LITHUANIAN UNIVERSITY OF HEALTH SCIENCES
Vilija Danilevičienė
HERPETIC EYE DISEASE:
MORPHOLOGICAL CHARACTERISTICS
OF THE CORNEA AND THEIR
RELATIONS WITH IMMUNE AND
GENETIC FACTORS
Doctoral Dissertation Medical and Health Sciences,
Medicine (M 001)
The dissertation has been prepared at the Department of Ophthalmology of the Medical Academy of Lithuanian University of Health Sciences over the period from 2015 through 2020.
Scientific Supervisor
Prof. Dr. Dalia Žaliūnienė (Lithuanian University of Health Sciences, Me-dical and Health Sciences, Medicine – M 001).
Dissertation will be defended at the Medical Research Council of the Lithuanian University of Health Sciences:
Chairperson
Prof. Dr. Ingrida Janulevičienė (Lithuanian University of Health Sciences, Medical and Health Sciences, Medicine – M 001).
Members:
Prof. Dr. Brigita Šitkauskienė (Lithuanian University of Health Sciences, Medical and Health Sciences, Medicine – M 001);
Assoc. Prof. Dr. Tomas Balsevičius (Lithuanian University of Health Sciences, Medical and Health Sciences, Medicine – M 001);
Prof. Dr. Janina Tutkuvienė (Vilnius University, Medical and Health Sciences, Medicine – M 001);
Prof. Dr. Andrzej Grzybowski (University of Warmia and Mazury in Olsz-tyn, Medical and Health Sciences, Medicine – M 001).
Dissertation will be defended at the open session of the Medical Research Council of the Lithuanian University of Health Sciences on the 29th of
Ja-nuary, 2021 at 1:00 PM in the auditorium No. A-202 of the Centre for the Advanced Pharmaceutical and Health Technologies of Lithuanian University of Health Sciences.
LIETUVOS SVEIKATOS MOKSLŲ UNIVERSITETAS
Vilija Danilevičienė
HERPINĖ AKIES LIGA:
RAGENOS MORFOLOGINIAI YPATUMAI
IR JŲ RYŠYS SU IMUNINIAIS BEI
GENETINIAIS VEIKSNIAIS
Daktaro disertacija
Medicinos ir sveikatos mokslai, medicina (M 001)
Disertacija rengta 2015–2020 metais Lietuvos sveikatos mokslų universiteto Medicinos akademijos Akių ligų klinikoje.
Mokslinė vadovė
prof. dr. Dalia Žaliūnienė (Lietuvos sveikatos mokslų universitetas, medi-cinos ir sveikatos mokslai, medicina – M 001).
Disertacija ginama Lietuvos sveikatos mokslų universiteto medicinos mokslo krypties taryboje:
Pirmininkė
prof. dr. Ingrida Janulevičienė (Lietuvos sveikatos mokslų universitetas, medicinos ir sveikatos mokslai, medicina – M 001).
Nariai:
prof. dr. Brigita Šitkauskienė (Lietuvos sveikatos mokslų universitetas, medicinos ir sveikatos mokslai, medicina – M 001);
doc. dr. Tomas Balsevičius (Lietuvos sveikatos mokslų universitetas, me-dicinos ir sveikatos mokslai, medicina – M 001);
prof. dr. Janina Tutkuvienė (Vilniaus universitetas, medicinos ir sveikatos mokslai, medicina – M 001);
prof. dr. Andrzej Grzybowski (Varmijos Mozūrų universitetas, medicinos ir sveikatos mokslai, medicina – M 001).
Disertacija ginama viešajame Medicinos mokslo krypties tarybos posėdyje 2021 m. sausio 29 d. 13:00 val. Lietuvos sveikatos mokslų universiteto Nau-jausių farmacijos ir sveikatos technologijų centro A-202 auditorijoje.
CONTENTS
ABBREVIATIONS ...7
INTRODUCTION ...8
1. ANALYSIS OF WORLDWIDE RESEARCH ...12
1.1. Prevalence of HSV keratitis ...12
1.2. The cornea ...12
1.3. The HSV entry into the human body ...13
1.4. The herpetic eye disease ...13
1.5. Characterization of HSV induced damage ...14
1.5.1. Corneal nerves damage ...14
1.5.2. Characterization of corneal endothelium damage ...15
1.6. The corneal reaction to HSV ...16
1.7. The immune response ...16
1.8. The role of dendritic and Langerhans cells ...17
1.9. The tear secretory immunoglobulin A ...18
1.10. The Cold Sore Susceptibility Gene ...18
2. METHODS ...20
2.1.The selection of subjects ...20
2.2. Clinical examination ...22
2.2.1. The examination of corneal nerves ...22
2.2.2. The examination of Langerhans and endothelium cells ...23
2.2.3. The detection of tear sIgA...24
2.2.4. Blood sample collection ...24
2.2.5. The detection of HSV specific antibodies and interleukins ...24
2.2.6. The C21orf91 gene genotyping ...25
2.3. The literature search ...25
2.4. The statistical analysis ...26
3. RESULTS ...27
3.1. The evaluation of the corneal sub-basal nerves ...27
3.2. The evaluation of the density of the Langerhans and endothelium cells ...35
3.3. The evaluation of the tear sIgA titers ...37
3.4. The evaluation of serum HSV IgM/IgG and IL-10, IL-17 titers ...40
3.5. The evaluation of the C21orf91gene genotypes ...42
4. DISCUSSION ...45
4.1. Confocal microscopy findings during herpetic keratitis ...45
4.1.1. Corneal sensitivity and the changes of the corneal sub-basal nerves ...46
4.1.2. Langerhans cells and the endothelium cells density ...48
4.2. The importance of tear secretory IgA ...50
4.3. The role of HSV immunoglobulins and interleukins ...50
CONCLUSIONS ...53
LIMITATIONS OF THE STUDY ...54
PRACTICAL RECOMMENDATIONS FOR DOCTORS OPHTHALMOLOGISTS ...55 SANTRAUKA ...56 REFERENCES ...81 A LIST OF PUBLICATIONS ...89 COPIES OF PUBLICATIONS ...90 CURRICULUM VITAE ...107
ABBREVIATIONS
APC – antigen presenting cells
C21orf91 – cold sore susceptibility gene
CNBD – corneal nerve branch density
CNFD – corneal nerve fibre density
CNFL – corneal nerve fibre length
CTBD – corneal nerve fibre total branch density
DC – dendritic cells
DNA – deoxyribonucleic acid
EC – endothelium cells
ELISA – enzyme-linked immunosorbent assay
HSV – herpes simplex virus
IgG – immunoglobulin G
IgM – immunoglobulin M
IL – interleukin
LC – Langerhans cells
LSCM – laser scanning confocal microscopy
mmHg – millimetre of mercury
NK – natural killers
OD – right eye (lat. oculus dexter)
OR – odds ratio
OS – left eye (lat. oculus sinister)
PCR – polymerase chain reaction
PVA – polyvinyl acetal
ROC – receiver operating characteristics
sIgA – secretory immunoglobulin A
SNP – single nucleotide polymorphism
SPSS – statistical package for the social sciences
Th – T helper cells
INTRODUCTION
Herpetic keratitis and its complications is a major cause of corneal blind-ness and a decreased quality of life. The disease manifests as a unilateral ker-atitis, tends to recur and causes hypoesthesia – a corneal sensory loss – which is associated with a reduced number of sensory nerves plexus at the corneal sub-epithelium layer, and tends to progress with herpes simplex virus (HSV) keratitis recurrences [1–4]. The disease recurs in 10% of patients after the occurrence of the first episode of infection during the first year, in 60% of patients – within 20 years [5]. Recurrent episodes of the infection lead to cor-neal scarring, opacities, irregular astigmatism, neurotrophic keratopathy for-mation and corneal blindness [6]. The nerve loss during HSV keratitis results in a decreased corneal sensitivity [7]. Corneal hypoesthesia presents an in-creased risk for the neurotrophic corneal damage [3]. It is currently unknown which of the corneal nerve parameters are responsible and how fast they tend to regenerate after herpetic keratitis and how they influence the changes of the corneal sensitivity. Even though the virus remains in a latent state after the first exposure to the individual, we do not know whether the corneal sensi-tivity and the corneal nerve density also decrease in healthy individuals with herpes labialis history.
Corneal infiltrate during HSV keratitis consists of neutrophils, macrophag-es, lymphocytmacrophag-es, polymorphonuclear cells, and plasma cells [8]. One of the neutrophil attractants is cytokine intereukin-17 (IL-17) [9]. IL-17 plays a pro-inflammatory role and is associated with a more severe herpetic keratitis [10, 11], while interleukin-10 (IL-10) plays a protective role during HSV ker-atitis [11, 12]. While the current evidence supports a role for cytokines acting locally in the cornea during the HSV keratitis pathogenesis, systemic IL-10 and IL-17 have not yet been characterised. To investigate the effect of HSV-1 infection systemically, we evaluated serum IL-10 and IL-17 and associated them with the clinical manifestation of herpetic keratitis, corneal swelling, ulceration and neovascularisation.
In the centre and the periphery of a healthy cornea, immature dendritic (DC) and Langerhans cells (LC) are detected [13]. The density of the den-dritic cells increases from the centre towards the periphery [14]. It is still not clear if the density of Langerhans cells during HSV keratitis correlates with the corneal sensitivity and the corneal nerve parameters, which can be detected by in vivo laser scanning confocal microscopy (LSCM). LSCM is a non-invasive method of examining the living human cornea in healthy and pathological sites at the cellular level, detecting epithelial defects, cor-neal sub-basal nerves changes, stromal edema and infiltration, endothelium
changes and keratic precipitates [15, 16]. In this study we used t he LSCM to assess the corneal sub-basal nerve changes, the Langerhans cell and endothe-lial cell density.
Herpetic keratitis involves not only systemic immune markers, but also local ones, one of which is tear secretory immunoglobulin A (sIgA). It is synthesized in the lacrimal glands and released into the tear ducts as a tear immunoglobulin against viral infections. Recently it has been suggested us-ing sIgA as a diagnostic and prognostic marker for HSV keratitis [17]. Yet, it is still not known if there are associations between the sIgA concentration in tears and the corneal clinical appearance, also, the corneal nerve damage during the course of presentation of HSV keratitis.
Not only immune, but also genetic factors are important in herpetic ker-atitis. C21orf91 gene, also known as the Cold Sore Susceptibility Gene 1, is located on chromosome 21 and is associated with the manifestation of herpes labialis [18, 19]. We have not found any studies that could reveal if the gene influences herpetic keratitis as the cause of both diseases is the same.
It is known that the corneal sub-basal nerves are damaged during the acute phase of herpetic keratitis and tend to regenerate later on, but there is a lack of information about the Langerhans cells density changes and the endothe-lium cells (EC) density changes in the course of the disease or their cor-relation with the corneal sub-basal nerve parameters, the appearance and a clinical manifestation of herpetic keratitis 6 months later. Moreover, we have not found any studies if the confocal microscopy findings correlate with the immunological factors such as tear sIgA, HSV immunoglobulins (Ig) and in-terleukins in blood serum and if they could influence the pathogenesis of the disease. Cold Sore Susceptibility Gene 1 was associated with herpes labialis, but there is no evidence if it can influence the appearance of herpetic keratitis. Taking into account that both of the diseases are caused by HSV, we decided to investigate whether the C21orf91 gene affects the onset of HSV keratitis. We decided to evaluate the corneal morphological changes by in vivo confo-cal microscopy over the course of the herpetic keratitis and the associations with the immunologic and genetic factors. We compared the results not only with healthy controls, but also with the controls who previously suffered from herpes labialis which might be a potential risk factor for the development of HSV keratitis, as, to our knowledge, this aspect has not been examined in previous scholarly studies.
The aim of the dissertation
To evaluate the corneal morphological characteristics observed over the course of the herpetic keratitis and determine their relations with the immune and genetic factors.
The objectives of the dissertation:
1. To evaluate the morphological changes of the corneal sub-basal nerves, corneal sensitivity and the intraocular pressure in patients with herpetic keratitis during the acute phase of the disease and 6 months later. 2. To evaluate the changes observed in the density of Langerhans cells
and endothelium cells during the acute phase of herpetic keratitis and 6 months later, and to determine their correlation with the corneal sub-ba-sal nerves parameters.
3. To determine the tear secretory IgA titre in patients with herpetic kera-titis, and to describe the association between sIgA titres and the clinical manifestation of HSV keratitis and the corneal confocal microscopy findings.
4. To evaluate the titers of serum HSV specific antibodies and interleukins in patients with herpetic keratitis.
5. To identify the association between C21orf91 gene polymorphism and herpetic keratitis.
Scientific novelty:
This study is one of a few longitudinal studies that has described and cor-related the data of corneal sensitivity and the sub-basal nerve, the changes in the Langerhans cell and endothelium cell density in patients with the herpetic keratitis. For the first time the research findings were compared not only with contralateral, clinically unafected eyes, but also with the eyes of the patients with a previous history of herpes labialis and healthy subjects during the acute phase of the disease and 6 months later.
This study is the first study that was deploying the corneal confocal mi-croscopy to find the association between the sIgA concentration in tears and the corneal morphological changes during the acute phase of the disease and 6 months later.
This is also the first study attempting to find any possible associations be-tween the C21orf91 gene and the herpetic eye disease. We have determined that the C21orf91 gene has an effect on the manifestation, frequency, corne-al morphologiccorne-al changes, cornecorne-al sensitivity, tear secretion and intraocular pressure of herpetic keratitis.
To our knowledge, so far, neither longitudinal studies have examined this type of data, or any other studies have compared the results between HSV keratitis and herpes labialis patients.
Author’s contribution:
The author performed the research literature analysis, prepared and sub-mitted project research proposals to the Lithuanian Univeristy of Health Sciences Science Foundation who granted funding. The funds received were successfully used to implement the study. The author performed clinical ex-amination of the eyes, measured intraocular pressure, corneal sensitivity, per-formed laser scanning confocal microscopy, collected tear samples and pre-pared blood serum for laboratory testing for all subjects. The author collected all the data of the patients, synthesized and interpreted the results, together with co-authors wrote four scientific articles, two of them had already been published, participated in local and international conferences where the au-thor presented the results of the study.
1. ANALYSIS OF WORLDWIDE RESEARCH
1.1. Prevalence of HSV keratitis
Herpes simplex virus (HSV) is widespread in the human population. Anti-HSV antibodies are found in 88% of the human population at the age of 40 years [20]. HSV-1 causes herpes labialis and herpetic keratitis. The inci-dence of herpetic keratitis is approximately 1.5 million worldwide, includ-ing 40.000 new cases of related blindness each year [21–23]. Herpes labialis manifests in about 40% of HSV-1 infected humans. Clinical manifestation recurs in 1/3 of them, and in half of them it recurs at least twice per year [5]. During herpetic keratitis, the cornea is damaged.
1.2. The cornea
The cornea consists of epithelium, Bowman’s layer, stroma, Descemet’s membrane and the corneal endothelium. The corneal epithelium is composed of three layers: the superficial cells, the wing cells and the basal cells. In LSCM images, the superficial cells can be seen as small bright rounded nu-clei, surrounded by a darker cytoplasm with a dark ring and a well-defined cell border [24]. The wing cells are characterized as bright cell borders usual-ly with a dimmer nucleus than that in the superficial cells or without a visible nucleus lying under the superficial cells [25]. The basal cells are smaller than the wing cells, with bright cell borders and a dark cytoplasmic mass, without a visable nucleus [26]. During the inflammation process Langerhans cells can be observed in the corneal basal or sub-basal layers. They appear as bright cellular images with a branching morphology [27]. The Bowman’s membrane is located posteriorily to the basal epithelium cell and contains collagen and nerve fibers which can be seen as bright, long structures that form a nerve fiber network [28]. The stroma is the thickest layer of the cornea which con-sists of collagen fibers and the interstitial substance which forms a grey back-ground in the confocal microscopy images, and keratocytes, which can be seen as large nuclei of different shapes in a grey background [29]. Descemet’s membrane is the basement membrane of the corneal endothelium and has a hazy appearance without a cellular structure in the confocal microscopy im-ages [30]. The corneal endothelium can be seen as bright cells with dark cell borders without recognizable nuclei [29]. The LSCM method is rapid, non-in-vasive, precise, and ensures a good interobserver variability [31].
1.3. The HSV entry into the human body
After entering the human body for the first time, HSV can cause primary herpetic keratitis, however, in most cases there is no clinical manifestation observed in the eye. Through the retrograde axonal path, the virus enters the sensory ganglia where remains lifelong in a latent state. An individual be-comes a HSV carrier and a source of contamination. During the latency time, even in the absence of the clinical manifestation of the virus, HSV can be isolated from tears and saliva. The viral latency and reactivation is controlled by the immune system [32]. No new replicable virus usually develops during the latency period, but immunosuppression of the host may reactivate the latently infected trigeminal ganglion [33]. The reactivation of the infection can be provoked by hormonal changes, fever, stress, exposure to ultraviolet radiation or other factors. After the reactivation of the virus takes place, the secondary acute herpetic keratitis occurs [34].
1.4. The herpetic eye disease
The primary herpetic eye disease usually manifests as an acute blepharo-conjunctivitis and only rarely as a herpetic keratitis. The inflammation is uni-lateral, characterized as a follicular conjunctivitis, therefore it can easily be mistaken with other viral conjunctivitis. During the primary HSV infection, the person is seronegative. After the virus reactivates, the secondary herpetic eye disease occurs [34]. It can manifest as herpetic keratitis, uveitis or retinal necrosis.
Herpetic keratitis can be divided into the superficial or deeper corneal lay-ers. Superficial keratitis can occur as an epithelial, dendritic or geographic keratitis. The virus replicates in the corneal epithelium cells which are vul-nerable to a direct viral cytopathic effect. Epithelial keratitis is characterized by multiple epithelial blisters filled with fluid which rupture form ulcers. In the course of dendritic keratitis, the epithelium and Bowman’s layer are vul-nerable, resulting in the appearance of the dendritic infiltrate. After dendritic infiltrates merge, the geographic herpetic keratitis is formed. It is character-ized by the ulceration of deeper corneal layers. When the virus invades into the deeper corneal layers, stromal keratitis is formed. Corneal stroma appears with opacities and destruction. At the presentation of stromal keratitis, ul-cerative infiltrates or neovascularisation can be found. HSV is able to reach corneal endothelium and induce endothelial keratitis which can be devided into disciform, linear or diffuse endothelial keratitis. In the cases of disciform endothelial keratitis there is a ring infiltrate and swelling in the stroma, pre-cipitates on the endothelium, which can be found as linear layout leading to
the diagnosis of linear endothelial keratitis. In the cases of diffuse endothe-lials keratitis, the entire cornea swelling and diffusely settled precipitates on the endothelium are found.
After the virus embraces the uveal tract or the retina, the anterior and/or posterior uveitis or the acute retinal necrosis begins. Uveitis is usually gran-ulomatous with sectoral iritis and high intraocular pressure. The appearance of a characteristic fundus involves the peripheral retinal periarteritis with multifocal, deep, yellow-whitish infiltrates which merge and retinal necrosis develops. The disease progresses rapidly. If left untreated, it affects the other eye in up to 70% of the cases.
1.5. Characterization of HSV induced damage 1.5.1. Corneal nerves damage
Nerve fibers enter the cornea from the terminal branches of the n.ophthal-micus branch of the trigeminal nerve. The nerves reach the cornea through the sclera, episclera and conjunctiva. Before entering the cornea, the nerve fibers connect to a dense, ring like meshwork which surrounds the peripheral cornea. The thick limbal nerve fibers pass radially into the central part of the anterior corneal stroma, divides into the stromal nerve fibers, run vertically and form the corneal sub-basal nerve plexus [35]. Each branch of a neuron supports from 200 to 3000 individual corneal nerve endings. Due to the dense concentration of sensory nerve endings, the cornea is one of the most sensi-tive parts in the human body [36].
Herpetic keratitis often presents with the decreased corneal sensitivity and is associated with the changes observed in the sensory nerve fibre and the branch density in the sub-basal corneal layer [37–39]. The normal corneal nerve values for clinical use in the population ranging from 56 to 65 year of age, was described being: CNFD – 14.20 in female, 12.29 in male, CNBD – 18.89 in female, 14.85 in male, CNFL – 12.90 in female, 13.12 in male and decrease with age [40]. Corneal sensitivity also decreases with age [41]. The central cornea sensation in 51–60 year old individuals, was found to be 56.4 mm, in those over 60–54.84 mm [42]. Desensitization begins in the corneal periphery and spreads toward the central zone [43].
Herpetic keratitis tends to recur during the lifetime. The recurrence of the disease may result in the corneal scarring, thinning and neovascularisation, more severe changes of the corneal nerves and hypesthesia which poses a major risk in the form of neurotrophic keratopathy due to the loss of the blink reflex, and a reduced influence of the sensory nerves to the normal corneal physiology [44].
Recently the corneal sensitivity and the changes in the sub-basal nerve parameters have been analysed in patients with other systemic and ophthal-mic diseases. Researchers have found that the decreased corneal sensitivi-ty resulting from the reduced densisensitivi-ty or the morphological changes in the corneal sub-basal nerves are observed in patients with diabetes [45], Fuchs endothelial dystrophy and keratoconus [46–49] in patients with reumathoid arthritis [50, 51], atopic keratoconjunctivitis [52], corneal infections, such as Acanthamoeba and fungal infections [53], tear deficiency diseases such as dry eye disease, Sjogren’s syndrome, meibomian gland diseases and conjunc-tivochalasis, where reduced corneal sensitivity is also associated with eyes irritation, tear instability, damaged ocular surface, increased blink rate and decreased epithelial cell density [54–56].
The corneal sensitivity increases when the corneal nerve regeneration is observed [57]. The corneal nerves tend to regenerate, but unfortunately, de-spite a significant nerve regeneration, the corneal nerves do not fully recover [38, 58]. The normal cornea has sensory nerves that enter the stroma, branch-es, form the subepithelial plexus and extend the terminis into the epithelium. The nerves from the infected cornea hyperinnervate the corneal stroma but do not form a plexus at the subepithelium layer or extend to the epithelium, leaving the cornea less sensitive. The hyperinnervation of the corneal stroma by sympathetic nerves determines the severity of HSV keratitis, the distur-bance of the sensory nerve plexus, and the regeneration of the epithelial nerve endings, which was demonstrated in the mouse model [3]. It is not fully clear how the morphological changes of the corneal sub-basal nerves associate with the clinical characteristics and the corneal sensitivity in the human cor-neas during herpetic keratitis and after the infection, if there are any changes in the corneas of the patients with the previous history of herpes labialis.
1.5.2. Characterization of corneal endothelium damage
The endothelium cells form a monolayer located at the Descemet’s mem-brane in the posterior cornea and plays a significant role in maintaining the visual function: controls the cornea hydration and transfers nutrients and oth-er molecules from the aqueous humour [59]. As a result of aging, diseases, injuries, or surgeries EC could be damaged and influence the corneal blind-ness [60]. The minimum density of the corneal EC for normal functioning is thought to be around 500 cells/mm2 [61]. Recently the hypothesis has been
formed that the peripheral nervous system may promote the corneal endothe-lial cell survival and homeostasis [62]. It is not fully clear if the endotheli-um damage during herpetic keratitis is associated with the corneal sub-basal nerve parameters and corneal damage.
1.6. The corneal reaction to HSV
HSV is a linear, double-stranded deoxyribonucleic acid (DNA) virus that belongs to the Herpesviridae family. During the first episode of HSV-1 infec-tion, the virus replicates in the corneal epithelium cells which are vulnerable to the direct viral cytopathic effect. When the infection recurs, the corneal stroma is exposed. Due to an inflammatory response, the cornea becomes cloudy and cicatrized. The stromal tissue swelling and the loss of integrity are influenced by the inflammatory cell infiltration which is detected in the cornea [9].
The corneal cell interaction with HSV causes an inflammatory cascade with an anti-inflammatory cell infiltration which is responsible not only for the clearance of the virus but also determines the corneal damage due to the corneal opacification, neovascularization, and the corneal nerve loss.
1.7. The immune response
The HSV enters the cell by fusing with the cell membrane. After the en-trance, the synthesis of cytokines begins, the cells involved in the immune response, such as natural killers (NK), DC, neutrophils, macrophages and lymphocytes, are attracted to the site of inflammation. The corneal epithelial cells and the plasma DC express toll-like receptors which detect the DNA of the virus [63].
There is lack of data on how the dendritic cells infiltration to HSV infected cornea influences the morphological changes of the corneal sub-basal nerves and the clinical appearance of herpetic keratitis.
Changes in pro-inflammatory and anti-inflammatory cytokines ratio occur in the cornea during herpetic keratitis. One of the pro-inflammatory cytokines is IL-17. Cytokine IL-17 which is secreted by the T helper (Th) cells Th1 and Th17, through the effect of other cytokines and chemokines, activates a mas-sive neutrophil infiltration into the cornea [9, 64]. Neutrophils primarily mi-grate from the limbal blood vessels to the response of chemokine signals [65]. At the onset of keratitis, neutrophils play a protective role and are responsible for the virus clearance, but later, during the clinical phase of keratitis they contribute to the corneal tissue damage and angiogenesis [66]. IL-17 cor-relates with the intensity of herpetic keratitis and plays a pro-inflammatory role [10]. Most of the cornea damage is the result of the neutrofil infiltration and neovascularisation, therefore after IL-17 inhibition, a milder course of keratitis is observed [63]. Unlike IL-17, IL-10 plays a protective role in cor-nea damage in the course of HSV keratitis. IL-10 reduces the production of pro-inflammatory cells and limits the corneal inflammation and damage [12,
67, 68]. Elevated levels of cytokines including the ones detected within the corneas following HSV-1 infection are important contributors to the devel-opment of HSK pathogenesis. It was found that the levels of cytokine IL-1β, the inducer of the production of IL-17, were increased in blood serum in the cases of HSV keratitis [69]. Elevated levels of IL-1β are associated with cor-neal inflammation. IL-1β does not inhibit viral replication, but is important in pathogenesis [10]. IL-1β promotes the production of IL-17 which activates the inflammation in the cornea [70]. We do not know if IL-10 and IL-17 con-centration in the peripheral blood associates with the corneal damage in the cases of HSV keratitis. Recently the studies were performed that described elevated levels of different cytokines in blood serum at presentation of differ-ent types of eye diseases [71–74].
1.8. The role of dendritic and Langerhans cells
The DC are the most potent antigen presenting cells (APC) of the immune system to induce antiviral immune responses. They phagocytose the viral par-ticles and the infected cells and deliver them to the draining lymph nodes to generate an adaptive immune response [75]. A decrease of DC in the cornea determines later NK, monocytes, and chemokines migration to the site of in-fection resulting in a delayed virus removal [63]. LC are a population of DC that mediate the antigen presentation. LC are a discrete leukocyte population of professional, specialized APCs with an exceptional capacity for initiating T-lymphocyte responses and an expression of major histocompatibility com-plex class II antigens in the corneal epithelium [2, 76, 77]. HSV tries to evade the host immune response by inducing apoptosis and eliminating the DCs resulting in a delayed activation of T cells and enabling HSV replicate for a longer time. HSV also downregulates chemokine receptors on the DCs and inhibits the migration to the draining lymph nodes and therefore interferes with the induction of an antiviral immune response induced by the DC [78].
The eye is a structure with the immune privilege and can tolerate foreign antigens, allergens, and pathogens without eliciting significant immune re-sponses. Although the cornea, due to its immune privilege, has historically been attributed to a lack of the immune cells, recent studies have found den-deritic cells and macrophages in the centre and periphery of the healthy cor-nea, limbus and conjunctiva [13, 76]. In the centre of the cornea the density of LC is three times less than in the conjunctiva and limbal areas [27]. Due to HSV, LC start migrating to the paracentral and central cornea and undergo maturation as early as 24 hours after the induction of inflammation [8, 79]. In the acute phase of HSV keratitis, LC are located at the level of the lower intermediate cells, basal cells, and sub-basal nerve plexus. During the active
infection, a wide net with the infiltration of LC can be found [80, 81]. Pre-viously the correlation between LC, corneal sensation and corneal sub-basal nerve changes were found in patients with diabetes, dry eye disease and kera-toconus [82–84]. It is still not clear if LC density correlate with the corneal sub-basal nerve parameters and influence the changes in the corneal sensation and the formation of neurotrophic keratopathy during HSV keratitis.
1.9. The tear secretory immunoglobulin A
sIgA is a dimeric molecule synthesized in the subepithelium plasma cells of the lacrimal glands, which binds to the receptors on the epithelial cells and is released into the tear ducts. It is known as tear immunoglobulin against viral pathogens and infections. sIgA inhibits the pathogen adhesion and pene-tration to the mucosal epithelium [85]. The peak in sIgA secretion is observed within the range from 3 to 5 days of the herpetic eye disease and is detected during the first 10 days [86]. For sIgA positive patients, the herpetic eye dis-ease recurs more often than for sIgA negative patients. Tear sIgA was detect-ed as the only potential marker for prdetect-edicting recurrences of herpetic keratitis [17]. It is not known if in the course of herpetic keratitis, sIgA is associated with the LC infiltration and the corneal sub-basal nerve changes. All these factors could be associated with a more severe course of herpetic keratitis, more frequent recurrences of the disease and neurotrophic keratopathy for-mation.
1.10. The Cold Sore Susceptibility Gene
The region of the cold sore susceptibility gene includes a position from 15.7 to 18.6 Mb. The C21orf91 gene encodes a cytosolic protein which is called the early undifferentiated retina and lens protein but its function is still not clear. It is unknown whether the effect is concerned with the production of more or less of the encoded protein. The ways in which the C21orf91 gene influences the pathogenesis and manifestation of herpes labialis have not been established yet [19]. The C21orf91 gene plays a role in the biologi-cal processes such as the differentiation of cerebral cortex neurons, the differ-entiation of cells and the regulation of the dendritic spine development. It is expressed within the developing central nervous system of mice and humans [87]. The researchers have found a connection between the C21orf91 gene and herpes labialis, between the gene and Down syndrome and also hepato-cellular carcinoma, but there are no studies if there is a connection between the gene and HSV keratitis [18, 19, 87, 88].
Immune and genetic factors are important in the onset, course, and prog-nosis of herpetic keratitis. Therefore, we chose to investigate the influence of both immune and genetic factors on corneal morphological characteristics during HSV keratitis.
2. METHODS
2.1.The selection of subjects
The study was performed at the Department of Eye Diseases in the Hos-pital of Lithuanian University of Health Sciences Kaunas Clinics. The study was approved by the Kaunas Regional Biomedical Research Ethics Commit-tee on 9 July 2015, No. BE-2-26 and 2017-01-26 No. P1-BE-2-26/2015.
A prospective clinical study enrolled the patients with HSV keratitis treat-ed in an outpatient or inpatient Department of Eye Diseases from 2016 until 2019. Herpes labialis subjects and healthy controls participated voluntarily. A written informed consent was obtained from all the subjects who participated in the study.
Calculation of required minimum sample size of 76 HSV keratitis patients was determined by the Paniott formula [89]:
n = 1 οమ ା భ
N
n – sample size;
∆ – the size of the sample error (for this study, the size of the sample error of 5% was chosen);
N – general whole (it was estimated that per year an average of 94 patients with HSV keratitis visit the department of eye diseases in Lithuanian Univer-sity of Health Sciences Kaunas Clinics);
ൌ ͳ
Ǥହమାభ Ͳ
ൌ .
A prospective clinical study included and examined 269 subjects: 170 women (63.2%), 99 men (36.8%) of the mean age range of 59.0 ± 8.5 years. The patients excluded from the study were those with the past history of other ocular infection, trauma, contact lens wearing, diabetes mellitus, glaucoma, with the previous intraocular or refractive surgery and the patients who did not present for the examination 6 months later.
Epithelial keratitis was diagnosed in patients (n = 48) with characteristic isolated or merged dendritic infiltrates, geographic atrophy. Herpes stromal keratitis was diagnosed in patients (n = 15) who had stromal opacities or a destruction of the corneal stroma, ulcerative infiltrates and/or neovasculariza-tion. The patients (n = 16) who presented with a characteristic swelling of the cornea and precipitates on the corneal endothelium had entohelial keratitis: disciform endothelial keratitis was defined as a central part stroma swelling with the ring infiltrate and precipitates on the corneal endothelium; linear
endothelial keratitis was defined as a stroma swelling and linear form of pre-cipitates on the corneal endothelium; diffuse endothelial keratitis was defined as an entire corneal swelling and precipitates on the corneal endothelium.
All the subjects were divided into three groups:
Group I: 79 patients (of whom there were 38 women (58.1%) and 41 men (51.9%), the mean age range of the subjects was 59.5±9.0 years) with an ac-tive unilateral herpetic keratitis (right eye (OD, lat. oculus dexter) n = 42, left eye (OS, lat. oculus sinister) n = 27 (p>0.05), including epithelial (n = 48), stromal (n = 15), or endothelial keratitis (n = 16), according to the clinical history and clinical examination.
Group II: 101 subjects (72 women (71.3%), 29 men (28.7%), mean age 58.9±8.5 years) with the previous clinical history of herpes labialis, but no clinical history of the herpetic eye disease.
Group III: 89 healthy controls (60 women (67.4%), 29 men (32.6%) of the mean age range of 58.5±7.9 years) with no clinical history of any HSV diseases.
The division of the subjects into groups is presented in Picture 2.1.1.
SUBJ ECTS n = 269 GROUP I HSV keratitis patients n = 79 n = 101 n = 15 n = 16 n = 89 n = 48 GROUP II Herpes labialis subjects GROUP III Healthy controls Epithelial keratitis Contralateral
eyes Contralateral eyes
Stromal keratitis Contralateral eyes HSV affected eyes HSV affected eyes Endothelial keratitis HSV affected eyes
Picture 2.1.1 Study groups
HSV affected eyes and contralateral eyes were investigated in patients with HSV keratitis. There was no difference between the eyes in the subjects with herpes labialis and in the healthy controls, the OD data were further evaluated. The study groups were homogeneous by age and gender.
2.2. Clinical examination
Both eyes of all patients with HSV keratitis, herpes labialis subjects and healthy controls underwent an ophthalmological examination:
• Biomicroscopy (Nidek SL-2000, Japan) of the anterior eye segment: special attention was paid to the cornea: checking for the corneal epi-thelium defects, opacities, infiltrates, edema, neovascularisation, and keratic precipitates.
• Cochet-Bonnet aesthesiometry (Cochet-Bonnet; Luneau Ophthalmolo-gie, Chartres, France) within the central 5 mm of the cornea, which measures the ophthalmic branch sensory fibers in the cornea of the fifth cranial nerve. The aesthesiometer contains retractable monofilament that extends from 5 mm to 60 mm. As the length is decreased, the pres-sure transmitted is increased from 11mm/grms to 200 mm/grms. The measurement started with a full length of the 6 cm filament which was retracted in 0,5 cm steps until the subject felt the contact. Healthy cor-nea can feel the contact with filament of 6 cm length.
• The measurement of intraocular pressure with a non-contact tonometer which uses a small puff of air to flatten the patients cornea and measure the intraocular pressure in a millimetre of mercury (mmHg) (Nidek non contact tonometer NT-530/510, Japan). The upper limit of the normal intraocular pressure was considered to be 21 mmHg.
• Laser scanning confocal microscopy (Heidelberg Retina Tomograph 3 with the Rostock Cornea Module, Heidelberg Engineering GmbH, Dossenheim, Germany) [90]. The images of all layers of the cornea were obtained and the most representative three images were selected for further evaluation. Special attention was paid to the morphology of the corneal sub-basal nerves, LC and endothelium.
After six months from the first clinical examination, the same ophthal-mological investigation was repeated for HSV keratitis patients, and the data were compared with the primary test results, herpes labialis subjects and healthy controls.
2.2.1. The examination of corneal nerves
The sub-basal nerves were categorised as the main nerves, the branches which branched from the main nerves, and the total nerves – the main nerves and the nerve branches. The analysis of the corneal nerves was performed using an automated Corneal Nerve Fibre Analyser ACCMetrics V.2. The software automatically extracts and quantifies the nerve fibre metrics in the images obtained using the Heidelberg HRT III corneal confocal microscope
with 384 × 384 pixels and the field of view of 400 × 400 μm2 (resolution:
400/384 = 1,0417 μm) (Picture 2.3.1.1). The cornea is examined by LSCM in 1-4 μm thick sections parallel to the corneal surface, the same direction as run corneal nerves.
Picture 2.2.1.1 Laser scanning confocal microscopy images of corneal
sub-basal nerves: A – herpetic keratitis with a severe sensation loss; B – a patient with Herpes labialis; C – a healthy patient
It quantifies the corneal nerve fibre length (CNFL) – the total length of nerves mm/mm2, the nerve fibre density (CNFD) – the number of fibres/mm2,
the nerve branch density (CNBD) – the number of branch points on the main fibres/mm2, and the nerve fibre total branch density (CTBD) – the total
num-ber of branch points/mm2 [91–97].
2.2.2. The examination of Langerhans and endothelium cells
Langerhans cells were morphologically identified as bright cellular imag-es with a branching morphology located at the sub-basal nerve plexus layer (Picture 2.3.2.1).
Picture 2.2.2.1 Laser scanning confocal microscopy images of:
The LC and EC density were counted manually using the software analysis of the instrument within a region of standardized dimensions (250 × 250 μm) and the result was given as cells/mm2 [14].
2.2.3. The detection of tear sIgA
The tear samples were investigated for the patients with HSV keratitis during the first visit (affected eyes and contralateral eyes separately), herpes labialis subjects and healthy controls. The tear samples for sIgA concentra-tion determinaconcentra-tion were collected from the lower fornices with Biotech sterile surgical polyvinyl acetal (PVA) eye spears with no use of local anaesthesia during the process. While using PVA eye spears we managed to collect larger amounts of tears compared to other previously described methods with cap-illary tubes and Schirmer strips, we decided to continue tear collection with PVA eye spears [98]. For samples extraction the spears with the collected tears were centrifuged with sponge up in the Eppendorf tubes for 7 minutes at 13300 rpm/min with high-speed microcentrifuge (Fisherbrand accuSpin Micro 17) immediately (up to 10 minutes) after collection in the Laboratory of Ophthalmology, Institute of Neurosciences. The samples were stored in –20 °C until used [99]. The thawed samples were diluted 1:100 with sample buffer and investigated with the tear sIgA enzyme-linked immunosorbent as-say (ELISA) kit (Cusabio, China, detection range 0.012 μg/mL–50 μg/mL) in the Laboratory of Immunology, Department of Immunology and Allergology, Hospital of Lithuanian University of Health Sciences Kaunas Clinics.
2.2.4. Blood sample collection
Blood from the peripheral vein was collected in BD Vacutainer® SSTTM II Advance blood tubes and left to coagulate. The tubes were then kept for 10 min. centrifuge at 1300 × g at room temperature to separate blood serum. Immediately after separation, the serum was collected and transferred to 1 mL refrigerated tubes and frozen at –20 °C for further protein analysis. ELISA measurements were performed by collecting a sufficient number of samples. Results were expressed as protein concentration per mL of serum. 96 plates pre-coated with primary antibodies against the test proteins were used for the study.
2.2.5. The detection of HSV specific antibodies and interleukins
For all patients with HSV keratitis (during the first visit), herpes labia-lis subjects and healthy controls the serology tests of the serum were per-formed in the Laboratory of Immunology, Department of Immunology and
Allergology, Hospital of Lithuanian University of Health Sciences Kaunas Clinics. aiming to detect HSV 1/2 IgG, IgM, IL-10 and IL-17. HSV 1/2 IgG, IgM were detected using the ELISA classic kit (Virion Serion, Germany, de-tection range 10–500 U/mL). The negative results were obtained when the IgG or IgM concentrations were <20 U/mL, possible 20–30 U/mL, positive >30 U/mL according to the manufacturer’s recommendations. The IgG po-sitive results were divided into two groups according to the median of the results: I – when the concentration was less than 10 000 U/mL, II – when the concentration was more than 10 000 U/mL.
IL-10 and IL-17 were determined by the ELISA kits (Elabscience, USA, detection range: IL-10 – 7.81–500 pg/mL, IL-17 – 31.25–2000 pg/mL) in the blood serum which was stored in the temperature of –20 °C until used. The concentration of each IL was measured and submitted by pg/mL.
2.2.6. The C21orf91 gene genotyping
The C21orf91 rs1062202 and rs10446073 were genotyped using the al-time polymerase chain reaction (PCR) method with the Rotor-Gene Q re-al-time PCR quantification system for HSV keratitis patients, herpes labialis subjects and healthy controls in the Laboratory of Ophthalmology, Institute of Neurosciences. The single nucleotide polymorphisms (SNPs) were deter-mined using the TaqMan genotyping assay according to the manufacturer’s manual.
2.3. The literature search
A search for literature was performed using electronic databases, check-ing reference lists of the published articles and searchcheck-ing trials registry. The search was applied to MEDLINE, EBSCO and Oxford journals electronic da-tabases, Cochrane Controlled Trials Register, Biomed central virtual, Black-well synergy digital and Wiley online libraries, and Springerlink online de-livery platform.
A specific literature search was performed describing the morphological changes of the cornea during HSV keratitis, influence of immune and genet-ic factors on HSV keratitis using combinations of keywords: HSV, herpes, herpetic, keratitis, labialis, infection, confocal microscopy, cornea, sub-basal nerves, Langerhans cells, endothelium, tear sIgA, IgM, IgG, IL-10, IL-17, C21orf91. The publication type was restricted to a clinical study. There was no limit for language and publishing years.
2.4. The statistical analysis
The statistical analysis was performed with the statistical package for the social sciences (SPSS) 23 programme. All the parametric data were expressed as the mean and standard deviation. Kolmogorov-Smirnov test was used for the determination of the quantitative data distribution. The results were an-alysed by the Kruskal-Wallis and Mann-Whitney tests. The Kruskal-Wallis test was applied to compare the scores for more than two independent groups and the Mann-Whitney test – for the scores of two independent groups. The differences between the dependent variables were analysed by the Wilcoxon signed-rank test. The Spearman correlation coefficient was calculated to de-termine the relationship between the nerve density parameters and the aesthe-siometry results in HSV affected eyes.
The frequencies of the qualitative variables were compared using the χ2
tests. In order to assess the minimally false negative and the minimally false positive results with greatest accuracy, the method of receiver operating char-acteristics (ROC) curve was used. The logistic regression analysis was per-formed to determine the odds ratio (OR) predictive value. The differences were considered statistically significant when p values <0.05. The patient age was taken into account during the statistical calculations of the data, some of the parameters tended to decline with the age, therefore it was excluded.
3. RESULTS
3.1. The evaluation of the corneal sub-basal nerves
The HSV affected eyes showed a reduction in the mean corneal sensitivity and the sub-basal nerve parameters, while an increase was observed in the mean intraocular pressure when compared with contralateral eyes, herpes la-bialis and healthy control group eyes (p < 0.05). The mean corneal sensitivity and the sub-basal nerve parameters of the contralateral eyes were also differ-ent when comparing with the herpes labialis patidiffer-ents and the healthy controls (p < 0.05). There was no difference of the parameters between the eyes in the herpes labialis and healthy control groups. The demographics of the subjects are presented in Table 3.1.1. The data of all the groups and subgroups are presented in Table 3.1.2.
Table 3.1.1 Demographic data of patients with HSV keratitis and control
groups
Parameters
HSV keratitis group Herpes
labialis group
Healthy controls group Epithelial Stromal Endothelial Total
n = 48 n = 15 n = 16 n = 79 n = 101 n = 89 Age (years) (mean ± SD) 61.0±8.9 55.7±8.1 57.2±11.4 59.9±9.0 58.9±8.5 58.5±7.9 Sex, n (female/ male) 29/19 4/11 5/8 38/41 72/29 60/29
Table 3.1.2 The quantitative analysis of the corneal sensitivity, intraocular
pressure and sub-basal nerve fibre parameters in the HSV keratitis subgro-ups, contralateral eyes and control groups
Parameters
HSV keratitis group Herpes
labialis group Healthy controls group
Epithe-lial Stromal Endo-thelial
HSV affected eyes Contra-lateral eyes n = 48 n = 15 n = 16 n = 79 n = 79 n = 101 n = 89 CNFD n/mm2 2.2± 5.4a 1.7±3.7 5.3±8.4a 5.82.6±abc 7.69.9±ade 19.1±12.2bd 15.8±10.8ce CNBD n/mm2 2.0± 5.5b 0.8±2.2c 13.18.2±bc 7.32.8±abc 13.511.7±ade 22.0±20.7bd 21.3±24.2ce CNFL mm/mm2 6.0± 4.2 6.2±3.7 8.4±5.3 4.36.5±abc 6.39.6±ade 12.8±5.2bd 11.9±4.9ce
Parameters
HSV keratitis group Herpes
labialis group Healthy controls group
Epithe-lial Stromal Endo-thelial
HSV affected eyes Contra-lateral eyes n = 48 n = 15 n = 16 n = 79 n = 79 n = 101 n = 89 CTBD n/mm2 19.8± 22.6 21.3±18.4 22.6±25.5 20.6±22.1ab 25.2±20.6cd 38.4±31.2ac 34.038.9±bd Corneal
sensiti-vity, cm 2.8±2.0 2.2±1.6d 3.7±1.3d 1.92.8±abc 1.45.1±ade 5.7±0.6bd 5.7±0.6ce
Intraocular pressure,
mmHg
17.1±
6.8e 19.5±9.1 25.4±10.6e 19.0±8.5abc 15.5±4.3a 14.1±3.0b 14.9±3.6c abcdep < 0.05, by Mann-Whitney test abcdep < 0.05, by Mann-Whitney test
Values reported as mean ±SD. CNFD – corneal nerve fibre density; CNBD – corneal nerve branch density; CNFL – corneal nerve fibre length; CTBD – corneal nerve total branch den-sity.
The intraocular pressure did not correlate with the corneal sub-basal nerve parameters or the corneal sensitivity. With the decreased CNFL and CNBD, the corneal sensitivity was lower (r = 0.247, p < 0.05). CNFD, CNBD and CTBD were not associated with corneal edema.
Six months later all of the examinations were performed repeatedly on the patients with the herpetic eye disease and compared with the primary ex-amination data. During the follow up, the corneal sensitivity increased in the HSV affected eyes but did not reach the contralateral eye or herpes labialis and healthy controls (p < 0.05). The changes in the corneal sensitivity and the intraocular pressure are demonstrated in Figure 3.1.1 and Figure 3.1.2.
0 1 2 3 4 5 6 Corneal sensitivity, cm HSV affected
eyes HSV affectedeyes after 6 months
Contralateral
eyes Contralateraleyes after 6 months * * ade abc bd ce
Figure 3.1.1 The corneal sensitivity during the first visit and
b 5 0 15 20 25 30 35 40 45 10 Intraocular pressure, mmHg * abc ad cd HSV affected
eyes HSV affectedeyes after 6 months
Contralateral
eyes Contralateraleyes after 6 months
Figure 3.1.2 Intraocular pressure during the first visit and during the follow
up 6 months later. abcdp < 0.05 by Wilcoxon test
Depending on the type of keratitis, the best recovery in the corneal sensi-tivity was observed in the endothelial type of keratitis and it was significant-ly higher than the epithelial type of keratitis (p < 0.05) (Table 3.1.3, Figure 3.1.3).
Table 3.1.3 The quantitative analysis of the corneal sensitivity and the
in-traocular pressure changes during the first visit and during the follow up 6 months later
Parameters
HSV affected eyes
during the first visit during the follow upHSV affected eyes Epithelial Stromal Endothe-lial Epithelial Stromal Endothe-lial
n = 48 n = 15 n = 16 n = 48 n = 15 n = 16
Corneal
sensiti-vity, cm 2.8±2.0a 2.2±1.6b 3.7±1.3c 3.5±2.0a 4.0±1.8b 5.2±1.0c
Intraocular
pressure, mmHg 17.1±6.8d 19.5±9.1 25.4±10.6e 13.6±5.3d 16.1±4.6 14.8±4.4e abcdep < 0.05 by Mann-Whitney test.
a a 6.5 6.0 –4.5 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 –0.5 –1.0 –1.5 –2.0 –2.5 –3.0 –3.5 –4.0 Stromal HSV keratitis Epithelial
Corneal sensitivity change during the follow up to 6 months later
Endothelial
Figure 3.1.3 Changes in the corneal sensitivity during the follow up
6 months later denpending on the type of HSV keratitis. HSV affected eyes: χ2 = 7.392, lls = 2, p = 0.025,
by Kruskal-Wallis test, ap = 0.02, by Mann Whitney test
During the follow up, the largest decrease in the intraocular pressure was observed in the endothelial type of HSV keratitis and it was significantly dif-ferent from other types of HSV keratitis (p < 0.05) (Table 3.1.2, Figure 3.1.4).
b ab a 40 35 30 25 20 15 10 5 0 –5 –10 –15 –20 Epithelial Endothelial
Intraocular pressure change during the follow up to 6 months later
Stromal
HSV keratitis
Figure 3.1.4 Changes in the intraocular pressure during the follow up 6
months later denpending on the type of HSV keratitis.
HSV affected eyes: χ2 = 6.988, lls = 2, p = 0.03, by Kruskal-Wallis test, abp < 0.05 by Mann Whitney test
During the follow up corneal sub-basal nerve parameters increased in the HSV affected and contralateral eyes, but did not reach the herpes labi-alis patients and the healthy controls (p < 0.05). The data of the changes in the sub-basal nerve parameters are presented in Table 3.1.4. The change of sub-basal nerve parameters was not associated with type of herpetic keratitis.
Table 3.1.4 The quantitative analysis of the corneal sub-basal nerve changes
during the first visit and during the follow up 6 months later
Parame- ters
HSV keratitis group Herpes
labialis group
Healthy controls group HSV affected eyes Contralateral eyes
First visit Follow up First visit Follow up
n = 79 n = 79 n = 79 n = 79 n = 101 n = 89
CNFD
Parame- ters
HSV keratitis group Herpes
labialis group
Healthy controls group HSV affected eyes Contralateral eyes
First visit Follow up First visit Follow up
n = 79 n = 79 n = 79 n = 79 n = 101 n = 89
CNBD
n/mm2 2.7±7.2ab1 10.78.1±ab6ce 11.7±13.5ab1 15.2±16.2ab6df 22.0±20.7cd 21.3±24.2ef
CNFL
mm/mm2 6.3±4.3b1 4.37.3±b6ce 9.6±6.3ab1 9.7±4.2ab6df 12.8±5.2cd 11.9±4.9ef
CTBD
n/mm2 20.1±21.9 22.2±18.5ce 25.2±20.6 29.7±25.6 38.4±31.2c 38.9±34e ab1b6cdefp < 0.05
CNFD – the corneal nerve fibre density; CNBD – the corneal nerve branch density; CNFL – the corneal nerve fibre length; CTBD – the corneal nerve total branch density. a – p value
between the parameters of the first visit and the follow up; b1 – p value between the
para-meters of the HSV affected and the contralateral eyes during the first visit; b6 – p value
be-tween the parameters of the HSV affected and the contralateral eyes during the follow up by Wilcoxon test; c – p value between the parameters of the HSV affected eyes and the herpes
labialis group during the follow up; d – p value between the parameters of the contralateral
eyes and the herpes labialis group during the follow up; e – p value between the parameters
of the HSV affected eyes and the healthy controls during the follow up; f – p value between
the parameters of the contralateral eyes and the healthy controls during the follow up by Mann-Whitney test.
Cornea edema was found in 61.5% eyes during the active phase of her-petic keratitis and in 16.4% eyes during follow up 6 months later. Keratic precipitates decreased from 60.8% eyes before to 17.9% after 6 months. Cor-nea neovascularization increased from 13.9% to 20.9% eyes. There were 7 patients (8.9%) who 6 months later had reactivation of herpetic keratitis with superficial epithelium defect or even corneal ulceration.
Based on the receiver operating characteristic (ROC) analysis, herpetic keratitis could be suspected when CNFD and CNBD is less than 3.13 n/mm2,
CNFL less than 9.92 mm/mm2, and CTBD less than 15.62 n/mm2 in the
eyes with keratitis of unknown ethiology, and CNFD less than 15.62 n/mm2,
CNBD less than 16.15 n/mm2, CNFL less than 10.07 mm/mm2, and CTBD
less than 37.5 n/mm2 in the contralateral eyes. The detailed ROC analysis is
presented in Table 3.1.5.
Table 3.1.5 The distribution of the predicted values and the characteristics of
the corneal sub-basal nerve parameters for the receiver operating characte-ristic test by groups of subjects
Parameter/ threshold value Area under the ROC curve (%) Sensitivity/ Specificity (%) HSV affected patients/ Herpes labi-alis and
heal-thy controls (%)
p value affected patients OR of HSV [95% CI] CNFD herpetic eye <3.13 87.5 78.5/91.1 78.5/8.9 <0.001 [17.847–77.184]37.114 CNFD healthy eye <15.62 70.0 80.6/50.5 80.6/49.5 <0.001 [2.173–8.282]4.242 CNBD herpetic eye <3.13 80.0 82.3/76.3 82.3/23.7 <0.001 [7.676–29.158]14.96 CNBD healthy eye <16.15 62.2 73.1/50.5 73.1/49.5 0.001 [1.51–5.119]2.78 CNFL herpetic eye <9.92 80.6 83.5/65.3 83.5/34.7 <0.001 [4.903–18.557]9.538 CNFL healthy eye <10.07 68.9 67.2/63.7 67.2/36.3 <0.001 [1.99–6.467]3.587 CTBD herpetic eye <15.62 67.1 58.2/72.6 58.2/27.4 <0.001 [2.136–6.407]3.699 CTBD healthy eye <37.5 61.3 80.6/41.6 80.6/58.4 0.001 [1.512–5.782]2.956 OR – odds ratio; CI – confidence interval; ROC – receiver operating characteristic; CNFD – corneal nerve fibre density; CNBD – corneal nerve branch density; CNFL – corneal nerve fibre length; CTBD – corneal nerve total branch density.
In all, 36 (45.6%) patients with herpetic eye disease previously suffered from herpes labialis: 30.6% rarely, 47.2% 1–2 times per year, 22.2% more than 2 times per year. In 22.2% of the patients, herpes labialis started in child-hood, 13.9% – during adolescence, 41.7% when they were 20-30 years old, 13.9% – 40 years or older, 8.3% could not remember. We did not find any evidence that quantity of herpes labialis recurrence could be a risk factor for herpetic eye disease, however, the comparison of those who contracted the disease showed the likelihood of more chances for the development of the herpetic eye disease if the patient contracted the first herpes labialis during childhood OR [95% CI], 4.524[1.448–14.135] (p < 0.05). The previous her-pes labialis did not influence the corneal sensitivity.
3.2. The evaluation of the density of the Langerhans and endothelium cells
The LSCM examination revealed that 48 HSV patients with epithelial ker-atitis had swollen or necrotic epithelial cells with decreased sub-basal nerve parameters and a large number of LC infiltration around the nerves. In 8 pa-tients with epithelial keratitis we could see the damage of anterior stroma, with enhanced reflectivity of the cells and the difficulty in seeing the cell nuclei. In 31 patients with stromal and endothelial HSV keratitis, we found an unclear structure of the stroma with a large number of LC and other inflam-matory cells infiltration. In 12 patients with endothelial HSV keratitis, we found pseudo guttata with enlarged intercellular gaps, spot holes and a loss of endothelium cell borders.
The HSV affected eyes showed an increase in the LC density when com-pared with the contralateral eyes, herpes labialis, and healthy control groups (357.9±308.7 vs. 197.1±258.5 vs. 62.9±72.6 vs. 81.7±88.3) (p < 0.05). The LC density of the contralateral eyes were also significantly different from the control groups (p < 0.05).
The EC density in the HSV affected eyes did not differ from the contralat-eral eyes, but it was lower when comparing with the herpes labialis patients and healthy controls (2300.7±530.6 vs. 2343.3±543.7 vs. 2672.9±268.1 vs. 2620.9±347.1) (p < 0.05). The LC and EC density did not differ between the HSV keratitis subgroups. There was no significant difference of the LC and EC density between the eyes with herpes labialis and healthy control groups. The data of all groups and subgroups are presented in Table 3.2.1.
Table 3.2.1 The quantitative analysis of the density of Langerhans cells and
endothelium cells in patients with HSV keratitis and the control groups
Parameters HSV keratitis group Herpes labialis group Healthy controls group HSV keratitis HSV affected eyes Contra lateral Eyes
Epithe-lial Stromal Endothe-lial
n = 48 n = 15 n = 16 n = 79 n = 79 n = 101 n = 89
LC
cell/mm2 358.4371.9±abc 234.5355.2±def 201.2304.5±ghi 357.1±312.8jkl 258.5197.1±adgjmn 72.662.9±behkm 88.381.7±cfiln
EC
cell/mm2 2420.3±467.6ab 2664.7±120.2 2127.3±235.7cd 2405.1±435.2ef 2398.2±506.6gh 2672.9±268.1aceg 2620.9±347.1bdfh abcdefghijklmnp < 0.05, by Mann-Whitney test
The values reported as mean ±SD. LC – Langerhans cells; EC – endothelium cells. Lan-gerhans cells: χ2 = 10.851 df = 3, p < 0.005; endothelium cells: χ2 = 2.614, df = 3, p > 0.05,
Table 3.2.2 The correlation between the corneal Langerhans cell density and
the corneal sub-basal nerve parameters in HSV affected eyes
Parameters
Corneal sub-basal nerve parameters
CNFD CNBD CNFL CTBD
r, p
LC cell/mm2 –0.249, < 0.001 –0.182, < 0.005 –0.246, < 0.001 –0.132, < 0.05
EC cell/mm2 0.076, < 0.05 0.218, < 0.05 0.186, < 0.05 0.332, < 0.05
r – Spearman correlation coefficient; p – p values; LC – Langerhans cells; EC – endothelium cells; CNFD – corneal nerve fibre density; CNBD – corneal nerve branch density; CNFL – corneal nerve fibre length; CTBD – corneal nerve total branch density.
There was no association between the LC and the EC density and the fre-quencies of the herpetic keratitis and the manifestation of herpes labialis. The LC or EC density during the active herpetic eye disease did not affect the corneal clinical signs (corneal edema, neovascularisation, opacities formation and keratic precipitates) 6 months later.
Six months later, the LC density in the HSV affected eyes and the con-tralateral eyes had decreased, but did not reach that of the herpes labialis group and healthy controls (p < 0.05). The EC density in the HSV affected eyes did not change (Table 3.2.3).
Table 3.2.3 The quantitative analysis of the changes in the density of
Lan-gerhans cells and endothelium cells during the follow up 6 months later in patients with HSV keratitis and the control groups
Parame-ters
HSV keratitis group Herpes
Labialis group
Healthy controls HSV affected eyes Contralateral eyes
First visit Follow up First visit Follow up
n = 79 n = 79 n = 79 n = 79 n = 101 n = 89
LC cell/
mm2 308.7357.9±ab1 195.1221.7±ab6c 258.5197.1±ab1 149.5143.5±ab6df 62.9±72.6cd 81.7±88.3ef
EC cell/
mm2 2300.7±530.6 2374.0±483.6ce 2343.3±543.7 2367.8±524.5df 2672.9±268.1cd 2620.9±347.1ef ab1b6cdefp < 0.05
LC – Langerhans cells; EC – endothelium cells. a – p value between the parameters
obtai-ned during the first visit and the follow up; b1 – p value between the parameters of the HSV
affected and the contralateral eyes during the first visit; b6 – p value between the parameters
of the HSV affected and the contralateral eyes during the follow up by Wilcoxon test; c – p
value between the parameters of the HSV affected eyes and the herpes labialis group during the follow up; d – p value between the parameters of the contralateral eyes and the herpes
labialis group during the follow up; e – p value between the parameters of the HSV affected
eyes and the healthy controls during the follow up; f – p value between the parameters of the
3.3. The evaluation of the tear sIgA titers
The tear sIgA was examined in 61 patients with an active herpetic eye disease, including epithelial (n = 14), stromal (n = 7) or endothelial keratitis (n = 7), 22 subjects with the previous history of herpes labialis and 11 healthy controls with no history of any HSV diseases.
The sIgA concentration was significantly higher in the HSV affected eyes and the contralateral eyes when compared with the herpes labialis pa-tients and healthy controls (2046.0±375.0 μg/mL vs. 606.3±147.8 μg/mL vs. 661.3±191.7 μg/mL) (p < 0.05) (Figure 3.3.1). 8000 6000 4000 2000 0 ab cd a c b d * * * * * * * * sIgA concentration, μ g/mL HSV keratitis
group Herpes labialisgroup Healthy controlsgroup HSV affected eyes Contralateral eyes
Figure 3.3.1 The difference in the tear secretory immunoglobulin A
con-centration between the groups. HSV affected eyes: χ2 = 21.3597, df = 2,
p < 0.001, by Kruskal-Wallis test, abp ≤ 0.01, by Mann-Whitney test.
Contralateral eyes: χ2 = 17.732, df = 2, p < 0.001, by Kruskal-Wallis test cdp < 0.01, by Mann Whitney test
There was no difference in the sIgA concentration between the herpes la-bialis patients and the healthy controls. The sIgA concentration also did not differ between the different types of HSV keratitis.
Higher concentrations of sIgA were found in patients with the first time keratitis compared with the patients who already had a previous history of herpetic keratitis (p < 0.05) (Figure 3.3.2).
2000 4000 6000 8000 10000 0
Incidence rate of herpetic keratitis1 >1
sIgA concentration in HSV af fected eyes, μ g/mL
Figure 3.3.2 The tear secretory immunoglobulin A concentration and the
incidence rate of herpetic keratitis. p = 0.001, by Mann-Whitney test There was a positive correlation between the sIgA concentration and LC in the HSV affected eyes during the acute phase of the disease: a higher concentration of sIgA correlated with a higher LC infiltration (p < 0.001) (Figure 3.3.3).
0 2000 4000 6000 8000 10000 1000 800 600 400 200 0
sIgA concentration in HSV affected eyes, μg/mL
Langerhans cell density
Figure 3.3.3 The correlation between the tear secretory immunoglobulin
A concentration and the Langerhans cell density in the HSV affected eyes. r = 0.586, p < 0.001; y = 114.066 + 0.052*x, x – tear secretory
immunoglo-bulin A concentration in the HSV affected eyes, y – Langerhans cell density in the HSV affected eyes
The sIgA concentration in both eyes negatively correlated with the corneal sub-basal nerve parameters during the acute phase of the disease: a higher concentration of sIgA agreed with lesser corneal sub-basal nerve parameters (p < 0.005) (Table 3.3.1).
Table 3.3.1 The correlation between the tear secretory immunoglobulin A
concentration and the corneal sub-basal nerve parameters
sIgA concentration
Corneal sub-basal nerve parameters
CNFD CNBD CNFL CTBD r, p HSV affected eyes –0.587,< 0.001 –0.541, < 0.001 –0.557,< 0.001 –0.398, 0.001 Contralateral eyes –0.56,< 0.001 –0.518, < 0.001 –0.559,< 0.001 –0.396, 0.004 r – Spearman correlation coefficient; p – p values; sIgA – tear secretory immunoglobulin A; CNFD – corneal nerve fibre density; CNBD – corneal nerve branch density; CNFL – corneal nerve fibre length; CTBD – corneal nerve total branch density.
