There appears to be little consensus on the extent and the exact nature of the dairy cow welfare problem. In the present paper the welfare of cows is described in the light of metabolic stress. The duration of lactation relative to level of production is highlighted as the main factor which sets the dairy cow apart from other lactating mammals. A conceptual diagram of the components and effects of metabolic load is given, a number of questions are raised and two aspects of dairy production are highlighted which relate to the welfare of modern dairy cows. These relate to decreased fertility in high yielding cows and the cumulative effects of successive lactations. Finally, the need for the dairy industry to participate actively and openly in the welfare debate is emphasized.

Ethological and ecological studies of wild animals are producing evidence for metabolic stress during courtship, breeding and parental care comparable with that of domestic livestock. Resistance to disease may be compromised by the demand for fatty acids and proteins during reproduction and even more during lactation. The adipose tissue around major lymph nodes is indistinguishable histologically from that in larger depots. In vitro and in vivo studies reveal that it is specialized to respond to lipolytic agonists secreted by lymphoid cells but is insensitive to the endocrine conditions of short-term fasting. These properties enable it to provision adjacent immune cells. Such adipose tissue may act as a forum for competing demands of mammary glands, muscles etc. and local defences against pathogens. Glutamine is essential to the nutrition of the immune system and is used by the mammary gland. Muscle is the best known source but adipose tissue also participates in glutamine metabolism and may become more important in animals in which the musculature is wasted through prolonged lactation.

The authors review their experiences of metabolic profiles in dairy herds, with a view to assessing whether metabolic stress is a problem in Great Britain at present. Many cows show elevated blood beta-hydroxybutyrate concentration, indicating energy deficit, or elevated urea levels, indicating an imbalance between energy and protein in the rumen but at present there is no evidence that high-yielding cows in commercial herds show more metabolic stress than low-yielding cows. The authors suggest that more cows could suffer metabolic stress in the future, unless farmers’ ability to feed and manage dairy cows develops as rapidly as genetic selection for high milk yield.

The strategy most widely adopted to improve milk production efficiency is to increase yield per cow. To date, this has been achieved primarily through genetic selection and improved nutrition. Achievement of very high individual yield has had its down-side, especially in terms of reduced reproductive efficiency and there is now quite widespread concern that the high genetic merit cow is at greater risk of metabolic disease than her unimproved counterpart. To quote from the recent Farm Animal Welfare Council Report on Dairy Cow Welfare (FAWC, 1997): ‘High metabolic turnover in cows can be associated with a greater risk of mastitis, lameness, infertility and other production diseases…’. Whilst there can be little doubt that metabolic turn-over is indeed higher in high merit cows, it is not safe to assume that this necessarily equates with more risk; metabolic turn-over is higher in an elephant than in a mouse but risk is certainly not. Metabolic load might be a better term to use. If we think, simplistically, of this being the ‘strain’ on a system it is logical to expect an inverse relationship between metabolic load and health. The extrapolation to high genetic merit cows being at greater risk then presupposes that they experience an increased metabolic load but there has been no rigorous evaluation of whether this is so. In this review we will consider what is meant by metabolic load, examine in qualitative theoretical terms what degree of load might be expected from different commercial systems and present some recently obtained data which addresses directly the question, is metabolic load greater in high genetic merit cows?

The dip in food intake, which starts in late pregnancy and continues into early lactation, has traditionally been interpreted as a depression in intake due to physical constraints. However, the rôle of physical constraints on intake has been overemphasized, particularly in early lactation. There is mounting evidence that the presence and mobilization of body reserves in early lactation play an important rôle in regulating intake at this time.

Conceptually, the dip in intake in early lactation observed when cows have access to non-limiting foods can be accounted for by assuming that the cow has a desired level of body reserves. When the cow is not compromised, the changes with time in body reserves and the dip in intake represent the normal case and provide the basis against which to assess true depressions in intake which may occur when the cow is compromised by limiting nutrition or environment.

The regulation of body reserves and intake in the periparturient cow is orchestrated through nervous and hormonal signals. Likely factors that are involved in intake regulation are reproductive hormones, neuropeptides, adrenergic signals, insulin and insulin resistance and leptin. Furthermore, oxidation of NEFA in the liver may result in feedback signals that reduce intake. The relative importance of these is discussed. A better understanding of the physiological signals involved in intake regulation and their interrelations with body weight regulation may provide important indicators of the degree of compromise that periparturient cows may experience.

There have been substantial advances in recent years in methods for genetic analysis of traits that are expressed repeatedly over time, for example milk yield on successive test days during lactation. The background to the methods, notably random regression, covariance functions and splines, are outlined. The utility of these methods for analysing functional data on which individual records on cows are few but sire family records that span the lactation, is reviewed. Methods of analyses for measures of herd life are discussed and that being adopted in the UK is outlined.

The objectives of this study were to investigate genetic variation for traits that are part of the food utilization complex and to investigate the scope for future genetic improvement of traits possibly linked to metabolic stress: live weight (change), condition score (change) and energy balance. Many aspects of the food utilization complex appear to be heritable and are affected by genetic selection for yield. In general, genetic selection for yield increases intake and body tissue mobilization and energy balance is expected to decrease. However, unfavourable effects of genetic selection can be compensated for by measuring additional traits to be included in breeding programmes. Food intake, live weight (change) and condition score (change) are all potential options. Which traits should be measured, at what lactation stages and in which (nutritional) environment will merely depend on the coheritability with health and fertility, the genetic correlation with milk yield and the cost of measuring the trait effectively in a breeding programme.

Lactation persistency, often simply called persistency, in general can be defined as the ability to maintain yields during the lactation. Persistency has an impact on food costs, health, and fertility. Of these three components affected by persistency, the impact of persistency on health, i.e. metabolic stress of the cow leading to health problems, may be most important nowadays. Numerous suggestions for criteria of persistency exist. Often simple ratios of part-lactation yields, e.g. the ratio of yield in the first and last 100 days of lactation, have been used. New approaches have used results from the application of random regression test day models developed for the genetic evaluation of yield traits. Many studies unfortunately have neglected the effect of gestation on persistency but acknowledged that an improved persistency should lead to an improved reproductive performance. Both relationships should be considered in genetic analyses and recommendations for improvement of management decisions. Today the correct description of persistency plays a prominent rôle to obtain correct genetic evaluations based on test day yields. But, although apparently trivial, a direct genetic analysis of lactation persistency and even more an inclusion of this trait into selection programmes clearly is a complicated task. A reason for this, amongst others, is that management strategies for feeding during the lactation and handling of the reproductive performance that are most often not recorded, are likely to mask the real persistency. Future studies on the genetics of persistency should also seek a strong interaction of geneticists and physiologists as persistency is fundamentally confounded with the problem of metabolic stress. Today, a recommendation of the inclusion of persistency in selection programmes appears to be premature and more studies, e. g. on the association of persistency with longevity, could aid in this process.

The overall objective of a series of experiments to investigate ‘metabolic stress’ was to examine the relationships between ‘metabolic load’, disease and other parameters associated with the welfare of the dairy cow. In the main, these used several well controlled herd based studies complimented with more basic and strategic investigations. In this paper we compare and contrast practical aspects of health and welfare in two high genetic merit herds managed at the extremes of inputs and outputs for dairy farming in south-west Scotland. The hypothesis was that high output herds would have more health and welfare problems than low input herds. Two herds (70 Holstein-Friesian cows each) at SAC Acrehead Dumfries of a similar genetic background (overall in the top 5% of UK cows by PIN and ITEM), were housed in identical buildings and tended by the same herdsman. Both herds had autumn- and spring-calving cattle. The ‘low input’ herd (LI) was given a minimum of concentrate (approx. 0.5 t per cow per year) and milked twice a day and had a restricted quota of 385 000 l. The ‘high output’ herd (HO) was managed for high yields (unrestricted quota) and was given concentrates (2 t per cow per year) and forage ad libitum and milked three times daily. In 1995-96 the sole source of winter forage was grass/clover silage (LI) or grass silage (HO) but in 1996-1998 ensiled cereal and fodder beet were included in both diets. ‘Metabolic load’ could only be inferred from overall inputs, milk outputs, weight loss, body condition score and behaviour. There were significant differences in 305-day lactation yields between herds, and season of calving especially in 1995-96 (LI autumn; 5952 l at 30 g/kg protein (P); LI spring; 5741 l, 32.5 g/kg P; HO autumn; 9541 l at 32.8 g/kg P; HO spring; 8402 l, 32.6 g/kg P). LI weight and body condition-score losses were greatest in this year and behavioural studies showed substantial differences in feeding time (HO < LI, P < 0.05) and total lying time (LI < HO; P < 0.05). However these differences were much less marked in subsequent years. There was a significant difference in the prevalence and incidence of clinical lameness between herds (HO > LI; P < 0.05) and season (autumn > spring P < 0.05) but not for mastitis or metabolic disease. An in-depth study of subclinical claw horn lesion development in first calving heifers showed significant differences between herds in 1996-97 (LI > HO, P < 0.05) but none in 1995-96. There was a significant difference for season in both years (autumn > spring, P < 0.05). Analysis of blood biochemistry parameters of samples taken at approximately 1 month after calving showed some significant differences between LI and HO generally indicating a greater ‘metabolic load’ for LI. Although the full effects of ‘metabolic load’ on immune function and reproduction are dealt with elsewhere our preliminary data showed no significant differences between herds for the former but some significant differences for the latter, in particular there were differences in aspects of the progesterone profiles between herds and more importantly between seasons. However these latter differences were not clearly reflected in conception rates. It was concluded that the hypothesis was not fully sustained and that both systems had pitfalls in terms of welfare. The three major areas causing difficulties for both systems were the need first to ensure adequate intake of forage; secondly to limit the environmental challenge to the feet and udder and finally to marry these systems to the factors limiting reproduction, primarily calving season and ability of reproduction management.

An effective method for enhancing milk production efficiency in dairy cows is to increase milk yield and significant progress has been achieved through intense selection, assisted by the application of new reproductive techniques. However this increased milk yield has been accompanied by a slow but steady decline in dairy cow fertility. The two main reasons for this reducing level of fertility appear to be selection for increased milk yield and large herd sizes, although the affect of the introduction of Holstein genes needs to be investigated. In addition, other negative consequences such as an increase in the incidence of metabolic diseases and lameness have been observed. This has given rise to public concern that the high-yielding dairy cow may be under a state of metabolic stress during peak lactation and therefore the welfare and performance of other body functions are compromised.

The reason for this decline in fertility is not well understood, although a nutritional influence on the initiation of oestrous cycles, follicular growth, oocyte quality and early embryonic development has been implicated. In early lactation dietary intake is unable to meet the demands of milk production and most cows enter a period of negative energy balance. Negative energy balance has a broadly similar effect to undernutrition leading to a mobilization of body reserves. Furthermore diets high in rumen degradable protein lead to an excess of rumen ammonia, which before it is converted to urea by the liver and excreted in the urine, may cause an alteration in the reproductive tract environment reducing embryo survival. Such major changes in the metabolic and endocrine systems can therefore influence fertility at a number of key points.

Possible reproductive sites where inadequate nutrition may have detrimental effects include: (i) the hypothalamic/pituitary gland where gonadotropin release may be impaired; (ii) a direct effect on the ovaries, where both follicular growth patterns and corpus luteum function may be directly influenced; (iii) the quality of the oocyte prior to ovulation may be reduced and coupled with an inadequate uterine environment will result in reduced embryo survival and (iv) there may be effects on subsequent embryo development. The initiation of normal oestrous cycles post partum is usually delayed in dairy cows with a higher genetic merit for milk production, confirming that intense selection towards high milk yield can compromise reproductive function. In addition, the effects of increased milk yield may include changes in circulating GH and insulin concentrations, which in turn alter both insulin-like growth factor (IGF) and IGF binding protein production. Nutrition has recently been shown to have a direct effect at the level of both the ovaries and the uterus to alter the expression of these growth factors.

In conclusion, further knowledge is required to determine how the metabolic changes associated with high milk output reduce fertility. Identification and understanding of the mechanisms involved and the key sites of action responsible for compromised reproductive function, will enable the identification of possible indices for future multiple-trait selection programmes.

Prolonged periods of stress have been associated with impaired immune function; the experiment reported here investigates a potential link between level of metabolic load and immune function in lactating dairy cattle. A group of 111 Holstein-Friesian dairy cows was used. The cows belonged to one of two genetic lines: a selection line (S) with high genetic merit for fat plus protein yield and an unselected control line (C). The cows were offered one of two silage-based total mixed diets containing either 200 g (LC) or 450 g (HC) of concentrate per kg dry matter. Combination of genetic selection and food gave four groups: S-LC, S-HC, C-LC and C-HC. All cows were inoculated with a live attenuated BHV-1 vaccine soon after parturition and the primary antibody response in whey monitored. The number of BHV-1 antibody positive cows was not significantly different between the four groups; but, the initial antibody response was lower in cows of high genetic merit which were given a low concentrate diet. Statistical analysis demonstrated that the contribution of diet to this effect was highly significant. One year later, again after parturition, the experiment was repeated, this time using serum as the test sample. The average antibody response of the BHV-1 antibody positive cows was not significantly different between the four groups but the number of antibody positive cows was group-dependent. In conclusion, diet type but not genetic merit for high fat plus protein yield made a highly significant contribution to the antibody response of dairy cows to BHV-1 vaccination, both initially and a year later.

Genetic selection for milk production has been very successful. However to achieve high yields, the metabolic load on dairy cows is believed to be substantial. If the size of this load is large enough then the animal may become ‘metabolically stressed’. Signs of this may include some sort of distortion of normal physiological function. There is evidence from both population studies and research herds to suggest that intense selection for milk yield has led to a deterioration in some aspects of health and fertility. Genetic correlation estimates between production and measures of fertility are unfavourable. As an example, calving intervals of high merit animals from Langhill are on average 12 days longer than those of average genetic merit, which is mostly due to a delay in days to first heat. It is suggested that some aspects of health and fertility problems in high genetic merit animals are a consequence, in part, of so-called metabolic stress. Future breeding goals should be broadened to include a broad spectrum of traits related to efficient milk production, in addition to either health and fertility traits themselves, or traits believed to be precursors of them, such as those related to metabolic stress. The complexity and subjectivity of metabolic stress and its components makes it very difficult to include in future breeding goals. However, traits related to energy balance, such as some measures of condition score, dry-matter intake and live weight may be useful in breeding programmes where one of the goals is to alleviate metabolic stress.

The aim of this paper is to define the critical periods within the lifetime of the cow where metabolic load is likely to have an influence on health, welfare and reproductive performance and to devise management strategies which will reduce the metabolic ‘load’ at those critical periods. The paper uses the data from the SAC Acrehead ‘metabolic stress’ study as its base. This systems study examines a low input (LI) and a high output (HO) strategy and their influence on financial performance, nutrient balance and the health and welfare of dairy cows. Milk yields ranged from 53371 (LI spring calving heifers) to 100191 (HO autumn-calving cows). By using weight and condition-score losses, reduced milk yield and metabolic profiles as indices of metabolic stress, two critical points were identified in the systems. These were in LI systems in years of low forage supply and spring-calving cows at grass in spring. The Acrehead study is remarkable in that it shows that cows of relatively high genetic merit have, in the main, a similar incidence of disease and level of immune function in both systems. However the general level of reproductive performance and incidence of lameness, although comparable with other studies, is disappointingly poor and reproductive performance at Acrehead has been declining over time. Thus there is reason for decreasing the degree of load in both systems. Strategies over the lifetime of the cow designed to overcome these problems in critical periods are examined. These periods are identified as rearing, pre-partum, calving and early lactation. Within each of these periods, management options to overcome load, such as feeding, housing and, where applicable, milking, are discussed. From the combination of systems results from Acrehead and reference to the literature it is concluded that, in the past, too much emphasis has been placed on examining the effects of nutrition alone on metabolic load and its implications. On the other hand, too little research has been conducted on the interaction between nutrition and the management of the cow, in terms of housing and grouping, food trough access and building design. It is also important to recognize that each day within the lifetime of the cow is not independent. Thus management during one period could have an influence during a subsequent period, not only on the likelihood of the cow to experience metabolic load but also on her ability to ameliorate its effects on health and reproduction.

The objective of this study was to investigate the effect of both cow genetic index (CGI) and feeding system on the performance of second lactation Holstein-Friesian dairy cows on grass-based feeding systems. There was no interaction between CGI and feeding system for any of the parameters measured. Cows of high genetic index (HGI) produced significantly higher yields of milk (P < 0.001), fat (P < 0.001) protein (P < 0.0001) and lactose (P < 0.001) than medium genetic index cows (MGI). CGI had no effect on the concentration of milk constituents. Averaged across four intake measurements the HGI cows had significantly (P < 0.001) higher grass dry-matter intake (GDMI) and total dry-matter intake (TDMI). Live weight was similar for both genotypes during lactation. The HGI cows had significantly (P < 0.05) higher live-weight loss in the first 10 weeks of lactation, significantly lower live-weight gain from week 10 to the end of lactation and higher (P < 0.05) live-weight gain during the dry period. Condition score was significantly lower with the HGI (P < 0.001) at all stages of lactation. There was a higher proportion (P < 0.05) of the HGI cows non-pregnant at the end of the 13-week breeding season. Feeding system had a significant effect on the yield of milk (P < 0.001), fat (P < 0.001), protein (P < 0.001) and lactose (P < 0.001). Feeding system B produced significantly higher (P < 0.05) milk yield and yield of constituents (when compared with feeding systems A and C). Over the period when feeding systems were being applied, feeding system C had significantly higher (P < 0.05) milk protein concentration. Feeding system had a significant effect (P < 0.001) on both GDMI and TDMI. Feeding system had no effect on live weight, condition score or fertility performance.

Energy balance is a function of dry-matter intake (DMI), live weight and milk yield over a certain time period. To investigate potential strategies to use genetic selection for the improvement of the negative energy balance, genetic co-variances were estimated among DMI, live weight and milk yield during the first 15 weeks of lactation (no.=628). Rather than estimating the full 45 by 45 matrix a random regression model was used to estimate a second order covariance functions for the additive genetic and permanent environmental effects. Fixed effects were test-day, a group effect and week in lactation. Estimates for the genetic covariance function demonstrated that a high level of milk yield is only moderately correlated with a high level of DMI (0.21) but very strongly correlated to an increase of intake (0.97) and a loss of live weight (-0.46) during the first 15 weeks of lactation. Levels of weight and intake were correlated strongly (0.81). Estimates for the genetic correlations between weeks 1 and 15 were 0.79, 0.34 and 0.83 for milk yield, DMI and live weight respectively. DMI during early lactation was negatively correlated with milk yield but DMI during the later weeks was positively correlated with milk yield. The implication is that when selection is for a linear combination of milk yield, DMI and live weight (i.e. energy balance or efficiency) the moment in lactation of measuring each trait on the cow is of importance

Disease incidence in dairy cattle is to be reduced for animal welfare and economical reasons. This should be achieved not only by improvement of management but preferably also by genetic means. This study looks at the possibility of decreasing disease incidence in first lactation cows by increasing food intake. The latter is not measured on the cows directly but on young bulls during their performance test. Data consisted of 2203 Danish Red, 4527 Danish Black and White and 1022 Danish Jersey potential AI-bulls and 56 494 Danish Red, 264107 Danish Black and White and 57 661 Danish Jersey first lactation cows. Measures of food intake were provided by two Danish performance test stations. Information on incidence of mastitis, retained placenta, metritis, sole ulcer and ketosis as well as calving interval and energy corrected milk yield of first lactation cows was based on data extracted from the national data base in Denmark. Genetic (co)variances were estimated using restricted maximum likelihood. Heritability estimates of disease incidence and calving interval were low, ranging from <0.01 to 0.13 depending on breed. Heritability estimates of energy corrected milk yield were in the range of 0.28 to 0.33. In all breeds, an unfavourable genetic relationship between milk yield and disease incidence was found, while genetic correlations between food intake and ketosis were favourable, ranging between -0.03 and -0.25. Fertility disorders had an inconsistent correlation with food intake traits across breeds. Food intake of bulls could be included in the selection process in order to avoid nutrition-related disorders like ketosis.

Functional traits related to costs are currently of interest for selection and management of dairy cattle. The present study was aimed to estimate heritability for body condition score (BCS) and heart girth (HG), to investigate the genetic relationships between BCS, HG and milk-yield traits using a test-day model and to analyse the consistency of the estimates in different lactation stages. Cows from 25 dairy herds were scored for BCS and measured for HG at 3-month intervals for 2 years. Approximately 5000 test-day observations on BCS, HG and milk fat and protein yield from 1429 Italian Friesian cows were analysed using two approaches: (1) repeated observations were treated as repeated measurements of the same trait, both within and across lactations; (2) observations collected in different stages of lactation (dry period, 1 to 75 days in milk (DIM), 76 to 130 DIM, 131 to 210 DIM, 211 to 300 DIM) were treated as different traits. (Co)variance components and related parameters were estimated using REML multiple-trait procedures and unequal design animal models.

Heritability estimates (approach 1) for fat and protein test-day yield, BCS and HG were 0.22, 0.18, 0.29 and 0.33, respectively. BCS was negatively correlated with yield traits (-0.43 and -0.48 for fat and protein yield, respectively) but positively correlated (0.33) with HG. Genetic relationships between HG and milk-yield traits were negligible. Heritability estimates (approach 2) were 0.28 and 0.27 for BCS recorded in the first half of lactation (1 to 75 and 76 to 130 DIM, respectively), 0.36 for BCS measured on cows in the second half of lactation and 0.32 for BCS recorded on dry cows. Heritability estimates for HG in different lactation stages ranged from 0.31 to 0.40. Genetic correlations between BCS measured in different lactation stages were generally high (0.85 or more), with the exception of the correlation between the first and the last stage of lactation (0.74) and of the relationships between the beginning of lactation and the dry period (0.7). Genetic correlations between HG measured in different lactation stages were mostly higher than 0.80.

Five size traits (body weight, heart girth, hip height, body depth and rump width) and three yield traits (milk, fat and protein) of 7192 heifers were analysed on a monthly and 305-day lactation basis. The additive genetic variation was moderate to high and changed over time. Correlations between observations measured in different months were all >0.89. Covariance methodology was used to reduce the number of parameters to estimate the variance components. Correlations between size and yield traits ranged from -0.14 to 0.26; this study did not indicate that these correlations were time-dependent.