The study protocol included an interview guide and an informed consent form for dog owners. The Research Ethics Committee of the National University of Huancavelica approved this protocol (Resolution No. 0791-2020-CU-UNH dated December 14, 2020). Dog owners signed the informed consent form, agreeing to participate in the survey, and authorizing the collection of fecal samples from their dogs. Owners’ participation was completely voluntary; therefore, there was no coercion or undue incentives to participate, and they understood the objectives, procedures, and potential risks or benefits of the study. Participation was anonymous to maintain community confidentiality when identifying disease outbreaks.

The protocol complied with the institutional animal welfare guidelines required by Peruvian regulations regarding the non-invasive handling of farm animals. The Peruvian Law on Animal Protection and Welfare (Law No. 30407) established a duty to ensure animal welfare and to avoid unnecessary suffering during any procedure, including sample collection. International guidelines from the WOAH (Terrestrial Code) were also followed as a basis for animal management and welfare, including pain and stress management and proper dog restraint to minimize discomfort. Furthermore, a refinement was implemented, as trained personnel performed rectal swab testing using techniques that minimized pain and stress. Finally, the Animal Research: Reporting of In Vivo Experiments 2.0 guidelines were considered (staff training and experience in sample collection, appropriate restraint and handling methods used with dogs, and appropriate procedures for fecal sample collection).

The study was conducted between April and December 2021. The district of Ascension has an area of 432.2 km² and is located in the province and region of Huancavelica, at an altitude of 3,650 m, at the coordinates: 12°47′02″S and 74°58′42″W [19] (Figure 1). The climate of Ascension is cold, with variations from semi-arid to humid depending on altitude, maintaining an average annual temperature of between 9°C and 11°C. It is characterized by a marked daily temperature range, with days of intense solar radiation and nights with sharp temperature drops. The annual cycle comprises two periods: the rainy season (December–April), with annual precipitation exceeding 800 mm [20], and the dry season (May–November), which is characterized by the absence of rain, clear skies, and the presence of intense frosts at night [21].

Vegetation is closely related to altitude and seasonal rainfall. During the rainy season, abundant herbaceous vegetation predominates, providing livestock with fodder. Bofedales (high Andean wetlands) serve as moisture and quality pasture reserves during drought in higher-altitude areas, along with xerophytic species, low shrubs, and grasses that are resistant to water and heat stress [22].

Recognized as the “Livestock Capital of the Andes,” the economy of Ascension is based on livestock farming. The predominant production system is semi-extensive, in which cattle graze freely on communal or private land with limited supplementation. Production efficiency is classified as low to medium, conditioned by traditional management and the scarcity of nutritious pastures during the dry season [23]. The main species identified are cattle (mainly Brown Swiss and Creole crosses, with a focus on dairy or dual-purpose production), sheep (for wool and meat production), and South American camelids (alpacas and llamas, strategically located in higher-altitude areas) [24].

The study design was non-experimental and analytical because interdependence was measured without manipulating the variables. Furthermore, this was a community-based cross-sectional study conducted between April and December 2021.

A simple random sample was taken from this population using the following formula: n = NZ²PQ / E² (N−1) + Z²PQ. The sample size was calculated based on a normal distribution, which required a sample of 270 households (N), with an estimated proportion of 50% (P), an accepted error of 5% (E), and a confidence level of 95% (Z). A sample of 159 households was estimated, considering that in the rural sector of the Peruvian highlands, there would be 0.6 dogs per household; thus, 95.4 dogs should have been sampled. However, fresh fecal matter from 453 dogs was collected in sterile containers [25], which were inactivated at −70°C for 72 h and stored at −20°C until processing.

Feces samples were collected from three to four dogs per household. The maximum number of dogs sampled per household was four. Only dogs of all ages with owners who permanently resided in the Ascension district were sampled. We excluded dogs with clinical signs or symptoms (apparently ill), stray dogs, puppies younger than 3 months, or dogs that had been dewormed with praziquantel in the last 30 days were excluded.

The inhabitants of the Ascension District have one of the lowest average incomes in the country and a high incidence of poverty, with an average monthly income of $359.2 for dependent household activities and $75.1 for independent activities. The poverty level in this population fluctuates between 32.9% and 36.2%, with extreme poverty between 6% and 7.7%. The main economic sector by employed population is agriculture (73.7%), followed by transportation (8.9%) and other services (17.3%). In addition, the illiteracy rate in the population aged 15 and over is 17.7%, which is significantly higher than the national average, with a notable gap between urban areas (9.2%) and rural areas (21.5%), as well as between men (8.4%) and women (26.2%). Conversely, the educational level of the population aged 15 and over is as follows: secondary education, 37%; primary education, 27.7%; higher education (non-university and university), 19.0%; and no educational level, 16% [19].

An epidemiological survey was conducted through an interview with heads of household to evaluate the factors associated with echinococcosis in dogs. The survey recorded personal data of the owners of the dogs (age, sex, level of education, place of residence, and characteristics of their home) and characteristics of the dogs (age, sex, coexistence with other animals, and coinfection with gastroenteric parasites), as well as risk practices related to animal breeding and visceral disposal. This study was conducted in coordination with the Regional Health Directorate and the District Municipality of Ascension. The questionnaire underwent initial pilot testing with 10% of the calculated sample in a community with characteristics similar to Ascension. Expert judgment was used to validate the content to ensure that the language was culturally appropriate for the area. The mode of administration was face-to-face (in-person) interviews conducted at the home of the participants. Each interview lasted 15–20 min. The responses were recorded on physical forms (paper) and then digitized into an electronic database (Excel, Office 365, Microsoft Corp., Washington, USA).

Sampling was conducted during the dry season, between April and November 2021, when camelid management is more active and often coincides with festivities or slaughter periods, during which dogs have greater access to raw offal. The feces were collected by a team of veterinarians and biologists who had been trained not only to administer the survey but also to handle dog fecal samples. The samples were obtained directly from the dogs’ rectum and from the superficial part of fresh feces found in the house, which were preserved in a disposable container and then labeled and transported, maintaining the cold chain, to the National Reference Laboratory of Parasitic Zoonosis of the National Center for Public Health (CNSP) of the National Institute of Health (INS), in the city of Lima, Peru. Fecal samples were collected from dogs of all ages in their respective homes. Dogs that showed signs or symptoms (apparently ill) were excluded.

To avoid direct contact between the handler and the feces, disposable, waterproof, and resistant gloves (nitrile or latex), a face mask, a disposable apron or gown, and safety goggles were required. It was also necessary to wash hands with soap and water, followed by an alcohol-based hand sanitizer, before and after removing personal protective equipment.

Rectal swab testing was used to ensure the sample belonged to the dog and was as fresh as possible to prevent DNA degradation. However, sampling of fresh feces from the ground was performed only when rectal sampling was not possible due to the animal’s aggressive temperament or if the dog had just defecated and the rectum was empty. Thermal management of the samples was critical to preserving the integrity of the eggs and genetic material. The samples were immediately placed in thermos flasks containing cooling gels (4°C–8°C). The maximum time between home collection and arrival at the local freezing center did not exceed 6 h. The samples were initially stored at −20°C for 48–72 h while the transport batch was completed. However, the samples were transported to the laboratory in validated thermal boxes. Upon arrival at the laboratory, the samples were transferred to −70°C ultra-freezers for long-term storage prior to polymerase chain reaction (PCR) analysis. Each sample was placed in sterile 60 mL polypropylene screw-cap bottles. Each bottle was labeled with indelible ink and a transparent tape-protected sticker that included the dog’s unique code, housing code, exact date and time of collection, and district sector.

Given the high infectivity of E. granulosus eggs to humans, disposable nitrile gloves (changed between each dog), N95 masks (to prevent inhalation of dry fecal dust in the yards), waterproof aprons, and eye protection were used during rectal manipulation. After sampling, 70% alcohol and hypochlorite solutions were used to disinfect the support materials and the external surfaces of the jars.

The study design focused exclusively on dogs with owners, collecting samples in home yards and interviewing the “heads of household.” Therefore, stray dogs (street dogs or vagrants) were not sampled because their existence in the area is unknown and there is no municipal slaughterhouse in the district where dogs could have access to raw offal.

To avoid cross-contamination of samples, trained professionals (veterinarians and biologists) collected the samples directly from the dog’s rectum. Therefore, the use of disposable containers and mass transport requires very strict handling to avoid contact between samples.

To reduce the PCR inhibition rate in fecal samples, stool-specific extraction kits and sample dilution protocols were used.

For the coproparasitological diagnosis, the simple flotation technique was used, sifting the feces and taking 2–3 g to be suspended in a 0.85% saline solution, which was filtered in a 0.12 μm metal strainer; then, 12 mL of sample was added to the 15 mL conical tubes to be centrifuged at 1,800 rpm for 5 min. The supernatant was removed, and 0.85% saline solution was added, then the mixture was centrifuged again. This process was repeated until the supernatant was translucent. The samples were observed under a microscope, and the identification and quantification of eggs per gram of feces were performed based on the general characteristics described by Hendrix [26].

The simple flotation technique is a standard method for screening eggs of the Taeniidae family, with an estimated sensitivity of 60%-80% in mild infections due to intermittent egg elimination in the feces. In terms of specificity, it is highly effective in identifying the genus. To ensure the reliability of coproparasitological findings, the following measures were implemented: A second expert analyst independently re-examined 10% of samples (n = 45); Cohen’s kappa index was calculated, obtaining a value of 0.82, which represents a “near-perfect” level of agreement according to the Landis and Koch scale (interobserver reliability); a blind reading was performed, in which analysts conducted the microscopic examination without prior knowledge of the sample’s origin or the results of cohabitation with livestock, thus eliminating possible observation biases; and duplicates of samples that initially tested negative but came from areas with high livestock density were analyzed to minimize the false negative rate.

A 0.85% saline solution was used exclusively for the initial homogenization of the sample and for sieving macroscopic debris during the washing and homogenization phases. A saturated solution of sodium nitrate (NaNO₃) or Sheather’s solution (saturated sugar) with a specific gravity of 1.20–1.27 was used for the recovery of Echinococcus/Taenia eggs. This ensures that the eggs, which have a lower density, float to the surface while the heavy debris settles. Observation was performed using a binocular optical microscope with an initial scanning magnification of 10× (to locate suspicious structures) and morphological confirmation at 40× (to identify internal hooks and the striated membrane characteristic of eggs of the Taeniidae family). Finally, a double-blind validation was performed to ensure inter-observer agreement, with each slide being independently read by two trained biologists. In case of discrepancy in the result (positive/negative), a third review was performed by a supervisor from the INS National Reference Laboratory.

For copro-PCR diagnosis, the sifted stool samples were thawed, and genomic DNA was extracted using the commercial SIGMA kit (Sigma-Aldrich, USA) following the manufacturer’s procedure, with minor modifications to concentrate the genomic DNA from the samples. A pair of primers was used to amplify E. granulosus DNA, selected from the mitochondrial gene cox1 (GenBank accession no. AF297617); primer sequences CO1-F (5′-TTTTTTGGCCATCCTGAGGTTTAT-3′) and CO1-R (5′-TAACGACATAACATAATGAAAATG-3′), producing an amplicon of 460 bp. These primers amplify DNA from almost all E. granulosus strains; therefore, the primer pair was selected for the identification of the cox1 gene segment, which is intended to serve as an amplification target for the copro-PCR diagnosis of E. granulosus spp. in dog feces [27].

The PCR reaction was performed in a final volume of 25 μL, containing 25 μL DreamTaq 2× Green PCR Master Mix (Thermo Scientific, Waltham, MA, USA), 18.5 μL nuclease-free water, 0.5 μL each primer, and 2 μL genomic DNA. PCR amplification was performed on the Mastercycler Pro S & Control Panel (Eppendorf, Hamburg, Germany) using the following thermal conditions: first, an initial denaturation cycle at 94°C for 4 min. Then, 38 denaturation cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s were performed. Third, the final elongation was performed at 72°C for 5 min, and the sample was stored at 4°C in the same instrument. The amplified products were confirmed and analyzed using 1.5% agarose gel electrophoresis. Electrophoresis was performed for 45 min, with 80 V for the first 15 min and 120 V for the next 30 min [1, 6].

To ensure the validity of the molecular diagnosis of E. granulosus, critical performance controls (positive and negative) were implemented: Positive Control, where genomic DNA from E. granulosus (locally confirmed strain) was included in each PCR run to validate the efficiency of amplification and the integrity of the reagents (master mix); negative control (NTC), where a contamination-free control (molecular biology grade water) was used in each plate/set of tubes to detect any traces of contamination during reaction preparation; and extraction control, where a known negative stool sample was processed during the DNA extraction phase to monitor the cleanliness of the kits and lysis reagents.

Area segregation was considered in terms of contamination prevention measures, where the workflow was carried out in separate and physically isolated rooms for the three critical stages: (a) preparation of the PCR master mix, (b) DNA extraction from fecal samples, and (c) amplification/analysis of products (post-PCR). Physical barriers were also used, with pipetting and reaction preparation carried out in a laminar flow hood under ultraviolet (UV) light for decontamination before and after use. Only filter tips (aerosol-resistant tips) were used.

Finally, all samples were processed in duplicate. In cases of discrepancy between the duplicates (one positive and one negative), a third confirmation run was performed. Only samples with consistent amplification in both tests were considered “positive.”

The presence or absence of Echinococcus spp. nucleic acids was determined by quantifying only the number of positive dogs to determine the prevalence of this parasitic disease; however, circulating genotypes were not identified.

Data analysis and processing were performed using SPSS v.25 (IBM Corp., Armonk, NY, USA). The categorical variables were statistically contrasted using the Chi-square independence test, with the asymptotic or exact method, and the odds ratio with 95% confidence intervals. A p-value ≤ 0.05 was used to indicate statistical significance.

Univariate and multivariate logistic regression analyses were performed to assess potential associations. Although variables with significant associations were identified for canine echinococcosis, multivariate analysis was not carried out due to collinearity among some factors evaluated. Missing data were handled using the Maximum Likelihood Models technique to avoid bias and loss of statistical power, using all available information to estimate the model parameters without the need for explicit imputation, assuming that the missing data were random.

The Hosmer–Lemeshow test was used to assess model goodness of fit and evaluate whether there was a significant difference between the observed and expected frequencies of positive cases of echinococcosis across the model-defined risk groups. Confounding factors, referring to external variables associated with both exposure and outcome, were controlled for by performing a multivariate logistic regression in which the confounding factors (owner’s gender and educational level, dog’s area of residence, type and materials of housing, water supply, type of drainage, open defecation, night light generator, coexistence with other animals, number of animals raised, and parasitic coinfection) were included as independent variables in the model along with the main variable of interest. The model calculated the adjusted odds ratio (OR), which represents the association of interest after “neutralizing” the effects of confounding variables.

For prevalence and biological factors, the unit of analysis was the dog, but for environmental and management factors (such as offal disposal), the unit was the household. Listwise deletion was used to handle missing data; that is, only subjects with complete information for all variables analyzed were included in the final models. An exploratory analysis was performed using Pearson’s Chi-square test to identify individual associations with E. granulosus positivity. Variables were coded to facilitate the interpretation of the OR: binary (dichotomous): presence of the condition = 1, absence = 0. Dummy variables were used for categorical/ordinal variables to establish a reference category.

A multivariate model (logistic regression) was used to adjust for possible confounding variables and identify independent predictors of infection. All variables with a p-value ≤ 0.20 in the univariate analysis were included. This preventive threshold ensured that variables, although not significant on their own, contributed to the model in the presence of other variables were not discarded. The backward stepwise method based on likelihood was used. A multicollinearity test was performed to assess the scientific validity of the inferences, with variance inflation factors (VIFs) calculated for the predictors. A VIF >5 was considered indicative of multicollinearity, and redundant variables were eliminated or combined. In addition, the Hosmer–Lemeshow test was applied, with a p-value >0.05 indicating that the model adequately fit the observed data. Figure 2 shows the methodological workflow for detecting E. granulosus in dogs.

Most dog owners were male (61.1%) and had a primary education (58.5%) (Table 1). The average age of the dogs was 3.9 (SD = 3.5) years and that of their owners was 53.9 (SD = 17) years. 68.5% (371/542) of the dogs were male and 31.5% (171/542) were female. All the dogs were mixed breeds, used for guarding and herding. The study showed that 47.3% of dogs were fed offal containing cysts at some point, and 7% were fed cooked offal. None of the dogs were confined because homes in rural areas do not usually have a complete perimeter fence, allowing dogs to move freely inside and outside the home. It was found that 65.4% of dogs had been dewormed at some point, and 89.6% had been dewormed within the last year.

On the other hand, as shown in Table 2, the prevalence of E. granulosus in dogs was 10.8% (49/453; 95% CI = 7.9–13.8). In addition, the villages with the highest prevalence of echinococcosis were Pastales and Sacsalla, with 29.4% (95% CI = 17.9–40.9) and 16.7% (95% CI = 6.4–26.9), respectively. The other villages had a prevalence of 8%. Sex was not associated with echinococcosis (p > 0.05).

Table 3 shows that none of the characteristics of the dwelling (housing material, type of dwelling, water supply, toilets, and energy source) were associated with echinococcosis in dogs (p > 0.05).

The coexistence of dogs with donkeys (p < 0.05) and alpacas (p < 0.01) in the same household was associated with echinococcosis in dogs (Table 4).

The stool examination showed that the dogs were infected with different parasites, such as Strongyloides sp. (7.2%; 39/536; 95% CI = 4.9–9.5), Taenia sp. (6.6%; 36/536; 95% CI = 4.5–8.8), Toxascaris sp. (1.8%; 10/536; 95% CI = 0.6–3.1) and Trichuris sp. (0.4%; 2/536; 95% CI = 0.1–1.3). Table 5 shows that the coinfection of E. granulosus with Taenia sp. and Strongyloides sp. showed a statistically significant association (p < 0.01). However, coinfections with Trichuris sp. and Toxascaris sp. were not associated (p > 0.05).

No statistically significant difference was found between the age of the owners and echinococcosis in dogs (p > 0.05), nor was there a difference between the number of inhabitants in the house and echinococcosis in dogs (p > 0.05). Table 6 shows the difference between the number of alpacas raised and the positivity of echinococcosis in dogs (p < 0.01). So much so that in the homes of dogs with echinococcosis, more alpacas were raised (x̅ = 293) compared with the homes of dogs that tested negative (x̅ = 215). The number of other bred animals showed no statistically significant difference with echinococcosis in dogs (p > 0.05).

The prevalence of Echinococcus granulosus in dogs in this study (10.8%) was similar to that found in Uganda [28], India [29], Iran [30, 31], Turkey [32], Kenya [33], and Israel [34]; it was only higher than that reported in the Falkland Islands [35] but lower than that reported in Iraq [36], Iran [37], and Morocco [38]; therefore, Ascension is recognized as a rural, remote, and epidemiologically underserved area. Demonstrating active parasite circulation fills an important knowledge gap in national surveillance.

The study showed that the average age of dog owners was 54 years, and half had only primary education (50.6%), unlike another study conducted in Andahuaylas, Peru, where most dog owners had higher education [39]. Consequently, having a primary education would imply that knowledge about dog ownership is precarious and that there is a greater risk of becoming infected with CEC [11].

Dogs in some villages had a higher prevalence of echinococcosis (Pastales and Sacsalla), because the localities in the Suni altitudinal zone (3500–4000 m.a.s.l.) have a higher prevalence than those in the Quechua zone (2300–3500 m.a.s.l.) [2]. Although altitude is not a direct factor in causing this disease, it does affect prevalence, particularly in mountainous areas where interactions between domestic and wild animals are common [40].

Keyhani et al. [20] found no association between dog sex and echinococcosis. However, in other research [41, 42], it is mentioned that sex is a risk factor for contracting E. granulosus, especially for male dogs, because people prefer to have male dogs as they better protect the home and livestock and because they do not reproduce unlike females [42].

House characteristics, such as housing material and type, water supply, toilets, and energy source, were not associated with echinococcosis in dogs. This result is consistent with reports from other Peruvian regions, likely because this parasitosis is endemic throughout the district [2].

The coexistence of dogs and alpacas in the same house was associated with echinococcosis in dogs, probably because dogs accompany livestock during grazing [41], resulting in continuous contact that increases the risk of disease [43]. However, the genotypes G7 (E. canadensis) [10] and G1 (E. granulosus sensu stricto) have been reported in alpacas from endemic regions of Peru, which is the most common strain identified in humans and animals in endemic regions around the world [44].

An association was also found between the coexistence of dogs and donkeys; however, this association may be due to chance, given the small number of donkeys raised in the Ascension houses. However, another study found that sheepdogs that lived with donkeys were less likely to be infected with echinococcosis, because people who raise donkeys tend to have higher incomes than those who do not. Consequently, dogs from higher-income households may be better fed and less likely to snoop around in the garbage [45].

Stool examination showed that the dogs were infected with Strongyloides sp., Taenia sp., Toxascaris sp., and Trichuris sp., as reported in previous studies [33, 34, 46, 47]. However, a statistically significant association was found between the coinfection of E. granulosus with Taenia sp. and Strongyloides sp. Coinfection with Taenia sp. was also reported by Hartnack et al. [47], who noted that dogs can be infected with both Taenia sp. and Echinococcus spp. through predation and local prey species are infected with metacestodes of both genera. However, no association between E. granulosus and other parasites has been reported [34]. The findings show a strong statistical association between E. granulosus infection and coinfection with Strongyloides sp. and Taenia sp. (p < 0.01). Few studies in Latin America have explored coinfections in detail. This adds originality by highlighting the ecological overlap between helminths in dogs. Coinfection profiles could serve as predictive markers of Echinococcus risk.

The study also showed that in the homes of dogs with echinococcosis, more alpacas were bred (x̅ = 293) compared to the homes of dogs that tested negative, in whose homes the number of alpacas bred was lower (x̅ = 215), which shows that the larger size of the alpaca herds would constitute a risk factor [28], since the breeding of these animals is inadequate in Huancavelica because it is done only as a source of food in pastures of low bearability and with deficient sanitary care [48]. A critical finding is the direct association between the number of raised alpacas and positivity in dogs. Unlike reports from other high Andean areas where sheep are the main reservoir, alpacas have emerged as a key epidemiological node in Ascension. The fact that households with infected dogs have larger herds suggests that, as production scale increases, the likelihood of home slaughter and accidental access by dogs to infected viscera also increases, reinforcing the importance of this camelid in the human transmission chain.