Introduction

Fusarium species may form reproductive structures in tissue, a phenomenon known as `adventitious sporulation’ [1]. If Fusarium-specific adventitious sporulation (phialides and phialoconidia morphologically suggestive of Fusarium species, and irregular hyphae with both 45° and 90º branching) is seen in a tissue lesion, a preliminary diagnosis of Fusarium infection can be made [2, 3], and specific therapy can be rapidly initiated while waiting for the culture results.

Fusarium species, the principal causes of fungal keratitis in many countries [4], are typically relatively resistant to most antifungal agents. Natamycin is currently the drug of choice to treat fusarial keratitis [5, 6], while topical and oral voriconazole may also be effective [7]. If keratitis due to Fusarium species is diagnosed rapidly, specific therapy can be started promptly.

This report describes how the phenomenon of Fusarium–specific adventitious sporulation in corneal scrapings allowed a rapid presumptive diagnosis of Fusarium keratitis in six patients.

Case report

Between July and December 2005, each of 152 patients with suspected microbial keratitis underwent scraping of the base and edges of the ulcerated corneal lesion [4]. The corneal material was stained by lactophenol cotton blue (LPCB wet film) and Gram’s stain (dry smear) for direct microscopic examination; fusoid macroconidia with foot cells were specially looked for, since this is an important feature of the genus Fusarium [8]. Fusarium-specific adventitious sporulation was said to be present in corneal scrapings exhibiting such reproductive structures. Corneal material was also inoculated as `C-streaks' onto sheep blood agar and Sabouraud glucose neopeptone agar and broth; the media were incubated at 30°C and 37°C. Fungi isolated were deemed significant if isolated on multiple media or on one medium with direct microscopic evidence of fungal hyphae. Filamentous fungi were identified according to standard procedures.

During the study period, a diagnosis of mycotic keratitis was established by isolation of fungi alone in culture in 49 patients (32.3 % of 152 patients). Fusarium species (20 strains) were isolated from 20 patients [Fusarium solani {13}, Fusarium dimerum {three}, undefined Fusarium species{four}].

Fusarium-specific adventitious sporulation (Fig. 1) was detected in the corneal scrapings of six patients. Five of these were males; all six were adults (age range: 30 to 52 years). Three patients had had the symptoms for 3 days or less, while the other three had had symptoms for 6 days or more. The principal risk factor was trauma in four patients (due to vegetable material in one, mud or dust in two and an unknown agent in one), while the other two patients did not report any specific risk factor. The right eye was affected in five patients, and the left in one patient. Two ulcers had a smallest diameter of 5 mm, while the smallest diameter of each of the other four ulcers was 3 mm or less. Four patients had good vision (20/60 or better) in the affected eye, while two patients had poor vision (20/200 or worse).

Fig. 1
figure 1

a Lactophenol cotton blue (LPCB) wet film of corneal scrape material showing Fusarium-specific adventitious sporulation (×400);Fusarium solani was isolated in culture of the corneal material. b Gram-stained smear of corneal scrape material exhibiting Fusarium-specific adventitious sporulation (×400); Fusarium solani was isolated in culture of the corneal material

Fungal hyphae were detected by Gram’s stain in corneal scrapings of all six patients, and in the LPCB wet films of corneal scrapings of five patients. Since Fusarium-specific adventitious sporulation was noted in the corneal scrapings, hourly topical natamycin (5%) was commenced without waiting for the culture results. With this therapy, the corneal ulceration completely healed in three patients, and was healing in one patient when last reviewed. One patient was lost to follow-up before the status could be evaluated, while in one patient, the ulcer was still active when last seen. Interestingly, Fusarium solani was isolated from the corneal scrapings of all six patients.

Discussion

Adventitious sporulation is known to occur in the course of infections due to Fusarium, Paecilomyces or Acremonium [2]. The exact significance of this phenomenon is yet to be established [9]. However, it can be used for a rapid presumptive diagnosis of certain fungal infections, although definitive identification requires culture. Liu et al. [2] reported adventitious sporulation in histological sections of nine of nine infections caused by Paecilomyces species, and in seven of ten caused by Fusarium species.

Direct microscopy is an important diagnostic modality in investigating suspected microbial keratitis, since the detection of fungal hyphae or yeast cells in corneal scrapings permits a rapid presumptive diagnosis of fungal keratitis and initiation of antifungal therapy. However, it is difficult to determine which genus of filamentous fungi is involved just by looking at the fungal hyphae in a corneal scraping. This has important implications for treatment. If specific anti-Fusarium therapy could be confidently initiated based on the findings in the corneal scraping, it would greatly benefit the clinician. Hence, we tried to assess the significance of Fusarium-specific adventitious sporulation in corneal scrapings as a diagnostic aid for presumed Fusarium keratitis.

We observed that in relation to a culture-proven diagnosis of Fusarium infection (the ‘gold’ standard), the occurrence of Fusarium-specific adventitious sporulation had a sensitivity of 30 percent, a specificity of 100 percent, a false-positive rate of 0 percent and a false-negative rate of 9 percent (Table 1). The results of direct microscopy (presence of Fusarium-specific adventitious sporulation suggesting Fusarium keratitis, and absence of such sporulation excluding Fusarium keratitis) were in agreement with the results of culture in 138 (91%) of the 152 patients investigated.

Table 1 Presence of Fusarium-specific adventitious sporulation as a diagnostic aid for Fusarium infection in patients with keratitis

Our results suggest that if Fusarium-specific adventitious sporulation is observed in corneal scrapings from a patient with keratitis, a relatively reliable diagnosis of presumed Fusarium keratitis can be made, and topical eyedrops active against Fusarium species (e.g., natamycin, econazole, voriconazole) can be rapidly instituted while awaiting the culture results. The low sensitivity of this technique is currently a concern, but improvement is possible with refinements in the techniques used.