Fungal keratitis (FK) is a serious eye condition caused by a variety of fungi (Ghosh et al., 2016). It can be caused by a variety of molds and yeasts, and the etiology and epidemiology of the condition may depend on geography and climatic conditions (Kredics et al., 2015; Ho et al., 2016). There are approximately 70 genera of fungi implicated in FK, but Fusarium and Aspergillus species account for 70% of the cases (Bharathi et al., 2003; Ahmadikia et al., 2021). While FK is relatively uncommon among developed countries, it is a major problem for ophthalmologist in developing countries that have tropical or subtropical climates (Kredics et al., 2015). The incidence of FK varies around the world. For instance, in Paraguay, fungi cause 50.06% of all microbial keratitis, while in Ireland, they cause 1.1% (Ahmadikia et al., 2021). FK usually results from trauma during cultivation or harvesting, so soil and plant-related workers are at a great risk (Kredics et al., 2015; Shah et al., 2017). Additionally, wide-spread use of broad-spectrum antibiotics and steroids, corneal surface diseases, frequent contact lens use or prior corneal surgery may all contribute to FK (Ng et al., 2013). The treatment of FK depends on the etiological agents, the extent of involvement, risk factors, and response to treatment of the causative agents (Puig et al., 2020). Delayed diagnosis, the limited number of available antifungal drugs, and drug resistance of the causative agents are contributing factors to treatment failure and blindness in some infected individuals (Henry et al., 2012; Al-Hatmi et al., 2014). An effective strategy for corneal ulcer management requires a thorough understanding of the local etiology of a given region. A few studies have reported the epidemiology of corneal ulceration, causative micro-organisms, and effective treatment in Iran (Mahmoudi et al., 2018; Ahmadikia et al., 2021). To provide comprehensive and valid background information and to facilitate future treatment of this infection, we conducted the present study to determine the proportion of FK among infective keratitis, risk factors, clinical outcomes, causative organisms and their drug sensitivity in the main referral center for eye disease in Tehran, Iran.

Of the 1189 patients with infective corneal ulcers, 86 (7.23%) were diagnosed as FK. Of the 86 patients, 57 (66.2%) were male, and 29 (33.8%) were female (P< 0.05). The male-to-female ratio was1.9:1 and the age of patients ranged from 22 to 103 years (mean; 52.9). The most frequent age group was 25-50 (41; 47.6%). The majority of patients (32/86; 37.2%) were farm workers. Trauma was noted as the predominant predisposing factor in 42 (48.8%) patients. Plant debris and stalks were the most common traumatic agents (21/86; 24.4%). Seasonal variation showed that autumn had a higher proportion of cases compared to other seasons. (34/86; 39.5%) (P<0.05). Table 1 presents the demographic characteristics, predisposing factors, and seasonal distribution.

Keratitis is the second leading cause of blindness worldwide and the most common cause of vision disorders and complications (Bharathi et al., 2003; Gupta et al., 2014). FK is a serious ophthalmological condition in all parts of the world. In the case of delayed diagnoses and improper treatment, FK can cause severe corneal damage and even blindness. Based on various published reports, FK accounts for 6 to 50% of ulcerative keratitis cases (Thomas, 2003; Srinivasan, 2004; Shokouhi et al., 2006). Although the disease is found worldwide, it is most prevalent in tropical and subtropical regions, especially among farmers (Shukla et al., 2008). The results of this study indicated that 86 (7.23%) of 1189 patients with suspected infectious keratitis were diagnosed with FK. According to the studies conducted in India, Egypt and Saudi Arabia FK accounted for 34.4%, 43.3% and 3.8% of microbial keratitis, respectively (Saha and Das, 2006; Alkatan et al., 2012; Khater et al., 2014). The incidence of FK can vary depending on the region of a country; for example, it was 32%, 38.9%, 32%, and 39.8% in North, East, West, and South India, respectively (Chowdhary and Singh, 2005). Studies conducted in Iran have found that the incidence of FK varies from region to region. In Sari (northern Iran), Shokouhi et al. (2006) reported an incidence of FK of 31.8% (Shokouhi et al., 2006), whereas another study from Tehran (2012) reported an incidence of 5.5% (Ebadollahi-Natanzi et al., 2016). Similarly, in Rasht in north of Iran (2017), the incidence of FK was reported to be 16% (Tighnavard et al., 2017). Houang et al. (2001) explored the relationship between FK and climate, they concluded that although higher levels of fungal keratitis could be expected in regions with more rainfall and temperatures, this was not always the case and depended on other factors (Houang et al., 2001). There are few cases of FK without associated predisposing factors (Mahmoudi et al., 2018). Some of the predisposing factors for FK include trauma, immunodeficiency, ocular surface disorders, ocular surgery, treatment with topical steroids, and long-term use of soft contact lenses (Cheikhrouhou et al., 2014). According to different studies, corneal trauma (primarily vegetative matter) is the predominant cause of FK in 40-60% of cases (Mahmoudi et al., 2018). The results of our study revealed that corneal trauma was the most important factor predisposing patients to FK (42/86, 48.8%) followed by eye surgery (10/86, 11.6%). Additionally, we found that plant organs were the most important traumatic factor among our patients. There have been similar findings reported in Melbourne, the southern United States, Singapore, India, and Bangladesh, as well as in previous studies in Iran (Shokouhi et al., 2006; Cheikhrouhou et al., 2014; Ebadollahi-Natanzi et al., 2016; Tighnavard et al., 2017; Lee et al., 2018; Mahmoudi et al., 2018). In contrast, eye trauma was reported as the second risk factor in the northern United States (Keay et al., 2011). Additionally, a study in Philadelphia noted that chronic eye disease, the habit of wearing contact lenses, and the use of topical corticosteroids are three common underlying factors (Tanure et al., 2000). A significant increase in FK occurs in men from rural areas following corneal damage (Bharathi et al., 2003). In this study, the male to female ratio was 1.9: 1 (male: 57/86, 66.2%). Furthermore, most FK patients (41/86, 47.6%) were aged 25 to 50 and were farmers. Different studies report higher incidences of FK in the 35-59 age group and among farmers, as well as a higher ratio of males to females (Saha and Das, 2006; Chander et al., 2008; Satpathy et al., 2019). Since this age group is more active outdoors and is in contact with soil and nature, they are more exposed to fungal agents (Cheikhrouhou et al., 2014). Similar findings have been observed in other studies conducted in Iran (Shokouhi et al., 2006; Ebadollahi-Natanzi et al., 2016; Tabatabaei et al., 2018). There is a higher prevalence of FK in the autumn and during the year when more agricultural activities occur (Bharathi et al., 2003). The highest proportion of cases in each season was noted in autumn (34/86, 39.5%) in our study. Several studies have demonstrated a significant increase in the number of reported cases of FK during harvest season and the monsoon winds (Gopinathan et al., 2002; Bharathi et al., 2003; Mahmoudi et al., 2018). Despite its widespread incidence, FK has no consistent clinical picture (Fong et al., 2004; Dahlgren et al., 2007). Clinical signs that can assist with diagnosing FK include severe hypopyon, lesions with feathery margins, corneal perforation, and satellite lesions (Mahmoudi et al., 2018; Satpathy et al., 2019). We found hypopyon, feathery margins, and satellite lesions on the slit-lamp examination in 56 (65.11%), 59 (68.6%) and 17 (19.7%) of our patients, respectively. Furthermore, the corneal ulcer diameter ranged from 1 to 11 mm (mean: 4.05 mm). As reported by Bharati et al., feathery margins were found in 786 (71.78%), satellite lesions in 110 (10.05%), and hypopyon in 609 (55.62%) cases of FK (Bharathi et al., 2003). Rosa et al. observed feathery margins in 62% of their patients as well as a rough texture in 47%, and satellite lesions in 41% (Rosa et al., 1994). In contrast, Shokohi et al. observed only one feathery pattern and did not observe hypopyon and satellite lesions among their FK patients (Shokouhi et al., 2006). As described above, these findings are not pathognomonic, and can lead to misdiagnosis and treatment with antiviral, antibacterial or corticosteroid drugs (Satpathy et al., 2019).

In most cases, FK is treated by topical administration of anti-fungal medications, but there are no gold standards (Iselin et al., 2017). Because of the limited availability of antifungal drugs and the emergence of resistant species, the treatment of FK has become a major challenge. Without proper treatment, it can lead to vision loss in cases of deep lesions (Sharma et al., 2011). Natamycin was previously an effective drug for treating FK, however, poor penetration of this drug into the corneal stroma has led to reports of treatment failure for this drug. As an alternative, topical amphotericin B 0.3% to 0.5% and voriconazole 1% are recommended (Sharma et al., 2011; Sharma et al., 2016). In the present study, 62.8% of patients received a combination of voriconazole, itraconazole, and amphotericin B. In 32 out of 86 patients, drug treatment did not improve the condition, so penetrating keratoplasty was employed. Among the patients who received antifungal and surgical treatment, 11 (3%) developed recurrent FK. A combination of systemic voriconazole and topical natamycin was found to be one of the most commonly recommended antifungal treatment regimens by Iselin et al. (Iselin et al., 2017). A report by Ghosh et al. noted that 95% of patients with FK were treated with either topical preparations of natamycin alone or in combination with other topical antifungals, including itraconazole, fluconazole, and voriconazole (Ghosh et al., 2016) . Tanure et al. used penetrating keratoplasty on six patients with acute fungal keratitis while they were being treated with combination of natamycin and amphotericin B (Tanure et al., 2000). According to previous studies, keratoplasty is required in 26 to 35% of cases of acute FK (Sanders, 1970). An accurate diagnosis of the causative agents for fungal infections is crucial for proper treatment. Improvements in diagnostic methods are essential for this (Fong et al., 2004). FK can be caused by yeasts or molds (Bharathi et al., 2003). FK due to molds is more common in rural areas among agricultural workers and in urban areas among construction workers. Yeast keratitis, on the other hand, usually affects patients with chronic systemic diseases, viral eye infections, and contact lens wearers (Satpathy et al., 2019). We found 72 filamentous fungi (83.7%) and 14 yeast cases (16.3%) in our study. As reported in most studies, Fusarium spp., Aspergillus spp., and Candida spp. are the most common pathogens associated with FK (Ahmadikia et al., 2021). In our study, Fusarium spp. were the most frequent isolates (34/86; 39.5%), followed by Aspergillus spp. (28/86; 32.5%) and Candida spp. (14/86; 16.2%). The most common cause of FK in India, Nepal, Sri Lanka, and Bangladesh has consistently been identified as Aspergillus spp. (Mahmoudi et al., 2018). Fusarium spp. were the most common causative agents in Africa and China (Xie et al., 2006; Thomas and Kaliamurthy, 2013). In European countries, FK is rare and Candida spp. are the most common isolates (Ahmadikia et al., 2021). There are other Iranian studies that report Aspergillus spp. and Fusarium spp. being the most common fungal pathogens in FK (Shokouhi et al., 2006; Ebadollahi-Natanzi et al., 2016; Tighnavard et al., 2017). There is a lack of information about the abundance of fungal species known to cause fungal keratitis in Iran due to the limited number of molecular identifications carried out in previous studies. In this study, PCR-based sequencing revealed that F. solani (26/86; 30.2%) and A. flavus (21/86; 24.4%) were the most frequently identified species. Additionally, MALDI-TOF results showed that the most frequent Candida species was C. albicans (10/86; 13%). Although in vitro antifungal susceptibility testing cannot predict clinical FK response, MIC data against different fungal isolates can help ophthalmologists determine the most suitable treatment (Kredics et al., 2015). Due to the wide spectrum of fungi causing FK, varying susceptibility patterns are not surprising (Mahmoudi et al., 2018). The MIC results in this study showed that amphotericin B had the lowest MIC₉₀ against Fusarium species (n=34), followed by voriconazole, posaconazole and natamycin. There were no significant differences in antifungal sensitivity between Fusarium species. In our study, we found posaconazole had the lowest MIC₉₀ against Aspergillus species (n=28), followed by caspofungin, voriconazole, and amphotericin B. Posaconazole had the lowest MIC₉₀ when tested on Candida species (n=14), subsequently followed by miconazole, voriconazole, flucytosine, fluconazole, amphotericin B and caspofungin. Compared to Fusarium species, Aspergillus and Candida species were inhibited at low MIC values by azoles. As expected, these results support a prior study on in vitro susceptibility of fungal isolates (Lalitha et al., 2007). Against S. apiospermum (n=4), voriconazole and miconazole had the lowest MIC₉₀ followed by posaconazole, caspofungin, natamycin, and amphotericin B. Natamycin had the lowest MIC₉₀ among the drugs tested on dematiaceous fungi, followed by posaconazole, caspofungin, voriconazole, amphotericin B, and miconazole. Results of our study are similar to those found by Xie et al. (Xie et al., 2008) and Al-Hatmi et al. (Al-Hatmi et al., 2016) regarding in vitro susceptibilities of isolates from keratitis. Contrary to our findings, O’day et al. (O'day et al., 1986) stated that amphotericin B was ineffective against Fusarium species (139). According to Hassan et al. (Hassan et al., 2015) amphotericin B, voriconazole, fluconazole, and natamycin proved to be potential antifungal agents for treating human Fusarium keratitis. Though susceptibility testing is more popular for bacterial than fungal disease, fungal susceptibility testing is becoming more credible in the literature. According to several studies, infections caused by susceptible isolates respond to therapy better than infections caused by resistant isolates (Lalitha et al., 2007). Therefore, drug susceptibility tests in vitro can provide valuable information on the proper treatment of fungal infections, particularly for specific antifungal drugs that interact significantly with fungi.

It is well-known that corneal damage is a major risk factor in FK. The most at risk group is young and middle-aged farmers due to the increased exposure to corneal damage caused by contaminated crops. Using a diagnostic method that is highly sensitive and specific can enable early initiation of antifungal therapy to allow for a complete recovery within a short period of time. The MIC results indicate that amphotericin B may be appropriate for treating FK caused by Fusarium species. Amphotericin B, voriconazole, posaconazole, and caspofungin may be good options in cases of Aspergillus keratitis. FK caused by Candida spp. can be treated with flucytosine, voriconazole, posaconazole, miconazole, and caspofungin and S. apiospermum can be treated with voriconazole and caspofungin. Keratitis caused by the dematiaceous fungi may require the use of antifungals such as amphotericin B, posaconazole, and caspofungin. Research that investigates the relationship between in vitro susceptibilities and in vivo clinical outcomes will be important for the treatment of fungal keratitis.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

This study was approved by the ethics committee of the Tehran University of Medical Sciences (IR.TUMS.MEDICINE.REC.1400.304). The patients/participants provided their written informed consent to participate in this study.

MS, Az, SK, FA, ZA, AM, MA, and SR collected the data and drafted the manuscript. S.K, MS reviewed the manuscript from the clinical point of view. SK supervised the work and reviewed the manuscript from the microbiological point of view. CO performed susceptibility testing. MT-K performed the molecular identification of fungal strain. RD, SH, FA, and SA reviewed the manuscript from the microbiological point of view. MA and ZA helped with collecting fungal strains and the data. SM and SK reviewed the manuscript from the microbiological point of view and helped with editing the manuscript. SR and MA helped with collecting the data. PV and SK reviewed the manuscript from the microbiological point of view and helped with editing the manuscript. All authors contributed to the article and approved the submitted version.