The healthy cornea is avascular due to corneal “immune privilege”, a process of homeostasis between the low level of angiogenic and high level of antiangiogenic factors.12 Inappropriate blood vessel formation in the normally transparent and avascular cornea is a major cause of vision loss and blindness due to herpetic keratitis.1;4 Currently, there is no epidemiological study that provides an accurate estimate of the incidence and prevalence of CoNV in the general population.3;12
Although mild CoNV may be asymptomatic, more severe forms predispose the cornea to inflammation, lipid exudation and scarring, leading to significant loss in visual function.8 Clinically, CoNV is subdivided into 3 groups based on the pattern of angiogenic invasion: 1) superficial vascularization, where vessels sprout from the superficial marginal arcade and extend beneath the epithelium; this is commonly seen in stromal keratitis; 2) vascular pannus which results from extension of vessels and fibrous tissues from the limbus onto the peripheral cornea and is mainly seen in ocular surface disorders when an insult is sustained for a long period of time; 3) deep stromal vascularization can occur at any level of the stroma, from beneath Bowman layer to Descemet membrane, as seen in herpetic and luetic interstitial keratitis.12;13 Although no one factor can explain all causes of CoNV, 3 pathological mechanisms have been identified: hypoxia, inflammation, and limbal barrier dysfunction. In human HSV type 1 corneal infection recent evidence suggests alteration of the normal balance between angiogenic and anti-angiogenic responses as the likely cause of corneal vascularization.4;12;14 After the first response of production of proinflammatory cytokines and chemokines and an invasion of the cornea by PMN, HSV infection can induce the production of many angiogenic factors such as thrombospondins 1 and 2, vascular endothelial growth factor, matrix metalloproteinases (MMP) 2 and 9, platelet-derived growth factor (PDGF) and beta fibrosing growth factor (bFGF).4;12;15;16
Knowing all these described mechanisms, in daily practice, and despite the importance of having an objective way of assessing abnormal corneal vessels, there has not yet a good noninvasive imaging technique to be widely used. Clinical applications are vast and include preoperative localization of CoNV for target intervention, monitoring treatment response using vascularization area or the diagnosis and prognostication of corneoscleral inflammmation.9 So an essential requirement for disease monitoring and evaluating the efficacy of any potential treatment is the ability to quantify CoNV before and after intervention.8;17
SLP of CoNV is limited by inconsistent vessel delineation from frequently coexisting corneal opacification, poor standardization, and the inability to perform quantitative measurements.5 Angiography techniques utilize intravenous injections of fluorescein and indocyanine green and require long acquisition times, with risks of serious adverse reactions.8 Since the recent advancement of the technology and its availability for clinical implementation, the interest of OCTA to assess AS vessels has grown rapidly.5 The group of Ang et al was a pioneer in presenting the application of OCTA in various clinical conditions with abnormal corneolimbal neovascularization such as postherpetic keratitis scar.6 Subsequent studies have been published: one have shown the ability to detect small and deep vessels in cases of previous herpetic keratitis9, other presented a small case series of herpetic corneal vascularization supporting the use of OCTA imaging technique for monitoring vascular changes after a variety of treatments8, and other showed good agreement and comparable results between OCTA and indocyanine green angiography for measurement of corneal vascularization.18 In this preliminary study, a bigger group of eyes with the same condition was selected to obtain measurements of the corneoscleral vessels using OCTA technology. 17 eyes with residual herpetic stromal scar with/without CoNV were included. It was possible to get a clear visualization of abnormal vessels invading the corneal stroma on the OCTA scans that were not as clearly seen on SLP, also taken in the same day. Farther that OCTA has proven to be an effective tool in quantification of vessels in patients with herpetic leucoma, with a mean total vessel area of 50.907±3.435. Mean VD was higher in the nasal quadrant (51.156±4.276) but there was no significant differences between the three analyzed areas (p=0.940), implying that herpetic inflammation may not reach a preferred corneal area. Superior corneoscleral margin was not analyzed in this study because of time consumption for image acquisition and patient collaboration for this localization. Additionally, OCTA can also provide simultaneous assessment of depth of the lesion, as well as its associated abnormal blood vessels and abnormal vascular loops. In many herpetic cases deep stromal vascularization occurs as it was mentioned before in this discussion. The en face scans can be in the future a useful instrument for corneal planes study, as it is now for retinal diseases. This article presents an example of this modality using the definitions created for retinal analysis (Figure 5). As it is known that HSV CoNV may be prominent, even reaching the central cornea, OCTA may become useful in monitoring patients with acute herpetic crisis and at risk of developing those CoNV.
Suppressing CoNV in severe cases is actually therefore a challenge as it is still no possible to pharmacologically remove pathological blood vessels from the cornea.15 So as the diagnostic techniques evolves, the mechanism and treatment of CoNV has been studied in more detail in recent years. From glucocorticosteroids that have traditionally been the mainstay of managing CoNV, with however, incomplete suppression, new drugs have been developed based on the mechanism neovascularization mentioned before.12;19 Anti- vascular endothelial growth factor (VEGF) trials were uncontrolled studies with small sample size, and the reported reduction in CoNV appeared to be transient and incomplete, hampered by the fact that herpetic keratitis may seem more resistant.12;15 Most often CoNV can be markedly reduced if the anti-VEGF strategy is begun early after infection, but the therapies are much less successful when commenced after CoNV is well established.15 Fine needle diathermy occlusion of CoNV has been reported as an effective and relatively easy procedure to perform, however, diathermy should only be applied to the afferent vessels.19 Finally corneal transplantation is the last option, despite being a leading risk factor for corneal graft rejection.20 Clinical studies have shown that there is a higher than-average risk of graft failure in patients who undergo corneal transplantation for HSV keratitis. The causes of failure are multifactorial and incompletely understood: the higher risk of graft failure results from a higher incidence of immunologic rejection caused by HSV-specific factors, or it may be merely an effect of the inflammation, manifest as vascularization and leukocyte infiltration.21 Clinical lesions attributed to HSV after transplantation frequently occur at the graft margin but may not show the typical dendritic appearance and other signs of HSV disease.20 More recently new studies are showing evidence that local gene therapy may be a promising universal treatment of CoNV although there are still current technical concerns.19;22 Compared to drug- or antibody-based treatments, that only provide short-term benefits and require repeated applications, a gene-based approach offers targeted treatments providing long-term therapeutic correction. The cornea has properties that make it an attractive target for gene-based manipulations: relative immune privilege and easy of access. A variety of vectors have been used for gene-based therapies for corneal angiogenesis;12 For exemple Lai et al have shown that an adenovirus vector carrying VEGFR genes caused regression of CoNV, but to date most studies of gene therapy for CoNV are still in the preclinical experimental stages.22
OCTA technology applied to AS presents some limitations that must be noted. Image distortions may occur due to patient movement, it doesn´t carry an eye-tracking system with registration which is necessary for comparisons in follow-up scans and it is unable to detect vessels without red cell flow. Some scans may contain artifacts that may appear as abnormal vessels, the ability to segment the images to separate the conjunctival vessels from scleral vessels is not optimal and image resolution is also not sufficient to distinguish normal from abnormal vessels. Thus, improvements in the segmentation software are needed to enhance the reliability of the data. Nonetheless, this research expand the importance of OCTA in the management of patients with herpetic eye leucoma. At present these patients are largely monitored with serial photographs, the only viable tool for rapid and non-invasive clinical use; developing standardized OCTA image protocols will help in both qualitative and quantitative patients follow-up and may become a non-invasive alternative to objectively monitor treatment response in eyes with CoNV.7;8; 23