Investigating the Relationship Between Somatic Cell Count and Milk Production

Reader Impact Factor Score
[Total: 1 Average: 5]

Published on International Journal of Agriculture & Agribusiness
Publication Date: June 18, 2019

Odumboni, A. A., Ekeocha, A. H. & Fernando Da’mata
Federal University Oye Ekiti
Ekiti State, Nigeria

Journal Full Text PDF: Investigating the Relationship Between Somatic Cell Count and Milk Production (Studied in Hartpury College Dairy Herd with Sub-Clinical Mastitis).

Abstract
Mastitis has been recognised as a major disease affecting dairy cattle and jeopardising milk production and quality in commercial herd especially in its subclinical and clinical form. It has been recognised to use Somatic Cell Count (SCC) in the detection of Sub-clinical mastitis. This research study evaluates the relationship between Somatic Cell Count and Milk yield at Hartpury College while considering main factors such as stages of lactation, parity, and effect of herd season, sanitary conditions and the transformed Logarithm of Somatic cell Count as a covariate in this study. 3,969 data records including dairy herd with suspected cases of sub-clinical mastitis were used and healthy cows as control. All data were analysed using mixed models analysis from SPSS software. From the results obtained, stages of lactation and sanitary conditions had a significant effect on milk yield (p < 0.05). Also there was percentage reduction of 27% from animals with SCC < 100,000 cells/mL; 33% milk reduction from animal having SCC > 100,000 cells/mL and 55% from animal having SCC > 350,000 cells/mL. The effect of herd season and parity showed no significant result on milk yield. However there could have been more significant result on parity and effect of herd season if other dairy farms were considered for this retrospective study that investigates the relationship between Somatic Cell Count and milk yield on the effect of parity and effect of herd season.

Keywords: Somatic cell count, milk yield, parity, effect of herd season, lactation and sanitary conditions.

1. INTRODUCTION
Mastitis has been recognised as a major disease affecting dairy cattle and jeopardising milk production and quality in commercial herd, especially in its subclinical and clinical form (Petrovski et al., 2006). The measurement of prevalence and incidence of mastitis, especially sub-clinical mastitis in the milk of cows is known as Somatic Cell Count (SCC). The Somatic cells present in milk are mostly macrophages, which resides in the teat cisterns. These macrophages help in recognising bacteria invasion in the milk (Philips, 2010). Somatic cell counts of milk samples have gained wide recognition in the detection of subclinical mastitis and clinical mastitis, which mostly are the cause of milk reduction in dairy herds.
Mastitis is an infection of the mammary gland, a disease of great economic importance resulting in significant economic losses in the dairy industry (Juozaitiene et al., 2006). The symptoms of mastitis includes normal or elevated body temperature, inflammation of the udder and on milk testing, there is high somatic cell count.
According to Hagnestam-Nielsen et al. (2009), Subclinical mastitis is associated with increased somatic cell count and has several negative consequences. The risk of clinical mastitis (CM) increases with increasing somatic cell count (Beaudeau et al., 1998; Steeneveld et al., 2008). Sub-clinical mastitis is 15 to 40 times more prevalent than clinical mastitis and causes high economic losses in most dairy herds (Schultz et al., 1978).
Subclinical mastitis is associated with increased somatic cell count and has several negative consequences; also the risk of clinical mastitis increases with increasing somatic cell count, a high somatic cell count is often considered as an indicator for an infection in the udder. The herd-level economic loss brought about by subclinical mastitis is significant and has been reported to have a larger negative effect on any dairy herd than that caused by clinical mastitis (Huijps et al., 2008).
In addition, high somatic cell count is associated with reduced processing ability for milking;
a. It is reduced because of damage to the sensory tissue of the teat,
b. It also decreases the shelf-life of consumer milk (Ma et al., 2000).
Reduced animal welfare, economic losses and poor milk quality, consequently, form the incentives to reduce incidence of sub-clinical mastitis. The main component of the economic loss associated with sub clinical mastitis is reduced milk production of the affected dairy cows (Huijps et al., 2008). Thus, the cost of clinical and sub clinical mastitis will largely depend on the extent of the associated milk yield loss.
The aim of this retrospective study is to investigate the relationship between Somatic Cell Count, milk productivity and subclinical mastitis and thus to identify possible areas where average monthly milk yield could be improved in order to gain economic turn-over for the dairy unit; and also to investigate whether the occurrence of sub-clinical mastitis in the dairy herd can be reduced by measuring the somatic cell count within the milk.
The objectives of this research study are;
a. To assess the reduction in milk yield associated with increased somatic cell count in the context of the different stages of lactation, parity and effect of herd season on the dairy herd.
b. To evaluate the test day milk loss/gain at a somatic cell count ≤ 500,000 cells / mL in order to bring about an increased economic turn-over for the dairy unit
c. To quantify the impact of sanitary conditions as a factor on how it affects somatic cell count and milk productivity.
d. To measure the amount of change in 305-day milk yield associated with an increase in lactation average somatic cell count using computer based records at the college dairy farm.
Null Hypothesis;
a. There is no significant difference between somatic cell count and milk productivity.
b. There is no significant difference between stages of lactation in relation to milk productivity.
c. There is no significant difference between parity and milk productivity.
d. There is no significant difference between the effect of herd season and milk productivity.
e. There is no significant difference between the levels of sanitary conditions and milk productivity.

2. LITERATURE REVIEW
2.1 The changing epidemiology of Bovine mastitis
In the early 1940s, the average dairy herd size in the United Kingdom was about 15 cows, which would have had an estimated 23 cases of mastitis annually due mainly to Streptococcus agalactiae and Streptococcus aureus (Bradley, 2003). The mean somatic cell count was probably around 750,000 cells/mL. However, there was great optimism that penicillin was about to eradicate mastitis, but it was not until the 1960s that real progress was made in the control of the disease. The plan called for a five pronged approach to the management of mastitis, namely rapid identification and treatment of both subclinical and clinical cases of mastitis, routine whole herd antibiotic dry cow therapy, post milking teat disinfection, culling of chronically affected cows and the routine maintenance/management of the milking machine.
It was the uptake of this plan that resulted in rapid progress in control of both clinical and sub-clinical mastitis in the UK (White, 2006). In addition to the initial benefit of implementation of the Five-Point Plan, a number of factors have added impetus to UK mastitis control programmes. The most notable of these were the implementation of EC Milk Hygiene Directive (92/46) that imposed an upper limit of 400 000 cells/mL in bulk milk for human consumption and the economic incentives offered to farmers, by milk purchasers, to produce milk of higher quality with a lower SCC (Andrew, 2009). The impact of the implementation of mastitis control strategies, and in particular the Five-Point Plan, has been very successful in controlling the contagious pathogens and has led to a substantial reduction in the incidence of clinical and subclinical mastitis and bulk milk somatic cell counts (BMSCC) (Booth, 1997). The historical change in both subclinical and clinical mastitis incidence and its causes is illustrated in Table 1.

Table 1: Incidence and aetiology of clinical and sub-clinical mastitis in UK dairy herds (Quarter cases/100 cows/year) (Green, 2001).
Over the years, there has been a dramatic fall of the overall incidence of mastitis especially sub-clinical mastitis from over 150 to 40 cases per 100 cows per year and the incidence of the contagious mastitis pathogens (Wilesmith et al., 1986). Over the same period of time, the average Bulk Milk Somatic Cell Count (BMSCC), fell from over 600,000 cells/mL to just 400, 000 cells/mL (booth). This fall is reflected in the distribution of herds between different cell count bands illustrated in table 2 below (M. Blanshard, personal communication, 2001).

2.2 Subclinical mastitis
In most surveys, approximately one third to one half of cows has inflammatory infections, with the vast majority of infections being subclinical. Sub-clinical infections is usually accompanied by an increase in somatic cell count and reduced milk productivity (Vernooy, 2004), and the extent and duration of which it depends on the causative pathogen, and the effectiveness of host defence mechanism. Ceron-Munoz, (2002) states that mastitis occurs as a response to invasive agents can be characterised by an increase in SCC or logarithmic transformation in Somatic cell scores (SCS).
It has been established that SCC higher than 283×103 cells mL-1 indicates the presence of mastitis (especially in the sub-clinical stage).The reduction in milk yield associated with subclinical mastitis is usually estimated by extrapolating from crude somatic cell count (SCC) and milk yield loss is non- linear, whereas logarithmic transformation results in a linear relationship. For instance, twofold increase in crude SCC above 50,000/mL on monthly test day reports was associated with an average reduction in milk yield of 0.6kg/day for multiparous cows and 0.3kg/day for heifers (Hortet, 1999). Each unit increase in log10SCC on weekly test day reports was associated with an average reduction in milk yield of 2kg/day for multiparous cows and 1.3 kg/day for heifers (Koldeweij, 1999). In the United Kingdom and the United States, linear score (LS), a log-based transformation of SCC that results in scores of 0-9; parameter is usually used as a measure. Each unit increase in lactation average LS above 2 results in an average milk yield reduction of 0.7kg/day or 180kg/lactation in multiparous cows, with losses for heifers being approximately half (Reneau, 1986). Such association provides a rough estimate of the magnitude of milk yield loss associated with sub- clinical mastitis in a given herd, but do not accurately account for the milk yield loss in individual cows.
Sometimes, herd milk yield loss is estimated from the SCC of the bulk tank (BTSCC). For example, a decrease in milk yield equating to a loss of at least £59/cow/year was demonstrated in herds with BTSCC greater than 200,000mL (Ott & Novak, 2001). However, one problem with using BTSCC to estimate milk yield loss is the dilutional effect of high yielding dairy cows on BTSCC.

2.3 Somatic Cell Count
Somatic cell count concentration measures can be collected from an individual quarter of the udder, from each cow or as a representative sample from the bulk milk tank. Measuring SCC in each quarter is the best method of obtaining information relating to SCC and sub-clinical mastitis. However, the short-coming of this method is that it is non-economically viable and therefore the other methods such as clinical mastitis test (CMT) are more predominantly used in UK dairy industries (Green et al., 2001)
SCC in dairy cattle is a tool for indicating infection status of individual cow. Somatic cell counts in dairy are particularly important to monitor and control as milk with higher SCC than 400,000 cells/ mL is not permitted for human consumption, thereby resulting in yield and financial loss for the herd (White, 2006).
There are three main demerits of high SCC in dairy cattle.
a. Milk is reduced because of damage to the sensory tissue of the teat, by about 2.5% for each 100,000 cell/mL above 200,000 cells/mL (Blowey & Edmondson, 1995).
b. The milk has increased lipase content and the lipid breakdown products give the milk a rancid taste.
c. The milk has a low casein content leading to reduced cheese yield.
If an animal has a high SCC, then it is assumed that the cow is fighting an infection. The cells in question are white blood cells (Leucocytes), which as in any animal, are a major part of the immune system response to infection entering the body. If the SCC of an animal is high but with no signs of mastitis, then the dairy cow is said to have sub-clinical mastitis. The opposite situation can also occur; if a cow has clinical signs of mastitis, its SCC can still remain fairly low or average (White, 2006). This makes recognition and diagnosis even more difficult. It must be noted that a cow’s cell count will fluctuate daily, which is the reason why daily and monthly recording of milk yield is very important (Rhodes, 2003).

2.4 Aim of the research study
The primary aim of this research study is to investigate the relationship between Somatic Cell Count (SCC) and milk yield, in context of the effects of stages of lactation, parity, effect of herd-season and sanitary conditions on milk yield at Hartpury College Dairy herd, Gloucester, United Kingdom
The specific relationships investigated were:
a. Somatic cell count and milk yield with BLnSCC (Transformed Somatic Cell Count).
b. Stages of lactation and milk yield.
c. Parity and milk yield.
d. Sanitary conditions and milk yield.
e. Investigating the reaction between parity and stages of lactation as both factors has effect on milk yield milk and the transformed SCC.
Percentage (%) reduction in milk with increased Somatic Cell Count

3. METHODOLOGY
3.1 Data
In total, 100 lactating cows suspected of being in a sub-clinical state of mastitis were chosen for the test experiment. Separately, 20 lactating cows for the control experiment were chosen from Hartpury College dairy cows with monthly records from 2007 to 2009, all resulting to an information of about 3500 points of data dairy cattle with suspected cases of sub-clinical mastitis and 700 clean/healthy dairy herd for control. All data resulting to 4200 statistical data information, but during the course of collecting the data, 231 data records were lost on milk yield and performance of both sub-clinical mastitic dairy cows and healthy cows. On each test day, SCC (×1000 cells/ml) and milk yield (from the udder quarters within 24 hour period) were recorded for all the lactating cows. All data were accessed from the computer system of Hartpury college dairy farm with the help of Alphro software programme which stores all weekly and monthly records of milk production, somatic cell count levels, milk quality and milk performance

3.2 Definitions of data
The statistical unit will be milk yield on a test month. Milk yield on the test month will be used as the dependent variable. SCC, stages of lactation on which SCC were recorded. SCC were divided into 1000 and then converted into natural logarithm (LnSCC) (Ali and Shook, 1980), to account for skewness to the right of SCC distribution. Stages of lactation were split into a variable comprising 3 classes (day 5 – day 105 for level 1; day 106 – day 207 for class 2 and day 208 – 305 for class 3). Parities were categorised into three levels (levels 1, 2 and 3). Three seasons were defined as (April to July for level 1; August to October for level 2 and November to March for level 3) in order to determine the effect of herd-season on milk yield. Sanitary conditions bactoscan units were also categorised into three levels (1 – 25% bactoscan for level 1; 16 – 35% bactoscan for level 2 and 36 – 50% bactoscan for level 3).
All data were analysed using analysis of variance (ANOVA) to look at the significant difference on the effect stages of lactation, parity, effect of herd-season, sanitary conditions on milk yield. Analysis of co-variance (ANCOVA) was used to analyse the significant difference on the effect of the transformed natural logarithm of somatic cell count (LNSCC) on milk yield. A mixed model comprising linear regression analysis was used to model all the effects of factors (LNSCC, stages of lactation, parity, effect of herd season and sanitary conditions) on milk yield.
The linear regression is written below,
Yijkl = m + BLnSCC + SLi + Hj + Pk + SCl + eijkl
Where Yijk is the milk yield on each test day,
m is the overall mean, lnSCC is the fixed effect of natural logarithm of SCC (X 1000 cells/ml),
SL is the fixed effect of class i of stages of lactation (3 levels),
Hj is the effect of herd-season j,
Pk is the effect of class k of parity (three classes),
SCl is the effect of sanitary conditions on milk yield and eijkl is the residual

3.3 Results for mixed model analysis
Result 1 – There is a significant difference between the stages of lactation and milk yield (p < 0.05). Result 2 – There is a significant difference between sanitary conditions and milk yield (p < 0.05). Result 3 – There is no significant difference between parity and milk (p > 0.05).
Result 4 – There is no significant difference between the effect of herd season and milk yield (p > 0.05).
Result 5 – There is no significant difference between the naturally transformed logarithm of SCC and milk yield (p > 0.05).

4. DISCUSSION
4.1 Stages of lactation
The result for stages of lactation using the mixed model analysis showed a significant difference between stages of lactation and milk yield (p < 0.05). As the lactation advances, the milk yield decreases signifying sufficient information for describing the graphical representation of the milk yield and stages of lactation (Thornley & France, 2007). Using the multivariate analysis (see appendix C) to describe the relationship of stages of lactation and the transformed SCC, there was significance difference (p < 0.05) between the two parameters meaning that as the stages of lactation advances from class 1 to 2, there was an increase in the Somatic Cell Count but as the stages of lactation advances from class 2 to 3), the Somatic Cell Count had a decrease. To further explain the result of the stages of lactation and the transformed SCC on milk yield, as stages of lactation advanced from early to mid-lactation (class 1 – 2), the SCC increased, signifying a decrease in the milk yield. However, from mid lactation to late lactation (class 2 – 3), there was a decrease in both milk yield and SCC. The latter statement does not support the results of Hagnestam-Nielson et al., (2009) which investigated the relationship between Somatic Cell Count and Milk Yield at different stages of lactation. Their findings showed that as SCC increased, the milk yield decreased towards the late lactation (Kelly et al., 2000; Rupp et al., 2000; Juozaitiene et al., 2004). Hagnestam-Nielson et al., (2009) further explained that bias is created if lactation mean values are calculated based on monthly Test Day (TD), because infections with a short duration may be missed. The data from this dissertation study was based on milk yield with monthly TD result and monthly SCC. Also, milk loss caused by environmental pathogens might therefore go undetected if estimates of lactational yield loss are based on monthly SCC records (De Haas et al., 2002). The decrease in monthly (TD) milk yield and Monthly SCC from the mid lactation to late lactation in the results of this study might be as a result of the undetected environmental pathogens.