Fusarioid keratitis and other superficial infections: A 10-years prospective study from Northeastern Brazil

Background Fusarium and allied genera (fusarioid) species are common colonizers of roots and aerial plant parts, or act as phytopathogens in forestry and horticultural or grain crops. However, they can also cause a wide range of infections in humans, including onychomycosis, cutaneous and invasive infections. Fusarioid keratitis is characterized by an infection of the cornea with a suppurative and ulcerative appearance, which may cause damage to vision and permanent blindness. The aim of the present study was to investigate the prevalence of fusarioid species, biofilm formation and antifungal susceptibility profiling of clinical isolates recovered from patients with keratitis and dermatomycoses. Methodology/Principal findings The study was performed between March, 2012-December, 2022. Demographic, clinical and epidemiological data of patients were also collected. In the present study, most of the patients with keratitis were male (74%), had a median age of 42 years old, worked with plant material or debris and 26% of them reported eye trauma. Regarding dermatomycosis, most of patients were female and exhibited toenail lesions. Forty-seven isolates belonged to the genus Neocosmospora (78.33%), nine to the Fusarium fujikuroi (15%) and four to the Fusarium oxysporum (6.66%) species complexes. Several strains were moderate biofilm producers, specifically among Fusarium annulatum. Most strains showed increased MICs to amphotericin B and ketoconazole and low MICs to itraconazole. MICs ranged from 0.25 to 16 μg/mL for amphotericin B, 0.0625 to >16 μg/mL for ketoconazole and 0.125 to 8 for itraconazole. Conclusions/Significance It is possible to conclude that fusarioid keratitis in Northeastern Brazil is an important and neglected disease, given the high number of cases, increased need for keratoplasty and poor outcome of the disease.


Introduction
Fusariosis is an infectious disease of worldwide distribution, caused by Fusarium spp.and related genera (fusarioid fungi) [1].The term "fusarioid" fungi refers to fungi not belonging to the Fusarium genera, but possess fusarium-like macroconidia [2].In fact, Fusarium is able to cause diseases in different living beings, including humans, other animals and plants.Therefore, these fungi have been recognized as "trans-kingdom" pathogens [3,4].In humans, fusariosis may range from superficial (including keratitis, nail and skin infections) to lifethreatening disseminated diseases, specifically among immunocompromised patients (neutropenic, leukemic patients, those under intense chemo or corticosteroid therapy, and transplant recipients) [5][6][7][8].
Keratitis is one of the most frequent diseases caused by fusarioid fungi and it is mostly related to trauma with plant debris or other organic matters, contact lens wear, ocular diseases and surgeries, mainly associated with the use of steroids [9,10].Clinical signs of fusarioid keratitis may involve a grayish-white ulcer accompanied by peripheral feathery corneal infiltrate, endothelial plaque formation and hypopyon [11].Fusarioid keratitis may progress to endophthalmitis or even cause corneal perforation and descemetocele, and in several occasions corneal transplantations and other surgeries are required [11][12][13][14].
Onychomycosis and other dermatomycosis represent important clinical entities caused by fusarioid spp.Although not considered life-threatening diseases, they may cause embarrassment and working-related challenges [15,16].Onychomycoses due to fusarioid spp.almost always occur on toenails, especially after traumatic and dystrophic abnormalities or even subsequent to dermatophyte onychomycosis [17].Risk factors for fusarioid onychomycosis may involve traumatic injury, diabetes mellitus, arterial hypertension, HIV and autoimmune illness [18].It is worth mentioning that in case of immunosuppression, nail and other skin infections caused by fusarioid spp.may become disseminated [19].
It has been previously reported that fusarioid spp.may produce biofilms, and this structure has been associated with contact lenses keratitis outbreaks, most often related to the contamination of lens solution [20][21][22].Biofilm formation in Neocosmospora (formerly the Fusarium solani species complex-FSSC) may vary among different strains, but usually consists of conidia, and a highly coordinated network of hyphal structures and an extracellular matrix (ECM) [21][22][23].It is important to emphasize that sessile cells of biofilm-forming Fusarium strains exhibit elevated minimum inhibitory concentrations (MICs) for amphotericin B (AMB), voriconazole (VRC), and posaconazole, when compared to their planktonic counterparts [24].
Fusarioid spp.may be intrinsically resistant to a few antifungal drugs used in the clinical practice [8].Nevertheless, it is important to emphasize that no clinical breakpoints have yet been established for them.On the other hand, epidemiological cutoff values (ECVs) have recently been proposed for some fusarioid species complexes [25].
The current treatment of choice for fusarioid keratitis is the use of eye drops containing 5% of natamycin (NAT).Topic AMB (0.3 to 0.5%) can also be successfully used, but it has the disadvantage of not being largely commercialized [26].Onychomycosis caused by fusarioid spp.may be treated with itraconazole (ITC) or sometimes with terbinafine (systemic) in combination with topic ciclopirox or amorolfine nail lacquers [8].
Therefore, the aims of this study were to accurately identify and study genetic relatedness of 60 isolates of fusarioid spp.obtained from cases of superficial infections (keratitis and dermatomycosis) in the Northeast region of Brazil.We also evaluated biofilm formation and performed antifungal susceptibility testing for these strains.

Ethics statement
All clinical and demographic data of the patients were collected in accordance with the Local Research Ethics Committee, from the Onofre Lopes University Hospital, Federal University of Rio Grande do Norte, approved under number 3.769.085.The written consent was waived because of data anonymization and commitment to preserve the identity of the patients and only secondary data obtained from medical records were used.

Patients demographic and clinical data
A prospective study was performed with 50 patients admitted at the Sector of Ophthalmology, Onofre Lopes University Hospital, Rio Grande do Norte state, Northeast Brazil, between March, 2012 and December, 2022.The patients enrolled in this study belonged to cases of suspected keratitis.In addition, 10 patients with suspected dermatomycoses were forwarded by dermatologists for mycological examination at the Laboratory of Medical and Molecular Mycology, Clinical and Toxicological Analyses Department, Federal University of Rio Grande do Norte.
For the patients with suspected keratitis, demographic and clinical data were recorded as follows: gender, age, occupation, admission date, presence of ocular trauma, affected eye (left, right or both), antifungal treatment, transplantation and clinical outcome.For the patients with suspected dermatomycoses, gender, age and body site affected were recorded.

Specimen collection and laboratory procedures
Corneal scrapes and swabs were collected from the base and edge of the ulcer by using a sterile surgical blade (# 15 on a Bard-Parker handle) and swab (Absorve-Stuart, Brazil) under topical anesthesia (oxybuprocaine hydrochloride, 0.4% w/v; Latinofarma, Brazil) and slit-lamp magnification.Direct examination of corneal smears was performed after Gram staining and 20% potassium hydroxide (KOH) with optical microscopy (CX21, Olympus, Japan).The smears were inoculated onto the surface of chocolate agar, 5% sheep blood agar, brain heart infusion and thioglycolate (Laborclin, Brazil) and incubated at 37˚C for bacterial isolation.For fungal isolation, the samples were seeded on Petri dishes containing Sabouraud Dextrose Agar (SDA), supplemented with 50 g/L chloramphenicol (Mumbai, Maharashtra, India), incubated at 25˚C-30˚C and 37˚C and discarded after 3-4 weeks if there was no growth [13].Skin and nail scrapings were treated with 20% KOH during 30 min for direct examination.The samples were inoculated on the surface of Petri dishes containing SDA, supplemented with 50 g/L chloramphenicol (Mumbai, Maharashtra, India) and the plates incubated at 25-30˚C, for a period of up to 4 weeks before being considered negative [29].Colonies showing macro and micromorphology compatible with Fusarium and related genera (morphology, pigmentation, types of conidia formed, etc.), were further identified with molecular biology techniques at the Westerdijk Fungal Biodiversity Institute, Utrecht, the Netherlands.

DNA extraction
Isolates were grown on Malt Extract Agar (MEA; 10 g Merck-Darmstadt, Germany malt extract, 20 g Merck agar, 1000 mL H 2 O), incubated at 24˚C, using a 12/12 h photoperiod (dark/light), under near UV light.Genomic DNA was extracted using the Wizard Genomic DNA purification Kit (Promega Corporation, Madison, WI, USA), following the manufacturer's instructions [2].

Phylogenetic analyses
Sequences of the individual markers, including introns, were aligned with ex-type and reference sequences via MAFFT v. 7.110 [36] using default parameters and adjusted manually, when necessary.Alignments were analyzed using Maximum Likelihood (ML) and Bayesian Inference (BI).For the ML analyses, concatenated phylogenies, where each marker was treated as a separate partition, were determined using IQ-TREE v. 2.1.2[37] with ultrafast bootstrapping [38] for estimation of branch support.Additional ML analyses were performed using RAxML v. 8.2.12 [39].To assess the robustness of the analyses, the Bootstrap support (BS) was determined automatically by the software using default parameters.The BI analysis was performed using MrBayes v. 3.2.6 [40] on the CIPRES science gateway portal [41], using four Markov chain Monte Carlo (MCMC) chains starting from a random tree topology.

Morphological studies
In order to check if the molecular identification was compatible with phenotypic characteristics of each species, micromorphological characters were observed after 7-10 days of incubation on carnation leaf agar (CLA; sterile leaf pieces, 20 g Merck agar, 1000 mL H 2 O) [42].All microscopic characters were examined using sterile water as a mounting medium with a Zeiss Axioskop 2 compound light microscope, including the shape of macroconidia produced in sporodochia, the shape and mode of formation of microconidia and chlamydospores on CLA.To facilitate the comparison of relevant micro-and macroconidial features, composite photo plates were assembled from separate photo micrographs using Adobe Photoshop CC [2].

Biofilm formation of Fusarium spp. and related genera
Biofilm formation assays were performed according to Sav et al. [24] with some modifications.The clinical isolates were previously cultivated in Potato Dextrose Agar (PDA; Merck, Darmstadt, Germany) slants at 25-30˚C, for 7 days.Then, 1 mL of phosphate buffered saline (PBS) was added to the cultures and further vortexed to make a conidial suspension.Next, the cell suspension of each isolate was adjusted to a final concentration of 1 × 10 6 conidia/mL) using a hemocytometer.Subsequently, 200 μL of the suspension was transferred to flat bottom, 96 well microtiter plates and incubated for 2 h at 37˚C for the adhesion phase in quintuplicate (test wells).As controls, eight wells of each microtiter plate were handled similarly, except that no conidial suspension was added.Following the adhesion phase, cell suspensions were aspirated and each well was washed twice with 150 μL of PBS to remove loosely adherent cells.A total of 200 μL of brain-heart infusion broth (BHIB; Difco Franklin Lakes, NJ, USA) supplemented with glucose (0.25%) was added to each of the washed wells and incubated at 37˚C in a shaker incubator at 100 rpm.Biofilms were allowed to develop for 24 h and quantified by a crystal   The inocula of all isolates tested were obtained from 48 to 72 h cultivation in PDA at 35˚C followed further incubation at 25 to 28˚C for 7 days.One milliliter of saline solution was added to each culture to prepare a conidial suspension.The suspension was adjusted to an optical density that ranged from 0.15 to 0.17 at 530 nm.Then, a serial dilution was made in RPMI (1:50), in order to obtain a final concentration of 0.4 to 0.5 × 10 4 cells/mL.Susceptibility to antifungal agents was evaluated by broth microdilution, as follows: Aliquots of 100 μL of the final inoculum solution were dispensed in microtiter plates of 96 wells containing 100 μL of various concentrations of the tested drugs.Finally, the plates were incubated at 37˚C and test reading taken after 72 h incubation.MIC was defined for KTC to the lowest drug concentration which showed approximately 50% reduction in turbidity as compared to the positive control well.For AMB and ITC, the MIC was defined as the lowest concentration able to inhibit any growth visually perceptible [43].

Statistical analysis
Data were analyzed using the statistical software GraphPad Prism version 8.0 (GraphPad Software, San Diego, CA, USA).Results were presented as mean ± standard deviation, and differences were analyzed by the Mann-Whitney test.For all the analyses, P was considered a default value of 0.05 and the confidence interval of 95%.

Clinical, demographic and epidemiological characteristic of the patients
A total of 50 patients with suspected keratitis were enrolled in this study between March 2012 and December 2022.The majority of individuals were male (n = 37; 74%), while 13 (26%) of them were female.Patient´s age ranged from four to 69 years old, average of 42.4 ± 16 years old, median of 42 years old and mode of 68 years old.Thirteen patients (26%) reported previous trauma of the eye, while 10 (20%) denied it, and the information was unknown for 27 (54%) of them.For the patients who reported trauma, most of the reported cases were related to injury with plant material or soil.The majority of patients were agricultural or plant related workers (approximately 67% of reported cases).Right or left eyes were relatively equally affected (50% versus 40%, respectively), while in 4% of the cases, both eyes were infected with fusarioid spp.Corneal transplant was performed in 26 patients (52%), while for 24 patients (12 each), the transplant was either not performed or nor informed (48%).The patients received the following antifungal drugs regimen: 10 of them (20%), AMB only.Ten (20%), a combination of AMB and KTC.Seven (14%), a combination of AMB and NAT.Six of them (12%) received a combination of AMB, NAT and KTC.Three patients (6%) each received either a combination of KTC and NAT or antibacterial agents only (totalizing 12%), while treatment information is missing for 10 (20%) of the patients.Eighteen patients showed clinical improvement (36%), while 16 of them (32%) did not show visual improvement after antifungal treatment, corneal transplantation or surgery.Loss of patient follow up/missing information was observed in 32% of patients.Only two patient had bacterial eye infections together with fusarioid keratitis, as follows: patient 33 (Acinetobacter baumannii) and 38 (Staphylococcus aureus).
Regarding patients with dermatomycoses, 10 individuals were included, being five females (50%) and three males (30%).Patient's age ranged from 38 to 67 years old, average of 51.5 ± 12.7 years old, median of 52.5 years old.Most of the onychomycosis lesions were found on the toenails (4; 40%), followed by fingernails (2; 20%).Tinea pedis accounted for 20% of cases, being one of them in the plantar and the other one in the interdigital region of the feet.Clinical and demographic data is missing for two (20%) patients.The collection period was between August, 2017 to September, 2018 (Table 1).

Phenotypic preliminary identification of the strains used in the present study
A total of 60 positive cultures were found (50 from corneal smears and 10 from nail or epidermal scales).All the strains produced fusarium-like colonies, macro and micromorphology with the presence of sickle-shaped septate macroconidia and/or smaller, usually non-septate microconidia.

Molecular identification and phylogenetic analysis of the strains used in the present study
For molecular species identification, DNA fragments of TEF1-α and RPB2 were amplified and the sequences obtained were lodged in the GenBank sequence database (NCBI, http://ncbi.nlm.nih.gov) using BLAST, to compare gene sequences.Phylogenetic analyses confirmed the species identification of 47 isolates as belonging to the genus Neocosmospora (78.33%), nine of them to the FFSC (15%) and four isolates to the FOSC (6.66%).The clinical isolates were identified as follows: 27 of them as N. falciformis (45%), all of them isolated from keratitis cases; thirteen isolates as N. keratoplastica (21.66%), of which seven were from keratitis cases and six from dermatomycosis.Nine isolates were identified as F. annulatum, seven from keratitis and two from dermatomycosis (15%).Five isolates were identified as N. suttoniana (8.33%), all of them obtained from corneal smears.Four isolates (6.6%) of F. nirenbergiae obtained from keratitis cases were found.Finally, two (3.3%) isolates obtained from onychomycosis were identified as N. gamsii.It is important to emphasize that preliminary BLAST searches had either identified some strains belonging to Neocosmospora as F. solani sensu stricto or were not able to distinguish the species within the Neocosmospora clade.In addition, F. nirenbergiae strains were misidentified as F. oxysporum sensu lato, while F. annulatum strains were identified as F. proliferatum.These uncertainties were solved with phylogenetic analyses (Fig 1).

Morphological studies
Micromorphological observations revealed that all the strains grown on CLA had fungal structures compatible with each species, as identified by DNA sequencing and phylogenetic analysis (Fig 2 ).

Evaluation of biofilm formation in Fusarium spp. and related genera
There was a large variation in biofilm biomass among all the clinical isolates evaluation.In N. falciformis, the OD 450nm ranged from 0.02±0.01(LMMM1528) to OD 450nm of 0.40 ± 0.01 (LMMM 130).The average OD 450nm for all the strains tested was 0.14 ± 0.08.With regard to tercile analysis, 17 isolates (62.96%) were considered weak, nine moderate (33.33%), while one (3.7%)isolate was a strong biofilm producer.In N. keratoplastica, the OD 450nm ranged from 0.08±0.01(LMMM1461) to OD 450nm of 0.21 ± 0.01 (LMMM1263).The average OD 450nm for all the strains tested was 0.14 ± 0.04.Most isolates were weak biofilm producers (8; 61.53%), while five of them (38.46%) were moderate biofilm producers.Eight-six of all F. annulatum keratitis isolates (n = 6) were moderate biofilm producers.When the strains obtained from nails and skin were included in the analysis, the percentage of moderate biofilm producers is still higher than the ones considered of low biofilm producers (66.66% versus 33.33%, respectively).The average OD 450nm for all the strains tested was 0.16 ± 0.03, ranging from 0.11 ± 0.01 (LMMM1058) to 0.21 ± 0.03 (LMMM440).Interestingly, the clinical isolate that showed the highest biofilm formation belonged to N. suttoniana (LMMM1355; OD 450nm 0.45 ± 0.02).However, there was a large variation in biofilm formation among the strains of this species, with 40% of them classified as weak and moderate biofilm producers, each.A similar trend of variation was observed among isolates of F. nirenbergiae (50% each of weak and moderate producers), with an average OD 450nm of 0.13 ± 0.02.The two isolates identified as N. gamsii were weak biofilm producers (Table 1 and Fig 3).The only statistically significant comparison was observed between F. annulatum and N. keratoplastica obtained from keratitis (0.17 ± 0.03 versus 0.12 ± 0.04, respectively; P = 0.04).This difference was also observed when comparing all strains (keratitis and dermatomycosis) from each SC (or Neocosmospora): FFSC versus Neocosmospora (0.16 ± 0.01 versus 0.14 ± 0.01, respectively; P = 0.04; Fig 3).

Antifungal susceptibility testing for Fusarium spp. and related genera
MIC ranges, MIC50, MIC90, MIC97.5 and geometric means (GMs) for all the antifungal drugs tested are displayed in Table 2. Interestingly, only a few strains had increased MICs if we consider the MIC97.5 as a cut-off value (equivalent to the ECV).For KTC, one isolate of N. falciformis and 2 isolates of N. keratoplastica showed MICs above 16 μg/mL.For all the other strains tested, the MICs were � MIC97.5.Nevertheless, several strains showed AMB MIC = 8 μg/mL which corresponds to a single dilution lower than the MIC97.5.

Discussion
Fungal keratitis is a serious problem that may lead to blinding and has been considered an ophthalmology emergency [10,44,45].It is one of the leading causes of blindness among young people.Currently, there is a call for attention for fungal keratitis to be considered a neglected tropical disease, recognized by the world health organization [46,47]  approximately 1.5 million new cases per year [48] with a greater prevalence in tropical and subtropical countries, where it accounts for 20 to 60% of microbial keratitis and are frequently related to agricultural-related eye trauma [49,50].Another remarkable fact is that approximately 10 to 25% of affected patients will require eye surgery, and around 60% of them may develop blindness even after transplantation and antifungal treatment [51,52] and 8-11% of patients having to have the eye removed.
In the present study, most of the patients with keratitis were male (74%) had a median age of 42 years old, worked with plant material or debris, while 26% of them reported trauma in the eye.This agrees with the literature, where this disease has been related to agricultural activities that are usually performed by males with ages ranging from 20 to 40 years old [13,[53][54][55], a demographic more commonly found among tropical and subtropical developing countries.By contrast, case series from developed, temperate nations frequently report contact lens wear as the main risk factor, with higher female prevalence [56][57][58], for which clinical outbreaks have been previously reported [59,60].Moreover, we did not encounter any case of fusarioid keratitis related to contact lens wear, which agrees with a previous Brazilian study, where the incidence of this risk factor was only 5% [54].
In the present study, the number of therapeutic keratoplasties was above 50%.This high percentage of necessity for corneal transplant is in agreement with reports from other developing countries [61].In a study conducted in Sao Paulo, Brazil, therapeutic grafts were needed for 22 patients (54%) with fusarioid keratitis [54].On the contrary, in the Netherlands, in a series of 89 cases of this infectious disease from 16 different hospitals, only 22% of the patients required corneal transplantation [57].
AMB was the most frequent antifungal drug used to treat the patients with Fusarium keratitis, either as a monotherapy or in combination with NAT and/or KTC.It is considered an alternative for keratitis treatment but the eye drop presentation has to be acquired from a compounding pharmacy and its toxicity limits its usage [26].On the other hand, NAT has been considered the therapy of choice to treat Fusarium keratitis, but it does not have effective penetration on the corneal stroma [62].Topical VRC may also be applied, but may result in increased corneal perforation after keratoplasty, while it is less effective than NAT [52].Oral KTC and ITC may be successfully used to treat fusarioid keratitis when VRC (a costlier antifungal agent) is not available [61,63].
Despite the low number of cases of dermatomycoses in the present study, our findings agree with the literature, where most of patients are female and the lesions found on the toenails.In a review conducted by Uemura et al. [18] regarding onychomycosis reported from 16 countries, 66.27% of the cases were found in women.Furthermore, in 46.5% of cases, the lesions were found on toenails, whereas in 10.46% of cases, both toenails and fingernails were affected.
Factors associated with higher prevalence of fusarioid onychomycosis on toenails compared to fingernails include larger ungueal size, slower growth rate, less blood flow and constant trauma associated with humidity [64,65].
The vast majority of the isolates in the present study were identified as Neocosmospora spp., independently of the body site affected.In fact, a study conducted by Mayayo et al. [66] revealed that the former F. solani strains were lethal in a murine model of systemic infection, whereas F. oxysporum, F. verticillioides and F. proliferatum were significantly less virulent.Neocosmospora strains could have been more frequently isolated from humans due to higher virulence to mammal hosts.
In a study conducted with 44 strains obtained from patients originated from different parts of the globe and collected from several body sites, most of the fusariosis cases were caused by Neocosmospora spp., corroborating our findings [67].The main species isolated were N. keratoplastica (31.8%), all obtained from dermatomycoses and N. falciformis (20%), most of them isolated from keratitis.N. keratoplastica was also the most isolated fusarioid species from 47 cases of onychomycosis in Taiwan (corresponding to almost 60%) [68].
In another case series of keratitis conducted in Taiwan, the main species found were also N. falciformis (32.6%) and N. keratoplastica (27.9%), among other species of Neocosmospora.Species belonging to the FDSC, FOSC (including one strain of F. nirenbergiae) and FFSC were also isolated, demonstrating a greater species diversity than the present study [69].N. falciformis corresponded to 95% of the isolates obtained from cases of keratitis in the South of India [70].That was also the main species found in our patients with keratitis.
In a German study with 22 patients with Fusarium keratitis, the clinical isolates were identified as belonging to Neocosmospora, FFSC and FOSC.The authors described a high prevalence of N. petroliphila (a species not found in the present study) and other species belonging to the FOSC (27%, each) [56].Interestingly, a high percentage of N. petroliphila from keratitis was also described in the Netherlands (another temperate country) [57].This fact could be somehow related to the variation of fusarioid species distribution in the environment, once they can be isolated from soil and plants [71].
In the present study, there was a high percentage of isolation of F. annulatum from both keratitis and dermatomycosis.This finding may be a peculiarity of the Northeast region of Brazil.This is a higher prevalence compared to other investigations even if we take into consideration that some of the publications might include strains misidentified as F. proliferatum when they are in fact F. annulatum.It is important to emphasize that both species belong to completely different phylogenetic clades within the FFSC [72].
Because of the large variation in the ability to form biofilm among Neocosmospora spp.strains, most of the comparisons did not reveal any statistically significant results.For instance, the only two strains that were strong biofilm producers belonged to N. suttoniana and N. falciformis, both obtained from keratitis.Conversely, some other Neocosmospora spp.strains showed very weak biofilm formation.Interestingly, most F. annulatum clinical isolates were consistently moderate biofilm producers, and the average values of optical densities for biofilm formation were statistically higher for the FFSC.It is recommended to check in future investigations if these results are not anecdotical.
Currently, there are not many studies comparing biofilm formation among different fusarioid species, specifically when we consider identification to the species level within the complexes.Recently, it has been demonstrated that N. falciformis may form biofilm on polystyrene plates and that conidial aggregates are important for the adhesion phase, whereas anastomosed and well-developed hyphae embedded in an extracellular matrix are found in mature biofilms [73].In addition, it has been shown that N. keratoplastica is able to form AMB-resistant biofilm on intravenous catheters [74].
A limitation in the interpretation of the antifungal susceptibility testing results is related to the lack of established clinical breakpoints for fusarioid spp.In addition, it was not possible to establish an ECV in the present study (despite the fact that we have determined MIC97.5 concentrations), because of the necessity to include at least 100 unrelated strains from three independent laboratories [75].Recently, ECVs were established for Neocosmospora (= FSSC) and FOSC, making it difficult to compare species within each complexes [25].
The results of GMs, MIC50 and MIC90 to AMB in the present study were generally higher than the other case series with Fusarium spp.isolated from both keratitis [56,61,69,70] and onychomycosis [18,68].The systemic AMB used here to treat patients with keratitis from the present study may have somehow resulted in increased MICs.
On the contrary, our ITC MICs were lower than most other studies [25,68].High MICs were also found for KTC.There are not many studies that include this drug in the panel of antifungal susceptibility testing for fusarioid spp., limiting our comparisons.Nevertheless, oral KTC may be used to treat keratitis caused by these pathogens [9,63,76] and its frequent use in our patients may also explain the increased MICs values found.
It is important to emphasize that the methodologies used to perform this study are the most appropriate and specific ones to evaluate fungi of the genus Fusarium and other allied genera.The CLSI methodology used for antifungal susceptibility testing is a reproducible procedure if the protocol is strictly used (culture medium, drug dilutions, test readings and interpretations) [25].The most important reason to evaluate antifungal susceptibility is when the fungal infection is invasive or cause severe damage to tissue and organs, when acquired drug resistance is suspected, or when the patient is unexpectedly failing therapy.For each of these scenarios, knowing the in vitro susceptibility pattern would inform the clinician when making therapeutic choices or changes for the adopted therapy [77].
We have found a high prevalence of fusarioid keratitis in Northeastern Brazil.Our study contributed to the knowledge of the epidemiology of Fusarium and allied genera infections and this information could be used to assist in the adoption of strategies to improve prevention, and treatment of fusarioid infections.We emphasize that this is the first study in Northeast Brazil that has made such a deep analysis in this regard, despite our limitations, mainly due to the incomplete information available in the medical records.Fusariosis is an important and neglected disease in this region and other developing and tropical countries, with specific peculiarities, given the high number of cases, increased need for keratoplasty and poor outcome, even after antifungal and surgical treatment.It is an infirmity essentially related to males that work with agriculture or other soil related professions.Species of Neocosmospora were the most prevalent etiological agents for both keratitis and dermatomycosis, but a remarkable number of consistently biofilm-producer F. annulatum isolates was found.Increased MICs (although possibly still of wild-type phenotype) were found against AMB and KTC.
Fusarium and Neocosmospora spp.Demographic, clinical, and antifungal treatment data of 60 patients diagnosed with keratitis and dermatomycoses from March 2012 to December 2022, in Rio Grande do Norte, Northeastern Brazil. of Medical and Molecular Mycology.

Table 1 . Identification and biofilm formation of
https://doi.org/10.1371/journal.pntd.0012247.t001violet assay.Briefly, the biofilm-coated wells of the microtiter plates were washed trice with 150 μL of PBS and then air dried for 30 min.Subsequently, each of the washed wells was stained with 200 μL of 1% aqueous crystal violet solution for 30 min.Next, each well was washed trice with 200 μL sterile distilled water and immediately distained with the addition of 200 μL of 33% acetic acid, for approximately 30 min until complete crystal violet solubilization.Subsequently, 100 μL of the distaining solution was transferred to a clean well and the amount of crystal violet stain in the referred solution was quantified with a microtiter plate reader (SpectraMAX 340 Tunable Microplate Reader; Molecular Devices Ltda., San Jose, CA, USA) at 450 nm.The absorbance values for the controls were subtracted from the values of the test wells to minimize background interference.The isolates were classified according to terciles distribution as follows: 0.01 � OD 450nm � 0.14, as weak; 0.15 � OD 450nm � 0.29, as moderate; 0.30 � OD 450nm � 0.45, as strong biofilm producers.
. Estimations indicate dermatomycosis clinical isolates from Rio Grande do Norte, Northeast Brazil isolated in this study are shown in bold font.The tree is rooted to Bisifusarium nectrioides (CBS 176.31) and Bisifusarium dimerum (CBS 108944).Ex-epitype, ex-neotype and ex-type strains are indicated with ET , NT , and T , respectively.LMMM stands for Laboratory of Medical and Molecular Mycology.https://doi.org/10.1371/journal.pntd.0012247.g001