- Email: [email protected]
Prevalence of Coxiella burnetii in cattle and buffalo populations in Punjab, India
Preventive Veterinary Medicine 166 (2019) 16–20
Contents lists available at ScienceDirect
Preventive Veterinary Medicine journal homepage: www.elsevier.com/locate/prevetmed
Prevalence of Coxiella burnetii in cattle and buffalo populations in Punjab, India
T
⁎
R. Keshavamurthya, B.B. Singha, , D.G. Kalambhea, R.S. Aulakha, N.K. Dhandb a b
School of Public Health & Zoonoses, Guru Angad Dev Veterinary and Animal Sciences University, Ludhiana, Punjab, 141004, India Sydney School of Veterinary Science, The University of Sydney, Camden, NSW, Australia
A R T I C LE I N FO
A B S T R A C T
Keywords: Buffalo Cattle Coxiella burnetii India Prevalence Punjab Q fever
Q fever is an important zoonosis of animal and public health significance but there is very limited information about its prevalence in the Punjab state of India. The current study was designed to estimate Q fever prevalence in cattle and buffalo populations of the state. We randomly selected 22 villages, one from each of the 22 districts of Punjab. Households in these villages were randomly selected using village voter lists to ensure representative sample collection. Blood, vaginal swab and milk samples were collected from the animals in these enrolled households. Serum samples were screened using Coxiella burnetii specific IgG ELISA whereas milk and genital swab samples were subjected to a Trans-PCR assay. The agreement (Cohan’s Kappa) between shedding of C. burnetii in milk and genital secretions and between ELISA and Trans-PCR was estimated. The selected PCR products were sequenced, and phylogenetic analyses were performed. We collected 610 blood samples, 610 genital swabs and 361 milk samples from 610 bovines (378 cattle and 232 buffaloes) in 179 households. Considering all tests in parallel and after adjusting for clustering, we estimated an overall individual animal prevalence of Q fever of 7.0% (95% CI: 4.7, 9.4). There was a low agreement between shedding of C. burnetii in milk and genital secretion (kappa: 14.3%; 95% CI: 5.6, 22.9) and between ELISA and Trans-PCR (10.3%; 95% CI: 3.2, 17.4%). Phylogenetic analysis confirmed all samples to be of C. burnetii. The results suggest that the disease is present in the state and further epidemiological information should be collected to determine its zoonotic potential and its impact on animal and public health in Punjab, India.
1. Introduction Q fever is a widely distributed zoonotic disease of animal and public health concern. The causative agent Coxiella burnetii is a Gram-negative intracellular bacterium distributed across the globe except in New Zealand and Antarctica (Hilbink et al., 1993; Kaplan and Bertagna, 1955). It is virtually present in all animal kingdom, including birds, arthropods, and rodents but it mainly affects cattle, sheep, goat and humans (Lang, 1990). Recent epidemiological studies have shown an increased prevalence of Q fever in livestock populations in many parts of the world raising public health concerns (Cruz et al., 2018; ArricauBouvery and Rodolakis, 2005; Cardinale et al., 2014). Q fever is a disease of economic significance in farm animals especially in sheep and goats and has a potential of causing considerable losses to the livestock industry (Asseldonk et al., 2013). In cattle, the disease is usually asymptomatic and dramatic clinical manifestations are rare (Arricau-Bouvery and Rodolakis, 2005), but if present, they are usually associated with reproductive complications such as
⁎
abortions, stillbirth, repeat breeding, anoestrus and premature birth (Lang, 1990). Human exposure to infection is mainly associated with the inhalation of contaminated aerosols generated during parturition, slaughter, cleaning and other activities at the farm. Role of cattle as a source for human infection is known in Australia but an association could not be found in Western Europe (Georgiev et al., 2013; SloanGardner et al., 2017). Infection in humans mostly causes acute and selflimiting flu-like illness with symptoms of headache, and atypical pneumonia. However, the chronic form can be severe and fatal without appropriate antibiotic therapy (Honarmand, 2012). Role of water buffalo (Bubalus bubalus) in Q fever epidemiology is not clearly understood but buffaloes are closely related to domesticated cattle and both are members are Bovidae family (Michelizzi et al., 2010). Buffaloes along with cattle are an important source of milk, raised together and are major contributors for the dairy industry in India. Q fever prevalence has been reported in buffaloes from some parts of the world including India and Africa (Klemmer et al., 2018; Vaidya et al., 2010) but the veterinary and economic significance of this
Corresponding author. E-mail address: [email protected] (B.B. Singh).
https://doi.org/10.1016/j.prevetmed.2019.03.003 Received 12 October 2018; Received in revised form 5 March 2019; Accepted 6 March 2019 0167-5877/ © 2019 Elsevier B.V. All rights reserved.
Preventive Veterinary Medicine 166 (2019) 16–20
R. Keshavamurthy, et al.
were replaced with the next individuals in the voter list. Individuals keeping bovines in their households were contacted and requested to participate in the study. If the selected individuals were not willing to participate in the study, they were considered as non-responders but were not replaced with other individuals as we had already considered an expected response rate of 40% in calculating sample size. If the farmer agreed to participate, we strived to collect samples from all of their animals although many farmers did not agree to provide samples from some of their animals. The details of the number of households contacted, responded and the calculated response rate from all villages are presented in Table S1. Blood and genital (vaginal or preputial) swab samples were collected from the selected cattle (crossbred/exotic and indigenous) and buffalo. Additionally, milk samples were collected from lactating females. The samples were transported to the laboratory on the same day and stored at −20 °C until further used. The sera were separated within 24 h in the sterile cryovials and stored at −20 °C until screened. The serum samples were screened for C. burnetii specific antibodies using commercial Q fever indirect ELISA kit containing test plate precoated with Phase I and Phase II antigenic extract (Q fever Monoscreeen Ab ELISA (IgG), Bio-X Diagnostics, Rochefort, Belgium). Milk, vaginal and preputial swabs were subjected to molecular screening to detect Coxiella DNA. Briefly, DNA was extracted using DNeasy Blood and Tissue kit (Qiagen, Maryland, USA). Trans-PCR assay, a PCR targeting the transposase gene of repetitive transposonlike elements of C. burnetii called IS1111 (Willems et al., 1994) was then employed using a published set of primers targeting 687 bp of IS1111a insertion sequence of C. burnetii (Berri et al., 2000). The referral positive DNA control for Coxiella was provided in-kind by the Department of Veterinary Public Health and Epidemiology, Nagpur Veterinary College, MAFSU. In our study, an animal was defined as seropositive if its serum sample was ELISA positive and as molecular positive if either milk or genital swab sample of the animal was Trans-PCR positive. An animal was declared overall positive by using a combination of tests in parallel, i.e. an animal was considered as overall positive if it was seroor molecular- positive. A selection of positive amplified products showing desired band size was purified and sequenced (n = 5) in both the directions (1st base, Apical Scientific, Selangor, Malaysia). The sequences were submitted to NCBI GenBank. The distance-based analysis was conducted using published Coxiella burnetii sequences (Gene Bank accession numbers KT954146.1, KR697576.1, EU000273.1, AB848993.1, KT391020.1, J N966900, JF972643, KP719167.1, MF197399.1, M80806, HM222935, KT381466.1 and EU009657.1). The sequence of Legionella pneumophila (AB594755.1) was used as an outgroup in the analysis. The phylogenetic tree was constructed using the maximum likelihood method based on the Tamura-Nei model by Mega 6.0 software. The bootstrap consensus tree inferred from 1000 replicates was taken to represent the evolutionary history of the taxa analysed. Prevalence estimates and their 95% confidence intervals were calculated separately for sero, molecular and overall prevalence using SAS Surveyfreq procedure after accounting for stratification by district and clustering by herd. Prevalence estimates were classified by species, age, breed and sex. Cohan’s Kappa and its 95% confidence intervals were calculated using SISA website (https://www.quantitativeskills.com/ sisa/statistics/twoby2.htm) to determine the agreement between shedding of C. burnetii in milk and genital secretions as well as to determine the diagnostic agreement between ELISA and Trans-PCR.
disease in buffaloes is not clearly known. Very limited information is available about Q fever prevalence in Punjab – an agriculture state home to 2.4 million dairy cattle and 5.2 million dairy buffaloes (Bubalus bubalus). In 1970s Randhawa et al. (1973) reported a high prevalence of 25.5% of Q fever in human patients suffering from a fever of unknown origin. They conducted further epidemiological investigations which revealed 16.2% prevalence of Q fever in cattle, 9.6% in buffaloes, 6.1% in goats and 3.7% in sheep, indicating the existence of the disease in Punjab. Later, Sodhi et al. (1980) tested serum samples from a farm with a high rate of abortions and found 23.2% and 24.1% prevalence of Q fever in cattle and buffaloes, respectively. However, we are not aware of any other studies conducted in Punjab to estimate the prevalence of Q fever in cattle and buffalo since then. Moreover, none of the studies conducted in Punjab, or in any other part of India for that matter, used probability sampling approaches for prevalence estimation. In Punjab, studies on the prevalence of zoonotic and economically significant diseases in livestock are mainly restricted to brucellosis. The studies on other infectious agents have not been given much consideration. Therefore, losses to livestock sector due to other pathogens are often overlooked and masked. The current study was, therefore, planned to estimate the prevalence of C. burnetii in the bovine populations of Punjab by collecting and testing representative samples from all districts. The results will provide information about the extent of the problem in the state to enable animal health authorities in making appropriate policy decisions in controlling the disease. 2. Methods This study was approved by the Institutional Animal Ethical Committee, Guru Angad Dev Veterinary and Animal Sciences University (GADVASU/2017/IAEC/42/02). The study was conducted using a multi-stage sampling design. We decided to select 22 villages, one per district of the state to ensure the collection of a representative sample from the entire state. Households were then selected using probability proportional to size sampling based on the number of households in a village. The target population of the study was the domestic bovine population in the state. The study population was the domestic bovine population of 22 villages selected to participate in the study. The sample size was calculated to be 414 to estimate Q fever prevalence in the bovine population of Punjab with 95% confidence and 5% precision, assuming a prevalence of 16% based on previous works (Randhawa et al., 1973; Vaidya et al., 2010) and after accounting for a design effect of 2, because animals are clustered within households and villages. Considering a response rate of 40% and an average herd size of 4.4 (see Appendix A), we estimated that we would have to contact 234 farmers to achieve the animal sample size of 414. Accounting for finite population only marginally reduced the number of farmers to 231. Using data about the number of households in 22 villages (5214) and our estimate of the proportion of households keeping livestock (0.54, see Appendix A), we determined that 2816 households in the selected 22 villages would keep livestock (called farmers from here onwards for brevity). To ensure probability proportional to size sampling of farmers, we calculated the proportion of farmers in each village of the total farmer population in the selected 22 villages (i.e. 2816). Multiplying this proportion with the required sample size of this study (i.e. 231 farmers) yielded the number of farmers required to be contacted in each village (see Table S1 for details). We randomly selected one village from each of the 22 districts using the Survey Toolbox program (Cameron, 1999) and then used the voter list of each village for a random selection of farmers. Briefly, sampling interval for systematic random sampling was calculated by dividing the total number of voters in each village by the sample size of each village and was then used in the selection of farmers from the voter list. Individuals not owning the animals or those belonging to the same family
3. Results We visited 247 randomly selected rural households in the selected 22 villages keeping cattle and/or buffaloes. Of these households, 179 smallholders (representing individual households) agreed to participate in the study (Table S1). Although we aimed to select all animals in the selected household, many of the smallholders allowed us to sample only 17
Preventive Veterinary Medicine 166 (2019) 16–20
R. Keshavamurthy, et al.
Table 1 Prevalence of Q fever in the bovine population in Punjab, India based on a study conducted in 2017-18. Serum samples obtained from cattle and buffalo were tested using indirect ELISA. Milk, vaginal and preputial swabs were tested using trans-PCR assay. Categories
Sample size
Number of ELISA positives
Sero-prevalence (95% CI)
Number of PCR positives
Molecular prevalence (95% CI)
Overall positives
Overall prevalence (95% CI)
Overall
610
33
5.4 (3.4-7.4)
13
2.1 (0.9-3.4)
43
7.0 (4.7-9.4)
Species Cattle Buffalo
378 232
24 9
6.3 (3.6-9.1) 3.9 (1.4-6.4)
11 2
2.9 (1.2-4.6) 0.9 (0.0-2.1)
33 10
8.7 (5.5-11.9) 4.3 (1.7-6.9)
Cattle breeds Crossbred/Exotic Indigenous
322 56
21 3
6.5 (3.4-9.6) 5.4 (0.0-11.3)
9 2
2.8 (0.9-4.7) 3.6 (0.0-8.1)
29 4
9.0 (5.4-12.6) 7.1 (0.6-13.6)
Sex Male Female
21 589
0 33
0 (-)* 5.6 (3.6-7.6)
0 13
0 (-)* 2.2 (0.9-3.5)
0 43
0.0 (-)* 7.3 (4.9-9.7)
Age Under 1 year 1 to 3 years Over 3 years
52 101 457
1 2 30
1.9 (0.0-5.6) 2.0 (0.0-4.6) 6.6 (4.1-9.1)
0 1 12
0 (-)* 1.0 (0.0-2.9) 2.6 (1.0-4.3)
01 3 39
1.9 (0.0-5.6) 3.0 (0.0-6.2) 8.5 (5.5-11.5)
* Inestimable.
of the five sequences, two from Amritsar district and one from Gurdaspur district (MH605306, MH605308 and MH598510) were found to be clustered together with C. burnetii sequence from IVRI, Bareilly, India representing the prevalence of common strain in the region and adjoining states. While the other two sequences, one each from Amritsar and Gurdaspur (MH605307 and MH598511) were grouped with rest of the sequences from the different part of the world indicating a common genotype/strain (Fig. 1).
some of their animals. Therefore, we ended up collecting 610 blood, 610 genital swabs and 361 milk samples from 610 of the 1142 bovines in 179 households, recording an overall response rate of 72.5% (179/ 247) at the household level and 53.4% (610/1142) at the animal level (Table S1). Overall prevalence was estimated to be 7.0% (95% CI: 4.7, 9.4). The overall prevalence was 8.7% (95% CI: 5.5, 11.9) in cattle and 4.3% (95% CI: 1.7, 6.9) in buffalo. Detailed prevalence estimates are presented in Table 1. The agreement between ELISA and Trans-PCR was 10.3% (95% CI, 3.2%, 17.4%) and between shedding of C. burnetii in milk (2.8%) and genital secretion (1.1%, for lactating animals) was 14.3% (95% CI, 5.6%, 22.9%). The NCBI GenBank accession numbers for the submitted sequences were obtained (MH598510- MH598511; MH605306- MH605308). The phylogenetic tree based on the alignment of partial IS1111a sequences indicated that all positive samples belonged to C. burnetii (Fig. 1). Out
4. Discussion This study was conducted to estimate the prevalence of Q fever in cattle and buffaloes in Punjab, India. We recorded an overall prevalence of 7.0% (95% CI: 4.7, 9.4) in the bovine population. Earlier studies conducted by Randhawa et al. (1973) reported a 16.2% prevalence of Q fever in cattle and 9.6% in buffaloes. In another study, Sodhi et al.
Fig. 1. Phenogram construction of the partial IS1111a gene sequences of C. burnetii isolates from naturally infected bovines in Punjab (India) along with reference strains. 18
Preventive Veterinary Medicine 166 (2019) 16–20
R. Keshavamurthy, et al.
animal level prevalence. The presence of C. burnetii infected bovine herds and shedding of the organism in the bovine milk indicates that the disease is present in India. Further information about disease epidemiology and mechanisms of spread should be obtained to help develop disease prevention and control programs. Information about the impact of the disease on animal and public health should also be collected to inform disease control development. We recommend that veterinary practitioners include Q fever in the list of differential diagnoses while investigating reproductive problems in dairy cattle in the Punjab state of India.
(1980) investigated Q fever in one dairy farm of Punjab and reported a prevalence of 23.2% and 24.1% in cattle and buffaloes respectively. Our prevalence estimates were considerably lower than the previous studies done in Punjab conducted three decades ago due to different study designs. We tried to collect a representative sample of the bovine population of the state whereas the study by Randhawa et al. (1973) mainly selected animals in contact with test positive human patients and Sodhi et al. (1980) tested samples from just one dairy farm and that too to investigate high rate of abortions. Therefore, our study designs are not comparable. The assessment of the shedding pattern of C. burnetii is essential since the route of shedding is often multiple and discontinuous. We found that milk is the more predominant route of shedding of C. burnetii than by vaginal mucous. This is in agreement with a French study which reported a slightly higher shedding of the organism by milk (24.4%) than by vaginal mucous (19.0%) in cows of Q fever infected herds (Guatteo et al., 2007). We found very poor agreement between ELISA and Trans-PCR and between the two shedding routes. Our study results are comparable with the study conducted by Natale et al. (2012) where low agreements of 0.15 and 0.21 were found between serological and molecular results obtained for vaginal swabs and individual milk respectively. This could be due to differences in the mechanisms of the two tests. Serological techniques only determine the exposure of the animals to C. burnetii and not the presence of the organism whereas PCR is used to study shedding status of the animal by detecting bacterial DNA. Antibodies may be present without current infection resulting in disagreement between the two tests. Likewise, shedding may not be continuous, and therefore, the test may not be able to detect a currently infected animal. Therefore, we recommend that serological response along with bacterial detection are necessary to study the epidemiology of the disease. We report an overall prevalence of 8.7% (95% CI: 5.5, 11.9) in cattle and 4.3% (95% CI: 1.7, 6.9) in buffalo. Many previous studies reported a lower seroprevalence of Q fever in buffalo than in cattle. A seroprevalence of 19.3% (n = 840) in cattle and 11.2% (n = 304) in buffaloes was reported from Egypt (Klemmer et al., 2018). Similarly, a study from Lao People's Democratic Republic reported 2.47% (n = 526) cattle samples to be positive for Q fever, but did not record seropositivity in buffaloes (n = 130) (Douangngeun et al., 2016). On the other hand, a lower overall Q fever prevalence was recorded in cattle (12.78%) than in buffaloes (16.66%) based on animals suffering from reproductive disorders in India (Vaidya et al., 2009). The epidemiology of Q fever in both the species needs to be further investigated in the state. The phylogenetic analysis resulted in two distinct clusters from sequences obtained in the current study. This indicates that there might be more than one C. burnetii strain circulating in Punjab state of India. Previous studies also report genetic diversity among C. burnetii strains in other states of India (Das et al., 2014). This study had several strengths. We made a lot of efforts to randomly select households to ensure representative selection of animals. Secondly, we achieved an overall response rate of 72.5% at the household level and 53.4% at the animal level. Both the response rates were higher than our expected response rate of 40%. Finally, we collected both serum and swab samples and employed both serological and molecular diagnostic tests for prevalence estimation. The study also had many limitations. Due to limited resources, we could only select one village per district, therefore, we were unable to determine the spatial distribution of the disease or calculate district level prevalence estimates with confidence. Secondly, no sampling frame of dairy farmers was available in the state. Therefore, we had to resort to the selection of households from the voter list. We thought it to be a better approach rather than just selecting households haphazardly but some selected households did not keep cattle or buffalo and so they had to be replaced. Further, refusal of the farmer to donate samples from some animals could also have biased our sample for calculation of
Funding This work was financially supported by the Indian Council for Agricultural Research, Government of India, New Delhi through “Outreach Programme for Zoonotic Disease Investigations”. Acknowledgement The authors wish to thank Miss Harpreet Kaur, Mandeep Singh and ST Reddy, School of Public Health & Zoonoses, Guru Angad Dev Veterinary & Animal Sciences University, Ludhiana for their help in sample collection. The authors also wish to thank veterinary officers and para-veterinarians, Department of Animal Husbandry, Dairying and Fisheries, Government of Punjab for facilitating sample collection from the selected herds. The co-operation received from dairy farmers is also thankfully acknowledged. Appendix A Based on the official data, there were 3,187,221 households in rural areas of Punjab. Out of these, 2,327,384 (73%) households keep either cattle or buffalo herds with a total bovine population of 7,587,438 bovines (DAHD, 2014). Given that some farmers keep multiple species, we adjusted the number of individual species holding after taking into account the proportion of mixed farms. As per the official data, there are 105,921,800 total operational livestock holdings in India (NSSO, 2013). However, the sum of individual species livestock holdings was found to be 142,420,600 in India (NSSO, 2013) indicating that 74% of the livestock holdings keep either cattle or buffaloes. Considering these two data, we estimated that 54% (0.73 × 0.74 × 100) of the rural households have either cattle or buffalo after adjusting for mixed farming and number of households keeping cattle/buffalo herds. The average herd size was estimated to be 4.4 (7,587,438 cattle and buffalo/ (3,187,221x0.54)) after considering the cattle and buffalo population, number of rural households and the proportion of households keeping cattle or buffalo in the state. Appendix B. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.prevetmed.2019.03. 003. References Arricau-Bouvery, N., Rodolakis, A., 2005. Is Q fever an emerging or re-emerging zoonosis? Vet. Res. 36 (3), 327–349. https://doi.org/10.1051/vetres:2005010. Asseldonk, M.A.P.M.Van., Prins, J., Bergevoet, R.H.M., 2013. Economic assessment of Q fever in the Netherlands. Prev. Vet. Med. 112 (1-2), 27–34. https://doi.org/10.1016/ j.prevetmed.2013.06.002. Berri, M., Laroucau, K., Rodolakis, A., 2000. The detection of Coxiella burnetii from ovine genital swabs, milk and fecal samples by the use of a single touchdown polymerase chain reaction. Vet. Microbiol. 72, 285–293. https://doi.org/10.1016/S03781135(99)00178-9. Cameron, A., 1999. Survey Toolbox for Livestock Diseases: a Practical Manual and Software Package for Active Surveillance in Developing Countries. Australian Centre for International Agricultural Research, Canberra, Australia.
19
Preventive Veterinary Medicine 166 (2019) 16–20
R. Keshavamurthy, et al.
fever in Egypt: epidemiological survey of Coxiella burnetii specific antibodies in cattle, buffaloes, sheep, goats and camels. PLoS One 13 (2). https://doi.org/10.1371/ journal.pone.0192188. e0192188. Lang, G.H., 1990. Coxiellosis (Q fever) in animals. In: In: Marrie, T.J. (Ed.), Q Fever The Disease. Vol 1. CRC Press, Boca Raton, pp. 23–48. Michelizzi, V.N., Dodson, M.V., Pan, Z., Amaral, M.E.J., Michal, J.J., Mclean, D.J., Womack, J.E., Jiang, Z., 2010. Water buffalo genome science comes of age. Int. J. Biol. Sci. 6 (4), 333–349. Natale, A., Bucci, G., Capello, K., Barberio, A., Tavella, A., Nardelli, S., Marangon, S., Ceglie, L., 2012. Old and new diagnostic approaches for Q fever diagnosis: correlation among serological (CFT, ELISA) and molecular analyses. Comp. Immunol. Microbiol. Infect. Dis. 35 (4), 375–379. https://doi.org/10.1016/j.cimid.2012.03.002. Randhawa, A.S., Dhillon, S.S., Jolley, W.B., 1973. Serologic prevalence of q fever in the state of Punjab. India. Am. J. Epidemiol. 97 (2), 131–134. https://doi.org/10.1093/ oxfordjournals.aje.a121490. Sloan-Gardner, T.S., Massey, P.D., Hutchinson, P., Knope, K., Fearnley, E., 2017. Trends and risk factors for human Q fever in Australia, 1991-2014. Epidemiol. Infect. 145 (4), 787–795. https://doi.org/10.1017/S0950268816002843. Sodhi, S.S., Joshi, D.V., Sharma, D.R., Baxi, K.K., 1980. Seroprevalence of brucellosis and q fever in dairy animals. Zentralblatt. Für. Veterinärmedizin. R. B. 27 (8), 683–685. https://doi.org/10.1111/j.1439-0450.1980.tb01733.x. Vaidya, V.M., Malik, S.V.S., Bhilegaonkar, K.N., Rathore, R.S., Kaur, S., Barbuddhe, S.B., 2010. Prevalence of Q fever in domestic animals with reproductive disorders. Comp. Immunol. Microbiol. Infect. Dis. 33 (4), 307–321. https://doi.org/10.1016/j.cimid. 2008.10.006. Willems, H., Thiele, D., Frölich‐Ritter, R., Krauss, H., 1994. Detection of Coxiella burnetii in cow’s milk using the polymerase chain reaction (PCR). J. Vet. Med. B Infect. Dis. Vet. Public Health 41, 580–587.
Cardinale, E., Esnault, O., Beral, M., Naze, F., Michault, A., 2014. Emergence of Coxiella burnetiiin Ruminants on Reunion Island? Prevalence and Risk Factors. PLoS Negl. Trop. Dis. 8 (8), e3055. https://doi.org/10.1371/journal.pntd.0003055. Cruz, R., Esteves, F., Vasconcelos‐Nóbrega, C., Santos, C., Ferreira, A.S., Mega, C., Coelho, A.C., Vala, H., Mesquita, J.R., 2018. Outbreaks of abortions by Coxiella burnetii in small ruminant flocks and a longitudinal serological approach on archived bulk tank milk suggest Q fever emergence in Central Portugal. Transbound. Emerg. Dis. 65 (4), 972–975. https://doi.org/10.1111/tbed.12913. Das, D.P., Malik, S.V., Das, S., Shoukat, S., Gandham, R.K., Saxena, S., Singh, R., Doijad, S.P., Barbuddhe, S.B., 2014. Isolation of Coxiella burnetii from bovines with history of reproductive disorders in India and phylogenetic inference based on the partial sequencing of IS1111 element. Infect. Genet. Evol. 22, 67–71. https://doi.org/10. 1016/j.meegid.2013.12.017. Douangngeun, B., Theppangna, W., Soukvilay, V., Senaphanh, C., Phithacthep, K., Phomhaksa, S., Yingst, S., Lombardini, E., Hansson, E., Selleck, P.W., Blacksell, S.D., 2016. Seroprevalence of q fever, brucellosis, and Bluetongue in selected provinces in Lao People’s Democratic Republic. Am. J. Trop. Med. Hyg. 95 (3), 558–561. https:// doi.org/10.4269/ajtmh.15-0913. Georgiev, M., Afonso, A., Neubauer, H., Needham, H., Thiéry, R., Rodolakis, A., Roest, H.J., Stärk, K.D., 2013. Q fever in humans and farm animals in four European countries, 1982 to 2010. Euro Surveill. 18 (8) pii=20407. Hilbink, F., Penrose, M., Kovacova, E., 1993. Q fever is absent from New Zealand. Int. J. Epidemiol. 22 (5), 945–949. Honarmand, H., 2012. Q fever: an old but still a poorly understood disease. Interdiscip. Perspect. Infect. Dis. 2012https://doi.org/10.1155/2012/131932. 131932. Kaplan, M.M., Bertagna, P., 1955. The geographical distribution of Q fever. Bull. World Health Organ. 13 (5), 829–860. Klemmer, J., Njeru, J., Emam, A., El-Sayed, A., Moawad, A.A., Henning, K., Elbeskawy, M.A., Sauter-Louis, C., Straubinger, R.K., Neubauer, H., El-Diasty, M.M., 2018. Q
20
Contents lists available at ScienceDirect
Preventive Veterinary Medicine journal homepage: www.elsevier.com/locate/prevetmed
Prevalence of Coxiella burnetii in cattle and buffalo populations in Punjab, India
T
⁎
R. Keshavamurthya, B.B. Singha, , D.G. Kalambhea, R.S. Aulakha, N.K. Dhandb a b
School of Public Health & Zoonoses, Guru Angad Dev Veterinary and Animal Sciences University, Ludhiana, Punjab, 141004, India Sydney School of Veterinary Science, The University of Sydney, Camden, NSW, Australia
A R T I C LE I N FO
A B S T R A C T
Keywords: Buffalo Cattle Coxiella burnetii India Prevalence Punjab Q fever
Q fever is an important zoonosis of animal and public health significance but there is very limited information about its prevalence in the Punjab state of India. The current study was designed to estimate Q fever prevalence in cattle and buffalo populations of the state. We randomly selected 22 villages, one from each of the 22 districts of Punjab. Households in these villages were randomly selected using village voter lists to ensure representative sample collection. Blood, vaginal swab and milk samples were collected from the animals in these enrolled households. Serum samples were screened using Coxiella burnetii specific IgG ELISA whereas milk and genital swab samples were subjected to a Trans-PCR assay. The agreement (Cohan’s Kappa) between shedding of C. burnetii in milk and genital secretions and between ELISA and Trans-PCR was estimated. The selected PCR products were sequenced, and phylogenetic analyses were performed. We collected 610 blood samples, 610 genital swabs and 361 milk samples from 610 bovines (378 cattle and 232 buffaloes) in 179 households. Considering all tests in parallel and after adjusting for clustering, we estimated an overall individual animal prevalence of Q fever of 7.0% (95% CI: 4.7, 9.4). There was a low agreement between shedding of C. burnetii in milk and genital secretion (kappa: 14.3%; 95% CI: 5.6, 22.9) and between ELISA and Trans-PCR (10.3%; 95% CI: 3.2, 17.4%). Phylogenetic analysis confirmed all samples to be of C. burnetii. The results suggest that the disease is present in the state and further epidemiological information should be collected to determine its zoonotic potential and its impact on animal and public health in Punjab, India.
1. Introduction Q fever is a widely distributed zoonotic disease of animal and public health concern. The causative agent Coxiella burnetii is a Gram-negative intracellular bacterium distributed across the globe except in New Zealand and Antarctica (Hilbink et al., 1993; Kaplan and Bertagna, 1955). It is virtually present in all animal kingdom, including birds, arthropods, and rodents but it mainly affects cattle, sheep, goat and humans (Lang, 1990). Recent epidemiological studies have shown an increased prevalence of Q fever in livestock populations in many parts of the world raising public health concerns (Cruz et al., 2018; ArricauBouvery and Rodolakis, 2005; Cardinale et al., 2014). Q fever is a disease of economic significance in farm animals especially in sheep and goats and has a potential of causing considerable losses to the livestock industry (Asseldonk et al., 2013). In cattle, the disease is usually asymptomatic and dramatic clinical manifestations are rare (Arricau-Bouvery and Rodolakis, 2005), but if present, they are usually associated with reproductive complications such as
⁎
abortions, stillbirth, repeat breeding, anoestrus and premature birth (Lang, 1990). Human exposure to infection is mainly associated with the inhalation of contaminated aerosols generated during parturition, slaughter, cleaning and other activities at the farm. Role of cattle as a source for human infection is known in Australia but an association could not be found in Western Europe (Georgiev et al., 2013; SloanGardner et al., 2017). Infection in humans mostly causes acute and selflimiting flu-like illness with symptoms of headache, and atypical pneumonia. However, the chronic form can be severe and fatal without appropriate antibiotic therapy (Honarmand, 2012). Role of water buffalo (Bubalus bubalus) in Q fever epidemiology is not clearly understood but buffaloes are closely related to domesticated cattle and both are members are Bovidae family (Michelizzi et al., 2010). Buffaloes along with cattle are an important source of milk, raised together and are major contributors for the dairy industry in India. Q fever prevalence has been reported in buffaloes from some parts of the world including India and Africa (Klemmer et al., 2018; Vaidya et al., 2010) but the veterinary and economic significance of this
Corresponding author. E-mail address: [email protected] (B.B. Singh).
https://doi.org/10.1016/j.prevetmed.2019.03.003 Received 12 October 2018; Received in revised form 5 March 2019; Accepted 6 March 2019 0167-5877/ © 2019 Elsevier B.V. All rights reserved.
Preventive Veterinary Medicine 166 (2019) 16–20
R. Keshavamurthy, et al.
were replaced with the next individuals in the voter list. Individuals keeping bovines in their households were contacted and requested to participate in the study. If the selected individuals were not willing to participate in the study, they were considered as non-responders but were not replaced with other individuals as we had already considered an expected response rate of 40% in calculating sample size. If the farmer agreed to participate, we strived to collect samples from all of their animals although many farmers did not agree to provide samples from some of their animals. The details of the number of households contacted, responded and the calculated response rate from all villages are presented in Table S1. Blood and genital (vaginal or preputial) swab samples were collected from the selected cattle (crossbred/exotic and indigenous) and buffalo. Additionally, milk samples were collected from lactating females. The samples were transported to the laboratory on the same day and stored at −20 °C until further used. The sera were separated within 24 h in the sterile cryovials and stored at −20 °C until screened. The serum samples were screened for C. burnetii specific antibodies using commercial Q fever indirect ELISA kit containing test plate precoated with Phase I and Phase II antigenic extract (Q fever Monoscreeen Ab ELISA (IgG), Bio-X Diagnostics, Rochefort, Belgium). Milk, vaginal and preputial swabs were subjected to molecular screening to detect Coxiella DNA. Briefly, DNA was extracted using DNeasy Blood and Tissue kit (Qiagen, Maryland, USA). Trans-PCR assay, a PCR targeting the transposase gene of repetitive transposonlike elements of C. burnetii called IS1111 (Willems et al., 1994) was then employed using a published set of primers targeting 687 bp of IS1111a insertion sequence of C. burnetii (Berri et al., 2000). The referral positive DNA control for Coxiella was provided in-kind by the Department of Veterinary Public Health and Epidemiology, Nagpur Veterinary College, MAFSU. In our study, an animal was defined as seropositive if its serum sample was ELISA positive and as molecular positive if either milk or genital swab sample of the animal was Trans-PCR positive. An animal was declared overall positive by using a combination of tests in parallel, i.e. an animal was considered as overall positive if it was seroor molecular- positive. A selection of positive amplified products showing desired band size was purified and sequenced (n = 5) in both the directions (1st base, Apical Scientific, Selangor, Malaysia). The sequences were submitted to NCBI GenBank. The distance-based analysis was conducted using published Coxiella burnetii sequences (Gene Bank accession numbers KT954146.1, KR697576.1, EU000273.1, AB848993.1, KT391020.1, J N966900, JF972643, KP719167.1, MF197399.1, M80806, HM222935, KT381466.1 and EU009657.1). The sequence of Legionella pneumophila (AB594755.1) was used as an outgroup in the analysis. The phylogenetic tree was constructed using the maximum likelihood method based on the Tamura-Nei model by Mega 6.0 software. The bootstrap consensus tree inferred from 1000 replicates was taken to represent the evolutionary history of the taxa analysed. Prevalence estimates and their 95% confidence intervals were calculated separately for sero, molecular and overall prevalence using SAS Surveyfreq procedure after accounting for stratification by district and clustering by herd. Prevalence estimates were classified by species, age, breed and sex. Cohan’s Kappa and its 95% confidence intervals were calculated using SISA website (https://www.quantitativeskills.com/ sisa/statistics/twoby2.htm) to determine the agreement between shedding of C. burnetii in milk and genital secretions as well as to determine the diagnostic agreement between ELISA and Trans-PCR.
disease in buffaloes is not clearly known. Very limited information is available about Q fever prevalence in Punjab – an agriculture state home to 2.4 million dairy cattle and 5.2 million dairy buffaloes (Bubalus bubalus). In 1970s Randhawa et al. (1973) reported a high prevalence of 25.5% of Q fever in human patients suffering from a fever of unknown origin. They conducted further epidemiological investigations which revealed 16.2% prevalence of Q fever in cattle, 9.6% in buffaloes, 6.1% in goats and 3.7% in sheep, indicating the existence of the disease in Punjab. Later, Sodhi et al. (1980) tested serum samples from a farm with a high rate of abortions and found 23.2% and 24.1% prevalence of Q fever in cattle and buffaloes, respectively. However, we are not aware of any other studies conducted in Punjab to estimate the prevalence of Q fever in cattle and buffalo since then. Moreover, none of the studies conducted in Punjab, or in any other part of India for that matter, used probability sampling approaches for prevalence estimation. In Punjab, studies on the prevalence of zoonotic and economically significant diseases in livestock are mainly restricted to brucellosis. The studies on other infectious agents have not been given much consideration. Therefore, losses to livestock sector due to other pathogens are often overlooked and masked. The current study was, therefore, planned to estimate the prevalence of C. burnetii in the bovine populations of Punjab by collecting and testing representative samples from all districts. The results will provide information about the extent of the problem in the state to enable animal health authorities in making appropriate policy decisions in controlling the disease. 2. Methods This study was approved by the Institutional Animal Ethical Committee, Guru Angad Dev Veterinary and Animal Sciences University (GADVASU/2017/IAEC/42/02). The study was conducted using a multi-stage sampling design. We decided to select 22 villages, one per district of the state to ensure the collection of a representative sample from the entire state. Households were then selected using probability proportional to size sampling based on the number of households in a village. The target population of the study was the domestic bovine population in the state. The study population was the domestic bovine population of 22 villages selected to participate in the study. The sample size was calculated to be 414 to estimate Q fever prevalence in the bovine population of Punjab with 95% confidence and 5% precision, assuming a prevalence of 16% based on previous works (Randhawa et al., 1973; Vaidya et al., 2010) and after accounting for a design effect of 2, because animals are clustered within households and villages. Considering a response rate of 40% and an average herd size of 4.4 (see Appendix A), we estimated that we would have to contact 234 farmers to achieve the animal sample size of 414. Accounting for finite population only marginally reduced the number of farmers to 231. Using data about the number of households in 22 villages (5214) and our estimate of the proportion of households keeping livestock (0.54, see Appendix A), we determined that 2816 households in the selected 22 villages would keep livestock (called farmers from here onwards for brevity). To ensure probability proportional to size sampling of farmers, we calculated the proportion of farmers in each village of the total farmer population in the selected 22 villages (i.e. 2816). Multiplying this proportion with the required sample size of this study (i.e. 231 farmers) yielded the number of farmers required to be contacted in each village (see Table S1 for details). We randomly selected one village from each of the 22 districts using the Survey Toolbox program (Cameron, 1999) and then used the voter list of each village for a random selection of farmers. Briefly, sampling interval for systematic random sampling was calculated by dividing the total number of voters in each village by the sample size of each village and was then used in the selection of farmers from the voter list. Individuals not owning the animals or those belonging to the same family
3. Results We visited 247 randomly selected rural households in the selected 22 villages keeping cattle and/or buffaloes. Of these households, 179 smallholders (representing individual households) agreed to participate in the study (Table S1). Although we aimed to select all animals in the selected household, many of the smallholders allowed us to sample only 17
Preventive Veterinary Medicine 166 (2019) 16–20
R. Keshavamurthy, et al.
Table 1 Prevalence of Q fever in the bovine population in Punjab, India based on a study conducted in 2017-18. Serum samples obtained from cattle and buffalo were tested using indirect ELISA. Milk, vaginal and preputial swabs were tested using trans-PCR assay. Categories
Sample size
Number of ELISA positives
Sero-prevalence (95% CI)
Number of PCR positives
Molecular prevalence (95% CI)
Overall positives
Overall prevalence (95% CI)
Overall
610
33
5.4 (3.4-7.4)
13
2.1 (0.9-3.4)
43
7.0 (4.7-9.4)
Species Cattle Buffalo
378 232
24 9
6.3 (3.6-9.1) 3.9 (1.4-6.4)
11 2
2.9 (1.2-4.6) 0.9 (0.0-2.1)
33 10
8.7 (5.5-11.9) 4.3 (1.7-6.9)
Cattle breeds Crossbred/Exotic Indigenous
322 56
21 3
6.5 (3.4-9.6) 5.4 (0.0-11.3)
9 2
2.8 (0.9-4.7) 3.6 (0.0-8.1)
29 4
9.0 (5.4-12.6) 7.1 (0.6-13.6)
Sex Male Female
21 589
0 33
0 (-)* 5.6 (3.6-7.6)
0 13
0 (-)* 2.2 (0.9-3.5)
0 43
0.0 (-)* 7.3 (4.9-9.7)
Age Under 1 year 1 to 3 years Over 3 years
52 101 457
1 2 30
1.9 (0.0-5.6) 2.0 (0.0-4.6) 6.6 (4.1-9.1)
0 1 12
0 (-)* 1.0 (0.0-2.9) 2.6 (1.0-4.3)
01 3 39
1.9 (0.0-5.6) 3.0 (0.0-6.2) 8.5 (5.5-11.5)
* Inestimable.
of the five sequences, two from Amritsar district and one from Gurdaspur district (MH605306, MH605308 and MH598510) were found to be clustered together with C. burnetii sequence from IVRI, Bareilly, India representing the prevalence of common strain in the region and adjoining states. While the other two sequences, one each from Amritsar and Gurdaspur (MH605307 and MH598511) were grouped with rest of the sequences from the different part of the world indicating a common genotype/strain (Fig. 1).
some of their animals. Therefore, we ended up collecting 610 blood, 610 genital swabs and 361 milk samples from 610 of the 1142 bovines in 179 households, recording an overall response rate of 72.5% (179/ 247) at the household level and 53.4% (610/1142) at the animal level (Table S1). Overall prevalence was estimated to be 7.0% (95% CI: 4.7, 9.4). The overall prevalence was 8.7% (95% CI: 5.5, 11.9) in cattle and 4.3% (95% CI: 1.7, 6.9) in buffalo. Detailed prevalence estimates are presented in Table 1. The agreement between ELISA and Trans-PCR was 10.3% (95% CI, 3.2%, 17.4%) and between shedding of C. burnetii in milk (2.8%) and genital secretion (1.1%, for lactating animals) was 14.3% (95% CI, 5.6%, 22.9%). The NCBI GenBank accession numbers for the submitted sequences were obtained (MH598510- MH598511; MH605306- MH605308). The phylogenetic tree based on the alignment of partial IS1111a sequences indicated that all positive samples belonged to C. burnetii (Fig. 1). Out
4. Discussion This study was conducted to estimate the prevalence of Q fever in cattle and buffaloes in Punjab, India. We recorded an overall prevalence of 7.0% (95% CI: 4.7, 9.4) in the bovine population. Earlier studies conducted by Randhawa et al. (1973) reported a 16.2% prevalence of Q fever in cattle and 9.6% in buffaloes. In another study, Sodhi et al.
Fig. 1. Phenogram construction of the partial IS1111a gene sequences of C. burnetii isolates from naturally infected bovines in Punjab (India) along with reference strains. 18
Preventive Veterinary Medicine 166 (2019) 16–20
R. Keshavamurthy, et al.
animal level prevalence. The presence of C. burnetii infected bovine herds and shedding of the organism in the bovine milk indicates that the disease is present in India. Further information about disease epidemiology and mechanisms of spread should be obtained to help develop disease prevention and control programs. Information about the impact of the disease on animal and public health should also be collected to inform disease control development. We recommend that veterinary practitioners include Q fever in the list of differential diagnoses while investigating reproductive problems in dairy cattle in the Punjab state of India.
(1980) investigated Q fever in one dairy farm of Punjab and reported a prevalence of 23.2% and 24.1% in cattle and buffaloes respectively. Our prevalence estimates were considerably lower than the previous studies done in Punjab conducted three decades ago due to different study designs. We tried to collect a representative sample of the bovine population of the state whereas the study by Randhawa et al. (1973) mainly selected animals in contact with test positive human patients and Sodhi et al. (1980) tested samples from just one dairy farm and that too to investigate high rate of abortions. Therefore, our study designs are not comparable. The assessment of the shedding pattern of C. burnetii is essential since the route of shedding is often multiple and discontinuous. We found that milk is the more predominant route of shedding of C. burnetii than by vaginal mucous. This is in agreement with a French study which reported a slightly higher shedding of the organism by milk (24.4%) than by vaginal mucous (19.0%) in cows of Q fever infected herds (Guatteo et al., 2007). We found very poor agreement between ELISA and Trans-PCR and between the two shedding routes. Our study results are comparable with the study conducted by Natale et al. (2012) where low agreements of 0.15 and 0.21 were found between serological and molecular results obtained for vaginal swabs and individual milk respectively. This could be due to differences in the mechanisms of the two tests. Serological techniques only determine the exposure of the animals to C. burnetii and not the presence of the organism whereas PCR is used to study shedding status of the animal by detecting bacterial DNA. Antibodies may be present without current infection resulting in disagreement between the two tests. Likewise, shedding may not be continuous, and therefore, the test may not be able to detect a currently infected animal. Therefore, we recommend that serological response along with bacterial detection are necessary to study the epidemiology of the disease. We report an overall prevalence of 8.7% (95% CI: 5.5, 11.9) in cattle and 4.3% (95% CI: 1.7, 6.9) in buffalo. Many previous studies reported a lower seroprevalence of Q fever in buffalo than in cattle. A seroprevalence of 19.3% (n = 840) in cattle and 11.2% (n = 304) in buffaloes was reported from Egypt (Klemmer et al., 2018). Similarly, a study from Lao People's Democratic Republic reported 2.47% (n = 526) cattle samples to be positive for Q fever, but did not record seropositivity in buffaloes (n = 130) (Douangngeun et al., 2016). On the other hand, a lower overall Q fever prevalence was recorded in cattle (12.78%) than in buffaloes (16.66%) based on animals suffering from reproductive disorders in India (Vaidya et al., 2009). The epidemiology of Q fever in both the species needs to be further investigated in the state. The phylogenetic analysis resulted in two distinct clusters from sequences obtained in the current study. This indicates that there might be more than one C. burnetii strain circulating in Punjab state of India. Previous studies also report genetic diversity among C. burnetii strains in other states of India (Das et al., 2014). This study had several strengths. We made a lot of efforts to randomly select households to ensure representative selection of animals. Secondly, we achieved an overall response rate of 72.5% at the household level and 53.4% at the animal level. Both the response rates were higher than our expected response rate of 40%. Finally, we collected both serum and swab samples and employed both serological and molecular diagnostic tests for prevalence estimation. The study also had many limitations. Due to limited resources, we could only select one village per district, therefore, we were unable to determine the spatial distribution of the disease or calculate district level prevalence estimates with confidence. Secondly, no sampling frame of dairy farmers was available in the state. Therefore, we had to resort to the selection of households from the voter list. We thought it to be a better approach rather than just selecting households haphazardly but some selected households did not keep cattle or buffalo and so they had to be replaced. Further, refusal of the farmer to donate samples from some animals could also have biased our sample for calculation of
Funding This work was financially supported by the Indian Council for Agricultural Research, Government of India, New Delhi through “Outreach Programme for Zoonotic Disease Investigations”. Acknowledgement The authors wish to thank Miss Harpreet Kaur, Mandeep Singh and ST Reddy, School of Public Health & Zoonoses, Guru Angad Dev Veterinary & Animal Sciences University, Ludhiana for their help in sample collection. The authors also wish to thank veterinary officers and para-veterinarians, Department of Animal Husbandry, Dairying and Fisheries, Government of Punjab for facilitating sample collection from the selected herds. The co-operation received from dairy farmers is also thankfully acknowledged. Appendix A Based on the official data, there were 3,187,221 households in rural areas of Punjab. Out of these, 2,327,384 (73%) households keep either cattle or buffalo herds with a total bovine population of 7,587,438 bovines (DAHD, 2014). Given that some farmers keep multiple species, we adjusted the number of individual species holding after taking into account the proportion of mixed farms. As per the official data, there are 105,921,800 total operational livestock holdings in India (NSSO, 2013). However, the sum of individual species livestock holdings was found to be 142,420,600 in India (NSSO, 2013) indicating that 74% of the livestock holdings keep either cattle or buffaloes. Considering these two data, we estimated that 54% (0.73 × 0.74 × 100) of the rural households have either cattle or buffalo after adjusting for mixed farming and number of households keeping cattle/buffalo herds. The average herd size was estimated to be 4.4 (7,587,438 cattle and buffalo/ (3,187,221x0.54)) after considering the cattle and buffalo population, number of rural households and the proportion of households keeping cattle or buffalo in the state. Appendix B. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.prevetmed.2019.03. 003. References Arricau-Bouvery, N., Rodolakis, A., 2005. Is Q fever an emerging or re-emerging zoonosis? Vet. Res. 36 (3), 327–349. https://doi.org/10.1051/vetres:2005010. Asseldonk, M.A.P.M.Van., Prins, J., Bergevoet, R.H.M., 2013. Economic assessment of Q fever in the Netherlands. Prev. Vet. Med. 112 (1-2), 27–34. https://doi.org/10.1016/ j.prevetmed.2013.06.002. Berri, M., Laroucau, K., Rodolakis, A., 2000. The detection of Coxiella burnetii from ovine genital swabs, milk and fecal samples by the use of a single touchdown polymerase chain reaction. Vet. Microbiol. 72, 285–293. https://doi.org/10.1016/S03781135(99)00178-9. Cameron, A., 1999. Survey Toolbox for Livestock Diseases: a Practical Manual and Software Package for Active Surveillance in Developing Countries. Australian Centre for International Agricultural Research, Canberra, Australia.
19
Preventive Veterinary Medicine 166 (2019) 16–20
R. Keshavamurthy, et al.
fever in Egypt: epidemiological survey of Coxiella burnetii specific antibodies in cattle, buffaloes, sheep, goats and camels. PLoS One 13 (2). https://doi.org/10.1371/ journal.pone.0192188. e0192188. Lang, G.H., 1990. Coxiellosis (Q fever) in animals. In: In: Marrie, T.J. (Ed.), Q Fever The Disease. Vol 1. CRC Press, Boca Raton, pp. 23–48. Michelizzi, V.N., Dodson, M.V., Pan, Z., Amaral, M.E.J., Michal, J.J., Mclean, D.J., Womack, J.E., Jiang, Z., 2010. Water buffalo genome science comes of age. Int. J. Biol. Sci. 6 (4), 333–349. Natale, A., Bucci, G., Capello, K., Barberio, A., Tavella, A., Nardelli, S., Marangon, S., Ceglie, L., 2012. Old and new diagnostic approaches for Q fever diagnosis: correlation among serological (CFT, ELISA) and molecular analyses. Comp. Immunol. Microbiol. Infect. Dis. 35 (4), 375–379. https://doi.org/10.1016/j.cimid.2012.03.002. Randhawa, A.S., Dhillon, S.S., Jolley, W.B., 1973. Serologic prevalence of q fever in the state of Punjab. India. Am. J. Epidemiol. 97 (2), 131–134. https://doi.org/10.1093/ oxfordjournals.aje.a121490. Sloan-Gardner, T.S., Massey, P.D., Hutchinson, P., Knope, K., Fearnley, E., 2017. Trends and risk factors for human Q fever in Australia, 1991-2014. Epidemiol. Infect. 145 (4), 787–795. https://doi.org/10.1017/S0950268816002843. Sodhi, S.S., Joshi, D.V., Sharma, D.R., Baxi, K.K., 1980. Seroprevalence of brucellosis and q fever in dairy animals. Zentralblatt. Für. Veterinärmedizin. R. B. 27 (8), 683–685. https://doi.org/10.1111/j.1439-0450.1980.tb01733.x. Vaidya, V.M., Malik, S.V.S., Bhilegaonkar, K.N., Rathore, R.S., Kaur, S., Barbuddhe, S.B., 2010. Prevalence of Q fever in domestic animals with reproductive disorders. Comp. Immunol. Microbiol. Infect. Dis. 33 (4), 307–321. https://doi.org/10.1016/j.cimid. 2008.10.006. Willems, H., Thiele, D., Frölich‐Ritter, R., Krauss, H., 1994. Detection of Coxiella burnetii in cow’s milk using the polymerase chain reaction (PCR). J. Vet. Med. B Infect. Dis. Vet. Public Health 41, 580–587.
Cardinale, E., Esnault, O., Beral, M., Naze, F., Michault, A., 2014. Emergence of Coxiella burnetiiin Ruminants on Reunion Island? Prevalence and Risk Factors. PLoS Negl. Trop. Dis. 8 (8), e3055. https://doi.org/10.1371/journal.pntd.0003055. Cruz, R., Esteves, F., Vasconcelos‐Nóbrega, C., Santos, C., Ferreira, A.S., Mega, C., Coelho, A.C., Vala, H., Mesquita, J.R., 2018. Outbreaks of abortions by Coxiella burnetii in small ruminant flocks and a longitudinal serological approach on archived bulk tank milk suggest Q fever emergence in Central Portugal. Transbound. Emerg. Dis. 65 (4), 972–975. https://doi.org/10.1111/tbed.12913. Das, D.P., Malik, S.V., Das, S., Shoukat, S., Gandham, R.K., Saxena, S., Singh, R., Doijad, S.P., Barbuddhe, S.B., 2014. Isolation of Coxiella burnetii from bovines with history of reproductive disorders in India and phylogenetic inference based on the partial sequencing of IS1111 element. Infect. Genet. Evol. 22, 67–71. https://doi.org/10. 1016/j.meegid.2013.12.017. Douangngeun, B., Theppangna, W., Soukvilay, V., Senaphanh, C., Phithacthep, K., Phomhaksa, S., Yingst, S., Lombardini, E., Hansson, E., Selleck, P.W., Blacksell, S.D., 2016. Seroprevalence of q fever, brucellosis, and Bluetongue in selected provinces in Lao People’s Democratic Republic. Am. J. Trop. Med. Hyg. 95 (3), 558–561. https:// doi.org/10.4269/ajtmh.15-0913. Georgiev, M., Afonso, A., Neubauer, H., Needham, H., Thiéry, R., Rodolakis, A., Roest, H.J., Stärk, K.D., 2013. Q fever in humans and farm animals in four European countries, 1982 to 2010. Euro Surveill. 18 (8) pii=20407. Hilbink, F., Penrose, M., Kovacova, E., 1993. Q fever is absent from New Zealand. Int. J. Epidemiol. 22 (5), 945–949. Honarmand, H., 2012. Q fever: an old but still a poorly understood disease. Interdiscip. Perspect. Infect. Dis. 2012https://doi.org/10.1155/2012/131932. 131932. Kaplan, M.M., Bertagna, P., 1955. The geographical distribution of Q fever. Bull. World Health Organ. 13 (5), 829–860. Klemmer, J., Njeru, J., Emam, A., El-Sayed, A., Moawad, A.A., Henning, K., Elbeskawy, M.A., Sauter-Louis, C., Straubinger, R.K., Neubauer, H., El-Diasty, M.M., 2018. Q
20