1. Introduction
Bovine respiratory disease (BRD) is regarded as the most common illness during the fattening period of cattle in Europe and North America [
1,
2,
3] and the most important cause of economic burden to the feedlot industry due to its high morbidity and mortality [
4,
5]. It has been estimated that it is responsible for 69% of the total casualties in a feedlot [
6,
7]. BRD’s economic costs are primarily due to metaphylactic and therapeutic use of antibiotics, and the loss of weight of affected animals [
1,
8,
9]. The negative effect of BRD on carcass composition and quality traits also has been demonstrated [
9,
10]. Costs can be up to 200 USD per calf [
5] and account for 7% of total production costs in North America, amounting to approximately 1 billion USD [
1,
2,
11].
Pneumonia is the principal feature of the BRD complex. Clinical signs include depressed demeanor, loss of appetite, increment of cough frequency, nasal discharge accompanied by febrile episodes and respiratory insufficiency [
12]. Pneumonia associated with BRD is considered a multifactorial disease resulting from interactions between infectious agents, such as bacteria or viruses [
4,
13], and extrinsic factors such as stress caused by deficient climatic conditions, dehorning, weaning, transportation or immunosuppression periods caused by other viral agents [
14,
15,
16].
In a study performed in Canadian feedlots, the most common infectious agents of bacterial etiology found in pneumonic lungs by polymerase chain reaction (PCR) were
Mycoplasma bovis (36%),
Mannheimia haemolytica (27%),
Pasteurella multocida (19%) and bovine viral diarrhea virus (BVD) (35%), bovine respiratory syncytial virus (BSRV) (9%), bovine herpesvirus serotype 1 (BHV-1) (6%) and parainfluenza virus serotype 3 (PI-3) (3%) regarding viruses [
6]. Mixed infections are frequently identified in BRD pneumonia, and the etiology of the infection is highly related to the type of pulmonary lesions and duration [
6,
8,
17,
18].
Assessment of the prevalence of pneumonic lesions at the slaughterhouse can be a good indicator of the relevance of BRD in beef cattle [
3,
19,
20]. However, there is a shortage of studies that assess the relationship between the presence of lesions within the individual with productive factors that can influence them at the end-point of fattening efficiency. There are scant works in relation to BRD in Europe [
19,
21] and particularly in Spain, where, despite the importance of the beef industry, which comprises approximately 2 million individuals [
22], only a few descriptive data of some local administration are available [
23].
The objectives of this research were to investigate the prevalence of pneumonia in clinically healthy veal calves and yearlings based on the examination of lesions related to BRD at the slaughterhouse and to determine the effect of production system, age, sex and the impact of subclinical pneumonia on weight. In addition, this study aimed to classify BRD-related pneumonia according to the type of lesion and also to identify, via PCR, the infectious causes associated with the different types.
2. Material and Methods
Experimental animals were not used in this work. An observational study has been performed with the data and with post mortem samples that are routinely collected in a slaughterhouse.
2.1. Animals
A total of 1101 beef-breed cattle, intended for consumption and without previous respiratory clinical signs, were post-mortem examined at an authorized slaughterhouse under the European Union regulations. All the animals came from feedlots located in Castile and León (Northwestern Spain). The visits at the slaughterhouse were carried out during four consecutive months (from September to December 2017). The animals participating in the study were subjected to the strictest traceability measures, and individual data were also provided by the feedlot administration. Upon arrival at the slaughterhouse, individual identification was verified. Animals were examined at arrival by official veterinarians in accordance with all the veterinary criteria scheduled for the antemortem examination included in Spanish legislation. The absence of clinical symptoms was established according to feedlot records and the antemortem veterinary official examination.
All animals under the study were categorized according to the most common production systems for beef cattle in Spain: intensively managed (animals were kept indoors with controlled feeding and no access to pasture), extensive management (animals remained in the pasture all the time) and mixed system (animals were on pasture for variable periods of their feeding and grazing seasons, but indoors the rest of the time). In every case, data related to the management system, gender and age type (veal calves up to 12 months old; yearlings between 12 and 24 months old) were recorded (
Table 1). For each animal, carcass weight was also measured after evisceration in the slaughterhouse. Detailed information on animal identification, weight and farming type was provided by the slaughterhouse veterinarians and the Livestock Department of the Castile and León regional government.
2.2. Sampling
Samples from the affected areas on every lung showing gross lesions were collected according to standardized procedures at the slaughterhouse and good laboratory practices to avoid contamination and ensure data quality. Lung samples were taken and stored in individual sterile freezer bags at −80 °C until analysis by PCR assay. Samples from the same areas were fixed in 10% buffered formalin for 48 h and dehydrated through a graded alcohol series before being embedded in paraffin wax. Cut sections (3.5 μm) were obtained from each sample and stained with hematoxylin and eosin (HE) for histological examination.
2.3. Macroscopic and Microscopic Inspection of the Lungs
Initial assessment of the lungs comprised the evaluation of the distribution and location of macroscopic changes consistent with pneumonia such as variations in color (from red to grey), presence of consolidation areas, or exudate. Examination was performed under the agreement of two pathologists (MF and MCF) in accordance with the guidelines and criteria for classification of bovine pneumonic lesions [
24,
25]. All the lungs with gross lesions, subjected to the same traceability as their carcass, were separated for more detailed examination, including cross-sectioning to ensure the optimal macroscopic evaluation and data control. Macroscopic lesions were post-hoc classified into two different types: chronic catarrhal, characterized by well-demarcated, purple to grey in color, cranioventral consolidated areas, with firm texture and no increase of volume; and acute fibrinous pneumonia, characterized by the presence of well-demarcated solid and swollen cranioventral areas with evident vascular reaction, such as congestion or fibrin deposition over the pleura of the affected parts. To classify the lesion expanse, a subjective assessment using a previously established scoring system was performed [
2]. Grade 1 or minor pneumonia was assigned to those cases in which the affected area was equal to or less than approximately 10% of the lung and grade 2 or extent pulmonary disease, in which the lesion concerned more than 10% of the pulmonary parenchyma.
Sections were microscopically analyzed and evaluated rigorously and independently by MF and MCF or, in case of disagreement, by VP. Histologically, lesions were post-hoc classified into two main types: chronic catarrhal and acute fibrinous pneumonia. The guidelines and microscopic findings already described and proposed for characterization of bovine pneumonic lesion [
25,
26] were employed.
2.4. Etiological Identification by PCR
This technique was only performed on a representative number of samples, chosen according and proportionally to the different type of lesion found. Etiological identification by RT-PCR was performed on a total of 50 randomly chosen samples, representative of each type of pneumonia: 40 from chronic catarrhal bronchopneumonia lesions and the remaining 10 from the acute fibrinous pneumonia group.
DNA extraction was carried out under sterile conditions in a vertical laminar flow cabinet. Tissue (25 mg) from each sample was cut with disposable sterile blades into small pieces and put into sterile Eppendorf tubes. Thereafter, the DNA and RNA extraction was performed using the Speedtools Tissue DNA Extraction Kit® (Biotools B&M Labs S.A, Madrid, Spain) and the Speedtools Total RNA Extraction kit® (Biotools B&M Labs S.A, Madrid, Spain), respectively, according to the manufacturer’s instructions. Eluted DNA was stored in a sterile eppendorf at −20 °C and total RNA at −80 °C until analysis by real-time PCR (RT-PCR).
Etiological agents assessed were those most commonly reported in BRD cases [
4,
18]. RT-PCR amplification of genomic regions of
Mycoplasma bovis,
Histophilus somni, Mannheimia haemolytica, Pasteurella multocida, Bovine Herpesvirus type 1 (BHV-1), Bovine Viral Diarrhea Disease Virus (BVDV), Bovine Respiratory Syncytial Virus (BRSV) and Parainfluenza Virus type 3 (PI-3) was carried out using commercial kits according to the instructions of each manufacturer on a conventional thermocycler ABI 7500 Real-Time PCR System (Applied Biosystems
®, Foster City, CA, USA) with the corresponding cycling parameters (
Supplementary Table S1). Samples were processed in duplicate for each kit, all at once, with the same equipment and by the same person (MF) to assure the laboratory measurements and avoid variability.
2.5. Statistical Analysis
The inter-observer agreements for the histopathological classification of pneumonia and the expanse of lesions were calculated through the weighted kappa (wκ) and Cohen’s kappa (κ) statistics, respectively.
Several models to fit the logit of the odds (log-probability of the event/probability of the no event) of different events (pneumonia vs. healthy animals, acute fibrinous vs. chronic pneumonia or severe vs. mild lesions) were constructed and tested using the GLIMMIX procedure of SAS version 9.4 (SAS Institute Inc, Cary, NC, USA). The models included the system of production (intensive, mixed and extensive), animal type (veal vs. yearling) and sex (male vs. female) as fixed effects and the farm nested to the system as a random effect. When the severity of the lesion was evaluated, the type of pneumonia was also included as fixed effect in the model. When possible, models including double interactions were tested, and those models whose Pearson chi-square/DF value was nearest to 1 were selected, in order to avoid the effect of overdispersion on probability values. Random effect of the farm (system) was dropped from the model when its variance was zero. Adaptive quadrature of the Gauss–Hermite method was used for computing the maximum likelihood (true). Pearson chi-square values were 0.99, 0.96 and 0.99 for the final selected to predict the log odds of pneumonia, type of pneumonia (acute fibrinous) and severity of lesions (grade II), respectively.
Carcass weight and age-at-slaughter data were subjected to analysis of variance using the MIXED procedure of SAS. Health status, production system, type of animal and sex were included as fixed effects in the statistical model, farm nested to system being included as a random effect. Double and triple interactions were also included in the models and dropped when the p-value was greater than 0.20. In order to get as much information as possible, a second analysis was performed to study separately the effects of the type of pneumonia and severity of the lesion.
4. Discussion
BRD is considered a common disease in livestock, particularly in beef feedlots, where it has been related to poor fattening and severe economic losses [
1,
2,
4]. This has been corroborated by the present study, where it has been shown that BRD-related pneumonic lesions have a significant prevalence (17.9% in this study) in slaughtered, clinically healthy beef cattle and are associated with a reduction in carcass weight and an increase in fattening days in the different management system studied.
Previous reports have stated that the prevalence of pneumonia varies depending on country, management system, etiological agents involved, breed or season entering the feedlot [
4,
5,
6,
7,
20,
21]. The findings of the present study suggest that BRD prevalence in Spain is similar to that previously described in different countries [
6,
7,
26] and close to the data gathered in post-mortem studies in the US [
19,
25] and Europe [
20,
27]. The high rate of BRD lesions in animals with apparently healthy status at the slaughterhouse time point found in this study supports preceding publications that have pointed out that only 25% of the animals suffering severe BRD lesions would show clinical signs of the disease [
2,
14].
These results reflect that BRD is a multifactorial syndrome linked to the influence of several variables. With respect to age, the prevalence tends to be higher in veal calves than in yearlings. This could be associated with a diversity of factors such as their naïve immune system, frequent contact with newly introduced animals in the feedlot, prophylaxis mistakes, poor weaning period and different production systems that seem to affect managed veal calves more intensely than yearlings [
14,
27,
28].
Some authors suggested that intensive co-housing boosts the exchange of airborne pathogens and facilitates nose-to-nose contact between calves, potentially increasing the risk of BRD [
29,
30]. However, in this study, when global data were considered, the highest prevalence rates of subclinical BRD were found in veal calves raised in mixed and extensive management systems. This fact could be due to some circumstances such as the mix of animals of different ages [
16], the existence of less controlled environmental conditions and poor daily observation [
28,
31] that would occur more frequently in mixed–extensive than in intensive management systems. All these aspects deserve further investigation since, at present, the majority of studies that deal with the impact of production systems on pneumonia focus on intensive systems, and no information on the relevance of other managements on BRD is available, highlighting the significance of this work. The season of entry to feedlot and the time of birth in the extensively reared animals could be a factor to take into account that can influence the prevalence of pneumonia [
32], particularly if the study were carried out at a single point. However, this study covers a long period and includes two types of animals with greater or lesser age (veals or yearlings). Thus, it does not seem likely that the season of entry to feedlot would have influenced the results since the animals included cover all times of the year.
In this study, differences were observed in the prevalence of BRD in relation to the sex of the animal on the whole. According to the results, pneumonia could be found with higher probability in males. However, when sex was analyzed together with the type of management and age, it was striking that BRD prevailed among females from the mixed and extensive systems. This could be likely related to several factors including hide thickness, fat depth, temperament, and response to stimuli that would make females prone to respiratory disease under these management circumstances [
33]. In relation to the effect that the presence of lesions due to a subclinical pneumonia may have on carcass weight, the results show a decrease similar to other studies including animals with a clinical history of pneumonia [
34]. According to these results, chronic catarrhal pneumonia is the most prevalent type found in slaughterhouse animals regardless of age group, gender or management systems. This was an expected finding since studied animals were fit for consumption without evident clinical signs, as occurs mostly in this type of chronic pneumonia [
7,
11]. Nevertheless, acute fibrinous pneumonia usually has an acute course with clinical signs such as fever or respiratory distress [
12] that were not observed in the present work, probably due to the limited extent of the fibrinous pneumonic lesions found [
7,
11]. Among mixed or extensively managed animals, acute fibrinous pneumonia acquired more relevance, a fact that could be linked to the environmental conditions or an insufficient observation by farmers that takes place in this type of system [
4,
16,
28].
In relation to the extent of lesions, differences were only found in animals with chronic catarrhal pneumonia, where those with grade 1 were the most frequent. This finding might be explained by the fact that a high number of cattle with this type of lesion showed changes associated with recovery as denotes the presence of areas of atelectasis, always with a low entity, affecting reduced areas of parenchyma. This could imply that chronic lesions are the consequence of events experienced in the past, mainly in calfhood [
20,
26], in which lesions tend to recover with tissue reorganization and are reduced in terms of extent and severity [
35]. However, no differences in lesion severity were found in cases of acute fibrinous pneumonia. Lesions never affected areas greater than 30% and were unrelated to clinical signs. Thus, it would seem feasible that in some animals, always in low numbers, stress or other factors perhaps related to the management [
20,
26] could have complicated and exacerbated a previously existing chronic catarrhal pneumonia, leading to a more severe acute form.
One noticeable finding of this study is the demonstration of the association of pneumonic lesions with a lower carcass weight in animals with the same average days on feeding. Supporting these results, it has been reported that subclinical BRD could cause a decrease between 0.125 to 0.350 kg in the average daily gain (ADG) during the finish feeding [
20,
36]. In addition, animals with pneumonia arrive at the slaughterhouse with less age than the healthy ones, which would aggravate the decrease in the average weight loss. The early slaughter of these animals could be due to the decrease in ADG due to a pneumonic process and a decrease in the daily conversion rate that leads farmers to an earlier slaughter [
34,
36], a fact observed among intensive-production animals, where decisions are usually made based on productive parameters.
Specifically, the significant reduction of the carcass weight was observed only in intensively reared males. This could be related to the fact that extrinsic factors affecting the ADG, such as pneumonia, would have a greater impact in intensive systems where males could reach the highest productive performance [
1,
32,
34]. Under extensive management conditions, the ADG is lower, and for pneumonia to have an effect, it should be associated with the extent of the lesion.
Indeed, when considering the macroscopic characteristics of the lesions, it was surprising to find that there were no differences in the influence of the different types of pneumonia on the carcass weight, even with those characterized by an acute course. This could be associated with the subclinical nature of all the lesions found and the limited extension of those acute cases whose clinical effect is related to the extent of pulmonary damage, and thus, cattle with mild disease would resemble the healthy cattle [
37]. The pathogen profiles regarding the type of the lesion are difficult to compare, since previous studies did not consider the different types of pneumonic lesions; rather, they only took into account the existence of consolidation in the lungs [
38].
It is clear that infectious agents play a key role in BRD pathogenesis [
13,
31], as demonstrated in this study. In Spain there are few publications regarding the etiology involved in BRD, and most of them are based on serological or swab PCR approaches. In this study, all of the pathogens screened by PCR were detected. Although PCR only indicates the presence of nucleic acids, it should be noted that samples came from affected tissue, suggesting that the pathology present was linked to the pathogen identified [
39]. In the effort to avoid variability, although an intra-rater agreement calculation was not made, samples were processed in duplicate, at the same time, by a single person and with the same equipment. In case of an inconclusive result, an additional test was performed.
In the current study, at the time of slaughter, bacterial agents were more commonly found than viruses in relation to BRD lesions. Particularly, among the bacteria and in agreement with other studies [
4,
11,
40],
Mycoplasma bovis and
Mannheimia haemolytica have been shown to be the most prevalent, either alone or coexisting with other bacteria or viruses. Besides, the importance of mixed infections with
Mannheimia haemolytica and
Mycoplasma bovis, also found by previous works, should be highlighted [
18,
41]. Other pathogens also reported as participating in BRD [
4,
17] such as
Pasteurella multocida or
Histophilus somni, have also been found in the current work, either alone or, more often, in coinfections with other agents. Our findings share with other studies the relevant involvement of
Mycoplasma bovis in chronic bronchopneumonic lesions but also in acute fibrinous cases in combination with
Mannheimia haemolytica [
17,
19]. Concerning viruses, finding such a low number of positive identifications, especially as unique agents was not expected, which is in line with some studies in the United States that have reported that viruses such as BRSV or BVDV are found in 14.1% of calves showing clinical respiratory disease [
4,
7,
17] or 9% in macroscopic lesions at the point of cull [
6]. The exact role of viruses on the pathogenesis of BRD is still unclear. It has been suggested that they could interfere with lung defense mechanisms and predispose calves to bacterial pneumonia [
15,
19]. It is then possible that viruses might have participated in the initiation of the lung lesions found in this study, but once the bacteria colonized the damaged tissue, most viruses were eliminated by the immune response of the host. The fact that viral infections could play a secondary role in BRD in this region is a possibility. This study has been conducted in the largest region of Spain with a significant beef cattle industry where, at this moment, no previous studies on the etiology of BRD have been conducted. In the present study, the only relation found between etiological agents assessed by PCR and the type of pneumonia was that acute fibrinous pneumonia was caused only by bacterial agents, mainly
Mannheimia haemolytica, in agreement with previous reports that found this bacterium as the unique agent of this type of lesion [
6]. In chronic catarrhal pneumonia cases, both bacteria and viruses were present, similar to what is described by previous studies [
4,
6,
41]. Likewise, there was not a clear relationship between the infectious agent identified in the samples and the grade of macroscopic lesions.