Livestock Research for Rural Development 31 (1) 2019 Guide for preparation of papers LRRD Newsletter

Citation of this paper

A novel pH-based pen-side test for detection of sub-clinical mastitis: validation in cattle and camels, Kenya

P N Ndirangu, E O Mungube, M W Maichomo, P K Nyongesa1, D Wamae, J O On’gala1, N W Gicheru2, G Keya3, H O Wesonga and D Siamba4

Veterinary Research Institute, Muguga North, P.O. Box 32-00902, Kikuyu
Erick.Mungube@kalro.org
1 Masinde Muliro University of Science and Technology, Department of Biological Sciences, P O Box 190-50100, Kakamega
2 Egerton University, Department of biochemistry and molecular biology, P O Box 536-20115, Njoro
3 Kenya Agricultural and Livestock Research Organization (KALRO), Headquarters, ASAL APRP Project Coordination Unit, P O Box 57811-00200, Nairobi
4 Kibabii University, Department of Agriculture & Veterinary Sciences, P.O. Box 1699-50200, Bungoma

Abstract

Mastitis is economically the most important disease for dairy producers. Existing diagnostic methods are complicated with limited application at farm level. The current study aimed at developing and validating a rapid, user friendly and cost-effective farm based mastitis diagnostic test for cattle and camels. During the development of the pH based test kit litmus paper concept was used in titrating 8 chemical pH-indicators before the best two indicators were selected. The selected indicators were impregnated on to absorbent paper to produce strips which were tested on-station for their ability to detect cows with sub-clinical mastitis using 25 lactating cows. After on-station testing, further refinement was done on the test kit before validating it on-farm in Laikipia North and West sub-counties and Loitokitok Sub-County of Kajiado, where 99 lactating camels and 136 lactating cattle were used. Udder quarter milk samples for each cow and camel were tested separately and sub-clinical mastitis (SCM) results of the pH-based test were compared to those of California Mastitis Tests (CMT) for agreement between them. On-station results showed that the pH based kit had SCM prevalence of 43.8 % which was significantly (p<0.05) higher than that for CMT (26.0%). The agreement for the on-station testing phase was 82.3% (Kappa = 0.62, p<0.0008). Overall, comparison of the results for the pH based kit during the validation phase showed remarkable agreement between the two tests. In camels, the pH based test detected SCM infection in 10.1% of the quarters examined while CMT detected 12.1% in the same quarters with an agreement of about 98% (Kappa = 0.96, p< 0.001). In cattle, the pH-based test detected 55% SCM cases of all quarters examined while CMT detected 56% SCM cases in the same quarters with an agreement of 99% (Kappa = 0.96, p< 0.001). The pH-based test showed high positive predictive value of 84.2% in camels and 99.5% among cattle. Negative predictive values were equally high at 99.5% among both camels and cattle tested on-farm. The results clearly demonstrated that the pH based pen-side test is reliable in detecting SCM in camels and cattle at farm level. Registration and commercialization of this kit needs to be fast tracked since its use and application of appropriate mastitis control measures will lower prevalence of mastitis in the study sites.

Key words: agreement, California Mastitis Test, pH based test kit, kappa, validation


Introduction

Kenya’s dairy sub sector is the largest agricultural sub-sector contributing about 8% of Gross Domestic Product (GDP) with an annual milk production of 3.43 billion litres, which is 3% of the 18% global production by Sub Saharan Africa (Odero-Waitituh 2017). It is one of the agricultural sub-sectors experiencing high growth, estimated at 3 to 4 % annually. In the country, milk based enterprises support over 1.8 million smallholder households engaged in dairy production. Despite this, several factors among them mastitis are constraining dairy productivity. Worldwide, mastitis is a disease of enormous economic importance but it is of particular concern in developing countries where milk and milk products are scarce (Abebe et al 2016). It causes great economic losses if not detected and treated promptly (Iyer et al 2014). In Kenya, the disease is considered a major constraint to milk production not only in dairy cattle but also in camels where SCM was reported to cause a milk production loss of about 200 ml per mastitis-infected quarter per day (Maichomo et al 2014). Mastitis not only reduces productivity as a result of health implications but also lowers the quality of both milk and milk products (Kasozi et al 2014). The sub-clinical form of the disease characterized by absence of visible changes in milk appearance is the most common and causes enormous losses through reduced milk production (Mungube et al 2005; Oliveira et al 2011).

Effective mastitis control is premised on early detection and treatment thus reducing losses attributed to mastitis such as reduced milk production, treatment costs and premature culling of cows (NMC 2013). Detection is made by using methods that examine milk quality, mammary gland status and pathogen involved. At farm level the strip cup identifies clinical mastitis by showing changes in milk colour and consistency. However, the strip cup does not detect sub-clinical cases due to the absence of detectable physical changes in milk. Screening of udders for SCM is done by the California Mastitis Test (CMT) which is a rapid pen-side test that indirectly estimates somatic cell counts (SCC) on the farm. It gives a semi-quantitative SCC measure by forming a stringy mass (gel) from the reagent (3% sodium lauryl sulphate) and DNA out of disrupted cells (Leslie et al 2002; Schalm et al 1971). Somatic cell counts are an important indicator of intra-mammary infections (Schukken et al 2003; Quinn et al 1999). However, it is not recommended for use to detect intra-mammary infections in cows earlier than four days post-calving (Dingwell et al 2003).

The most sensitive test for diagnosing mastitis remains the detection of causative agents through culture (Viguier et al 2009). However, culturing is costly, laborious, time consuming, requires a laboratory and the results are realized after several days. Furthermore, culture only detects viable cells and thus the clinical relevance of negative culture results from samples positive by CMT requires further research (Koskinen et al 2010). It is thus clear that the existing mastitis diagnostic tests although accurate and reliable are expensive, technical and suitable for research purposes. In most developing countries including Kenya, the role played by farmers in controlling mastitis is minimal since they cannot timely detect mastitic animals in their herds. They rely on assistance from the technical experts who are not only unavailable but sometimes charge expensively for mastitis screening services (Ndirangu et al 2013). In order to actively involve dairy farmers in the effective mastitis control programmes, there is need to avail a cheap and easy to use pen-side kit for farmers to use on their dairy herds. This justified the development and validation of a pH-based test that is easy-to-use, cheap and farmer friendly to help in the timely detection of mastitis infected animals in dairy herds. The pH-based test is made of cheap and readily available chemical pH-indicators, impregnated onto available absorbent papers and results easy to read since they are based on colour-change that corresponds with changes in milk pH. Further the test is suitable for ASALs where water is scarce since the test does not require cleaning of the puddle with water after each test. This paper reports results of validation of the pH-based pen-side mastitis test in the detection of SCM in lactating cattle and camels alongside the CMT, an existing conventional on-farm screening method.


Materials and Methods

Study area

The study on cattle was carried out in Naivasha and Kajiado, while that on camels was undertaken in Laikipia, Kenya. In Naivasha, the study was carried out at KALRO, DRI-Naivasha situated along longitude 36o 42’ East and latitude 00 o7’ South. Naivasha is a sub-county of Nakuru County. The area receives bimodal rainfall with long rains occurring from March to May while the short rains are in October and November. It receives an average of 650mm of rain per year. In Kajiado County, the study was carried out in Loitokitok Sub-County, situated between longitudes 36º 5’ and 37º 5’ East, and between latitudes 1 º0’ and 3 º0’ South. Out of the 5 wards in Loitokitok (Kuku, Rombo, Imbirikani/Eselenkei, Entonet/Lenkisim and Kimana), Kuku, Entonet and Kimana were randomly selected for the study (Figure 1). Loitokitok has a bimodal rainfall pattern of between 900 and1250mm annually, where short rains are received between October and December, while short rains fall between March and May (MoALF 2015).

The study in camels was undertaken in Laikipia North and Laikipia West sub-counties situated along longitudes 36o45’ East and between latitudes 00o25’ North on the leeward sides of Mt. Kenya and Aberdare Ranges. Laikipia receives bimodal rainfall of between 400mm and 750mm annually. The long rains occur from March to May while the short rains are in October and November (Laikipia County report 2016).

Figure 1. Map of Kenya showing the study sites
Study design

The study was carried in three phases namely; development and production of the pH-based mastitis test strips in the laboratory, on-station trial of the test kit and field validation of the test kit. The production of the pH-based test kit was undertaken at Maside Muliro University of Science and Technology- Kakamega, Kenya in 2015 before on-station testing on the Sahiwal and dairy herds at the Kenya Agricultural and Livestock Research Organization (KALRO), Dairy Research Institute (DRI), Naivasha, Kenya in 2016.

Development of the pH-based sub-clinical mastitis test detection kit

Prior to commencing the work, extensive literature search was undertaken to identify possible chemical pH-indicators. They were then screened so as to establish indicators with pH ranges that can be used in milk testing for mastitis. A total 8 chemical pH indicators including bromothymol blue, bromocresol purple, bromophenol blue, neutral red, phenol red, phenolphthalein, brilliant blue and rosolic acid were screened. The screening process involved titration of the identified suitable indicators while noting the colour change at different pH levels.

Briefly, the chemical indicators were first dissolved in different diluents (ethyl alcohol, distilled water and sodium hydroxide) to prepare solutions for titration. Titration was carried out using a calibrated pH meter (112211-02 series, HANNA® instruments, Romania-Europe) where hydrochloric acid was added to increase acidity and sodium hydroxide to lower acidity/increase alkalinity of the indicators as and when necessary. The end-point of the titration process was a colour change near the pH transition point for mastitis (≈6.8). This process led to the selection of two pH indicators of choice (not disclosed since they are currently undergoing patenting). The selected mixture of indicators was impregnated on absorbent paper (Whatman® filter paper grade 3, GE Health Care, America) by immersion and baking at 22oC and 27 oC (representative of room storage temperatures and ambient temperatures in the field) for 12 hours. The absorbent paper with indicator was then cut into strips of 2cm wide by 10cm long ready for use during the on-station testing for accuracy to detect sub-clinical mastitis.

Description of the novel pH-based pen-side sub-clinical mastitis (SCM) test

The pH-based pen-side SCM detection test, which was developed by the current study, works on the basis of changes in colour attributed to changes in pH (hydrogen ion concentration) in milk. The test is thus meant to detect variations in the normal pH range of cattle milk which is 6.5 to 6.8 while that of camel milk is 6.5 to 6.7. Mastitic milk is alkaline and thus causes an increase in pH following increased leakage of various ions and salts into milk as a result of increased permeability of vascular membranes following inflammatory reactions (Hussain et al 2012). The test strip is made up of two pH-chemical colour indicators impregnated onto absorbent paper. This strip, changes colour after approximately 5 seconds, from yellow to blue with increasing pH. Results of screened cow udders are interpreted on the basis of rising alkalinity. Thus, in those with SCM there is a colour change to blue, indicating that the udder quarter(s) screened is positive for SCM or no colour change in those udder quarters which are negative for SCM.

On-station testing of the pH-based sub-clinical mastitis test detection kit

After the pH-based mastitis test kit was ready, test strips were produced and used during the on-station trial on dairy herds at KALRO, DRI-Naivasha, Kenya. A total of 25 lactating cattle were randomly selected for the on-station study to validate the pH-based mastitis test in the dairy setting. The study animals used were 11 Sahiwals, 9 crosses and 5 Friesians. With regard to stage of lactation, 10 were at mid lactation (0-3 months), 8 at late lactation (>9 months) and 7 at early lactation (>3-9 months). The pH kit was tested alongside CMT, a conventional test for comparison purposes. The interpretation of the CMT results was as described by Quinn et al. (1999). CMT was used as a comparative test because it is rapid and can be used ‘on-site’ or in the laboratory (Viguier et al 2009).

Briefly, SCM was diagnosed using CMT and the pH-based pen-side mastitis test. The two tests were conducted in parallel on the same quarter milk sample before comparing the results. Just before collecting milk, milker’s hands and teats were washed with warm clean water and teats swabbed with cotton dipped in 70% ethyl alcohol. The first few squirts of milk from each quarter were discarded before obtaining a few drops (about 2 ml) of milk from each teat into a plastic CMT-puddle that had 4 shallow wells corresponding to the four udder quarters. Thereafter, individual pH-based mastitis detection strips were dipped into shallow well of CMT-puddle containing milk sample from each quarter and any colour change recorded after approximately 5 seconds. A positive result was recorded when the strip changed from yellow to blue while negative results were recorded when there was no colour change.

In order to compare the ability for the pH strip to correctly and accurately detect SCM, CMT testing was performed on the same quarter milk samples by adding an equal amount of CMT reagent and swirling for 10 seconds. This was observed for gel formation. Each milk sample in the CMT puddle was scored as 0 (negative), trace, 1, 2 and 3 (all positive) depending on the consistency of agglutination/gel formed where the score increased with the viscosity of the gel (Quinn et al 1999). An animal or udder quarter was diagnosed as having SCM when it was found to be positive to either or both of the two tests.

On-farm validation of the pH based sub-clinical mastitis kit

After the on-station testing of the pH-based kit, refinements were done and mastitis detection strips produced for use during validation under field conditions. These refinements included overcoming the problem of the indicator solution bleeding out of the absorbent paper by adding amine salts as well as changing the type of absorbent paper from grade 1 to grade 3. The validation was designed as a cross-sectional study and was carried out on selected herds of dairy cattle and camels. Validation study on cattle was carried out in Loitokitok Sub-county of Kajiado County in November, 2016, while that on camel was carried out in Laikipia North and West sub-counties of Laikipia County in December 2016.

In Loitokitok sub-County, 52 dairy herds were randomly selected from a sampling frame of 170 active members of Loitokitok Dairy Farmers Cooperative. This was inclusive of four manyattas (Maasai traditional temporary livestock structures) which were conveniently included depending on accessibility and willingness of the owners to participate in the study. Each selected herd acted as a cluster where all lactating cattle were examined for mastitis. In total, 136 cattle were examined for clinical and sub-clinical forms of mastitis. In this area, the breeds were Friesians (78), aryshire (24), crosses (19), indigenous cattle (13), jersey (1) and Guernsey (1). The cows were at different stages of lactation where 72, 35 and 29 being at late, early and mid stages of lactation.

In Laikipia, the camel herds were purposively selected on the basis of accessibility and willingness of camel owners to participate in the study. A total of 99 lactating camels from six herds (three camel ranches and three manyattas) were used for purposes of validating the kit. All the 6 camel herds were treated as clusters from where all available lactating camels were selected and examined for mastitis. The 99 camels consisted of mainly Somali breed (82), followed by Turkana (13) and crossbreeds (4). With regard to their stages of lactation 65, 19 and 15 were in early, late and mid lactation stages respectively.

Diagnosis of SCM in the dairy cattle in Kajiado County and camel herds in Laikipia County was done using the pH-based test kit and CMT as described for the on-station testing.

Diagnosis of clinical mastitis

In both the on-station and on-farm validation phases, study animals were also examined for clinical mastitis through physical and clinical examination of the udders. In addition, visual observation of milk from udders was also used. The udders of the selected study animals were palpated for swellings and other signs of inflammation. Milk was visually examined for changes in colour and consistency. An animal was diagnosed with clinical mastitis if the udder was swollen, hardened and painful to touch or when milk extracted from that udder was watery, had clots, flakes and milk colour had changed in appearance to reddish or brownish (Radostits et al 2007). Further physical examination of the udders enabled detection of blind teats, udder deformities and other teat defects. Blind teats were characterized by presence of lost or blocked teats and/or atrophied non-functional udder quarters.

Data management and statistical analysis

Data for both the pH-based mastitis strip and CMT tests were entered at two levels namely the animal and udder quarter levels and stored in MS-Excel 2007. The data were then exported to R-program where descriptive statistics and associations between pH-based test and CMT test were performed through chi-square test at 95% confidence interval (CI) and p set at 0.05 (5%). A Kappa Test (Viera and Garnett 2005) for agreement between pH-based mastitis test and CMT results in both camels and cattle was calculated. Agreement was calculated as the proportion (%) of milk samples where the two tests gave the same results. Positive predictive value, which is the probability that the disease is present when the test is positive, was calculated as the number of milk samples that were positive using the new pH-based test divided by the total number of samples that tested positive to both tests. Negative predictive value (NPV), which is the probability that the disease is not present when the test is negative, was calculated as number of milk samples that tested negative on pH-based test divided by total number of samples that tested negative on both tests. Prevalence of sub-clinical mastitis was calculated as number of animals or udder quarters testing positive on either or both of the two tests divided by total number of lactating animals or quarters tested and expressed as a percentage either at animal or udder quarter level. Additionally, prevalence of clinical mastitis was calculated as the number of animals or quarters with clinical mastitis divided by number of animals or udder quarters examined.


Results

On-station testing

During the on-station testing, the pH-based test detected more SCM infected cattle as compared to the CMT. Apparent SCM prevalence results for the pH based kit was 43.8 % which was significantly higher than that for CMT (26.0%) as shown in Table 1.

Table 1. Sub-clinical mastitis test results using CMT and pH-based mastitis test for udder quarter milk samples of cattle at DRI Naivasha

Test

SCM
positive

Total milk
samples

Prevalence %
(95% CI)

pH-based

42

96

43.8 (34.3 - 53.7)

CMT

25

96

26. 0 (18.3 - 35.7)

95% CI denotes confidence interval.

When the tests for agreement were calculated, there was 82.3% agreement between the pH based test and CMT. However, the Kappa statistic was 62%. The resultant positive predictive value (PPV) and negative predictive value (NPV) were 59.5% (95% CI: 43.3-74.4) and 100% (95% CI: 93.4-100).

Validation results in cattle and camels

A comparison of the SCM apparent prevalence results between the pH-based test and CMT for cattle and camels showed a high agreement (ranging from to 82.3 to 99%) between the two screening methods in their ability to correctly detecting cows and camels positive for SCM. Overall, the pH-based test was as efficient (ability to pick mastitis cases) as CMT in correctly detecting SCM in cattle and camels despite slight species variations observed (Table 2).

Table 2. On-farm validation results for SCM at udder quarter level in cattle at Kajiado and camels in Laikipia

Animal
Species

Test

SCM
positive

Total milk
samples

% (95% CI)

Cattle

pH based

149

386

38.6 (33.9 – 43.6)

CMT

151

386

39.1 (34.4 – 44.1)

Camels

pH based

19

374

5.1 (3.2 – 7.8)

CMT

18

374

4.8 ( 3.0 – 7.5)

The perfect agreement (> 98 %) between SCM results by pH-based kit and CMT is also shown by the high Kappa statistic for both cattle and camels (Table 3). The Kappa statistic for cattle was 0.96 (p<0.00), which is equivalent to that of camels. However, there was a variation in the estimated Positive Predictive Value (PPV) with that for cattle being significantly (p<0.05) higher than that in camels. On the contrary, the Negative Predictive Value (NPV) for the two species was over 99% and similar for both cattle and camels (Table 3).

Table 3. Agreement, Kappa, Positive predictive values and negative values for the pH based test kit and CMT in camels and cattle using functional udder quarters as sample size

Statistics

Camels
(Laikipia, n=374)

Cattle
(Kajiado, n=386)

Agreement

98.4%

98.5%

Kappa

0.96 (p<0.00)

0.96 (p<0.00)

Positive predictive value

84.2% (60.4-96.6)

100% (97.6-100)

Negative predictive value

99.5% (98.1-99.9)

99.5% (98.2-99.4)

Parentheses are 95% confidence intervals

Non-functional and atrophied quarters with blind teats were detected at animal level in 4% (4/99) of the camels, 12% (3/25) of cattle in Naivasha and 3% (4/136) of cattle in Kajiado. The same was noted (considering each animal has four udder quarters) for 1% (4/396) in camel quarters, 4% (4/100) in Naivasha cattle quarters and 1.3% (7/544) cattle quarters in Kajiado. Clinical mastitis was only diagnosed in cattle in Kajiado County where at cow level, 6% (4/136) of the lactating cattle were infected and with udder quarter infection rate of 3% (16/544).


Discussion

The results of the current study clearly demonstrated that it is possible to simplify mastitis diagnosis by developing and availing to farmers an easy to use, cheap and rapid pen side test kit for the timely detection and control of mastitis at farm level. The kit described in the current study was simplified for use by smallholder dairy farmers as well as pastoralists. The pH-based test correctly and accurately detected SCM cases in near total agreement with CMT. This agreement was manifested for both cattle and camels. A comparison of agreement between different SCM screening tests reported that Kappa values <0.4 indicate poor agreement, values between 0.4 and 0.75: fair to good agreement, and values >0.75: excellent agreement (Sharma et al 2010). The on-station testing on cattle at the DRI Naivasha resulted in Kappa statistic estimate of 0.62 demonstrating good agreement between the pH based test kit and CMT as per the cut-offs established by Sharma et al (2010). On the other hand, the on-farm validation resulted in excellent agreement between the test kit and CMT in cattle and camels since the kappa static was >0.75. The low Kappa value for Naivasha can be attributed to the fact that the pH-based test was able to diagnose more cases of SCM compared to CMT. Further, it should be noted that the study animals in Naivasha were composed of different breeds and it could not be determined what effects this has on the performance of the pH-based test. A future study will investigate the effects of breed on the kit output.

In this study CMT was used to validate the pH-based pen-side mastitis test in detecting SCM in both camels and cattle. Although CMT has been used for more than 50 years and continues to be the most accurate farm screening test for SCM (Ruegg and Reinemann 2002), its adoption by farmers in developing countries has been low. Although CMT is also useful in detecting SCM in camels (Wanjohi et al 2013), its application is limited due to inherent challenges such as unavailability of potable water for cleaning the puddle between tests. It is thus expected that the pH based test kit will not only be an alternative test to the CMT but will be the test of choice in areas without potable water. Previous studies involving measurement of milk pH as an indicator of SCM involved use of pH-meter to directly determine the pH of milk (Leslie et al 2002; Shahid et al 2011). Comparative SCM diagnostic studies that had used pH-chemical indicators to indirectly determine milk pH, like was the case in the current study, were unavailable.

It was worth noting that this study did not endeavor to estimate the sensitivity and specificity of the newly developed pH-based mastitis test particularly for camels because in this species there is no established gold standard like somatic cell counts (SCC) used in cattle. It is thought that the somatic cell count (SCC) could be the appropriate gold standard even in camels. However, since our sampling technique did not define the health status of the study animals by establishing a baseline for SCC cut-offs prior to commencement of the study, further research work in camels to recommend SCC thresholds for health and diseased udders is recommended for use as standards for future work.

In absence of a gold standard then predictive values for the pH-based test were determined. In general terms, predictive values are used to interpret the results of a test by examining the correct classification of individuals by the test. This measure is valuable because whether an animal is truly a case or non-case is difficult to know (without determining sensitivity or specificity), but a positive or negative result of a test is known (Kelly and Firestein 2017). The results of this study showed that the pH-based test had a high positive predictive value in cattle in Kajiado (100%) and camels in Laikipia (84.2%). This implied that the pH-based test was able to detect all the cases of SCM in cattle in Kajiado that were diagnosed using both tests, while for camels it missed 15.8% of SCM cases diagnosed using both tests. This variation may be attributed to differences in prevalence of SCM among the two species where it was higher in cattle in Kajiado than camels in Laikipia, since predictive values of diagnostic or screening tests, are influenced by the prevalence of disease (Mausner et al 1985). On the other hand a negative predictive value is a proportion of the non-cases identified out of all negative test results. The results of the current study revealed that the pH-based test had the same negative predictive value in both cattle and camels of 99.5%. This means that the pH-based test and CMT had equal ability to pick non-mastitic cases in cattle and camels. Generally, when positive and negative predictive value of a test is higher (as close to 100 as possible), then it suggests that this new test is doing as good as conventional test, which in this case is CMT (Parikh et al 2008).

The findings of the current study further demonstrated that mastitis is a problem in both camels and cattle in the three study sites. The two mastitis diagnostic tests were able to detect SCM in lactating animals from all the three study sites. Prevalence of SCM by CMT in camels and cows was comparable to that by the pH-based test. Actually SCM prevalence results obtained using pH-based test were thus consistent or comparable with those from past studies where CMT was used (Matofari et al 2003; Ondieki et al 2013; Iyer et al 2014; Ndirangu et al 2017).

Additionally, clinical mastitis (CM) was diagnosed only in cattle from Kajiado. In this herd, sub-clinical form of mastitis, as detected using both pH-based test and CMT, was more common than CM. This is consistent with findings of Ondieki et al (2013).The most probable reason for this is that SCM usually receives little attention and efforts are concentrated on treatment of clinical cases. This implies that cows with clinical mastitis are promptly treated and thus clinical cases are not routinely found in the herd during most cross-sectional studies such as the current one which was a one-off study. In addition, since SCM is not visible to the naked eyes, farmers cannot easily decide on what order to milk mastitic animals. This can lead to the spread of SCM within the herd and the ensuing severe economic losses.

Study results further revealed presence of blind teats, which was characterized by presence of atrophied and dysfunctional udder quarters, among camels and cattle from the three study sites. The high occurrence of blind and dysfunctional mammary quarters in camels and cattle reported in this study, which results in reduced milk production with subsequent impact on food security, signifies the importance of the problem. Lack of screening and appropriate treatment of SCM, inadequate follow-up of clinical cases and persistent challenges of mammary glands by microbial pathogens could be the main predisposing factors to quarter blindness. More studies to determine productivity losses incurred by mastitis are recommended, given the high proportion of non-functional udder quarters.


Conclusions


Recommendations


Acknowledgements

The study was funded by European Union, ASAL-APRP. The authors are grateful to the management of KALRO-DRI Naivasha for allowing us to use their cattle during on-station validation. In addition, the management and technical staffs in KALRO-Veterinary Research Centre, Muguga are highly appreciated for their inputs during the field survey.


References

Abebe R, Hatiya H, Abera M, Megersa B and Asmare K 2016 Bovine mastitis: prevalence, risk factors and isolation of Staphylococcus aureus in dairy herds at Hawassa milk shed, South Ethiopia. BMC Veterinary Research 12, 270-281.

Dingwell R T, Leslie K E, Schukken Y H 2003 Evaluation of the California Mastitis test to detect an intrammamary infection with a major pathogen in early lactation. Canadian Veterinary Journal 44, 413-415.

Hussain R, Javed M T and Khan A 2012 Changes in some biochemical parameters and somatic cell counts in the milk of buffalo and cattle suffering from mastitis. Pakistan Veterinary Journal 32: 418-421

Iyer A P, Albaik M and Baghallab I 2014 Mastitis in Camels in African and Middle East Countries. Journal of Bacteriology and Parasitology  5:188. doi: 10.4172/2155-9597.1000188

Kasozi K I, Matovu E, Tayebwa D S, Natihwera J, Mugezi I and Mahero M 2014 Epidemiology of increasing hemo parasite burden in Uganda cattle. Open Journal of Veterinary Medicine 4, 220-231.

Kelley and Firestein 2017 Textbook of Rheumatology (Tenth Edition) Volume 1, 2017, Pages 717-733

Koskinen M T, Wellenberg G J, Sampimon O C, Holopainen J and Rothkamp A 2010 Field comparison of real-time polymerase chain reaction and bacterial culture for identification of bovine mastitis bacteria. Journal of Dairy Science 93: 5707-5715.

Laikipia County report 2016 Directorate of Livestock and Fisheries, Laikipia County 2015/16 annual report.

Leslie K E, Jansen J T and Lim G H 2002 Opportunities and implications for improved on-farm cow-side diagnostics. Proceedings of De Laval Hygiene Symposium, pp. 147.

Maichomo M W, Kibugu J K, Kurgat R, Malonza V M and Youman M 2014 Effect of subclinical mastitis caused by ss-haemolytic streptococci on milk yield in Kenya camels (Camelus dromedarius). Bulletin of Animal Health and Production in Africa 62, 275-282.

Matofari J W, Mario Y, Mwatha E W and Okemo P O 2003 Microorganisms associated with subclinical mastitis in Kenyan camels (Camelus dromedarius). Journal of Tropical Microbiology, 2:11-16.

Mausner J S and Kramer S 1985 In Mausner and Bahn Epidemiology: An Introductory Text. Philadelphia, WB Saunders, p. 221.

MoL&F 2015 Ministry of Livestock and Fisheries, Loitokitok sub-County Livestock department annual report for 2015.

Mungube E O, Tenhagen B A, Regassa F, Kyule M N, Sheferaw Y, Kassa T and Baumann M P O 2005 Reduced milk production in udders with sub-clinical mastitis and associated economic losses in crossbred dairy cows in Ethiopia. Tropical Animal Health and Production 37(6): 503-512.

Ndirangu P N, Wesonga H O, Siamba D, Mugambi J M and Nginyi J M 2013 Mastitis therapy in Western and Central parts of Kenya: veterinary service providers’ perspective. In: Proceedings of EAAPP mini-conference and stakeholders open day (Muinga R, Wandera F, Ouda J, Maina J, Ilatsia E, Kariuki J, Onyango T.and Wainaina R eds) 12-15 November, 2013. RDCoE/Morendat Training Centre, Naivasha, Kenya. Pp 149-154.

Ndirangu P N, Siamba D, Wesonga H O, Mungube E O, Maichomo M W and Mugambi J M 2017 Prevalence of bovine mastitis and multi-antibiotic resistant Staphylococcus and Streptococcus species in a research centre farm at Naivasha, Kenya. Bulletin of Animal Health and Production in Africa 65: 289-301.

NMC, National Mastitis Council 2013 Recommended mastitis control program. http://www.nmconline.org

Odero-Waitituh J A 2017 Smallholder dairy production in Kenya; a review. Livestock Research for Rural Development. Volume 29, Article #139. Retrieved October 31, 2018, from http://www.lrrd.org/lrrd29/7/atiw29139.html

Oliveira C M C Sousa M G S, Silva N S, Mendonca C L, Silveira J A S, Oaigen R P, Andrade S J T and Barbosa J D 2011 Prevalence and etiology of bovine mastitis in dairy region of Rondon do para. Pesquisa Veterinária Brasileira 31(2): 104-110.

Ondieki J O, Ogore P B, Shakala E K and Kaburu G M 2013 Prevalence of bovine mastitis, its therapeutics and control in Tatton Agriculture Park, Egerton University, Njoro District of Kenya. Basic Research Journal of Agricultural Science and Review 2(1): 15-20.

Parikh R,Mathai A,Parikh, S, Sekhar, C and Thomas R 2008 Understanding and using sensitivity, specificity and predictive values. Indian Journal of Ophthalmology 56(1): 45–50.

Quinn P J, Carter M E, Markey B and Carter G R 1999 Clinical Veterinary Microbiology, Mosby: London, UK. pp. 21-66.

Radostits O M, Gay C C, Hinchliff K W, Kelton D F and Scholl T 2007 Veterinary Medicine . A Textbook of the diseases of cattle, sheep, pigs, goats and horses. 10 th ed., Saunders Elsevier, Spain. Pp. 1045-1046.

Ruegg P L and Reinemann D J 2002 Milk quality and mastitis tests. Bovine Practice 36(1): 41-55.

Schalm O W, Carroll E J and Jain N C 1971 Bovine mastitis: A symposium. Lea and Febiger, Philadelphia, PA, USA, 24: 182-208.

Schukken Y H, Wilson D J and Welcome F 2003 Monitoring udder health and milking quality using somatic cell count Veterinary Research 34: 579-596.

Shahid M, Sabir N, Ahmed I, Khan R, Irshad M, Rizwan M and Ahmed S 2011 Diagnosis of subclinical mastitis in bovine using conventional methods and electronic detector Asian Research Publishing Network (ARPN) Journal of Agricultural and Biological Science 6(11): 18-22.

Sharma N, Pandey V and Sudhan N A 2010 Comparison of some indirect screening tests for detection of subclinical mastitis in dairy cow. Bulgarian Journal of Veterinary Medicine, 13(2): 98-103.

Viera A J and Garnett J M 2005 Understanding inter-observer Agreement: The Kappa statistics. Research series for Family Medicine 37(5): 360-366.

Viguier C, Arora S, Gilmartin N, Welbeck K and O’Kennedy R 2009 Mastitis detection: current trends and future perspectives. Trends in Biotechnology 27: 486-493.

Wanjohi M, Gitao C G and Bebora L 2013 Subclinical mastitis affecting hygienic quality of marketed camel milk from North-Eastern Province, Kenya. Microbiology Research International 1(1): 6-15.


Received 28 November 2018; Accepted 19 December 2018; Published 1 January 2019

Go to top