|Year : 1994 | Volume
| Issue : 4 | Page : 171-192
Current perspectives in infectious keratitis
Vinay Agrawal1, Jyotirmay Biswas2, HN Madhavan2, Gurmeet Mangat3, Madhukar K Reddy4, Jagjit S Saini3, Savitri Sharma4, M Srinivasan5
1 Bombay City Eye Institute, Bombay, L.V. Prasad Eye Institute, Hyderabad, India
2 Sankara Nethralaya, Madras, India
3 Postgraduate Institute of Medical Education and Research, Chandigarh, India
4 L.V. Prasad Eye Institute, Hyderabad, India
5 Aravind Eye Hospital, Madurai, India
Madhukar K Reddy
L.V. Prasad Eye Institute, Road No. 2, Banjara Hills, Hyderabad 500 034
Source of Support: None, Conflict of Interest: None
|How to cite this article:|
Agrawal V, Biswas J, Madhavan H N, Mangat G, Reddy MK, Saini JS, Sharma S, Srinivasan M. Current perspectives in infectious keratitis. Indian J Ophthalmol 1994;42:171-92
|How to cite this URL:|
Agrawal V, Biswas J, Madhavan H N, Mangat G, Reddy MK, Saini JS, Sharma S, Srinivasan M. Current perspectives in infectious keratitis. Indian J Ophthalmol [serial online] 1994 [cited 2020 Mar 28];42:171-92. Available from: http://www.ijo.in/text.asp?1994/42/4/171/25566
| I. INTRODUCTION|| |
Corneal blindness is a major public health problem in India and infections constitute the most predominant cause. Infectious keratitis has for long been the Achille' heel of most ophthalmic surgeons. A wide spectrum of microbial organisms can produce corneal infections and consequently the therapeutic strategies may be variable [Figure - 1]. Proper public health initiatives can effectively prevent sight-threatening corneal infections and aggressive initial treatment for clinical cases of infectious keratitis can minimize the incidence of post-infectious corneal scars. One of the key elements in this effort is a proper understanding of the microbiological and clinical characteristics of this disease entity which will enable the ophthalmologist to initiate appropriate antimicrobial therapy.
A recent survey carried out in Southern California covering a representative population of ophthalmologists had very revealing figures. A total of 47.8 percent of the patients with a clinical diagnosis of infectious keratitis were treated with antibiotics without any cultures being obtained.  If this is the case in countries with advanced care, a thought about the possible situation in developing countries like ours would indeed be revealing. It is time, however, that we stop using the excuse of limited resources to hide our sins of omission about the recommended steps in the approach to the treatment of any disease. More often than not, it is the lack of proper training and outlook rather than the paucity of resources which is responsible for irregular treatment modalities.
This review is an attempt to provide the reader with a survey of all aspects of infectious keratitis. Emphasis is given to the most recent information. Although a lot of progress has been made in the medical therapy of bacterial keratitis, fungal keratitis still remains a therapeutic challenge.
| II. CLINICAL FEATURES OF INFECTIOUS KERATITIS|| |
This section will cover some salient features of infectious keratitis that may help in the initial diagnosis.
II. A. Bacterial Keratitis
Although the clinical characteristics of bacterial keratitis caused by specific individual organisms have been previously described, it is not generally possible to identify the causative agent by clinical examination alone. The microorganisms by virtue of their toxins, adherence capabilities, invasiveness and even strain differences within each species may produce different types of clinical picture. The clinical picture may also vary or alter in these patients using contact lenses or with a previous corneal scar due to viral infections, trauma, or surgery. However, some general clinical descriptions may be useful. For example, grampositive organisms tend to produce discrete, small abscess like lesion and gram-negative bacteria are more likely to cause diffuse, rapidly spreading necrotic lesions, but exceptions do occur. The bacterial or fungal infection may strongly be suspected on the basis of clinical criteria, but confirmation by scrapings and culture is essential. Herpetic keratitis, noninfectious keratitis, fungal keratitis and even toxic keratitis can all be confused with bacterial keratitis without proper laboratory diagnostic studies.
Patients with infectious corneal ulceration complain of pain, watering, foreign body sensation, redness and decreased vision. Pain is a prominent symptom but it varies with the size of the ulcers and nature of the organism. Watering and pain are more severe in rapidly spreading ulcer caused by Pseudomonas and Streptococcus pneumoniae species. Indolent ulcers due to Moraxella and Staphylococcus may be quiet and less symptomatic [Figure - 2]. Marked lid oedema and conjunctival chemosis are commonly seen in corneal ulceration due to gram-negative organisms especially following trauma and gonococcal infection. However, one should remember that patients who had topical steroids will exhibit signs and symptoms in milder form.
Hypopyon, one of the most important clinical signs of any type of infectious keratitis could be considered as the most important sign in establishing the aetiological diagnosis. For example, a definitive diagnosis of pneumococcal ulcer was based on hypopyon. It was once called as central hypopyon corneal ulcer. Haemorrhagic hypopyon is attributed to either Pneumococcal or Herpes simplex viral keratitis. It is also seen in fungal ulcer. One should remember that hypopyon could form with viral, fungal and parasitic ulcers. Non-bacterial or nonfungal problems such as Bechet's syndrome, abuseof topical anaesthetic agents and severe bums may also give rise to hypopyon. In general, corneal ulceration associated with purulent or mucopurulent discharge from lacrimal sac of the same side gives diagnostic clue in favour of pneumococcal ulcer; whereas, greenish discharge makes one think of Pseudomonas ulcer. Gonococcal ulcer in infants which, is rare now, presents with lid oedema, marked chemosis and purulent discharge.
One of the most dreaded corneal ulcers is that caused by Pseudomonas species. It is typically characterized by rapid evolution, primary involvement of corneal stroma and rapid spread to involve the entire cornea, unusual size of hypopyon and mucous consistency and often greenish colour of the pus [Figure - 3].
II. B. Fungal Keratitis
Corneal infection of fungal aetiology is very common and may represent 30 to 40 percent of all cases of culture-positive infectious keratitis in South India. Of these, Aspergillus and Fusarium are responsible for 70 percent of cases. These affect young, immunocompetent healthy adults, more often from rural areas. A history of trauma with organic matter is elicited in a significant percentage of cases. The ulcer commences insidiously and runs an indolent course. It begins at the midperiphery of healthy cornea in the exposed areas. The epithelium may show defect at the site of infiltration or epithelial defect would have healed with deep stromal infiltrate or endothelial plaque. The ulcer spreads towards the centre of the cornea. Moderate hypopyon or cheesy hypopyon is frequently noticed. In rare cases one may see a haemorrhagic hypopyon. The ulcer base has a raised, wet, soft creamy to greyish-white or yellowish-white infiltrate without mucous or exudate [Figure - 4]. It has feathery or hyphate borders. In the early stages a dendritic pattern may be seen leading to misdiagnosis and treatment with antiviral drugs. In advanced stages with involvement of the whole cornea, the typical clinical signs of fungal ulcer become obliterated. Satellite lesions and immune ring, which are infrequent, may assist in diagnosing fungal ulcer. Endothelial plaque and posterior corneal abscess are seen more frequently than described earlier. In pigmented ulcers (chromomycosis) the brown, dark brown or black pigment covering the ulcer base is the unique clinical sign caused by pigmented fungi (dematiaceous). In some of these eyes the slough is dry, tough and leathery and one usually needs a blade to remove it. These ulcers heal very slowly.
Keratitis caused by Candida is extremely rare in India and should be differentiated from staphylococcal and Moraxella ulcers. Stromal herpes and other low virulent bacterial ulcers should be considered in the differential diagnosis. All these clinical features mentioned earlier may be masked or altered by using native medicines, steroids or/and minor surgical procedures. Although fungal ulcers spread very slowly, corneal perforation can occur within 5 to 6 days from the onset as happens in Pseudomonas ulcer.
II. C. Acanthamoeba Keratitis
This parasitic infection is now reported with increasing frequency throughout the world. Of the 44 cases diagnosed by one of us (MS), only two were contact lens wearers. The experience of others in India has been similar (unpublished data). Trauma with organic matter, exposure to muddy or brackish water are the major predisposing factors. Delayed diagnosis is common. Even though severe pain is considered as the characteristic clinical symptom, we have not seen any marked difference in this symptom between Acanthamoeba keratitis and fungal keratitis. A history of unsuccessful treatment by several ophthalmologists with various ophthalmic topical preparations is more often the rule than an exception. Acanthamoeba keratitis is usually never suspected or diagnosed during the first visit. Ring of stromal infiltration at midperiphery of the cornea without involving the pupillary area with intact gray or hazy epithelium is noticed in a high percent of cases [Figure - 5]. Superficial punctate keratitis, small dendrites, subepithelial keratopathy, satellite lesions could be other variants [Figure - 6]. Radial keratoneuritis is very rare. Usually there is no hypopyon but it may be seen in about 20 percent of patients. The ulcer remains superficial for several weeks [Figure - 7]. Typically the disorder evolves over several weeks as a gradually worsening keratitis with periods of temporary remission. A higher index of suspicion is the key to diagnosing Acanthamoeba keratitis. Keratitis due to Herpes simplex, atypical Mycobacteria and fungi should be thought of in the differential diagnosis.
II. D. Viral Keratitis
Among the causative agents of viral keratitis, Herpes simplex virus (HSV) infection is the most important one as it often leads to blindness. a Among the two types of HSV, type I is more commonly associated with this condition. a Type II virus keratitis is found in 20 percent of infants born with HSV infection .  In adults the recurrence rates of Herpes simplex keratitis (HSK) are about 25 percent within 1 year and 33 percent within 2 years.  Various nonspecific stress factors, e.g., trauma, fever, menstruation, psychological stress, climatic changes are implicated as precipitating factors [Table - 1]. The type of treatment at the beginning of the disease has no apparent effect on recurrence. Recurrent HSV infection manifests in the following varied forms [Table - 2].
II. D. 1. Corneal Epithelial Lesions
II. D. 1.a. Dendritic Keratitis
The most frequent manifestation of herpetic ocular infection is the branching linear epithelial ulcer referred to as dendritic keratitis [Figure - 8]. Superficial herpetic infections of the cornea, appearing as first or recurrent attacks, are almost invariably accompanied by partial or complete loss of corneal sensation. Although this sensory loss can usually be detected within a day or two of the onset of the attack, it can be delayed for as long as 11 days. In antibiotic- or placebo-treated patients, most cases of dendritic keratitis last 7 to 14 days, although some may persist for 25 days or longer. Unfortunately, dendritic keratitis may usher in some of the more prolonged forms of the disease. Atypical and resistant forms of dendritic keratitis are seen in immunocompromised and AIDS victims.[ 4],,
II. D. 1.b. Geographic or Amoeboid Herpetic Ulcers
Occasionally a linear dendritic figure progresses to a broad area of epithelial involvement with irregular angulated borders ("geographic" or "amoeboid" ulcer). These lesions have a much longer clinical course, often of many months and often follow the injudicious use of topical corticosteroids for the treatment of dendritic keratitis.
II. D. 1.c. Marginal Herpetic Keratitis
Dendritic and geographic forms near the corneal limbus tend to run a longer course and to respond less readily to antiviral chemotherapy than other herpetic lesions. The marginal lesions- often have underlying corneal infiltration as well as epithelial staining, and thus are mistaken for bacterial or other marginal ("catarrhal") infiltrates. In India, marginal herpetic lesions are more frequently reported after penetrating keratoplasty and cataract surgery . 
II. D. 1.d. Indolent Keratitis
Necrotic stromal keratitis may be associated with relatively large epithelial defects or it may evolve from treated geographic ulcers. These indolent forms have a particularly prolonged course and usually have profound corneal anaesthesia, do not respond to therapy with topical antiviral drugs, and are not associated with infectious HSV. Along with the loss of the epithelium, there is marked swelling of the cornea, accompanied by folds in Descemet's membrane and marked discomfort. The term "trophic" ulcer is often applied to these indolent ulcers because of the marked loss of sensation.
In patients with severe localized stromal keratitis, necrosis of the deep tissues leads to loss of the overlying epithelium. These round or oval, relatively deep ulcers tend to have straight borders and were referred to by Gunderson as "metaherpetic ulcers." They are usually accompanied by marked corneal anaesthesia and run a prolonged course.
All these indolent ulcers bear some resemblance to recurrent epithelial erosions of the cornea in which there is a failure of the epithelium to attach to underlying basement membrane. Some of these indolent ulcers are complicated by microperforation. 
II. D. 2. Herpetic Stromal Keratitis
II. D. 2.a. Superficial Keratitis With Epithelial Lesions
It is not uncommon to see opacification of the anterior corneal stroma directly beneath the site of dendritic lesion. This superficial opacity has the same form as the ulcer and persists long after the epithelial lesion has healed, particularly in cases treated with antiviral agents.
II. D. 2.b. Disciform Keratitis
This is a central round (Disciform) lesion of the cornea, with opacity and swelling of the corneal stroma. It may follow an epithelial lesion immediately or may appear long after the original epithelial lesions have healed. Disciform keratitis due to HSV is associated with marked anaesthesia and with keratic precipitates immediately beneath the lesion on the corneal endothelium. It may also follow infections with vaccinia, herpes zoster, mumps and varicella but is most frequently associated with HSV ocular infections. It may run a course of a few weeks to several months. The milder cases tend to heal without sequel, but severe cases sometimes progress to permanent stromal scarring.
II. D. 2.c. Diffuse Stromal Keratitis
In patients with previous ocular herpetic infection, the deep layers of the corneal stroma may be diffusely affected, often without any typical epithelial herpetic lesions. This is most likely to occur after the use of corticosteroids, and the lesion may have a prolonged course.
II. D. 3. Endothelial Involvement
Herpes simplex endothelitis is a recently recognized clinical entity. Clinically endothelitis presents with mild stromal oedema, few medium sized keratic precipitates, aqueous flare and cells.  The condition is usually associated with secondary glaucoma due to trabeculitis. Herpes simplex virus antigens have been demonstrated in human endothelial cells in patient suffering from endothelitis, disciform keratitis and anterior uveitis.
II. D. 4. Herpetic Limbitis
This is characterized by the presence of localized inflammation of the deep corneal stroma at the limbus with an adjacent sector involved by scleritis. The corneal lesion often has a wide base at the limbus and narrows towards the centre of the cornea. These lesions may represent sclerokeratitis or may evolve from marginal epithelial herpetic lesion.
II. D. 5. Herpes Zoster Ophthalmicus (HZO)
It occurs due to activation of latent varicella zoster virus infection which persists after primary varicella infection. Corneal complications occur in 40 percent of cases of HZO. The most common findings are dendrites and punctate keratitis  [Figure - 9]. The dendritiform figures are not excavated, but are made of swollen, heaped cells and have a gray plaque-like appearance; unlike the delicate pattern of HSV dendrites, the zoster dendrite is more coarse, ropy and stellate, also the terminal bulbs seen in simplex dendrites are absent. They resolve without treatment within one month.
Other findings in cases with corneal involvement in HZO are punctate keratitis, disciform keratitis and mucous plaques.
| III. PATHOLOGY OF CORNEAL ULCER|| |
The pathology of corneal ulcer can be divided into three stages: (A) stage of infiltration (B) stage of ulceration (C) stage of healing.
III. A. Stage of Infiltration
Normally tear film and intact corneal epithelium act as effective barriers against invasion by organisms. However, this barrier can sometimes get disrupted, leading to invasion of organisms. Initially the organisms adhere to the defective epithelial surface. Exceptions to this rule are Neisseria gonorrhoeae, Corynebacterium diptheriae and Acanthamoebae which can invade the intact corneal epithelium directly. Invasion of the organism even in the presence of a damaged epithelium depends on the amount of inoculum and virulence of the organism. The most common corneal pathogens like Staphylococcus aureus, Staphylococcus epidermidis, or Psuedomonas aeruginosa are known to possess adhesiveness to a breached epithelium. The glycocalyx (the slimy envelope) assists in the adhesion of the organism to the epithelium. sub Scraping of the cornea at this stage is often helpful in demonstrating the pathogenic organism. Once the organisms gain entry, they proliferate in the epithelium and superficial stroma leading to swelling and necrosis which clinically cause a white or yellowish lesion in the cornea. Infiltration of acute inflammatory cells (mostly polymorphs) occurs following invasion of the organism in the tissue. Necrosis of the tissue occurs due to the toxins and enzymes liberated by the organism. Even after the microorganisms die or are killed, their residues release endotoxins which can perpetuate inflammation. Toxins liberated by most of the bacteria do not have collagenolytic activity except in the case of Pseudomonas which produces a protease which can destroy the collagen. Polymorphonuclear (PMN) leucocytes liberate several toxins causing tissue damage. Complement components (C3 & C5) help in chemotaxis of polymorphonuclear leucocytes, their adherence and activation. 
III. B. Stage of Ulceration
If the infection is not controlled, the inflammatory reaction progresses relentlessly leading to deeper stromal invasion and perpetuation of ulcer. There is sloughing of the epithelium and stroma leading to tissue loss thereby causing a crater. The margin of the crater is surrounded by oedematous corneal epithelium and the stroma is infiltrated by acute and chronic inflammatory cells. There can be an outpouring of acute inflammatory cells (hypopyon) into the anterior chamber. In the case of a bacterial corneal ulcer the hypopyon is usually sterile as the organism does not ordinarily penetrate an intact Descemet's membrane. However, fungi can often penetrate such a barrier and can be demonstrated in the hypopyon material. Progression of such ulcerative processes can lead to perforation of the cornea.
III. C. Stage of Repair
In this stage, both humoral and cellular immunity result in neutralization of the organisms by phagocytosing them as well as the cellular debris. The corneal epithelium grows over the crater of the ulcer from its margin. A leash of blood vessels as well as fibroblasts and macrophages encroach the subepithelial space, laying down collagen tissue, resulting in scar formation. Histopathologically, fibrocytes with characteristic deep staining nuclei and cytoplasm, with dense collagen fibres and absence of normal lamellar clefts indicate past inflammation. Since Bowman's membrane is not capable of regeneration, it is often found replaced by fibrous tissue. The stage of healing is often complicated by keratectasia (bulging forward of the Descemet's membrane and adherent leucoma (adherence of intraocular tissue, e.g., iris on the posterior surface of the cornea). Histopathologic changes in keratitis due to various organisms differ and are discussed below.
III. D. Histopathology of Corneal Ulcers III.
D. 1. Bacterial Corneal Ulcer
Common bacterial organisms which cause keratitis usually affect the central two-thirds of the cornea. The extent of damage by the organism depends on the virulence of the organism, e.g., Pseudomonas causes a rapid and nonreactive necrosis.  In 50 percent of cases, histopathological demonstration of the bacteria is possible from the specimen.  There is more likelihood of demonstration of the organism if it is removed in the early stage. Bacteria are seen as colonies and appear as clumps of basophilic material [Figure - 10]. Infiltration of acute and chronic inflammatory cells is usually seen within the stroma. In the case of crystalline keratopathy there is a characteristic absence of inflammatory cells. Grampositive cocci are seen as clumps by special stain or transmission electron microscopic study. The organisms identified are usually Streptococcus viridans or Peptostreptococcus. ,
III. D. 2. Fungal Corneal Ulcer
The pathologic study of fungal keratitis in contrast to bacterial keratitis exhibits less marked purulent inflammatory cellular reaction.  The peripheral cornea is rarely involved. The filamentous fungus is seen mostly lying parallel to the corneal stromal lamellae [Figure - 11]. They may be found in areas without inflammation and can be seen in the deeper stroma skipping the superficial portion. Inflammatory cells seen are often lymphocytes and plasma cells with variable degree of polymorphonuclear leucocytes. There is coagulative necrosis of the stroma resulting in stromal abscess. Satellite microabscesses are also seen with focal necrosis of corneal stroma with clusters of acute inflammatory cells. Healing of the epithelium is seen with the active proliferation of the fungus in the deeper stroma.  Corneal scrapings at this stage often fail to demonstrate the organisms. Corneal biopsy is recommended in cases of suspected fungal keratitis with negative results from scrapings. The destruction of the stroma is caused not only by the fungus but also by the toxins, enzymes and antigens liberated from it. An intact Descemet's membrane often prevents the spread of the fungus to the anterior chamber. However, some fungi, especially Fusarium solani can penetrate such a barrier and can be seen in the anterior chamber. In the tissue specimens (corneal biopsy, corneal button) fungus is well demonstrated by PAS, GMS stain or Gram's stain. One can use fluorescent dyes like calcofluor white or fluorescein conjugated lectins. Selective screening of several fungal species has been demonstrated when a panel of commercially available lectins were used. 
III. D. 3. Acanthamoeba Keratitis
This organism has been demonstrated in both contact lens wearers and others , The diagnosis is made from corneal scrapings in initial stages, but in some cases a corneal biopsy is needed. The organism can be seen in haematoxylin-eosin stain, but special stains like PAS, GMS, calcoflour  and fluorescein conjugated lectins  can also help in their identification .  Ultrastructural studies reveal that the cyst measured 10 to 25 urn in size and has a characteristic double-layered cell wall, outer wall of which is wrinkled (ectocyst) and the inner wall is round, polygonal or stellate (endocyst) [Figure - 12]. The nucleus is large with centrally located densely staining nucleolus. Trophozoites are difficult to identify by light microscopy alone, as they are of irregular outline, their sizes and shapes vary and they resemble histiocytes and keratocytes. The surrounding corneal tissue shows variable degree of inflammatory cellular infiltration, comprising of neutrophils and lymphocytes. The epithelium is often absent or detached. Immunohistochemical studies done in two cases of Acanthamoeba keratitis showed majority of inflammatory cells to be neutrophils and HLA-DR positive macrophages 
III. D. 4. Viral Keratitis
In epithelial disease caused by Herpes simplex virus, active viral replication occurs in the ulcer margin.  In the early stages, polymorphonuclear leucocytes form the main inflammatory reaction later replaced by lymphocytes, macrophages and multinucleated giant cells with intranuclear acidophilic inclusions. The titre of virus is highest before the development of dendritic lesion and progressively falls as the lesion progresses .  Stromal involvement occurs in recurrent attacks and is often referred to as disciform keratitis. This is an immune-mediated disease as evidenced by the presence of empty viral capsids and incomplete virions in stromal cells, activity of lymphokines and a favourable outcome of steroid therapy. , These immunopathologic lesions have also been demonstrated in experimental animals and have been shown to be due to CD4+ T lymphocytes.  Recurrent HSV keratitis occurs when latent virus in trigeminal ganglion is reactivated, replicates and travels along the trigeminal nerve to the eye.
Latency of HSV in cornea is proposed. Demonstration of HSV-DNA in human cornea without any apparent active disease has been made using polymerase chain reaction. Recovery of HSV from human cornea by co-cultivation techniques has also been reported.  Latency of HSV in cornea of rabbit  and mice  has been experimentally demonstrated. Further studies are needed for final acceptance of HSV in cornea.
Adenovirus is the most common cause of acute viral keratitis. Several serotypes (frequently 8, 19 and 37 and infrequently 3, 4, 7, 10, 11, 21) cause epidemic keratoconjunctivitis. Viral replication occurs in the corneal epithelial cells and later with the development of delayed hypersensitivity, subepithelial infiltration appear which persists for an extended period of time.
| IV. MICROBIOLOGICAL DIAGNOSIS OF INFECTIOUS KERATITIS|| |
Owing to the considerable overlap in the clinical appearances of corneal ulcers due to various micro organisms, a standard basic laboratory methodology should encompass techniques that allow for the recognition of as large a number of offending organisms as possible. There are now enough studies available to justify a mandatory decision to perform proper laboratory studies in every patient seen with corneal ulcer or infiltrate , [Figure - 13] depicts a procedural flowchart to isolate the causative organism.
Collection and Processing of Clinical Samples
IV. A.1. Conjunctival and Lid Swabs
It has generally been recommended to collect conjunctival and lid cultures from the ipsilateral and contralateral eyes before the corneal material is obtained.  The utility of these cultures is, however, controversial. In a recent study, designed to determine if the organism isolated from conjunctiva reflects the causative organism of infectious keratitis, it was found that a correlation existed only in 3 percent of cases.  Based on this, it is our feeling that microbiologic evaluation of scrapings from the ulcer is mandatory.
IV. A.2. Corneal Scrapings
Kimura's spatula is traditionally used to collect scrapings from a corneal ulcer though Bard-Parker blade no. 15 is equally popular. Cotton swabs are not recommended for collection of corneal material.  Multiple scraping of the ulcer bed and margins is done under topical anaesthesia (0.5% proparcaine hydrochloride or 4% lignocaine hydrochloride) with the aid of a slit-lamp or operating microscope after removal of debris or discharge in the vicinity. Several scrapings are collected and used in a sequence to prepare smears and inoculate culture media [Table - 3]. The blade or spatula may be reused when a sterile medium has been streaked. However, a blade must be changed (spatula should be flamed) when a smear has been made on a slide which may not be sterile.
Solid agar media are inoculated on the surface making multiple "C" shaped marks without cutting the agar. In the liquid media the spatula or blade is swirled to allow the sample to be transferred. In the case of thioglycolate broth deep inoculation of the medium is ensured by transferring the sample to a swab tip and dropping the swab in the tube allowing it to settle at the bottom. The incubation period and conditions of growth for various culture media are presented in [Table - 4].
Techniques of Gram stain, Kinyoun stain (Cold carbol Fuchsin) or Giemsa stain are employed to study the corneal material which is spread as a thin layer on several clean glass slides within an area defined with a wax pencil on the reverse. In the preparation of wet mounts such as potassium hydroxide, lactophenol cotton blue or calcoflour white, the scrapings can be placed on the slide in a demarcated area and covered with a drop of the solution followed by a coverslip. Special stains and media may be included whenever usual procedures have yielded negative results.
IV. A.3. Corneal Biopsy
If infectious keratitis is suspected clinically and twice repeated microscopic evaluation of smears and culture results are negative and no clinical improvement is noted on the initial broad spectrum antibiotic therapy, we recommend corneal biopsy. A partial-thickness trephination employing a trephine of sufficient size, to guarantee adequate material for the laboratory, is required. Harvesting the material for diagnosis should include an adequate area of cornea affected clinically by the inflammatory or ulcerative process. Corneal scrapings for microbiological evaluation from the site of biopsy may help in isolating the organism.
The biopsied tissue is preferably removed enbloc. It is bisected, half being sent to microbiology laboratory for homogenization and culture, and the remaining half placed in 10% buffered formalin to be transported to a pathology laboratory.
IV. A.4. Anterior Chamber Paracentesis
This procedure is rarely indicated in the diagnostic evaluation of a patient with a corneal ulcer. However, this procedure may be indicated in the instance where keratomycosis is strongly suspected clinically, yet corneal scrapings and biopsy have been negative, the damage to the cornea is progressive and a hypopyon is present or increasing.
IV. B. Interpretation of Smears and Cultures IV. B.1. Smears
These methods yield a rapid result and form the basis for a provisional diagnosis and therefore determine the initial choice of an antimicrobial agent to be instituted for therapy.
IV. B.1.a. Potassium Hydroxide (KOH) Wet Preparation
A 10 to 20% solution of KOH has been used to visualize fungal elements in corneal scrapes.  Though later reports sub have discounted the value of KOH mount it has been found to be an useful diagnostic aid for both fungi and Acanthamoeba. sub Owing to the chitin in their cell wall, fungal filaments [Figure - 14] and cysts of Acanthamoeba are clearly delineated in a homogenous background of corneal tissue digested by KOH. Addition of 10% glycerol to the KOH acts as a mordant and serves to preserve the smear for as long as six months.
IV. B.1.b. Gram Stain
The Gram stain is utilized to identify bacteria, fungi as well as Acanthamoeba. It has been reported to yield an accuracy of 60 to 75 percent in identifying the responsible organisms.  However, one should be aware that indigenous bacterial flora in the tear film can occasionally be detected in the corneal smear.
Fungal filaments exhibit variability in their staining pattern in Gram stain with the cell wall and septae remaining unstained and only the protoplasm being stained or an empty protoplasm with faint outline of fungus. Yeasts, on the other hand, stain typically blue.
The Giemsa technique renders all bacteria dark blue. Acanthamoeba cyst wall stains dark blue, and the cytoplasm stains blue. The value of the Giemsa stain for distinguishing infectious from non-infectious keratitis by type of inflammatory cells has not been established.
IV. B.1.c. Calcofluor White (CFW) Stain
CFW is a fluorescent brightener with great affinity for certain polysaccharides such as cellulose and chitin, thus providing the basis for demonstration of fungal cell walls as well as cysts of Acanthamoeba spe cies. sub The preparation is viewed under fluorescence microscope using exciter and barrier filters. The cysts of Acanthamoeba and fungal filaments appear bright apple green in a corneal scraping stained with CFW. Acanthamoeba trophozoites and bacteria such as Nocardia and Actinomyces do not stain with CFW . 
IV. B.1.d. Other Stains
A modification of Gomori methenamine silver stain may be helpful for the identification of fungal elements and Acanthamoeba cysts in corneal scrapings. Fungi and Acanthamoeba cysts stain black on a light green background.
The periodic acid-schiff (PAS) stain may also be used to visualize fungal elements as well as Acanthamoeba especially in tissue sections.
Bacteria, fungi as well as Acanthamoeba have been demonstrated in smears and tissues using lectins conjugated with fluorescein or peroxidase. 
Nonspecific fluorescent stains like blankophor and uvitex 2B have been introduced for the rapid detection of fungal elements in corneal scrapings.
Lactophenol cotton blue stain, which is generally used for the microscopic examination of fungal cultures, has been effectively used for the demonstration of fungal elements and Acanthamoeba cysts in corneal scrapings . , Ziehl-Neelsen stain or its modification (Kinyoun's stain) are indicated for the detection of Mycobacteria and Nocardia species, respectively.
IV. B.2. Cultures
A daily monitoring of all culture media is essential, coupled with standard procedures for description of colony morphology, selection of colonies for processing and antibiotic susceptibility testing. The use of solid media has definite advantages. Growth away from a streak is assumed to be a contaminant. A rough quantitation is also possible on a solid medium.
The amount of growth depends on many factors. Ophthalmologists can vary in their enthusiasm when collecting material to inoculate plates. Occasionally, many bacteria may be seen in a smear of the scraping, but very little is grown on culture. This may be due to previous antimicrobial treatment or most of the infected corneal tissue being used for preparing the smears.
An isolate is more likely to be considered significant if:
1. it is consistent with the clinical signs
2. smear results are consistent with culture
3. the same organism is grown on more than one media
4. the same organism is grown from repeated scrapings
Identification of bacteria may be accomplished within 48 hours along with its antibiotic susceptibility pattern. Standardized disk diffusion or dilution techniques should be utilized for antibiotic susceptibility testing of bacteria. One should be aware, however, that results of disk diffusion susceptibility tests relates to levels of drug achievable in serum and do not relate directly to concentration of drug produced in the preocular tear film and ocular tissues.
Bacterial contamination of fungus cultivating media (e.g., Sabourand's dextrose agar, potato dextrose agar) can be avoided by incorporating chloramphenicol in a concentration of 50 ug/ml. Cycloheximide, an inhibitor of saprophytic fungal growth, is a frequent component of some fungal media. It should not be added to SDA used in ocular microbiology laboratory since most cases are due to so-called saprophytes. The majority of fungi causing keratitis can be detected on SDA within 72 hours. Aspergillus and Fusarium species grow on blood agar, SDA and brain-heart infusion broth within 48 hours. However, appreciably characterisitic colonies develop after 1 to 2 weeks. Culture media should be observed for at least 2 weeks before they are considered negative.  It is not unusual to come across strains of fungus that do not show spores and thus are unidentifiable. Such cultures should be sent to reference centres for identification. The value of antifungal susceptibility testing for treatment of mycotic keratitis is not yet established.
Non-nutrient agar (NNA) is the standard medium used with an overlay of Escherichia coli for the growth of Acanthamoeba The specimen is simply touched to the surface of the plate without streaking or breaking the surface. Two plates may be inoculated for incubation at 25 and 37 o C since some species do not grow at the higher temperature  and the plates are examined for trophozoites and cysts directly under the microscope (100 X). Trophozoites may be seen in 24 to 48 hours. They move and cover the entire plate surface on further incubation and turn into cysts. The plates should be observed for at least 7 days.
Special media, selective and non-selective, may be indicated in certain clinical situations. Lowenstein-Jensen medium is used when mycobacterial infection is suspected. Nocardia organisms can grow, though slowly, on blood agar as well as other bacterial media.
IV. C. Diagnosis of Contact Lens-associated Keratitis
A variety of organisms have been reported from keratitis associated with contact lens wear, most notable of them being Pseudomonas aeruginosa and Acanthamoeba species. Contact lenses, lens cases and lens solutions should be collected for culture apart from the corneal scrapings. Contact lenses, if present on the eye, should be removed aseptically and placed in sterile saline or buffered saline and sent to the laboratory, where they can be cultured by agar-sandwich method .  Fluid from the lens cases can be cultured on standard media such as blood agar, MacConkey agar, NNA and SDA. Microscopy of the lens deposit, centrifuged deposit of the lens care solutions may help detect the offending organism.
IV. D. Newer Methods in the Diagnosis of Infectious Keratitis
The need for rapid diagnosis has led to modification of various conventional techniques and introduction of new techniques such as immunohistochemistry, fluorescent microscopy, enzyme immunoassays, radioimmunossay and molecular biology. Most ocular infections can now be diagnosed by these modern techniques within 1 to 6 hours. 
Most immunoassays are based on the availability of specific antisera against infectious agents. Highly specific antibodies have been made available by hybridoma technology. Such monoclonal antibodies have been used in the diagnosis of viral, chlamydial and bacterial infections.
Recent introduction of nucleic acid hybridisation technique has revolutionized the field of diagnostic microbiology. This technique utilizes the methods of molecular biology and is now being marketed for various diseases as diagnostic test kits. They are highly sensitive and specific. Some of these test kits use non-radioactive DNA probes and are easy to perform. This method not only detects the species of microorganisms but also the strain of the organism, thus providing information on antibiotic susceptibility of the organism. Presently, probes are available for limited organisms such as viruses, Chlamydiae and Mycobacteria. The availability of probes is increasing with time and this technique is going to have a tremendous impact on the rapid diagnosis of infectious keratitis and many other ocular diseases.
IV. E. Laboratory Diagnosis of Viral Keratitis
Basic approaches to laboratory diagnosis of viral keratitis are proper collection and transport of clinical specimens (corneal scrapings), rapid methods for detection of viral antigens or viral products in the clinical material with isolation and identification of the virus. Direct demonstration of viral antigens or viral products are possible by Giemsa stain cytology, electron microscopy, immunodiagnostic methods such as enzyme immune assay (EIA) or immunofluorescence (IF) and nucleic acid hybridisation. Scraped material in adequate amount is placed in one ml of Hank's balanced salt solution (HBSS) containing 3% foetal calf serum (FCS) and material collected later is used for preparing 4 to 5 smears on clean microscopic slides. If conjunctivitis co-exists, as in adenovirus infections, smears of conjunctival scrapings and conjunctival swab in HBSS with 3% FCS are collected. Smears are fixed in cold acetone or methyl alcohol. The specimens in HBSS are transported in an ice chest to the laboratory.
IV. E.1. Methods for Detection of HSV and Adenoviruses
Rapid presumptive diagnosis of the viral aetiology is possible by Giemsa stain cytology on the smears. Cytopathologic changes such as syncytial giant cells with Cowdry type A intranuclear acidophilic inclusions in HSV and VZV infections and intranuclear inclusions in adenovirus infections are demonstrable. Giemsa stain cytology is an useful screening procedure to arrive at an aetiological diagnosis taking into account the type of inflammatory cells seen in the smear . 
With the availability of specific monoclonal and polyclonal antibodies against viruses and their antigens, immunodiagnostic methods allow detection of virus in the direct smears. Enzyme Immune Assay (EIA) and immunofluorescence are commonly used for examination of smears and they are rapid, specific and sensitive. Monoclonal antibodies to differentiate HSV type 1 and type 2 have been developed. 
Exudates collected in buffers and smears are used for EIA. Several EIA formats have been described for detection of adenoviruses which can be applied for detection of this virus in specimens from cornea. Immunofluorescence (IF) has gained wide acceptance as a highly sensitive, specific and rapid diagnostic procedure for detection of viral antigen in direct smears from clinical material. For detection of HSV in corneal ulcer, IF is found to be highly reliable but results have to be considered with caution when patient is on antiviral therapy.  IF on direct smears has been found to be more sensitive than culture for detection of HSV in corneal ulcers particularly in those which are on treatment with acyclovir. This is likely to be due to the presence of HSV antigens rather than replicating virus in the lesions.
Electron microscopy is not a common diagnostic procedure for rapid detection of viruses causing keratitis. Polymerase chain reaction (PCR) for detection of HSV has recently been used by several workers. sub Kowalski et al found that PCR was not significantly more sensitive than slit-lamp examination. Hayashi et a1  found that PCR amplification technique provided a quick, sensitive method for the detection of the specific herpes virus in keratitis of unknown aetiology.
The most sensitive and specific method for diagnosis of HSV and adenovirus infection is the conventional method of isolation and identification of viruses in cell cultures, which usually requires upto 2 to 4 weeks. This is more sensitive than direct methods described above for their detection because the virus content is amplified by growth in susceptible host system such as cell cultures [Figure - 15]. Both primary and established cell lines are extensively used for isolation and identification of HSV and adenoviruses.
Application of immune assays such as EIA, IF and immune electron microscopy (IEM) has made a significant advance in rapid identification of isolated viruses though neutralisation tests may be used for final typing of the isolate. PCR can also be applied to the virus isolate in cell cultures for rapid identification. An isolate can be specifically identified in 48 to 96 hours by shell vial culture method which has been widely applied for detection of HSV  and adenoviruses. 
IV. E.2. Serological Methods of Diagnosis
The traditional serological methods for diagnosis HSV infection are generally of little value, since most viral infections of cornea are secondary to a primary infection elsewhere in the body. For diagnosis of adenovirus infection, neutralisation test, complement fixation test and ELISA test are useful serological tests.
IV. E.3. Other Viruses
Mumps virus keratitis is almost always one of the ocular complications of the systemic disease. Virus antigen can be detected in conjunctival epithelial cells by immunofluorescence.  Chick embryo and chick embryo fibroblast cell line are used for primary isolation of the virus. Being a generalised disease, serological responses to mumps and measles viruses are excellent for demonstration of four-fold rise in virus specific antibody titre in paired sera of the patients.
| V. MEDICAL MANAGEMENT OF INFECTIOUS KERATITIS|| |
V. A. Bacterial Keratitis
Once a provisional diagnosis of bacterial keratitis is made, one should institute therapy with appropriate antibiotics. The selection of antibiotics can be in the form of specific agents  or a combination therapy. , This latter strategy aims at using antibiotics that can be effective against both gram-positive and gramnegative bacteria. One guideline that may be helpful is the epidemiologic information on the microbiologic basis for infectious keratitis in a given geographic region.
Amongst the currently available antibiotics the most commonly employed is a combination of a cephalosporin with an aminoglycoside. The critical factor to enhance the efficacy of these drugs are the concentration of the drug and frequency of application. These drugs have to be made as fortified preparations and applied as frequently as every 30 minutes to 1 hour. The details of the actual drug regimen that we recommend are given in [Table - 5]. Alternatively, the newer fluoroquinolone drugs are often employed; the most popular being ciprofloxacillin. The major advantages claimed for its usage include ready availability as a commercial preparation, efficacy against broad spectrum of bacteria and relatively low toxicity and resistance.
V.A.1. Drug Delivery
At the concentration and frequency recommended, the antibiotics used topically reach the cornea in adequate concentration, and so other routes of administration are rarely required. The usage of painful subconjuctival injections is almost eliminated except in situations where administration of drugs cannot be assured round-the-clock.
Systemic antibiotics are indicated only if there is a perforation and/or if there is any indication of endophthalmitis.
A grading of the severity of the keratitis is an useful adjunct in estimating the urgency of treatment and risk of perforation [Table - 6]. In a mild infection standard strength antibiotics used topically are probably as effective as the fortified versions.
V. A.2. Assessment of Progress
A careful slit-lamp examination at each visit [Table - 7] is an essential feature. Early signs suggesting an improvement are:
1. increased comfort
2. decreased discharge
3. reduced "fluffiness" of the infiltrate
4. blunting of the edges of the ulcer
5. reduced oedema in adjacent stroma
6. reduced anterior chamber inflammation
Most often a lack of progression in the first 24 to 48 hours is an indication of effective treatment. In some cases, especially Pseudomonas infections, early progression can occur inspite of appropriate antibiotic therapy.
The initial therapy should be reviewed at the end of 24 to 48 hours based on the progression of the keratitis and culture results. If there is an improvement, the initial therapy should be continued regardless of the culture reports. This is so because it often happens that the organisms may not be sensitive to an agent in vitro but are sensitive at the high stromal levels achieved with topical administration of fortified antibiotics. If a combination of drugs has been commenced and the organism is sensitive to only one of them, the second drug may be discontinued or replaced with a more appropriate one. The whole plan can be designed as shown in [Figure - 16].
Infections with gram-negative bacteria, especially Pseudomonas can lead to formation of ring-shaped infiltrates .  ]These appear to be an immunologic reaction to bacterial endotoxin, and respond to corticosteriods. They are usually seen about 10 to 14 days after the onset of the infection and it is critical to differentiate them from recurrence or worsening of the infection. 
V. A.3. Altering the Frequency of Drug Administration
It is a good policy to commence the antibiotic therapy in the dose of 1 drop every minute for 5 minutes and then every half hourly. This helps in achieving high concentrations of the antibiotic drops in the corneal stroma. The tapering of the antibiotics is based on the response in each individual case. Duration and severity of infection, depth of invasion and causative bacteria are some of the determining factors. Epithelialization alone is not a good criterion to suggest resolution of infection as any of the antibiotics may deter epithelial healing due to toxic or allergic reactions.
V. B. Fungal Keratitis
The antimicrobial agents available today to combat fungal keratitis are not as well developed as those available against bacterial infections. All the available agents only inhibit growth of the fungus, the host defence mechanisms must eradicate the infection. Most of these drugs are irritant and toxic to the ocular surface and have a limited penetration of the cornea. The main drugs in use currently are: (1) amphotericin B, (2) natamycin, (3) imidazole compounds (ketoconazole, miconazole, clotrimazole), (4) flucytosine.
In the warmer countries like India it is the filamentous fungi which are a major source of infection. The mainstay of therapy have been the polyene group (amphotericin B, natamycin and nystatin).
Natamycin 5% suspension is the drug of choice for filamentous fungi.  It alters the permeability of the fungal cell elements, is unstable, but less toxic than amphotericin B.  A 5% solution of natamycin when applied topically usually "adheres" to the ulcer bed and forms a rope like strand in the inferior fornix which may serve as a reservoir. Therapy with natamycin should be initiated as 1 drop every half hourly for the first 3 to 4 days and then decreased to 6 to 8 times daily .  Like all other antifungal drugs it penetrates the deepithelialized cornea well. Unlike amphotericin B which cannot penetrate the intact epithelium, natamycin can penetrate intact epithelium,to a small extent . 
Imidazole compounds available to us are ketoconazole and clotrimazole. Ketoconazole as a 1-2% solution is well tolerated as is systemic administration. An oral dosage of 200 to 600 mg per day may be given. It has been shown to be an effective agent in limited trials. It is active against most yeasts and many filamentous fungi. Systemic usage requires careful monitoring of liver status.
Clotrimazole is most significant in treatment of Aspergillus group of fungi. Topical 1% concentration in arachis oil or the dermatologic cream are well tolerated.
Antifungal therapy should not be initiated without laboratory evidence of fungal keratitis since clinical history and appearance alone are not diagnostic, therapy is prolonged, response is slow and agents are often toxic.
The selection of antifungal agents is based mainly on clinical response. There are no standardized sensitivity determinants for antifungal drugs. If the patient is doing well on a drug, treatment should not be changed unless toxicity to the drugs is seen. Improvement is often not visible till weeks after initiating treatment. One should look for reduced size of the central corneal infiltrate, disappearance of the satellite lesions and rounding off of the feathery margins. The conjunctiva usually reacts to the antifungal agents and shows chemosis and injection. The presence or absence of these features should not be used as an indication of the success of therapy. Persistent epithelial staining may be noticed. Similar to the conjunctival reaction, this too, often indicates toxicity of the medication.
The duration of treatment should be long enough to allow the body defence mechanisms to eradicate the organisms. Long-term therapy of at least 6 weeks is usual. Negative cultures are not a guide to eradication of the fungus. A close watch after discontinuing treatment is essential to look for any signs of recurrence. An important point usually not highlighted but which needs to be stressed upon is the need to scrape the corneal epithelium to within 2 to 3 mm of the limbus everyday, especially in case of stromal fungal infiltrates. This not only acts as a debulking procedure by removing the necrotic tissue but more importantly it enhances the penetration of the antifungal drugs . 
V. C. Acanthamoeba Keratitis
The current approach to treatment of Acanthamoeba keratitis lacks consistent efficacy. Some cases may progress despite medical therapy. Recurrence of infection in grafts after penetrating keratoplasty has been demonstrated .
A host of drugs have been cited as being useful in treatment of Acanthamoeba keratitis. The most extensive experience has been with propamidine isethionate and some medical -cures have been reported in combination with neomycin - polymyxin B - gramicidin (neosporin) drops. Propamidine is freely available in the United Kingdom. The recommended treatment regime is outlined in [Table - 8]. The duration of treatment should be up to one year in most cases. In the case of lack of response or stromal involvement 1% clotrimazole may be added .
Recent additions to the armamentarium are benzethonium chloride and polyhexamethylene biguanide.  Other drugs which may be useful are: ketoconazole, miconazole, paromomycin, natamycin, amphotericin B, polymyxin B and metronidazole.
V. D. Viral Diseases
We shall restrict our discussion to the treatment of Herpes simplex (HSV) and Herpes zoster infections.
V. D.1. Herpes Simplex Virus (HSV) Keratitis
Of the various agents used to treat HSV keratitis (idoxuridine, trifluorothymidine, adenine arabinoside, acyclovir) only idoxuridine and acyclovir are available locally. Idoxuridine was the first agent to be made available as an antiviral agent. It is used as a 0.1% solution and 0.5% ointment and is useful primarily in corneal epithelial disease due to poor topical penetration. The use of this drug as a first line drug has declined considerably due to its high incidence of toxicity [Table - 9], need for higher frequency of instillation and the availability of safer drugs like acyclovir. Acyclovir (acycloguanosine) has specific activity against HSV types I and II. After topical application it penetrates the stroma and reaches the aqueous. It is least toxic to the ocular surface among the currently available drugs. One of the drawbacks with this drug is its greater potential for developing resistant strains. Oral acyclovir though effective even in epithelial keratitis has a more useful role in necrotizing stromal disease, endothelitis and keratouveitis.  There is some evidence to suggest that it reduces the risk of recurrence .
V. D.1.a. Specific Treatment
V. D.1.a.i. Recurrent Epithelial Keratitis Without Stromal Disease
Debridement may be equal to or superior to antiviral treatment for epithelial herpes.  A combi-nation of debridement and antiviral treatment is more effective than antiviral treatment alone.
V. D.1.a.ii. Limbitis
This is usually recalcitrant to treatment. Artificial tears are probably as effective as any other therapy. Corticosteroids should be used only if visual axis is involved.
V. D.1.a.iii. Disciform Keratitis
Some prefer to adopt conservative management with topical cycloplegics. The usage of corticosteroids is deferred since these may prolong the disease course without having any effect on the final outcome. The other school of thought believes that use of steroids offer both short- and long-term visual benefits. We recommend the use of steroids when vision is significantly reduced or pain is a significant concern. It must be remembered however, that once steroids are used, control of recurrence may not be possible without further recourse to steroids, thus very dilute doses may be sufficient. Treatment may be initiated at two to four times daily and tapered to as low a 1:200 dilution daily or 1:125 dilution to be used once a week.  Concurrent antiviral shedding or reinfection may occur.
V. D.1.a.iv. Persistent Epithelial Defects With or Without Stromal Ulceration
One should first determine the exact cause of persistent epithelial defect. These may be due to:
(1) Recurrence, viral resistance or poor compliance with therapy
(2) Antiviral agent toxicity
(3) Persistent anterior stromal inflammation
(4) Impaired epithelial healing due to recurrent attacks (5) Associated bacterial/fungal infections
The dose of antiviral drugs should be reduced if no active disease exists. In the absence of bacterial infection one should discontinue the antibiotics. An attempt should be made to look for any associated adnexal problems like lagophthalmos and trichiasis. Use of topical corticosteroids under careful supervision may help reduce stromal inflammation. One should be cognizant of the potential for stromal degradation and perforation. Lubricating agents, patching or a bandage contact lens facilitates reepithelialization. A temporary tarsorraphy or conjunctival flap may be needed.
Necrotising Stromal Keratitis (Stromal Ulcer With Infiltration): One should always look for and rule out secondary infection. After several days of antiviral therapy, topical steroids can be cautiously added initially in low doses. Oral prednisolone (0.5 - 1 mg/ kg) is preferable to topical steroids till the epithelium has healed.
V. D.1.a.v. Endothelitis
Active infection is a possibility in these cases. Topical acyclovir is able to achieve an adequate drug level in the aqueous. Topical steroids can be used in addition, if the epithelium is intact. Both, acyclovir and steroids should be tapered very slowly.
No treatment can prevent latency or eliminate latent infection. However, early treatment of recurrence seem to have beneficial effect on the severity and duration of attack.
V. D.2. Herpes Zoster Ophthalmicus
The treatment for the various forms of this disease are summarised in [Table - 10]. The management of pain in herpes zoster can be a serious problem. The administration of levodopa (100 mg tds)  and cimetidine (300 to 400 mg daily x 14 days) sub have been shown to be effective in relieving pain. Other agents have not been very successful. Intractable pain may need stellate ganglion block. 
| VI. SURGICAL MANAGEMENT|| |
The decision to surgically intervene in a case of active infectious keratitis should be made after proper appraisal of the clinical progress. The role of surgery may be diverse and as follows:
1. Aid in medical management
(a) by increasing drug penetration
(b) by bringing in blood vessels in the form of conjunctival flaps
2. Stabilise the corneal epithelial surface by conjunctival flaps
3. Excise the infected corneal tissue and eliminate or reduce the microbial load
4. Tectonically support the globe where the integrity is threatened as in cases of thinning or perforation of the cornea.
The various modalities of treatment available in such cases are:
A. Removal of epithelium and anterior lamellar keratectomy
B. Conjunctival flaps
C. Tissue adhesives
D. Penetrating keratoplasty
VI. A. Epithelial Removal and Anterior Lamellar Keratectomy
This modality of treatment is useful particularly in cases of fungal keratitis. Regular debridement of the base of the ulcer helps in elimination of organisms and necrotic material. This procedure facilitates penetration of antifungal drugs.  This can be done under topical anaesthesia leaving a margin of 1 to 2 mm at the limbus with a no.15 Bard-Parker blade.
Anterior lamellar keratectomy helps in removal of the thick mat of fungal filaments on the cornea and facilitates increased drug penetration in cases of dematiaceous fungal filaments. Anterior stromal corneal infiltrates can also be ablated with excimer laser for therapeutic purposes . 
VI. B. Conjunctival Flaps
Conjunctival flaps help in achieving a stable epithelial surface in cases of persistent or recurrent epithelial defects and progressive ulceration especially in viral keratitis.
These are particularly helpful in chronic peripheral disease where the flap does not encroach onto the visual axis .  In peripheral fungal corneal ulcers, the blood vessels brought in by conjunctival flaps help in healing of the ulcer. A superficial lamellar keratectomy with removal of necrotic stroma is to be done with anchoring of a thin conjunctival flap over the ulcerated site [Figure - 17].
VI. C. Tissue Adhesives
Tissue adhesive (cyanoacrylates) helps in supporting corneal thinning and sealing corneal perforation upto 2mm. , Cyanoacrylate adhesive is bacteriostatic for gram-positive bacteria .  sub Necrotic stroma or epithelium and other debris must be removed from the base of the ulcer before the adhesive is applied. Usually a bandage contact lens must be fitted after the application. The adhesive is left in place until it loosens spontaneously, or the bed becomes vascularized [Figure - 18], [Figure - 19] or keratoplasty is performed. This modality of treatment has a valuable role in the management of infectious keratitis.
VI. D. Penetrating Keratoplasty
The indications for penetrating keratoplasty are (a) perforation, (b) descemetocele or impending perforation [Figure - 20][Figure - 21], (c) continued progression despite maximal medical treatment, (d) post-infectious corneal scar. The results of keratoplasty in acutely infected or inflamed eyes are relatively poor, the risk of rejection and glaucoma greater especially in larger grafts. 
In all these cases at least 0.5 mm of clear tissue all around the infected area is to be excised to decrease the incidence of recurrence. Postoperative antimicrobial treatment is to be continued. In fungal keratitis, postoperative topical steroids are to be used with caution. Surgery when performed with 8 mm or smaller diameter donor grafts had better results than larger grafts.  Hence, penetrating keratoplasty is to be considered early when fungal ulcers do not respond to antifungal medication. The results of penetrating keratoplasty for Acanthamoeba keratitis are poor and surgery is to be considered only in patients with gross corneal thinning or perforation.
| VII. CONCLUSION|| |
While infectious keratitis continues to have the potential for high rate of ocular morbidity, advances in diagnosis and treatment have changed the outlook for the management of this problem. Rapid methods of isolation of causative organisms, newer drugs with greater potency and specificity of action delivered to ensure appropriate dosage constantly make the prognosis better for these cases. One could be optimistic that the role of surgery will be relegated to only visual rehabilitation of central corneal scars.
| References|| |
McDonnel PJ, Nobe J, Gauderman WJ, et al. Community care of corneal ulcers. Am J Ophthalmol 114:531538, 1992.
a. Pavan-Langston D. Viral diseases: Herpetic infection. In: Smolin F, Thoft RA (eds). The Cornea: Scientific Foundations and Clinical Practice (2nd ed). Boston, Little Brown & Co., pp. 240-266, 1987.
Nahmias AJ, Hagler W. Ocular manifestations of herpes simplex in the newborn. Int Ophthalmol Clin 95:1798-1799, 1972.
Shuster JJ, Kaufman HE, Nesburn AB. Statistical analysis of rate of recurrence of herpes virus ocular epithelial disease. Am J Ophthalmol 91:328-331, 1981.
Young TL, Robin JB, Holland GN. Herpes simplex, keratitis in patients with acquired immunodeficiency syndrome. Ophthalmology 96:1476-1479, 1989.
Kremer I, Wagner A, Shmuel D, et al. Herpes simplex keratitis in renal transplant patients. Br J Ophthalmol 75:94-96, 1991.
Brik D, Dunkel E, Pavan-Langston D. Herpetic keratitis: Persistence of viral particles despite topical and systemic antiviral therapy. Arch Ophthalmol 111:522-527, 1993.
Saini JS, Sharma A. Post-surgery herpetic keratitis. Proc All India Ophthalmol Soc 1992.
Saini JS, Sharma A, Grewal SPS. Chronic cornea perforations. Ophthalmic Surg 23:394-402, 1992.
Holbach LM, Font RI, Neumann G. Herpes simplex stromal and endothelial keratitis. Ophthalmology 97:722-728,
Leisegang TJ. Corneal complications of herpes zoster ophthalmicus. Ophthalmology 92:316, 1985.
Costerton JW, Brown MRW, Sturgess JM. The cell envelope. In: Doggett RG (ed), Pseudomonas aeruginosa:
Clinical manifestations of infection and current therapy. New York, Academic, 1979.
Mondino BJ, Brown SI, Robin BS. Role of complement in corneal inflammation. Trans Ophthalmol Soc UK 98:363, 1978.
Hyndiuk RA, et al. Experimental Pseudomonas
Keratitis Clinical and pathological observations. Cornea 2:103, 1983.
Jones DB. A plan for antimicrobial therapy. Trans Am Acad Ophthalmol Otol 79:95, 1975.
Samples JB, Baumgartner SD, Binder PS. Infectious crystalline keratopathy - An electron microscope analysis. Cornea 4:118-126, 1985/1986.
Eiferman RA, Ogden LL, Snyder J. Anaerobic peptostreptococcal keratitis. Am J Ophthalmol 100:335-336, 1985.
Jones DB. Pathogenesis of bacterial and fungal keratitis. Trans Ophthalmol Soc UK 98:367, 1978.
Naumann G, Green WR, Zimmerman LE. Mycotic keratitis. A histo-pathologic study of 73 cases. Am J Ophthalmol 64:668, 1967.
Robin JB, Arffa RC, Avni I, et al. Rapid visualization of three common fungi using fluorescein-conjugated lectins. Invest Ophthalmol Vis Sci 27:500-506, 1986.
Auran JD, Starr MB, Jakobiec FA. Acanthamoeba Keratitis - A review of the literature. Cornea 6:2-26, 1987.
Sharma S, Srinivasan M, George C. Acanthamoeba Keratitis in non-contact lens wearers. Arch Ophthalmol 108:676-678, 1990.
Wilhemus KR, Osato MS, Font RL, et al. Rapid diagnosis of Acanthamoeba keratitis using calcofluor white. Arch Ophthalmol 104:1309-1318, 1986.
Robin JB, Chan R, Rao NA, et al. Fluorescein conjugated lectin visualization of fungi and Acantha
moeba in infectious keratitis. Ophthalmology 96:11981202, 1989.
Mathers W, Stevens G Jr, Rodrigues M, et al.Immuno pathology and electron microscopy of Acanthamoeba
keratitis. Am J Ophthalmol 103:626-635, 1987.
Hyndiuk RA, Glasser DB. Herpes simplex keratitis in infections of the eye: Diagnosis and management. In: Tabbara K, Hynduik RA (eds). Infections of the Eye. Boston, Little Brown & Company, pp. 343-368, 1986.
Mayers-Elliott RH, Pettit TH, Maxwell WA. Viral antigens in the immune ring of herpes simplex stromal keratitis. Arch Ophthalmol 98:897-904, 1980.
Hendricks RL. Role of lymphokines in HSV-1 corneal stromal disease. Invest Ophthalmol Vis Sci (ARVO Abstracts). 32:854, 1991.
Doymaz MT, Rouse BT. Herpetic stromal keratitis: An immunopathologic disease mediated by CD4+ T lymphocytes. Invest Ophthalmol Vis Sci 33:2165-2173, 1992.
Cantin EM, Chen J, McNeill J, et al. Detection of herpes simplex virus DNA sequences in corneal transplant recipients by polymerase chain reaction assays. Curr Eye Res 10(Suppl):15-22, 1991.
Kaye SB, Lynas C, Patterson A, et al. Evidence for herpes simplex virus latency in the human cornea. Br J Ophthalmol 75:195-200, 1991.
Cook SD, Hill JM, Lynas C, et al. Latency associated transcripts in corneas and ganglia of HSV-1 infected rabbits. Br J Ophthalmol 75:644-648, 1991.
Tumpey TM, Glorioso JG, Hendriks RL. Expression of the HSV-1 latency associated transcript (LAT) promotor in ocular tissues of the mouse. (ARVO Abstract). Invest Ophthalmol Vis Sci 31:1534, 313, 1990.
Foulks GN. Bacterial infections of the conjunctiva and cornea. In: Albert DM, Jakobiec FA (ed). Principles and Practice of Ophthalmology (vol. 1). Philadelphia, WB Saunders Company, 1994.
Wilson LA, Sexton RR. Laboratory aids in diagnosis. In: Duane TD (ed). Clinical Ophthalmology, vol 4. Philadelphia, Harper and Row, 1986.
Jones DB, Liesegang TJ, Robinson NM. Laboratory diagnosis of ocular infections. Cumitech-13, American Society for Microbiology, Washington DC, 1981.
Sharma S, Sankaridurg PR, Ramachandran L, et al. Is the conjunctival flora a reflection of the pathogenic bacteria causing corneal ulceration? Invest ophthalmol Vis Sci (suppl) 35(4) : 1947, 1994.
Wilson LA, Sexton RR. Laboratory diagnosis of fungal keratitis. Am J Ophthalmol 66:646-653, 1968.
Brinsser JH, Burd EM. Principles of diagnostic ocular microbiology. In: Tabarra KF, Hynduik RA (eds). Infections of the Eye. Boston, Little Brown & Company, pp. 73-92, 1986.
Jones DB. Initial therapy of suspected microbial corneal ulcers. II. Specific antibiotic therapy based on corneal smears. Surv Ophthalmol 24:97-116, 1979.
Marines HM, Osato MS, Font RL. The value of calcofluor white in the diagnosis of mycotic and Acanthamoeba
infections of the eye and ocular adnexa. Ophthalmology 94:23-26, 1987.
Robin JB, Chan R, Rao NA, et al. Fluoresceinconjugated lectin visualisation of fungi and Acanthamoeba
in infectious keratitis. Ophthalmology 96:11981202, 1989.
Thomas PA, Kuriakose T, Kirupashanker MP, et al. Use of lactophenol cotton blue mounts of corneal scrapings as an aid to the diagnosis of mycotic keratitis. Diagn Microbiol Infect Dis 14:219-224, 1991.
Thomas PA, Kuriakose T. Rapid detection of Acanthamoeba cysts in corneal scraping by lactophenol cotton blue staining. Arch Ophthalmol 108:168, 1990.
O'Day DM, Akrabawi PL, Head SW, et al. Laboratory isolation techniques in human and experimental fungal infections. Am J Ophthalmol 87:688-693, 1979.
Jones DB, Visvesvara GS, Robinson NR. Acanthamoeba
polyphaga keratitis and Acanthamoeba
uveitis associated with fatal meningoencephalitis. Trans Ophthalmol Soc UK 95:221-232, 1975.
Ma P, Visvesvara GS, Martinez AJ, et al. Naeglaria and Acanthamoeba
infections: Review. Rev infect infect Dis 12:490-513, 1990.
Dart JK, Badenoch PR. Bacterial adherence to contact lenses. CLAO J 12:220-224, 1986.
Rao NA. A laboratory approach to rapid diagnosis of ocular infections and prospects for the future. Am J Ophthalmol 107:283-291, 1989.
Goldstein LC, Corey L, McDougall JK, et al. Monoclonal antibodies to herpes simplex viruses. Use in antigenic typing and rapid diagnosis. J Infect Dis 147:829-837, 1983.
Ruiz C, Labella F, Duran JA, et al. Accuracy of direct immunofluorescene for the diagnosis of herpes simplex keratitis (ARVO Abstracts). Invest Ophthalmol Vis Sci 32:805, 1991.
Kowalski RP, Romanowski EG, Cruz TA, et al. Immuno-assay, PCR or slit-lamp diagnosis. Which is superior for detecting HSV disease? (ARVO Abstracts). Invest Ophthalmol Vis Sci 32:805, 1991.
Hayashi S, Hanashiro R, Nishi M, et al. Polymerase chain reaction for detection of herpes family genome in keratitis of unknown etiology. (ARVO Abstracts). Invest Ophthalmol Vis Sci 33:788, 1992.
Salmon VC, Turner RB, Spirenza MJ, et al. Rapid detection of herpes simplex virus in clinical specimens by centrifugation and immunoperoxidase staining. J Clin Microbiol 23:683-686, 1986.
Grandien M, Pettersson CA, Gardner PS, et al. Rapid viral diagnosis of acute respiratory infections: comparison of enzyme-linked immunosorbent assay and the immunofluorescence technique for detection o f viral antigens in nasopharyngeal secretions. J Clin Microbiol 22:757, 1985.
Jones DB. Initial therapy of suspected microbial corneal ulcers. II. Specific antibiotic therapy based on corneal smears. Surv Ophthalmol 24:97, 1979.
Baum J. Therapy for ocular bacterial infection. Trans Ophthalmol Soc UK 105:69, 1986.
Asbell P, Stenson S. Ulcerative keratitis: Survey of 30 years laboratory experience. Arch Ophthalmol 100:77, 1982.
Laibson PR. Cornea and sclera. Arch Ophthalmol 88:553, 1972.
Belmont JB, et al. Noninfectious ring-shaped keratitis associated with Pseudomonas aeruginosa.
Am J Ophthalmol 93:338, 1982.
O'Day DM, Robinson RD, Head WS. Efficacy of antifungal agents in the cornea. I. A comparative study. Invest Ophthalmol Vis Sci 24:1098, 1983.
Taab WP. Natamycin (pimaricin): Its properties and possibilities in medicine. Stuttgart, Georg, Thieme Verlag, 1972.
O'Day DM, et al. In vitro and in vivo susceptibility of candida keratitis to topical polyenes. Cur Eye Res 6:36, 1987.
O'Day DM, et al. Corneal penetration of amphotericin B and natamycin. Cur Eye Res 5:877, 1986.
O'Day DM, Ray WA, Head WS. Influence of the corneal epithelium on the efficacy of topical antifungal agents. Invest Ophthalmol Vis Sci 25:855, 1984.
Ficker LA, Kirkness C, Wright P. Prognosis for keratoplasty in Acanthamoeba
keratitis. Ophthalmology 100:105, 1993.
Driebe WT, et al. Acanthamoeba
keratitis: Potential role for topical clotrimazole in combination chemotherapy. Arch Ophthalmol 106:1196, 1988.
Varga TH, et al. Treatment of Acanthamoeba
keratitis. Am J Ophthalmol 115:466, 1993.
Schwabb I. Oral acyclovir in the management of herpes simplex ocular infections. Ophthalmology 95:423, 1988.
Coster DJ, Jones BR, Falson MG. Role of debridement in treatment of herpetic keratitis. Trans Ophthalmol Soc UK 97:314, 1977.
Laibson PR. Current therapy of herpes simplex virus infection of the cornea. Int Ophthalmol Clin 13:39, 1973.
Kernbaum S, Hauchecome J. Administration of levodopa for relief of herpes zoster pain. JAMA 246:132, 1981.
Mavligit GM, Talpaz M. Cimetidine for herpes zoster. N Engl J Med 310:318, 1984.
Olson ER, Ivy HB. Stellate block for trigeminal zoster. J Clin Neuro Ophthalmol 1:53, 1981.
O'Day DM, Head WS, Robinson RD. Efficacy of antifungal agents in the cornea 3. Influence of the corneal epithelium. (ARVO abstracts). Invest Ophthalmol Vis Sci 24(Suppl):223, 1983.
Serdarevic 0, Darrell RW, Kruger RR, et al. Excimer laser therapy for experimental candida keratitis. Am J Ophthalmol 99:534-538, 1985.
Portnoy SL, Insler MS, Kaufman HE. Surgical management of corneal ulceration and perforation. Surv Ophthalmol 34(1) 47-58, 1989.
Arffa RC. Grayson's Diseases of the Cornea. St Louis, Mosby Year Book, pp. 194-212, 1991.
Leahey AB, Gottsch JD, Stark WJ. Clinical experience with n-butyl cyanoacrylate (Nexacryl) tissue adhesive. Ophthalmology 100:173-181, 1993.
Eiferman RA, Snyder JW. Antibacterial effect of cyanoacrylate glue. Arch Ophthalmol 101:958-960, 1963.
Killingsworth DW, Stern GA, Driebe WT,et al. Results of therapeutic penetrating keratoplasty. Ophthalmology 100:534-599, 1993.
[Figure - 1], [Figure - 2], [Figure - 3], [Figure - 4], [Figure - 5], [Figure - 6], [Figure - 7], [Figure - 8], [Figure - 9], [Figure - 10], [Figure - 11], [Figure - 12], [Figure - 13], [Figure - 14], [Figure - 15], [Figure - 16], [Figure - 17], [Figure - 18], [Figure - 19], [Figure - 20], [Figure - 21]
[Table - 1], [Table - 2], [Table - 3], [Table - 4], [Table - 5], [Table - 6], [Table - 7], [Table - 8], [Table - 9], [Table - 10]
|This article has been cited by|
||Distinguishing infective versus noninfective keratitis
| ||Srinivasan, M., Mascarenhas, J., Prashanth, C.N. |
| ||Indian Journal of Ophthalmology. 2008; 56(3): 203-207 |
||Mycotic keratitis: An overview of diagnosis and therapy
| ||Shukla, P.K., Kumar, M., Keshava, G.B.S. |
| ||Mycoses. 2008; 51(3): 183-199 |
||Fusarium and Candida albicans biofilms on soft contact lenses: Model development, influence of lens type, and susceptibility to lens care solutions
| ||Imamura, Y., Chandra, J., Mukherjee, P.K., Lattif, A.A., Szczotka-Flynn, L.B., Pearlman, E., Lass, J.H., (...), Ghannoum, M.A. |
| ||Antimicrobial Agents and Chemotherapy. 2008; 52(1): 171-182 |
||Phenotypic and plasmid pattern analysis of Staphylococcus epidermidis in bacterial keratitis
| ||Nayak, N., Satpathy, G., Vajpayee, R., Mrudula, S. |
| ||Indian Journal of Ophthalmology. 2007; 55(1): 9-13 |
||Amniotic membrane transplantation for persistent corneal ulcers and perforations in acute fungal keratitis
| ||Chen, H.-C., Tan, H.-Y., Hsiao, C.-H., Huang, S.C.-M., Lin, K.-K., Ma, D.H.-K. |
| ||Cornea. 2006; 25(5): 564-572 |
||Comparison of two methods of PCR followed by enzymatic restriction digestion for detection and typing of herpes simplex viruses isolated from patients with mucocutaneous or cutaneous lesions
| ||Herrera-Martínez, E., Ondarza-Aguilera, R., Estrada-Parra, S., Pérez D., G., Barrón, B.L. |
| ||Revista Latinoamericana de Microbiologia. 2005; 47(3-4): 76-81 |
||Bacterial and fungal profile of corneal ulcers - A prospective study
| ||Bashir, G., Shah, A., Thokar, M.A., Rashid, S., Shakeel, S. |
| ||Indian Journal of Pathology and Microbiology. 2005; 48(2): 273-277 |
||Epidemiological Characteristics and Laboratory Diagnosis of Fungal Keratitis. A Three-year Study
| ||Bharathi, M.J., Ramakrishnan, R., Vasu, S., Meenakshi, R., Palaniappan, R. |
| ||Indian Journal of Ophthalmology. 2003; 51(4): 315-321 |
||Current Perspectives on Ophthalmic Mycoses
| ||Thomas, P.A. |
| ||Clinical Microbiology Reviews. 2003; 16(4): 730-797 |
||Nocardia asteroides keratitis: Report of seven patients and literature review
| ||Rao, S.K., Madhavan, H.N., Sitalakshmi, G., Padmanabhan, P. |
| ||Indian Journal of Ophthalmology. 2000; 48(3): 217-221 |
||Corneal ulceration in the elderly in Hyderabad, south India
| ||Kunimoto, D.Y., Sharma, S., Garg, P., Gopinathan, U., Miller, D., Rao, G.N. |
| ||British Journal of Ophthalmology. 2000; 84(1): 54-59 |
||Detection of Herpes simplex virus (HSV) genome using polymerase chain reaction (PCR) in clinical samplesComparison of PCR with standard laboratory methods for the detection of HSV
| ||Madhavan, H.N., Priya, K., Anand, A.R., Lily Therese, K. |
| ||Journal of Clinical Virology. 1999; 14(2): 145-151 |
||Clinical Pattern of Recurrent Herpes Simplex Keratitis
| ||Saini, J.S., Agarwala, R. |
| ||Indian Journal of Ophthalmology. 1999; 47(1): 11-14 |
| ||Abad, J.-C., Foster, C.S. |
| ||International Ophthalmology Clinics. 1996; 36(3): 1-15 |