Epidemiology and Economic Impact Assessment of Caprine Brucellosis in Penang State, Malaysia: A Retrospective Analysis

1Department of Clinical Studies, Faculty of Veterinary Medicine, Universiti Malaysia Kelantan, Pengkalan Chepa 16100 Kota Bharu, Kelantan, Malaysia Department of Clinical Studies, Faculty of Veterinary Medicine, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia Department of Preclinical Studies, Faculty of Veterinary Medicine, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia


Introduction
Brucella melitensis organism remains as versatile as ever and continues to be one of the most fascinating microorganisms with multiple mysteries deeply buried in its genome (O'callaghan and Whatmore, 2011). B. melitensis, a Gram negative bacterial pathogen that presents varied clinical diseases in livestock, causes abortions during the trimester often followed by retained placenta, weak offspring and metritis in sheep, goats and cattle, orchitis, epididymitis and polyarthritis in rams, fever and focal forms with osteoarticular and genitourinary forms 37 in humans and multi-systematic infections in camels, dogs and sporadic cases in horses, pigs and other wildlife species. Infections of B. melitensis are associated with a wide range of diseases, termed 'brucellosis'. In some situations, B. melitensis also complicates pathogenesis of other diseases as secondary pathogen.
Infections related to B. melitensis occur worldwide and thus a significant contributor to Disability-Adjusted Life Years (DALYS) and loss associated with zoonotic disease (Osman et al., 2016). This includes Mediterranean countries (south and east Europe), north and east Africa, central and south America, Asia and the Middle East (Corbel, 2006). The disease has a high global index as an animal health constraint to poor farmers (Seleem et al., 2010;Samadi et al., 2011). Subsequently, several notable achievements in the realm of the organism's characteristics, virulence, pathogenicity, diagnosis and development of vaccines have been reported (Doganay and Aygen, 2003;Corbel, 2006;Osman et al., 2016). Brucella melitensis is generally considered to involve infection through ingestion, inhalation and through direct contact with broken skin and often severely terminates in septicaemia with isolation of the organism from multiple organs (Grilló et al., 2012;Díaz et al., 2013;Osman et al., 2017). Livestock farmers, veterinarians and laboratory workers are at risk of being affected by this organism due to their constant and close contact with animals and animal related materials (Samadi et al., 2011). In Malaysia, brucellosis has been occurring in the livestock population for many decades although at a relatively low prevalence compared to other countries in Asia have been reported (Bamaiyi et al., 2014). In Penang, B. melitensis surveillance for eradication purpose in still ongoing. However, there are no published reports to date on B. melitensis in goat farms in Penang and the works on the epidemiological aspects are very limited in Malaysia. Therefore, in this study, we describe the trends and pattern of caprine brucellosis between 2012 and 2014 from a retrospective analysis of data collected from a nationwide brucellosis active surveillance programme. We believe that information from this study provides insight on the epidemiology of caprine brucellosis in Penang, Malaysia and assist the authorities in improving their disease-control strategies. Malaysia (4.1936° N, 103.7249° E) is located in Southeast Asia and comprises East Malaysia (peninsular Malaysia) and West Malaysia (Sabah and Sarawak on Borneo Island). The two regions are separated by the South China Sea. Penang is located on the northwest coast of Peninsular Malaysia with an area of 651.784 square kilometers bordering Kedah to the north and the east and Perak to the south. Penang has an average rainfall of 340 mm of rain and experiences hot and humid weather throughout the year with two monsoon seasons; the north-east monsoon from November to March and the south-east monsoon from May to September (http://www.met.gov.my). Malaysia has a relatively small goat population size with estimated number of 482,280 heads (Malaysia: Livestock Population, 2009-2013).

Data Sources
Brucellosis serosurveillance activities were performed regularly by the state veterinary departments in Malaysia, as described in the Protokol Veterinar Malaysia Penyakit Brucella. The program allows for serological screening of goats at least twice a year. Once confirmed, all seroreactors must be culled in the government abattoir. Slaughter under the supervision of the veterinary officer is required to ensure compensation of culled goats. The livestock sampling and serological testing for the serosurveilance programme was performed via the state's veterinary departments and its regional veterinary laboratories that are located at Penang, Malaysia. Accordingly, serum samples from goats were tested for evidence of Brucella antibodies using the Rose Bengal Plate Test (RBPT) and the Complement Fixation Test (CFT), using the protocols and guidelines described previously (Jacobson, 2004). The confirmatory diagnosis for Brucella antibodies using the CFT was performed at the Veterinary Research Institute (VRI), Ipoh. The study was approved to be conducted by the Department of Veterinary Services, Putrajaya Malaysia.
We obtained data generated from the serological testing for ovine brucellosis from the Epidemiology and Surveillance Unit at the Department of Veterinary Services (DVS), Putrajaya and the database at VRI, Ipoh, for years 2012 to 2014. These serosurveillance data have not been extensively analysed in the past. The data were compared and collated to improve their integrity. The data were thoroughly checked for accuracy in entry, coding and typing errors and repeated entry of a farm in the course of one year was eliminated to ensure that a herd or farm was not over represented in a given year of study. The information obtained from the data includes farm names and addresses, date of sampling, locations, breed, age range, number of animals tested and the number of animals within the tested herd.

Data Analysis
The data were managed and stored in a Microsoft Excel® (Microsoft Corporation) spreadsheet and frequency tables were used to calculate prevalence based on districts within Penang, year, herd, animal and breed. Seroprevalences rates over the 2 years were determined as the number of seropositive goats divided by the total number of goats sampled and confidence intervals were calculated at a 95% level. The differences between/among proportions were tested using Chisquare and univariate logistic regression statistics. All statistical analyses were performed using SPSS (version 16, Chicago, IL) at a significance level of α = 0.05 and the figures were plotted using Sigma Plot (version 11.0)
For the year 2012, the highest percentage of goat positive for B. melitensis infection was in September with 60 cases (Fig. 2). The goats positive for B. melitensis infection showed variation in number of cases from February until August with 9 to 2 cases. There were increased cases of positive B. melitensis infection from August to September rising from 9 to 60 cases. The number of goats' positive for B. melitensis infection decreased from October to December from 2 to zero (Fig. 2).
For the year 2013, the highest number of goat positive for B. melitensis infection was in August with 6 cases (Fig. 3). The cases of goats positive with B. melitensis infection fluctuated from January until September from 6 to 1 cases. There were slightly increase in prevalence from July to August from 1 to 6 positive of B. melitensis infection. The percentage of goats' positive for B. melitensis infection decreased from September to December from 2 to zero case (Fig. 3).
For the year 2014, the highest number of goat positive for B. melitensis infection was in March with 197 cases (Fig. 4). The cases of goats positive with B. melitensis infection fluctuated from January until October which ranged from 197 cases to 2 cases. There was a major increase from June until July from 6 to 48 cases positive for B. melitensis infection. It later decreased by 4 cases in the month of August. The number of goats positive for B. melitensis infection decreased between October and December from 6 cases to zero (Fig. 4).

Economic Losses to Farmers Due to B. melitensis Infection
Economic loss to the goat farmers due to B. melitensis infection in Penang from 2012-2014 was calculated based on market value of slaughter and culled positive animal. The Malaysian government provides compensation for the goat farmer with a standard compensation value of RM5.60 per kg which only accounts for 23% of the value of the live weight of the animal.
In the year 2012, 97 goats were positive for B. melitensis infection at which the market value was approximately RM 65,960 but the compensation given by the DVS Penang was RM 29,550. Therefore, the economic losses in 2012 due to B. melitensis infection were RM 36,410 (Table 1). Similarly, in the year of 2013, 14 goats were positive for B. melitensis infection at which the market value was approximately RM 9,520 but the compensation given by the DVS Penang, Malaysia was RM 464. Therefore, the economic losses in 2013 due to B. melitensis infection were RM 9,056 ( Table 1). The highest annual economic losses for the study period was observed in 2014. In the later year, 356 goats were positive for B. melitensis infection at which the market value was approximately RM 242,080 but the compensation given by the DVS Penang was RM 45,128 making the total of economic losses due to B. melitensis infection to a total of RM 196,952 (Table 1)   The total compensation given by government due to B. melitensis infection cases between 2012 and 2014 was estimated to RM 75,142 as the average market value of live goat is RM17 per kg within that time frame. Assuming that the average live weights of positive goats are 40 kg; therefore, the total current value should be RM 317,560. The calculated losses of the farmer due to B. melitensis infection would be approximately RM 242,418 (Table 1). The total economic losses were about 76.34%, however the government only compensated for 23.66% of the total value of the animals. This implies that the government compensated only 23.66% of the actual value of the animals culled in Penang during the study period.

Discussion
The results showed that the prevalence of B. melitensis infection in Penang from year 2012 to 2014 was 3.4% which showed that the prevalence was considerably high when compared to the national prevalence which was 0.8%. According to a recent research paper published in Malaysia, the prevalence of B. melitensis infection in Penang was at 2.01%. The prevalence increased within one year by 1.39% from 2013 to 2014. The observed increase in the prevalence could be attributed to mass importation of goats from endemic countries (Zamri-Saad and Kamarudin, 2016). This is the major cause for the persistent infection of B. melitensis infection in Penang state, Malaysia. The increase in the prevalence of B. melitensis in this state was consistent with the increasing number of human brucellosis cases as a result of consumption of raw goats milk (Leong et al., 2015).
Recently, the number of small holders had dramatically increased over the last years as well as the importation of goats from neighboring countries (Bamaiyi et al., 2014). The increased number of small holders, however, is believed to be associated with the recent initiatives in which the Malaysian government promotes small holders to participate national economy reform. Moreover, the increased seroprevalence observed in Boer breed rather than indigenous species such as Katjang, Jamnapari and Sannen is strongly related to the importation of animals from other countries (Zamri-Saad and Kamarudin, 2016). Several factors influence the observed incidence of the disease and include the failure of quarantine management upon arrival of animals and also failure of herd health management of clinical and sub-clinical cases. The mass importation of Boer goats from endemic counties throughout the year became a major issue in Malaysia. Most of the farmers were rearing more Boer goats for meat purposes than any other breed. In addition, farmers preferred natural breeding rather than practicing Artificial Insemination (AI) as means of breeding which can contribute to the increasing number of B. melitensis cases in Malaysia.
The proportion of sero-reactor animals remained relatively high (95.5%) in Seberang Perai Utara (SPU). This phenomena reflects the geographical location as well as occupational factors where residents in this area are extensively involved in agricultural and livestock based industries. Moreover, large scale farms, sampled from this area, might contributed to the increased number of positive cases of B. melitensis infection. The number of goat's population was two fold higher in SPU as compared to the other districts. We believed that the smaller numbers of farms sampled may not have given the correct picture as compared to other districts in Penang and may have contributed to the difference observed. It was also possible that the high seroprevalence of SPU was caused by its geographical area as it is mostly covered with agriculture land and also located at the border of Kedah DarulAman and Thailand which are areas known to be endemic for the disease. At the same time, we believed that the reallocation of time and resources by the authorities to control other pandemics and re-emerging infections, notably the multiple outbreaks of FMD (Edwards, 2004;Wongsathapornchai et al., 2008), bird flu (H5N1) (Morris et al., 2005) and swine flu (H1N1) (Goodwin et al., 2011) led to asubstantial reduction in of nationwide brucellosis surveillance and control program. Previous studies have also reported that location, region or area have a significant correlation with brucellosis seropositivity, which, according to the authors, is attributable to management practices and other agroecological factors (Megersa et al., 2011). It is also possible that the difference was due to other confounding variables unaccounted for in this study.
Larger herd farmers are also more likely to be actively purchasing or introducing new animals into their herds thereby increasing the risk of acquiring Brucella 43 melitensis seropositive animals. These infected goats could be carriers of the bacteria for lifetime and while shedding the organism, they may show little or no sign of infection (Al-Majali, 2005;Seleem et al., 2010).
The association of months of sampling with seropositivity to Brucella infection is in agreement with other previously published reports in which rainfall or season was reported to play a major role in the epidemiology of the disease (Muñoz et al., 2010).
The increase in number of positive cases of B. melitensis infection in March and September could be ascribed due to of multiple factors that include socioeconomic, seasonal factors among other environmental conditions. Seasonal variation in the epidemiology of infectious diseases are common phenomenon in both temperate and tropical climates (Grassly, 2006). However, the mechanism of the change is poorly understood (Osman et al., 2016) and has been linked to the interaction of several intrinsic and extrinsic factors (Pascual, 2005;Altizer et al., 2006;Grassly, 2006). In the epidemiology of B. melitensis infection, seasonal factors have been reported to be associated with human brucellosis which, in most cases, coincide with the period of parturition among farm animals and, hence, increased exposure to farmers when attending to animals and consuming their milk (Corbel, 2006). This was also evidenced by recent outbreaks of human Brucellosis following consuming of raw goats' milk in Penang, Malaysia (Leong et al., 2015).
In the analysis of economic loses, the observed pattern varied also over the study period with highest loses recorded in 2014 followed by 2012 and 2013, respectively. The total of financial loss was estimated to be around RM 242,418 during the study period.
We believe that the high economic losses reported in our study would provide awareness to the different sectors regarding this zoonotic pathogen, hence extreme measure could also be exercised. The losses to the farmer are usually from culling of positive animals with or without minimum compensation. However, we strongly believe that the figures of economic losses associated with Brucella infection are always underestimated due to the fact that most of the cases remains unreported. This was evidenced by the persistence nature of the infection and thus invalidating to arrive a definitive conclusion in terms of the economic losses due to this devastating disease.

Conclusion
In conclusion, the prevalence of goat brucellosis is significanty high among goats in Penang state, Malaysia. Breed, locality and seasonality are considered as potential risk factors for contracting brucellosis. This study would be expected to reduce the impact of the disease and its economic consequences on the livelihood of small holders and ultimately avoid a disease spread in human during a possible brucellosis outbreak. Further molecular characterization studies such as, complete bacterial genome sequencing and subsequent phylogeny would be necessary to determine the genetic lineages of Brucella strains circulating in infected hosts in Malaysia.