Secondary Logo

Journal Logo

Systematic Review on Therapeutic Strategies to Minimize Corneal Stromal Scarring After Injury

Kwok, Sum S.; Shih, Kendrick C. FCOphth (HK); Bu, Yashan B.Sc.; Lo, Amy C.-Y. Ph.D.; Chan, Tommy C.-Y. F.R.C.S. (Ed.); Lai, Jimmy S.-M. M.D., FRCOphth; Jhanji, Vishal M.D., FRCOphth; Tong, Louis M.D., Ph.D.

doi: 10.1097/ICL.0000000000000584
Review Article

Objectives: To evaluate recent studies on available and experimental therapies in preventing or minimizing corneal stromal scarring after injury.

Methods: We performed an Entrez PubMed literature search using keywords “cornea,” “scarring,” “haze,” “opacity,” “ulcer,” “treatments,” “therapies,” “treatment complications,” and “pathophysiology” resulting in 390 articles of which 12 were analyzed after filtering, based on English language and publication within 8 years, and curation for relevance by the authors.

Results: The 12 articles selected included four randomized control trials (RCTs) (two were double-blinded placebo-controlled RCTs, one was a prospective partially masked RCT, and one was an open-label RCT), two retrospective observational studies, and six laboratory-based studies including two studies having in vivo and in vitro experiments, one was in vivo study, one was ex vivo study, and the last two were in vitro studies. The current mainstay for preventing or minimizing corneal scarring involves the use of topical corticosteroids and local application of mitomycin C. However, supportive evidence for their use in clinical practice from well-designed RCTs is lacking. Laboratory studies on topical rosiglitazone therapy, vitamin C prophylaxis, gene therapy, and stem cell therapy have shown promising results but have yet to be translated to clinical research.

Conclusion: There is a need for more robust randomized controlled trials to support treatments using topical corticosteroids and mitomycin C. Furthermore, their clinical efficacy and safety profile should be compared with new treatments that have shown promising results in the laboratory setting. Ultimately, the goal should be to personalize cornea scarring treatment according to the most effective treatment for the specific underlying pathology.

Department of Ophthalmology (S.S.K., K.C.S., Y.B., A.C.-Y.L., and J.S.-M.L.), Li Ka Shing Faculty of Medicine, University of Hong Kong, Hong Kong, China; Department of Ophthalmology (T.C.-Y.C.), Hong Kong Sanatorium and Hospital, Hong Kong, China; Department of Ophthalmology (V.J.), University of Pittsburgh Medical Centre, Pittsburgh, PA; Cornea and External Eye Disease Service (L.T.), Singapore National Eye Centre, Singapore; and Ocular Surface Research Group (L.T.), Singapore Eye Research Institute, Singapore.

Address correspondence to Kendrick C. Shih, FCOphth (HK), Department of Ophthalmology, Li Ka Shing Faculty of Medicine, University of Hong Kong, 301B Cyberport 4, 100 Cyberport Road, Pokfulam Road, Island South, Hong Kong, China; e-mail:

The authors report no conflicts of interest.

All authors attest that they meet the current ICMJE criteria for authorship. S. S. Kwok, K. C. Shih, V. Jhanji, and L. Tong were involved in study design, data collection, data analysis, manuscript writing, and editing. Y. Bu, A. C.-Y. Lo, T. C.-Y. Chan, and J. S.-M. Lai were involved in data collection, data analysis, manuscript writing, and editing.

The authors alone are responsible for the content and writing of the paper. The authors agree to make all materials, data, and associated protocols promptly available to readers without undue qualifications in material transfer agreements.

Accepted December 30, 2018

Back to Top | Article Outline


According to the World Health Organization (WHO), corneal opacity is the fourth common cause of blindness globally behind cataract, glaucoma, and age-related macular degeneration. Corneal opacities account for 5.1% of blindness globally, with trachoma, ocular trauma, and corneal ulceration being the leading underlying causes.1 An estimated 4.9 million people globally have been blinded by trachoma, and 1.5 to 2 million new cases of ocular trauma and corneal ulceration-related blindness occur every year.1 The Corneal Opacity Rural Epidemiological (CORE) study, a population-based study conducted in rural India, studied 12,899 participants and showed 3.7% had corneal disease with the top 3 causes being pterygium (34.5%), trauma (22.3%), and infectious keratitis (14.5%).2 In a cross-sectional study on 10,384 participants in Heilongjiang China, 0.3% of the population had been found to have corneal blindness, with the leading causes being keratitis in childhood (40%), ocular trauma (33.3%), and keratitis of adulthood (20%).3 It was also noted that corneal disease was the third commonest cause of bilateral blindness. In Hong Kong, it was discovered that in a sample of 1,270 adults above 40 years old, about 7% of cases classified with visual disability were a result of corneal disease.

Back to Top | Article Outline

Burden of Corneal Opacity

Current management of cornea diseases with drugs aims to prevent corneal opacity, but when this fails, the last resort is often a corneal transplant, which carries risks including rejection, infection, and graft failure. Because of advancements in the last two decades, penetrating keratoplasty has been largely supplanted by lamellar keratoplasty as the most common corneal transplantation technique. Posterior lamellar keratoplasty can be divided into Descemet's stripping endothelial keratoplasty and Descemet's membrane endothelial keratoplasty, where the former contains posterior corneal stroma within the graft and the latter does not.4 Lamellar keratoplasty allows for faster recovery, and lower postoperative astigmatism and rejection rates while minimizing endothelial cell loss.5 Aside from issues of graft rejection, a major limitation in cornea transplantation is the shortage of donor corneas. As of 2013, the organ donation rate was 6.1 corneas per million in Hong Kong, one of the lowest organ donation rates among developed countries.6 Apart from the surgical risks, the vision-related quality of life in those with corneal pathology would be significantly impacted and the extent would depend on the severity of disease, whether it is unilateral or bilateral pathology, and the nature of the original insult.

Back to Top | Article Outline

Mechanism of Corneal Scarring/Opacification

Corneal scarring often occurs toward the end of an infective, thermal, or chemical injury process. The associated inflammatory response in these conditions damages the corneal stromal layer and initiates the cascade of corneal haze formation.7 When stromal keratocytes are damaged, inflammatory cytokines and fibrogenic growth factors such as TGFβ1, TGFβ2, connective tissue growth factor (CTGF), coupled with downstream suppressor of mothers against decapentaplegic protein (SMAD) signaling, and SMAD-independent signaling activate the stromal keratocytes and trigger a wound healing response.8,9 The profibrotic factors induce the differentiation of activated stromal keratocytes to myofibroblasts, which in turn actively secrete extracellular matrix, thus resulting in stromal scar formation. Unlike activated fibroblasts, myofibroblasts have elevated expression of TGFβ1 receptors and cadherins. Myofibroblasts also have increased antiapoptotic properties, allowing them to persist during corneal stromal remodeling.8,10 Many new treatments aim to neutralize or inhibit the downstream signaling of TGFβ1: the main profibrotic cytokine in myofibroblast differentiation.9

This scarring mechanism is an attempt to prevent infections by increasing the barrier property of the cornea to microbes. However, the same scarring mechanism results in an uneven, light-scattering cornea, and therefore a detrimental effect on visual quality and quality of life.

Current clinical practice relies heavily on the application of topical medications, including 1% prednisolone acetate ophthalmic suspension and 0.02% mitomycin C (MMC) to minimize or reduce corneal scarring after injury. However, these medications have associated visually significant long-term complications, including cataract formation, glaucoma, and corneoscleral melting. In this article, we will systematically review and critically appraise the clinical and experimental treatments used to prevent corneal scarring, including rosiglitazone, gene therapy, and stem cell therapy.

Back to Top | Article Outline


We performed an Entrez PubMed search on August 7, 2018, using the following keywords: “cornea,” “scarring,” “haze,” “opacity,” “ulcer,” “treatments,” “therapies,” “treatment complications,” and “pathophysiology” to search for relevant clinical trials in humans as well as “in vivo” studies in animals and “in vitro” studies in the laboratory. This search strategy yielded 390 publications. We further limited the results to publications in English and those published within the last 8 years. Further curation by KSS and KCS were performed for relevance, and this resulted in 12 studies that were finally selected for review (Fig. 1).

FIG. 1

FIG. 1

Back to Top | Article Outline


We reviewed 12 articles published in the last 8 years. These included six clinical studies, with four being randomized control trials (RCT) (two were double-blinded placebo-controlled RCTs, one was a prospective partially masked RCT, and one was an open-label RCT) and two were retrospective observational studies. In six laboratory-based studies, two studies had in vivo and in vitro experiments, one was an in vivo study, one was an ex vivo study, and the last two were in vitro studies.

We have highlighted the various strengths and limitations of each study (Table 1). Overall, well-designed RCT on the subject are scanty. Three of the four RCTs (and both of the double-blinded ones) evaluated topical corticosteroids, and the remaining RCT (which was open-label) evaluated topical MMC. Retrospective observational studies in the use of vitamin C show promising results, but this should now be evaluated in larger, prospective clinical trials. In vitro studies have demonstrated antiscarring effects of vitamin C at a cellular and molecular level. However, in clinical scenarios, one needs to consider issues such as drug penetration, toxicity to surrounding cells, and the optimal dosing formulation, regime, and route of administration. Topical rosiglitazone and limbal stem cell transplantation have shown benefits in the in vitro and in vivo studies and should be further evaluated in human trials. We have outlined the various characteristics and mechanisms of each type of antiscarring therapy (Table 2).





Back to Top | Article Outline

Topical Corticosteroids

Topical application of prednisone acetate ophthalmic suspension is a well-established treatment for minimization of corneal scarring. It possesses anti-inflammatory and antifibrotic effects. Corticosteroids have demonstrated their ability to reduce TGFβ expression in a human fetal lung fibroblast model, which is the underlying principle for antifibrotic therapy in the cornea.11

The Steroids for Corneal Ulcers Trial (SCUT) is a large-scale randomized placebo-controlled double-masked clinical trial, which aimed to evaluate the effectiveness of adjunctive topical prednisolone with topical moxifloxacin in the treatment of bacterial keratitis-related corneal ulcers. The study showed no significant difference in the best-corrected visual acuity (BCVA) between the two treatment groups, over 3 months after treatment. It also showed no significant difference in the incidence of corneal perforations and epithelial defect healing rates. However, corticosteroid treatment was safe, in that there was no clinically significant elevation of intraocular pressure (IOP) after treatment.12 In the secondary aim of the SCUT trial, which evaluated the outcome at 12 months, topical corticosteroids have been shown to improve visual outcome in corneal ulcers unrelated to Nocardia organisms, with a mean improvement in one line of BCVA, compared with placebo. For Nocardia-related ulcers, there was no improvement of long-term visual outcomes when using adjunctive corticosteroid therapy.13 In addition, corticosteroid use may be linked to larger scars in Nocardia-related ulcers with reported cases of recurrent infection and prolonged epithelial wound healing time in such eyes. It should be noted in cases of corneal scarring related to bacterial infections, the use of corticosteroid is controversial and the timing of steroid treatment needs to be delayed until there is adequate control of the infection. Using corticosteroids too early can lead to uncontrolled infection and worsening of the disease. Although Nocardia is not a very common pathogen in bacterial keratitis, the SCUT results suggest that knowledge of the underlying microorganism by culture is an important prerequisite for topical steroid use in corneal ulcers. In the SCUT trial, steroid treatment was delayed for safety reasons while awaiting the microbiological results, and this may have altered its efficacy.

The effect of topical corticosteroids has also been investigated in surface ablation laser refractive therapy. A prospective partially masked RCT compared 3-month corneal healing outcomes in 132 participants.14 The IOP and corneal haze grading were assessed by masked observers and ophthalmologists, respectively, with other parameters such as manifest refraction, BCVA, and uncorrected visual acuity (UCVA) being assessed by unmasked observers. A group of participants used loteprednol etabonate 0.5% gel use and then tapering down, whereas another group was treated with prednisolone acetate 1% suspension and fluoromethalone (FML) 0.1% solution before tapering. Loteprednol etabonate is quickly broken down into inactive metabolites thus reducing its side effects such as raised IOP.15 The study ultimately found no significant postoperative corneal haze between groups and no significant difference in IOP elevation between groups. Nearly all eyes in the study achieved an UCVA of 20/20 or better by 3 months.14 This study shows that both forms of corticosteroids were efficacious in terms of visual outcomes but did not include a noncorticosteroid comparison group.

It is worth noting that these are very different studies. Corneas suffering from bacterial keratitis are likely to have a less-predictable and more-intense stromal inflammatory response than corneas undergoing photorefractive keratectomy (PRK). Prophylactic antibiotic and steroid treatment can also begin promptly in the context of laser refractive surgery compared with delayed treatment in the setting of bacterial keratitis. In clinical practice, one concern with all topical eye drops is patient compliance and ability to properly instill eye drops, which may be less-important issues in properly conducted and tightly monitored clinical trials.

Back to Top | Article Outline

Topical Mitomycin C

MMC is a powerful mitotic inhibitor capable of forming covalent linkage with DNA to inhibit DNA, RNA, and protein synthesis. It has been used in chemotherapeutic treatment of numerous tumors but is now widely applied as eyedrops because of its ability to induce apoptosis of myofibroblasts and prevention of keratocyte differentiation.16 The use of MMC is more common in prevention of recurrence of pterygium after excision surgery, and for treatment of conjunctival-corneal intraepithelial neoplasia and squamous cell carcinoma, compared with its use in corneal scarring.17

Equine corneal fibroblasts cultured with TGFβ1 showed significantly increased αSMA staining compared with those not exposed to TGFβ1. When 0.02% MMC was applied early after adding TGFβ1, that is, day 1, there was significant inhibition of TGFβ1-induced keratocyte differentiation to myofibroblast (69% ± 5.8) compared with late treatment, that is, day 11 (28% ± 9.8). In fact, administration on day 1 showed a 14-fold reduction of αSMA expression.18 It has also been noted that when different concentrations of MMC was evaluated, 0.02% MMC for 2 min was found to be most effective for inhibition of keratocyte differentiation. The same in vitro study demonstrated that the cellular viability after instilling MMC 0.02% for 2 min is unchanged, but higher concentrations such as MMC 0.05% showed mild to moderate cellular toxicity when treated for 2 min.18 This demonstrates the significant therapeutic potential of MMC as well as its toxicity in terms of inducing cell apoptosis and tissue necrosis. The earlier MMC is started after cornea insult, the more effective its antiscarring effects are. Moreover, a study shows that despite its ability to deplete the stromal keratocytes, the effect is reversed by stopping the treatment after 6 to 12 months after refractive surgery.19 This is reassuring because it implies its side effects could be prevented by limiting the treatment duration and close monitoring.

Regarding re-epithelization, human corneal in vitro studies have shown an exposure time-dependent delay in re-epithelialization, but treatment with MMC still resulted in eventual complete closure of defect.20 The same study showed MMC having minimal effects on corneal endothelial cells that may possibly be attributed to their lack of amitotic activities. This suggests the potential suitability of using MMC in patients with concurrent endothelial dysfunction.20 However, there are other side effects of MMC that are not limited to the cornea. These include lacrimal punctal stenosis, scleromalacia and perforation, cataract, intractable glaucoma, anterior uveitis, pyogenic granulomas, and calcific scleral plaques.21

A randomized control trial involving 261 patients, investigating intraoperative topical MMC onto the ablation area during laser-assisted subepithelial keratectomy (LASEK) surgery, found that those given 0.04% MMC had 1% incidence of corneal haze while those given 0.02% had a 13% incidence at 1 year post surgery, with no significant difference in the endothelial cell count between groups.22 However, it is worth noting that the MMC was applied for 30 to 110 sec depending on the patient's refractive error and was then washed out with saline. The short exposure times may not be long enough to induce harmful effects on the cells.22 Moreover, patients were given topical dexamethasone, a corticosteroid, and tobramycin postoperatively for 1 week, which may have affected the outcome.22 However, the results from this study do show the superiority of a higher dose of MMC as a one-time treatment during LASEK surgery. More studies are needed to show similar superiority in other pathologies such as infectious keratitis or chemical injuries, and potential cytotoxic effects on other cells such as limbal stem cells and corneal epithelial cells should be investigated.

Back to Top | Article Outline

Topical Rosiglitazone

Rosiglitazone, a thiazolidinedione, is used in the treatment of type 2 diabetes mellitus by acting as an insulin sensitizer. It has since become less popular due to serious associated adverse effects, including congestive heart failure and increased risk of bladder cancer. Like other thiazolidinediones, rosiglitazone possess anti-inflammatory effects making it useful to treat conditions such as rheumatoid arthritis and inflammatory bowel disease, and in addition, its antifibrotic effects have been demonstrated in lung and corneal tissues in vitro.23,24

Rosiglitazone is a ligand of peroxisome proliferator–activated receptor gamma (PPARγ), and studies have shown that PPARγ overexpression reduces myofibroblast differentiation and blocks the nuclear translocation of SMAD2 thus exerting antifibrotic effect through signaling downstream of TGFβ.25 An in vivo study on feline corneas evaluated the effect of rosiglitazone in dimethylsulfoxide (DMSO)/Celluvisc (RefreshCelluvisc, Allergan10%), DMSO/Celluvisc alone, DMSO alone, or Celluvisc alone, given daily for 2 weeks after PRK and assessing the progress at 2, 4, 8, and 12 weeks after PRK.26 At 2 weeks after PRK, immunostaining for αSMA showed the least detection of αSMA for rosiglitazone-treated corneas followed by corneas treated with the DMSO/Celluvisc without rosiglitazone. Optical coherence tomography and in vivo confocal images demonstrated reduced backscatter of the anterior stroma and reduced reflectivity of cornea in rosiglitazone-treated cases compared with other groups, and these changes returned to basal levels 12 weeks after PRK. In the case of the DMSO-/Celluvisc-treated eyes, the level of haze was even more than preoperative levels.26

Rosiglitazone's lipophilic properties were advantageous because of solubility in DMSO, thus enhancing drug penetration through membranes and tissues.27 Lipophilic drugs in general penetrate the corneal stroma much more effectively. Moreover, DMSO itself is an anti-inflammatory agent that may allow the DMSO–rosiglitazone vehicle solution to work synergistically.

In short, rosiglitazone demonstrated its ability to allow the cornea to return to its original state of clarity and refractive properties after 3 months of treatment without inhibiting re-epithelization and stromal rethickening.26 This would make rosiglitazone a significantly safer treatment than prednisolone and MMC. The damage of MMC to the corneal epithelial cells and stromal cells is a significant concern. There is also little concern about systemic uptake as demonstrated by previous studies. However, there is a lack of understanding in the degree of systemic uptake for topical rosiglitazone usage in the cornea, and if there is significant systemic uptake, what would the effects be and what the optimum concentration or dosage of topical rosiglitazone would allow us to balance the desired effect while minimizing systemic effects. Regarding the cost effectiveness of this treatment, we would need to compare its efficacy with our existing treatments and assess whether it is at least equally effective to what we currently use. Moreover, it is a treatment that can be easily introduced into our current clinical practice.

Back to Top | Article Outline

Systemic or Intravenous Vitamin C

Ascorbic acid or vitamin C has been known to facilitate corneal healing by promoting the synthesis of parallel arrays of extracellular matrix fibrils in cultured human keratocytes.28 Moreover, its antioxidant properties reduce corneal neovascularization and increase collagen synthesis.29 Both of these factors are very important in preventing corneal haze because the composition and three-dimensional organization of the corneal stroma play a major role in the clarity and normal functioning of the cornea.7 A study on 82 patients with infectious keratitis was performed, and participants were divided into control, oral vitamin C (3 g/day), and intravenous (IV) vitamin C (20 g/day) groups (systemic antibiotics were used in all groups) while in hospital. The study measured the size of epithelial defect and corneal haze on admission, at discharge, and after 3 months. The study showed that vitamin C was able to reduce the size of the epithelial defect and the size of the corneal haze, with IV vitamin C being the most effective in both cases. It was also found that the reduction of opacity was more significant when a hypopyon was present at admission, and the patient was under 60 years old.30

However, a nonrandomized clinical trial involving 48 people compared the effect of additional systemic vitamin C (500 mg, twice daily 1 week before to 2 weeks post op) with the standard perioperative topical MMC alone when undergoing LASEK and found no significant reduction in corneal haze. It was reported that 33.3% and 37.5% of the patients in the treatment and control groups, respectively, developed corneal haze.31

The evidence demonstrates that vitamin C may be effective for fibrosis after infective keratitis, but its use in iatrogenic causes such as laser therapy is still questionable and needs further RCTs to verify. Different etiologies of corneal injury should be investigated for the newer approaches to treating scarring. In addition, there is a need to understand the mechanisms of certain therapies, which have been found to be efficacious.

Nevertheless, vitamin C is a promising supplemental treatment for infective keratitis with its very good safety profile and easy introduction into common clinical practice with vitamin C supplements and IV formulation readily available at a low cost. There is little to no concern about side effects because these would be given over a short course. There is more work required to optimize the therapeutic dosage and the route and duration of administration.30

Back to Top | Article Outline

Triple siRNA Therapy

An ex vivo study rabbit corneal model assessing use of triple combination siRNA targeting scarring genes such as TGFβ1, TGFβ receptor (TGFBR2), and CTGF to inhibit their expression was performed. The effect of siRNA was compared to that of 1% prednisolone acetate ophthalmic suspension, in corneal scarring after laser ablation.8 This study also evaluated the effectiveness of nanoparticle carriers in targeted ocular drug delivery. The carriers have been known to be nonimmunogenic, relatively nontoxic, unlikely to be absorbed into the bloodstream, and stable in an enzymatic environment.8 The study demonstrated that prednisolone reduced SMA expression by 85% through reduction of mRNA expression of scarring genes. The triple combination siRNA demonstrated a similar reduction in SMA expression, by 77%.8 Eventually corneas without any form of treatment developed a haze percentage of 22% compared with 9.6% for siRNA-treated corneas and 7% for steroid-treated corneas, which was statistically significant (P<0.05).8 The siRNA can be delivered using a topical formulation, but data on long-term safety are not available. When comparing nanoparticle carried siRNA with siRNA alone, fluorescein-labeled siRNA showed no fluorescence in the stroma. The fluorescein-labeled nanoparticle siRNA showed fluorescence in several superficial epithelial cells but not in the stroma or endothelium. These findings suggest that even with the nanoparticles, the penetration of the siRNA is significantly improved, and may account for the slight inferiority of siRNA compared with steroids for reduction of corneal haze. More work is necessary to find a more effective way to deliver the siRNA to maximize its potential.

Back to Top | Article Outline

AAV5-Smad7 Gene Therapy

The cornea is an ideal structure for gene therapy given its superficial and easily accessible location. This is perhaps why gene therapy is a promising potential treatment for corneal scarring.32,33 Gene therapy targets the TGFβ signaling that plays a key role in fibrogenesis, including the Smad proteins that mediate wound repair and scarring. Smad7 inhibits TGFβ signaling through the TGFβ1 receptor by competing with R-Smad-Smad four complex for DNA binding, consequently blocking stromal keratocyte differentiation into myofibroblasts.34–36 Smad7 is delivered to the cornea using recombinant adeno-associated virus serotype 5, a nonpathogenic virus with no known toxicity, and can be potentially transduced into a wide range of tissues.37 An in vivo rabbit cornea model was used to study cornea scarring in corneas introduced with topical AAV5-Smad7 gene therapy immediately after PRK.38 It was found that Smad7 overexpression inhibited myofibroblast formation and thus reduced αSMA staining in TGFβ1-treated HCF.38 Moreover, rabbit corneas treated with AAV5-Smad7 demonstrated significantly less fibrosis as evidenced by less cloudiness and haze under slit-lamp examination and stereo microscopic imaging 4 weeks after PRK.38 This treatment also maintained a normal, stable IOP, suggesting that this treatment is safer than the use of long-term steroids. TdT-mediated dUTP nick end labeling (TUNEL) immunostaining demonstrated normal cell turnover of corneal epithelial cells, suggesting that the AAV5-Smad7 is noncytotoxic. However, there may be issues concerning the cost of the treatment and limited expertise among clinicians. Further studies on the safety profile and human trials are necessary.

Back to Top | Article Outline

Limbal Stem Cell Biopsy and Transplant

Autologous limbal stem cell transplants (LSCT) are also being considered to prevent corneal scarring. Allogeneic LSCTs can be performed, but these have a high risk of rejection. A limitation of LSCT is the requirement for an eye with adequate limbal stem cells, and thus, it is usually applicable in cases of unilateral pathology in one eye, such as microbial infection or chemical injuries. Limbal stem cell transplants poses a slight risk to the healthy eye, which often may be the patient's only functioning eye. Should there be decompensation in the healthy eye, there would be catastrophic effects on quality of life.

Limbal stem cells were traditionally believed to generate only epithelial cells of the cornea. However, studies have demonstrated that ex vivo expansion of a limbal stem cell biopsy can also generate cornea stromal stem cells, the progenitor of keratocytes.39 A study obtained human limbal stem cells and transplanted them in mice cornea 1 week after wounding. This was compared to mice without an LSCT, and it was found that corneas with LSCT had less-fibrotic elements present in the corneal extracellular matrix, such as fibronectin, tenascin C, biglycan, hyaluronan, type III collagen, and SPARC (secreted protein acidic and rich in cysteine).39 Assessment of the corneal transparency by studying the organization of collagen after LSCT showed that it is similar to native uninjured corneas. Thus, the efficacy of LSCT is demonstrated at the clinical and cellular levels. However, high costs, limited expertise, and lack of infrastructure are serious concerns. With further advancements in this field, there is potential to benefit numerous patients all over the world.

Back to Top | Article Outline

Induced Pluripotent Stem Cells

Using a series of reprogramming, induced pluripotent stem cells (iPS) have been successfully differentiated to regenerate retinal pigment epithelial cells, cardiomyocytes, neurons, and even corneal epithelium.40 These iPS have been obtained from precursors such as human dermal fibroblasts (HDF), keratinocytes, and neural precursor cells. However, some iPS retain their original epigenetic characteristics that will limit their differentiation capability, including attempts to regenerate corneal epithelium.41 To obtain limbal stem cells, it is necessary to perform a biopsy at the corneoscleral rim, which is rather invasive, and may damage a healthy eye. On the other hand, HDF are much more readily available and safer to harvest.

A study comparing the differentiation of iPS derived from human limbal stem cells and HDF and their ability to differentiate into corneal epithelial stem cells showed that both types of cells could generate corneal epithelial stem cells, using long-term culture with the stromal cell–derived inducing activity method.40 However, it is necessary to assess iPS differentiation to corneal stromal cells as well. As of now, limbal stem cells can be induced to cornea stromal keratocytes in vitro by fibroblast growth factor β2,42 but there has yet to be evidence of any success through iPS. Although iPS has been promising for certain pathologies such as spinal injury, we believe that its clinical potential is still limited in the field of corneal scarring, given the technical demands and limited expertise, which are likely to result in an expensive treatment.

Back to Top | Article Outline


Currently, clinicians rely on topical corticosteroids and MMC, which have very significant side effects while lacking the high-level evidence needed to demonstrate clinical benefits. The basis of the use of MMC and corticosteroid in treating corneal ulcers relies heavily on reduction of inflammation and decrease of profibrotic cytokine synthesis. We would highlight this gap in our understanding, with caution to clinicians when prescribing topical corticosteroids or MMC for treating corneal ulcers. The potential benefits and risks to using these agents should be carefully weighed. In addition, when using MMC and topical corticosteroids, it is essential to monitor the IOP and corneal status closely, while educating the patient on the correct use.

Because vitamin C supplements are nontoxic and an economical supplement, we believe that clinicians can consider following the dosage mentioned in the literature and offer as prophylactic or short-term add on treatment (e.g., 1 week) especially for infectious keratitis (because we do not have to wait for culture results). Oral/IV vitamin C treatment should be compared with current treatment in RCTs, and topical administration should be compared with oral/IV administration. Use of vitamin C in other pathologies should be evaluated further. Other proposed antiscarring therapies still lack the safety profile evaluation and evidence in large-scale clinical trials to warrant clinical use.

Topical rosiglitazone has demonstrated its antifibrotic effects, but its effectiveness in doing so compared with topical steroids and MMC has yet to be investigated. A possibility is treating with both rosiglitazone and either steroid/MMC, which would lower the necessary dose of steroids/MMC. Among the options, topical rosiglitazone has the greatest potential to be introduced in the near future because there is a better understanding of the drug safety profile when used systemically in diabetes mellitus.

There are still no large RCTs for the use of MMC, unlike topical corticosteroids, to evaluate effectiveness in treating corneal scarring. This is very important as the SCUT trial demonstrated questionable efficacy of steroids, although some effect was shown in improving the outcome of non-Nocardia ulcers after 12 months. It is important to consider whether the benefits outweigh the risks. The time for commencement of steroid treatment in infectious keratitis is critical, and cultures are necessary to help the decision.

We described a variety of newer treatments that have been proposed. More promising treatments such as topical rosiglitazone and AAV5-gene therapy have yet to be performed on HCFs or in clinical trials (of large sample size). The former should be much more accessible in the near future and with less hurdles. Reproducibility of the results is important and so is the verification of the safety profile. It is important to bear in the mind the limitations of many of the above-stated treatments especially in terms of affordability, especially for gene therapy and stem cell therapy, although they are very promising. To incorporate gene therapy or stem cell therapy into corneal scar treatment, we would need to thoroughly evaluate their efficacy in scarring from various pathologies. Use of various disease models (postlaser ablation, infectious keratitis, and chemical injury) is necessary as certain pathologies may respond better to certain treatments. This also allows for deeper investigation into the mechanism of corneal scarring and the intricacies of cytokine signaling and other downstream events, which can explain disease-specific events. Clinicians can more easily choose the most cost-effective treatment for patients, taking into consideration healthcare resources and the patient's ability to afford certain treatments thus tailoring or personalizing the treatment.

Supplements such as Wolfberry or Lycium barbarum, a well-known antifibrotic and antioxidizing agent used for thousands of years in Traditional Chinese Medicine, should also be investigated. Alternative medicine is increasingly popular, and Wolfberry has been shown to have protective effects in retinal ganglion cells43 and corneal epithelial cells.44 Such supplements have potential as an alternative remedy and may reduce our dependence on rather toxic agents with questionable efficacy. Similarly, blueberry extract also demonstrates antioxidizing and anti-inflammatory effects on corneal epithelial cells.45 Apart from the pharmaceutical treatment, the future of cornea scar treatment may extend to the field of alternative medicine.

Back to Top | Article Outline


Corneal scarring is a serious complication of corneal injury. Depending on the degree of corneal haze, size, and location of the lesion, it variably impacts on quality of life of patients. Currently, we rely on corticosteroids and MMC, and for many decades, there has been no other widely available treatment. However, potentially safer topical agents such as rosiglitazone or vitamin C prophylaxis could potentially be exploited in clinical practice. More work needs to be performed with gene therapy and stem cell therapy to determine safety and role in clinical practice. It is very important to understand the underlying etiology in the development of corneal haze/scarring even when a decision has been made to use corticosteroids or MMC. There is still an unmet need to understand whether certain therapies are more effective for specific etiologies. Further work in this may facilitate personalize medicine in the management of cornea scarring.

Back to Top | Article Outline


1. Whitcher JP, Srinivasan M, Upadhyay MP. Corneal blindness: A global perspective. Bull World Health Organ 2001;79:214–221.
2. Gupta N, Vashist P, Tandon R, et al. Prevalence of corneal diseases in the rural Indian population: The Corneal Opacity Rural Epidemiological (CORE) study. Br J Ophthalmol 2015;99:147–152.
3. Wang H, Zhang Y, Li Z, et al. Prevalence and causes of corneal blindness. Clin Exp Ophthalmol 2014;42:249–253.
4. Gorovoy MS. DMEK complications. Cornea 2014;33:101–104.
5. Krumeich JH, Knulle A, Krumeich BM. Deep anterior lamellar (DALK) vs. penetrating keratoplasty (PKP): A clinical and statistical analysis. Klin Monbl Augenheilkd 2008;225:637–648.
6. Cheung CY, Pong ML, Au Yeung SF, et al. Factors affecting the deceased organ donation rate in the Chinese community: An audit of hospital medical records in Hong Kong. Hong Kong Med J 2016;22:570–575.
7. Qazi Y, Wong G, Monson B, et al. Corneal transparency: Genesis, maintenance and dysfunction. Brain Res Bull 2010;81:198–210.
8. Sriram S, Gibson DJ, Robinson P, et al. Assessment of anti-scarring therapies in ex vivo organ cultured rabbit corneas. Exp Eye Res 2014;125:173–182.
9. Tandon A, Tovey JC, Sharma A, et al. Role of transforming growth factor Beta in corneal function, biology and pathology. Curr Mol Med 2010;10:565–578.
10. Jester JV, Petroll WM, Cavanagh HD. Corneal stromal wound healing in refractive surgery: The role of myofibroblasts. Prog Retin Eye Res 1999;18:311–356.
11. Wen FQ, Kohyama T, Skold CM, et al. Glucocorticoids modulate TGF-beta production. Inflammation 2002;26:279–290.
12. Srinivasan M, Mascarenhas J, Rajaraman R, et al. Corticosteroids for bacterial keratitis: The Steroids for Corneal Ulcers Trial (SCUT). Arch Ophthalmol 2012;130:143–150.
13. Srinivasan M, Mascarenhas J, Rajaraman R, et al. The steroids for corneal ulcers trial (SCUT): Secondary 12-month clinical outcomes of a randomized controlled trial. Am J Ophthalmol 2014;157:327–333.e323.
14. Mifflin MD, Betts BS, Frederick PA, et al. Efficacy and safety of a 3-month loteprednol etabonate 0.5% gel taper for routine prophylaxis after photorefractive keratectomy compared to a 3-month prednisolone acetate 1% and fluorometholone 0.1% taper. Clin Ophthalmol 2017;11:1113–1118.
15. Holland EJ, Bartlett JD, Paterno MR, et al. Effects of loteprednol/tobramycin versus dexamethasone/tobramycin on intraocular pressure in healthy volunteers. Cornea 2008;27:50–55.
16. Galm U, Hager MH, Van Lanen SG, et al. Antitumor antibiotics: Bleomycin, enediynes, and mitomycin. Chem Rev 2005;105:739–758.
17. Prabhasawat P, Tarinvorakup P, Tesavibul N, et al. Topical 0.002% mitomycin C for the treatment of conjunctival-corneal intraepithelial neoplasia and squamous cell carcinoma. Cornea 2005;24:443–448.
18. Buss DG, Sharma A, Giuliano EA, et al. Efficacy and safety of mitomycin C as an agent to treat corneal scarring in horses using an in vitro model. Vet Ophthalmol 2010;13:211–218.
19. Santhiago MR, Netto MV, Wilson SE, et al. Biological effects and use in refractive surgery. Cornea 2012;31:311–321.
20. Rajan MS, O'Brart DP, Patmore A, et al. Cellular effects of mitomycin-C on human corneas after photorefractive keratectomy. J cataract refractive Surg 2006;32:1741–1747.
21. Khong JJ, Muecke J. Complications of mitomycin C therapy in 100 eyes with ocular surface neoplasia. Br J Ophthalmol 2006;90:819–822.
22. Shi J, Yuan Y, Zhao S, et al. Effects of different mitomycin C concentrations on laser-assisted subepithelial keratectomy. Exp Ther Med 2014;7:1591–1594.
23. Burgess HA, Daugherty LE, Thatcher TH, et al. PPARgamma agonists inhibit TGF-beta induced pulmonary myofibroblast differentiation and collagen production: Implications for therapy of lung fibrosis. Am J Physiol Lung Cell Mol Physiol 2005;288:L1146–L1153.
24. Pan HW, Xu JT, Chen JS. Pioglitazone inhibits TGFbeta induced keratocyte transformation to myofibroblast and extracellular matrix production. Mol Biol Rep 2011;38:4501–4508.
25. Saika S, Yamanaka O, Okada Y, et al. Effect of overexpression of PPARgamma on the healing process of corneal alkali burn in mice. Am J Physiol Cel Physiol 2007;293:C75–C86.
26. Huxlin KR, Hindman HB, Jeon KI, et al. Topical rosiglitazone is an effective anti-scarring agent in the cornea. PLoS One 2013;8:e70785.
27. Kolb KH, Jaenicke G, Kramer M, et al. Absorption, distribution and elimination of labeled dimethyl sulfoxide in man and animals. Ann New York Acad Sci 1967;141:85–95.
28. Grobe GM, Reichl S. Characterization of vitamin C-induced cell sheets formed from primary and immortalized human corneal stromal cells for tissue engineering applications. Cells Tissues Organs 2013;197:283–297.
29. Lee MY, Chung SK. Treatment of corneal neovascularization by topical application of ascorbic acid in the rabbit model. Cornea 2012;31:1165–1169.
30. Cho YW, Yoo WS, Kim SJ, et al. Efficacy of systemic vitamin C supplementation in reducing corneal opacity resulting from infectious keratitis. Medicine 2014;93:e125.
31. Yulish M, Beiran I, Miller B, et al. Ascorbate prophylaxis with mitomycin-C for corneal haze after laser-assisted sub-epithelial keratectomy. Isr Med Assoc J 2012;14:382–385.
32. Mohan RR, Tovey JC, Sharma A, et al. Gene therapy in the cornea: 2005—Present. Prog Retin Eye Res 2012;31:43–64.
33. Mohan RR, Sinha S, Tandon A, et al. Efficacious and safe tissue-selective controlled gene therapy approaches for the cornea. PLoS One 2011;6:e18771.
34. Hayashi H, Abdollah S, Qiu Y, et al. The MAD-related protein Smad7 associates with the TGFbeta receptor and functions as an antagonist of TGFbeta signaling. Cell 1997;89:1165–1173.
35. Nakao A, Afrakhte M, Moren A, et al. Identification of Smad7, a TGFbeta-inducible antagonist of TGF-beta signalling. Nature 1997;389:631–635.
36. Zolotukhin S, Potter M, Zolotukhin I, et al. Production and purification of serotype 1, 2, and 5 recombinant adeno-associated viral vectors. Methods 2002;28:158–167.
37. Samulski RJ, Muzyczka N. AAV-mediated gene therapy for research and therapeutic purposes. Annu Rev Virol 2014;1:427–451.
38. Gupta S, Rodier JT, Sharma A, et al. Targeted AAV5-Smad7 gene therapy inhibits corneal scarring in vivo. PLoS One 2017;12:e0172928.
39. Basu S, Hertsenberg AJ, Funderburgh ML, et al. Human limbal biopsy-derived stromal stem cells prevent corneal scarring. Sci translational Med 2014;6:266ra172.
40. Hayashi R, Ishikawa Y, Ito M, et al. Generation of corneal epithelial cells from induced pluripotent stem cells derived from human dermal fibroblast and corneal limbal epithelium. PLoS One 2012;7:e45435.
41. Kim K, Doi A, Wen B, et al. Epigenetic memory in induced pluripotent stem cells. Nature 2010;467:285–290.
42. Funderburgh JL, Funderburgh ML, Du Y. Stem cells in the limbal stroma. Ocul Surf 2016;14:113–120.
43. Mi XS, Feng Q, Lo AC, et al. Protection of retinal ganglion cells and retinal vasculature by Lycium barbarum polysaccharides in a mouse model of acute ocular hypertension. PLoS One 2012;7:e45469.
44. Du S, Han B, Li K, et al. Lycium barbarum polysaccharides protect rat corneal epithelial cells against ultraviolet B-induced apoptosis by attenuating the mitochondrial pathway and inhibiting JNK phosphorylation. Biomed Res Int 2017;2017:5806832.
45. Li J, Ruzhi D, Hua X, et al. Blueberry component pterostilbene protects corneal epithelial cells from inflammation via anti-oxidative pathway. Sci Rep 2016;6:19408.

Cornea stromal scarring; Prevention; Minimization; Systematic review; Topical treatment; Systemic treatment

© 2019 Contact Lens Association of Ophthalmologists, Inc.