A Retrospective Analysis to Identify Epidemiologic Patterns of the Infectious Coryza Outbreak in California 2016–22 Open Access

Avibacterium paragallinarum (AP), the pathogen responsible for infectious coryza (IC), is a nonmotile, pleomorphic, Gram-negative bacterium that belongs to the Pasteurellaceae family. Natural infections occur in chickens of all ages, with increased severity of clinical disease in mature birds. The main reservoirs of IC are chronic asymptomatic carriers, which are typically subclinically infected chickens that recovered from acute disease. Exposure to physiologic and environmental stress factors can precipitate bacterial shedding from the respiratory tract and oral cavity (1). The disease is spread by horizontal transmission via direct contact with infected or carrier birds and/or indirect contact by aerosolization of AP and contamination of water and feed (2,3). Consequently, direct and indirect routes of transmission of IC from older asymptomatic carriers to successive age groups are key components for recurrence of IC outbreaks in multi-age production systems (2,4).

The incubation period for this disease is relatively short at approximately 24–72 hr. Clinical signs of IC are mainly acute upper respiratory signs such as facial edema, nasal discharge, and conjunctivitis. IC can also result in a 10%–40% reduction in egg production in layers, and disease in susceptible broilers can result in severe mortality due to airsacculitis. In broilers, production losses are associated with increased condemnation, commonly due to airsacculitis at processing, especially when co-infections are present (2).

Diagnosis of IC is achieved via AP isolation and identification. While the gold standard is isolation and identification, PCR detection from choanal, nasal, or tracheal swabs has been incorporated into the detection tests, most importantly to identify isolates recovered from culture (2,4).

A previous study of the 2017 IC outbreak in central California analyzed 54 positive cases in the California Animal Health and Food Safety (CAHFS) laboratory database (4). Within those cases, the majority were commercial broilers followed by layers and backyard chickens. Additionally, most of the layer cases were localized in Merced County and Stanislaus County, and broiler cases were localized in Merced County. In contrast, the backyard cases were localized in Alameda County and Sacramento County (4). Although these findings are confounded by variables such as IC vaccination in layers and potentially increased representation of broilers due to increased submissions to the CAHFS system, there is still value in evaluating IC trends within the CAHFS database as positive IC cases have remained elevated when compared to the number of cases seen in 2016. Additionally, IC remains a significant disease throughout North America: Current outbreaks are occurring throughout the midwestern states, and outbreaks have now reached Iowa as of September 2023 (M. El-Gazzar, pers. comm., February 23, 2024). This current retrospective analysis aimed to evaluate a larger sample size of positive IC cases over a longer period (7 yr) to expand the assessment of the epidemiologic behavior of IC and identify significant risk factors involved.

Cases were obtained from the CAHFS database from January 1, 2016, through December 31, 2022. A case was defined as a sample submitted to our laboratory from chickens with clinical signs suggestive of IC, where samples were positive for AP via isolation and identification and/or PCR. Each case submission at CAHFS can comprise up to 8 chickens from the same flock for commercial poultry cases. Backyard bird submissions can consist of up to 2 chickens per case.

The date, age, poultry production type, and the originating county were obtained from the submission form associated with each case. For seasonality analysis, the following definitions were used for each season: winter (December–February), spring (March–May), summer (June–August), and fall (September–November). Poultry types consisted of backyard poultry, commercial broilers, and commercial layers.

Bacterial isolation techniques previously described by Crispo et al. were used to culture AP (4). In summary, based on antemortem and postmortem findings, swabs were collected mainly from the sinus and trachea. Other samples or swabs included the coelomic cavity, lungs, conjunctiva, nasal cavity, respiratory exudate, pericardium, oropharynx, and air sacs. These samples were then cultured on MacConkey agar, 5% sheep blood agar, and chocolate blood agar that were cross streaked with Staphylococcus aureus ATCC 25923. The plates were then incubated at 37 C and 7% CO₂ for 48 hr. Colonies consistent with AP were then confirmed via PCR as previously described (4,5,6).

For cases determined to be positive via PCR alone, and in some cases in conjunction with bacterial isolation, samples from the infraorbital sinus and trachea were collected in 1-mL vials of phosphate-buffered solution. Real-time PCR was then performed on these samples using a primer-probe assay as previously described (7,8).

To investigate the risk factors associated with the occurrence of IC in California from 2016 to 2022, a two-step approach using a univariable and a multivariable logistic regression model was performed. Univariable analysis was conducted to test the association between a positive IC diagnosis and an explanatory variable suspected as a risk factor. Explanatory variables included age, seasonality, location, and poultry type. Multivariable logistic regression was then applied to the selection of explanatory variables that were statistically significant in the univariable analysis (9). A forward stepwise variable selection was used to add the variable with the lowest p value to construct a final model with a significance level of p ≤ 0.05. The models were compared using a likelihood-ratio test after the addition of each variable. Model fit was evaluated using the Hosmer-Lemeshow test. Odds ratios (ORs), 95% confidence intervals (CIs), and p values were estimated using maximum likelihood methods. Analysis was performed using SPSS Statistics 29.0 (10).

The average age was 37.6 days ± 10.1 days for affected commercial broilers and 41.8 wk ± 24.6 wk for affected commercial layers. The average age for backyard poultry was not included as many of the accessions had unknown ages or multiple ages within one submission. Additionally, the ages that were known for backyard submissions varied greatly from weeks to years.

There were 674 IC-positive cases identified in the CAHFS database from 2016 to 2022. Most of the positive case submissions were from backyard flocks (50.6%, 341/674) followed by commercial layers (31.5%, 212/674) and finally commercial broilers (18.0%, 121/674).

Fig. 1 shows the spatial distribution of backyard and commercial poultry cases throughout California. IC-positive birds from backyard flock cases were detected in 42 California counties, but most case submissions were from Los Angeles County (n = 37), San Joaquin County (n = 26), and Riverside County (n = 24) (Fig. 1a). Most commercial broiler cases were localized to Merced County (n = 100); commercial layer cases were mainly found in Merced County (n = 72) followed by Stanislaus County (n = 50) (Fig. 1b). The percentage of each poultry type tested for IC in Northern California, Central Valley, and Southern California was also calculated. In Northern California, backyard poultry made up 82.6% (252/305) of the submissions tested, and commercial broilers and layers made up 0.7% (2/305) and 16.7% (51/305), respectively. In the Central Valley, the respective percentages composition for backyard poultry, commercial broilers, and commercial layers were 26.5% (335/1264), 48.7% (615/1264), and 24.8% (314/1264). In Southern California, backyard poultry, commercial broilers, and commercial layers accounted for 78.7% (288/366), 0.27% (1/366), and 21.0% (77/366) of positive cases, respectively.

Fig. 2a shows that most of the cases occurred during spring (33.0%) followed by summer (29.0%), fall (22.0%), and then winter (16.0%). When further divided by poultry type, there was a slight variation in this seasonality trend. For backyard flocks, most of the cases occurred in the summer; however, the majority of both commercial layer and broiler cases occurred in the spring (Fig. 2b). The total number of IC tests performed, irrespective of their results, was also assessed per season. Testing was the highest in the spring (35.2%) followed by the summer (28.3%), fall (20.2%), and then winter (16.3%).

Figure 3 shows submission trends associated with IC over the years. Positive and negative results are differentiated by colors. There has been a general upward trend in testing in all three poultry types. This is best shown in 2018, when 351 tests were performed on backyard birds with only 90 (25.6%) being positive for IC.

For statistical analysis, the number of IC tests performed (n = 1935) and the number of positive IC cases (n = 674) were evaluated for four explanatory variables (age, poultry type, spatial distribution, and seasonality) in a univariable logistic regression model. This was used to select statistically significant variables to include in the final model (i.e., multivariable logistic regression model). In the univariable logistic regression model, age, poultry type, and location were significantly associated with IC (p < 0.05). Commercial layers were 1.4 times more likely to have IC than backyard poultry (95% CI 1.14–1.81, OR = 1.4, p < 0.002), and birds from Southern California were two times more likely to have IC than birds located in Northern California (95% CI 1.50–2.80, OR = 1.4, p < 0.001). Last, birds 5 mo or older were 2.30 times more likely to have IC than birds less than 5 mo old (95% CI 1.50–2.80, OR = 2.3, p < 0.001) (Table 1).

The explanatory variables that were statistically significant in the univariable model (i.e., age, poultry type, location) were then placed into a multivariable logistic regression model. This was used to determine if poultry type with age and poultry type with location were significantly associated with an IC diagnosis. Poultry type was used to construct the final model as this was the variable with the lowest p value. In this model, poultry type and location were statistically significant variables, and age was no longer significant.

The multivariable logistic regression model showed that commercial layers were 1.3 times more likely to have IC than backyard poultry (95% CI 1.01–1.65, OR = 1.3, p = 0.039); in contrast, commercial broilers were less likely to have IC than both backyard poultry and layers (95% CI 0.24–0.42, OR = 0.31, p < 0.001).

When analyzing location, samples from Central Valley (95% CI 1.50–2.73, OR = 2.00, p < 0.001) and Southern California (95% CI 1.48–2.79, OR = 2.00, p < 0.001) were 2 times more likely to have IC than those from Northern California (Table 2).

This retrospective analysis evaluated 674 IC-positive cases found in the CAHFS database from 2016 through 2022 and categorized positive cases by age, poultry type, seasonality, and location.

Age for each of the cases was analyzed because IC can affect chickens of all ages (2), and this study aimed to determine if either juvenile or adult birds were at an increased risk for IC infections. For this study, birds of all poultry types 5 mo or older were categorized as adults, and birds less than 5 mo old were considered juveniles. This age cutoff was used since laying hens begin to produce eggs at approximately 20 wk of age (11). It is important to note that this definition placed all broiler cases evaluated into the juvenile category. This is because broiler chickens are usually marketed prior to 2 mo of age (12). In the univariable analysis, adult chickens were 2.3 times more likely than juvenile birds to be positive for IC (p < 0.001); however, age was insignificant in the multivariable analysis when included with poultry type. These results may be due to the particularities described below. As seen in Table 1, the ages of 244 of the 1935 submissions tested for IC were unknown, with most of these unknown ages originating from backyard poultry submissions. Additionally, in multi-age layer complexes, affected birds may range in age, but only a subset of these is submitted for testing, thus not representing all the affected ages. Another factor to consider is that age is strongly associated with poultry type. Broiler chickens will always be categorized as juvenile given the definition used for this investigation. Finally, an overrepresentation of broilers/juveniles can be seen due to the large numbers of submissions of broiler/juvenile samples to CAHFS compared with layer-type birds.

Another area of focus of this study was poultry type and location. Studies have shown that backyard poultry may act as a reservoir, perpetuating poultry diseases (13). Given the concern for backyard poultry being a reservoir for pathogens, cases were characterized by poultry type and location in this current study. The results showed that layer-type birds were more likely to have IC than backyard poultry, while broilers were the least likely. Considering that the major reservoir for IC transmission is asymptomatic carriers, layers are likely more at risk for IC than backyard poultry because they are housed at higher density than backyard birds, and unlike broiler production systems, layers are often in multi-age facilities that allow for the existence of more asymptomatic carriers (2,3,4).

While not analyzed as a risk factor in this study, proper vaccination should be part of a successful IC prevention and control strategy. In California, while most producers use two doses of killed IC products, either autogenous or commercial, or a combination, there are still some producers using only one dose (R. Gallardo pers. comm.). This practice might contribute to increased bacterial shedding by infected birds, posing a higher risk to neighboring flocks. The results of this study suggest that a revision of these vaccination programs, not only to the doses but also application method effectiveness, is needed. The increased incidence in commercial layers does not eliminate the possibility that backyard birds might be important in the introduction of the disease to commercial facilities, due to the proven absence of biosecurity and their proximity. When analyzing spatial distribution, results demonstrated that Central Valley and Southern California were more likely to have IC-positive cases than Northern California. This can likely be explained by the fact that the top poultry production counties within California are located in the Central Valley (14). In contrast, Southern California has a large population of backyard poultry. This was shown in a 2013 retrospective study that evaluated CAHFS submissions from 2007 to 2012. It was concluded that the county with the most backyard poultry submissions was Los Angeles County (15). Therefore, it is reasonable for the Central Valley and Southern California to show more IC-positive cases because they are locations in which poultry are more prevalent. Another consideration is the location of CAHFS laboratory branches, since proximity to clients can have an impact on the likelihood of submissions due to both time and logistic constraints. There are currently four CAHFS laboratory branches, three located within the Central Valley (i.e., Yolo County, Stanislaus County, Tulare County) and one in Southern California (i.e., San Bernardino County).

The last variable assessed was seasonality, which was not shown to be statistically significant in this study. A 2023 systematic review of the association between particulate matter and respiratory disease discussed how ventilation frequency is often reduced in the winter; thus, higher concentrations of particulates such as ammonia are increased in intensive operations and can compromise the respiratory tract and predispose a host to respiratory disease (16,17,18). Therefore, it is known that seasonality can indirectly impact the prevalence of respiratory pathogens. Another driver to investigate seasonality trends was to better understand if IC is significantly affected by environmental changes such as humidity or temperature changes. AP is not considered to be an environmentally persistent bacterium. It has been shown to be inactivated within 4 hr when in tap water at room temperature. Additionally, this bacterium can be inactivated within 2 to 10 min of exposure to temperatures of 45 C to 55 C (2). Therefore, evaluating seasonality trends may offer insight on seasonal conditions that increase the persistence of AP in an environment. In the study of the 2017 IC outbreak (4), positive IC cases were more prevalent in the winter and spring months for layers, and most broiler cases were identified in May. The lack of statistical significance between seasonality and an IC diagnosis in this study was likely affected by the representation of poultry types included, where backyard poultry consisted of 50.6% of the submissions compared with 5.5% in the 2017 study. Backyard poultry operations have lower bird stocking densities, different enclosure designs, and different cleaning and disinfection protocols than commercial operations. These differences in management can have a profound impact on disease risk and transmission and can allow for backyard poultry to not be influenced by seasonality in the same way as are intensive commercial operations.

In addition, as mentioned previously, longer-lived birds (i.e., layers and backyard poultry) provide the means for the development of chronic asymptomatic carriers (2). In contrast to acute IC cases, chronic asymptomatic carriers will not be affected by seasonality. Therefore, the inclusion of backyard poultry and commercial layers may have affected the results and may not provide a true reflection of the seasonality of acute IC infections.

The other variable to consider is that all the cases in this study originated from the CAHFS database, and the data were extrapolated to represent the entirety of California. However, as discussed earlier, the likelihood of submissions may be influenced by proximity to a CAHFS laboratory branch, and thus some areas of California may be overrepresented. Additionally, analysis of seasonality trends in California in general has limitations. The climate throughout California varies drastically based on distance from the Pacific Ocean, elevation level, and latitude, and thus even within this state, seasonal temperatures and humidity can be vastly different (19). However, IC is endemic in California due to temperate weather throughout the whole state. Additionally, biosecurity measures due to current highly pathogenic avian influenza cases might influence IC risk (20,21). Another factor to consider is that by using the CAHFS database, the amount of positive IC cases can be affected by the number of submissions that CAHFS receives per season and during outbreaks of other pathogens. This is supported by comparison of the seasonality of positive IC cases to the seasonality of the total number of submissions tested for IC. Interestingly, both follow very similar seasonality trends, with the majority for both occurring in the spring (positive IC cases: 33.0.99%, total IC tests: 35.2%), followed by the summer (positive IC cases: 29.0%, total IC tests: 28.3%), fall (positive IC cases: 22.0%, total IC tests: 20.2%), and then winter (positive IC cases: 16.0%, total IC tests: 16.3%). Overall, this study demonstrated that birds in layer operations and birds located in Central Valley and Southern California locations are at an increased risk for IC. In contrast to commercial poultry, backyard flocks concentrate most of the detections in summer. This might be related with the poultry show season in California, which runs from November to May, allowing for increased submissions in the summer, when clients are able to better monitor their birds for disease and submit samples to the laboratory.

Other limitations to this study, which afford areas of future research, are vaccination history, litter management practices, and biosecurity practices, which all can impact disease risk. Future research should focus on cases that include this type of information at the time of submission in addition to biosecurity surveys of commercial and backyard operations that have submitted IC-positive samples. This study provided an update on the epidemiologic patterns of an IC outbreak in California and allows for comparison of results to those reported in other geographic locations and production settings. The results of this study highlight the need for effective biosecurity, vaccination, and management to control and further prevent the spread of IC in poultry populations.

We would like to acknowledge CAHFS for providing the data to investigate these trends, specifically Jamie Nunez for her help in retrieving the data.

AP =

Avibacterium paragallinarum;

CAHFS =

California Animal Health and Food Safety Laboratory System;

CI =

confidence interval;

IC =

infectious coryza;

OR =

odds ratio