Confoscan: an ideal therapeutic aid and screening tool in acanthamoeba keratitis.

Although present worldwide, Acanthamoeba keratitis (AK) is a rare condition. It is a protozoal infection of the eye that is generally caused by wearing contaminated contact lenses or lens solutions. Confoscan and confocal scanning laser tomography (CSLT) are in vivo noninvasive diagnostic tools which provide high definition images of corneal microstructures. Laser in situ keratomileusis (LASIK) is a very common refractive surgery. We report a case series in which the first patient had contact lens induced Acanthamoeba keratitis with corneal epitheliopathy that was unresponsive to conservative treatment. Epithelial debridement was performed based on confoscan findings which confirmed the presence of Acanthamoeba cysts. Subsequently, the cornea re-epithelialized over two days. Another patient had CSLT prior to the LASIK which showed stromal cyst-like structures suggestive of Acanthamoeba keratitis. Four months after medical therapy, repeat CSLT was negative for Acanthamoeba cysts. Third patient was diagnosed with Acanthamoeba infection after undergoing lamellar keratoplasty. CSLT should be used as a screening procedure prior to any corneal refractive surgery to detect and treat protozoal and other infections preoperatively.

INTRODUCTION

Acanthamoeba keratitis (AK) is a rare condition with an infection rate of 0.2 per 10,000 contact lens wearers yearly.1 Acanthamoeba is a protozoan, present in soil, almost all sources of water, sewer, insect vectors, overhead water tanks, and as a commensal in human nasopharynx. This protozoan is the most common cause of keratitis in contact lens wearers (90%). The most common symptoms include pain, photophobia, redness, reduced vision and tearing. Early diagnosis (within 3 weeks of onset of symptoms) can restore visual acuity of 6/12 in 90% patients, whereas late diagnosis can be devastating including loss of the eye.

Current laboratory techniques include cultures, stains, microscopy and molecular analysis. Culture may be positive in 0-68% cases only, making it mandatory to rely on non-laboratory techniques as well. The advent of polymerase chain reaction (PCR) is encouraging but not yet firmly established. The risk factors include soft contact lenses, hard gas permeable lenses, overnight use of contact lenses, poor personal hygiene and trauma. Biguanides and diamidines form the mainstay of treatment. CSLT is an in vivo noninvasive diagnostic tool that provides high definition images of corneal microstructures as small as 4 μm. It is particularly useful when organisms are ≥15 μm in size, which makes it useful in detecting Acanthamoeba trophozoites (25-40 μm) and cysts (15-28 μm).2

Laser in situ keratomileusis (LASIK) is a very common refractive procedure selected mainly by young adults for correction of their refractive errors. Young adults also are usual contact lens wearers and hence at risk of Acanthamoeba keratitis. This case series illustrates the significance of confocal biomicroscopy in the diagnosis and treatment of this infection.

CASE REPORTS

Case 1

A 27-year-old female physician and soft contact lens wearer (monthly-disposable) presented with a history of pain, photophobia, foreign body sensation and redness in the right eye. There was no history of fatigue or prior labial cold sores or herpes virus infection of the eye.23 The chronology of events of the disease process was as follows:

First visit

There was corneal epithelial irregularity with punctate staining in the paracentral area without any corneal epithelial defect. An initial diagnosis of contact lens induced epitheliopathy was made and lubricating drops were prescribed.

Second visit (five days)

Patient did not appreciate any improvement, complained of increased severity of pain. Slit-lamp examination revealed a ring-shaped lesion with corneal haze and a pseudo-dendrite configuration [Figure 1], stromal edema, radial keratoneuritis and anterior chamber cells (2+). A provisional diagnosis of Acanthamoeba keratitis was made, mainly on the basis of history, ring-shaped lesion and radial keratoneuritis.4 Corneal scraping was sent for smears, wet mount potassium hydroxide (KOH) stain, Giemsa stain and culture on non-nutrient agar with Escherichia coli overlay. The patient was prescribed topical moxifloxacin and lubricating drops awaiting the outcome of the smear and culture tests.

Figure 1

(Case 1) Fluorescein staining of corneal pseudodendrite

Third visit (eight days)

Smears did not show Acanthamoeba, however, the patient symptoms improved. The density of the corneal dendritic infiltrates regressed. The likely diagnosis now shifted to herpes simplex keratitis and she was prescribed oral and topical acyclovir.

Fourth visit (12 days)

The epithelial defect had healed and topical fluoromethalone drops were prescribed to reduce stromal haze and the patient was instructed to continue acyclovir. Culture results of the corneal scrapings were negative for acanthamoeba.

Fifth visit (26 days)

The patient presented with a relapse of symptoms including pain, photophobia and redness. She had lid edema and could not open the affected eye. Slit-lamp examination showed ciliary congestion, diffuse corneal superficial punctate keratitis, stromal edema, anterior chamber flare (1+), anterior chamber cells (2+) and reduced corneal sensation. Corneal confocal microscopy (Confoscan 4, Nidek Co. Ltd., Gamagori, Japan) was performed which revealed multiple Acanthamoeba cysts in the anterior epithelium [Figure 2a] with a highly irregular epithelial surface. The corneal nerves were enlarged and keratocytes showed increased reflectivity indicating activation [Figure 2b]. This was highly suggestive of Acanthamoeba keratitis. Corneal epithelial debridement was performed and she was prescribed three medications (propamidine isethionate 0.1%, polyhexamethylene biguanide 0.02% and chlorhexidine 0.02%). All these medications were to be instilled on an hourly basis. Topical cyclopentolate 1% and oral ketoconazole (200 mg OD) were added as well.

Figure 2

(a) (Case 1) Confoscan image showing characteristic trophozoite Acanthamoeba cysts with double halo sign (b) Highly refractile activated keratocytes (c) Repeat corneal confoscan after 3 months showed only one cyst remaining

Sixth visit (76 days)

The patient continued on triple therapy for acanthamoeba infection. However, she had reduced the frequency of eye drops to twice daily on her own. She was asymptomatic and slit-lamp examination revealed a healed epithelium with no staining.

Seventh visit (90 days)

The patient stopped medications on her own. Repeat Confoscan 4 examination showed marked reduction in the number of cysts (only one remaining - Figure 2c), with a regular epithelial surface, reduced keratocyte activity and a healthy endothelium. The patient was completely symptom free and slit-lamp examination showed no signs suggestive of epithelial keratitis. She was advised to continue at least with one anti-acanthamoeba drug (propamidine isethionate 0.1%) for 6 months to prevent recurrence, which she refused. The patient continues to be symptom-free with normal appearing corneal epithelium and mild peripheral anterior stromal haze after 18 months follow up.

Case 2

A 25 year old male underwent bilateral LASIK in August 2010 at a private center. Six days postoperatively the patient developed bilateral diffuse lamellar keratitis. Visual acuity with correction (decimal notation) was 0.05 in the right eye and 0.2 in the left eye. Slit-lamp examination was as following:

Right eye

Missing corneal flap, severe stromal corneal haze and edema [Figure 3].

Figure 3

(Case 2) Diffuse lamellar keratitis six days after laser in situ keratomileusis

Left eye

Edema of the corneal epithelial flap, anterior stromal infiltrates and a button hole in the flap just below the visual axis.

CSLT (Heidelberg Retina Tomograph Rostock Corneal Module; Heidelberg Engineering, Heidelberg Germany) in both eyes revealed stromal cyst-like structures [Figure 4a] suggestive of Acanthamoeba keratitis. Some of the cysts had characteristic double-walled structures [Figure 4b]. The patient was put on triple topical anti-Acanthamoeba medication (propamidine hydrochloride 0.1%, polyhexamethylene biguanide 0.02% and chlorhexidine 0.02%, every 2 hours; the drops were obtained from Moorfields Eye Hospital, United Kingdom) and systemic ketoconazole (200 mg daily). Repeat CSLT at 4 months showed disappearance of the Acanthamoeba cysts and the patient was maintained on topical propamidine, twice daily for 3 months. At 9 months follow up the patient had gained vision to 1.0 with correction in both eyes.

Figure 4

(a) (Case 2) CSLT: A characteristic cyst with surrounding halo-suggestive of Acanthamoeba keratitis. (b) (Case 2): Multiple double-walled cystic structures suggestive of Acanthamoeba keratitis. (c) (Case 3): Corneal stromal cyst suggestive of Acanthamoeba keratitis

Case 3

A 23-year-old male patient underwent deep lamellar anterior keratoplasty in September 2010 without any intraoperative or postoperative complications. He was advised to use contact lenses after the procedure. Six months postoperatively, CSLT was performed to evaluate the endothelial cell count as a routine follow up procedure. There was an incidental finding of Acanthamoeba cyst-like structures in the anterior stroma bilaterally [Figure 4c]. The patient had been asymptomatic all along. Since the patient was on corticosteroids, he was advised to start propamidine 0.1% eye drops as a prophylactic measure.

DISCUSSION

Acanthamoeba keratitis can be a diagnostic challenge. Since the signs and symptoms masquerade as other infections, the specific treatment is often delayed.3 The response to treatment is rewarding only when the infection is diagnosed early and treated within four weeks of the onset of symptoms.5 Clinically it is important to look for epithelial lesions such as pseudodendrites, edema and necrosis in the early stages as these can differentiate Acanthamoeba lesions from herpetic infections. This can save time for diagnosis, investigations and treatment. Pain in Acanthamoeba keratitis is generally disproportionate to the lesion. Even when a firm diagnosis is made, response to the treatment may be frustrating. Acanthamoeba is difficult to culture in routine media. The tissue should be cultured in non-nutrient agar seeded with E. coli which remains the standard culture for this organism. Microbiology remains an extremely important test for the diagnosis of Acanthamoeba keratitis but it may not always be helpful, as illustrated in our first case. In this case series it was only corneal confocal microscopy that confirmed the diagnosis of Acanthamoeba keratitis that allowed the appropriate treatment. Quite often the role of corneal epithelial debridement is overlooked. Over-reliance on medications can lead one to overlook the advantage of epithelial debridement in these cases. Epithelial debridement removes the infected and necrotic epithelium along with trophozoites and cysts, thereby reducing the microbial load and allows better penetration and effectiveness of topical therapy. A combination of corneal confoscan and epithelial debridement played a crucial role in the diagnosis and treatment of this patient. There was minimal residual corneal haze and reduction in the frequency and duration of the topical medication. The role of epithelial debridement in Acanthamoeba keratitis is supported by Brooks.6

Laser in situ keratomileusis (LASIK) has been associated with a variety of infections including fungal keratitis, infectious keratitis, mycobacterial keratitis, nocardia keratitis, and endophthalmitis. A detailed integration of the published literature on LASIK associated infections is available.7 There have been very few associations of LASIK with Acanthamoeba keratitis and our search in Pubmed revealed only two reports of Acanthamoeba keratitis post-LASIK.

Confoscan/CSLO has been an asset in aiding the diagnosis and treatment response in Acanthamoeba keratitis - a condition that can frustrate a very well-established eye center and the laboratory. The use of antibiotics, steroids and anti-virals is quite common while awaiting laboratory investigations and other diagnostic methods. It is difficult to determine how long the patients in our series had been harboring Acanthamoeba cysts in the cornea. The findings in our patients should encourage us to remember that, whatever the type of corneal refractive surgery, preoperative screening for Acanthamoeba should be mandatory, knowing that these patients may have to use contact lenses for a prolonged period of time. Whether it was diagnosing Acanthamoeba infection after (case 1) or before LASIK (case 2), confoscan/CSLT really made an impact on diagnosis and treatment in these patients. The decision of performing LASIK in patients, who had Acanthamoeba keratitis earlier, is doubtful. However Lim and Wei have reported no recurrence of Acanthamoeba keratitis 3 months after LASIK procedure in a case that was treated successfully for Acanthamoeba keratitis in the same eye.8 Acanthamoeba keratitis after LASIK has been reported from India but in this case no confocal microscopy was done.9

CONCLUSION

In conclusion, Acanthamoeba keratitis continues to be a great mimicker, and challenging in terms of diagnosis, laboratory investigations and management. Confoscan/CSLT should not be a substitute for clinical examination, smear, culture or biopsy procedures in Acanthamoeba keratitis; however, whenever possible, it should be used as a screening procedure before any type of corneal refractive surgery is done so that latent subclinical protozoal infections can be diagnosed and treated, making surgery safer. A larger study comparing response to treatment with and without debridement will help to identify the role of ‘confoscan/CSLT guided debridement’ in corneal infections with Acanthamoeba keratitis. Also more studies and clinical experience would be valuable in determining the viability and feasibility of recognizing “confoscan screening” as an acceptable routine procedure in all refractive surgeries.