**2.1.8 Graft clarity and visual acuity**

At the last follow up visit the visual acuity was 0.43. In 3 cases it was only light perception, but the other 3 patients had very good visual acuity (0.7-1.0).

In the emergency cases the visual acuity was much worse (light perception in 3 cases and 0.9 in one case), then in the proposed cases (0.7 and 1.0). In 5 cases the cornea graft was clear, rejection and decompensation developed only in one case.

## **2.2 Corneal graft surgery/keratoplasty 2.2.1 Emergency keratoplasty**

Emergency keratoplasty or keratoplasty à chaud is often the only possible intervention to prevent complete and irreversible vision loss in patients with severe corneal disorders. The main indications for this urgent surgical procedure are corneal perforations or imminent perforations, maintaining the integrity of the globe, as well as persisting infectious keratitis. Due to acute inflammation of the anterior ocular segment, emergency keratoplasty is supposed to have a worse outcome and more postoperative complications especially more immune reactions than scheduled keratoplasty. Keratoplasty for management of the acute complications of Acanthamoeba keratitis has with few exceptions in small numbers of patients (Illingworth & Cook, 1998; Nguyen et al., 2010; Maier et al., 2007; Shi et al., 2009), generally been reported to have poor results. Graft failure due to recurrent infection is common when keratoplasty is performed in an inflamed eye. In most cases, however, keratoplasty has been successful in maintaining the integrity of the globe, and a second procedure has often resulted in a good visual outcome (Illingworth & Cook, 1998; Maier et al., 2007).

Keratoplasty in Contact Lens Related Acanthamoeba Keratitis 45

Kitzmann 2009 30 22 45.5 8 100

Ficker 1993 13 7 20 6 100

Awwad 2005 13 13 100

Qian 2010 8 8 100

Bacon 1993 34 21 35 13 69

be kept to the minimum size required to excise all ulcerated and necrotic tissue, retaining clinically healthy (but usually subclinically infected) tissue. This is because of the risk of rejection with large grafts and because repeating grafting may be needed as a result of recurrence; a further graft represents a new food source for the organism and can be used to attract residual amoebae. Recurrence of disease in a graft was frequent in the first 2 weeks after surgery when keratoplasty was performed in an inflamed eye before the introduction of biguanides; it typically involved the donor periphery, usually without clinical involvement of the host. Late recurrences, several months after surgery, may also occur. The use of large grafts worsens the prognosis because it increases the chances of an immune response to the graft. Furthermore, the use of a large graft indicates a more severe

Host trephination size ranged from 7 to 11 mm depending on the extent of the corneal infiltration. The trephination should be performed beyond the clinical areas of infiltrate, including satellite lesions. Confocal microscopy can be used to outline the extent of involvement or the size of the trephination (Kashiwabuchiet al., 2008; Nguyen et al., 2010). Meier et al. reported about a trend toward improved clear graft survival and fewer immune reactions following emergency keratoplasty using smaller grafts (≤8.00 mm) (Awwad et al.,

Ideally, a corneal scar should be debulked as much as possible because residual cysts might still be present and might lead potentially to a recurrence of the disease when exposed to topical corticosteroids after PK. In those cases, rather than jeopardizing graft survival by trephining a large donor button (sizes of 9 mm and beyond), we suggest waiting for a longer period of time after the discontinuation of medical treatment, with repetitive confocal examination when available, before committing the patient to surgery

Por 2009 15 12 3

Nguyen 2010 9 9 100

Maier 2007 13 13 73 Kashiwabuchi 2008 32 32 43.8

Tanhehco 2010 8 8 50

Illengworth 1995 9 9 100

Sharma 2000 3 3 0

Radford 1998 31

Table 2. Success rate of keratoplasty

preoperative infection (Ficker et al., 1993).

2005; Maier et al., 2007).

(Awwad et al, 2005).

author year total acute success % optical success %

Since the introduction of the biguanides as medical therapy, PK has not been recommended as a treatment for the elimination of organisms; emergency keratoplasty should therefore be reserved as a treatment for corneal perforation, imminent perforation and fulminant corneal abscess. Many of these eyes will be severely inflamed with uncontrolled scleritis and limbitis, which should be treated before surgery with systemic immunsuppression using prednisolone (0.5 to 1 mg/kg/day) and/or cyclosporine (3 to 7.5 mg/kg/day), which is tapered as inflammation is controlled in the post graft period (Dart et al. 2009; Maier et al., 2007). Performing PK before scleral or peripheral corneal extension can minimize the risk of recurrence and poor outcome (Nguyen et al., 2010; Tanhehco & Colby, 2010).

#### **2.2.2 Therapeutic/optical keratoplasty**

Several authors (Awwad et al., 2005; Por et al., 2009) recommend observing for at least 3 months of clinical inactivity after completion of antiamoebic therapy before attempting PK. Most recently, Kitzmann et al. (Kitzmann et al., 2009) compared outcomes of 22 eyes with Acanthamoeba keratitis undergoing emergency penetrating or anterior lamellar keratoplasty and 9 eyes undergoing optical penetrating or anterior lamellar keratoplasty. Although all eyes ultimately achieved microbiological cure, there was a 41% recurrence rate of Acanthamoeba keratitis after emergency keratoplasty compared with 22% recurrence rate after optical keratoplasty. The Kaplan–Meier graft survival was 37.5% (95% confidence interval, 16.8–58.4) at 5 years in the therapeutic group compared with 100% in the optical group. Shi et al. (Shi et al., 2009) reported 28% recurrence after emergency PK. For acanthamoeba keratitis, Ficker et al. (Ficker et al., 1993) found that clear graft survival was only 20% if penetrating keratoplasty was performed during the acute, inflamed state of the disease compared to 100% clear graft survival when penetrating keratoplasty was performed in a quiet state of the disease. The reason for the difference compared to other results might possibly be differences in case selection at the time of diagnosis, preoperative treatment and the time point of the decision for emergency keratoplasty seem to be most important for the outcome of the graft (Maier et al., 2007). Dart et al. (Dart et al., 2009) present cases, when therapeutic keratoplasty is rarely used but lamellar or PK to improve vision is carried out in patients with scarred corneas and/or irregular astigmatism. The outcome of corneal transplantation is good in this group of patients. Exacerbations of scleritis and limbitis may occur following graft surgery in these eyes and may need to be treated with systemic anti-inflammatory treatment (Dart et al., 2009). Quian et al. (Qian et al., 2010) performed 8 optical perforating keratoplasty after at least 4 months medical treatment, with great success. Awwad et al. (Awwad et al., 2005) published an article about 13 eyes which underwent optical PK for visual rehabilitation after acanthamoeba keratitis. They mean that PK for visual restoration after resolution of *Acanthamoeba* keratitis in quiet eyes that are judged cyst-free by preoperative confocal microscopy appears to have an excellent long-term prognosis. Waiting at least 3 months after the discontinuation of medical therapy also appears helpful to rule out latent infection because of cyst reactivation and to allow the active inflammation to cease.

The success of keratoplasty in the literature are summarised in Table 2.

#### **2.2.3 Donor size**

The extent of the infected corneal tissue cannot be identified and should be assumed to include the entire cornea so that, unlike grafts for other corneal infections, which should be large enough to remove all contaminated tissue, the graft for Acanthamoeba keratitis should

Since the introduction of the biguanides as medical therapy, PK has not been recommended as a treatment for the elimination of organisms; emergency keratoplasty should therefore be reserved as a treatment for corneal perforation, imminent perforation and fulminant corneal abscess. Many of these eyes will be severely inflamed with uncontrolled scleritis and limbitis, which should be treated before surgery with systemic immunsuppression using prednisolone (0.5 to 1 mg/kg/day) and/or cyclosporine (3 to 7.5 mg/kg/day), which is tapered as inflammation is controlled in the post graft period (Dart et al. 2009; Maier et al., 2007). Performing PK before scleral or peripheral corneal extension can minimize the risk of

Several authors (Awwad et al., 2005; Por et al., 2009) recommend observing for at least 3 months of clinical inactivity after completion of antiamoebic therapy before attempting PK. Most recently, Kitzmann et al. (Kitzmann et al., 2009) compared outcomes of 22 eyes with Acanthamoeba keratitis undergoing emergency penetrating or anterior lamellar keratoplasty and 9 eyes undergoing optical penetrating or anterior lamellar keratoplasty. Although all eyes ultimately achieved microbiological cure, there was a 41% recurrence rate of Acanthamoeba keratitis after emergency keratoplasty compared with 22% recurrence rate after optical keratoplasty. The Kaplan–Meier graft survival was 37.5% (95% confidence interval, 16.8–58.4) at 5 years in the therapeutic group compared with 100% in the optical group. Shi et al. (Shi et al., 2009) reported 28% recurrence after emergency PK. For acanthamoeba keratitis, Ficker et al. (Ficker et al., 1993) found that clear graft survival was only 20% if penetrating keratoplasty was performed during the acute, inflamed state of the disease compared to 100% clear graft survival when penetrating keratoplasty was performed in a quiet state of the disease. The reason for the difference compared to other results might possibly be differences in case selection at the time of diagnosis, preoperative treatment and the time point of the decision for emergency keratoplasty seem to be most important for the outcome of the graft (Maier et al., 2007). Dart et al. (Dart et al., 2009) present cases, when therapeutic keratoplasty is rarely used but lamellar or PK to improve vision is carried out in patients with scarred corneas and/or irregular astigmatism. The outcome of corneal transplantation is good in this group of patients. Exacerbations of scleritis and limbitis may occur following graft surgery in these eyes and may need to be treated with systemic anti-inflammatory treatment (Dart et al., 2009). Quian et al. (Qian et al., 2010) performed 8 optical perforating keratoplasty after at least 4 months medical treatment, with great success. Awwad et al. (Awwad et al., 2005) published an article about 13 eyes which underwent optical PK for visual rehabilitation after acanthamoeba keratitis. They mean that PK for visual restoration after resolution of *Acanthamoeba* keratitis in quiet eyes that are judged cyst-free by preoperative confocal microscopy appears to have an excellent long-term prognosis. Waiting at least 3 months after the discontinuation of medical therapy also appears helpful to rule out latent infection because of cyst reactivation and to

recurrence and poor outcome (Nguyen et al., 2010; Tanhehco & Colby, 2010).

**2.2.2 Therapeutic/optical keratoplasty** 

allow the active inflammation to cease.

**2.2.3 Donor size** 

The success of keratoplasty in the literature are summarised in Table 2.

The extent of the infected corneal tissue cannot be identified and should be assumed to include the entire cornea so that, unlike grafts for other corneal infections, which should be large enough to remove all contaminated tissue, the graft for Acanthamoeba keratitis should


Table 2. Success rate of keratoplasty

be kept to the minimum size required to excise all ulcerated and necrotic tissue, retaining clinically healthy (but usually subclinically infected) tissue. This is because of the risk of rejection with large grafts and because repeating grafting may be needed as a result of recurrence; a further graft represents a new food source for the organism and can be used to attract residual amoebae. Recurrence of disease in a graft was frequent in the first 2 weeks after surgery when keratoplasty was performed in an inflamed eye before the introduction of biguanides; it typically involved the donor periphery, usually without clinical involvement of the host. Late recurrences, several months after surgery, may also occur. The use of large grafts worsens the prognosis because it increases the chances of an immune response to the graft. Furthermore, the use of a large graft indicates a more severe preoperative infection (Ficker et al., 1993).

Host trephination size ranged from 7 to 11 mm depending on the extent of the corneal infiltration. The trephination should be performed beyond the clinical areas of infiltrate, including satellite lesions. Confocal microscopy can be used to outline the extent of involvement or the size of the trephination (Kashiwabuchiet al., 2008; Nguyen et al., 2010).

Meier et al. reported about a trend toward improved clear graft survival and fewer immune reactions following emergency keratoplasty using smaller grafts (≤8.00 mm) (Awwad et al., 2005; Maier et al., 2007).

Ideally, a corneal scar should be debulked as much as possible because residual cysts might still be present and might lead potentially to a recurrence of the disease when exposed to topical corticosteroids after PK. In those cases, rather than jeopardizing graft survival by trephining a large donor button (sizes of 9 mm and beyond), we suggest waiting for a longer period of time after the discontinuation of medical treatment, with repetitive confocal examination when available, before committing the patient to surgery (Awwad et al, 2005).

Keratoplasty in Contact Lens Related Acanthamoeba Keratitis 47

treatment plants, bottled water, dental treatment units, hospitals and dialysis units, eyewash stations, and contact lenses and lens cases and as contaminants in bacterial, yeast, and mammalian cells. It mainly affects contact lens wearers with poor hygiene. In non contact lens wearers history of trauma in a garden is a risk factor. Patients' ages range from 4 to 64 years, with a mean age of 30 years, with no difference in genders. Infection usually affects only one eye, although it is occasionally bilateral. Most patients complain of symptoms of photophobia, pain, and tearing. The earliest signs may be non-specific and may take the form of epithelial micro erosions, irregularities, opacities or microcystic oedema, often with patchy anterior stromal infiltrates. There may be no fluorescein staining at the onset. Commonly, there is a dendriform keratitis that is often initially treated as herpes simplex infection. A ring infiltrate, usually with an overlying epithelial defect, is commonly seen. The ring may be incomplete, or occasionally it is double and concentric. The importance of early diagnosis cannot be overemphasized. We have to consider Acanthamoeba keratitis when symptoms associated with trauma especially involving vegetable matter or exposure to contaminated water, such as lake-, sea-, or spring water are mentioned. Acanthamoebic keratitis has to be differentiated from bacterial or fungal or herpes simplex virus keratitis. Diagnosis may be achieved by using different methods, including non invasive confocal microscopy, staining corneal scrapings with acridine orange, corneal scrapes or corneal biopsy specimens onto no nutrient agar containing *E. coli.,* cytological various staining methods, like indirect immunofluorescent-antibody assay, polymerase chain reaction (PCR) based method. The goals of medical therapy in Acanthamoeba keratitis include the eradication of viable cysts and trophozoites and rapid resolution of the associated inflammatory response. There is no single drug capable of eliminating the infection therefore drug combinations have been suggested as a treatment regimen. There are currently no drugs that are effective as monotherapy in Acanthamoeba keratitis, hence combinations are suggested. Acanthamoeba trophozoites are sensitive to most available chemotherapeutic agents (antibiotics, antiseptics, antifungals, antiprotozoals including metronidazole, antiviral, and antineoplastic agents). The diamidines and biguanides are currently the most effective cysticidal antiamoebics. With respect to the severity of the disease as well as the complications accompanying it there is a range of surgical methods to choose from. Corneal abrasion, cryosurgery, deep lamellar keratectomy with a conjunctival flap, amnion membrane transplantation, keratoplasty, deep lamellar keratoplasty. In some cases enucleation or evisceration is needed, because of severe inflammation, infection or secondary glaucoma. Antiamoebic therapy should be used before surgery and be continued postoperatively using drugs and doses that will minimize or avoid signs of toxicity. PHMB 0.02% has low clinical toxicity in most patients and is clinically less than that with either of the diamidines. We use PHMB 0.02% 6 to 8 times daily immediately after surgery, with an adequate level of topical steroid to control inflammation. This should be continued for at least 3 weeks while results of culture of the host keratectomy specimen are awaited. If viable organisms are cultured it is prudent to continue antiamoebic therapy 4 times daily while high-doses of steroids are needed usually for 6 months after surgery, as recurrent Acanthamoeba keratitis has occurred up to 3 months after an initially successful transplant. If culture of the excised host cornea is negative after 3 weeks we assume that most viable amoebae have been treated, and the topical antiamoebic therapy is reduced to 4 times daily

and stopped after 1 month.
