Fungal keratitis is one of the most challenging types of infectious keratitis [
The ultimate goal of fungal keratitis treatment is to preserve or improve visual acuity [
Fifty-four eyes of 54 consecutive patients admitted to our hospital were analyzed retrospectively. These patients were treated with LKIIF from January 2010 to April 2013 at the Shandong Eye Hospital. The ages of the patients ranged from 21 to 73 years, and the mean age was 46.5 years. Thirty-three patients were men and 21 were women. These patients had no other systemic diseases, except for six patients with diabetes mellitus and four patients with hypertension. No patients had a history of bacterial infections, viral infections, amebic infections, or previous corneal scars. The time from the appearance of symptoms to the first visit was 3−60 days with a mean time of 13.5 days. The best-corrected visual acuity (BCVA) was finger count (FC) to 20/30 (
Information about the cornea lesions in our recruited patients is summarized as follows. The maximum diameters of the lesions including ulcers and infiltrations were from 2.0 to 5.0 mm (
All 54 cases received 0.5% fluconazole drops every 15 min and 0.25% amphotericin B or 5% natamycin drops every hour. All drops were used every day until 22:00. Ointment of 1% fluconazole or 1% amphotericin B was applied to the cornea at bedtime. At the same time, all patients received oral itraconazole (0.2 g daily) or an intravenous injection of fluconazole (the first dose was 400 mg followed by 200 mg daily). Two hours after operation, local chemotherapy started again and then gradually reduced, depending on the wound condition.
All operations were performed under local anesthesia. First, the conjunctival sac and corneal surface were washed with 0.2% fluconazole and the necrotic tissues on the ulcer surface were removed with a corneal epithelium knife. An oblique incision was made 0.5–1.0 mm away from the edge of the lesion. The diseased corneal tissues were gradually moved to the stroma with a blade tilted at 45°. The infected lamellae were pulled up with 0.12 mm forceps, which maintained tension. The blade was held parallel to the stroma to remove the white fiber connecting the lamellae and the stroma. A gentle transition was made at the edge of lesion to the clear area. Next, we washed the wound and inspected the infiltration with an operating microscope. The same procedure was repeated until no residual gray infiltration could be seen. At this point, the lesion including the ulcer and infiltration tissues were completely removed. Corneal scissors were used to trim the edge to make a smooth wound edge. An insulin syringe (1 mL, 30-gauge needle, Becton Dickinson Company, Franklin Lakes, New Jersey, USA) was used for intrastromal injections. The injection was administered from the clear cornea of the lesion edge. The needle was advanced towards the lesion center in parallel with the stroma. When the bevel of the needle was completely in the cornea, 0.2% fluconazole solution was slowly pushed into the stroma until the inflated stroma exceeded the edge of the lesion by more than 0.5 mm. The needle was quickly removed when the injection was completed. This procedure is shown in Figures
(a) When removing the ulcer, the knife should be kept in parallel with the cornea surface to cut the white fiber between the lesion and the residual cornea. (b) The margin is a slope and the surface is smooth after keratectomy. The injection is from the clear cornea of the lesion edge. The needle is inserted toward the lesion center in parallel with the stroma. The bevel of the needle is in the corneal stroma completely.
Surgery and intrastromal injection. (a) The corneal flap is held tense with the aid of toothed forceps, and the filamentous white collagen fiber could be seen clearly. (b) Corneal scissors were used to trim the edge to make the edge of wound smooth. (c) The needle was advanced towards the lesion center in parallel with the stroma. (d) Liquid was slowly pushed into the stroma until the inflated stroma exceeded the edge of the lesion by more than 0.5 mm.
The corneal infection, healing time of the ulcer, and visual acuity outcomes were observed and recorded. The scars were measured using AS-OCT. Corneal endothelial cells were observed with a corneal endothelial microscope. Intraocular pressure was measured with a Tono-Pen tonometer. Fungal cultures were made from the corneal scraping. If the fungal culture was negative before surgery, another culture was prepared with corneal tissue. Fungal species identification was performed if the fungi culture was positive [
Application of the LFIIK treatment resulted in 46 of the 54 eyes being cured (85.2%) (Figures
The lesion was located in the peripheral cornea. (a) Before surgery. (b) The first day after surgery. (c) The wound healed 10 days after surgery.
The lesion was located in the paracentral cornea. (a) Before surgery. (b) The first day after surgery. (c) The wound healed 7 days after surgery.
The lesion was located in the central cornea. (a) Before surgery. (b) The first day after surgery. (c) The wound healed 12 days after surgery.
(a) Fungal keratitis with hypopyon. (b) Seven days after antifungal chemotherapy, the hypopyon disappeared but the ulcer had not healed yet. (c) The first day after surgery. (d) The wound healed 7 days after surgery. (e) An OCT image of the lesion before surgery. (f) An OCT image of the lesion when followed up after more than 90 days.
We next checked the visual acuities of the cured eyes, and the results are shown in Table
The numbers distribution of BCVA and UCVA (BCVA, UCVA) in 46 cured cases.
Visual acuity | Preoperation | When ulcers were cured | 90 days after surgery |
---|---|---|---|
<20/400 | 11, 4 | 0, 0 | 0, 0 |
≥20/400 |
2, 11 | 1, 0 | 1, 0 |
≥20/200 |
27, 22 | 21, 20 | 16, 13 |
≥20/40 |
6, 9 | 21, 23 | 24, 27 |
≥20/25 |
0, 0 | 3, 3 | 5, 6 |
Clinical characteristics of 8 failed cases.
Patient | The second |
The max diameter |
The max depth |
Fungus |
UCVA† | BCVA† |
---|---|---|---|---|---|---|
1 | LKT + CF | 4.0, 4.0 | 41.5, 37.3 |
|
20/60, |
20/60, |
2 | LKT + CF | 3.5, 3.5 | 30.6, 53.7 |
|
FC/BE, |
FC/BE, |
3 | LKT + CF | 4.0, 4.5 | 49.2, 50.0 | Negative | FC/1.5 m, |
FC/1.5 m, |
4 | LKT + CF | 5.0, 5.0 | 46.5,49.5 | Negative | 20/125, |
20/125, |
5 | LKT + CF | 4.5, 5.0 | 43.5, 40.5 |
|
20/100, |
20/100, |
6 | LKP | 4.0, 5.0 | 45.0, 70.0 |
|
20/200, |
20/200, |
7 | LKP | 5.0, 5.5 | 40.2, 63.0 |
|
20/200, |
20/100, |
8 | PKP | 4.5, 5.0 | 39.4, 92.6 |
|
FC/1 m, |
FC/1 m, |
†BCVA values are before the first surgery and 90 days after the second surgery and when followed up for 1 year in the keratoplasty cases (first, second).
LKT + CF, lamellar keratectomy and inlay conjunctival flap; LKP, lamellar keratoplasty; PKP, penetrating keratoplasty; FC, finger count; BE, before eye.
Correlation between preoperative and postoperative UCVA when the wound healed, most of the cases improved their UCVA than preoperation.
When followed up for more than 90 days, the depth of the residual cornea scar was 360.5–427.4
Corneal scrapings were positive in 46 cases (85.2%). Fungal cultures were positive in 42 cases (77.8%). The results of the species identifications are shown in Table
Fungal species identified in the 38 positive cases in fungal culture.
Fungus | Number of cases | % |
---|---|---|
Fusarium genu |
23 | 54.8 |
|
7 | 16.7 |
|
8 | 19.0 |
|
1 | 2.4 |
|
3 | 7.1 |
|
2 | 4.8 |
|
1 | 2.4 |
Unspecified | 1 | 2.4 |
|
9 | 21.4 |
|
1 | 2.4 |
|
1 | 2.4 |
|
7 | 16.7 |
|
2 | 4.8 |
|
3 | 7.1 |
|
2 | 4.8 |
|
1 | 2.4 |
|
1 | 2.4 |
Total | 42 | 100 |
Clinical characteristics of all 5 hypopyon cases.
Patient | Hypopyon height (mm) | Max diameter of lesion (mm) | Max depth of lesion (%) | Healing time (days) | Fungus identified | UCVA |
BCVA |
---|---|---|---|---|---|---|---|
1 | 1 | 3.5 | 38.0 | 8 |
|
20/200, |
20/200, 20/40 |
2 | 4 | 2.5 | 37.0 | 9 | Negative | FC/20 cm, |
FC/20 cm, |
3 | 1 | 3.0 | 28.5 | 7 |
|
20/160, |
20/160, |
4 | 3 | 5.0 | 22.3 | 3 |
|
FC/BE, |
FC/BE, |
5 | 1 | 5.0 | 45.0 | 9 |
|
FC/BE, |
FC/BE, |
Neovascularization was observed in one eye. The cornea in the lesioned area became thinner but no perforation or other complications were found.
Surgeries for fungal keratitis include debridement, inlay conjunctival flaps, lamellar keratectomy, and corneal transplantation [
The effect of lamellar keratoplasty for fungal keratitis was consistent with the observation that fungi were not always present throughout the whole cornea [
When the ulcers were cured, the refraction and vision stabilized quickly. The healing time of the wound in our study was 3–16 days, and, in 28 cases, it was less than 7 days (51.9%). In contrast, chemotherapy alone lasted 7 days. UCVA and BCVA in three patients with small graft keratoplasty were lower than the average level of the cured cases in the follow-up. Even if the visual acuity was improved later, it needed more time to improve (Table
There are a number of factors affecting vision and refraction after the eye is cured. In theory, keratectomy in the central cornea changes the spherical lens, and peripheral excision changes the cylindrical lens (Figure
(a) Schematic diagram of the refraction change in an emmetropic eye when a keratectomy is performed in the central cornea. (b) Schematic diagram of the refraction change in an emmetropic eye when a keratectomy is performed in the peripheral cornea.
To cure hypopyon, preoperative antifungal chemotherapy should be used to effectively control the inflammation. Hypopyon is one of the most significant risk factors involving failure [
We believe it is important to carefully inspect the cornea to accurately measure the depth of the lesion before surgery. Use of AS-OCT is increasing in surgical planning and evaluation [
Except for neovascularization in one eye, no perforation or other complications were found. Perforation occurred in some cases, excised to not more than half of the corresponding corneal full thickness in PTK, and was considered to be related to the vacuum aspiration during the surgery [
The possible reasons for failure may be attributed to hyphal growth patterns in corneas. In the 54 cases, the pathogenic fungi were
Our results described a treatment that was effective for the treatment of fungal keratitis not healed or aggravated after general and local fungal chemotherapy. The maximum diameter of lesions, including ulcers and infiltrations, should not be less than 5 mm or not more than 3 mm if the lesion is located in the central cornea. The maximum depth of the lesion should be not more than half of the corresponding corneal full thickness, as determined by slit lamp and AS-OCT, and not more than 40% if the lesion is located in the central cornea. The lesion should not have corneal limbus involvement. In cases of hypopyon, inflammation should be controlled by drugs to quickly remove hypopyon. Surgery was more suitable for fungi growing horizontally in the cornea, and caution should be taken if the fungi grow vertically. The approach described in the present study recovered valuable vision with less damage and fewer complications and needed no donors or special equipment, making it an effective and efficient treatment for fungal keratitis.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors thank Professor Dongsheng Yang of the Department of Ophthalmology, Liaocheng People’s Hospital, for the extraordinary kindness and help with the paper.