Major poultry respiratory diseases, such as avian influenza virus (AIV), infectious bronchitis virus (IBV), infectious laryngotracheitis virus (ILTV), Newcastle disease virus (NDV), and avian metapneumoviruses (AMPV), are ubiquitous worldwide, causing significant economic losses in the commercial poultry sector. These diseases can be rapidly disseminated within flocks because such diseases utilize the respiratory tract as the primary route of infection, causing high morbidity and mortality. They can also adversely affect bird performance, leading to various management issues. AIV, belonging to the family Orthomyxoviridae, is one of the major respiratory viruses prevalent in birds that can mutate rapidly, leading to interspecies transmission that affects humans and other mammalian species [1-5].

The viral genome of these single-stranded, negatively sensed RNA viruses is segmented, allowing the easy exchange of genetic material and causing antigenic shifts and drifts, making them difficult to combat. The hemagglutinin (HA) and neuraminidase (NA) proteins are responsible for viral entry and progeny release within cells. These proteins are also the primary targets of antigenic determination and identification. The highly pathogenic form of AIV (HPAI) is the most profound concern because it is prevalent mainly in migratory wild birds, posing significant threats to the domestic poultry industry [6]. The most popular HPAI H5N1 subtype alone had 878 cases documented from 2002 to 2023 [3]. The IBV is a highly contagious virus that primarily affects the upper respiratory tract of chickens but is also capable of causing nephropathogenecity, depending on the strain. As a member of the Coronaviridae family, genus Gammacoronavirus, this positive single-stranded RNA virus contains a hypervariable region within the spike (S) protein, particularly in the S1 regions, leading to the generation of variants [7,8]. The IBV has worldwide prevalence, affecting almost every continent [9]. Their high mutation rates complicate intervention strategies against these viruses [10], like those against the AIV, although infection is mainly limited to chickens. As indicated by its name, ILTV infections are confined to the upper respiratory tract. Belonging to the Gallid herpesvirus-1 family, the clinical manifestations are usually acute and exhibit a narrow host range [11-13]. The clinical signs include the development of hemorrhagic conjunctivitis and bloodstained mucous, with varying mortality rates. The NDV is part of the Paramyxoviridae family known as Avian avulavirus-1. The host range of NDV is quite diverse, affecting up to 236 species, and cases have been found in many species of birds, including domestic poultry species [14]. The mortality can reach up to 100% depending on the virulence of the infected strain. In particular, new cases of NDV have emerged in North America in 2019 [15]. Lastly, the AMPV, a virus from the Paramyxoviridae family, is prevalent in turkeys, but it was later confirmed that wild birds and chickens were also susceptible [16-20]. AMPVs can exacerbate co-infections with other bacterial pathogens, leading to high morbidity and mortality [21]. Recently, in addition to the re-emergence of traditional subtypes [18,22], new serotypes of AMPV have been confirmed [23,24].

The timely detection and control of these viral diseases is essential because they can be devastating as stand-alone pathogens and aggravate the pathogenicity of other diseases in the event of co-infection. Moreover, uncontrolled outbreaks of these viruses can lead to trade restrictions and embargoes, significantly impacting a nation’s economy.

Controlling these diseases requires demonstrating the presence of the virus and identifying the etiologic agent. Furthermore, differentiating between these viruses is crucial, as they can exhibit similar early-stage pathogenesis [25]. Over the past decades, diagnostic methodologies have advanced significantly, with several methods being developed, including detecting or isolating the viral agent and targeting specific antibody responses. Table 1 lists the most diagnostic methodologies [26-37].

Virus isolation (VI) is a conventional gold standard for routine use (Fig. 1). The two most widely used methods are propagating the virus in cell culture or embryonated chicken eggs. Embryonated chicken eggs are generally used to isolate the AIV, IBV, NDV, and AMPV [38-41]. In contrast, the ILTV can be isolated in cell cultures and embryonated chicken eggs [42,43]. The inoculation site in embryonated eggs for the AIV, IBV, and NDV is the allantoic cavity [8,40,44], while the AMPV uses the yolk sac [45], and the ILTV targets the chorioallantoic membrane (CAM) [46]. In ILTV diagnostics, in addition to a direct examination of the CAM for plaque and membrane thickening, CAM samples are often subjected to histological and fluorescent antibody analysis [42]. In the case of the AIV, the allantoic fluid is collected from inoculated embryonated chicken eggs and further tested for the presence of hemagglutinating antigens by the hemagglutination (HA) assay.

The HA assay is one of the most commonly used methods to screen hemagglutinating pathogens such as the AIV [47]. Amplification in embryonated chicken eggs is generally required before HA activity detection. The advantage of a HA assay is that it can, to some extent, quantify the virus based on the hemagglutinating units (HAU) (1 HAU = 5-6 log10 of viruses [48], in addition to being cost-effective and relatively easy to conduct. The hemagglutination inhibition (HI) assay has been used widely for further antigenic characterization and subtype identification [26]. This assay relies on the antigen-antibody (ag-ab) reaction against the HA protein and is used to detect and quantify antibodies against AIV in serum. The serum is serially diluted two-fold on a 96-well microtiter plate, with the addition of a constant 4HAU standard live AIV virus and red blood cells to visualize the inhibition of HA activity. If antibodies exist within the collected serum, they will form the ag-ab complex, preventing hemagglutination of the AIV. After a short incubation period, the highest dilution of the serially diluted serum that inhibits hemagglutination is determined; it is called the HI titer. Antibodies within a collected serum can be confirmed and quantified according to the HI titer. The HI assay is inexpensive and easy to conduct, and the results are obtained promptly, but it generally requires a broad panel of standardized live AIV viruses. Hence, HI assays are slowly being replaced because of the development of lower-cost gene sequencing technology. In addition, antibodies against IB can also be quantified by these assays depending on the strain. Nevertheless, an extra step in adding NA is needed to cleave the S protein to expose the hemagglutinating domains [49].

Despite the widespread use of VI, this technique has several drawbacks, highlighting the need for further research and development. VI assays depend strongly on optimal sample collection and handling, and some viruses are unstable outside their host environments. VI can also be labor-intensive and time-consuming, with prolonged turnaround times, especially if the virus requires multiple in vitro passages [27]. The availability of suitable propagation systems is also a concern because not all viruses grow in the same system. For example, primary isolation methods involving the direct introduction of virus-containing materials into conventional monolayer cell cultures are ineffective for the IBV [40]. In contrast, the NDV [50], AMPV [39], and ILTV [42,43] can readily adapt to grow in various cell culture systems. Although VI assays have strengths over molecular techniques, such as the ability to identify new viral agents because of their broader specificity, they are slow and labor-intensive. Consequently, alternative assays are continually being developed in viral diagnostics.

Several detection methods focus on the antibody response to the target virus, including virus neutralization and HI. These tests operate on the same fundamental principle: forming an ag-ab complex. Typically, a known virus or viral protein is used alongside samples containing antibodies, usually serum collected from birds. HI tests are used widely to differentiate and serotype the IBV [40] and other paramyxoviruses [38,51,52]. For the IBV, HI tests require pre-treatment with NA to remove sialic acid residues from the spike protein of the virus envelope, which is essential for inducing HA activity [53]. Although HI tests are straightforward and economical, they require a broad panel of antigens for each serotype, usually live viruses. This requirement can decrease specificity because whole virus preparations contain conserved viral proteins that may cross-react with the antibodies of different serotypes.

Enzyme-linked immunosorbent assay (ELISA) has become popular because of its simplicity and low cost. Conventional ELISA consists of an array of wells coated with either the target virus as the antigen or a specific viral protein representing the virus of interest. If the antibodies in the serum sample bind to the antigens in the well, a color change can be observed when a secondary antibody labeled with an enzyme and a substrate is added. These changes are analyzed using a spectrophotometer and displayed as the optical density. ELISA has been developed and used to detect the IBV [54], NDV [55], AMPV [56,57], and ILTV [58] in the poultry industry, with commercial kits widely available. Other molecular-based methods, such as polymerase chain reaction (PCR) and quantitative real-time PCR (qPCR), have significant advantages because they do not require a live virus but target a viral genome segment, enabling the sensitive detection of viral agents. Coupled with refined and standardized nucleic acid extraction and purification techniques and supported by viral sequence data from various databases, PCR has become a ubiquitous tool in virus diagnostics regardless of the virus type. The discovery of reverse transcriptase (RT), an enzyme capable of generating complementary DNA (cDNA) from an RNA template used by retroviruses, has allowed PCR reactions to include a reverse transcription step. During this step, cDNA synthesized from the viral RNA of the target virus becomes a template for exponential amplification in the PCR process, facilitating the identification of RNA viruses [59]. Quantitative real-time PCR represents a more advanced diagnostic methodology wherein amplification of the DNA target template is monitored during the PCR process rather than post-amplification, as in conventional end-point PCR. New fluorescent DNA labeling techniques make this advance possible, enabling the real-time monitoring and detection of amplified DNA copies. RT-PCR is the current gold standard in viral diagnostics. This technique accurately quantifies gene copies without requiring post-PCR processing steps, such as adding ethidium bromide and densitometric analysis of PCR bands using ultraviolet rays [28]. This method offers significant advantages in preventing contamination and generating high-throughput results [29].

Quantitative real-time PCR has established a new standard in viral diagnostics because of its ability to detect and quantify precise amounts of gene copies across a wide range of samples. The speed, practicality, high sensitivity, and reduced risk of carry-over contamination with this procedure make it superior to conventional PCR [60]. This method is particularly advantageous for prompt decision-making in clinical settings because it allows real-time monitoring of gene copies or amplicon accumulation using fluorescently labeled primers, probes, or amplicons [61]. The development of amplifiable hybridization probes for gene quantification has contributed significantly to the foundation and advancement of this technique [62].

Hydrolysis (5’ nuclease probes) is commonly used in qPCR assays to monitor amplicon accumulation, paired with a primer set for amplifying the target DNA sequence. The first detection of amplicons was achieved by tracking the 5’ → 3’ endonuclease activity of Taq DNA polymerase on the oligoprobe-bound target DNA templates [58]. Fluorophore-labeled probes, commonly referred to by their proprietary name, TaqMan probes, were introduced based on this concept [63]. A typical TaqMan probe consists of a short oligonucleotide with a fluorescent dye reporter at the 5’ end and a quenching dye, or quencher, at the 3’ end (Fig. 2). The production and detection of a fluorescence signal bound to target DNA occur in two main sequential events: first, during PCR, the probe binds to the target-complementary strand (cDNA); second, the Taq polymerase cleaves the 5’ end of the TaqMan probe via 5’ → 3’ endonuclease activity, releasing the fluorescent reporter from the quencher [64]. The real-time PCR instrument then monitors the unquenched emissions of the reporter dye, resulting in fluorescence signals. The cleaved fluorescent dye accumulates after each PCR cycle, enabling detection at any point during thermo-cycling, and allowing the real-time representation of amplification. The signal emission increases proportionally to the quantities of PCR product synthesized during amplification. The optimal design of TaqMan probes requires specific criteria. The probe length should be limited to 20 to 40 nucleotide base pairs with a 40% to 60% guanine-cytosine content. They should avoid repeated sequence motifs and have a melting temperature (Tm) at least 5°C higher than the primers. This ensures that the probes can readily bind to the target template before the primer extension step [65]. The thermo-cycling conditions typically range between 60°C and 95°C because Taq polymerase and TaqMan probes require a minimum of 60°C for efficient 5’ → 3’ endonuclease activity.

Data processing in qPCR involves handling a large amount of raw numerical data collected during the amplification process, generating a standard curve, and evaluating the amplification efficiency. A guideline known as the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) was developed to enhance the uniformity, reliability, and transparency of reporting real-time PCR data [66]. According to the MIQE, the key assay performance characteristics must be determined, with the PCR efficiency being one of the most critical. The PCR efficiency, defined as the rate of increase of the target DNA template per amplification cycle, is calculated using standard curves [67], which provide a simple and intuitive measure of the mean PCR efficiency. These curves are used to evaluate the overall performance of real-time PCR assays, including assessing the dynamic range and detection limit [68]. The PCR efficiency is typically determined from the slope of the log-linear portion of the standard curves, calculated using the formula E = [10(−1 / slope)] - 1 [66]. A standard curve is generated by preparing a series of samples with controlled relative quantities of the target template, usually through serial 10-fold dilutions, and analyzing each sample in triplicate [69]. These standard samples are then analyzed by real-time PCR, and the quantification cycle (CT) values are measured.

A plot of the CT values as a function of the logarithm of the target template concentrations generates a standard curve, which is expected to be linear with a negative slope (Fig. 3). Without interfering or inhibitory factors in the sample matrix, an adequately designed real-time PCR assay should exhibit efficiencies ranging between 80% and 115% [69,70]. For 10-fold serial dilutions of samples, the slope of the standard curve should be −3.33 when the PCR efficiency is 100%, suggesting a perfect doubling of the target template with each amplification cycle. Therefore, assays should aim for standard curves with slopes between −3.0 (E = 115%) and −3.9 (E = 80%), though some studies accept slopes up to −2.9 (E = 122%) [71]. The slope of the standard curve is correlated directly with the CT values, as the interval (ΔCT) between 10-fold diluted target templates at 100% efficiency would be 3.33. Nevertheless, an efficiency exceeding 115% may indicate PCR inhibition in the sample matrix.

Excessive amounts of DNA/RNA or carry-over materials like sodium dodecyl sulfate or phenol in the sample matrix can function as PCR inhibitors, flattening the efficiency plot and reducing the slope, paired with a PCR efficiency exceeding 115%. Inhibitors are also diluted with other materials, so the least diluted samples are typically the most affected. This can be explained further through the ΔCT values. If inhibition occurs in the most concentrated (least diluted) sample, a higher-than-predicted CT value appears as the starting point of the standard curve. As inhibition decreases with higher dilution, assuming 100% efficiency, ΔCT will revert to 3.33 at some point. On the other hand, the average ΔCT between concentrated samples will be smaller than 3.33 because the CT starting point is higher than expected, flattening the standard curve slope and translating to unrealistically higher efficiency. A practical solution is to omit such concentrated samples when calculating the amplification efficiency.

In addition to conventional DNA-based diagnostic methodologies, new techniques are being introduced, such as the widely used next generation sequencing (NGS) and the later developed biosensing strategies focusing on antibodies, proteins, aptamers, and portable lateral flow tests optimized for on-site diagnosis. Viral diagnostics also evolve by incorporating the deep learning imaging of on-site samples [72-74]. These novel methodologies offer rapid and efficient detection of infectious diseases and potentially enable real-time application. The development of point-of-care testing (POCT), which can be applied directly on-site, is becoming prominent in avian diagnostics [74,75]. Laboratory diagnostics of animal viruses have advanced rapidly in recent decades, and potential techniques that do not require complex instruments, laboratory settings, and a highly skilled workforce are gaining focus. POCT allows test diagnostics outside the laboratory, particularly at the farm site, leading to faster decision-making. Most early animal POCT techniques originate from human diagnostics, which are antigen-based assays.

The widely used NGS provides powerful tools for characterizing various poultry pathogens and incredibly complex infections faster than conventional methodologies [76,77]. In addition, NGS can simultaneously characterize multiple viruses in the same sample. The NGS settings require viral sequences in each sample to be fragmented and converted into sequencing libraries. The libraries are constructed by adding adapter sequences on the end of each fragmented DNA, leading to clonal amplification for obtaining genetic data. Although NGS has a tremendous throughput of sequencing up to billions of viral sequences in a single run, paired with high sensitivity and accuracy, it requires expensive instrumentation and skill to analyze complex data. Biosensors recognize biomarkers via specific receptors, such as monoclonal antibodies or DNA glycans, enabling the identification of viral characteristics through paired biochemical reactions [30,78-80]. These reactions are translated into measurable signals, allowing qualitative and quantitative virus identification. The majority of portable, antibody-based methods for virus detection use lateral flow tests [31,81,82]. These tests detect specific viral biomarkers in complex media by solubilizing samples with detergents and depositing them on a sample conjugate pad, initiating lateral flow of the sample components. Viral biomarkers are then detected using the antibodies packed in gold nanoparticles preabsorbed in the sample pad. Two test lines (positive, control) with pre-immobilized antibodies, each explicitly recognizing the target viruses, bind to different epitopes of the virus. This binding results in the accumulation of immune-gold nanoparticles that carry the respective viruses. Positive lines appear when viruses bind to these immune-gold nanoparticles, indicating positive results. On the other hand, only the control line appears if no viruses are present, indicating negative results. The simplicity and rapid detection capability of this test have led to widespread use. In particular, it does not require sample pre-processing steps, which are often necessary in conventional molecular diagnostics. Artificial intelligence (AI) is also being integrated into virus detection, with deep-learning techniques showing promise in early diagnosis [72]. AI deep-learning techniques can analyze fecal images to distinguish healthy from unhealthy poultry, potentially enabling early disease detection (Fig. 4). Although in the testing phase, these techniques have satisfactory test accuracies and can be implemented in smart poultry farm concepts [83].

Nucleic acid amplification tests are becoming more popular regarding portable PCR tests. Among them, loop-mediated isothermal amplification (LAMP) is attracting attention for avian virus detection [84]. LAMP assays use a particular type of polymerase that does not require a denaturation step during the RNA/DNA amplification process, in which reactions can proceed at a constant temperature in a single-step reaction. Simplifying the process can lead to faster results and simple instruments that make on-site farm testing possible.

Timely and accurately detecting poultry respiratory viruses is crucial because of their rapid spread and significant economic impacts on the poultry sector. The morbidity and mortality of affected flocks rely on various factors, but co-infection with other respiratory diseases can exacerbate the clinical signs and prognosis. For example, the AIV, ILTV, and IBV enhance replication and pathogenicity when predisposed to other respiratory avian pathogens [85]. In particular, there are reports that some low-pathogenic AIV strains can become virulent and cause high mortality when infected with the IBV and ILTV [86]. This increased pathogenicity in low-pathogenic AIV-infected birds was attributed to the significant decrease in the humoral and cellular immune responses against the IBV and ILTV. Co-infection typically leads to immunosuppression, affecting the performance of birds, decreased weight gain, and reduced egg production. This makes it essential to detect and differentiate these viruses in infected birds promptly so countermeasures against these viral diseases can be implemented. Because the viruses above share similarities in the early manifestation of the disease, identifying the causative agent and differentiating between possible candidates is crucial.

Over the decades, diagnostic methodologies have evolved from traditional wet-laboratory techniques to popular molecular detection methods. With the advances in AI, diagnostic data interpretation using deep learning predictions will become popular alongside automated real-time surveillance systems in smart poultry farms. Nevertheless, efforts to efficiently detect respiratory diseases will continue, and it is hoped that more advanced methodologies will be introduced, complementing contemporary techniques.