurtz at the University of Michigan, Ann Arbor, 30 48,49 MI, USA and was extensively used in the surgical field of ophthalmology for its increased safety, precision and predictability over conventional microkeratomes and reduced dry eye symtoms. Femtosecond lasers emit light pulses of short dura- 33-35 tion (approximately 10-15 s) at 1053 nm wavelength that cause photodisruption of the tissue with minimum collateral dam- 48-51 age. This enables bladeless incisions to be performed within the tissue at various patterns and depth with high precision. A new corneal refractive procedure that does not require stromal ablation using an excimer laser called Refractive lenticule extraction (ReLEx) has been discussed by Ang et al.52 In ReLEx, a femtosecond laser is used to create an intrastromal refractive lenticule to correct the refractive error. There are 2 versions of this. In the original ReLEx procedure, femtosecond lenticule
Table 1. Mean Corneal Sensation (in millimeters) at Baseline and after the Procedures.
57 FS-LASIK ±7.15 Wei and Wang 25.44 Central FS-LASIK group (n=54) (18 to 49)
ReLEx flex ±5.72 ReLEx 24.45 group (n=40) (18 to 37) smile
ReLEx smile ±6.52 27.44 Superior FS-LASIK group (n=61) (18 to 43)
55 F-LASIK ±4.4 Demirok et al 26.2 Central F-LASIK group (n=28) (21 to 34)
SMILE group ±4.4 26.2 (n=28) (21 to 34)
All values are Mean+SD standard deviation * refers to changes of corneal sensitivity values post-op in the FS-LASIK group were significantly different from pre-op values with p<0.05, ** refers to changes of corneal sensitivity values post-op in the ReLEx flex group were significantly ***refers to changes of corneal sensitivity values post-op in the ReLEx flex group were significantly different from values in the FS-LASIK group with p< 0.05 W refers to week; M refers to month; pre-op refers to pre-surgery; post-op refers to post surgery
extraction (FLEx), mimics LASIK with the creation of an anterior hinged flap. The lenticule is peeled away after the flap is lifted. Small incision lenticule extraction (SMILE) is a refined version of ReLEx and does not need flap-creation. It involves lenticule dissection and extraction from a small curved bow-like incision (2.5-3 mm) positioned superiorly. In a prospective, randomized clinical trial (contralateral-eye study), 28 patients with myopia or myopic astigmatism in both eyes were enrolled. One eye of each patient was treated by SMILE, and the fellow eye was treated by F-LASIK. One of the mean outcome measures for corneal sensation was Cochet-Bonnet esthesiometry54 and patients were evaluated pre-operatively as well as 1 week, 1 month, 3 months, and 6 months after surgery. This study by Demirok et al55 evaluated the effects of SMILE and F-LASIK on corneal sensation and dry eye parameters revealed that although the dry eye parameters were similar in both surgical groups, there was a significant decrease in corneal sensation measured using a Cochet-Bonnet corneal esthesiometer after both types of surgery with more pronounced effects after F-LASIK sur-
5.81±0.43 2.21±1.28* 2.62±1.72* 3.79±1.44* ReLEx flex 5.88±0.22 2.95±1.41**,*** 3.00±1.24** 4.52±0.96**,*** 5.66±0.45 4.75±1.21**,*** 5.11±1.05**,*** 5.73±0.51***
5.25±0.69 3.61±1.35* 4.09±1.35* 4.63±1.05* ReLEx flex 5.21±0.85 4.43±1.16**,*** 4.39±1.22** 4.98±1.03 ReLEx 5.33±0.56 4.70±0.90**,*** 5.19±0.61*** 5.55±0.57*** smile
Inferior FS-LASIK 5.56±0.56 2.28±1.40* 2.81±1.80* 4.19±1.32* ReLEx flex 5.39±0.68 2.46±1.31** 2.93±1.37** 4.95±0.99*** ReLEx 5.66±0.47 4.92±0.79**,*** 5.37±0.66**,*** 5.63±0.57*** smile
56.2±5.0 30.3±15.3 31.2±14 37.5±14.8 53.7±5 SMILE 56.8±4.7 45.6±11.5 45.3±10.5 49.3±9.9 55.9±4.9
Superior F-LASIK 54.3±4.4 34.3±12.2 35±12 41.2±10.8 53.4±4.7 SMILE 55.3±4.6 44±9.1 44±10.2 48.7±9.5 55.3±4.9
Inferior F-LASIK 55.0±4.8 29.0±15 30.6±14 36.2±15 53.1±6 SMILE 55.6±4.4 46.8±11.9 46.2±10.8 49.7±10 55.6±5.1
different from pre-op values with p< 0.05
gery as compared to SMILE surgery. This difference could be attributed to the fact that LASIK disrupts both the dense subbasal nerve plexus and stromal corneal nerves in the creation of the anterior stromal flap and excimer laser ablation of the cornea whereas in SMILE there is less damage to the corneal nerve 52,53 since the refractive change in SMILE is not obtained by excimer laser-induced photoablation but rather by a femtosecond laserinduced refractive cut.56 Another non-randomized clinical trial by Wei and Wang et al57 evaluated corneal sensitivity between FS-LASIK and femtosecond lenticule extraction (ReLEx flex) or small-incision lenticule extraction (ReLEx smile) for myopic eyes. Twenty-seven subjects (54 eyes) underwent FSLASIK, 22 subjects (40 eyes) underwent ReLEx flex, and 32 subjects (61 eyes) underwent ReLEx smile surgery. Corneal sensitivity was evaluated by Cochet-Bonnet esthesiometry preoperatively as well as at 1 week and 1 and 3 months after surgery.58 In both trials, randomized and non-randomized, better DRY Eye outcomes were observed after SMILE as compared to femtosecond LASIK (femto LASIK) and recovery to baseline corneal sensi-
Table 2. Mean Corneal Nerve Morphology (from in-vivo confocal microscopy; IVCM) and Corneal Sensation (in centimeters) at Baseline and after the Procedures.
56 Corneal nerve morphology Vestergaard et al FLEX group (n=34) 35±7 (25 to 45)
SMILE group (n=34) 35±7 (25 to 45)
All values are Mean+SD (standard deviation) *refers to a statistically significant difference between femtosecond lenticule extraction (FLEX) and small-incision lenticule extraction (SMILE) with p<0.05 **refers to a statistically significant difference between femtosecond lenticule extraction (FLEX) and small-incision lenticule extraction (SMILE) with p<0.01 W refers to week; M refers to month; pre-op refers to pre-surgery; post-op refers to post surgery
tivity was faster with SMILE as compared to both femto-LASIK and Femtosecond Lenticule Extraction (FLEx). Other studies by Jodhbir Mehta’s group59 have also evaluated corneal nerve changes after small incision lenticule extraction (SMILE) and laser in situ keratomileusis (LASIK). They found that more subbasal nerves were disrupted and undergoing regeneration after LASIK as compared to the SMILE group which in comparison demonstrated greater subbasal nerve length and density and higher subbasal nerve recovery at different time-points postsurgery (Table 1 and 2).
New advances in imaging technology and disease models (in vitro cell and tissue-based as well as in vivo transgenic animalbased models) for studying corneal nerves and reinnervation after nerve injury or disease have further enhanced our knowledge of the corneal structure and architecture of corneal nerves in normal and diseased states.60 The thy1-YFP transgenic mouse model developed by Joshua Sanes’ group61 which exhibits yellow fluorescent nerves in the cornea has provided an amazing tool to basic and translational scientists to study corneal nerves in vivo and studies using this model have yielded numerous breakthroughs and publications in the field of corneal nerve injury 7,62-64 and regeneration research. These studies have highlighted neurotoxicity in the eye due to preservatives like benzalkonium chloride and augmented the move towards non-neurotoxic preservative-free eye-drops,62 the presence of inflammatory CD11b+GR1+myeloid-derived suppressor cells which play an important role in nerve regeneration,65 the importance of VEGF- B in stimulating peripheral nerve growth, etc.63
FLEX 19.00±5.51 4.78 ± 3.91 (n=31 patients) Density (mm/mm , mean±SD) SMILE 17.62±5.27 8.41±7.01*
Number (/mm , mean±SD) FLEX 80.3±25.8 32.7±22.4 SMILE 78.3±19.4 53.8±37.5**
Tortuosity (grade, mean±SD) FLEX 1.65±0.54 1.51±0.62
SMILE 1.69±0.49 1.60±0.54 Corneal sensation FLEX 5.87±0.20 5.49±0.45 (n=34 patients) Cochet-Bonnet esthesiometry SMILE 5.88±0.19 5.78±0.34** (cm, mean±SD)
This greater clarity and continued progress in our understanding of the functional and structural alterations of nerves in normal and disease states and their correlation with clinical signs and symptoms is crucial for the further development of targeted drug therapy and treatments for debilitating corneal diseases. The window to the future appears truly bright indeed!
The author would like to thank Ruth Zelkha from the UIC Ophthalmology Core for her help with artwork and schematic representations. Supported by National Eye Institute (NEI) Core Grant EY001792 and Research to Prevent Blindness.
The author does not have any financial or proprietary interest in any material or method mentioned.