Silent shedding of zoonotic enteric protozoa in domestic pigs for human consumption in selected farms in the Iberian Peninsula

Abstract

Enteric protozoa such as Balantioides coli, Giardia duodenalis, and Cryptosporidium scrofarum are of veterinary importance and may pose occupational exposure risks, yet data on their occurrence in pigs (Sus scrofa domesticus) raised for human consumption in the Iberian Peninsula are limited. A total of 400 stool samples were collected at a slaughterhouse in northern Portugal, originating from farms in Portugal (n = 200) and Spain (n = 200), and screened by PCR and qPCR. B. coli was highly prevalent in both countries (Portugal: 93%, Spain: 89.5%), with variant B predominating and variant A detected only in Portugal. C. scrofarum was detected exclusively in Portuguese pigs (4.5%), whereas G. duodenalis occurred more frequently in Portugal (8.5%) than in Spain (0.5%). These findings indicate that enteric protozoa of veterinary and potential occupational relevance are widespread in Iberian pig production, highlighting the importance of biosecurity measures for workers in contact with pigs and slaughterhouse environments.

Introduction

The Iberian Peninsula is Europe’s leading region for pork production, with Spain and Portugal together generating more than 5.3 million tons of carcass weight in 2024 according to Eurostat (European Commission Directorate-General for Agriculture and Rural Development 2025). Such intensive production has important implications for food safety and public health because pigs (Sus scrofa domesticus) can carry several gastrointestinal protozoa with recognized zoonotic potential (Stensvold et al. 2021). Among these, Balantioides coli, Giardia duodenalis and Cryptosporidium spp. are notable for their ability to infect both animals and humans and for their role in gastrointestinal disease across host species (Cacciò et al. 2018; Gerace et al. 2019; Ponce-Gordo and García-Rodríguez 2021). These organisms share fecal-oral transmission pathways and can persist in farm environments, a circumstance which facilitates their circulation within swine populations.

Balantioides coli, formerly known as Balantidium coli and Neobalantidium coli, is the only ciliate capable of infecting humans (Ponce-Gordo and García-Rodríguez 2021; Allievi et al. 2025). It can also infect other animals, including pigs, which are considered the main reservoir, as well as various mammals such as non-human primates, and has a documented potential for zoonotic transmission (da Silva et al. 2021; Ibañez-Escribano et al. 2025). Balantioides coli is divided into two main variants, A and B, with additional proposed subvariants A0, A1, A2, B0 and B1 (Ponce-Gordo et al. 2011). Among these, variants A0 and A2 have been detected in humans, although overall prevalence is generally low, ranging from to be 0.02 to 1% and even 30% in endemic areas especially in people in close contact with pigs and with a compromised immune system (Schuster and Ramirez-Avila 2008; Allievi et al. 2025). Human infections can be asymptomatic or symptomatic, with symptomatic cases presenting as chronic, with intermittent diarrhea, or as acute dysenteric forms, which can be life-threatening (Ponce-Gordo et al. 2011). In pigs, this protozoan achieves prevalence values of up to 100%, with infection being typically subclinical, but it can act as an opportunistic trigger of mucosal ulceration, dysentery and growth retardation, leading to substantial economic losses in intensive pig production systems (Ahmed et al. 2020; Giarratana et al. 2021; Szczotka-Bochniarz et al. 2021; Allievi et al. 2025).

Cryptosporidium spp. can infect the intestinal tract of both humans and animals (Alarcón et al. 2023). In humans they are a leading cause of diarrhea, morbidity and mortality in children younger than 5 years. Cryptosporidium hominis and Cryptosporidium parvum are the most important species infecting humans worldwide (Ugboko et al. 2020; Sone et al. 2025). Cryptosporidium spp. can cause infection even at low parasitic load and are notably resistant to disinfectants like chlorine (Kosek et al. 2001). Infections are often mild or asymptomatic or subclinical, which contributes to them being frequently undetected. Disease severity depends on the host’s age, nutritional and immune condition, and possibly the specific species or subtype of the parasite (Kosek et al. 2001; Cama et al. 2008; Khalil et al. 2018). At least 44 species and more than 120 genotypes within the genus Cryptosporidium have been identified, with 13 species of the genus reported to infect pigs. Among them, Cryptosporidium scrofarum and Cryptosporidium suis infections account for more than 90% of cryptosporidiosis in pigs (Feng et al. 2018). This parasitic infection in pigs is often subclinical, yet documented human infections with C. scrofarum and C. suis indicate that both species may have zoonotic potential (Kvác et al. 2009; Moore et al. 2016; Sannella et al. 2019). Their ability to cause disease in humans, however, is still poorly defined, which means they continue to represent a possible public health concern (Chen et al. 2023).

Giardia duodenalis (synonym with Giardia intestinalis and Giardia lamblia) is a parasite capable of infecting the upper intestinal tract of humans and a wide range of other mammals worldwide (Dixon 2021). It represents significant public health concern, contributing to chronic malnutrition and impaired growth, especially in affected children (Asghari et al. 2023). Molecular analyses using multilocus genotyping of the genes encoding triose phosphate isomerase (tpi), β-giardin (bg), and glutamate dehydrogenase (gdh) indicate that G. duodenalis is classified into eight assemblages, labeled A through H (Chang et al. 2023). Assemblages A and B are the main ones associated with human infections, while the remaining groups, C through H, are typically linked to various animal hosts and are only rarely detected in humans (Li et al. 2025). In pigs, infection with G. duodenalis may result in self-limited illness with weight loss and malabsorption, causing a decline in production (Zou et al. 2021). Six of the eight known assemblages have been detected, with assemblage E being the most prevalent, followed by A and B, and with C, D and F appearing only sporadically (Zou et al. 2021).

Pigs shed infective stages of protozoa in feces, but infections in pigs are nearly always subclinical, potentially contaminating housing facilities, manure, and slaughterhouse environments, creating opportunities for exposure among farm and slaughterhouse workers. Their ability to circulate silently within herds, combined with their documented or potential zoonotic relevance, reinforces the importance to protect immunocompromised humans from exposure.

In Portugal, information on the occurrence and molecular diversity of G. duodenalis, Cryptosporidium spp., and B. coli in domestic pigs is inexistent and only reported in wild boar (Figueiredo et al. 2023; Santos-Silva et al. 2023) and in Spain, studies have been mostly restricted to the northern and southern regions (Rivero-Juarez et al. 2020; Dashti et al. 2022; Garza-Moreno et al. 2025). Given the limited number of investigations assessing the molecular diversity of these enteric parasites in Portugal and northwestern Spain, this study aims to evaluate the distribution, genetic diversity, and zoonotic potential of B. coli, Cryptosporidium spp. and G. duodenalis, in domestic pigs intended from human consumption.

Materials and methods

Sampling

Sampling was carried out in a commercial slaughterhouse in the northern region of Portugal, in the city of Porto. Stool samples were obtained directly from the small intestine of non-diarrheic fattening pigs over a 2-month period, from December 2021 to January 2022, a total of 400 samples were collected, 200 from Portugal and 200 from Spain, originating from three Portuguese farms and two Spanish farms. Stool samples could not be allocated to specific farms and the proportions of stool samples from individual farms within the sample size are unknown. All sampling material consisted of byproducts that would otherwise have been discarded, so no animals were slaughtered specifically for this study. Samples were kept at 4 °C and processed within 12 h, after which they were stored at −80 °C. This sampling set had also been used in previous work (Santos-Silva et al. 2025; Gomes-Gonçalves et al. 2025).

DNA extraction and purification

Genomic DNA extraction and purification were carried out automatically on the QIAcube^® platform (Qiagen) using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany), following the manufacturer’s instructions. Before automated extraction, stool samples were prepared by diluting in PBS to obtain a 10% suspension. The suspensions were vortexed thoroughly to ensure complete homogenization, then centrifuged at 8,000×g for 5 min (Eppendorf, Hamburg, Germany) to pellet debris and concentrate particulate material. The resulting ∼140 µl supernatant was transferred to the QIAcube workstation. The automated protocol then performed cell lysis, binding to silica columns, washing, and final elution of purified DNA. genomic DNA, minimizing manual handling and reducing the risk of cross-contamination.

Molecular detection of zoonotic protozoa

The molecular detection of zoonotic protozoa was performed using conventional, nested, or semi-nested PCR depending on the target organism.

Balantioides coli

Detection of B. coli was carried out by conventional PCR targeting the 3′ end of the SSU rRNA gene (~ 117 bp), the ITS region, and the 5′ end of the LSU rRNA gene (~ 28 bp). Amplification used the primer pair B5D (5′-GAGCTTTTTAACTGCAACAACG-3′) and B5RC (5′-ATCTGGTTGATCCTGCCAGT-3′) following the protocol described by (Ponce-Gordo et al. 2011). Thermal cycling consisted of an initial denaturation at 95 °C for 5 min followed by 40 cycles of 94 °C for 2 s, 60 °C for 5 s, and 72 °C for 5 s, with a final extension at 72 °C for 10 min. Each 25 µl reaction contained 12.5 µl Speedy Supreme NZYTaq 2× Green Master Mix, 0.4 µM of each primer, and 2 µl DNA template.

Cryptosporidium spp.

Detection of Cryptosporidium spp. was performed by nested PCR targeting the SSU rRNA region using the outer primers CR-P1 (5′-CAGGGAGGTAGTGACAAGAA-3′) and CR-P2 (5′-TCAGCCTTGCGACCATACTC-3′) and the inner primers CR-P3 (5′-ATTGGAGGGCAAGTCTGGTG-3′) and CPB-DIAGR (5′-TAAGGTGCTGAAGGAGTAAGG-3′) (Tiangtip and Jongwutiwes 2002). Cycling conditions for both primary and secondary reactions consisted of an initial denaturation at 95 °C for 5 min followed by 40 cycles of 94 °C for 2 s, 50 °C annealing for 5 s, and 72 °C for 5 s, with a final extension at 72 °C for 10 min. The primary PCR reaction was carried out in a 15 µl reaction containing 7.5 µl Speedy Supreme NZYTaq 2× Green Master Mix, 0.4 µM of each primer, and 2 µl DNA template, and the secondary PCR was carried out in a 25 µl reaction containing 12.5 µl Speedy Supreme NZYTaq 2× Green Master Mix, 0.4 µM of each primer, and 2 µl DNA template.

Giardia duodenalis

Detection of G. duodenalis was performed by qPCR targeting a 62-bp region of the SSU-rRNA gene employing the primer set Gd-80F (5’-GACGGCTCAGGACAACGGTT-3’) and Gd127-R (5’- TTGCCAGCGGTGTCCG-3’) and Probe (5’-FAM–CCCGCGGCGGTCCCTGCTAG–BHQ1-3’) (Verweij et al. 2003). The qPCR reaction mix employed a total volume of 20 µl, 10 µl Xpert Fast Probe 2× Mastermix (uni) (Grisp^®, Porto, Portugal), 0.4 µM of each primer, 0.2 µM of Probe and finally 4 µl of DNA template. Thermal conditions scored an initial denaturation at 95 °C for 3 min, followed by 45 cycles at 95 °C for 5 s and annealing at 60 °C for 20 s.

Following screening by qPCR sequence-based multilocus genotyping (MLST) strategy was performed targeting the genes encoding for the triose phosphate isomerase (tpi), β-giardin (bg) and glutamate dehydrogenase (gdh) proteins. Separate nested PCRs were performed for tpi and bg. For tpi, the outer primers were AL3543 (5′-AAATIATGCCTGCTCGTCG-3′) and AL3546 (5′-CAAACCTTITCCGCAAACC-3′), and the inner primers were AL3544 (5′-CCCTTCATCGGIGGTAACTT-3′) and AL3545 (5′-GTGGCCACCACICCCGTGCC-3′) (Lalle et al. 2005). For bg, the outer primers were G7_F (5′-AAGCCCGACGACCTCACCCGCAGTGC-3′) and G759_R (5′-GAGGCCGCCCTGGATCTTCGAGACGAC-3′), and the inner primers were G99_F (5′-GAACGAACGAGATCGAGGTCCG-3′) and G609_R (5′-CTCGACGAGCTTCGTGTT-3′) (Sulaiman et al. 2003). Detection of gdh was performed using a semi-nested PCR with GDHeF (5′-TCAACGTYAAYCGYGGYTTCCGT-3′) and GDHiR (5′-GTTRTCCTTGCACATCTCC-3′) for the primary reaction, and GDHiF (5′-CAGTACACCTCYGCTCTCGG-3′) with GDHiR for the secondary reaction (Read et al. 2004).

Cycling conditions consisted of an initial denaturation at 95 °C for 5 min followed by 40 cycles of 94 °C for 2 s, locus-specific annealing for 5 s, and 72 °C for 5 s, with a final extension at 72 °C for 10 min. For tpi and gdh, both primary and secondary reactions used annealing temperatures of 50 °C and 55 °C respectively. For bg, the primary PCR used an annealing temperature of 65 °C and the secondary PCR used 55 °C. Reaction mixtures followed the same composition of nested PCR described above.

All PCR runs included both positive control and a no-template control (PCR mix with RNase-free water instead of DNA) to validate amplification and detect potential contamination. DNA fragments were separated by electrophoresis on a 1.5% agarose gel stained with Xpert Green Safe DNA gel dye (GRiSP^®, Porto, Portugal) for 30 min and visualized under UV light. Reactions were carried out on a T100 thermocycler (Bio-Rad, Hercules, CA, USA).

Sequencing and phylogenetic analysis

PCR products of the expected size were purified using the Exo/SAP Go PCR purification kit (Grisp^®, Porto, Portugal) and sequenced bidirectionally by Sanger sequencing. The resulting sequences were edited, aligned, and analyzed using BioEdit Sequence Alignment Editor (v7.2.5), and consensus sequences were compared against the NCBI GenBank database using BLASTN 2.17.0+ (NCBI 2025).

For phylogenetic analysis, reference sequences were retrieved from GenBank and aligned using MAFFT v7.490 with the L-INS-i algorithm to ensure high accuracy for diverse sequences (Katoh and Standley 2013). Maximum Likelihood phylogenetic trees were constructed in IQ-TREE (Nguyen et al. 2015) using 1000 bootstrap replicates and the best-fit substitution model for each. The resulting tree was further annotated and visually enhanced using the Interactive Tree of Life (iTOL) platform v7 (Letunic and Bork 2024).

Statistical analysis

The occurrence of zoonotic protozoa in domestic pigs was calculated as the proportion of positive samples relative to the total number analyzed, with a 95% confidence interval (95% CI). Data processing and preliminary analyses were conducted using Microsoft Excel^® for Microsoft 365 (Version 2312, Build 16.0.17126.20132, 64-bit).

Results

A total of 400 domestic pig samples from Portugal and Spain were screened for the three gastrointestinal protozoa. Balantioides coli was detected in 91.3% of the samples (385/400), G. duodenalis in 4.5% (18/400), and Cryptosporidium spp. in 2.3% (9/400). The proportion of samples positive for G. duodenalis and Cryptosporidium spp. was higher in Portuguese samples than in Spanish samples, whereas the prevalence of B. coli was similar between samples from both countries. Detailed occurrence data are provided in Table 1. Among the 400 samples, co-infection with B. coli and Giardia was detected in 18 samples, while co-infection with B. coli and Cryptosporidium was observed in nine sample. No triple co-infections (B. coli + Giardia + Cryptosporidium) were detected. All co-infections involved B. coli, which is consistent with its near-universal prevalence in the sampled population.

Table 1.

Occurrence of Balantioides coli, Cryptosporidium scrofarum and Giardia duodenalis in swine samples from Portugal and Spain. Values are shown as the number of positive samples, percentage with 95% confidence interval (CI), detected variants or assemblages, and corresponding p-values for country comparisons

	Total n = 400			Portugal Samples n = 200			Spain Samples n = 200
Protozoa	Pos. (n)	% (95% CI)	Variant/ Assemblage	Pos. (n)	% (95% CI)	Variant/ Assemblage	Pos. (n)	% (95% CI)	Variant/ Assemblage
B. coli	365	91.2	A and B	186	93	A and B	179	89.5	B
		(88.0-93.8)			(88.5–96.1)			(84.4–93.3)
C. scrofarum	9	2.25	-	9	4.5	-	0	-	-
		(1.0-4.2)			(2.1–8.4)
G. duodenalis	18	4.5	ND	17	8.5	ND	1	0.5	ND
		(2.7-7.0)			(5.0-13.2)			(0.01–2.8)

ND not defined, Pos positive

For variant analysis of B. coli, 20 positive samples from each country were selected, totaling 40 samples. Thirty sequences (n = 15 per country) were successfully assigned to a genotype, while 10 sequences displayed complex electropherogram patterns with multiple heterozygous sites, preventing genotype assignment. BLAST analysis identified variant B in both countries, present in 12 isolates from Portugal and 15 isolates from Spain. Variant A was detected only in Portugal (n = 3). Phylogenetic reconstruction confirmed that all assigned sequences clustered with previously described reference sequences for variants A and B, as shown in Fig. 1.

Fig. 1.

The phylogenetic tree of Balantioides coli variants was constructed using IQ-TREE software, based on the complex of SSU rRNA gene the ITS region, and the 5′ end of the LSU rRNA gene. The representative sequence generated in this study is indicated by bold formatting. Reference sequences retrieved from GenBank are shown without formatting, along with their respective accession numbers and geographic location. The tree was inferred using the TPM2u + F substitution model, with 1000 bootstrap replicates to ensure robustness. Bootstrap values of 50% or higher are displayed in the figure. Color codes represent the different variants and sub-variants. Additional annotations and visual enhancements were produced using the Interactive Tree of Life (iTOL) platform v7

Cryptosporidium was detected exclusively in Portuguese pigs, yielding nine DNA isolates. Sequence analysis identified all isolates as C. scrofarum, with sequence identities ranging from 99.07% to 100% relative to reference sequences in GenBank (PQ846466 and MT114474). Phylogenetic analysis of SSU-rRNA sequences grouped all isolates within the established C. scrofarum clade as illustrated in Fig. 2.

Fig. 2.

The phylogenetic tree of Cryptosporidium spp. was constructed using IQ-TREE software, based on SSU-rRNA. The representative sequence generated in this study is indicated by bold formatting and blue circles. Reference sequences retrieved from GenBank are shown without formatting, along with their respective accession numbers, host and geographic location. The tree was inferred using the TPM2u + F substitution model, with 1000 bootstrap replicates to ensure robustness. Bootstrap values of 50% or higher are displayed in the figure. Color codes represent the different variants and sub-variants. Additional annotations and visual enhancements were produced using the Interactive Tree of Life (iTOL) platform v7

A total of 18 isolates were positive for G. duodenalis by qPCR (Table 2), including 17 from Portugal and one from Spain. All isolates were subjected to MLST targeting the tpi, bg, and gdh genes. However, none yielded PCR products for any of the loci tested.

Table 2.

Positive qPCR cycle threshold (Ct) values for Giardia duodenalis detection in pig fecal samples from farms in Portugal and Spain

Country	Sample ID	Ct
Portugal	B23_46	32.52892
Portugal	B23_47	33.61882
Portugal	B23_50	29.1195
Portugal	B23_52	30.51407
Portugal	B23_54	38.95592
Portugal	B23_59	35.98275
Portugal	B23_123	32.6367
Portugal	B23_128	37.46122
Portugal	B23_129	42.37925
Portugal	B23_134	39.9143
Portugal	B23_135	32.01701
Portugal	B23_137	31.29288
Portugal	B23_140	42.29376
Portugal	B23_153	31.28556
Portugal	B23_154	29.36118
Portugal	B23_155	31.91853
Portugal	B23_156	44.38257
Spain	F23_106	36.0935

Discussion

This study demonstrates, to the best of our knowledge, the first detection of B. coli, G. duodenalis, and C. scrofarum in domestic pigs raised for human consumption in selected farms Portugal and northwestern Spain. Regarding Balantioides coli prevalences of 93% and 89.5%, were obtained from Portugal and Spain respectively. These occurrence values are among the highest reported in European pig populations, exceeding those documented in Southern Italy (46.9%) (Giarratana et al. 2021), Denmark (57%) (Stensvold et al. 2021), Switzerland (29.6%) (Schubnell et al. 2016), and Southern Spain (45.5–52.7%) (Rivero-Juarez et al. 2020; Dashti et al. 2022) and closely approaching levels reported in Northern Italy (92.4%) (Allievi et al. 2025) and Germany (80%) (Wylezich et al. 2019). As previously described, B. coli is primarily regarded as an opportunistic ciliate in pigs, often detected in apparently healthy animals, and is widely recognized as a commensal organism in swine (Schuster and Ramirez-Avila 2008). Molecular analysis revealed a predominance of genotype B in samples from both countries, with genotype A detected only in Portugal, aligning with previous reports that genotype B is the most frequent in pigs. The detection of variant A, which has documented zoonotic potential and has previously been reported in humans (Oberhuber et al. 1993; Sharma and Harding 2003; Schuster and Ramirez-Avila 2008), underscores the need for continued molecular surveillance, particularly in farms producing pigs for human consumption and in slaughterhouses, where occupational exposure may increase the risk of B. coli infection in people. Overall, these findings underscore the widespread distribution and genetic diversity of B. coli in Iberian swine populations and confirm its relevance as a protozoan of veterinary importance and occupational exposure risk.

Cryptosporidium scrofarum was detected exclusively in samples from Portuguese farms, with an occurrence of 4.5%. This is consistently reported as the dominant species of Cryptosporidium infecting older pigs across Europe, including Denmark (92% and 40.9%) (Petersen et al. 2015; Stensvold et al. 2021), Sweden (25%) (Pettersson et al. 2020), the Czech Republic (21.8%) (Němejc et al. 2013), Australia (22.1%) (Johnson et al. 2008), and Spain (Northeast 22.5%, Southern 5.5% and 8.2%) (Rivero-Juarez et al. 2020; Dashti et al. 2022; Garza-Moreno et al. 2025), a situation that aligns with previous observations, although its absence in Spanish samples in this study contrasts with earlier reports. Despite the infection of humans with C. scrofarum has been previously reported, the species is considered to be predominantly host-specific (Kváč et al. 2009). However the present findings underscore the need for targeted investigations to clarify local transmission dynamics and inform effective biosecurity and management strategies.

Giardia duodenalis was detected at a low frequency, with a higher occurrence in Portuguese pigs (8.5%) than in Spanish pigs (0.5%). The consistently high qPCR cycle threshold values and the lack of MLST suggest a low parasite burden, or potentially chronic infections (Rivero-Juarez et al. 2020). Similar challenges in amplification have been reported in previous studies, reflecting the difficulty of characterizing G. duodenalis in low-load infections. Although these limitations prevent robust conclusions regarding the genetic diversity of G. duodenalis assemblage in Iberian pigs, the findings indicate that swine can carry this protozoan (Asghari et al. 2023). However, in the absence of genotyping, the zoonotic potential cannot be assumed. Even low-intensity infections may result in the shedding of cysts into the environment, representing a potential source of contamination and a risk for indirect transmission to other animals or humans under certain conditions.

Some limitations of this study should be acknowledged. The present findings represent baseline data from northern Iberian pig production but cannot be extrapolated to Portuguese and Spanish production systems as a whole. The slaughterhouse sampling design avoided on-farm selection bias, but limits interpretation at the herd level. Sampling was restricted to a single slaughterhouse in northern Portugal and a limited number of farms in Spain, which may not fully represent the diversity of pig production systems across the Iberian Peninsula. Control of farm distribution across sampling dates was not possible due to logistical constraints and the functioning of the slaughterhouse, which made assessing temporal clustering by farm or batch difficult and that potentially influenced prevalence estimates. The small number of farms and animals per country reduces the ability to generalize prevalence estimates or detect regional differences. Low parasite burden and the inability to perform MLST for G. duodenalis limited assessment of its genetic diversity and transmission dynamics. All sampled pigs were apparently healthy, a circumstance that potentially differs from infections of higher parasitic loads or clinical severity. The cross-sectional design captures only a single time point, and seasonal or batch-level variations in parasite shedding were not addressed. Moreover, the detection of these agents was not correlated with the presence of diarrheal manifestations in pigs, indicating that their occurrence does not necessarily reflect clinical disease.

Differences in parasite occurrence between Portuguese and Spanish samples might reflect farm-level heterogeneity rather than true country-level patterns. The present study included samples from five farms (three Portuguese and two Spanish), limiting the assessment of key factors such as husbandry, biosecurity, stocking density, and water sources, which strongly influence protozoan transmission. Due to the fact, that pigs could not be allocated to the farm of origin and that only a low number of farms was included in the study, it was impossible to determine prevalences for the pathogens. Nonetheless, our results indicate that enteric protozoa of veterinary and potential zoonotic interest are present in Iberian pig production. Moreover, the detection of these agents was not correlated with the presence of diarrhea, indicating that these protozoa are rather commensal organisms belonging to the physiological gut flora of pigs than pathogens.From a public health perspective, zoonotic potential of B. coli variant A and C. scrofarum should be interpreted in the context of plausible transmission routes, such as contaminating slaughterhouse environments and manure management systems.

To address these limitations and clarify the drivers of parasite transmission, future studies should include more farms, systematically characterize farm-level factors, increase sampling intensity, incorporate multi-seasonal collection, and apply statistical methods that account for farm-level clustering. Such approaches would support more accurate prevalence estimates and targeted surveillance strategies across Iberian pig farms, particularly given the seasonal variation in shedding patterns.

Conclusion

In conclusion, this study confirms the widespread presence of B. coli, G. duodenalis, and C. scrofarum in pigs from selected farms in Portugal and northwestern Spain. While these protozoa are usually of no veterinary relevance and the general zoonotic risk is neglectable, an exposure of immunocompromised humans should be avoided. However, the findings highlight potential occupational exposure, particularly for slaughterhouse and farm workers who may come into contact with feces or contaminated environments. Differences between countries suggest that local management practices on farms influence parasite circulation. Future research with larger, multi-seasonal datasets and detailed farm-level characterization is needed to clarify transmission dynamics and support targeted surveillance and biosecurity measures aimed at reducing occupational risk.

Authors’ contributions

S.G-G(Conceptualization, Methodology, Writing – original draft, Writing – review & editing), R.J.F. (Methodology, Investigation, Writing – review & editing) S.S-S(Resources, Methodology, Writing – review & editing), A.V. S. C.(Resources, Methodology, Writing – review & editing), G.M. (Resources, Methodology), L.C. (Supervision, Writing – review & editing) and J.R. M. (Funding acquisition, Methodology, Project administration, Resources, Writing – review & editing). All authors reviewed the manuscript.

Funding

Open access funding provided by FCT|FCCN (b-on). This work was supported by the Fundação para a Ciência e Tecnologia (FCT) through doctoral grants 2022.15408.BD, 2025.04664.BD, 2025.05151.BD and under project reference 2021.09461.BD (DOI: https://doi.org/10.54499/2021.09461.BD, accessed 2 October 2025).

Data availability

All data supporting the findings of this study are available within the paper.

Declarations

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare no competing interests.