Tear Production, Intraocular Pressure, Ultrasound Biometric Features and Conjunctival Flora Identification in Clinically Normal Eyes of Two Italian Breeds of Chicken (Gallus gallus domesticus)

Simple Summary In Italy, chickens are used for egg production and as courtyard/domestic animals and consequently veterinarians need to know their general and specialist characteristics. One key area is normal ocular measurements in order to understand any pathological changes affecting the eyes. For an accurate diagnosis and better management of ophthalmic diseases in chickens, this paper describes the normal values for the evaluation of ocular tear production, intraocular pressure, and biometric measurements of the eyes and on the microbial and cultural flora normally present in the conjunctival sac in two Italian chicken breeds. Abstract Given the abundance of chickens in Italy, it is important for veterinarians to know the normal state of chickens’ eyes in order to identify any ophthalmic pathological changes. The aim of this study was to determine the normal values of select ocular parameters and to evaluate conjunctival microflora in two Italian chicken breeds. Sixty-six healthy chickens underwent a complete ophthalmic examination, which included a phenol red thread test (PRTT) for the evaluation of tear production and the assessment of intraocular pressure by rebound tonometry. B-mode ultrasound biometric measurements and conjunctival microflora identification were also performed in twenty-seven chickens. Mean PRTT was 23.77 ± 2.99 mm/15 s in the Livorno breed and 19.95 ± 2.81 mm/15 s in the Siciliana breed. Mean intraocular pressure was 14.3 ± 1.17 mmHg in the Livorno breed and 14.06 ± 1.15 mmHg in the Siciliana breed. Reference ranges for morphometric parameters were reported in the two breeds. Twenty-three chickens (85.18%) were bacteriologically positive. Chlamydia spp. antigen was detected in 14.81% of chickens. No positive cultures were obtained for fungi. Normal reference range values for selected ophthalmic parameters were obtained in clinically healthy chickens, which could facilitate accurate diagnosis and better management of ophthalmic diseases in these animals.


Introduction
In Italy there are various breeds of chicken (Gallus gallus domesticus-Linnaeus, 1758) and in addition to intensive chicken farms, it is very common to own a small number of chickens both for egg production and as courtyard/domestic animals. For this reason, Tear production was assessed in each eye using a commercial phenol red thread test (PRTT) kit (PRT-TEST; Tianjin Jingming New Technological Development Co. Ltd., London, UK). The end with the 3 mm angle for the test was placed inside the lower fornix at a point approximately one third of the eyelid width from the temporal cantus. Tear production was recorded in millimeters, wetting after 15 s.
The adnexa and anterior segment of both eyes were examined with a portable slitlamp biomicroscope (Kowa SL-17 ® , Kowa Company, Tokyo, Japan). This procedure as well as the assessment of pupillary light reflexes were carried out in a dark room. Intraocular pressure (IOP) was measured using rebound tonometry (TonoVet ® , iCare, Vantaa, Finland) with the P setting, which is not specific to any species. The P setting was chosen due to the lack of a specific tonometer internal calibration table for measuring IOP in birds. During each IOP assessment, birds were gently restrained, and any pressure on the neck, globe, and periocular region was avoided.
The first three IOP measurements with low/no error were recorded for each eye, each measurement being the average that the rebound tonometer automatically performed on six valid bounces against the cornea. The average of the three measurements was considered the IOP value per eye. The IOP was taken starting each time with the left eye. The ocular fundus was observed using a binocular indirect ophthalmoscope (Omega 180; Heine, Berlin, Germany) with a 20 or 30 D lens, without pharmacological pupil dilation. The fundus examination was performed in a dark room without the use of a mydriatic agent, only minimizing the intensity of the examination light. Fluorescein staining was also performed after completion of fundoscopy. All the measurements were performed by the same operator (GB).
Ocular ultrasonography was performed in twenty-seven chickens and specimens were collected for cultural evaluation of the flora of the conjunctival sac.

Ultrasound Biometric Measurements
In twenty-seven chickens (54 eyes), transpalpebral B-mode ultrasonographic examinations were performed using an ultrasound system, a high frequency linear probe (12 MHz) and abundant ultrasound gel. The chickens were manually restrained. No general and/or topical anesthesia was required for transpalpebral ultrasonographic examinations since the eyelids were closed, and the cornea was not exposed. Axial and transvers scans of the chicken globe were performed. The left eye was assessed before the right eye.
The following morphometric parameters were measured: the distance between the cornea and the anterior capsule of the lens (D1), the distance between the posterior capsule of the lens and the optical papilla (D2), the thickness of the lens (D3), and the axial (D4) and transverse (D5) lengths of the globe. All ultrasound examinations were performed by the same experienced radiologist (SC) in order to reduce interobserver variability.

Conjunctival Bacterial and Fungal Flora Identification
In twenty-seven chickens (54 eyes), materials for bacterial and fungal culture were collected at the inferior conjunctival fornix. The chickens were restrained manually, and three swabs (TRANSYSTEM AMIES Agar GelWith Charcoal, Biolife Italiana Srl, Milan, Italy) were taken from both eyes. One swab was for bacterial culture, one for Chlamydia spp. antigen identification, and the third swab for fungal flora culture was executed one week after material collection for bacterial culture. All the samples were collected by the same operator (GB), without topical anesthesia by rolling the sterile swab on the conjunctival surface, thus avoiding contact with the eyelid margin and cornea.
Enumeration of mesophilic aerobic and heterotrophic anaerobe microorganism (total bacterial count-TBC) was determined for each eye: one swab was mixed with 9 mL of sterile saline solution (1:10 dilution), vortexed and processed. Ten-fold serial dilutions (1:100-1:10,000) were performed, and 1 mL from each of them was inoculated in a sterile Petri dish. Next, 15 mL of Tryptic Glucose yeast Agar (PCA) (Biolife Italiana Srl, Milan, Italy) cooled to 45 • C was added to each plate. Inoculum and medium were carefully mixed, left to solidify, and another 4 mL of APC was added.
After incubation at 30 • C for 72 h, colonies on plates containing more than 10 and less than 300 colonies were counted. TBC was expressed as colony forming units for mL (CFU/mL). In addition, the same swab was inoculated onto a blood-agar plate, incubated at 37 • C, and examined for bacterial growth after 24, 48, and 72 h. Developed colonies were examined by Gram staining, and submitted to typing employing specific media and biochemical tests.
One swab for each eye was employed to detect the presence of Chlamydia spp. antigen using the commercial immunoenzymatic test Clearview Chlamydia MF (Inverness Medical, Waltham, MA, USA).
Samples for mycological culture were seeded onto malt extract agar (MEA, Oxoid, Milan, Italy), and gentamicin was added to avoid bacterial contamination. The samples were then incubated at 25 • C and examined daily from day 4 post-incubation, over a 10-day period to identify mycotic growth.

Statistical Analysis
For each parameter analyzed, the average values +/− the standard deviation, the median, the maximum and the minimum values were reported. The D'Agostino and Pearson test was used to investigate the Gaussian distribution of quantitative variables. The differences relating to D1, D2, D3, D4, D5, PRTT, and IOP between OD (oculus dexter) and OS (oculus sinister) and between the various breeds were investigated by Student's t test for parametric data or by the Mann-Whitney test for nonparametric data.
The mean PRTTs, IOPs, D1s, D2s, D3s, D4s, and D5s of the right eye and left eye were considered as a single value for each bird (single value per animal), because there were no differences between the two eyes. A possible correlation between the weight of every chicken and each biometric parameter was evaluated by the Pearson correlation coefficient for parametric data or by the Spearman correlation coefficient for nonparametric data. In the same way, a possible correlation between the weight and the age of each chicken and PRTT and IOP values was investigated. p values < 0.05 were considered significant.

Ophthalmologic Examination
The mean PRTT of all the chickens was 21.98 ± 3.46 mm/15 s, with 23.77 ± 2.99 mm/15 s in the Livorno breed, and 19.95 ± 2.81 mm/15 s in the Siciliana breed. There were no significant differences in PRTT between OD and OS in the Livorno (p = 0.2787) and the Siciliana (p = 0.4448) breeds, but there was a significant difference between the two breeds themselves (p < 0.0001). Descriptive and inferential statistics of PRTT data are presented in Table 1a-c.  No correlations between the weight and PRTT (p = 0.1419) or the weight and IOP (p = 0.6865) were detected. No correlations between the age and IOP (p = 0.6182) were observed in either breed. However, there was a direct correlation between the age of chickens and PRTT (p = 0.00079) in both breeds.

Ultrasound Biometric Measurements
When an ultrasound is performed in B-mode, the cornea is hyperechoic and convex, and the anterior chamber is anechoic. The lens appears anechoic and ovoid and is delimited by two curvilinear hyperechoic lines corresponding to the anterior and posterior lens capsule. The vitreous chamber is anechoic with the pecten visible as a hyperechoic, millimetric, tubular-shaped structure, projecting into the vitreous chamber from the retina. The retina, the choroid, and the sclera are not ultrasonographically distinguishable since they appear as a single concave hyperechoic line at the posterior limit of the globe. The distance between the cornea and the anterior lens capsule (D1) ( Figure 1A), the distance between the posterior lens capsule and the optical papilla (D2) ( Figure 1B No correlations between the weight and PRTT (p = 0.1419) or the weight and IOP (p = 0.6865) were detected. No correlations between the age and IOP (p = 0.6182) were observed in either breed. However, there was a direct correlation between the age of chickens and PRTT (p = 0.00079) in both breeds.

Ultrasound Biometric Measurements
When an ultrasound is performed in B-mode, the cornea is hyperechoic and convex, and the anterior chamber is anechoic. The lens appears anechoic and ovoid and is delimited by two curvilinear hyperechoic lines corresponding to the anterior and posterior lens capsule. The vitreous chamber is anechoic with the pecten visible as a hyperechoic, millimetric, tubular-shaped structure, projecting into the vitreous chamber from the retina. The retina, the choroid, and the sclera are not ultrasonographically distinguishable since they appear as a single concave hyperechoic line at the posterior limit of the globe. The distance between the cornea and the anterior lens capsule (D1) ( Figure 1A), the distance between the posterior lens capsule and the optical papilla (D2) ( Figure 1B    No correlations between the weight and PRTT (p = 0.1419) or the weight and IOP (p = 0.6865) were detected. No correlations between the age and IOP (p = 0.6182) were observed in either breed. However, there was a direct correlation between the age of chickens and PRTT (p = 0.00079) in both breeds.

Ultrasound Biometric Measurements
When an ultrasound is performed in B-mode, the cornea is hyperechoic and convex, and the anterior chamber is anechoic. The lens appears anechoic and ovoid and is delimited by two curvilinear hyperechoic lines corresponding to the anterior and posterior lens capsule. The vitreous chamber is anechoic with the pecten visible as a hyperechoic, millimetric, tubular-shaped structure, projecting into the vitreous chamber from the retina. The retina, the choroid, and the sclera are not ultrasonographically distinguishable since they appear as a single concave hyperechoic line at the posterior limit of the globe. The distance between the cornea and the anterior lens capsule (D1) ( Figure 1A), the distance between the posterior lens capsule and the optical papilla (D2) ( Figure 1B      No differences were observed in the ocular ultrasound appearance, both within the same breed and between the two breeds. There was a significant difference for D1 between  No differences were observed in the ocular ultrasound appearance, both within the same breed and between the two breeds. There was a significant difference for D1 between the Livorno and the Siciliana breeds (p = 0.0301), while no significant differences were

Conjunctival Bacterial and Fungal Flora Identification
Among the 27 chickens examined, 23 (85.18%) were bacteriologically positive for one or both eyes. Twelve (44.44%) chickens had a positive culture (including presence of Chlamydia spp.) in both eyes, and of these twelve chickens, ten (37.03%) had the same bacterial genus in both eyes.
Chlamydia spp. antigen was detected in both eyes of four (14.81%) chickens. The results obtained for each eye are reported in Table 4.
No positive cultures were obtained for fungi.

Discussion
Although there are many studies regarding the normal values of tear production and intraocular pressure (IOP) in many bird species, in chickens there are few studies with differing results [1,7,29,33].
In the present study, we used a commercially available phenol red thread test (PRTT) kit to determine tear production in chickens for two reasons: firstly, because the palpebral fissure is small and PRTT is easier to use than other tests; secondly, it takes less time (15 s) to get the results compared with the Schirmer Tear Test (1 min).
Our results indicate that the mean normal tear production in adult chickens with PRTT is 21.98 ± 3.46 mm/15 s: 23.77 ± 2.99 mm/15 s in the Livorno breed and 19.95 ± 2.81 mm/15 s in the Siciliana breed, with a significant difference between the two breeds (p < 0.0001).
Although normal tear production in birds can be evaluated with several methods, such as PRTT and STTI, they are not equivalent and reference values should be established for each method [8,9,22,29,34,38,46].
There is also an intraspecific difference among chicken breeds. In fact, we found significant differences between the Livorno and the Siciliana breeds. Tear production in the two breeds was evaluated at the same time and in the same environmental conditions. This indicates that tear production differences were only due to breed.
Although Table 5 highlights that there are many articles reporting tear production values in birds, we can only partially compare our results with those by Fornazari et al. (2018) who performed PRTT in 42-day-old chicks and showed values of 25.58 ± 4.8 mm/15 s. The difference in tear production could be due to different breeds of chicken tested and to environmental factors, such as humidity, temperature, wind, dust, and ammonia levels. Our population and Fornazari's were born and reared in different environments, with different temperatures and humidity. However, we do not know the exact levels of dust and ammonia. In fact, a limitation of our study was the lack of any ammonia level measurements. However, because we used an open-air rearing system, it is likely that the ammonia levels were low and that they did not influence tear production.
This means that it is not possible to compare the results from different devices and between different species. Instead, comparisons are possible when the same type of tonometer is used, with the same calibration, and in the same bird species. We can thus only compare our mean results (14.19 ± 1.16 mmHg) with those of Prasher et al. (2007). They reported three different results (expressed as mean value ± standard error) for three different chicken lines: 16.3 ± 0.2 mmHg for an egg-layer line, 16.1 ± 0.2 mmHg for a broiler line, and 17.51 ± 0.13 mmHg for an advanced layer-broiler intercross line.
Our results and Prasher's (2007) are similar, but do not overlap. This may depend on the different ages and weights between our chicken population and Prasher's. In fact, our animals were more than 39-weeks-old (39 to 53 weeks) and can be considered as young adults/adults, while Prasher's were only 3 weeks old.
Another factor to assess is the time at which the IOP was measured in the subjects examined. In fact, the IOP in normal eyes of chickens is high during the day, and low in the middle of the night [1]. Therefore, the variation between our results and Prasher's could also be due to the different time at which the IOP measurements were performed: from 10:00 to 12:00 a.m. and from 12:00 to 16:00 p.m., respectively.
To the best of our knowledge, there are currently no data on conjunctival bacterial flora in poultry. Previous studies have been limited to the bacteria present in the eyes of wild birds. Our results are in reasonable agreement with those found by Dupont et al. (1994), who studied the bacterial and fungal flora in the healthy eyes of Falconiform and Strigiform raptors from Canada. They isolated bacteria of the genera Bacillus, Corynebacterium, Staphylococcus, Pasteurella, and Enterobacter with prevalences, in some cases, similar to those detected in our investigation. In particular, Gram-positive bacteria were the microorganisms most frequently isolated in both studies, although Dupont et al. found Staphylococcus spp. coagulase negative with the highest prevalence (49.5%).
The detection of these bacterial genera in healthy eyes is not surprising, considering that they are usually non-pathogenic. However, in some circumstances, such as injuries or inflammations, they can cause ocular infections. Some bacteria in healthy eyes seem to play an important role in their defense mechanisms, taking nutrients from invading organisms or secreting substances with antimicrobial properties [27,30,57].
On the other hand, in our study the detection of Chlamydia spp. in the examined eyes suggests the circulation of this pathogen in the environment where the chickens lived. Chlamydia psittaci frequently infects domestic and wild birds without inducing disease. Birds contract chlamydias through oral and/or inhalation, but also through conjunctival contamination by dust. Infected birds can develop conjunctivitis and respiratory and enteric clinical signs, mainly when the birds are stressed [58].
In the present study, no positive fungal cultures were achieved. Our data cannot be compared with similar studies from the literature, since to the best of our knowledge, no papers have investigated fungal flora in chickens. There appear to be only two studies dealing with the occurrence of fungi in birds' eyes and both were carried out on birds of prey. In those studies, Cladosporium spp. and Aspergillus spp. were cultured from one and 2 eyes, respectively, out of 97 subjects [30], while Aspergillus fumigatus and Candida spp.
were obtained from 1 and 2 birds, respectively, out of 51 raptors [34]. The occurrence of fungi in the conjunctivae is believed to be transitory and resulting from environmental exposure. These features could explain their low prevalence [59] considering that the chickens in our study live outdoors and probably there are no environmental conditions for the stabilization of the fungi in the conjunctival sac. Fungi have been more frequently isolated from horses [60][61][62][63][64], donkeys [65,66], cattle [67], dogs [68], rabbits [69], turtles, and tortoises [70] and the genera and species recovered seem to reflect a transient seeding from the environment.

Conclusions
In conclusion, our study performed on clinically healthy chickens reported normal reference range values for selected ophthalmic parameters, including ultrasound biometric measurements. These reference ranges can facilitate accurate diagnosis and better management of ocular diseases, preventing diagnostic misinterpretation during the ophthalmic examination. In healthy chickens we also found evidenced of conjunctival microflora composed of bacteria. Knowledge of the conjunctival flora in chickens is important to increase the accuracy of treatments in the case of corneal injuries. In fact, corneal ulcers can often become secondarily infected by opportunistic flora of the conjunctival sac.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.