Cryptosporidium animal species in Iran: a systematic review and meta-analysis

Background Cryptosporidiosis is an acute and short-term infection which can lead to severe diarrhea (intestinal cryptosporidiosis) associated with a persistent cough in the host with immune system defect. This systematic review and meta-analysis was conducted to estimate the prevalence of animal Cryptosporidium species and the corresponding epidemiological aspects in Iran. Methods In this study, all original research articles relating to the animal cryptosporidiosis in Iran were collected from reliable databases using keywords. A meta-analysis was conducted separately for each subgroup, and heterogeneity among the studies was performed using the Q and I2 tests. Furthermore, it should be noticed that the significance level in the statistical analysis with the Comprehensive Meta-analysis software was considered to be less than 0.05. Finally, meta-analysis results were shown in forest plot with a 95% CI. Results In total, 4795 studies were included in the initial screening. Duplicated or non-original studies and the ones which did not meet our considered criteria were excluded from the list. Out of the 100 articles included in our first list for the meta-analysis, 40, 16, 13, 10, 9, 7, and 5 were done on cattle and calves, birds, dogs, sheep, rodents, camels, and horses, respectively. The prevalence rate of cryptosporidiosis among the birds, horses, rodents, camels, dogs, cattle, and sheep in Iran was estimated to be 7.5%, 19.5%, 20.8%, 8.4%, 4.9%, 14.4%, and 9.1%, respectively. Conclusion The different Cryptosporidium species have been found in different regions of Iran. Geographical region, climate, and domestic animals are considered as factors responsible for animal cryptosporidiosis prevalence in the area. Moreover, this parasite is zoonotic which causes disease in animals as well as humans which can result in economic loss. Supplementary Information The online version contains supplementary material available at 10.1186/s41182-020-00278-9.


Introduction
Intestinal parasites are considered an important public health problem in humans and animals in developing and low-income countries [1][2][3]. Cryptosporidium is one of the most common intestinal protozoan parasites which is located in the phylum of Apicomplexa and causes cryptosporidiosis [2]. Cryptosporidium is spread easily in the environment due to its simple transmission via contaminated water, air, and dust. Cryptosporidiosis is considered as a major economic problem in many countries including Iran, and there are annually numerous reports of this infection in immunocompromised and young children [2][3][4][5]. Different methods have been used to detect the protozoan parasite including the molecular diagnostic method which is considered one of the most useful diagnostic tools.
This method has identified up to 30 species and more than 50 genotypes of Cryptosporidium [6][7][8]. Different species of Cryptosporidium have been reported in various hosts including birds, horses, cattle, sheep, camels, rodents, and dogs. Cryptosporidium parvum, C. hominis, C. canis, C. felis, C. meleagridis, and C. muris were distinguished from gastrointestinal diseases as well as diarrhea in humans [9]. However, the infection typically occurs in a short-term and acute form in immunocompromised and HIV-positive individuals. In these cases, it tends to remain in the lower intestine for up to 6 weeks with severe diarrhea and persistent cough [9]. This protozoan could develop its life cycle in one host without the requirement of other animals as intermediate or reservoir hosts [9,10]. Despite the many studies which were conducted in Iran on investigating the prevalence of cryptosporidiosis in different animal hosts, these data have not shown the overall prevalence of animals in Iran. Since cryptosporidiosis causes irreversible economic damages to domestic animals, critical screening programs and epidemiological aspects should be considered by authorities. This systematic review and meta-analysis was conducted to estimate the prevalence of animal Cryptosporidium species and their epidemiological aspects in Iran.

Study protocol
The present systematic review focused on the estimates of the prevalence of animal Cryptosporidium species according to the PRISMA guidelines for systematic review and meta-analysis [11] (Supplementary 1).

Search strategy
In order to select the suitable articles for this study, all records since 1991 up to February 2018 were investigated using seven international databases in English including PubMed, Web of Science, Scopus, Science Direct, and Google Scholar search engine as well as national databases in Persian including Magiran (http://www.magiran.com/) and Scientific Information Database (SID) (http://www.sid.ir/). Furthermore, references of each article were screened manually, and the authors were contacted for additional references.

Eligibility criteria
All original descriptive studies which investigated the prevalence of Cryptosporidium in animals in Iran, both in English and Persian, were included in this study. Duplicates, qualitative studies, review articles, case reports, case series, and studies out of Iran or those performed on humans were excluded. Finally, articles with epidemiological parameters of interest were selected, and a total of 100 collected articles fulfilled the considered criteria.

Quality assessment
The scoring system based on the 8-item modified Newcastle Ottawa Scale (NOS) for non-randomized studies was used for assessing the quality of the studies. In this system, each question has a score between 0 and 1, and the maximum point summation is 9. Studies with point summation 5 or less, 6-7, and 8-9 were considered low, moderate, and high quality, respectively [12].

Screening and data extraction
All records were evaluated based on their title and abstract and according to the inclusion and exclusion criteria by two researchers (MM and NB) independently. The kappa index showed an agreement of 91% between the findings of two researchers. The full-text version of the papers was obtained through library resources and online databases. Finally, the difference between records among the researchers was corrected by re-examining the articles. The agreement was reached by group discussion with a third researcher (SK).
Data extraction was conducted independently by two researchers (MM and NB) and imported to the preprepared form. Data including authors, year of study, publication year, kind of animal, geographical area of the

Quality assessment studies
The methodological quality of the studies was examined based on the guidelines of the Newcastle and Ottawa Statements [12]. This guideline sets the criteria for selecting people to study, comparing and accepting them, as well as exposure and consequences where a maximum of 9 stars can be allocated to each study. Studies with 7 stars or more are classified as highquality studies, and studies with 6 stars and less are considered as low-quality studies. Investigating the probability of an error in the results of the studies is performed separately by two researchers. The disagreement between the parties is resolved through negotiations.

Data analysis
The meta-analysis method was adopted to a 95% confidence interval (CI) in order to assess the pooled prevalence of Cryptosporidium infection in animals using the random effect model. Various subgroup analyses were separately conducted based on animal type and the associated species. Finally, meta-analysis results were displayed in forest plot (reported as effect estimates (ES) with a 95% CI). We also performed a sensitivity analysis to verify the stability of the data. In order to assess the sensitivity analysis, the effect estimate was estimated irrespective of one study at a time, and the robustness of the pooled estimate was assessed. Heterogeneity was calculated among the studies by the Q and I 2 tests [6,8].
Cochran's Q test (Q statistic, p < 0.10) showed statistically significant heterogeneity, and I 2 statistic (I 2 > 50%) indicates a large heterogeneity. Statistical analysis and data analysis were performed using the second version of the Comprehensive Meta-analysis software. The significance level was considered to be less than 0.05.

Description of studies
In total, 4795 studies were collected in the initial screenings from the published articles and their references in     Fig. 1 and Table 1.
Within these 100 articles, 40,16,13,10,9,7, and 5 studies were performed on cattle, birds, dogs, sheep, rodents, camels, and horses, respectively. The most frequent studies were performed on cattle and the least ones on horses. Considering the various projects in searching cryptosporidiosis on cattle in Iran, the distribution of positive cases relating to cattle is presented in Fig. 9. The quality assessment of studies using the guideline of the Newcastle Ottawa Scale showed that 27%, 65%, and 8% of the studies have low, medium, and high quality, respectively.      39.93% for sheep, respectively. Low heterogeneity was reported in studies which were conducted on horses as well as the ones on dogs. The statistic factors (Q statistic, df, and I 2 ) were 3.86, 4, 0.00% for horses and 6.07, 12, and 0.00% for dogs.
The prevalence rate of Cryptosporidium infection in cattle is shown in Fig. 9. The most positive cases of cryptosporidiosis were reported in cattle of West Azerbaijan, Tehran, Khuzestan, Chaharmahal and Bakhtiari, Kohgiluyeh and Boyer-Ahmad, and Kerman provinces. Considerable positivity rates of cryptosporidiosis in cattle were identified in Razavi Khorasan, South Khorasan, Semnan, Hamadan, Alborz, and East Azerbaijan provinces. There were no positive reports of cryptosporidiosis in cattle in other provinces.

Discussion
Cryptosporidiosis is one of the most important zoonotic diseases which is reported in humans and animals with a worldwide distribution in more than 106 countries and especially in developing countries [91,92]. To the best of our knowledge, this is the first systematic review and meta-analyses on the prevalence of animal cryptosporidiosis in Iran.
The present study showed that the average prevalence rate of cryptosporidiosis in birds was 7.5% in Iran. Additionally, the prevalence rate of cryptosporidiosis in animals in Ahvaz, southwestern of Iran was reported as 50% [86] while it was shown that in Gilan, north of Iran, this rate was 17% [84]. In a study by Jasim and Marhoon, it has shown that in Iraq, which is a neighboring region of Iran, the cryptosporidiosis prevalence rate in wild and domestic birds was 58.1% [93]. Even though this region is near the southwest of Iran, the prevalence was higher than in Iran. Moreover, a prevalence of 49% was shown in Mexico [94], and in Brazil, 76% of birds were infected by Cryptosporidium [95]. Changes in the prevalence seen in various reports indicate that the probability of transmitting the parasite is higher among animals living together on farms and next to each other compared to other studies that have examined individual specimens. Preventive efforts by Iranian authorities related to awareness of zoonotic diseases, control of stray dogs, and a low population of pet dogs have increased the possibility of transmission of the disease from livestock [96]. On the one hand, the stray dogs are the largest group of dogs in both rural and urban areas in Iran which usually become infected by roaming in human neighborhoods and feeding on contaminated residues. On the other hand, domestic dogs are not restricted to the limited area of houses or farms. Stray dogs are allowed to wander around, so it increases the risk of zoonotic infections in rural habitats. In this study, the overall prevalence of cryptosporidiosis between dogs was found to be 4.9% in Iran. There are various reports of cryptosporidiosis prevalence in the different geographical regions of Iran. Mohaghegh et al. reported a prevalence of 21.7% and 25.4% of cryptosporidiosis respectively in domestic and stray dogs of Kermanshah [61]. Furthermore, the 12.3% prevalence rate of cryptosporidiosis in dogs was observed in Ahvaz [60]. These results are higher than the data which were obtained in other regions of Iran, specifically 5% in Chenaran, northeast of Iran [56]; 7% in Ilam [97]; 2% in Kerman, southeast of Iran [98]; 2.9% in Urmia, northwest of Iran [99]; 2.14% in Isfahan, center of Iran [62]; and 3.8% in Hamadan, west of Iran [17]. The high prevalence of cryptosporidiosis in some areas, for instance, Kermanshah Province, indicated that humans are at serious risk of Cryptosporidium infection. Furthermore, the infection can spread vastly and cause severe problems in the community.
Epidemiological studies on cryptosporidiosis infection indicated that the prevalence of Cryptosporidium species in dogs is very different in various countries changing from 0 to 52.7%. These differences might be attributed to several factors, such as geographical area, sample size, keeping a dog, correlation with other hosts (such as goat, sheep, horse, cattle, and pig), different species of Cryptosporidium, and sampling procedures as well as diagnostic methods [100,101]. The current results imply that the prevalence rate of cryptosporidiosis in Iran is higher than in countries such as the Czech Republic with 1.4% [102], Thailand with 2.1% [103], Brazil with 2.4% [104], Japan with 3.9% [105], and Spain with 4.1% [106], but lower than in Nigeria with 18.5% [107] and Romania with 52.7% [108].
The prevalence rate of cryptosporidiosis in cattle was 1.5% in Japan [119], 35.7% in Vietnam [120], 20.6% in Turkey [121], 40.6% in Canada [122], and 40.6% in the USA [123]. However, according to this systematic review and meta-analysis, the prevalence of cryptosporidiosis in cattle and calves was 14.4% in Iran, and the prevalence rate in various geographical regions was as follows: 2.1% in Ahvaz [73], 9.07% in Lorestan [18], 16.45% in Qazvin [26], 22.3% in the city of Urmia [48], and 28.3% in the city of Mashhad [47]. Furthermore, our study showed that the prevalence rate of cryptosporidiosis in camels and horses was 8.4% and 19.5%, respectively.
It was suggested in this study that the distribution of Cryptosporidium differs among geographical regions. Therefore, the study location might be one of the most determinant factors in cryptosporidiosis distribution. The highest prevalence rate (50%) of cryptosporidiosis was observed in Khuzestan Province [86]. This high prevalence might be attributed to the high temperature and humidity of the southwestern regions of Iran as well as the people's lifestyle, who have a high level of seafood consumption compared to other regions. Additionally, the immigration of birds to the south of Khuzestan Province may transmit parasite protozoan infection.

Conclusion
The relatively high prevalence of cryptosporidiosis infection among animals in Iran, mostly among sheep, cattle, and calves, shows the enzootic status of cryptosporidiosis in the investigated areas and may be a threat to the inhabitants. Our data offer important information about the epidemiology of cryptosporidiosis among animals in Iran, which could be useful for managing and controlling programs for the disease. Further investigation and monitoring will be required to expand the surveillance and control policies in order to reduce the prevalence of Cryptosporidiosis among livestock and consequently decrease the economic damages and public health hazards in Iran.

Abbreviations
HIV: Human immunodeficiency viruses; CI: Confidence interval