Listeria Contamination of Raw Bovine Milk and the Factors Influencing Its Occurrence in Greater Luweero District, Uganda

Aim: To establish the occurrence of Listeria contamination of raw bovine milk and the contributory factors in Greater Luweero. Methods: A cross-sectional study was carried out during 2013; in Luweero, Nakaseke and Nakasongola districts, in Central Uganda. All the 16 milk collecting centers in the study area were sampled and the supplying farms identified for trace back. A total of 100 bulk raw milk samples, each representing a farm, were analysed using the VIDAS® Listeria monocytogenes II (LMO2) enzyme-linked fluorescent immunoassay (ELFA) kit (BioMérieux, Durham NC, USA). The supplying farms were systematically random sampled; and the managers together with those of the milk collecting centers were interviewed to establish the management practices and environmental risk factors associated with Listeria contamination of the milk. Results: High level of Listeria contamination of raw milk from farms (72%); knowledge gap and significant non-adherence to milk trade guidelines (50% and 31.25%; and 88% and 39% among milk collecting center and farm managers, respectively) were encountered. Among the factors influencing occurrence of Listeria in raw milk, the significant ones ( P =.05) included improper hygienic practices; such as poor faecal disposal, improper cleaning of milking utensils and of hands before milking; and non-adherence to Dairy Development Authority (DDA) guidelines; lack of access to dairy extension services; and absence of farm entry restriction and biosecurity measures. Conclusions: There is a high occurrence of Listeria contaminated raw bovine milk from farms in Greater Luweero district. The risk factors that were significantly associated with the contamination can be minimized through sensitization and training of farmers and center managers. Listeriosis is of great public health significance, hence effective inspections to assess compliance to guidelines for quality and safety is recommended. Establishing Listeria monocytogenes carrier status of cattle; and microbial levels in milk will inform on policies for prevention of contamination.


INTRODUCTION
Listeria monocytogenes is a psychrotolerant food-borne zoonotic bacterial pathogen, which survives heating, and drying, thus creating safety hazards to the food industry [1,2]. The organism causes Listeriosis, which is associated with consumption of various contaminated food categories, including dairy products [3]. Particular groups of consumers, such as the elderly, children, pregnant women and those with weakened immunity are more at risk [4]. Among the livestock, sheep seem to be more susceptible to L. monocytogenes than cattle [5]. Other Listeria species exist but are of less significance, although, L. seeligeri and L. innocua have been reported in humans; while L. ivanovii has occasionally been associated with abortions in sheep and cows, or septicemia in sheep and some affect humans too [6]. Infected animals have been reported to persistently excrete the microorganisms in their milk [7,8]. Hence, the sources of Listeria include the production (farm) and processing environments of various foods, especially those of animal origin [9].
Livestock production is one of the major activities in Greater Luweero, which currently comprises of Luweero, Nakaseke and Nakasongola districts, in Uganda. In Luweero, milk is mainly consumed in unpasteurised forms, while the rest is sold to the collecting centers; and eventually to processing plants, often located in other districts. Outbreaks of Listeriosis, more so due to L. monocytogenes, have been reported in developed countries, however limited information exist for most African countries [9][10][11]. Listeriosis among humans could be one of those febrile illnesses often clinically mis-diagnosed as malaria, which is very prevalent and endemic in Uganda. Previously, Listeria species and Listeria monocytogenes contamination of raw milk was reported, at 60% and 13% respectively, in Kampala [12]. The milk sold in Kampala is sourced from various districts of Uganda including Luweero, Nakaseke and Nakasongola, but no information on Listeria contamination exists in the areas of origin. Hence, this study was undertaken to establish the occurrence of Listeria contaminated milk and the contributory factors, in the greater Luweero district.

MATERIALS AND METHODS
A cross-sectional study was carried out during 2013; in Luweero, Nakaseke and Nakasongola districts, in Central Uganda. All the 16 milk collecting centers in the study area were sampled and the supplying farms identified for trace back.
Basing on previous Listeria prevalence in raw milk of 60% in Kampala Uganda [12]; and at 90% level of confidence with a desired absolute precision of 10%; a sample size estimate of 92 was obtained, using the formula n=1.96 2 P exp (1-P exp )/d 2 by Thrushfield [13]. However, a total of 100 bulk milk samples, each representing a farm, proportionately distributed for each district, were obtained. Therefore, the number of farms from each district was 24, 48 and 28 for Luweero, Nakaseke and Nakasongola districts, respectively; however, 11, 64 and 25 milk samples were obtained, respectively.
The supplying farms were systematically random sampled and assessed for the management practices and environmental risk factors associated with the contamination. In addition, both the farmers and the milk collecting center managers were interviewed. The checklist for the interview comprised of anticipated risk factors including sanitation, interaction of different animals on farm, disease control strategies, housing, feeding, milk containers and transportation means, knowledge of and challenges faced in adherence to Dairy Development Authority (DDA) guidelines and regulations. The latter include quality, safety, hygiene practices for milk and dairy products, dairy facilities and premises (www.dda.or.ug/ visited 19/7/2014).
Raw milk samples were aseptically collected on delivery at the collection centers and transported on ice to the laboratory within three hours. The samples were analysed in duplicate using the VIDAS ® Listeria monocytogenes II (LMO2) enzyme-linked fluorescent immunoassay (ELFA) kit (BioMérieux, Durham NC, USA) following the manufacturer's instructions. A test value was generated for each of the samples and compared to internal references (thresholds) allowing interpretation as positive or negative where it was ≥ 0.05 or <0.05 respectively.
The Epi Info statistical package version 3.4.3 and SPSS version 16.0 allowed establishment of the statistical associations using Odds Ratio and Chi-square and significance was defined as a Pvalue <.05, at 95% confidence interval. Table 1 present data from the milk collecting center in relation to the managers' education level, knowledge of the DDA guidelines and practices; and access to the extension services. The level of education of milk collecting center managers ranged from no formal education to tertiary level; with 50% having gone beyond the primary level. Fifty per cent (50%) of the center managers had knowledge and copies of the guidelines issued by the DDA, however, five (62.5%) did not follow them. The challenges faced by center managers in following the DDA guidelines, included use of inappropriate milk containers (37.5%) by the farmers and traders; and adulteration by farmers (18.8%). A statistical association between knowledge of DDA guidelines among center managers and access to dairy extension services existed; with those not accessing the services being about 12 times unlikely to know the guidelines compared to those who did (χ 2 =4.65, P=.04 and Odds Ratio=11.68).

RESULTS
The licensing authorities included the Town Councils and DDA, each licensing 50% of the milk collecting centers. For those centers licensed by the Town Councils, none was inspected compared to those by DDA that had varying frequencies of inspection as indicated in Table 1. There was a relationship between the licensing body and inspection (χ 2 =21.26, P=.00); and of inspection and access to extension services (χ 2 =7.71, P=.01).
Out of the 68 milk samples from farms supplying the DDA licensed collection centers, 47 (69.1%) were contaminated with Listeria whereas of the 32 samples delivered to those centers not licensed by DDA, 25 (78.1%) were contaminated. Farms that supplied non-DDA licensed milk collection centers were 1.13 times more likely to have Listeria contamination. Table 2 presents the occurrence of Listeria in raw milk and the association with various practices on the farms. From the laboratory analysis, 72% of the farms (n=100) had Listeria spp detected in their milk. Majority (64%) of the farm managers/attendants did not go beyond the primary level of education. Majority (69%) of the milking personnel used disinfectants and soap for cleaning their hands before milking; and not water only. Un-boiled water was commonly (65%) used for cleaning utensils and hands and such farms were 3.6 times more likely to have Listeria contaminated milk than those that used disinfectants and soap (χ 2 =5.67, P=.02; Odds Ratio=3.6). Most of the farms (86%) used water from communal dams, followed by boreholes (11%) while some (3%) used piped water. Farms that used water from communal dams were twice more likely to have milk contamination than those that used boreholes (χ 2 =11.66, P=.00 and Odds Ratio=2.07). There was a significant relationship between occurrence of Listeria and milking personnel hygienic practices (hand washing) (χ 2 =61.72, P=.00); and with water source (χ 2 =10.06, P=.01). On majority of farms (97%) cows grazed and fed on pastures within paddocks while 3% reported use of silage, hay and concentrates in addition to pastures. All the three farms that offered silage had Listeria contaminated milk. On majority of the farms (61%) faecal matter was left within the housing environment compared to 39% that had manure .51, P=.00 and Odds Ratio=5.57). Majority of the farms (69%) had no paddocks designated for lactating cows; while 31% had such. On most farms (70%) cows were milked from outside in the kraal, 22% on the paddocks, 7% in milking parlour and 1% in crush. A significant relationship between milking place and Listeria contamination existed (χ 2 =12.75, P=.01 at 95% CI) and milk from animals milked from the kraal was more than twice likely to be contaminated (χ 2 =11.99 and P=.00 and Odds Ratio=2.5). Only 38.5% of the farms reported cleaning the milking area daily, while 39.7% and 21.8% of farms did so after a week and more than two weeks respectively. The frequency of cleaning the milking area was associated with Listeria contamination (χ 2 =4.66 and P=.05 at 95% CI) especially those that cleaned after two weeks (37.2%). Although many of the farms (56%) had no farm entry restriction and biosecurity measures, where they existed, it included fencing off the area with specific entrances, prohibited entry of non-farm workers/strangers and changing of attire and footwear on re-entry; and cleaning of hands and boots with disinfectants before entry. There was a relationship between farm entry restriction/biosecurity measures and prevalence of Listeria where farms without such were 3.14 times more likely to have contaminated milk (χ 2 =6.51, P=.01 and Odds Ratio=3.14). Interaction between wildlife and domestic livestock was reported on 83% of the farms. The wildlife species were reported to originate from forest foci around the farm areas and a gazetted wildlife (rhino) ranch in Nakasongola district.
Considering the type of milk containers and milk handling practices during transportation from farms to the collection centers, majority of the farms (52%) used plastic jerry cans, 42% used Aluminium cans, while 6% used both; and none of the farms had refrigeration facilities. There was a strong relationship between use of plastic jerry cans as milk containers and contamination by Listeria spp (Odds Ratio=2.72, χ 2 =10.2 and P=.00). Farms that used plastic jerry cans were 2.72 times more likely to have Listeria contamination compared to those that used Aluminium cans. Majority of the farms (63%) had no access to dairy extension services. Those farms that accessed dairy extension services, were provided by NGOs (62.2%); DDA (21.6%), area government veterinarians (13.5%); and private farm veterinarians (2.7%). There was a significant relationship between lack of access to extension services and prevalence of Listeria (χ 2 =12.42, P=.00 and Odds Ratio=5.02) as farms that had no access to extension services were five times more likely to have Listeria contaminated raw milk than those that did.
Majority of the managers/farmers (88%) knew the milk trade guidelines as set by DDA while 12% were ignorant. Out of the 88 managers/farmers that knew the guidelines, 69.3% had Listeria contamination in their milk compared to 91.7% of 12 that did not know the guidelines. Among the farm managers who knew DDA guidelines, 39 (44.3%) followed them, 18 (20.5%) partially did while 31 (35.2%) did not. The major challenges faced by the farmers/managers in adhering to DDA guidelines included high costs of white coats/overall (13%); and of aluminium cans, which were also claimed to be very heavy for transportation by bicycles and motorcycles (63%). Another challenge included adulteration of milk by herdsmen, which leads to poor quality of milk that is eventually rejected by milk collecting centers (1%). However, 23% of the managers/farmers did not report any challenges to adhering to the DDA guidelines. There was a relationship between farm non-adherence to DDA guidelines and Listeria contamination (Odds Ratio=6.81, χ 2 =14.62, P=.00), where farms that were not following guidelines were 6.81 times more likely to have contamination of raw milk than those that followed them.
As presented in Table 2, of the 100 farms, 72% had their milk contaminated by Listeria. By district of origin, Luweero had the highest with all the milk from the 11 farms contaminated, followed by Nakaseke with 70% (n=64) and lastly Nakasongola with 64% (n=25). There was an association between prevalence of Listeria contamination and origin of the samples which was more statistically significant for samples that originated from Nakaseke district (χ 2 =4.8 and P=.03).

DISCUSSION
This study reported a prevalence of Listeria contaminated raw milk at 72%, which was higher than 60% in bulked raw milk that was reported earlier in Kampala by Mugampoza et al. [12], 50% in USA by Jackson et al. [14]; but slightly lower than 75% in Portugal by Pintado et al. [15]. The differences in contamination compared to the study by Jackson et al. [14]; could be attributed to the test methods employed; since the latter employed, a combination of preenrichment; VIDAS test and sub-culture on selective agars. VIDAS®LMO2 used in this study was approved as AOAC Official Method 2004.02 for detection of L. monocytogenes in dairy products; and is also an acceptable alternative method for detection of L. monocytogenes in various foods [16]. However, it is recommended to confirm and validate the results by isolation on selective media, that is Palcam and Oxford medium; and confirm the isolates using biochemical tests. Comparable contamination levels were detected although cultural methods were employed by Mugampoza et al. [12] and Pintado et al. [15]. During this study, further analyses to confirm L. monocytogenes were not done. Reports from previous research, indicate that Listeria monocytogenes contamination rates of raw milk at 8.3% in Morocco; 6.5% in Ethiopia and 13% in Kampala, Uganda; 12.5% in USA; 46% in Portugal are lower than that for Listeria spp in general. Hence, it is likely that Listeria monocytogenes contamination of milk obtained from farms from greater Luweero are also lower [12,14,15,17,18]. According to Meyer et al. [19], the VIDAS® LMO2 test is suitable for screening Listeria negative samples and samples strongly positive (<2) for Listeria by this test are positive for L. monocytogenes. Although the test is advocated for screening for Listeria monocytogenes in various foods; false negative samples are encountered due to presence of non-monocytogenes Listeria spp which have been reported to complicate its recovery during selective enrichment [20,21]. Hence, there is need to confirm VIDAS ® LMO2 positive samples using cultural methods for confirming the Listeria species.
Listeria contamination of the raw milk is likely to have originated from the farm due to management practices such as improper faecal disposal systems, since infected animals usually shed the organisms in their faeces [5,22,23]. In addition, the poor hygienic practices such as use of water from communal dams; and lack of use of disinfectants, are likely to have contributed to contamination of the milk handling equipment as reported by some previous researchers [24]. The significant association between milking hygienic practices and prevalence of Listeria monocytogenes has also been reported [25]. In addition, the variation in prevalence of Listeria contamination of milk, especially in USA are most likely to be due to differences in hygienic practices, knowledge of prescribed sanitary, use of treated piped water in production process, knowledge status of guidelines directed towards prevention of contamination; and difference in education levels.
In general, variability in frequency of contamination reflects difference in farm management practices, geographical locations, yearly seasons, sampling or analytical methods [17,26]. In addition, the education status of the farmers and lack of access to dairy extension services by both the farmers and milk collecting centers could have contributed to the high frequency of contamination as reported by previous researchers [17,27]. The absence of farm entry restriction and biosecurity measures; as encountered in this research has been reported to be a risk factor for farm environmental contamination with Listeria monocytogenes elsewhere [28][29][30][31].
Silage has been reported as a major risk factor to Listeriosis; however only 3% of the farms used it to feed the animals; and all milk samples were contaminated with Listeria. This is in agreement with what was reported earlier, that ensiling and stored forage is a risk factor for presence of L. monocytogenes on farms [32][33][34]. However, for this study, it was not possible to establish whether silage feeding was significantly associated with Listeria contamination of the milk because of the rarity of the practice among sampled farms.
Considering the knowledge of DDA guidelines, there was a difference between milk collection center managers and farm managers; and also prevalence of Listeria contaminated raw milk. Lack of knowledge about clean milk production and unclean milk equipment as a factor for Listeria monocytogenes contamination of raw milk has been reported elsewhere [27,35]. Lack of refrigeration facilities coupled by the practice of transportation of milk in non-recommended containers, like plastic jerry cans, are a contributory factor since the latter are difficult to clean, are conducive for the formation of biofilms; and L. monocytogenes has been reported to adhere to polymeric materials, especially at temperatures greater than 30°C and low pH (4 to 7) [36][37][38]. Apart from the costly milking utensils, the low rate of inspection may have contributed to the high level of non-adherence to guidelines among milk collection centers and farms.

CONCLUSION
A high proportion (72%) of raw milk supplied to the collecting centers was contaminated with Listeria; and therefore consumption of raw milk from this area is of health risk. The factors that contributed to the high frequency of Listeria contaminated milk included absence of proper faecal disposal methods on farms; poor unhygienic practices like cleaning of utensils and hands without disinfectants and soap; lack of farm entry restriction and biosecurity measures; use of untreated water from communal dams; absence of established milking areas; infrequent cleaning of milking area; and lack of access to dairy extension services, sensitization and subsequent non-adherence to DDA guidelines.
We therefore recommend that the farmers and milk collecting center managers should be sensitized and trained as far as the DDA guidelines for quality, safety, hygiene practices for milk and dairy products, dairy facilities and premises; and to carry out effective inspections to assess compliance. In addition, confirming the Listeria species contaminating milk from various farms; assessing Listeria monocytogenes carrier status of cattle; the microbial levels in milk; as well as the influence of silage feeding will inform on policies for prevention of contamination. Food processing environs are of particular importance as far as L. monocytogenes contamination is concerned; hence the status of the milk collecting centers needs to be explored.

STATEMENT OF ANIMAL RIGHTS
"All authors hereby declare that "Principles of laboratory animal care" (NIH publication No. 85-23, revised 1985) were followed, as well as specific national laws where applicable. All experiments have been examined and approved by the appropriate ethics committee". However, this research did not involve direct manipulation of the animals.

ETHICAL APPROVAL
The research was approved by the College of Veterinary Medicine, Animal Resources and Biosecurity Higher Degrees Committee and the College Institutional Review Board.