Aspergillus spp. Isolated from Lungs of Poultry (Gallus gallus) at the Mycology Laboratory, School of Veterinary Medicine, Universidad Nacional, Heredia, Costa Rica between 2008 and 2021 and Associated Factors

Aspergillosis is a disease caused by some species of the fungus Aspergillus, occurring in both mammals (including humans) and birds, the latter being the most susceptible group. Aspergillus must be considered a public health concern as it affects the poultry industry economically and is an occupational risk to its workers. A retrospective study of fungal isolates from the lungs of chickens (Gallus gallus), analyzed between 2008 and 2021 at the Mycology Laboratory, School of Veterinary Medicine, Universidad Nacional, Heredia, Costa Rica was performed to report the prevalence of Aspergillus spp. in poultry farms in Costa Rica and their associated factors. A total of 1113 cases were received, of which 31% (n = 392; 95% CI: 28.3–33.7) were positive for fungal isolation. Aspergillus was the most frequently detected genus, and the most frequent sections were Fumigati (n = 197/392, 50.3%), Flavi (n = 90/392, 22.9%), and Nigri (n = 50/392, 12.7%). Significant effects (p < 0.05) related to the year, geographical origin, purpose, and age were identified in relation to the Aspergillus infection. The identified factors are explained by climatic variations in the tropics and the particularities of the birds. Future research including molecular characterization and antifungal susceptibility tests in animals, humans, and the environment, are needed to better understand the risks of the diseases caused by those fungi in this country.


Introduction
Aspergillus is a genus of filamentous saprophytic fungi, which is found in various environmental substrates such as soil, grains, seeds, as well as in decomposing vegetation [1]; it has a worldwide distribution [2,3]. This genus is composed of six subgenus, 27 sections, 86 series, and more than 695 species [4]. The vast majority inhabit the environment and do not represent a risk to public health, except for species from the sections Fumigati, Flavi, Nigri, Circumdati, Terrei, Nidulantes, Ornati, Warcupi, Candidi, Restricti, Usti, Flavipedes, and Versicolores, which have been described as agents of disease [5,6].
Aspergillosis is the disease caused by some species of this genus [3,7] that occur in birds and mammals (including humans) [5,7], in which infection happens by the inhalation or ingestion of conidia (asexual reproductive structures) from the environment or, in the case of eggs, by penetration of the shell [2,3]. Inhalation is the main route of entry; thus, respiratory symptoms are the most common symptoms in all affected species [5].
From all over the animal kingdom, avian species are the most susceptible group, including commercial birds such as chickens, turkeys, ducks, geese, ostriches, rheas, quails, and pigeons [8], as well as captive and free-ranging wild species [2,7]. This susceptibility

Sample and Culture Processing
The lungs were directly extracted from the coelomic cavity using dissection scissors and tweezers; later, they were placed in sterile plastic Petri dishes, making pools of up to five pair of organs per plate. They were transported to be immediately cultured or cultured for a maximum of 12 h; in the latter case, the samples were kept refrigerated between 2 and 8 • C. The dissection was performed in the UPDA on a stainless-steel necropsy table accompanied by a Bunsen burner to create a sterile area. The lungs were extracted with forceps and dissecting scissors that were flamed between each bird. At the LMIC, in a type II laminar flow chamber (Labogard, Nuaire, Plymouth, MN, USA), the lungs were cut into portions spanning 1 cm wide at the maximum, then they were passed through the flame three times and cultivated, keeping at least 2 cm of distance between each portion, up to a maximum of seven portions per plate. The samples were cultivated in Sabouraud dextrose agar (Oxoid, Basingstoke, UK; Liofilchem, Roseto Degli Abruzzi, Italy) and were incubated at 37 • C for five days. Once the incubation time had passed, each plate was analyzed for the growth of filamentous hyaline fungi on the surface of the tissue portion; any growth outside this area was considered contamination. The colony-forming units (CFU) were counted and registered in each culture; a case was considered positive when there was growth of at least one CFU of fungus, and a positive case could have more than one type of fungus (mixed isolates).

Fungal Identification
Macroscopic and microscopic morphological identification was carried out following the available guidelines, depending on the year in which the samples were received [4,[16][17][18][19][20][21][22]. The macroscopic characteristics that led to the identification of the Aspergillus genus included hyaline colonies, circumscribed on the reverse, sometimes with yellowish to reddish pigmentation. On the obverse, they were hyaline colonies ranging from cottony to velvety, which varied in color depending on the conidial pigmentation: blue-green, olive-green, dark brown, black, cinnamon, yellow, and white. Regarding the evaluation of the microscopic morphology, the hyphae were hyaline septate, the conidiophore was rough or smooth, the shape of the vesicle varied from round to semi-clavated, the type of phialide disposition were uniseriate and/or biseriate, and the arrangement of the phialides on the vesicle were radiate, semi-radiate, or columnar. Sexual reproduction structures such as cleistothecia or other structures such as Hülle cells were also considered for the identification. Because the molecular identification was not performed on the isolates, they were classified in the following sections: Fumigati, Flavi, Nigri, Terrei, Circumdati, and Candidi. In the cases that fungi other than Aspergillus were isolated, only those whose growth was pure and directly came from the tissue surface were considered important, and the identification was made following the same references.

Registration of Information and Statistical Analysis
A database was created using Excel software (Microsoft Corporation, Redmond, WA, USA, version 18.2205.10910) with the information from the receipt sheets for each sample, considering the following data: case number of the sample, year of receipt, month it was received, purpose of the bird (incubator, broiler, layer, or reproduction), geographical area of remission such as province (San José, Alajuela, Cartago, Heredia, and Puntarenas) and county of origin, whether or not it came from the airport (if so, that means that the bird lot came from another country), breed, age (in days), sex, and isolated fungus (genus if possible or section in the case of Aspergillus), with their respective CFU counts (Supplementary Table S1).
Regarding the CFU counts, a table of frequency was created in Excel, and the results were displayed on a heat map created using KNIME (Konstanz Information Miner: Version 4.7.0). Considering that the National Animal Health Service (SENASA) of Costa Rica requires that in cases where at least one CFU of Aspergillus is isolated, corrective actions must be performed in the poultry farm; however, no further analysis was performed with that information.
A logistic regression analysis was carried out using the SAS/STAT ® program to evaluate the effect of the following factors: year, month, county of origin, purpose, breed, sex, and age of the referred cases, on the occurrence of Aspergillus infection. In some cases, due to the low number of samples in certain categories, they were grouped as "others". The same analysis was specifically performed for the three most frequently isolated sections of Aspergillus: Fumigati, Flavi, and Nigri.

Study Limitations
Because the laboratories in which the birds and the samples were received are in the facilities of a university campus, the reception is not carried out during the holiday periods, which includes Holy Week, two weeks in July, the month of December, and one week in January, impacting the number of samples received in those months. Likewise, due to laboratory opening hours, on some occasions the farms delivered the chick from the incubator on day two as they did not have time to deliver it on day one; for this reason, the incubation area goes from 0 to 2 days, although on farms it is up to one day old. Furthermore, the data collected were the information that was filled out by the senders in the receipt sheet; therefore, on several occasions it is not complete. The identification of the genus Aspergillus prior to 2016 was not performed by sections, so in cases where it could not be categorized into any of the most common sections, the isolated fungi was reported as Aspergillus spp.

Colony-Forming Units (CFU)
The count of colony-forming units (CFU) of the isolates showed that most of the fungi growth occurred as a single isolate (n = 216); however, in some cases, Aspergillus section Fumigati (n = 3) and Aspergillus section Flavi (n = 2) growth up exceeded 10 CFUs up to a maximum of 24 CFUs (Figure 2).

Logistic Regression Analysis of Aspergillus spp.
The logistic regression analysis showed significant effects (p < 0.05) of the year of reception, county of origin, purpose, and age of the birds on Aspergillus infection. The categories that pose a greater risk were cases received in 2014 (OR 3.48; IC 95% 1.64-7.37), broilers (OR 2.07; IC 95% 1.18-3.63), poultry intended for breeding (OR 2.94; IC 95% 1.60-5.41), chickens from the county of Puntarenas (OR 10.61; IC 95% 3.90-28.88), and poultry older than 15 days (OR 2.96; IC 95% 1.35-6.50) ( Figure 3). There were no significant differences about the months when the samples were submitted or the sex of the chicken.

Colony-Forming Units (CFU)
The count of colony-forming units (CFU) of the isolates showed that most of the fungi growth occurred as a single isolate (n = 216); however, in some cases, Aspergillus section Fumigati (n = 3) and Aspergillus section Flavi (n = 2) growth up exceeded 10 CFUs up to a maximum of 24 CFUs ( Figure 2).

Colony-Forming Units (CFU)
The count of colony-forming units (CFU) of the isolates showed that most of the fungi growth occurred as a single isolate (n = 216); however, in some cases, Aspergillus section Fumigati (n = 3) and Aspergillus section Flavi (n = 2) growth up exceeded 10 CFUs up to a maximum of 24 CFUs ( Figure 2).
The logistic regression analysis showed significant effects (p < 0.05) of the year of reception, county of origin, purpose, and age of the birds on Aspergillus infection. The categories that pose a greater risk were cases received in 2014 (OR 3.48; IC 95% 1.64-7.37), broilers (OR 2.07; IC 95% 1.18-3.63), poultry intended for breeding (OR 2.94; IC 95% 1.60-5.41), chickens from the county of Puntarenas (OR 10.61; IC 95% 3.90-28.88), and poultry older than 15 days (OR 2.96; IC 95% 1.35-6.50) (Figure 3). There were no significant differences about the months when the samples were submitted or the sex of the chicken. Regarding the breed significant differences, they were noted in "others" (OR 3.14; IC 95% 1.31-7.56) and for poultry in which the breed was not specified (OR 1.79; IC 95% 1.06-3.02) (Figure 4).  5.41), chickens from the county of Puntarenas (OR 10.61; IC 95% 3.90-28.88), and poultry older than 15 days (OR 2.96; IC 95% 1.35-6.50) (Figure 3). There were no significant differences about the months when the samples were submitted or the sex of the chicken. Regarding the breed significant differences, they were noted in "others" (OR 3.14; IC 95% 1.31-7.56) and for poultry in which the breed was not specified (OR 1.79; IC 95% 1.06-3.02) (Figure 4).

Isolations from Bird Lots Coming from the International Airport
During the study period, a total of 129 samples from bird lots were received from other countries, of which 46 positive cases (35.7%) were obtained without finding a significant difference in relation to chickens from Costa Rica. The Fumigati section was the one that was most frequently found (n = 37), and the Flavi, Nigri and Terrei sections were found in equal proportion (n = 5 each).

Aspergillus Section Flavi
In relation to the positive isolates of Aspergillus section Flavi, significant differences were found in the years 2012 (OR 9.20; 95% CI 2.27-37. 40 (Figure 6).

Aspergillus Section Flavi
In relation to the positive isolates of Aspergillus section Flavi, significant differences were found in the years 2012 (OR 9.20; 95% CI 2.27-37. 40 (Figure 6).

Aspergillus Section Nigri
Regarding the positive isolates of Aspergillus section Nigri, significant differences were observed in chickens intended for breeding ( (Figure 7).

Discussion
Reports on the prevalence of Aspergillus in bird lungs are scarce, even though it is a causative agent of an important disease in these animals. The available research varied in terms of the geographic location (the majority are in the tropic), types of samples taken (including bedding, environment surfaces of the enclosures, and food, as well as in various organs such as the lungs, air sacs, brain, gizzard, and liver), bird species considered (productive and wild), and diagnostic methodology (clinical and/or laboratory).
The prevalence obtained in Costa Rica is lower than the prevalence reported in Egypt [23], where it was obtained that 74% of Aspergillus isolates that came from animals had signs suggestive of aspergillosis. A review of common respiratory diseases in poultry carried out in Bangladesh mentioned that 6.14% of 1981 birds with respiratory signs were diagnosed with aspergillosis, but only by clinical findings without laboratory confirmation [24]. More recently and also in Bangladesh, a 36% (18/50) prevalence of this fungus was reported in chicken lung samples, confirmed by culture and polymerase chain reaction (PCR) analyses [25].
The occurrence of Aspergillus in environmental samples from poultry farms in the Apulia region of southern Italy ranged from 31.6 to 55.1% depending on whether it came from air samples (30/57), feces pools (38/69), or feeders (6/19) [1]. In Egypt, a study was carried out on 100 samples of chicken lungs and 80 samples of feed, bedding, water, and air from farms, obtaining an occurrence of 24% and between 10 and 55%, respectively [26]. Also in Egypt, Aspergillus was isolated from a sampling carried out in 88 broilers at a prevalence of 43.2% (38/88) [27]. In that same country, laryngeal swabs were taken from poultry, and the obtained positivity for this fungus was 72.5% (29/40) [28].
In a study carried out in Indonesia with lungs from chickens offered in the market, 66.7% (20/30) were positive for Aspergillus [29]. In a study from Iran conducted on 600 broilers, Aspergillus was isolated from tracheal swab samples in 51.6% (310/600) of the samples [30]. In Mexico, a sampling was carried out in 73 birds with clinical signs of aspergillosis, and an occurrence of 9.6% of A. fumigatus was obtained [31].
Paecilomyces variotii has not been associated with disease in poultry, but in human medicine, there are some reports of pneumonia, mainly in immunocompromised patients [32]. Considering its thermotolerance and presence in the dust of poultry farms [33], we highlight this finding to be followed for animal health and occupational risk purposes.
Fungi of the genus Aspergillus more easily reproduce when environmental conditions are favorable, especially in warm and humid climates [34]. In the present research, a positive effect of Aspergillus infection was detected in the year 2014. During that year, at the Sabana Larga Pluviometry Station located in Atenas, Alajuela, Costa Rica (one of the geographical areas where most of the poultry farms in this country are concentrated), a record of very low rainfall was recorded in July, with an anomaly of rainfall of −85.7 mm compared with the historical average of 1940-2008. Moreover, the temperatures in the province of Alajuela were higher than the historical average reported [35]. Added to this, in the first half of 2014 and 2015, lower than normal temperatures were reported in the tropical sector of the Atlantic Ocean and the Caribbean Sea, and this cooling caused variations in the climate, significantly influencing the rainfall on the Pacific slope. In addition, in the second semester of both years, there was an increase in significant rainfall, and the effect was greater in 2014 due to the warming of the Atlantic waters and the low intensity of El Niño; in 2015, this increase was overshadowed by the strong intensity of El Niño [36].
Aspergillus infections have been documented in all ages of birds, but the extremes of age are the most affected [37]. In the present study, a higher risk was obtained in birds older than 16 days, such as in the research in Bangladesh [25], in which a higher infection of Aspergillus flavus was reported in animals older than 3 weeks. In addition, also in Bangladesh, a higher mortality was observed and associated with aspergillosis in chickens of Isa Brown and Hy-Line breeds from three to eight weeks of age [34].
Regarding the productive purpose of the birds, in the cases obtained in Costa Rica, a higher risk was observed in animals destined to reproduction and broilers, contrary to the Nigerian research made in clinically sick chickens, where the most affected were the layers and where serious economic losses occur in the growth stage and just before the start of production [38]. Regarding broilers, in a study which compared the effect of an experimental infection by A. fumigatus in a line of White Leghorns (layers) and a line of broilers, it was observed that the layers did not present mortality compared with the broilers, which presented a mortality greater than 50%. Furthermore, it was shown that the layers were able to resolve the infection more easily than the broilers [39]. This may be due to the immunological differences of each line of production as layers show two types of well-developed immune response (antigenic and cellular) as opposed to broilers, which only show a developed antigen-specific immune response [40].
In reference to a greater risk detected in the geographical area of Puntarenas, Costa Rica, a possible explanation could be that it is the only coastal province where samples were submitted, so the salinity, humidity, and temperature conditions are different from the greater metropolitan area (located in the central valley of the country). At the same time, the greatest risk of isolating section Nigri in this province could be attributed to this fungus halotolerance [21,22], which coincides with the study carried out in Payakumbuh, Indonesia (a coastal zone) in broiler chicken lungs, where A. niger was most frequently isolated [29]. Furthermore, this fungus was the second most isolated section in samples of laryngeal swabs from poultry and the most frequently isolated in sputum samples of poultry workers in Quena, Egypt [29]; in Bangladesh, this fungus was found in higher numbers in samples from 2-3-week-old birds [25].
Detection of Aspergillus in the lungs, environment, surface, food, and litter of birds has implications for public health, the occupational health of farmworkers, and animal health and welfare [1,33,37,38]. Aspergillosis is a sapronotic disease, which means that the infection is acquired by inhalation or ingestion of conidia from a contaminated environment, which can cause respiratory signs in immunocompromised patients and an asthma-like allergic reaction in people exposed to high concentrations or people who are repeatedly exposed to this fungus [22,41]. Poultry farm personnel are an occupational risk group due to the continuous exposure to contaminated environments [1,33,37,38]; the dust from avian farms contains a combination of microorganisms and organic materials from the animals [42]. Biosecurity measures on farms are aimed at avoiding contagion between birds due to movement of personnel, but they must also include the use of personal protective equipment for the workers at risk of environmental exposure.
According to the present study, future research must be intended to identify those fungi at the species level using PCR and sequencing analyses to survey the presence of cryptic species that could be more virulent or refractory to treatment. Moreover, it must involve antifungal susceptibility tests to create a perspective of the antifungal resistance of the isolates, including samples from animals, humans, food, and the environment. All this information will improve the understanding of the epizootiology and the epidemiology of the diseases caused by them in this country, and it will contribute to minimizing the loses of the poultry industry while supporting public health policies with a One Health background.
Mycotic diseases (both from human and animal medicine) are prevalent in tropical areas; however, many of the countries located in that geographical zone are under developing economies, and the reports of fungal agents and diseases are underestimated and/or lacking molecular methods to provide robust information for diagnosis and epidemiology. Furthermore, the difficulty in standardizing and acquiring antifungal susceptibility tests or the availability of less harmful antifungal treatments enhances the importance of funding opportunities and international collaboration to train and integrate professionals of those countries to close the gap regarding diagnosis and research in those populations.