A case of fungal keratitis: A clinical and in vivo confocal microscopy assessment
, a, aaWilliam L. Miller, Amber Gaume Giannoni and Judith Perrigin
aUniversity of Houston, College of Optometry, 505 J. Davis Armistead Building, Houston, TX 77204–2020, United States
Available online 1 July 2008.
A presumed case of fungal keratitis is presented. Confocal images highlight the advantage of using such technology to follow therapeutic progress, particularly when standard laboratory testing is inconclusive. This case also demonstrates the importance of continued patient education regarding overall compliance and lens care.
Keywords: Fungal keratitis; Confocal microscopy; Contact lenses; Contact lens solution Article Outline
2. Case report
The cornea is equipped with several protective mechanisms to defend itself from microbiological and fungal infection. For a variety of reasons, this protection can be weakened or overwhelmed, resulting in an ulcerative keratitis. Common causative agents for microbial keratitis include bacteria, viruses, acanthamoeba and fungus. Erie et al.  reported the incidence of ulcerative
keratitis in the United States as approximately 11 in 100,000, however another report studied cases from 1969 to 1998 and found an incidence of infectious keratitis at 3.1 in 1000 clinical visits .
A synopsis of contact lens-related complications worldwide indicates that gas permeable contact lenses carry the lowest risk of microbial keratitis, with extended wear soft contact lenses carrying a much higher risk . Incident rates for microbial keratitis range from 9.3 to 20.9 per 10,000. This risk is elevated even when the patient is wearing a silicone hydrogel (SiHy) soft contact lens in an extended wear (EW) modality although it is less than EW hydrogel soft contact lens wearers  and . Although elevated in EW SiHy, the risk has been shown to be different when comparing non-severe cases of keratitis to severe cases with the former being nearly zero while the relative risk of a non-severe keratitis is around 4.0 . Of the etiologies known to cause
microbial keratitis, fungal entities are the least likely, comprising between 6% and 20% of cases
. Fungal infection is also relatively rare in contact lens wearers, typically constituting about 5% of all corneal infections ,  and . However, a resurgence of fungal keratitis cases occurred
in 2005 and 2006, even in the Northwest United States and the West Coast, which had not previously been seen as regions prone to fungal corneal infection. The incidence during this time period was 2.35 per 10,000 contact lens wearers and included among the factors related to the increasing numbers were patient compliance and contact lens care systems ,  and .
Some suggest that up to 30% of fungal keratitis cases also have an associated bacterial infection .
2. Case report
A 24-year-old male was immediately referred to the University Eye Institute (UEI) by a nearby University health center in October of 2007 for full care of a “paracentral corneal ulcer OS.” He reported awaking to left eye irritation after a 2 h nap 48 h prior to his visit to the UEI while wearing his 2 week-old Acuvue Oasys (senofilcon A) silicone hydrogel contact lenses. After lens removal, he repeatedly splashed tap water into his eye, but stopped after experiencing tremendous pain. He subsequently disposed of this pair of lenses prior to visiting our clinic, so the lenses were not available for visual inspection or microbiological analysis. He reported occasional extended wear and discarded his lenses every 2 weeks. The patient also denied trauma to the ocular surface. His current contact lens case was 6 months old and he habitually used either Renu Multiplus or Renu with MoistureLoc multi-purpose contact lens solution for daily cleaning and soaking of his lenses. The patient reported severe pain with tearing and denied recent activity around lakes, pools or other bodies of water. In addition, he was currently taking hydrocodone/acetominophen (Vicodin) as needed for pain, as prescribed by the health care center physician. The patient's entering Snellen visual acuity (VA) was not attainable due to severe tearing. Despite mild blepharitis, examination of his right eye was unremarkable. Biomicroscopy of his left eye revealed severe, diffuse bulbar injection, a central corneal ulcer and 4 clusters of peripheral infiltrates (Fig. 1). A moderate amount of corneal edema was present and the anterior chamber was quiet without evidence of cells, flare or hypopyon. The lesion was scraped with a Kimura spatula and the specimen was plated on Sabouraud's media and sent to a lab for processing. The diagnosis at that visit was a central corneal ulcer with peripheral infiltrates of unknown etiology. The patient was instructed to alternate drops of tobramycin and moxifloxacin (Vigamox) every 30 min and to use bacitracin ointment at night in the left eye only. In-office drops of Vigamox and 0.25% Scopolamine were given and he was asked to return in 24 h.
Fig. 1. Initial visit to the
University Eye Institute. Biomicroscopy demonstrates the primary and satellite infiltrative lesions. (Main annular lesion, arrows signify representative satellite lesions.) Bulbar conjunctival grade 4 hyperemia is also evident in the biomicroscopic beam.
The next day, the patient reported adhering to the prescribed medication regimen and stated that although his eye was still very red and the vision was poor, he was experiencing less pain and
?1tearing. Entering VA was 20/70 in the left eye with habitual spectacles. All infiltrates remained flat and showed slight improvement from the previous visit. A grade 4 (CCLRU scale) conjunctival hyperemia was still present . He was instructed to continue with the current
course of therapy and return in 2 days, or sooner if symptoms worsened. He was educated on the possibility of central scarring and permanent reduction of VA.
The patient was seen five times over the next week, including a visit to a corneal specialist who re-cultured the lesions on two different occasions. In vivo confocal microscopy (Confoscan 3, Nidek, Greensboro NC) was also performed at the start of this visit sequence time demonstrating hyper-reflective branching hyphae-like bodies present in the anterior stroma (Fig. 2a and b). The
width of the branching hyphae appeared to be between 5 and 10 μm which is consistent with what
is found with fungal agents. The in vivo confocal images were suggestive of the fusarium species. In addition, the contact lens case and solution bottle were also cultured. Fig. 3 shows the status of
the patient's solution bottle with evident non-compliance in both the case and bottle. A positive result for fungal growth was never obtained on any cultured media; however, the ophthalmologist added natamycin 5% (Natacyn) every 2 h to the current treatment regimen due to a high suspicion that the infection was fungal in nature. Biomicroscopy and confocal microscopy showed a slight improvement in the infiltrative response at the 1 week visit. (Fig. 4a and b).
Fig. 2. (a) Confocal images of the anterior stroma. Hyper-reflective areas representing hyphae are present during the return visits in week 1. (Arrows show fungal hyphae, shortened arrow demonstrates hyphae branching and stars indicate representative keratocytes in the field.) (b) Confocal images of the basal epithelium/anterior stroma demonstrating infiltrates during the return visits in week 1. (Long arrow shows corneal infiltrates while the shortened arrow shows basal epithelial cells. A suspected Langerhan cell is located at the asterisk.) Image section represents an oblique view showing basal epithelial cells in the lower left and anterior stroma on the upper right.
Full-size image (39K)
Fig. 3. Solution used by the patient exhibiting a soiled lid and cap top.
Fig. 4. (a) One week after initiation of antifungal therapy. Image demonstrates that the satellite
lesions are decreasing in intensity and density. (b) Confocal images showing scarring and residual infiltrates at the level of the basal epithelium after 1 week of antifungal treatment.
This patient had 13 additional visits (November 2006 through March 2007) over the next several months and continued to show extremely slow, but steady improvement at each examination. His right spectacle lens was updated to provide the best possible vision during the long course of healing. In vivo confocal microscopy was performed several times throughout this period. It revealed hyphae-like hyper-reflective linear structures throughout the sub-epithelial and anterior corneal stroma that slowly lessened in quantity and density during the course of healing. The hyper-reflective intensity of each hyphae also decreased over the course of these multiple confocal exams.
By late November, all corneal lesions were almost fully healed, although sub-epithelial corneal scars were present (Fig. 5a and b). Fig. 5b shows that the density of the scarring had intensified
and grown larger when compared with Fig. 4b. Fig. 5c shows that very few residual hyphae
remained as evidenced with confocal microscopy. Tobramycin and Vigamox were decreased to four times per day while Natacyn was reduced to three times per day. The reduction in Natacyn was necessary to decrease the medicamentosa effects which were affecting the corneal surface and the ultimate visual outcome of the patient. It was felt that since so few hyphae remained and their hyper-reflectivity was decreased the benefit of decreasing the dosage of the Natacyn outweighed the risk of possible reactivitation.
Fig. 5. (a) Biomicroscopy 3 weeks after initiation of antifungal therapy. Most of the satellite lesions have disappeared, however the primary lesion is still evident in the center of the image. (b) Residual scarring has intensified and increased in size (arrow) in the anterior stroma. A corneal stromal nerve is also present (star). (c) Confocal images showing fungal hyphae remnants (arrows) located in the anterior stroma with keratocytes scattered in the field.
On December 4, 2006, the patient reported irritation of his left eye and itching upon drop instillation. Moderate bulbar injection was observed along with diffuse superficial punctuate corneal and conjunctival staining; however no staining was observed overlying the old lesions. He was diagnosed with mild medicamentosa and all topical medications were reduced to twice per day. Seven days later, the symptoms remained the same and tobramycin was discontinued. Vigamox and Natacyn were reduced to one drop daily for the next month and then discontinued. All symptoms of medicamentosa resolved without sequelae.
The patient returned in March, 2007 for a full eye examination with a main complaint of ocular dryness and glare at night OS. His best corrected VA was 20/15 in the right eye and 20/20 in the left eye. Biomicroscopy revealed several old scars and he was educated that they were likely the cause of his symptoms. New spectacles with anti-reflective coating were prescribed and artificial tears were recommended for dryness. The patient indicated a strong desire to begin contact lens wear again which may be considered in the future. He was re-evaluated in June 2007 where the
cornea was found to be stable with a slight increase in the density of the scar in the left eye. Best corrected VA was stable in the right eye and had decreased to 20/25 in the left eye. No fluorescein staining was found in either eye.
Early diagnosis of fungal keratitis can be challenging because some patients may exhibit severe symptoms yet show little to no inflammation initially in the disease process. Trauma with vegetative matter as well as other corneal traumatic events can provide the fungal species access to a normally intact corneal surface. Other associated risk factors may include HIV, chronic ocular surface disease, herpes simplex keratitis and contact lens use . Biomicroscopy often reveals a
grayish-white infiltrative lesion with ill-defined borders. The excavated lesion, or in cases with yeast infections, elevated lesions, will stain with sodium fluorescein. Satellite lesions, often cited as a hallmark sign, may or may not be present. Fungal keratitis cases may mimic other infectious entities such as bacterial, herpes simplex or acanthamoebic. Therefore, in vivo confocal microscopy can be used as an adjunct diagnostic tool to evaluate the cornea under high magnification and aid in distinguishing the causative entities involved. Its non-invasive approach may be preferable in some early cases to the invasive evaluation that is necessary for a corneal biopsy. The use of the confocal microscope has been advocated and used to diagnose fungal and acanthamoebic keratitis early in the diseases course and in turn lead to a quicker initiation of therapy . In our particular case, the confocal microscopy was used not to only diagnose the keratitis but also to aid in the management of our patient by providing clues as to the eventual waning of the fungal species within the intact cornea. Especially in fungal and acanthamoebic cases which demonstrate unique characteristics using confocal microscopy, that when present, provide the necessary proof to begin appropriate therapy . Confocal microscopy has also been
shown to demonstrate the presence of fungal hyphae and acanthamoeba, and in addition to differentiate between the two . Even though cultures and smears continue to be the most
common method used to diagnosis both bacterial and fungal keratitis, the use of an in vivo confocal microscope can give an immediate picture of any suspected fungal involvement and may serve as a useful adjunctive diagnostic tool for detection . Additionally, a culture may not
yield results and in fact in a recent study a culture positive rate of 43% was found after corneal scraping . In a review of contact lens changes seen with in vivo confocal microscopy, Efron succinctly details the image attributes of infectious agents such as acanthamoeba and fungus .
Confocal microscopy can also aid in assessing the viability of the fungus during the treatment phase allowing for an accurate determination of when to halt treatment. Unless the level of suspicion undeniably suggests a fungal etiology, treatment is usually empirical, using antibiotics such as fluoroquinolones. Although specifically targeting bacterial entities, fluoroquinolones may also exhibit some antifungal properties by targeting prokaryotic topoisomerase IV as was seen in a recent report by Munir et al. . They report on at least 4 cases of fungal keratitis which resolved without antifungal treatment. However in the present case, the patient did not get better when treated with the fluoroquinolone moxifloxacin and was later started on antifungal treatment. In severe or traumatic cases, robust antibiotics such as vancomycin in combination with gentamicin or tobramycin should be used. Each antibiotic is usually dosed hourly and the patient should be monitored within 24 h. Cycloplegics are often used to help manage pain due to ciliary body inflammation.
In cases of unresolved or worsening levels of infective keratitis, other possible causes must be promptly addressed. To assess definitively, a deep corneal scrape or biopsy must be performed. Examination of the specimen after staining with Gram, Giemsa or calcofluor white provides an almost immediate assessment of likely causes. Cultures can be directly plated; however, this may take several days, if not a few weeks, to demonstrate fungal growth. In vivo confocal microscopy may also aid diagnosis and is a rapid, non-invasive means of visualizing organisms in the tissue. It may be the sole determinant in starting treatment in cases such as this when other methods are inconclusive.
Management typically begins with hourly administration of a 5% suspension of commercially available natamycin (Natacyn). Natamycin is effective against filamentous fungi such as Fusarium
solani but less effective against yeasts such as the Candida species. Other antifungals include
amphotericin B 0.15% and miconazole 1%. Topical therapy may continue for days or weeks with careful monitoring to assess the cornea's response to the given medication. Systemic therapy may be needed if the fungal keratitis is deep or has penetrated the cornea; however, in these cases a prudent assessment of liver function is necessary, both before and during therapy. Referral for surgical intervention may be necessary in severe cases or if there has been a substantial delay in treatment.
Our case, as has been reported in the literature, might have been tentatively linked to the contact lens solution (Renu MoistureLoc) the patient was using  and . A worldwide fungal
keratitis outbreak was noted in 2006 and ultimately linked primarily to Renu MoistureLoc. This solution was ultimately pulled from the marketplace; however our patient continued to use the solution even after it had been removed. He denied hearing or seeing any warnings about using the solution and at the time of the first visit was unsure of the exact solution being used. Of course, active disinfection practices should be highly recommended to all contact lens patients who use multi-purpose solutions: Rub and rinse lenses, immediately report any redness or pain, discard lenses as directed and change lens cases often. Compliance plays a large role in the risk of contracting a microbial keratitis . It is also suggested that daily disposable lenses be prescribed for infrequent wearers or for use during activities such as swimming or gardening. Although daunting, practitioners should to the best of their ability inform patients of contact lens and contact lens solution recalls. So in addition to the manufacturer's attempts at informing the public, the practitioner can use variety of venues including postcards, letters and web information to inform and educate the patient. However, despite our collective best efforts some patients will continue to be non-compliant in their contact lens care.
John Goosey, MD. This paper was supported in part by a grant from the Association of Optometric Contact Lens Educators (AOCLE) through their 2007 Annual Workshop.
 M.P. Nevitt, D.O. Hodge and D.J. Ballard, Incidence of ulcerative keratitis in a defined population from 1950 through 1988, Arch Ophthalmol 111 (12) (1993), pp. 1665–1671.
 R.K. Forster, Conrad Berens Lecture. The management of infectious keratitis as we approach the 21st century, Clao J 24 (3) (1998), pp. 175–180. View Record in Scopus | Cited By in Scopus
 L. Keay and O. Schein, Epidemiology of contact lens-related inflammation and microbial keratitis: A 20-year perspective, Eye Contact Lens 33 (6) (2007), pp. 346–353. Full Text via
CrossRef | View Record in Scopus | Cited By in Scopus (4)
 P.B. Morgan, N. Efron, E.A. Hill, M.K. Raynor, M.A. Whiting and A.B. Tullo, Incidence of keratitis of varying severity among contact lens wearers, Br J Ophthalmol 89 (2005), pp. 430–436.
Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (88)
 U.V. Jurkunas, D.P. Langston and K. Colby, Use of voriconazole in the treatment of fungal keratitis, Int Ophthalmol Clin 47 (2) (2007), pp. 47–59. View Record in Scopus | Cited By in
 O.D. Schein, L.D. Ormerod, E. Barraquer, E. Alfonso, K.M. Egan and B.G. Paton et al.,
Microbiology of contact lens-related keratitis, Cornea 8 (4) (1989), pp. 281–285. View Record in
Scopus | Cited By in Scopus (72)
 D.S. Lam, E. Houang, D.S. Fan, D. Lyon, D. Seal and E. Wong et al., Incidence and risk
factors for microbial keratitis in Hong Kong: comparison with Europe and North America, Eye 16
(5) (2002), pp. 608–618. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus
 K.R. Wilhelmus, N.M. Robinson, R.A. Font, M.B. Hamill and D.B. Jones, Fungal keratitis in contact lens wearers, Am J Ophthalmol 106 (6) (1988), pp. 708–714. View Record in Scopus |
Cited By in Scopus (66)
 W.B. Khor, T. Aung, S.M. Saw, T.Y. Wont, P.A. Tambyah and A.L. Tan et al., An outbreak
of Fusarium keratitis associated with contact lens wear in Singapore, JAMA 295 (24) (2006), pp.
2867–2873. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (76)
 D.C. Chang, G.B. Grant, K. O’Donnell, K.A. Wannemuehler, J. Noble-Wang and C.Y. Rao
et al., Multistate outbreak of Fusarium keratitis associated with use of a contact lens solution, JAMA 296 (8) (2006), pp. 953–963. Full Text via CrossRef | View Record in Scopus | Cited By in
 M.D. Bernal, N.R. Acharya, T.M. Lietman, E.C. Strauss, S.D. McLeod and D.G. Hwang et
al., Outbreak of Fusarium keratitis in soft contact lens wearers in San Francisco, Arch Ophthalmol
124 (7) (2006), pp. 1051–1053. Full Text via CrossRef | View Record in Scopus | Cited By in
 J.C. Pate, D.B. Jones and K.R. Wilhelmus, Prevalence and spectrum of bacterial co-infection during fungal keratitis, Br J Ophthalmol 90 (3) (2006), pp. 289–292. Full Text via CrossRef |
View Record in Scopus | Cited By in Scopus (6)
 CCLRU grading scales In: A.J. Phillips and L. Speedwell, Editors (4th ed.), Appendix D in
Contact Lenses, Butterworth-Heinemann, Oxford (1997).
 D.C. Ritterband, J.A. Seedor, M.K. Shah, R.S. Koplin and S.A. McCormick, Fungal keratitis at the New York eye and ear infirmary, Cornea 25 (3) (2006), pp. 264–267. Full Text via
CrossRef | View Record in Scopus | Cited By in Scopus (25)
 E. Brasnu, T. Bourcier, B. Dupas, S. Degorge, T. Rodallec and L. Laroche et al., In vivo
confocal microscopy in fungal keratitis, Br J Ophthalmol 91 (12) (2007), pp. 588–591. Full Text
via CrossRef | View Record in Scopus | Cited By in Scopus (21)
 N. Efron, Contact lens-induced changes in the anterior eye as observed in vivo with the
confocal microscope, Prog Retinal Eye Res 26 (2007), pp. 398–436. Article | PDF (4930 K) |
View Record in Scopus | Cited By in Scopus (13)
 S.C. Kaufman, D.C. Musch, M.W. Belin, E.J. Cohen, D.M. Meisler and W.J. Reinhart et al.,
Confocal microscopy: a report by the American Academy of Ophthalmology, Ophthalmology 111
(2) (2004), pp. 396–406. Article | PDF (427 K) | View Record in Scopus | Cited By in Scopus (74)
 Parmer DN, Awwad ST, Petroll WM, Bowman RW, McCulley JP, Cavanagh HD, et al.
Tandem scanning confocal corneal microscopy in the diagnosis of suspected Acanthamoeba
keratitis. Ophthalmology 113:538–547.
 W.M. Munir, S.I. Rosenfeld, I. Udell, D. Miller, C.L. Karp and E.C. Alfonso et al., Clinical
response of contact lens-associated fungal keratitis to topical fluoroquinolone therapy, Cornea 26
(6) (2007), pp. 621–624. View Record in Scopus | Cited By in Scopus (6)
 E.C. Alfonso, J. Cantu-Dibildox, W.M. Munir, D. Miller, T.P. O’Brien and C.L. Karp et al.,
Insurgence of Fusarium keratitis with contact lens wear, Arch Ophthalmol 124 (2006), pp.
941–947. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (56)