Detection of Antibodies Against Toxoplasma gondii in Filter Paper-Dried Blood Dot Spots Compared with Serum in Pigs and Assessment of Variation Associated with Packed Cell Volume

Introduction

Toxoplasmosis is caused by Toxoplasma gondii, an obligate intracellular protozoan parasite. Domestic and wild felids (Felidae) are the definitive hosts of the parasite, being able to shed environmentally resistant oocysts. Different species of warm-blooded mammals, birds, as well as human (Dubey, 2010, 2021; Smith et al., 2021) comprise the wide range of intermediate hosts. T. gondii transmission occurs mainly through soil, water, or feed contaminated with oocysts as well as through consumption of undercooked or raw infected meat containing tissue cysts (Jones and Dubey, 2012).

Toxoplasmosis is a significant zoonotic disease with worldwide distribution in humans and animals with varying prevalence among the different species and countries (Flegr et al., 2014; Papatsiros et al., 2021). Although T. gondii infections in pigs are commonly subclinical, several cases of clinical disease after natural infection have been recorded worldwide most frequently in neonatal and weaned pigs while also reproductive failure may appear in affected sows (Dubey, 2009). Previous studies suggest that the global seroprevalence in the domestic swine population is 19%, slightly lower than the T. gondii seroprevalence reported in wild boar (23%) (Limon et al., 2017; Rostami et al., 2017; Foroutan et al., 2019).

Serological methods used for T. gondii include Enzyme-Linked Immunosorbent Assay, Indirect Immunofluorescence Antibody assay (IFA), Modified Agglutination Test (MAT) (Foroutan et al., 2019; Dubey et al., 2020). However, there is increasing evidence on the usefulness of dried blood samples on filter paper (FP), which have been commonly used in human medicine for 50 years (Mei et al., 2001).

FPs have been introduced the last 20 years in veterinary medicine as well, being used for the detection of antibodies in epidemiological studies in wild (Dubay et al., 2006; Yu et al., 2007; Curry et al., 2011, 2014; Jakubek et al., 2012; Aston et al., 2014; Elmore et al., 2014) and domestic animals (Beard and Brugh, 1977; Hopkins et al., 1998; Duscher et al., 2009; Bolais et al., 2017; Simon et al., 2021). FPs are a convenient and cost-efficient type of sample, which can be collected even from stressed or aggressive animals in the clinic or in the farm and from wild animals in the field.

This is of great importance for pigs, as in this species blood sampling presents unique difficulties; identification of major blood vessels is challenging due to a large amount of subcutaneous fat and muscle mass; they can be easily excited and difficult to handle, the handling and sampling procedures themselves comprise important stress factors for this species leading to poor meat quality, and the operator should be adequately experienced (Hu et al., 1993). However, the main disadvantage for this type of sample is the lack of validation with a reference method before being used for the evaluation of the serological status of a given species (Simon et al., 2021).

Taking into consideration that farm animals represent a source of infection for humans and other animals, the reduction of T. gondii infection in livestock is a necessity (Stelzer et al., 2019; Dubey et al., 2020). This is especially true for swine, as globally, pork meat comes first in demand and its consumption is projected to increase to 127 metric tons over the next 10 years and to account for 33% of the total increase in meat consumption (OECD and Food and Agriculture Organization of the United Nations, 2021). Apart from the economic significance of this species, the swine population is of public health concern for several zoonotic diseases. Importantly, the European Food Safety Authority recognized T. gondii as one of the public health hazards in swine that should be evaluated during meat inspection (EFSA, 2011; van Asseldonk et al., 2017).

The aim of this study was (1) to assess the agreement between antibody detection in serum and FP blood spots using the IFA and (2) to evaluate the potential impact of the packed cell volume (PCV) on antibody detection in FP blood spots, due to the influence of hematocrit on spreading parameters of FP sampling (Chao et al., 2014) by assessment of agreement in samples grouped by PCV.

Materials and Methods

Ethics approval

All procedures were done according to the ethical standards in the Helsinki Declaration of 1975, as revised in 2000, as well as the national law, and after receiving approval (nr: 129/29-1-2021) from our Institutional Animal Use Ethics Committee of the Faculty of Veterinary Medicine, University of Thessaly.

Animals

A pair of a serum and a FP sample was collected from 96 sows at various farms in Greece, with previously identified high seropositivity (Athanasiou et al., 2021) and/or risk factors associated with high seropositivity against T. gondii, such as small size, poor biosecurity measures, and absence of vaccination against Porcine Circovirus Type 2 (PCV2) (Papatsiros et al., 2016, 2021; Athanasiou et al., 2021).

Blood collection in tubes

Blood samples were collected through puncture of the vena jugular externa with an 18-gauge needle into one vacutainer (Venoject, Terumo Europe, Leuven, Belgium) with no anticoagulant for serum retrieval and one ethylenediamine tetraacetic acid (EDTA)-coated vacutainer (Venoject, Terumo Europe). Samples were transferred to the laboratory pending further analysis.

PCV determination

The PCV value was determined using the microhematocrit method (Bull et al., 2000). PCV was measured as the height of the red cell column in the tube after centrifugation.

FP blood collection

A drop of blood was collected by puncture of the marginal ear vein onto a Whatman No. 4 FP, allowed to dry, placed in a self-sealing polythene bag containing a few crystals of silica gel, and stored at −20°C pending analysis. A paper punch was used to obtain FP discs 6 mm in diameter from the dried blood samples.

Antibody detection by IFA

For the detection of antibodies against T. gondii, Indirect Immunofluorescence Antibody Assay Test Kits using commercially available slides coated with parasite tachyzoites (Fuller Laboratories, Fullerton, CA) and a goat polyclonal fluorescein isothiocyanate (FITC)-conjugated anti-pig IgG (Porcine IgG FITC conjugate; VMRD, Inc., Pullman, Washington) were used. After the addition of 1:64 (cut-off value) diluted sera in phosphate-buffered saline (PBS), the slides were incubated for 30 min at 37°C. Likewise, 15 μL of PBS solution and the discs were transferred to the antigen-coated slides incubated under moist conditions for 1 h at room temperature. The FP discs were removed with a jet of PBS and the slides washed twice 5 min each, with PBS solution as previously described (Sharudin et al., 2019).

The slides were then washed to remove unreacted serum or eluted blood proteins, and fluorescence-labeled anti-porcine IgG (conjugate) was added. After incubation for the same time and at the same conditions with those used in the first incubation for each type of sample, the slides were washed again to remove unreacted conjugate. A Nikon Eclipse E-400 fluorescence microscope was used for the observation (objective 40 × ).

Statistical analyses

A sample size of 96 was found adequate to observe at least 40 seropositive samples given probability of occurrence equals prevalence with 95% probability using free online software (http://sampsize.sourceforge.net/iface/).

For the statistical analyses of the data, MedCalc Statistical Software version 14.8.1 (MedCalc Software bvba, Ostend, Belgium; www.medcalc.org; 2014) was used, to calculate the sensitivity, the specificity, Positive likelihood ratio (PLR), Negative likelihood ratio (NLR), Positive Predictive Value (PPV), and Negative Predictive Value (NPV) using IFA as the reference method. PLR values >10 and NLR values <0.1 are indicative of good test performance (Gardner and Greiner, 2006).

Moreover, the agreement between the results of the tests performed in the two different biological samples was measured using the Cohen's Kappa (κ) value. The same analyses were performed in the two groups formed by PCV, namely one group of animals with PCV <32 (anemic animals) and another with PCV ≥32 (nonanemic animals) based on previously reported reference values for pigs (Jackson and Cockcroft, 2002). A value of 0 indicates poor agreement, while a value of 1 indicates an almost perfect agreement (Landis and Koch, 1977; Watson and Petrie, 2010; McHugh, 2012). A value of p ≤ 0.05 was considered significant in all comparisons.

Results

T. gondii-specific antibodies were detected by IFA in 44/96 serum samples (45.8%) and 40/96 FP samples (41.6%).

An almost perfect agreement, with Cohen's Kappa (κ) value of 0.915, was found between serum and FP samples for the detection of anti-T. gondii antibodies in swine (Table 1). When samples were divided in PCV-based categories (< or ≥32%), the agreement between serum and FP samples remained almost perfect, with κ value ranging from 0.88 to 0.91 (Table 1).

As shown in Table 2, detection of anti-T. gondii IgG antibodies in FP samples presented high sensitivity (87.1–92.8%) and excellent specificity (100%) when compared with serum. The recorded sensitivity and specificity values were high when all the samples were taken into account as well as when they were divided in PCV-based categories. The estimated PLR and NLR values further supported the high accuracy of detecting anti-T. gondii antibodies in FP samples. The PPVs were 100% when all the samples were taken into account, as well as when the samples were categorized based on the PCVs, while the NPVs ranged between 90.0% and 92.7%.

Discussion

In this study, we evaluated the agreement between serum and FP samples for the detection of anti-T. gondii antibodies in swine using IFA, the diagnostic accuracy of detecting antibodies in FP compared with serum and the potential impact of PCV values in the detection of antibodies in FP samples.

An almost perfect agreement was found between the two sample types that were used for the determination of seropositivity to T. gondii in swine (κ values ranged 0.88–0.91). Similarly, an almost perfect agreement was found in a previous study (mean κ values ranged 0.89–0.92) between feline FP and serum samples for the detection of T. gondii antibodies using MAT (Simon et al., 2021). The lower agreement (κ value 0.88) was found in the group of animals with PCV value ≥32%. However, this difference was not practically considered as significant between the PCV-based categorized samples and the samples in total (without taking into account the PCV value), suggesting that PCV does not really affect the detection of the antibodies.

The storage of the FP samples in self-sealing polythene bag containing a few crystals of silica gel and in low temperature (−20°C) may have also positively affected the stability of the antibodies (Nogami et al., 1992; Sharudin et al., 2019). In fact, the long-term preservation of T. gondii antibodies' activity in feline FP samples that had been stored at 25°C with silica gel for at least 6 months, has been shown previously (Nogami et al., 1992).

The diagnostic accuracy of FP samples for the detection of IgG antibodies against T. gondii was assessed in terms of sensitivity and specificity values. FPs presented excellent specificity (100%) and high sensitivity (90.1%) compared with serum samples suggesting that this convenient sample type is reliable, and it could be used in field conditions. Our results are in agreement with a previous study conducted in feline serum and FP samples for the detection of anti-T. gondii antibodies using MAT, which showed high specificity (90.1–98.6%) and sensitivity (94.7–100%) values (Simon et al., 2021).

To evaluate the potential impact of PCV value on the detection of antibodies in FP samples, PCV-based sample categories were created using the reference interval (RI) of PCV value (RI: 32–55%) of the laboratory that performed the examination. The PCV values were determined using the reference method, the microhematocrit method, which is also easy to perform and cost-effective as it does not require any special equipment, except from a centrifuge. The higher sensitivity (92.8%) was recorded in the animals with PCV value <32%, comprising the animal group with anemia. The lower sensitivity (87.1%) was recorded in the nonanemic group of animals with PCV value ≥32%.

However, the sensitivity values in anemic and nonanemic animals were close suggesting that PCV value does not affect the antibody concentration in FP samples. Our results are in agreement with a previous study conducted in feline serum and FP samples for the detection of anti-T. gondii antibodies, which showed high specificity and sensitivity values (Simon et al., 2021).

Due to 100% specificity, the PLR value was undefined, which in all cases was indicative of good performance; anti-T. gondii antibodies are likely to be detected in FP samples from a seropositive animal. Regarding the animals with PCV values <32% and all the animals regardless of the PCV value status, the NLR values (<0.1) indicated a good performance as well suggesting that in seronegative samples, it is likely to detect the absence of anti-T. gondii antibodies in FP samples. In the group of animals with PCV value ≥32%, the NLR value (0.13) indicated a moderate power to identify seronegative animals when antibodies are not detected in FPs.

The PPV of 100% is the ideal value for a perfect test, and it indicates that the positive result of the detection of anti-T. gondii antibodies in FPs is true. The NPV indicates that there is possibility that a negative result in the detection of anti-T. gondii antibodies in FPs, may not be true. However, even though the seroprevalence of T. gondii is known in swine herds in Greece, the sample size is not representative of the general swine population. The PPV and NPV calculated are valid only for the seropositivity found in the present study.

The high seropositivity herein reported (45.8% for serum samples and 41.6% for FP samples) does not reflect the true seroprevalence of T. gondii in swine herds in Greece. The samples included in this study were purposely selected from swine farms that high T. gondii seropositivity (Athanasiou et al., 2021) and/or risk factors associated with high seropositivity, such as small size, poor biosecurity measures, and absence of vaccination against PCV2 (Papatsiros et al., 2016, 2021; Athanasiou et al., 2021) were previously identified. The abovementioned criteria were applied so as to increase the probability of detecting seropositive samples and make feasible the comparison between the two sample types. In fact, a low seroprevalence of 4.3% was previously reported in swine, in Greece (Papatsiros et al., 2016).

Although serum is the preferred sample to be used in serological studies for the detection of antibodies, other kind of samples, more convenient, cost-effective, and easy to obtain can be used to enable the collection of samples in large-scale epidemiological studies. FP presents many advantages and could be considered as a good alternative to blood collection. One of the foremost advantages is that the procedure is less stressful for the animals. This is particularly important for pigs as there are several technical difficulties during sample collection and handling, due to their excitability as well as the relative lack of superficial blood vessels (Hu et al., 1993).

Blood sample collection in pigs requires special equipment and adequate restraint from specially educated staff (Buzzard et al., 2012). The procedure can be stressful for the animals sampled with several potential adverse effects varying from the development of hematomas to syncope, shock, and death secondary to vasovagal response or laceration of anterior vena cava or jugular vein (Baldi et al., 1989; von Altrock and von Holleben, 1999). Besides, it is well known that stress negatively affects production and meat quality in pigs.

The reduction of stress induced by sampling and handling procedures should be prioritized in this species so as to ensure the welfare of pigs, a subject of growing concern. Among other important advantages, no special equipment and training are required, only a small amount of blood on strips or spots of an absorbent paper is needed, it does not require drawing blood with a needle and syringe, and there is no need for centrifugation (Martínez-Miró et al., 2016). Moreover, FP storage is easy, it only needs to get dry, and then they can be stored in different conditions and time periods, without significant impact on their quality (Behets et al., 1992).

On the other hand, because of the small amount of collected blood and the elution step that needs to be performed, the quantity of antibodies might be lower compared with serum (Portejoie et al., 2009). Thus, validation of FPs with a reference method before being used for the evaluation of the serological status of a given species is necessary.

FPs have been used in veterinary medicine for the detection of T. gondii antibodies in wild and domestic animals using serological techniques such as IFA and MAT (Etheredge et al., 2004; Jordan et al., 2005; Mercier et al., 2013; Aston et al., 2014; Elmore et al., 2014, 2016; Bolais et al., 2017; Brouat et al., 2018; Simon et al., 2018; Sharma et al., 2019). However, the majority of these studies lack validation of FPs to assess the serological status of infected animals. Exceptionally, there are two studies that compared the detection of T. gondii antibodies in feline FP and serum samples using MAT (Bolais et al., 2017; Simon et al., 2021). Those studies are in concordance with the results of this study that we used IFA, suggesting that this type of sample could be considered a reliable alternative to serum samples for the detection of anti-T. gondii antibodies.

Conclusions

The findings of this study support the reliability of FPs as an alternative sample type to serum, for the evaluation of the serological status of swine against T. gondii. Using IFA, the FPs showed an almost perfect agreement with serum samples for the detection of anti-T. gondii antibodies in swine. This method presents excellent specificity and high sensitivity regardless of the PCV values and high probability to detect seropositive and seronegative samples. The PCV value was not found to significantly affect the detection of the anti-T. gondii antibodies and the diagnostic accuracy of the FP samples in overall. FPs could be a good alternative sample type to be used in large-scale epidemiological studies for T. gondii in swine herds.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.